Hindawi Publishing CorporationArchaeaVolume 2013 Article ID 456318 17 pageshttpdxdoiorg1011552013456318

Research ArticlePH1 an Archaeovirus of Haloarcula hispanica Related toSH1 and HHIV-2

Kate Porter1 Sen-Lin Tang2 Chung-Pin Chen2 Pei-Wen Chiang2

Mei-Jhu Hong2 and Mike Dyall-Smith3

1 Biota Holdings Limited 10585 Blackburn Road Notting Hill VIC 3168 Australia2 Biodiversity Research Center Academia Sinica Nankang Taipei 115 Taiwan3 School of Biomedical Sciences Charles Sturt University Locked Bag 588 Wagga Wagga NSW 2678 Australia

Correspondence should be addressed to Mike Dyall-Smith mikedyallsmithgmailcom

Received 7 December 2012 Accepted 7 January 2013

Academic Editor Shaun Heaphy

Copyright copy 2013 Kate Porter et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Halovirus PH1 infects Haloarcula hispanica and was isolated from an Australian salt lake The burst size in single-step growthconditions was 50ndash100 PFUcell but cell density did not decrease until well after the rise (4ndash6 hr pi) indicating that the viruscould exit without cell lysis Virions were round 51 nm in diameter displayed a layered capsid structure and were sensitive tochloroform and lowered salt concentration The genome is linear dsDNA 28064 bp in length with 337 bp terminal repeats andterminal proteins and could transfect haloarchaeal species belonging to five different genera The genome is predicted to carry49 ORFs including those for structural proteins several of which were identified by mass spectroscopy The close similarity ofPH1 to SH1 (74 nucleotide identity) allowed a detailed description and analysis of the differences (divergent regions) between thetwo genomes including the detection of repeat-mediated deletions The relationship of SH1-like and pleolipoviruses to previouslydescribed genomic loci of virus and plasmid-related elements (ViPREs) of haloarchaea revealed an extensive level of recombinationbetween the known haloviruses PH1 is a member of the same virus group as SH1 and HHIV-2 and we propose the namehalosphaerovirus to accommodate these viruses

1 Introduction

Viruses of Archaea (archaeoviruses [1]) show considerablediversity and encompass novel morphotypes not seen inBacteria or Eukarya Relatively few have been examined indetail partly because of the demanding growth requirementsof many extremophilic Archaea (particularly thermophiles)and also because genetic analysis is often technically difficultcompared to bacterial systems such as Escherichia coliAlthough viruses of thermophilic Archaea show the mostinnovative capsids and replication strategies [1] the viruses ofhalophilic Archaea (haloarchaea) are of increasing attentionas new isolates are found with unexpected properties Manyof the earliest reported haloviruses including all thosedescribed before 1998 are bacteriophage-like (Caudovirales)with typical head-tail capsids and linear dsDNA genomesThese include groups of related viruses such as the ΦH-likegenus (ΦH ΦCh1 and BJ1) [2ndash4] and the unassignedvirus group comprising of HF1 and HF2 [5] The first

spindle-shaped halovirus His1 was reported in 1998 [6 7]and the first round virus SH1 was described in 2003 [8 9]and electron microscopic studies indicate that these mor-photypes dominate in natural waters [10ndash12] Over the last 5years there has been a wonderful increase in the number andtypes of described haloviruses including further examplesof SH1-like viruses (eg HHIV-2 [13]) and a range of His2-related viruses that are now classified as pleolipoviruses (egHRPV-1 and HHPV-1 [14 15]) One example of the biologicalnovelty displayed by these archaeoviruses is the geometryof the SH1 capsid which was found to be of a previouslyundescribed type 119879 = 28 dextro [16]

Halovirus His2 has a 16 kb dsDNA genome with terminalproteins and probably replicates via an encoded protein-primed DNA polymerase [7] It is now known to be apleomorphic virus distinct from the spindle-shaped His1[17] When the genome sequence was first described it wasshown to be related to a cryptic plasmid (pHK2) ofHaloferaxlucentense [18 19] and to a number of genomic loci found in

2 Archaea

several different haloarchaea Later descriptions of halo-viruses HRPV-1 HHPV-1 and several others revealed a spec-trum of related viruses with very different genome structures(circular ds- and ssDNA) lengths and replication strategies[14 15] but they all share a similar set of capsid proteinsThe mechanisms underlying the movement of the capsidgenes between viruses having such different characteristicsand modes of replication (protein primed versus rollingcircle) remain unclear One possibility is that such modularrecombinations could be facilitated by previously describedgenomic loci of virus and plasmid-related elements (ViPREs)[20] Initially these did not appear to be simple provirusgenomes in varying states of decay but the description ofpleolipoviruses with genomes much smaller than His2 anddiffering replication strategy can explain many of them asvirus integrants On the other hand some of these genomicloci have expanded in length as more virus and plasmidsequences become available and their gene homologs can berecognised in flanking sequences For these we retain theViPRE epithet From the decades of the study of bacterio-phages it is well documented that related phages frequentlyrecombine (by both legitimate and illegitimate recombina-tion events) giving rise to mosaic genomes with varyingevolutionary histories [21 22] This process can be assistedby the modular arrangement of genes (eg capsid formationor replication) commonly found inmany virus genomes [23]The same appears true of haloviruses such as the large recom-bination event evident in the comparison of halovirus HF1and HF2 genomes [5] More generally there is good evidenceof widespread recombination from the study of archaealMCM helicase genes (often associated with mobile geneticelements) [24] and in the comparative genomics of archaealcaudoviruses [25]

The group of round haloviruses exemplified by SH1 hasbeen recently expanded by the descriptions of HHIV-2 andSNJ1 [13 26] Members of the SH1 virus group are related bytheir capsid proteins but differ in genome length and repli-cation strategy SH1 and HHIV-2 have linear dsDNA gen-omes (sim305 kb) with terminal proteins indicative of repli-cation by protein priming while SNJ1 has a circular dsDNAgenome of only 163 kbThis diversity of genome types withinthe same virus group closely parallels that of the pleolipo-viruses described above SH1 has been the most intensivelystudied including host range particle stability virion struc-ture genome sequence transcription mapping transfectionand the establishment of a method of genetic manipulation[8 13 16 27ndash31] The aim of the current study is to describea new member of this group PH1 Its virological characteris-tics genome sequence and major proteins are presented andcompared to other members of the SH1 virus group and togenomic loci containing related genes For convenience weprovide the group name halosphaerovirus to encompass theSH1 virus group currently consisting of SH1 PH1 HHIV-2and SNJ1

2 Results

21 Virus Isolation and Host Range A water sample fromPink Lake a hypersaline lake (32∘001015840 S and 115∘301015840 E) in

100 nm

Figure 1 Purified PH1 virus examined by negative-stain electronmicroscopy Particles were stained with 2 (wv) uranyl acetateScale bar represents 100 nm

western Australia was screened for haloviruses by platingdirectly on lawns of Har hispanica using overlay plates ofmodified growth medium (MGM) with 12 or 18 salts(wv) A high titre of similar plaque morphology was ob-served (12 times 105 PFUmL) on the plates with 18 (wv) saltsA novel halovirus was isolated from one of these plaquesand designated PH1 based on the source (Pink Lake) and theisolating host (Har hispanica)

Plaques were fully developed after two days at 37∘Cusing overlay plates of 18 (wv) MGM and were 1-2mm indiameter clear and with ragged edges At 30∘C plaques tookthree days to develop and at 25∘C plaques were hazy andtook seven days to develop (data not shown)

The host range of PH1 was identical to that of SH1 [8] itwas unable to plaque on lawns of 12 different species belong-ing to six different genera of the Halobacteriaceae (Haloar-cula Halobacterium Haloferax Halorubrum Haloterrigenaand Natrialba) but could plaque on Halorubrum strain CSW2094 (described in [8]) an uncharacterized Australianisolate (Table 1)

22 Virus Purification and Particle Morphology Virus waspurified by a slight modification of the method describedpreviously for halovirus SH1 ([8] and Section 4) PH1 bandedat a density of 129 gmL in CsCl gradients and gave a finalspecific infectivity of sim5 times 1011 PFUA

260 Negative-stain

electron microscopy revealed spherical particles with anaverage diameter of sim51 nm (Figure 1) The capsid displayedtwo layers with a compact core particle of sim43 nm indiameterThismorphology is similar to that of halovirus SH1a closely related virus (see below) that was isolated at the sametime as PH1 but from a neighbouring salt lake [8]

23 Single-Burst and Single-Step Growth of PH1 Around 30of Har hispanica cells could be infected by PH1 and singleburst experiments [32] indicated an average burst size of87 PFUcell (data not shown) This value was supported bysingle-step growth curves which gave burst sizes of between50 and 100 PFUcellThe number of infectious centres usuallyincreased at 4ndash6 hr pi but cell lysis did not appear to begin

Archaea 3

Table 1 Strains used in this study

Speciesisolate Strain ReferenceCSW 2094 Original isolate [51]Haloarcula hispanica ATCC 33960 [52]Haloarcula marismortui ATCC 43044 [53]ldquoHaloarcula sinaiiensisrdquo ATCC 33800 [54]Halobacterium salinarum NCIMB 763 [55]Haloferax gibbonsii ATCC 33959 [52]Haloferax lucentense NCIMB 13854 [56]Haloferax mediterranei ATCC 33500 [54]Haloferax volcanii ATCC 29605 [57]Halorubrum coriense ACAM 3911 [58]Halorubrum lacusprofundi ACAM 34 [59]Halorubrum saccharovorum NCIMB 2081 [60]Haloterrigena turkmenica NCIMB 784 [61]Natrialba asiatica JCM 9576 [62]

until well after this usually between 14 and 24 hr pi Arepresentative example of a single-step growth curve for PH1is shown in Figure 2 where the rise begins at sim6 hr pi andvisible cell lysis begins at sim14 hr pi This figure shows that asecond round of growth occurs at around 18 hr pi reflectingthe initial infection of only 30 of cells and it is during thissecond round of virus release that the cell density decreasesmost rapidly reaching a value of about 01 A

550 which is close

to the initial cell density In this example the average burstsize for the first growth step was 95 PFUcell

24 Virus Stability The stability of PH1 virus was testedunder several conditions (Figures 3(a)ndash3(d))When stored inHVD at 4∘C the infectivity of PH1 remained unaltered forseveral months (data not shown) A thermal stability curveis presented in Figure 3(a) and shows that PH1 is stable upto 56∘C above in which it rapidly loses titre Particles weresensitive to a reduced salt environment (Figure 3(b)) and tochloroform (Figure 3(d)) PH1 was most stable between pH8 and pH 9 (Figure 3(c)) In general the stability of PH1 wassimilar to that described previously for SH1 [8]

25 PH1 Structural Proteins and Protein Complexes Theproteins of purified PH1 virus were separated by SDS-polyacrylamide gel electrophoresis alongside the proteins ofSH1 virus (Figure 4) Nine PH1 protein bands were detectedwith molecular weights from 7 to 185 kDa (Figure 4(a)) andthese were designatedwith the prefixVP and a number corre-sponding to the homologous protein of SH1 [8 27] (see later)This nomenclature is consistent with that of the recentlydescribed HHIV-2 virus a member of the same virus group[13]The protein profile of PH1 was very similar to that of SH1[8] and with similar relative masses of the protein bands Onenotable difference was that PH1 proteins VP9 and VP10 ranclosely together (calculatedMWs are 165 and 167 kDa resp)compared to their SH1 homologs (165 and 169 kDa resp)It is probable given the strong sequence similarity to SH1(see later) that at least six additional PH1 structural proteinswere unable to be visualized using our staining techniques

0 5 10 15 200

01

02

Cell

den

sity

(A55

0)

Time (hr)

Viru

s titr

e (PF

Um

L)

1010

109

108

107

106

105

Figure 2 Growth curve of PH1 An early exponential culture ofHar hispanica in 18 (wv) MGM was infected with virus (MOI50) washed to remove unbound virus and incubated at 37∘C Atregular intervals samples were removed the absorbance at 550 nmwere measured (circles) and the number of infectious centres weredetermined by plaque assay (squares) Error bars represent onestandard deviation of the average titre

The potential glycosylation of virus proteins was examinedby staining similar protein gels either with a periodate-acid-Schiff (PAS) stain (GelCode Glycoprotein Staining Kit PierceBiotechnology USA) or the more sensitive fluorescent stain(Pro-Q Emerald 488 Glycoprotein Gel and Blot Stain KitMolecular Probes USA) No glycoproteins were detected

The major protein bands of PH1 (asterisked in Figure 4)were excised from gels digestedwith trypsin and analysed byMALDI-TOFMS and the results are summarized in Table 2From these data the structural proteins VP 1ndash4 7 9 10 and12 were found to be specified by ORFs 12 24 28 21 20 27 26and 19 respectively (see later)

26 Characteristics of the PH1 Genome Nucleic acid wasextracted from purified PH1 virus preparations treated withproteinase K and incubated with various nucleases to deter-mine the characteristics of the genome (Table 3) The PH1genome was sensitive to dsDNA endo- and exonucleasesbut not to ssDNA nuclease (mung bean) or RNase A Thisindicated that the PH1 genome is linear dsDNA with free(ie not covalently closed) termini Restriction endonucleasedigestions of the PH1 genome gave a length of approximately29 kb (data not shown)

The presence of terminal bound proteins was examinedusing a silica binding assay [33] Nonproteinase K-treatedPH1 DNA was restricted with AseI and the four resultingfragments passed through GFC filters under conditionswhere proteins bind firmly to the glass As shown in Figure 5the two internal AseI fragments (23 and 76 kb) passedthrough the filter but the right terminal fragment (174 kb)was bound indicating it carried an attached protein Theleft terminal fragment was too small (062 kb) and faintly

4 Archaea

0 20 40 60 80 100

Titre

(PFU

mL)

Temperature1011

1010

109

108

107

106

Temperature (∘C)

le105

(a)

0 5 10 15

0

Time (min)

Salt concentration

Titre

(PFU

mL)

107

106

105

104

minus5

(b)

4 5 6 7 8 9 10pH

pH

Titre

(PFU

mL)

108

107

(c)

Exposure time (min)0 5 10 15 20

Chloroform

Titre

(PFU

mL)

108

107

(d)

Figure 3 Stability of PH1 virus to various treatments and conditions Virus preparations (infected-cell supernatants in 18 (wv)MGM)wereexposed to various conditions after which the virus titre was determined (in duplicate) onHar hispanica cells (a) The effect of temperatureVirus was incubated for 1 hr with constant agitation at temperatures between 4 and 100∘C (b) The effect of lowered salt concentration Viruswas diluted 1 1000 in double-distilled H

2

O and incubated at room temperature with constant agitation Samples were removed at regularintervals (c)The effect of pHViruswas diluted 1 100 inTris-HCl buffers at the different pHs and incubatedwith constant agitation for 30min(d)The effect of chloroformChloroformwasmixedwith virus (1 4 ratio) and incubated at room temperaturewith constant agitation Sampleswere removed at regular intervals Open square symbols indicate where virus titres were undetectable Error bars represent one standarddeviation of the average titre

Archaea 5

Table 2 PH1 virus proteins identified by mass spectroscopy of tryptic peptides

Proteina Locus tag ORF MW (kDa) Matching peptide massesbObserved (calculated)

VP1 HhPH1 gp12 12 185 (158) 16VP2 HhPH1 gp24 24 100 (78) 15VP3 HhPH1 gp28 28 40 (38) 5VP4 HhPH1 gp21 21 35 (26) 5VP7 HhPH1 gp20 20 24 (20) 5VP9 HhPH1 gp27 27 16 (17) 3VP10 HhPH1 gp26 26 15 (17) 4VP12 HhPH1 gp19 19 7 (99) 4aVirus capsid proteins are numbered according to their similarity with SH1 capsid proteinsbThenumber of peptidemasses between 7500 and 35130mz identified byMALDI-TOFMS that correspond to the theoretical tryptic peptides of the predictedvirus protein

212 -

(kD

a)

PH1

158 -97 -66 -56 -42 -35 -

27 -

20 -

14 -

65 -

lowast

lowast

lowast

lowast

lowast

lowast

lowast

VP1VP2

VP3VP4

VP7

VP9

VP10

VP12

(a)

SH1

(b)

Figure 4 Structural proteins of halovirus PH1 Viral structuralproteinswere separated by SDS-PAGEon a 12 (wv) acrylamide geland stained with Brilliant Blue G (a) They were run in parallel withthe proteins of purified SH1 virus (b) The sizes of protein standardmarkers are indicated on the left side (in kDa) Asterisks denoteproteins bands of PH1 that were identified by mass spectroscopyThe numbering of proteins of PH1 (VP1ndashVP12) follows that of theSH1 homologs seen in (b) (and explained also in the text)

staining to be detected Further evidence was obtained by anuclease protection assay as previously used for SH1 DNA[31] As shown in Table 3 exonuclease III (a 31015840 exonuclease)was able to digest nonproteinase K-treated PH1 DNA butT7 exonuclease (a 51015840 exonuclease) and Bal31 (51015840 and 31015840exonucleases) were unable to digest PH1 DNA unless it hadbeen previously treated with proteinase KThis indicated thatboth 51015840 termini of the genome had protein attached

27 Sequence of the PH1 Genome The complete PH1 genomesequence was determined (accession KC252997) using a

1 2 3

-23

-76

-17410 -

6 -

3 -

2 -

15 -

05 -

4 -

(kB)

Figure 5 Detection of proteins bound to the termini of PH1 DNADNA was extracted from purified PH1 particles without proteinaseK digested with AseI then passed through a silica filter (GFC)under conditionswhere proteinswould bind to the filter Lane 1AseIfragments that did not bind and were eluted Lane 2 AseI fragmentsthat bound and were eluted only after protease treatment Lane 3AseI digest of PH1 DNA The positions of DNA size standards areindicated on the left side in kb The calculated sizes of the AseIfragments of PH1 DNA in lane 3 are indicated on the right side (alsoin kb) The predicted 062 kb PH1 AseI fragment was not detected

combination of cloned fragments PCR primer walkingon virus DNA and 454 whole genome sequencing (seeSection 4) It was found to be 28072 bp in length 676G + C with inverted terminal repeat sequences (ITRs) of337 bp Using GLIMMER [34] manual BLAST searching atthe GenBank database and comparison to related virusesthe PH1 genome sequence was predicted to contain 49 ORFs(Table 4) Most ORFs were closely spaced or overlappinggiving a gene density of 174 geneskb (average of 572 ntgene)The cumulative AT-skew plot of the PH1 genome shown atthe bottom of Figure 6 shows inflection points (circled) thatare consistent with the changes in transcription direction

6 Archaea

6 8 10 12 14 16 18 20 22 24 kb

1 2 4 6 8 10 12 14 162 kb

VP1

VP1-like fragments

A

A

VP2

VP2

VP7

VP1

2

20 21 22

VP4

VP1

3

VP5 p2

2

p9p4 p11

p12

p13

NP5

356A

Hmuk

p26

VP3

VP1

7

1728Hmuk

0477

Locus tag ZOD2009 19038

Locus tag HlacAJ 010100019872 NZ AGFZ01000123 (reversed)

NZ AEMG01000027 contig29

TransposaseHlacAJ 010100019767

ZOD2009 19153

Hbf lacisalsi

Hap paucihalophilus

(a)

VP18VP2VP1 A

VP1

VP18VP1

VP1

2

VP1

VP7

VP4

VP1

3

VP5

VP2

VP9

VP1

0

VP3

VP1

6

VP1

7V

P6 VP18

ATPa

se

kb1 2 4 6 8 10 12 14 16 18 20 22 24 26 28

VP2

VP2

VP18A

AA

2211 1312

94 26

HHIV-2

SH1

PH1

0

2

4

6

30 578 nt

30 889 nt

28 072 nt

(times102

)Cu

m A+

T sk

ew

Cumulative A + T skew of PH1

(b)

Figure 6 Genome alignments of haloviruses PH1 SH1 and HHIV-2 along with two related genomic loci (a) Genomic loci of Happaucihalophilus andHbf lacisalsi that contain genes related to halosphaeroviruses SH1 PH1 andHHIV-2The names andGenBank accessionsfor these contigs are given on the far right andORFs are coloured and labeled to indicate the relationships of theseORFs to those of the virusesbelowThe locus tag numbers for the first and last ORFs shown in each locus are given nearby their respectiveORFs In addition grey colouredORFs represent sequences that do not match any of the haloviruses and green coloured ORFs represent protein sequences that are closelyrelated to ORFs found within or very close to previously described virusplasmid loci so called ViPREs [20] The scale bars shown aboveeach contig show the position of the described region within the respective contig (b) The three virus genomes are labeled at the left withscale markers below (in kb) and the total length indicated at the far right At the bottom is a cumulative AT-skew plot of the PH1 genome(httpmolbiol-toolscaJie Zheng) with inflection points circledThe grey shaded bands between the genome diagrams indicate significantnucleotide similarity (usingACT [63]) AnnotatedORFs are represented by arrows with colours indicating structural proteins (red or brown)nonstructural proteins (yellow or orange) or the packaging ATPase (blue) The names of structural protein ORFs are indicated either withinthe arrow (eg VP1) or in text nearby The numbered orange coloured ORFs of HHIV-2 are homologous to ORFs found in the genomic loci(probably proviruses or provirus remnants) pictured in (a)

indicated by the annotated ORFs as has been seen in otherhaloviruses [7] There are no GATC motifs in the genomeand the inverse sequence CTAG is strongly underrepre-sented with just 2 motifs (68 expected)

When compared to the related viruses SH1 and HHIV-2(Figure 6) the PH1 genome is seen to be of similar length andis much more closely related to SH1 than to HHIV-2 (74and 54 nt identity resp) Like PH1 the genomes of SH1and HHIV-2 lack GATC motifs and CTAG is either absent(SH1) or underrepresented (HHIV-2) The ITRs of PH1 andSH1 share 785nucleotide identity but the PH1 ITR is longerthan that of SH1 (337 and 309 nt resp) partly because SH1has replaced an 18 bp sequence (gtcgtgcggtttcggcgg) found atthe internal end of its left-hand ITR with a sequence at thecorresponding position of its right-hand ITR that shows littleinverted sequence similarity In the PH1 ITR this sequenceis retained at both ends Whether this difference represents arecombination event or a mistake in replication of the ITRs

is unclear An alignment of the ITRs of all three virusesidentified a number of highly conserved regions (labeled 1ndash9 in Supplementary Figure 2 see Supplementary Materialavailable online at httpdxdoiorg1011552013456318) A16 bp sequence at the termini (region 1) is conserved con-sistent with the conservation seen at the termini of linearStreptomyces plasmids [35] A 16 bpGC-rich sequence aroundthe middle of the ITR (region 3) is also conserved Shortermotifs of either C-rich (region 8) or AT-rich sequence(regions 7 and 9) are found near the internal end of the ITRs

The grey shading between the schematic virus genomesin Figure 6(b) indicates regions of high nucleotide similarityrevealing that the differences between PH1 and SH1 are notevenly distributed but vary considerably along the lengthof the aligned genomes A comparison between SH1 andHHIV-2 has been published recently [13] and since the PH1and SH1 genomes are so similar we will focus the followingdescription on the differences between PH1 and SH1

Archaea 7

Table 3 Digestion of the PH1 genome by nucleasesa

Nuclease (amount) Proteinase K-treated UntreatedControl PH1 DNA Control PH1 DNA

DNase I (RNase-free)(100U) +b + +b +

Exonuclease III(100U) +b + +b +

Mung bean nuclease(10U) +c

minus +cminus

Nuclease BAL-31(1 U) +b + +b

minus

RNase A(5 120583gmL) +d

minus +dminus

T7 exonuclease(10U) +b + +b

minus

aExtracted nucleic acid from purified PH1 virus was treated with variousnucleases and then analysed by gel electrophoresis to detect whether thenuclease digested (+) or did not digest (minus) the genome Control nucleicacids used to confirm activity of the nucleases were b

120582 DNA ca DNAoligonucleotide and dyeast transfer RNA

Within corresponding ORFs there are blocks of codingsequence with low (or no) similarity These may represent insitu divergence or short recombination events (eg indels)There are also cases where an entire ORF in one virus hasno corresponding homolog in the other because of an inser-tiondeletion event (indel) Lastly there are replacementswhere the sequences within the corresponding regions showlow or no similarity but are flanked by sequence with highsimilarity The largest visible differences seen in Figure 6 aredue to sequence changes within or near the long genes encod-ing capsid proteins VP1 and VP18 A summary of the mainsequence differences between SH1 and PH1 and their loca-tions is given in Table 5 where these divergent regions (DV)are numbered from DV1 to DV18

The sequences of the two viruses are close enough that inmany cases the mechanism for the observed differences canbe inferred For example DV2 is likely the result of a deletionevent that has removed the SH1 ORF10 gene homolog fromthe PH1 genome At either end of the SH1 ORF are almostperfect direct repeats (cggcctgaccggcatgac) that would allowrepeat-mediated deletion to occur removing the interveningsequence Small direct repeats leading to deletions have beendescribed previously in the comparative analysis of Halo-quadratumwalsbyi genomes [20] DV5 appears to be an indelwhere SH1 ORFs 14ndash16 are absent in PH1 and an invertedrepeat (AGCCATG) found at each end of the SH1 divergentregion may be significant in the history of this changeThe proteins specified by SH1 ORFs 14ndash16 are presumablydispensable for PH1 (or their functions supplied by otherproteins) but a homolog of SH1 ORF14 is found also inHHIV-2 (ORF 6) and a homolog of SH1 ORF15 occurs inhalovirus His1 (ORF13) Replacement regions are also com-mon and can occur within ORFs (eg DV3 in capsid proteinVP1) or provide additional or alternative genes (eg DV12and DV13) Several divergent regions occur within the genefor capsid protein VP2 a hot spot for change because of the

repetitive nature of the coding sequence which specifies aprotein with runs of glycines andmany heptapeptide repeatsas described previously for SH1 VP2 [27] DV18 is a largereplacement that significantly alters the predicted length ofthe minor capsid protein VP18 in the two viruses (865519aa for SH1PH1 resp) and also provides SH1 with 3 ORFsthat are different or absent in PH1 ORFs 52 53 and 54 Forexample a homolog of SH1 ORF52 is not present in PH1 butis present in HHIV-2 (putative protein 40) While ORF48 ofPH1 shows no aa sequence homology to SH1 ORFs 53 and 54all of these predicted proteins carry CxxC motifs suggestiveof related functions such as DNA binding activity [36]

The genes of the PH1 genome are syntenic with those ofSH1 and are found in similarly oriented blocks of closelyspaced or overlapping ORFs that suggest that transcription isorganised into operons The programme of transcription ofthe SH1 genome has been reported previously [30] and thesequences in PH1 corresponding to the six promoter regions(P1ndashP6) determined in SH1 showed that five of these (P1ndashP5)are well conserved Only the region corresponding to SH1 P6was poorly conserved (data not shown) but this promoter isstrongly regulated in SH1 and only switches on late in infec-tion [30]

28 Structural Protein Genes The genes coding for the majorvirus structural proteins of PH1 (VP 1ndash4 7 9 10 and 12)were identified byMALDI-TOF (see above) and the genes forthese proteins are found in the same order and approximatepositions as in the SH1 genome (Figure 6(b) red ORFs)Genes coding for minor structural proteins can be deducedby sequence comparison with SH1 so that VP5 and VP6 arelikely to be encoded by PH1ORFs 25 and 29 respectively andVP13 and VP18 by ORFs 23 and 49 respectively This wouldgive PH1 a total of 11 structural proteins the same as SH1Of these VP7 is the most conserved between the two viru-ses (98 aa identity) followed by VP4 VP9 VP3 and VP12(91ndash94 identity) VP5 VP6 and VP13 (82ndash88 iden-tity) VP2 (77) VP1 (65) and lastly the least conservedprotein was VP18 (31)

29 Genomic Loci Related to PH1 and Other Halosphaeroviru-ses Clusters of halosphaerovirus-related genes are present inthe genomes of two recently sequenced haloarchaea Happaucihalophilus and Hbf lacisalsi (Figure 6(a)) Neither ofthese regions appears to represent complete virus genomebut they retain a number of genes that share a similar seq-uence and synteny with the virus genomes depicted belowthem (Figure 6(a)) Both loci show a mixed pattern ofrelatedness to SH1PH1 and HHIV-2 as indicated by ORFsof matching colour and gene name If these genomic locirepresent provirus integrants that have decayed over timethen the relationships they show suggest not only that halo-sphaeroviruses are a diverse virus group but also that asignificant level of recombination occurs between them lead-ing to mosaic gene combinations The two ORFs colouredgreen in theHap paucihalophilus locus (Figure 6) are relatedto ORFs found within or very close to genomic loci ofvirusplasmid genes found in other haloarchaea suggestingadditional links between (pro)viruses [20]

8 Archaea

Table4ORF

anno

tatio

nsof

theh

aloviru

sPH1g

enom

e

ORF

aPo

sitionb

Locustag

Leng

thc(aa)

pKi

Similaritycharacteris

tics

144

1ndash605

HhP

H1gp01

5043

76aa

similarityto

SH1O

RF2(YP271859)

2602ndash802

HhP

H1gp02

66109

80aa

similarityto

SH1O

RF3(YP271860)

3802ndash1059

HhP

H1gp03

8539

85aa

similarityto

SH1O

RF4(YP271861)44

sim

ilarityto

HHIV-2

protein1

41056ndash1283

HhP

H1gp04

7539

85aa

identityto

SH1O

RF5(YP271862)Predictedcoiled-coilregion

51276ndash1686

HhP

H1gp05

136

49

80aa

similarityto

SH1O

RF6(YP271863)Tw

otransm

embraned

omainsA

lsorelated

toHHIV-2

protein2(38

AFD

02283)

andto

Halobiform

alacisalsiORF

ZP09950018

(35

)6

1683ndash1829

HhP

H1gp06

4861

81aa

similarityto

SH1O

RF7(YP271864

)Predictedsig

nalsequence(Sign

alP)

71822ndash2130

HhP

H1gp07

102

58

91aa

similarityto

SH1O

RF8(YP271865)a

nd86to

HHIV-2

putativ

eprotein

3(A

FD02284)O

ther

homologsinclude

Nocardiaproteinpn

f2110(YP1220601)H

rrlacusprofund

iHlac0751

(YP0025654211)ORF

sinactin

ophage

VWB

(AAR2

9707)Frankia(EAN116

57)andgp40

ofMycobacteriu

mph

ageD

ori(AER

47690)

82123ndash2383

HhP

H1gp08

8642

90aa

similarityto

SH1O

RF9(YP271866)

92380ndash2714

HhP

H1gp09

111

52

88aa

similarityto

SH1O

RF11(YP271868)and

also

similarityto

putativ

eprotein

4of

HHIV-2

(AFD

02285)

HlacA

J19877of

Hbflacisa

lsiAJ

5(ZP09950016)

102861ndash300

4HhP

H1gp10

47104

Weaksim

ilarityto

SH1O

RF12

(YP271869)o

nlyover

theN

-term

inal18

resid

ues

113029ndash3100

HhP

H1gp11

23123

Not

presentinSH

1orH

HIV-2

123134ndash7453

HhP

H1gp12

1439

42

Capsid

protein

VP172sim

ilarityto

SH1V

P1(O

RF13Y

P271870)HHIV-2

VP1

(AFD

02286)H

lacA

J19872Hbflacisa

lsiAJ

5Predictedhelix-tu

rn-helixandRu

vA-like

domain(In

terproScan)

137505ndash8227

HhP

H1gp13

240

54

Predicted

P-loop

ATPa

sedo

main(C

OG0433)92aa

similarityto

SH1O

RF17

(YP271874)a

ndalso

similarityto

putativ

eAT

Pase

ofHHIV-2

(AFD

02288)A

TPaseo

fHaladapatus

paucihalophilusD

X253

(ZP0804

6180)HlacA

J19867of

Hbf

lacisalsi(ZP09950014)

148419ndash8228c

HhP

H1gp14

6351

43aa

similarityto

SH1O

RF18

(YP271875)Alanine-richCentraltransm

embraned

omain(Pho

bius)

158856ndash8

416c

HhP

H1gp15

146

41

92aa

similarityto

SH1O

RF19

(YP271876)a

ndalso

toHHIV-2

putativ

eprotein

9(A

FD02290)H

lacA

J19857of

Hbf

lacisalsiAJ

5ZO

D2009

19093of

Happau

cihalophilus

168853ndash9

500c

HhP

H1gp16

215

39

65aa

similarityto

SH1O

RF20

(YP271877)a

ndalso

topu

tativ

eprotein

11of

HHIV-2

(AFD

02292)Z

OD2009

19098of

Hap

paucihalophilusHlacA

J19852Hbflacisa

lsiAJ

5C-

term

inaltransm

embraned

omain(Pho

bius)

179500ndash9

919c

HhP

H1gp17

139

41

79aa

similarityto

SH1O

RF21

(YP271878)a

ndalso

topu

tativ

eprotein

12of

HHIV-2

(AFD

02293)hypotheticalprotein

HlacA

J19847of

Hbflacisa

lsi(ZP09950010)Transm

embraned

omain(Pho

bius)

189923ndash10594c

HhP

H1gp18

223

43

83aa

similarityto

SH1O

RF22

(YP271879)a

ndalso

topu

tativ

eprotein

13of

HHIV-2

(AFD

02294)Z

OD2009

19108of

Hap

paucihalophilusHlacA

J19842of

Hbflacisa

lsiP

redicted

signalsequence(sig

nalP)Con

tains4

CxxC

motifs

1910659ndash10943

HhP

H1gp19

94104

Capsid

protein

VP12

(ORF

19)91aa

similarityto

SH1V

P12(O

RF23Y

P271880)a

ndalso

toVP12of

HHIV-2

(AFD

02295)

andHlacA

J19837of

Hbflacisa

lsiZ

OD2009

19113of

Happau

cihalophilusTw

otransm

embraned

omains

(Pho

bius)

2010960ndash

11517

HhP

H1gp20

185

44

Capsid

protein

VP7(O

RF20)98aa

similarityto

SH1V

P7(O

RF24Y

P271881)a

ndalso

toVP7

ofHHIV-2

(AFD

02296)

ZOD2009

19118

ofHappau

cihalophilusHlacA

J19832of

Hbflacisa

lsiAJ

5

2111519ndash

12217

HhP

H1gp21

232

41

Capsid

protein

VP4(O

RF21)94aa

similarityto

SH1O

RF25

(YP271882)a

ndalso

toVP4

ofHHIV-2

(AFD

02297)

ZOD2009

19123of

Happau

cihalophilusHlacA

J19827of

Hbflacisa

lsiAJ

522

12233ndash12454

HhP

H1gp22

7339

81aa

similarityto

SH1O

RF26

(YP271883)a

ndalso

topu

tativ

eprotein

17of

HHIV-2

(AFD

02298)

2312458ndash12697

HhP

H1gp23

7948

78aa

similarityto

SH1capsid

protein

VP13

(ORF

27Y

P271884)a

ndalso

toVP13of

HHIV-2

(AFD

02299)P

redicted

coil-coildo

main

PredictedC-

term

inaltransm

embraned

omain(Pho

bius)

Archaea 9

Table4Con

tinued

ORF

aPo

sitionb

Locustag

Leng

thc(aa)

pKi

Similaritycharacteris

tics

2412701ndash1506

4HhP

H1gp24

787

39

Capsid

protein

VP2(O

RF24)75aa

similarityto

SH1V

P2(O

RF28Y

P271885)a

ndto

VP2

ofHHIV-2

(AFD

02300)

ZOD2009

19133of

Happau

cihalophilus(ZP

0804

6189)

2515065ndash15961

HhP

H1gp25

298

45

Putativec

apsid

protein

VP5(O

RF25)81aa

similarityto

SH1V

P5(O

RF29Y

P271886)a

ndto

HHIV-2

VP5

(AFD

02301)

HlacA

J19807of

HbflacisalsiAJ

5ZO

D2009

19133of

Happau

cihalophilus

261596

4ndash1644

6HhP

H1gp26

160

46

73aa

similarityto

SH1capsid

protein

VP10

(ORF

30Y

P271887)a

ndto

VP10of

HHIV-2

(AFD

02302)Z

OD2009

19138of

Happau

cihalophilusHlacA

J19802of

Hbflacisa

lsiP

redicted

C-term

inaltransm

embraned

omain(TMHMM)

271644

6ndash16895

HhP

H1gp27

149

42

Capsid

protein

VP9(O

RF27)94aa

similarityto

SH1V

P9(O

RF31Y

P271888)a

ndto

ZOD2009

19143of

Hap

paucihalophilus(ZP

0804

6191)

2816908ndash17921

HhP

H1gp28

337

43

Capsid

protein

VP3(O

RF28)91aa

similarityto

SH1V

P3(O

RF32Y

P271889)a

ndto

NJ7G

2365

ofNa

trinemaspJ7-2

2917928ndash18617

HhP

H1gp29

229

42

Capsidprotein

VP6(O

RF29)83aa

similarityto

SH1V

P6(O

RF33Y

P271890)Also

toNJ7G

3194

ofNa

trinemaspJ7-2and

Hmuk

0476

ofHalom

icrobium

mukohataeiPredictedcarboxypeptid

aser

egulatory-lik

edom

ain(C

arbo

xypepD

reg

pfam

13620)

3018614ndash

18943

HhP

H1gp30

109

49

71aa

similarityto

SH1O

RF34

(YP271891)a

ndto

putativ

eprotein

27of

HHIV-2

(AFD

02308)P

redicted

signalsequence

(signalP)

3119054ndash

19176c

HhP

H1gp31

4050

TwoCx

xCmotifsN

oho

molog

inSH

1orH

HIV-2

3219173ndash19289c

HhP

H1gp32

3841

68aa

similarityto

SH1O

RF37

(YP271894)

3319286ndash

19552c

HhP

H1gp33

8853

58aa

similarityto

SH1O

RF39

(YP271896)a

ndto

HHIV-2

putativ

eprotein

29(A

FD02310)Fou

rCxxCmotifs

3419549ndash

19731c

HhP

H1gp34

6070

Con

tainsC

xxCmotif

Noho

molog

inSH

1orH

HIV-2

3519728ndash20219c

HhP

H1gp35

163

56

93aa

similarityto

SH1O

RF41

(YP271898)a

ndto

putativ

eprotein

30of

HHIV-2

(AFD

02311)

3620216ndash

21415c

HhP

H1gp36

399

50

95aa

similarityto

SH1O

RF42

(YP271899)a

ndto

putativ

eprotein

31of

HHIV-2

(AFD

02312)

3721419ndash

21778c

HhP

H1gp37

11941

91aa

similarityto

SH1O

RF43

(YP271900)a

ndto

putativ

eprotein

32of

HHIV-2

(AFD

02313)

3821762ndash2300

6cHhP

H1gp38

414

42

83aa

similarityto

SH1O

RF44

(YP271901)a

ndto

putativ

eprotein

33of

HHIV-2

(AFD

02314)P

redicted

coil-coiland

helix-tu

rn-helixdo

mains

3923003ndash23158c

HhP

H1gp39

5165

69aa

similarityto

SH1O

RF45

(YP271902)

4023161ndash23370c

HhP

H1gp40

6935

74aa

similarityto

SH1O

RF46

(YP271903)and55sim

ilarityto

putativ

eprotein

35of

HHIV-2

(AFD

02316)

4123422ndash23586c

HhP

H1gp41

5476

86aa

similarityto

SH1O

RF47

(YP271904)Con

tains2

CxxC

motifsand

show

ssim

ilarityto

proteindo

mainfamily

PF14206Arginine-ric

h

4223583ndash24023c

HhP

H1gp42

48

77aa

similarityto

SH1O

RF48

(YP271905)a

ndto

putativ

eprotein

36of

HHIV-2

(AFD

02317)H

ham114540of

Hcc

hamelinensis(ZP11271999)Ph

iCh1p72of

Natrialba

phageP

hiCh

1(NP665989)DUF4

326(pfam14216)

family

domain

4324020ndash

24538c

HhP

H1gp43

172

47

79aa

similarityto

SH1O

RF49

(YP271906)and56sim

ilarityto

putativ

eprotein

37of

HHIV-2

(AFD

02318)

4424535ndash25053c

HhP

H1gp44

172

43

92aa

similarityto

SH1O

RF50

(YP271907)and72sim

ilarityto

putativ

eprotein

38of

HHIV-2

(AFD

02319)

4525050ndash

25277c

HhP

H1gp45

7543

Noho

molog

inSH

1orH

HIV-2

4625274ndash

25387c

HhP

H1gp46

3748

Noho

molog

inSH

1orH

HIV-2

4725533ndash25904

HhP

H1gp47

123

43

61aa

similarityto

SH1O

RF51

(YP271908)66sim

ilarityto

putativ

eprotein

39of

HHIV-2

(AFD

02320)P

redicted

COG1342

domain(D

NAbind

inghelix-tu

rn-helix)

4825901ndash26092c

HhP

H1gp48

6349

Con

tainsC

xxCmotif

Noho

molog

inSH

1orH

HIV-2

4926089ndash

2764

8cHhP

H1gp49

506

40

81aa

similarityto

SH1O

RF55

(YP271912)(bu

tonlyin

theN

-term

inalhalf)H

omolog

ofvirusstru

cturalprotein

VP18

ofHHIV-2

(AFD

02323)

a ORF

swerep

redicted

either

byGLIMMER

orby

manualsearching

forh

omologsintheG

enBa

nkdatabase

b Startandendpo

sitions

ofORF

sare

give

inbp

numbera

ccording

totheP

H1sequenced

epositedatGenBa

nk(KC2

52997)O

RFso

nthec

omplem

entary

strandared

enoted

bythes

uffixc

c Lengthof

thep

redicted

ORF

innu

mbero

faminoacids

10 Archaea

Table 5 Main differences (divergent regions) between the SH1 and PH1 genomes

RegionaSH1 startb SH1 stop Length (bp) PH1 start PH1 stop Length (bp) Comment

DV1 527 650 123 538 604 66

Replacement Both regions have direct repeats at their bordersPH1 has two sets GACCCGGC and CGCTGC while SH1 hasCCCGAC In SH1 this replacement region includes theC-terminal region of ORF2 and the N-terminal region ofORF3 In PH1 it covers only the C-terminal region of ORF1

DV2 2433 2588 155 2386 2387 mdash RMDc from PH1 SH1 repeat at border is TGACCGThisremoves the homolog of SH1 ORF10 from PH1

DV3 3324 3598 274 3137 3416 279 Replacement at the beginning of SH1 ORF13PH1 ORF12(capsid protein VP1)

DV4 6656 7128 472 6478 7085 607 Replacement near the C-terminus of SH1 ORF13PH1 ORF12(capsid protein VP1)

DV5 7491 8341 850 7446 7447 mdash

Indel that results in ORFs14-16 of SH1 not being present inPH1 SH1 ORF14 has a close homolog in HHIV-2 (ORF6) in asimilar position just after the VP1 homolog SH1 ORF15 is aconserved protein in haloarchaea (eg NP 2552A andHmuk 2978) and halovirus His1 (ORF13) Possible invertedrepeat (AGCCATG) at border of SH1 region

DV6 13705 13706 mdash 12818 12851 33RMD from SH1 PH1 repeat at border is CAGCGG(gt)G Thisremoves a part of capsid protein VP2 sequence from the SH1protein

DV7 13789 13863 74 12934 12941 7 Replacement This occurs within VP2 gene of both viruses

DV8 15142 15186 44 14220 14221 mdash RMD from PH1 SH1 repeat at border is TG(tc)CCGACGAand occurs within capsid protein VP2 gene of both viruses

DV9 15303 15317 14 14337 14338 mdash RMD from PH1 SH1 repeat at border is GCCGACGA andoccurs within capsid protein VP2 gene of both viruses

DV10 15504 15527 23 14529 14530 mdash RMD from PH1 SH1 repeat at border is GACGA and occurswithin capsid protein VP2 of both viruses

DV11 20026 20405 379 19019 19173 154Replacement beginning at the start of SH1 ORF35PH1 ORF31This region is longer in SH1 where it includes ORF36 an ORFthat has no homolog present in PH1

DV12 20440 20661 221 19208 19286 78 Replacement This is unequal and includes an ORF in SH1(ORF38) that is not present in PH1

DV13 20943 21085 142 19562 19728 166 Replacement This covers SH1 ORF40PH1 ORF34 Thepredicted proteins are not homologous

DV14 23541 23542 mdash 22171 22197 27 Probable RMD in SH1 PH1 repeat at border is CGTCTCGGand occurs in SH1 ORF44PH1 ORF38

DV15 24506 24521 15 23152 23153 mdash Probable RMD in PH1 SH1 repeat at border is CTCGGT andoccurs near the end of SH1 ORF45PH1 ORF39

DV16 24951 24964 13 23579 23580 mdash Probable RMD in PH1 SH1 repeat at border is CGGTC andoccurs in SH1 ORF47PH1 ORF41

DV17 26449 26450 mdash 25050 25274 224Probable RMD in SH1 PH1 repeat at border is TCATGCG andoccurs near the start of SH1 ORF50PH1 ORF44 and providesan extra ORF for PH1 (ORF45)

DV18 27074 29764 690 25895 26933 1038

Replacement Left border at start of SH1 ORF51PH1 ORF47and extends rightwards into SH1 ORF55PH1 ORF49 (VP18gene) It is an unequal replacement and SH1 has two moreORFs in this region than PH1

aDV Divergent regions between the genomes of SH1 and PH1bStart and stop positions refer to the GenBank sequences of the two viruses SH1 NC 0072171 PH1 KC252997cRMD repeat-mediated deletion event as described in [20]

The recently described halovirus SNJ1 carries manyORFsthat have homologs adjacent to a previously described ViPREofHmc mukohataei (from here on denoted by ViPREHmuk1)a locus that contains genes related to His2 (and other ple-olipoviruses) as well as toHaloferax plasmid pHK2 (probablya provirus) and to two small plasmids ofHqr walsbyi (sim6 kbpL6A and pL6B)The latter relationship provides wider linksto other viruses because one of the PL6 genes (Hqrw 6002)

has homologs in some pleolipoviruses (HPRV-3 and HGPV-1) while another gene (Hqrw 6005) is related to ORF16 ofthe spindle-shaped halovirus His1 [15] After adding the SNJ1gene homologs to ViPREHmuk1 it is extended significantlyand a close inspection of the flanking regions revealed atRNA-ala gene at one end and a partial copy of the sametRNA-ala gene (next to a phage integrase gene) at the otherend (supplementary Figure 1) Genes beyond these points

Archaea 11

Table 6 Transfection of haloarchaea by PH1 DNA

Speciesa Efficiency of transfectiontransformation withPH1 DNAb (PFU120583g of DNA) pUBP2 DNAbc (CFU120583g of DNA)

Har hispanica 53 plusmn 05 times 103 14 plusmn 08 times 104

Har marismortui 45 plusmn 07 times 103 18 plusmn 01 times 104

ldquoHar sinaiiensisrdquo 32 plusmn 06 times 102 mdashHalobacterium salinarum mdash 54 plusmn 42 times 103

Haloferax gibbonsii 28 plusmn 08 times 103 40 plusmn 26 times 103

Haloferax lucentense mdash 57 plusmn 11 times 103

Hfx volcanii mdash 21 plusmn 07 times 105

Hrr lacusprofundi 70 plusmn 22 times 102 95 plusmn 15 times 103

Haloterrigena turkmenica 16 plusmn 17 times 102 24 plusmn 06 times 103

Natrialba asiatica 35 plusmn 39 times 102 31 plusmn 10 times 104aOnly those species positive for transfection andor plasmid transformation are shownbRates of transfection by (nonprotease treated) PH1 DNA or transformation by plasmid pUB2 are averages of three independent experiments each performedin duplicate (plusmn standard deviation)cTransformants were selected on plates with 2 4 or 6120583gmL simvastatin depending on the strainmdash no plaques or colonies observed

appear to be related to cellular metabolism ViPREHmuk1is now 39379 nt in length flanked by a 59 nt direct repeat(potential att sites) and includes 54 ORFs many of which arerelated to known haloviruses are virus-like (eg integrasesmethyltransferases transcriptional regulators DNA methyl-transferase and DNA glycosylase) are homologs in or nearother knownViPREs or show little similarity to other knownproteins This locus also includes an ORC1CDC6 homolog(Hmuk 0446) which could provide a replication function

The mixture of very different virus and plasmid genesseen within ViPREHmuk1 is remarkable but it also revealshomologs found in or adjacent to genomic loci of otherspecies such as the gene homologs of Har marismortuiseen in the left end of ViPREHmuk1 (supplementary Figure1) When these genes are added to the previously describedHaloarcula locus (ViPREHmuk1) it is also extended signifi-cantly It becomes 25550 bp in length encompasses genesfrom Hmar 2382 to Hmar 2404 and is flanked by a fulltRNA-ala gene at one end and a partial copy at the other Itcarries an ORC1CDC6 homolog an integrase and a phiH-like repressor as well as pleolipovirus homologs

210 Transfection of Haloarchaea by PH1 DNA PH1 DNAwas introduced into Har hispanica cells using the PEGmethod [37] and the cells were screened for virus productionby plaque assay Figure 7 shows the increase in transfectedcells with input (nonproteinase K-treated) virus DNA Theestimated efficiencywas 53times 103 PFUper120583gDNA PH1DNAthat had been treated with proteinase K or with DNAase I(RNase-free) did not produce plaques (data not shown) PH1DNA was found to transfect six species of haloarchaea otherthanHar hispanica includingmembers of the generaHaloar-cula Haloferax Halorubrum Haloterrigena and Natrialba(Table 6) For these experiments the virus production fromtransfected cells was detected by plaque assay on indicatorlawns of Har hispanica

0

0

200 400 600 800 1000

Titre

(PFU

mL)

104

103

102

PH1 DNA (ng)

Figure 7 Transfection ofHar hispanica cells with PH1 DNA Vary-ing amounts of nonproteinase K-treated PH1 DNAwere introducedinto cells of Har hispanica using the PEG method [64] Cells werethen screened for infective centres by plaque assay Data shown is theaverage of three independent experiments performed in duplicateError bars represent one standard deviation of themean If protease-treated DNA was used no transfectant plaques were observed

3 Discussion

PH1 is very similar in particlemorphology genome structureand sequence to the previously described halovirus SH1 [816] Like SH1 it has long inverted terminal repeat sequencesand terminal proteins indicative of protein-primed replica-tion Purified PH1 particles have a relatively low buoyant

12 Archaea

density and are chloroform sensitive and show a layeredcapsid structure consistent with the likely presence of aninternalmembrane layer as has been shown for SH1HHIV-2and SNJ1 [13 26]The particle stability of PH1 to temperaturepH and reduced salt was similar to SH1 [8] The structuralproteins of PH1 are very similar in sequence and relativeabundance to those of SH1 so the two viruses are likely toshare the same particle geometry [16]

Viruses that use protein-primed replication such as bac-teriophageΦ29 carry a viral type BDNApolymerase that caninteract specifically with the proteins attached to the genomictermini and initiate strand synthesis [38] ArchaeovirusesHis1 [7] and Acidianus bottle-shaped virus [39] probablyuse this mode of replication However a polymerase genecannot be found in the genomes of PH1 SH1 or HHIV-2DNA polymerases are large enzymes and the only ORF ofSH1 that was long enough to encode such an enzyme andhad not previously been assigned as a structural protein wasORF55 which specified an 865 aa protein that contained noconserved domains indicative of polymerases [27]The recentstudy of HHIV-2 showed that the corresponding ORF in thisvirus (gene 42) is a structural protein of the virion and byinference this is likely also for the corresponding proteinsof SH1 and PH1 (no homolog is present in SNJ1) Withouta polymerase gene these viruses must use a host enzymebut searches for a viral type B DNA polymerase in thegenome of the hostHar hispanica or in the genomes of othersequenced haloarchaea did not find anymatchesThis arguesstrongly for a replication mechanism that is different to thatexemplified byΦ29 An attractive alternative is that displayedby Streptomyces linear plasmids which have 51015840-terminal pro-teins and use a cellular polymerase for replication [40] Theterminal proteins are not used for primary replication but forend patching [41] and it has been shown that if the term-inal repeat sequences are removed from these plasmids andthe ends ligated they can replicate as circular plasmids [42]This provides a testable hypothesis for the replication of halo-sphaeroviruses and also offers a pathway for switchingbetween linear (eg PH1) and circular (eg SNJ1) forms Useof a host polymerase would also fit with the ability of SH1 andPH1 DNA to transfect many different haloarchaeal species[31] as these enzymes and theirmode of action are highly con-served

The growth characteristics of PH1 inHar hispanica bothby single-cell burst and single-step growth experiments gavevalues of 50ndash100 virusescell significantly less than that ofSH1 (200 virusescell) [8] and HHIV-2 (180 virusescell) [13]The latter virus is lytic whereas SH1 PH1 and SNJ1 start pro-ducing extracellular viruswell before any decrease in cell den-sity indicating that virus release can occur without lysisThishas been most clearly shown with SH1 where almost 100of cells can be infected [8] Infection ofHar hispanica by PH1was less efficient (sim30 of cells) but the kinetics of virus pro-duction in single-step growth curves of SH1 and PH1 aresimilar with virus production occurring over several hoursrather than at a clearly defined time after infection The twoviruses are also very closely related and infect the same hostspecies so it is likely that they use the same scheme for cellexit For both viruses the cell density eventually decreases in

single-step cultures showing that virus infection does resultin cell death This mode of exit appears to maximise theproduction of the virus over time as the host cells survive foran extended period In natural hypersaline waters cell num-bers are usually high but growth rates are low [12] and thehigh incident UV (on often shallow ponds) would damagevirus DNA so reducing the half-life of released virus part-icles [43] These factors may have favoured the exit strategydisplayed by these viruses improving their chance of trans-mission to a new host

The motif GATC was absent in the genomes of PH1 SH1andHHIV-2 and themotif CTAGwas either absent or greatlyunderrepresented All three viruses share the same host butGATC and CTAG motifs are plentiful in the Har hispanicagenome (accessions CP002921 and CP002923) Avoidance ofthesemotifs in halovirusesHis1 His2 HF1 andHF2 has beenreported previously [7] Such purifying selection commonlyresults from host restriction enzymes and inHfx volcanii itis exactly these two motifs that are targeted [44] One Hfxvolcanii enzyme recognises A-methylated GATC sites [45]and another recognises un-methylated CTAG sites (whichare protected by methylation in the genome) In the currentstudy this could explain the negative transfection results forHfx volcanii as the PH1 genome contains two CTAGmotifsBy comparison the SH1 genome contains no CTAG motifsand is able to transfect Hfx volcanii [31]

Comparison of the PH1 and SH1 genomes allowed adetailed picture of the natural variation occurring betweenclosely related viruses revealing likely deletion events medi-ated by small repeats a process described previously in astudy of Hqr walsbyi and termed repeat-mediated deletion[20] There are also many replacements including blocks ofsequence that occur within long open reading frames (suchas the capsid protein genes VP1 and VP18) VP3 and VP6 areknown to form the large spikes on the external surface of theSH1 virion and presumably one or both interact with host cellreceptors [16] PH1 and SH1 have identical host ranges con-sistent with the high sequence similarity shown between theircorresponding VP3 and VP6 proteins

A previous study of SH1 could not detect transcriptsacross annotated ORFs 1ndash3 or 55-56 [30] ORFs 1 and 56occur in the terminal inverted repeat sequence and in thepresent study it was found that there were no ORFs cor-responding to these in the PH1 genome Given the close simi-larity of the two genomes the comparative data are in agree-ment with the transcriptional data indicating that these SH1ORFs are not used SH1 ORFs 2 and 3 are more problematicas good homologs of these are also present in PH1 The con-flict between the transcriptional data and the comparativegenomic evidence requires further experimental work toresolveThe status of SH1ORF55 has recently been confirmedby studies of the related virus HHIV-2 (discussed above)

The pleolipovirus group has expanded dramatically inthe last few years and it now comprises a diverse group ofviruseswith different genome types and replication strategiesWhat is evenmore remarkable is that a similar expansion cannow be seen with SH1-related viruses which includes viruseswith at least two replication modes and genome types (linearand circular) Evidence from haloarchaeal genome sequences

Archaea 13

show cases of not only provirus integrants or plasmids (egpKH2 and pHH205) but also genomic loci (ViPREs) thatcontain cassettes of virus genes from different sources andat least two cases (described in this study) have likely attsites that indicate circularisation and mobility While thereis a clear relationship between viruses with circular ds- andssDNA genomes that replicate via the rolling circle method(ie the ssDNA form is a replication intermediate) it ismore difficult to explain how capsid gene cassettes can movebetween these viruses and those with linear genomes andterminal proteins Within the pleolipoviruses His2 has alinear genome that contains a viral type B DNA polymeraseand terminal proteins while all the other described membershave circular genomes that contain a rep homolog and prob-ably replicate via the rolling circle method [14 19] Similarlyamong the halosphaeroviruses PH1 SH1 and HHIV-2 havelinear dsDNA with terminal proteins and probably replicatein the same way but the related SNJ1 virus has a circulardsDNA genome The clear relationships shown by the capsidgenes of viruses within each group plus the connectionsshown to haloviruses outside each group (eg the DNA poly-merases of His1 and His2) all speak of a vigorous meansof recombination one that can readily switch capsid genesbetween viruses with radically different replication strategiesHow is the process most likely to operate One possibility issuggested by ViPREs which appear to be mobile collectionsof capsid and replication genes from different sources Theyoffer fixed locations for recombination to occur providegene cassettes that can be reassorted to produce novel virusgenomes and some can probably recombine out as circularforms For example ViPREHmuk1 carries genes related topleolipoviruses halosphaeroviruses and other haloviruses Itis yet to be determined if any of the genes carried in theseloci are expressed in the cell (such as the genes for capsidproteins) or if ViPREs provide any selective advantage to thehost

4 Materials and Methods

41 Water Sample A water sample was collected in 1998from Pink Lake (32∘001015840 S and 115∘301015840 E) a hypersalinelake on Rottnest Island Western Australia Australia It wasscreened in the same year for haloviruses using the methodsdescribed in [8] A single plaque on a Har hispanica lawnplate was picked and replaque purified The novel haloviruswas designated PH1

42 Media Strains and Plasmids The media used in thisstudy are described in the online resource The Halohand-book (httpwwwhaloarchaeacomresourceshalohandbookindexhtml) Artificial salt water containing 30 (wv) totalsalts comprised of 4M NaCl 150mM MgCl

2 150mM

MgSO4 90mM KCl and 35mM CaCl

2and adjusted to pH

75 using sim2mL 1M Tris-HCl (pH 75) per litre MGM con-taining 12 18 or 23 (wv) total salts and HVD (halo-virus diluent) were prepared from the concentrated stock aspreviously described [46] Bacto-agar (Difco Laboratories)was added to MGM for solid (15 gL) or top-layer (7 gL)media

Table 1 lists the haloarchaea used in this study Allhaloarchaea were grown aerobically at 37∘C in either 18 or23 (wv)MGM(depending on the strain) andwith agitation(except for ldquoHar sinaiiensisrdquo)The plasmids used were pBlue-script II KS+ (Stratagene Cloning Systems) pUBP2 [47]and pWL102 [48] pUBP2 and pWL102 were first passagedthrough E coli JM110 [49] to prevent dam-methylationof DNA which has been shown to reduce transformationefficiency in some strains [45]

43 Negative-Stain Electron Microscopy The method fornegative-stain TEM was adapted from that of V Tarasovdescribed in the online resource The Halohandbook (httpwwwhaloarchaeacomresourceshalohandbookindexhtml)A 20120583L drop of the sample was placed on a clean surfaceand the virus particles were allowed to adsorb to a Formvarfilm 400-mesh copper grid (ProSciTech) for 15ndash2min Theywere then negatively stained with a 20120583L drop of 2 (wv)uranyl acetate for 1-2min Excess liquid was absorbed withfilter paper and the grid was allowed to air dry Grids wereexamined either on a Philips CM 120 BioTwin transmissionelectron microscope (Royal Philips Electronics) operating atan accelerating voltage of 120 kV or on a Siemens Elmiskop102 transmission electron microscope (Siemens AG) operat-ing at an accelerating voltage of 120 kV

44 Virus Host Range Twelve haloarchaeal strains from thegenera Haloarcula Halobacterium Haloferax HalorubrumandNatrialba (Table 1)Thirteen naturalHalorubrum isolates(H Camakaris unpublished data) and five uncharacterizedhaloarchaeal isolates fromLakeHardy Pink Lake (inWesternAustralia) and Serpentine Lake (D Walker and M K Seahunpublished data) were screened for PH1 susceptibilityLysates from PH1-infected Har hispanica cultures (1 times1011 PFUmL) were spotted onto lawns of each strain (usingmedia with 12 or 18 (wv)MGM depending on the strain)and incubated for 2ndash5 days at 30 and at 37∘C

45 Large Scale Virus Growth and Purification Liquid cul-tures of PH1 were grown by the infection (MOI 005) of anearly exponential Har hispanica culture in 18 (wv) MGMCultures were incubated aerobically at 37∘C with agitationfor 3 days Clearing (ie complete cell lysis) of PH1-infectedcultures did not occur and consequently cultures were harv-ested when the absorbance at 550 nm reached the minimumand the titre (determined by plaque assay) was at the maxi-mum usually sim1010ndash1011 PFUmL Virus was purified usingthe method described previously for SH1 [8 31] except thatafter the initial low speed spin (Sorvall GSA 6000 rpm30min 10∘C) virus was concentrated from the infected cul-ture by centrifugation at 26000 rpm (13 hr 10∘C) onto acushion of 30 (wv) sucrose in HVDThe pellet was resus-pended in a small volume of HVD and loaded onto a pre-formed linear 5ndash70 (wv) sucrose gradient followed byisopycnic centrifugation in 13 gmL CsCl (Beckman 70Ti60000 rpm 20 hr 10∘C) The white virus band occurredat a density of 129 gmL and was collected and diluted inhalovirus diluent (HVD see Section 4) and the virus pelleted(Beckman SW55 35000 rpm 75min 10∘C) and resuspended

14 Archaea

in a small volume of HVD and stored at 4∘C Virus recoveryat the major stages of a typical purification is given insupplementary Table 1 The specific infectivity of pure virussolutions was determined as the ratio of the PFUmL to theabsorbance at 260 nm

46 PH1 Single-Step Growth Curve An early exponentialphase culture of Har hispanica grown in 18 (wv) MGMwas infected with PH1 (MOI 50) Under these conditions thepercentage of infected cells was approximately 30 After anadsorption period of 1 hr at 37∘C the cells were washed threetimes with 18 (wv) MGM (at room temperature) resus-pended in 100mL 18 (wv) MGM (these methods ensuredthe removal of all residual virus) and incubated at 37∘Cwith shaking (100 rpm) Samples were removed at hourlyintervals for measurements of absorbance at 550 nm andthe number of infective centres Immediately after samplingtitres were determined by plaque assay with an indicator lawnof Har hispanica Each experiment was performed in tripli-cate

47 Halovirus Stability After various treatments sampleswere removed and diluted in HVD and virus titres weredetermined by plaque assay on Har hispanica Each experi-ment was performed in triplicate and representative data areshown Chloroform sensitivity was examined by the exposureof PH1 lysates in 18 (wv) MGM to chloroform in a volumeratio of 1 4 (chloroform to lysate) Incubation was at roomtemperature with constant agitation At appropriate timepoints the mix was allowed to settle and a sample wasremoved from the upper layer and diluted in HVDThe effectof a low ionic environment was examined by diluting PH1lysates (in 18 (wv) MGM) into double distilled H

2O in a

volume ratio of 1 1000 (lysate to double distilledH2O) Incu-

bation was at room temperature with constant agitationSamples were removed at various times and diluted in HVDThe pH stability of PH1 was determined by dilution of PH1lysates in 18 (wv)MGMin the appropriate pHbuffer (HVDbuffered with appropriate Tris-HCl) in a volume ratio of1 100 (lysate to buffer) Incubation was at room temperaturewith constant agitation After 30min samples were removedand diluted in HVD Thermal stability of PH1 was examinedby a 1 hr incubation of a virus lysate in 18 (wv) MGM atdifferent temperatures after which they were brought quicklyto room temperature diluted in HVD and titrated

48 Protein Procedures To remove salts purified virus prepa-rations were mixed with trichloroacetic acid (10 (vv) finalconcentration) and incubated on ice for 15min to allowthe proteins to precipitate After centrifugation (16000 g15min room temperature) the precipitate was washed threetimes in acetone dried and resuspended in double distilledH2O Proteins were dissolved in Laemmli sample buffer with

48mM 120573-mercaptoethanol [50] heated in boiling water for5min then separated on 12 (wv) NuPAGE Novex Bis-Tris Gels using MES-SDS running buffer according to themanufacturerrsquos directions (Invitrogen) After electrophoresisgels were rinsed in double distilled H

2O and stained with

01 (wv) Brilliant Blue G in 40 (vv) methanol and

10 (vv) acetic acid Gels were destained with severalchanges of 40 (vv) methanol and 10 (vv) acetic acidAlternatively gels were stained with GelCode GlycoproteinStaining Kit Pro-Q Emerald 488 Glycoprotein Gel and BlotStain Kit SYPRO Ruby Protein Gel Stain according tothe manufacturerrsquos directions (Invitrogen Molecular ProbesPierce Biotechnology)

Protein bands were cut from the gels and sent to theAustralian Proteome Analysis Facility (Macquarie Univer-sity) for trypsin digestion and analysis by matrix assistedlaser desorption ionisation time of flight mass spectrometry(MALDI-TOF MS) on an Applied Biosystems 4700 Pro-teomics Analyser (Applied Biosystems)

49 DNA Procedures Proteinase K-treated and nonpro-teinaseK-treatedDNApreparationsweremade frompurifiedPH1 using the methods described previously for SH1 [8]DNA was separated on 1 (wv) agarose gels in Tris-acetate-EDTA electrophoresis buffer and was stained with ethidiumbromide (Sigma-Aldrich)120582 DNA DNase I (RNase-free) exonuclease III mung

bean nuclease nuclease BAL-31 T4 DNA polymerase T4DNA ligase T7 exonuclease and type II restriction endonu-cleases were purchased from New England Biolabs RNase Aand yeast transfer RNA were purchased from Sigma-AldrichProteinase K was purchased from Promega Oligonucleotideprimers were purchased from Geneworks Australia

To clone fragments of the PH1 genome purified virusDNA was digested either with AccI Eco0109I and MseI orSmaI restriction endonucleases blunted with DNA Poly-merase I Large (Klenow) Fragment where appropriate andligated into AccI- Eco0109I- or SmaI-digested pBluescript IIKS+ using T4 DNA ligase according to the manufacturerrsquosinstructions (New England Biolabs) The DNA was intro-duced intoE coliXL1-Blue and transformants grown onLuriaagar containing 15 120583gmL tetracycline and 100 120583gmL ampi-cillinThe resulting cloneswere sequenced and the sequenceswere used to design specific oligonucleotide primers for PCRamplification andor primer walking using the virus genome(or specific restriction fragments) as templates

To amplify PH1 nucleic acid approximately 10 ng virusDNA was combined with 500 nM primers 100120583M of eachdNTP 1U Deep Vent DNA Polymerase (New England Bio-labs) and 1 timesThermoPol buffer (New England Biolabs) ina total reaction volume of 50 120583L Template was denatured at95∘C for 10min followed by 30 cycles of denaturation at 95∘Cfor 30 sec annealing at 56∘C for 30 sec extension at 75∘C for2min and a final extension at 75∘C for 10min PCR reactionswere performed on a PxE02ThermoCycler (Thermo ElectroCorporation) Sequencing reactions using 32 pmol primerwere performed by the dideoxy chain termination methodusing ABI PRISM Big Dye Terminator Mix version 31 onan ABI 3100 capillary sequencer (Applied Biosystems) bythe Applied Genetic Diagnostics Sequencing Service at theDepartment of Pathology (The University of Melbourne)

To obtain terminal genomic fragments PH1 DNA wasdigested by the enzyme Pas1 and the head (sim700 bp) andthe tail (sim1300 bp) fragments were purified by the QIAEXII gel extraction kits and treated with 05M piperidine for

Archaea 15

2 hr at 37∘C to remove protein residues The piperidine-treated fragments were sequenced by the Genomics BiotechCompany (Taipei Taiwan) using two primers TGACCA-ATTAATTAGGCCGGTTCGCC (PH1 head-R) and GTGC-CATACTGCTACAATTCT (PH1 tail-F)

Four hundred fifty-four whole genome sequencing PH1DNA samples (sim5 120583g) were sequenced using parallel pyrose-quencing on a Roche 454 Genome Sequencer System atMission Biotech (Taipei Taiwan) The largest contig was27399with 8803 reads (Newbler version 27) with 27387 posi-tions being of Q40 quality The average depth of coverage foreach base was 829TheGenBank accession for the entire PH1sequence is KC252997

Acknowledgments

The authors thank Carolyn Bath Mei Kwei (Joan) Seahand Danielle Walker for their early work in PH1 characteri-zation Simon Crawford Jocelyn Carpenter and Anna Fried-huber for their assistance with the transmission electronmicroscopy and Tiffany Cowie and Voula Kanellakis fortheir sequencing assistance This paper has been facilitatedby access to the Australian Proteome Analysis Facility estab-lished under the Australian Governmentrsquos Major NationalResearch Facilities program SLT was supported by a grantfrom the Biodiversity Research Center Academia SinicaTaiwan

References

[1] M Krupovic M F White P Forterre and D PrangishvilildquoPostcards from the edge structural genomics of archaealvirusesrdquo Advances in Virus Research vol 82 pp 33ndash62 2012

[2] P Stolt andW Zillig ldquoGene regulation in halophage 120601Hmdashmorethan promotersrdquo Systematic and Applied Microbiology vol 16no 4 pp 591ndash596 1994

[3] R Klein U Baranyi N Rossler B Greineder H Scholz and AWitte ldquoNatrialba magadii virus 120593Ch1 first complete nucleotidesequence and functional organization of a virus infecting ahaloalkaliphilic archaeonrdquo Molecular Microbiology vol 45 no3 pp 851ndash863 2002

[4] E Pagaling R D Haigh W D Grant et al ldquoSequence analysisof an Archaeal virus isolated from a hypersaline lake in InnerMongolia Chinardquo BMC Genomics vol 8 article 410 2007

[5] S L Tang SNuttall andMDyall-Smith ldquoHalovirusesHF1 andHF2 evidence for a recent and large recombination eventrdquo Jour-nal of Bacteriology vol 186 no 9 pp 2810ndash2817 2004

[6] C Bath and M L Dyall-smith ldquoHis1 an archaeal virus of theFuselloviridae family that infectsHaloarcula hispanicardquo Journalof Virology vol 72 no 11 pp 9392ndash9395 1998

[7] C Bath T Cukalac K Porter andM L Dyall-Smith ldquoHis1 andHis2 are distantly related spindle-shaped haloviruses belongingto the novel virus group Salterprovirusrdquo Virology vol 350 no1 pp 228ndash239 2006

[8] K Porter P Kukkaro J K H Bamford et al ldquoSH1 a novelspherical halovirus isolated from an Australian hypersalinelakerdquo Virology vol 335 no 1 pp 22ndash33 2005

[9] M Dyall-Smith S L Tang and C Bath ldquoHaloarchaeal viruseshow diverse are theyrdquo Research in Microbiology vol 154 no 4pp 309ndash313 2003

[10] A Oren G Bratbak and M Heldal ldquoOccurrence of virus-likeparticles in the Dead Seardquo Extremophiles vol 1 no 3 pp 143ndash149 1997

[11] T Sime-Ngando S Lucas A Robin et al et al ldquoDiversityof virus-host systems in hypersaline Lake Retba SenegalrdquoEnvironmental Microbiology vol 13 no 8 pp 1956ndash1972 2010

[12] N Guixa-Boixareu J I Calderon-Paz M Heldal G Bratbakand C Pedros-Alio ldquoViral lysis and bacterivory as prokaryoticloss factors along a salinity gradientrdquoAquaticMicrobial Ecologyvol 11 no 3 pp 215ndash227 1996

[13] S T Jaakkola R K Penttinen S T Vilen et al ldquoClosely relatedarchaeal Haloarcula hispanica icosahedral viruses HHIV-2 andSH1 have nonhomologous genes encoding host recognitionfunctionsrdquo Journal of Virology vol 86 no 9 pp 4734ndash47422012

[14] E Roine P Kukkaro L Paulin et al ldquoNew closely related halo-archaeal viral elements with different nucleic acid typesrdquo Jour-nal of Virology vol 84 no 7 pp 3682ndash3689 2010

[15] A Sencilo L Paulin S KellnerMHelm and E Roine ldquoRelatedhaloarchaeal pleomorphic viruses contain different genometypesrdquo Nucleic Acids Research vol 40 no 12 pp 5523ndash55342012

[16] H T Jaalinoja E Roine P Laurinmaki H M Kivela D HBamford and S J Butcher ldquoStructure and host-cell interactionof SH1 a membrane-containing halophilic euryarchaeal virusrdquoProceedings of the National Academy of Sciences of the UnitedStates of America vol 105 no 23 pp 8008ndash8013 2008

[17] M K Pietila N S Atanasova VManole et al ldquoVirion architec-ture unifies globally distributed pleolipoviruses infecting halo-philic archaeardquo Journal of Virology vol 86 no 9 pp 5067ndash50792012

[18] M L Holmes and M L Dyall-Smith ldquoA plasmid vector witha selectable marker for halophilic archaebacteriardquo Journal ofBacteriology vol 172 no 2 pp 756ndash761 1990

[19] M L Holmes F Pfeifer andM L Dyall-Smith ldquoAnalysis of thehalobacterial plasmid pHK2 minimal repliconrdquo Gene vol 153no 1 pp 117ndash121 1995

[20] M L Dyall-Smith F Pfeiffer K Klee et al ldquoHaloquadratumwalsbyi limited diversity in a Global Pondrdquo PLoS ONE vol 6no 6 Article ID e20968 2011

[21] S Casjens ldquoProphages and bacterial genomics what have welearned so farrdquoMolecular Microbiology vol 49 no 2 pp 277ndash300 2003

[22] R W Hendrix M C M Smith R N Burns M E Ford andG F Hatfull ldquoEvolutionary relationships among diverse bacte-riophages and prophages all the worldrsquos a phagerdquo Proceedings ofthe National Academy of Sciences of the United States of Americavol 96 no 5 pp 2192ndash2197 1999

[23] D Botstein ldquoA theory ofmodular evolution for bacteriophagesrdquoAnnals of the New York Academy of Sciences vol 354 pp 484ndash491 1980

[24] M Krupovic S Gribaldo D H Bamford and P Forterre ldquoTheevolutionary history of archaeal MCM helicases a case study ofvertical evolution combinedwithHitchhiking ofmobile geneticelementsrdquo Molecular Biology and Evolution vol 27 no 12 pp2716ndash2732 2010

[25] M Krupovic P Forterre and D H Bamford ldquoComparativeanalysis of the mosaic genomes of tailed archaeal viruses andproviruses suggests common themes for virion architecture andassembly with tailed viruses of bacteriardquo Journal of MolecularBiology vol 397 no 1 pp 144ndash160 2010

16 Archaea

[26] Z Zhang Y Liu S Wang et al et al ldquoTemperate membrane-containing halophilic archaeal virus SNJ1 has a circular dsDNAgenome identical to that of plasmid pHH205rdquoVirology vol 434no 2 pp 233ndash241 2012

[27] D H Bamford J J Ravantti G Ronnholm et al ldquoConstituentsof SH1 a novel lipid-containing virus infecting the halophiliceuryarchaeonHaloarcula hispanicardquo Journal of Virology vol 79no 14 pp 9097ndash9107 2005

[28] H M Kivela E Roine P Kukkaro S Laurinavicius P Somer-harju andDH Bamford ldquoQuantitative dissociation of archaealvirus SH1 reveals distinct capsid proteins and a lipid corerdquoVirology vol 356 no 1-2 pp 4ndash11 2006

[29] H Jaalinoja Electron Cryo-Microscopy Studies of Bacteriophage8 and Archaeal Virus SH1 University of Helsinki HelsinkiFinland 2007

[30] K Porter B E Russ J Yang andM L Dyall-Smith ldquoThe trans-cription programme of the protein-primed halovirus SH1rdquoMi-crobiology vol 154 no 11 pp 3599ndash3608 2008

[31] K Porter and M L Dyall-Smith ldquoTransfection of haloarchaeaby the DNAs of spindle and round haloviruses and the useof transposon mutagenesis to identify non-essential regionsrdquoMolecular Microbiology vol 70 no 5 pp 1236ndash1245 2008

[32] S E Luria General Virology John Wiley and Sons New YorkNY USA 1953

[33] C A Thomas K Saigo E McLeod and J Ito ldquoThe separationofDNA segments attached to proteinsrdquoAnalytical Biochemistryvol 93 pp 158ndash166 1979

[34] A L Delcher D Harmon S Kasif OWhite and S L SalzbergldquoImproved microbial gene identification with GLIMMERrdquoNucleic Acids Research vol 27 no 23 pp 4636ndash4641 1999

[35] R Zhang Y Yang P Fang et al ldquoDiversity of telomere palin-dromic sequences and replication genes among Streptomyceslinear plasmidsrdquo Applied and Environmental Microbiology vol72 no 9 pp 5728ndash5733 2006

[36] X Wang H S Lee F J Sugar F E Jenney M W W Adamsand J H Prestegard ldquoPF0610 a novel winged helix-turn-helix variant possessing a rubredoxin-like Zn ribbon motiffrom the hyperthermophilic archaeon Pyrococcus furiosusrdquoBiochemistry vol 46 no 3 pp 752ndash761 2007

[37] S W Cline L C Schalkwyk and W F Doolittle ldquoTransfor-mation of the archaebacterium Halobacterium volcanii withgenomic DNArdquo Journal of Bacteriology vol 171 no 9 pp 4987ndash4991 1989

[38] S Kamtekar A J Berman J Wang et al ldquoThe 12059329 DNA poly-merase protein-primer structure suggests a model for the ini-tiation to elongation transitionrdquo EMBO Journal vol 25 no 6pp 1335ndash1343 2006

[39] X Peng T Basta M Haring R A Garrett and D PrangishvilildquoGenome of theAcidianus bottle-shaped virus and insights intothe replication and packaging mechanismsrdquo Virology vol 364no 1 pp 237ndash243 2007

[40] C C Yang C H Huang C Y Li Y G Tsay S C Lee and CW Chen ldquoThe terminal proteins of linear Streptomyces chrom-osomes and plasmids a novel class of replication primingproteinsrdquo Molecular Microbiology vol 43 no 2 pp 297ndash3052002

[41] C C Yang Y H Chen H H Tsai C H Huang T W Huangand C W Chen ldquoIn vitro deoxynucleotidylation of the term-inal protein of Streptomyces linear chromosomesrdquo Applied andEnvironmental Microbiology vol 72 no 12 pp 7959ndash79612006

[42] P C Chang and S N Cohen ldquoBidirectional replication froman internal origin in a linear Streptomyces plasmidrdquo Science vol265 no 5174 pp 952ndash954 1994

[43] S W Wilhelm W H Jeffrey C A Suttle and D L MitchellldquoEstimation of biologically damaging UV levels in marinesurface waters with DNA and viral dosimetersrdquo Photochemistryand Photobiology vol 76 no 3 pp 268ndash273 2002

[44] T Allers and M Mevarech ldquoArchaeal geneticsmdashthe third wayrdquoNature Reviews Genetics vol 6 no 1 pp 58ndash73 2005

[45] M L Holmes S D Nuttall and M L Dyall-Smith ldquoConstruc-tion and use of halobacterial shuttle vectors and further studieson Haloferax DNA gyraserdquo Journal of Bacteriology vol 173 no12 pp 3807ndash3813 1991

[46] S D Nuttall and M L Dyall-Smith ldquoHF1 and HF2 novelbacteriophages of halophilic archaeardquo Virology vol 197 no 2pp 678ndash684 1993

[47] U Blaseio and F Pfeifer ldquoTransformation of Halobacteriumhalobium development of vectors and investigation of gas vesi-cle synthesisrdquo Proceedings of the National Academy of Sciences ofthe United States of America vol 87 no 17 pp 6772ndash6776 1990

[48] W L Lam and W F Doolittle ldquoShuttle vectors for the archae-bacterium Halobacterium volcaniirdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 86 no14 pp 5478ndash5482 1989

[49] C Yanisch-Perron J Vieira and J Messing ldquoImproved M13phage cloning vectors and host strains nucleotide sequences ofthe M13mp18 and pUC19 vectorsrdquo Gene vol 33 no 1 pp 103ndash119 1985

[50] U K Laemmli ldquoCleavage of structural proteins during theassembly of the head of bacteriophage T4rdquo Nature vol 227 no5259 pp 680ndash685 1970

[51] D G Burns H M Camakaris P H Janssen and M L Dyall-Smith ldquoCombined use of cultivation-dependent and culti-vation-independent methods indicates that members of mosthaloarchaeal groups in anAustralian crystallizer pond are culti-vablerdquo Applied and Environmental Microbiology vol 70 no 9pp 5258ndash5265 2004

[52] G Juez F Rodriguez-Valera A Ventosa and D J KushnerldquoHaloarcula hispanica spec nov and Haloferax gibbonsii specnov two new species of extremely halophilic archaebacteriardquoSystematic and Applied Microbiology vol 8 pp 75ndash79 1986

[53] A Oren M Ginzburg B Z Ginzburg L I Hochstein and BE Volcani ldquoHaloarcula marismortui (Volcani) sp nov nomrev an extremely halophilic bacterium from the Dead SeardquoInternational Journal of Systematic Bacteriology vol 40 no 2pp 209ndash210 1990

[54] M Torreblanca F Rodriguez-Valera G Juez A Ventosa MKamekura and M Kates ldquoClassification of non-alkaliphilichalobacteria based on numerical taxonomy and polar lipidcomposition and description of Haloarcula gen nov andHaloferax gen novrdquo Systematic and Applied Microbiology vol8 pp 89ndash99 1986

[55] A Ventosa and A Oren ldquoHalobacterium salinarum nom cor-rig a name to replace Halobacterium salinarium (Elazari-Vol-cani) and to include Halobacterium halobium and Halobac-terium cutirubrumrdquo International Journal of Systematic Bacte-riology vol 46 no 1 p 347 1996

[56] M C Gutierrez M Kamekura M L Holmes M L Dyall-Smith and A Ventosa ldquoTaxonomic characterization of Halo-ferax sp (ldquoH alicanteirdquo) strain Aa 22 description of Haloferaxlucentensis sp novrdquo Extremophiles vol 6 no 6 pp 479ndash4832002

Archaea 17

[57] M F Mullakhanbhai and H Larsen ldquoHalobacterium volcaniispec nov a Dead Sea halobacterium with a moderate saltrequirementrdquo Archives of Microbiology vol 104 no 1 pp 207ndash214 1975

[58] S D Nuttall and M L Dyall-Smith ldquoCh2 a novel halophilicarchaeon from an Australian solar salternrdquo International Jour-nal of Systematic Bacteriology vol 43 no 4 pp 729ndash734 1993

[59] P D Franzman E Stackebrandt K Sanderson et al ldquoHalobac-terium lacusprofundi sp nov a halophilic bacterium isolatedfrom Deep Lake Antarcticardquo Systematic and Applied Microbi-ology vol 11 pp 20ndash27 1988

[60] G A Tomlinson and L I Hochstein ldquoHalobacterium saccharo-vorum sp nov a carbohydrate metabolizing extremely halo-philic bacteriumrdquoCanadian Journal ofMicrobiology vol 22 no4 pp 587ndash591 1976

[61] A Ventosa M C Gutierrez M Kamekura and M L Dyall-Smith ldquoProposal to transfer Halococcus turkmenicus Halobac-terium trapanicum JCM9743 and strainGSL-11 toHaloterrigenaturkmenica gen nov comb novrdquo International Journal ofSystematic Bacteriology vol 49 no 1 pp 131ndash136 1999

[62] M Kamekura and M L Dyall-Smith ldquoTaxonomy of the familyHalobacteriaceae and the description of two new genera Halo-rubrobacterium and Natrialbardquo Journal of General and AppliedMicrobiology vol 41 no 4 pp 333ndash350 1995

[63] T J Carver KM RutherfordM BerrimanM A RajandreamB G Barrell and J Parkhill ldquoACT the Artemis comparisontoolrdquo Bioinformatics vol 21 no 16 pp 3422ndash3423 2005

[64] S W Cline W L Lam R L Charlebois L C Schalkwyk andW F Doolittle ldquoTransformationmethods for halophilic archae-bacteriardquo Canadian Journal of Microbiology vol 35 no 1 pp148ndash152 1989

2 Archaea

several different haloarchaea Later descriptions of halo-viruses HRPV-1 HHPV-1 and several others revealed a spec-trum of related viruses with very different genome structures(circular ds- and ssDNA) lengths and replication strategies[14 15] but they all share a similar set of capsid proteinsThe mechanisms underlying the movement of the capsidgenes between viruses having such different characteristicsand modes of replication (protein primed versus rollingcircle) remain unclear One possibility is that such modularrecombinations could be facilitated by previously describedgenomic loci of virus and plasmid-related elements (ViPREs)[20] Initially these did not appear to be simple provirusgenomes in varying states of decay but the description ofpleolipoviruses with genomes much smaller than His2 anddiffering replication strategy can explain many of them asvirus integrants On the other hand some of these genomicloci have expanded in length as more virus and plasmidsequences become available and their gene homologs can berecognised in flanking sequences For these we retain theViPRE epithet From the decades of the study of bacterio-phages it is well documented that related phages frequentlyrecombine (by both legitimate and illegitimate recombina-tion events) giving rise to mosaic genomes with varyingevolutionary histories [21 22] This process can be assistedby the modular arrangement of genes (eg capsid formationor replication) commonly found inmany virus genomes [23]The same appears true of haloviruses such as the large recom-bination event evident in the comparison of halovirus HF1and HF2 genomes [5] More generally there is good evidenceof widespread recombination from the study of archaealMCM helicase genes (often associated with mobile geneticelements) [24] and in the comparative genomics of archaealcaudoviruses [25]

The group of round haloviruses exemplified by SH1 hasbeen recently expanded by the descriptions of HHIV-2 andSNJ1 [13 26] Members of the SH1 virus group are related bytheir capsid proteins but differ in genome length and repli-cation strategy SH1 and HHIV-2 have linear dsDNA gen-omes (sim305 kb) with terminal proteins indicative of repli-cation by protein priming while SNJ1 has a circular dsDNAgenome of only 163 kbThis diversity of genome types withinthe same virus group closely parallels that of the pleolipo-viruses described above SH1 has been the most intensivelystudied including host range particle stability virion struc-ture genome sequence transcription mapping transfectionand the establishment of a method of genetic manipulation[8 13 16 27ndash31] The aim of the current study is to describea new member of this group PH1 Its virological characteris-tics genome sequence and major proteins are presented andcompared to other members of the SH1 virus group and togenomic loci containing related genes For convenience weprovide the group name halosphaerovirus to encompass theSH1 virus group currently consisting of SH1 PH1 HHIV-2and SNJ1

2 Results

21 Virus Isolation and Host Range A water sample fromPink Lake a hypersaline lake (32∘001015840 S and 115∘301015840 E) in

100 nm

Figure 1 Purified PH1 virus examined by negative-stain electronmicroscopy Particles were stained with 2 (wv) uranyl acetateScale bar represents 100 nm

western Australia was screened for haloviruses by platingdirectly on lawns of Har hispanica using overlay plates ofmodified growth medium (MGM) with 12 or 18 salts(wv) A high titre of similar plaque morphology was ob-served (12 times 105 PFUmL) on the plates with 18 (wv) saltsA novel halovirus was isolated from one of these plaquesand designated PH1 based on the source (Pink Lake) and theisolating host (Har hispanica)

Plaques were fully developed after two days at 37∘Cusing overlay plates of 18 (wv) MGM and were 1-2mm indiameter clear and with ragged edges At 30∘C plaques tookthree days to develop and at 25∘C plaques were hazy andtook seven days to develop (data not shown)

The host range of PH1 was identical to that of SH1 [8] itwas unable to plaque on lawns of 12 different species belong-ing to six different genera of the Halobacteriaceae (Haloar-cula Halobacterium Haloferax Halorubrum Haloterrigenaand Natrialba) but could plaque on Halorubrum strain CSW2094 (described in [8]) an uncharacterized Australianisolate (Table 1)

22 Virus Purification and Particle Morphology Virus waspurified by a slight modification of the method describedpreviously for halovirus SH1 ([8] and Section 4) PH1 bandedat a density of 129 gmL in CsCl gradients and gave a finalspecific infectivity of sim5 times 1011 PFUA

260 Negative-stain

electron microscopy revealed spherical particles with anaverage diameter of sim51 nm (Figure 1) The capsid displayedtwo layers with a compact core particle of sim43 nm indiameterThismorphology is similar to that of halovirus SH1a closely related virus (see below) that was isolated at the sametime as PH1 but from a neighbouring salt lake [8]

23 Single-Burst and Single-Step Growth of PH1 Around 30of Har hispanica cells could be infected by PH1 and singleburst experiments [32] indicated an average burst size of87 PFUcell (data not shown) This value was supported bysingle-step growth curves which gave burst sizes of between50 and 100 PFUcellThe number of infectious centres usuallyincreased at 4ndash6 hr pi but cell lysis did not appear to begin

Archaea 3

Table 1 Strains used in this study

Speciesisolate Strain ReferenceCSW 2094 Original isolate [51]Haloarcula hispanica ATCC 33960 [52]Haloarcula marismortui ATCC 43044 [53]ldquoHaloarcula sinaiiensisrdquo ATCC 33800 [54]Halobacterium salinarum NCIMB 763 [55]Haloferax gibbonsii ATCC 33959 [52]Haloferax lucentense NCIMB 13854 [56]Haloferax mediterranei ATCC 33500 [54]Haloferax volcanii ATCC 29605 [57]Halorubrum coriense ACAM 3911 [58]Halorubrum lacusprofundi ACAM 34 [59]Halorubrum saccharovorum NCIMB 2081 [60]Haloterrigena turkmenica NCIMB 784 [61]Natrialba asiatica JCM 9576 [62]

until well after this usually between 14 and 24 hr pi Arepresentative example of a single-step growth curve for PH1is shown in Figure 2 where the rise begins at sim6 hr pi andvisible cell lysis begins at sim14 hr pi This figure shows that asecond round of growth occurs at around 18 hr pi reflectingthe initial infection of only 30 of cells and it is during thissecond round of virus release that the cell density decreasesmost rapidly reaching a value of about 01 A

550 which is close

to the initial cell density In this example the average burstsize for the first growth step was 95 PFUcell

24 Virus Stability The stability of PH1 virus was testedunder several conditions (Figures 3(a)ndash3(d))When stored inHVD at 4∘C the infectivity of PH1 remained unaltered forseveral months (data not shown) A thermal stability curveis presented in Figure 3(a) and shows that PH1 is stable upto 56∘C above in which it rapidly loses titre Particles weresensitive to a reduced salt environment (Figure 3(b)) and tochloroform (Figure 3(d)) PH1 was most stable between pH8 and pH 9 (Figure 3(c)) In general the stability of PH1 wassimilar to that described previously for SH1 [8]

25 PH1 Structural Proteins and Protein Complexes Theproteins of purified PH1 virus were separated by SDS-polyacrylamide gel electrophoresis alongside the proteins ofSH1 virus (Figure 4) Nine PH1 protein bands were detectedwith molecular weights from 7 to 185 kDa (Figure 4(a)) andthese were designatedwith the prefixVP and a number corre-sponding to the homologous protein of SH1 [8 27] (see later)This nomenclature is consistent with that of the recentlydescribed HHIV-2 virus a member of the same virus group[13]The protein profile of PH1 was very similar to that of SH1[8] and with similar relative masses of the protein bands Onenotable difference was that PH1 proteins VP9 and VP10 ranclosely together (calculatedMWs are 165 and 167 kDa resp)compared to their SH1 homologs (165 and 169 kDa resp)It is probable given the strong sequence similarity to SH1(see later) that at least six additional PH1 structural proteinswere unable to be visualized using our staining techniques

0 5 10 15 200

01

02

Cell

den

sity

(A55

0)

Time (hr)

Viru

s titr

e (PF

Um

L)

1010

109

108

107

106

105

Figure 2 Growth curve of PH1 An early exponential culture ofHar hispanica in 18 (wv) MGM was infected with virus (MOI50) washed to remove unbound virus and incubated at 37∘C Atregular intervals samples were removed the absorbance at 550 nmwere measured (circles) and the number of infectious centres weredetermined by plaque assay (squares) Error bars represent onestandard deviation of the average titre

The potential glycosylation of virus proteins was examinedby staining similar protein gels either with a periodate-acid-Schiff (PAS) stain (GelCode Glycoprotein Staining Kit PierceBiotechnology USA) or the more sensitive fluorescent stain(Pro-Q Emerald 488 Glycoprotein Gel and Blot Stain KitMolecular Probes USA) No glycoproteins were detected

The major protein bands of PH1 (asterisked in Figure 4)were excised from gels digestedwith trypsin and analysed byMALDI-TOFMS and the results are summarized in Table 2From these data the structural proteins VP 1ndash4 7 9 10 and12 were found to be specified by ORFs 12 24 28 21 20 27 26and 19 respectively (see later)

26 Characteristics of the PH1 Genome Nucleic acid wasextracted from purified PH1 virus preparations treated withproteinase K and incubated with various nucleases to deter-mine the characteristics of the genome (Table 3) The PH1genome was sensitive to dsDNA endo- and exonucleasesbut not to ssDNA nuclease (mung bean) or RNase A Thisindicated that the PH1 genome is linear dsDNA with free(ie not covalently closed) termini Restriction endonucleasedigestions of the PH1 genome gave a length of approximately29 kb (data not shown)

The presence of terminal bound proteins was examinedusing a silica binding assay [33] Nonproteinase K-treatedPH1 DNA was restricted with AseI and the four resultingfragments passed through GFC filters under conditionswhere proteins bind firmly to the glass As shown in Figure 5the two internal AseI fragments (23 and 76 kb) passedthrough the filter but the right terminal fragment (174 kb)was bound indicating it carried an attached protein Theleft terminal fragment was too small (062 kb) and faintly

4 Archaea

0 20 40 60 80 100

Titre

(PFU

mL)

Temperature1011

1010

109

108

107

106

Temperature (∘C)

le105

(a)

0 5 10 15

0

Time (min)

Salt concentration

Titre

(PFU

mL)

107

106

105

104

minus5

(b)

4 5 6 7 8 9 10pH

pH

Titre

(PFU

mL)

108

107

(c)

Exposure time (min)0 5 10 15 20

Chloroform

Titre

(PFU

mL)

108

107

(d)

Figure 3 Stability of PH1 virus to various treatments and conditions Virus preparations (infected-cell supernatants in 18 (wv)MGM)wereexposed to various conditions after which the virus titre was determined (in duplicate) onHar hispanica cells (a) The effect of temperatureVirus was incubated for 1 hr with constant agitation at temperatures between 4 and 100∘C (b) The effect of lowered salt concentration Viruswas diluted 1 1000 in double-distilled H

2

O and incubated at room temperature with constant agitation Samples were removed at regularintervals (c)The effect of pHViruswas diluted 1 100 inTris-HCl buffers at the different pHs and incubatedwith constant agitation for 30min(d)The effect of chloroformChloroformwasmixedwith virus (1 4 ratio) and incubated at room temperaturewith constant agitation Sampleswere removed at regular intervals Open square symbols indicate where virus titres were undetectable Error bars represent one standarddeviation of the average titre

Archaea 5

Table 2 PH1 virus proteins identified by mass spectroscopy of tryptic peptides

Proteina Locus tag ORF MW (kDa) Matching peptide massesbObserved (calculated)

VP1 HhPH1 gp12 12 185 (158) 16VP2 HhPH1 gp24 24 100 (78) 15VP3 HhPH1 gp28 28 40 (38) 5VP4 HhPH1 gp21 21 35 (26) 5VP7 HhPH1 gp20 20 24 (20) 5VP9 HhPH1 gp27 27 16 (17) 3VP10 HhPH1 gp26 26 15 (17) 4VP12 HhPH1 gp19 19 7 (99) 4aVirus capsid proteins are numbered according to their similarity with SH1 capsid proteinsbThenumber of peptidemasses between 7500 and 35130mz identified byMALDI-TOFMS that correspond to the theoretical tryptic peptides of the predictedvirus protein

212 -

(kD

a)

PH1

158 -97 -66 -56 -42 -35 -

27 -

20 -

14 -

65 -

lowast

lowast

lowast

lowast

lowast

lowast

lowast

VP1VP2

VP3VP4

VP7

VP9

VP10

VP12

(a)

SH1

(b)

Figure 4 Structural proteins of halovirus PH1 Viral structuralproteinswere separated by SDS-PAGEon a 12 (wv) acrylamide geland stained with Brilliant Blue G (a) They were run in parallel withthe proteins of purified SH1 virus (b) The sizes of protein standardmarkers are indicated on the left side (in kDa) Asterisks denoteproteins bands of PH1 that were identified by mass spectroscopyThe numbering of proteins of PH1 (VP1ndashVP12) follows that of theSH1 homologs seen in (b) (and explained also in the text)

staining to be detected Further evidence was obtained by anuclease protection assay as previously used for SH1 DNA[31] As shown in Table 3 exonuclease III (a 31015840 exonuclease)was able to digest nonproteinase K-treated PH1 DNA butT7 exonuclease (a 51015840 exonuclease) and Bal31 (51015840 and 31015840exonucleases) were unable to digest PH1 DNA unless it hadbeen previously treated with proteinase KThis indicated thatboth 51015840 termini of the genome had protein attached

27 Sequence of the PH1 Genome The complete PH1 genomesequence was determined (accession KC252997) using a

1 2 3

-23

-76

-17410 -

6 -

3 -

2 -

15 -

05 -

4 -

(kB)

Figure 5 Detection of proteins bound to the termini of PH1 DNADNA was extracted from purified PH1 particles without proteinaseK digested with AseI then passed through a silica filter (GFC)under conditionswhere proteinswould bind to the filter Lane 1AseIfragments that did not bind and were eluted Lane 2 AseI fragmentsthat bound and were eluted only after protease treatment Lane 3AseI digest of PH1 DNA The positions of DNA size standards areindicated on the left side in kb The calculated sizes of the AseIfragments of PH1 DNA in lane 3 are indicated on the right side (alsoin kb) The predicted 062 kb PH1 AseI fragment was not detected

combination of cloned fragments PCR primer walkingon virus DNA and 454 whole genome sequencing (seeSection 4) It was found to be 28072 bp in length 676G + C with inverted terminal repeat sequences (ITRs) of337 bp Using GLIMMER [34] manual BLAST searching atthe GenBank database and comparison to related virusesthe PH1 genome sequence was predicted to contain 49 ORFs(Table 4) Most ORFs were closely spaced or overlappinggiving a gene density of 174 geneskb (average of 572 ntgene)The cumulative AT-skew plot of the PH1 genome shown atthe bottom of Figure 6 shows inflection points (circled) thatare consistent with the changes in transcription direction

6 Archaea

6 8 10 12 14 16 18 20 22 24 kb

1 2 4 6 8 10 12 14 162 kb

VP1

VP1-like fragments

A

A

VP2

VP2

VP7

VP1

2

20 21 22

VP4

VP1

3

VP5 p2

2

p9p4 p11

p12

p13

NP5

356A

Hmuk

p26

VP3

VP1

7

1728Hmuk

0477

Locus tag ZOD2009 19038

Locus tag HlacAJ 010100019872 NZ AGFZ01000123 (reversed)

NZ AEMG01000027 contig29

TransposaseHlacAJ 010100019767

ZOD2009 19153

Hbf lacisalsi

Hap paucihalophilus

(a)

VP18VP2VP1 A

VP1

VP18VP1

VP1

2

VP1

VP7

VP4

VP1

3

VP5

VP2

VP9

VP1

0

VP3

VP1

6

VP1

7V

P6 VP18

ATPa

se

kb1 2 4 6 8 10 12 14 16 18 20 22 24 26 28

VP2

VP2

VP18A

AA

2211 1312

94 26

HHIV-2

SH1

PH1

0

2

4

6

30 578 nt

30 889 nt

28 072 nt

(times102

)Cu

m A+

T sk

ew

Cumulative A + T skew of PH1

(b)

Figure 6 Genome alignments of haloviruses PH1 SH1 and HHIV-2 along with two related genomic loci (a) Genomic loci of Happaucihalophilus andHbf lacisalsi that contain genes related to halosphaeroviruses SH1 PH1 andHHIV-2The names andGenBank accessionsfor these contigs are given on the far right andORFs are coloured and labeled to indicate the relationships of theseORFs to those of the virusesbelowThe locus tag numbers for the first and last ORFs shown in each locus are given nearby their respectiveORFs In addition grey colouredORFs represent sequences that do not match any of the haloviruses and green coloured ORFs represent protein sequences that are closelyrelated to ORFs found within or very close to previously described virusplasmid loci so called ViPREs [20] The scale bars shown aboveeach contig show the position of the described region within the respective contig (b) The three virus genomes are labeled at the left withscale markers below (in kb) and the total length indicated at the far right At the bottom is a cumulative AT-skew plot of the PH1 genome(httpmolbiol-toolscaJie Zheng) with inflection points circledThe grey shaded bands between the genome diagrams indicate significantnucleotide similarity (usingACT [63]) AnnotatedORFs are represented by arrows with colours indicating structural proteins (red or brown)nonstructural proteins (yellow or orange) or the packaging ATPase (blue) The names of structural protein ORFs are indicated either withinthe arrow (eg VP1) or in text nearby The numbered orange coloured ORFs of HHIV-2 are homologous to ORFs found in the genomic loci(probably proviruses or provirus remnants) pictured in (a)

indicated by the annotated ORFs as has been seen in otherhaloviruses [7] There are no GATC motifs in the genomeand the inverse sequence CTAG is strongly underrepre-sented with just 2 motifs (68 expected)

When compared to the related viruses SH1 and HHIV-2(Figure 6) the PH1 genome is seen to be of similar length andis much more closely related to SH1 than to HHIV-2 (74and 54 nt identity resp) Like PH1 the genomes of SH1and HHIV-2 lack GATC motifs and CTAG is either absent(SH1) or underrepresented (HHIV-2) The ITRs of PH1 andSH1 share 785nucleotide identity but the PH1 ITR is longerthan that of SH1 (337 and 309 nt resp) partly because SH1has replaced an 18 bp sequence (gtcgtgcggtttcggcgg) found atthe internal end of its left-hand ITR with a sequence at thecorresponding position of its right-hand ITR that shows littleinverted sequence similarity In the PH1 ITR this sequenceis retained at both ends Whether this difference represents arecombination event or a mistake in replication of the ITRs

is unclear An alignment of the ITRs of all three virusesidentified a number of highly conserved regions (labeled 1ndash9 in Supplementary Figure 2 see Supplementary Materialavailable online at httpdxdoiorg1011552013456318) A16 bp sequence at the termini (region 1) is conserved con-sistent with the conservation seen at the termini of linearStreptomyces plasmids [35] A 16 bpGC-rich sequence aroundthe middle of the ITR (region 3) is also conserved Shortermotifs of either C-rich (region 8) or AT-rich sequence(regions 7 and 9) are found near the internal end of the ITRs

The grey shading between the schematic virus genomesin Figure 6(b) indicates regions of high nucleotide similarityrevealing that the differences between PH1 and SH1 are notevenly distributed but vary considerably along the lengthof the aligned genomes A comparison between SH1 andHHIV-2 has been published recently [13] and since the PH1and SH1 genomes are so similar we will focus the followingdescription on the differences between PH1 and SH1

Archaea 7

Table 3 Digestion of the PH1 genome by nucleasesa

Nuclease (amount) Proteinase K-treated UntreatedControl PH1 DNA Control PH1 DNA

DNase I (RNase-free)(100U) +b + +b +

Exonuclease III(100U) +b + +b +

Mung bean nuclease(10U) +c

minus +cminus

Nuclease BAL-31(1 U) +b + +b

minus

RNase A(5 120583gmL) +d

minus +dminus

T7 exonuclease(10U) +b + +b

minus

aExtracted nucleic acid from purified PH1 virus was treated with variousnucleases and then analysed by gel electrophoresis to detect whether thenuclease digested (+) or did not digest (minus) the genome Control nucleicacids used to confirm activity of the nucleases were b

120582 DNA ca DNAoligonucleotide and dyeast transfer RNA

Within corresponding ORFs there are blocks of codingsequence with low (or no) similarity These may represent insitu divergence or short recombination events (eg indels)There are also cases where an entire ORF in one virus hasno corresponding homolog in the other because of an inser-tiondeletion event (indel) Lastly there are replacementswhere the sequences within the corresponding regions showlow or no similarity but are flanked by sequence with highsimilarity The largest visible differences seen in Figure 6 aredue to sequence changes within or near the long genes encod-ing capsid proteins VP1 and VP18 A summary of the mainsequence differences between SH1 and PH1 and their loca-tions is given in Table 5 where these divergent regions (DV)are numbered from DV1 to DV18

The sequences of the two viruses are close enough that inmany cases the mechanism for the observed differences canbe inferred For example DV2 is likely the result of a deletionevent that has removed the SH1 ORF10 gene homolog fromthe PH1 genome At either end of the SH1 ORF are almostperfect direct repeats (cggcctgaccggcatgac) that would allowrepeat-mediated deletion to occur removing the interveningsequence Small direct repeats leading to deletions have beendescribed previously in the comparative analysis of Halo-quadratumwalsbyi genomes [20] DV5 appears to be an indelwhere SH1 ORFs 14ndash16 are absent in PH1 and an invertedrepeat (AGCCATG) found at each end of the SH1 divergentregion may be significant in the history of this changeThe proteins specified by SH1 ORFs 14ndash16 are presumablydispensable for PH1 (or their functions supplied by otherproteins) but a homolog of SH1 ORF14 is found also inHHIV-2 (ORF 6) and a homolog of SH1 ORF15 occurs inhalovirus His1 (ORF13) Replacement regions are also com-mon and can occur within ORFs (eg DV3 in capsid proteinVP1) or provide additional or alternative genes (eg DV12and DV13) Several divergent regions occur within the genefor capsid protein VP2 a hot spot for change because of the

repetitive nature of the coding sequence which specifies aprotein with runs of glycines andmany heptapeptide repeatsas described previously for SH1 VP2 [27] DV18 is a largereplacement that significantly alters the predicted length ofthe minor capsid protein VP18 in the two viruses (865519aa for SH1PH1 resp) and also provides SH1 with 3 ORFsthat are different or absent in PH1 ORFs 52 53 and 54 Forexample a homolog of SH1 ORF52 is not present in PH1 butis present in HHIV-2 (putative protein 40) While ORF48 ofPH1 shows no aa sequence homology to SH1 ORFs 53 and 54all of these predicted proteins carry CxxC motifs suggestiveof related functions such as DNA binding activity [36]

The genes of the PH1 genome are syntenic with those ofSH1 and are found in similarly oriented blocks of closelyspaced or overlapping ORFs that suggest that transcription isorganised into operons The programme of transcription ofthe SH1 genome has been reported previously [30] and thesequences in PH1 corresponding to the six promoter regions(P1ndashP6) determined in SH1 showed that five of these (P1ndashP5)are well conserved Only the region corresponding to SH1 P6was poorly conserved (data not shown) but this promoter isstrongly regulated in SH1 and only switches on late in infec-tion [30]

28 Structural Protein Genes The genes coding for the majorvirus structural proteins of PH1 (VP 1ndash4 7 9 10 and 12)were identified byMALDI-TOF (see above) and the genes forthese proteins are found in the same order and approximatepositions as in the SH1 genome (Figure 6(b) red ORFs)Genes coding for minor structural proteins can be deducedby sequence comparison with SH1 so that VP5 and VP6 arelikely to be encoded by PH1ORFs 25 and 29 respectively andVP13 and VP18 by ORFs 23 and 49 respectively This wouldgive PH1 a total of 11 structural proteins the same as SH1Of these VP7 is the most conserved between the two viru-ses (98 aa identity) followed by VP4 VP9 VP3 and VP12(91ndash94 identity) VP5 VP6 and VP13 (82ndash88 iden-tity) VP2 (77) VP1 (65) and lastly the least conservedprotein was VP18 (31)

29 Genomic Loci Related to PH1 and Other Halosphaeroviru-ses Clusters of halosphaerovirus-related genes are present inthe genomes of two recently sequenced haloarchaea Happaucihalophilus and Hbf lacisalsi (Figure 6(a)) Neither ofthese regions appears to represent complete virus genomebut they retain a number of genes that share a similar seq-uence and synteny with the virus genomes depicted belowthem (Figure 6(a)) Both loci show a mixed pattern ofrelatedness to SH1PH1 and HHIV-2 as indicated by ORFsof matching colour and gene name If these genomic locirepresent provirus integrants that have decayed over timethen the relationships they show suggest not only that halo-sphaeroviruses are a diverse virus group but also that asignificant level of recombination occurs between them lead-ing to mosaic gene combinations The two ORFs colouredgreen in theHap paucihalophilus locus (Figure 6) are relatedto ORFs found within or very close to genomic loci ofvirusplasmid genes found in other haloarchaea suggestingadditional links between (pro)viruses [20]

8 Archaea

Table4ORF

anno

tatio

nsof

theh

aloviru

sPH1g

enom

e

ORF

aPo

sitionb

Locustag

Leng

thc(aa)

pKi

Similaritycharacteris

tics

144

1ndash605

HhP

H1gp01

5043

76aa

similarityto

SH1O

RF2(YP271859)

2602ndash802

HhP

H1gp02

66109

80aa

similarityto

SH1O

RF3(YP271860)

3802ndash1059

HhP

H1gp03

8539

85aa

similarityto

SH1O

RF4(YP271861)44

sim

ilarityto

HHIV-2

protein1

41056ndash1283

HhP

H1gp04

7539

85aa

identityto

SH1O

RF5(YP271862)Predictedcoiled-coilregion

51276ndash1686

HhP

H1gp05

136

49

80aa

similarityto

SH1O

RF6(YP271863)Tw

otransm

embraned

omainsA

lsorelated

toHHIV-2

protein2(38

AFD

02283)

andto

Halobiform

alacisalsiORF

ZP09950018

(35

)6

1683ndash1829

HhP

H1gp06

4861

81aa

similarityto

SH1O

RF7(YP271864

)Predictedsig

nalsequence(Sign

alP)

71822ndash2130

HhP

H1gp07

102

58

91aa

similarityto

SH1O

RF8(YP271865)a

nd86to

HHIV-2

putativ

eprotein

3(A

FD02284)O

ther

homologsinclude

Nocardiaproteinpn

f2110(YP1220601)H

rrlacusprofund

iHlac0751

(YP0025654211)ORF

sinactin

ophage

VWB

(AAR2

9707)Frankia(EAN116

57)andgp40

ofMycobacteriu

mph

ageD

ori(AER

47690)

82123ndash2383

HhP

H1gp08

8642

90aa

similarityto

SH1O

RF9(YP271866)

92380ndash2714

HhP

H1gp09

111

52

88aa

similarityto

SH1O

RF11(YP271868)and

also

similarityto

putativ

eprotein

4of

HHIV-2

(AFD

02285)

HlacA

J19877of

Hbflacisa

lsiAJ

5(ZP09950016)

102861ndash300

4HhP

H1gp10

47104

Weaksim

ilarityto

SH1O

RF12

(YP271869)o

nlyover

theN

-term

inal18

resid

ues

113029ndash3100

HhP

H1gp11

23123

Not

presentinSH

1orH

HIV-2

123134ndash7453

HhP

H1gp12

1439

42

Capsid

protein

VP172sim

ilarityto

SH1V

P1(O

RF13Y

P271870)HHIV-2

VP1

(AFD

02286)H

lacA

J19872Hbflacisa

lsiAJ

5Predictedhelix-tu

rn-helixandRu

vA-like

domain(In

terproScan)

137505ndash8227

HhP

H1gp13

240

54

Predicted

P-loop

ATPa

sedo

main(C

OG0433)92aa

similarityto

SH1O

RF17

(YP271874)a

ndalso

similarityto

putativ

eAT

Pase

ofHHIV-2

(AFD

02288)A

TPaseo

fHaladapatus

paucihalophilusD

X253

(ZP0804

6180)HlacA

J19867of

Hbf

lacisalsi(ZP09950014)

148419ndash8228c

HhP

H1gp14

6351

43aa

similarityto

SH1O

RF18

(YP271875)Alanine-richCentraltransm

embraned

omain(Pho

bius)

158856ndash8

416c

HhP

H1gp15

146

41

92aa

similarityto

SH1O

RF19

(YP271876)a

ndalso

toHHIV-2

putativ

eprotein

9(A

FD02290)H

lacA

J19857of

Hbf

lacisalsiAJ

5ZO

D2009

19093of

Happau

cihalophilus

168853ndash9

500c

HhP

H1gp16

215

39

65aa

similarityto

SH1O

RF20

(YP271877)a

ndalso

topu

tativ

eprotein

11of

HHIV-2

(AFD

02292)Z

OD2009

19098of

Hap

paucihalophilusHlacA

J19852Hbflacisa

lsiAJ

5C-

term

inaltransm

embraned

omain(Pho

bius)

179500ndash9

919c

HhP

H1gp17

139

41

79aa

similarityto

SH1O

RF21

(YP271878)a

ndalso

topu

tativ

eprotein

12of

HHIV-2

(AFD

02293)hypotheticalprotein

HlacA

J19847of

Hbflacisa

lsi(ZP09950010)Transm

embraned

omain(Pho

bius)

189923ndash10594c

HhP

H1gp18

223

43

83aa

similarityto

SH1O

RF22

(YP271879)a

ndalso

topu

tativ

eprotein

13of

HHIV-2

(AFD

02294)Z

OD2009

19108of

Hap

paucihalophilusHlacA

J19842of

Hbflacisa

lsiP

redicted

signalsequence(sig

nalP)Con

tains4

CxxC

motifs

1910659ndash10943

HhP

H1gp19

94104

Capsid

protein

VP12

(ORF

19)91aa

similarityto

SH1V

P12(O

RF23Y

P271880)a

ndalso

toVP12of

HHIV-2

(AFD

02295)

andHlacA

J19837of

Hbflacisa

lsiZ

OD2009

19113of

Happau

cihalophilusTw

otransm

embraned

omains

(Pho

bius)

2010960ndash

11517

HhP

H1gp20

185

44

Capsid

protein

VP7(O

RF20)98aa

similarityto

SH1V

P7(O

RF24Y

P271881)a

ndalso

toVP7

ofHHIV-2

(AFD

02296)

ZOD2009

19118

ofHappau

cihalophilusHlacA

J19832of

Hbflacisa

lsiAJ

5

2111519ndash

12217

HhP

H1gp21

232

41

Capsid

protein

VP4(O

RF21)94aa

similarityto

SH1O

RF25

(YP271882)a

ndalso

toVP4

ofHHIV-2

(AFD

02297)

ZOD2009

19123of

Happau

cihalophilusHlacA

J19827of

Hbflacisa

lsiAJ

522

12233ndash12454

HhP

H1gp22

7339

81aa

similarityto

SH1O

RF26

(YP271883)a

ndalso

topu

tativ

eprotein

17of

HHIV-2

(AFD

02298)

2312458ndash12697

HhP

H1gp23

7948

78aa

similarityto

SH1capsid

protein

VP13

(ORF

27Y

P271884)a

ndalso

toVP13of

HHIV-2

(AFD

02299)P

redicted

coil-coildo

main

PredictedC-

term

inaltransm

embraned

omain(Pho

bius)

Archaea 9

Table4Con

tinued

ORF

aPo

sitionb

Locustag

Leng

thc(aa)

pKi

Similaritycharacteris

tics

2412701ndash1506

4HhP

H1gp24

787

39

Capsid

protein

VP2(O

RF24)75aa

similarityto

SH1V

P2(O

RF28Y

P271885)a

ndto

VP2

ofHHIV-2

(AFD

02300)

ZOD2009

19133of

Happau

cihalophilus(ZP

0804

6189)

2515065ndash15961

HhP

H1gp25

298

45

Putativec

apsid

protein

VP5(O

RF25)81aa

similarityto

SH1V

P5(O

RF29Y

P271886)a

ndto

HHIV-2

VP5

(AFD

02301)

HlacA

J19807of

HbflacisalsiAJ

5ZO

D2009

19133of

Happau

cihalophilus

261596

4ndash1644

6HhP

H1gp26

160

46

73aa

similarityto

SH1capsid

protein

VP10

(ORF

30Y

P271887)a

ndto

VP10of

HHIV-2

(AFD

02302)Z

OD2009

19138of

Happau

cihalophilusHlacA

J19802of

Hbflacisa

lsiP

redicted

C-term

inaltransm

embraned

omain(TMHMM)

271644

6ndash16895

HhP

H1gp27

149

42

Capsid

protein

VP9(O

RF27)94aa

similarityto

SH1V

P9(O

RF31Y

P271888)a

ndto

ZOD2009

19143of

Hap

paucihalophilus(ZP

0804

6191)

2816908ndash17921

HhP

H1gp28

337

43

Capsid

protein

VP3(O

RF28)91aa

similarityto

SH1V

P3(O

RF32Y

P271889)a

ndto

NJ7G

2365

ofNa

trinemaspJ7-2

2917928ndash18617

HhP

H1gp29

229

42

Capsidprotein

VP6(O

RF29)83aa

similarityto

SH1V

P6(O

RF33Y

P271890)Also

toNJ7G

3194

ofNa

trinemaspJ7-2and

Hmuk

0476

ofHalom

icrobium

mukohataeiPredictedcarboxypeptid

aser

egulatory-lik

edom

ain(C

arbo

xypepD

reg

pfam

13620)

3018614ndash

18943

HhP

H1gp30

109

49

71aa

similarityto

SH1O

RF34

(YP271891)a

ndto

putativ

eprotein

27of

HHIV-2

(AFD

02308)P

redicted

signalsequence

(signalP)

3119054ndash

19176c

HhP

H1gp31

4050

TwoCx

xCmotifsN

oho

molog

inSH

1orH

HIV-2

3219173ndash19289c

HhP

H1gp32

3841

68aa

similarityto

SH1O

RF37

(YP271894)

3319286ndash

19552c

HhP

H1gp33

8853

58aa

similarityto

SH1O

RF39

(YP271896)a

ndto

HHIV-2

putativ

eprotein

29(A

FD02310)Fou

rCxxCmotifs

3419549ndash

19731c

HhP

H1gp34

6070

Con

tainsC

xxCmotif

Noho

molog

inSH

1orH

HIV-2

3519728ndash20219c

HhP

H1gp35

163

56

93aa

similarityto

SH1O

RF41

(YP271898)a

ndto

putativ

eprotein

30of

HHIV-2

(AFD

02311)

3620216ndash

21415c

HhP

H1gp36

399

50

95aa

similarityto

SH1O

RF42

(YP271899)a

ndto

putativ

eprotein

31of

HHIV-2

(AFD

02312)

3721419ndash

21778c

HhP

H1gp37

11941

91aa

similarityto

SH1O

RF43

(YP271900)a

ndto

putativ

eprotein

32of

HHIV-2

(AFD

02313)

3821762ndash2300

6cHhP

H1gp38

414

42

83aa

similarityto

SH1O

RF44

(YP271901)a

ndto

putativ

eprotein

33of

HHIV-2

(AFD

02314)P

redicted

coil-coiland

helix-tu

rn-helixdo

mains

3923003ndash23158c

HhP

H1gp39

5165

69aa

similarityto

SH1O

RF45

(YP271902)

4023161ndash23370c

HhP

H1gp40

6935

74aa

similarityto

SH1O

RF46

(YP271903)and55sim

ilarityto

putativ

eprotein

35of

HHIV-2

(AFD

02316)

4123422ndash23586c

HhP

H1gp41

5476

86aa

similarityto

SH1O

RF47

(YP271904)Con

tains2

CxxC

motifsand

show

ssim

ilarityto

proteindo

mainfamily

PF14206Arginine-ric

h

4223583ndash24023c

HhP

H1gp42

48

77aa

similarityto

SH1O

RF48

(YP271905)a

ndto

putativ

eprotein

36of

HHIV-2

(AFD

02317)H

ham114540of

Hcc

hamelinensis(ZP11271999)Ph

iCh1p72of

Natrialba

phageP

hiCh

1(NP665989)DUF4

326(pfam14216)

family

domain

4324020ndash

24538c

HhP

H1gp43

172

47

79aa

similarityto

SH1O

RF49

(YP271906)and56sim

ilarityto

putativ

eprotein

37of

HHIV-2

(AFD

02318)

4424535ndash25053c

HhP

H1gp44

172

43

92aa

similarityto

SH1O

RF50

(YP271907)and72sim

ilarityto

putativ

eprotein

38of

HHIV-2

(AFD

02319)

4525050ndash

25277c

HhP

H1gp45

7543

Noho

molog

inSH

1orH

HIV-2

4625274ndash

25387c

HhP

H1gp46

3748

Noho

molog

inSH

1orH

HIV-2

4725533ndash25904

HhP

H1gp47

123

43

61aa

similarityto

SH1O

RF51

(YP271908)66sim

ilarityto

putativ

eprotein

39of

HHIV-2

(AFD

02320)P

redicted

COG1342

domain(D

NAbind

inghelix-tu

rn-helix)

4825901ndash26092c

HhP

H1gp48

6349

Con

tainsC

xxCmotif

Noho

molog

inSH

1orH

HIV-2

4926089ndash

2764

8cHhP

H1gp49

506

40

81aa

similarityto

SH1O

RF55

(YP271912)(bu

tonlyin

theN

-term

inalhalf)H

omolog

ofvirusstru

cturalprotein

VP18

ofHHIV-2

(AFD

02323)

a ORF

swerep

redicted

either

byGLIMMER

orby

manualsearching

forh

omologsintheG

enBa

nkdatabase

b Startandendpo

sitions

ofORF

sare

give

inbp

numbera

ccording

totheP

H1sequenced

epositedatGenBa

nk(KC2

52997)O

RFso

nthec

omplem

entary

strandared

enoted

bythes

uffixc

c Lengthof

thep

redicted

ORF

innu

mbero

faminoacids

10 Archaea

Table 5 Main differences (divergent regions) between the SH1 and PH1 genomes

RegionaSH1 startb SH1 stop Length (bp) PH1 start PH1 stop Length (bp) Comment

DV1 527 650 123 538 604 66

Replacement Both regions have direct repeats at their bordersPH1 has two sets GACCCGGC and CGCTGC while SH1 hasCCCGAC In SH1 this replacement region includes theC-terminal region of ORF2 and the N-terminal region ofORF3 In PH1 it covers only the C-terminal region of ORF1

DV2 2433 2588 155 2386 2387 mdash RMDc from PH1 SH1 repeat at border is TGACCGThisremoves the homolog of SH1 ORF10 from PH1

DV3 3324 3598 274 3137 3416 279 Replacement at the beginning of SH1 ORF13PH1 ORF12(capsid protein VP1)

DV4 6656 7128 472 6478 7085 607 Replacement near the C-terminus of SH1 ORF13PH1 ORF12(capsid protein VP1)

DV5 7491 8341 850 7446 7447 mdash

Indel that results in ORFs14-16 of SH1 not being present inPH1 SH1 ORF14 has a close homolog in HHIV-2 (ORF6) in asimilar position just after the VP1 homolog SH1 ORF15 is aconserved protein in haloarchaea (eg NP 2552A andHmuk 2978) and halovirus His1 (ORF13) Possible invertedrepeat (AGCCATG) at border of SH1 region

DV6 13705 13706 mdash 12818 12851 33RMD from SH1 PH1 repeat at border is CAGCGG(gt)G Thisremoves a part of capsid protein VP2 sequence from the SH1protein

DV7 13789 13863 74 12934 12941 7 Replacement This occurs within VP2 gene of both viruses

DV8 15142 15186 44 14220 14221 mdash RMD from PH1 SH1 repeat at border is TG(tc)CCGACGAand occurs within capsid protein VP2 gene of both viruses

DV9 15303 15317 14 14337 14338 mdash RMD from PH1 SH1 repeat at border is GCCGACGA andoccurs within capsid protein VP2 gene of both viruses

DV10 15504 15527 23 14529 14530 mdash RMD from PH1 SH1 repeat at border is GACGA and occurswithin capsid protein VP2 of both viruses

DV11 20026 20405 379 19019 19173 154Replacement beginning at the start of SH1 ORF35PH1 ORF31This region is longer in SH1 where it includes ORF36 an ORFthat has no homolog present in PH1

DV12 20440 20661 221 19208 19286 78 Replacement This is unequal and includes an ORF in SH1(ORF38) that is not present in PH1

DV13 20943 21085 142 19562 19728 166 Replacement This covers SH1 ORF40PH1 ORF34 Thepredicted proteins are not homologous

DV14 23541 23542 mdash 22171 22197 27 Probable RMD in SH1 PH1 repeat at border is CGTCTCGGand occurs in SH1 ORF44PH1 ORF38

DV15 24506 24521 15 23152 23153 mdash Probable RMD in PH1 SH1 repeat at border is CTCGGT andoccurs near the end of SH1 ORF45PH1 ORF39

DV16 24951 24964 13 23579 23580 mdash Probable RMD in PH1 SH1 repeat at border is CGGTC andoccurs in SH1 ORF47PH1 ORF41

DV17 26449 26450 mdash 25050 25274 224Probable RMD in SH1 PH1 repeat at border is TCATGCG andoccurs near the start of SH1 ORF50PH1 ORF44 and providesan extra ORF for PH1 (ORF45)

DV18 27074 29764 690 25895 26933 1038

Replacement Left border at start of SH1 ORF51PH1 ORF47and extends rightwards into SH1 ORF55PH1 ORF49 (VP18gene) It is an unequal replacement and SH1 has two moreORFs in this region than PH1

aDV Divergent regions between the genomes of SH1 and PH1bStart and stop positions refer to the GenBank sequences of the two viruses SH1 NC 0072171 PH1 KC252997cRMD repeat-mediated deletion event as described in [20]

The recently described halovirus SNJ1 carries manyORFsthat have homologs adjacent to a previously described ViPREofHmc mukohataei (from here on denoted by ViPREHmuk1)a locus that contains genes related to His2 (and other ple-olipoviruses) as well as toHaloferax plasmid pHK2 (probablya provirus) and to two small plasmids ofHqr walsbyi (sim6 kbpL6A and pL6B)The latter relationship provides wider linksto other viruses because one of the PL6 genes (Hqrw 6002)

has homologs in some pleolipoviruses (HPRV-3 and HGPV-1) while another gene (Hqrw 6005) is related to ORF16 ofthe spindle-shaped halovirus His1 [15] After adding the SNJ1gene homologs to ViPREHmuk1 it is extended significantlyand a close inspection of the flanking regions revealed atRNA-ala gene at one end and a partial copy of the sametRNA-ala gene (next to a phage integrase gene) at the otherend (supplementary Figure 1) Genes beyond these points

Archaea 11

Table 6 Transfection of haloarchaea by PH1 DNA

Speciesa Efficiency of transfectiontransformation withPH1 DNAb (PFU120583g of DNA) pUBP2 DNAbc (CFU120583g of DNA)

Har hispanica 53 plusmn 05 times 103 14 plusmn 08 times 104

Har marismortui 45 plusmn 07 times 103 18 plusmn 01 times 104

ldquoHar sinaiiensisrdquo 32 plusmn 06 times 102 mdashHalobacterium salinarum mdash 54 plusmn 42 times 103

Haloferax gibbonsii 28 plusmn 08 times 103 40 plusmn 26 times 103

Haloferax lucentense mdash 57 plusmn 11 times 103

Hfx volcanii mdash 21 plusmn 07 times 105

Hrr lacusprofundi 70 plusmn 22 times 102 95 plusmn 15 times 103

Haloterrigena turkmenica 16 plusmn 17 times 102 24 plusmn 06 times 103

Natrialba asiatica 35 plusmn 39 times 102 31 plusmn 10 times 104aOnly those species positive for transfection andor plasmid transformation are shownbRates of transfection by (nonprotease treated) PH1 DNA or transformation by plasmid pUB2 are averages of three independent experiments each performedin duplicate (plusmn standard deviation)cTransformants were selected on plates with 2 4 or 6120583gmL simvastatin depending on the strainmdash no plaques or colonies observed

appear to be related to cellular metabolism ViPREHmuk1is now 39379 nt in length flanked by a 59 nt direct repeat(potential att sites) and includes 54 ORFs many of which arerelated to known haloviruses are virus-like (eg integrasesmethyltransferases transcriptional regulators DNA methyl-transferase and DNA glycosylase) are homologs in or nearother knownViPREs or show little similarity to other knownproteins This locus also includes an ORC1CDC6 homolog(Hmuk 0446) which could provide a replication function

The mixture of very different virus and plasmid genesseen within ViPREHmuk1 is remarkable but it also revealshomologs found in or adjacent to genomic loci of otherspecies such as the gene homologs of Har marismortuiseen in the left end of ViPREHmuk1 (supplementary Figure1) When these genes are added to the previously describedHaloarcula locus (ViPREHmuk1) it is also extended signifi-cantly It becomes 25550 bp in length encompasses genesfrom Hmar 2382 to Hmar 2404 and is flanked by a fulltRNA-ala gene at one end and a partial copy at the other Itcarries an ORC1CDC6 homolog an integrase and a phiH-like repressor as well as pleolipovirus homologs

210 Transfection of Haloarchaea by PH1 DNA PH1 DNAwas introduced into Har hispanica cells using the PEGmethod [37] and the cells were screened for virus productionby plaque assay Figure 7 shows the increase in transfectedcells with input (nonproteinase K-treated) virus DNA Theestimated efficiencywas 53times 103 PFUper120583gDNA PH1DNAthat had been treated with proteinase K or with DNAase I(RNase-free) did not produce plaques (data not shown) PH1DNA was found to transfect six species of haloarchaea otherthanHar hispanica includingmembers of the generaHaloar-cula Haloferax Halorubrum Haloterrigena and Natrialba(Table 6) For these experiments the virus production fromtransfected cells was detected by plaque assay on indicatorlawns of Har hispanica

0

0

200 400 600 800 1000

Titre

(PFU

mL)

104

103

102

PH1 DNA (ng)

Figure 7 Transfection ofHar hispanica cells with PH1 DNA Vary-ing amounts of nonproteinase K-treated PH1 DNAwere introducedinto cells of Har hispanica using the PEG method [64] Cells werethen screened for infective centres by plaque assay Data shown is theaverage of three independent experiments performed in duplicateError bars represent one standard deviation of themean If protease-treated DNA was used no transfectant plaques were observed

3 Discussion

PH1 is very similar in particlemorphology genome structureand sequence to the previously described halovirus SH1 [816] Like SH1 it has long inverted terminal repeat sequencesand terminal proteins indicative of protein-primed replica-tion Purified PH1 particles have a relatively low buoyant

12 Archaea

density and are chloroform sensitive and show a layeredcapsid structure consistent with the likely presence of aninternalmembrane layer as has been shown for SH1HHIV-2and SNJ1 [13 26]The particle stability of PH1 to temperaturepH and reduced salt was similar to SH1 [8] The structuralproteins of PH1 are very similar in sequence and relativeabundance to those of SH1 so the two viruses are likely toshare the same particle geometry [16]

Viruses that use protein-primed replication such as bac-teriophageΦ29 carry a viral type BDNApolymerase that caninteract specifically with the proteins attached to the genomictermini and initiate strand synthesis [38] ArchaeovirusesHis1 [7] and Acidianus bottle-shaped virus [39] probablyuse this mode of replication However a polymerase genecannot be found in the genomes of PH1 SH1 or HHIV-2DNA polymerases are large enzymes and the only ORF ofSH1 that was long enough to encode such an enzyme andhad not previously been assigned as a structural protein wasORF55 which specified an 865 aa protein that contained noconserved domains indicative of polymerases [27]The recentstudy of HHIV-2 showed that the corresponding ORF in thisvirus (gene 42) is a structural protein of the virion and byinference this is likely also for the corresponding proteinsof SH1 and PH1 (no homolog is present in SNJ1) Withouta polymerase gene these viruses must use a host enzymebut searches for a viral type B DNA polymerase in thegenome of the hostHar hispanica or in the genomes of othersequenced haloarchaea did not find anymatchesThis arguesstrongly for a replication mechanism that is different to thatexemplified byΦ29 An attractive alternative is that displayedby Streptomyces linear plasmids which have 51015840-terminal pro-teins and use a cellular polymerase for replication [40] Theterminal proteins are not used for primary replication but forend patching [41] and it has been shown that if the term-inal repeat sequences are removed from these plasmids andthe ends ligated they can replicate as circular plasmids [42]This provides a testable hypothesis for the replication of halo-sphaeroviruses and also offers a pathway for switchingbetween linear (eg PH1) and circular (eg SNJ1) forms Useof a host polymerase would also fit with the ability of SH1 andPH1 DNA to transfect many different haloarchaeal species[31] as these enzymes and theirmode of action are highly con-served

The growth characteristics of PH1 inHar hispanica bothby single-cell burst and single-step growth experiments gavevalues of 50ndash100 virusescell significantly less than that ofSH1 (200 virusescell) [8] and HHIV-2 (180 virusescell) [13]The latter virus is lytic whereas SH1 PH1 and SNJ1 start pro-ducing extracellular viruswell before any decrease in cell den-sity indicating that virus release can occur without lysisThishas been most clearly shown with SH1 where almost 100of cells can be infected [8] Infection ofHar hispanica by PH1was less efficient (sim30 of cells) but the kinetics of virus pro-duction in single-step growth curves of SH1 and PH1 aresimilar with virus production occurring over several hoursrather than at a clearly defined time after infection The twoviruses are also very closely related and infect the same hostspecies so it is likely that they use the same scheme for cellexit For both viruses the cell density eventually decreases in

single-step cultures showing that virus infection does resultin cell death This mode of exit appears to maximise theproduction of the virus over time as the host cells survive foran extended period In natural hypersaline waters cell num-bers are usually high but growth rates are low [12] and thehigh incident UV (on often shallow ponds) would damagevirus DNA so reducing the half-life of released virus part-icles [43] These factors may have favoured the exit strategydisplayed by these viruses improving their chance of trans-mission to a new host

The motif GATC was absent in the genomes of PH1 SH1andHHIV-2 and themotif CTAGwas either absent or greatlyunderrepresented All three viruses share the same host butGATC and CTAG motifs are plentiful in the Har hispanicagenome (accessions CP002921 and CP002923) Avoidance ofthesemotifs in halovirusesHis1 His2 HF1 andHF2 has beenreported previously [7] Such purifying selection commonlyresults from host restriction enzymes and inHfx volcanii itis exactly these two motifs that are targeted [44] One Hfxvolcanii enzyme recognises A-methylated GATC sites [45]and another recognises un-methylated CTAG sites (whichare protected by methylation in the genome) In the currentstudy this could explain the negative transfection results forHfx volcanii as the PH1 genome contains two CTAGmotifsBy comparison the SH1 genome contains no CTAG motifsand is able to transfect Hfx volcanii [31]

Comparison of the PH1 and SH1 genomes allowed adetailed picture of the natural variation occurring betweenclosely related viruses revealing likely deletion events medi-ated by small repeats a process described previously in astudy of Hqr walsbyi and termed repeat-mediated deletion[20] There are also many replacements including blocks ofsequence that occur within long open reading frames (suchas the capsid protein genes VP1 and VP18) VP3 and VP6 areknown to form the large spikes on the external surface of theSH1 virion and presumably one or both interact with host cellreceptors [16] PH1 and SH1 have identical host ranges con-sistent with the high sequence similarity shown between theircorresponding VP3 and VP6 proteins

A previous study of SH1 could not detect transcriptsacross annotated ORFs 1ndash3 or 55-56 [30] ORFs 1 and 56occur in the terminal inverted repeat sequence and in thepresent study it was found that there were no ORFs cor-responding to these in the PH1 genome Given the close simi-larity of the two genomes the comparative data are in agree-ment with the transcriptional data indicating that these SH1ORFs are not used SH1 ORFs 2 and 3 are more problematicas good homologs of these are also present in PH1 The con-flict between the transcriptional data and the comparativegenomic evidence requires further experimental work toresolveThe status of SH1ORF55 has recently been confirmedby studies of the related virus HHIV-2 (discussed above)

The pleolipovirus group has expanded dramatically inthe last few years and it now comprises a diverse group ofviruseswith different genome types and replication strategiesWhat is evenmore remarkable is that a similar expansion cannow be seen with SH1-related viruses which includes viruseswith at least two replication modes and genome types (linearand circular) Evidence from haloarchaeal genome sequences

Archaea 13

show cases of not only provirus integrants or plasmids (egpKH2 and pHH205) but also genomic loci (ViPREs) thatcontain cassettes of virus genes from different sources andat least two cases (described in this study) have likely attsites that indicate circularisation and mobility While thereis a clear relationship between viruses with circular ds- andssDNA genomes that replicate via the rolling circle method(ie the ssDNA form is a replication intermediate) it ismore difficult to explain how capsid gene cassettes can movebetween these viruses and those with linear genomes andterminal proteins Within the pleolipoviruses His2 has alinear genome that contains a viral type B DNA polymeraseand terminal proteins while all the other described membershave circular genomes that contain a rep homolog and prob-ably replicate via the rolling circle method [14 19] Similarlyamong the halosphaeroviruses PH1 SH1 and HHIV-2 havelinear dsDNA with terminal proteins and probably replicatein the same way but the related SNJ1 virus has a circulardsDNA genome The clear relationships shown by the capsidgenes of viruses within each group plus the connectionsshown to haloviruses outside each group (eg the DNA poly-merases of His1 and His2) all speak of a vigorous meansof recombination one that can readily switch capsid genesbetween viruses with radically different replication strategiesHow is the process most likely to operate One possibility issuggested by ViPREs which appear to be mobile collectionsof capsid and replication genes from different sources Theyoffer fixed locations for recombination to occur providegene cassettes that can be reassorted to produce novel virusgenomes and some can probably recombine out as circularforms For example ViPREHmuk1 carries genes related topleolipoviruses halosphaeroviruses and other haloviruses Itis yet to be determined if any of the genes carried in theseloci are expressed in the cell (such as the genes for capsidproteins) or if ViPREs provide any selective advantage to thehost

4 Materials and Methods

41 Water Sample A water sample was collected in 1998from Pink Lake (32∘001015840 S and 115∘301015840 E) a hypersalinelake on Rottnest Island Western Australia Australia It wasscreened in the same year for haloviruses using the methodsdescribed in [8] A single plaque on a Har hispanica lawnplate was picked and replaque purified The novel haloviruswas designated PH1

42 Media Strains and Plasmids The media used in thisstudy are described in the online resource The Halohand-book (httpwwwhaloarchaeacomresourceshalohandbookindexhtml) Artificial salt water containing 30 (wv) totalsalts comprised of 4M NaCl 150mM MgCl

2 150mM

MgSO4 90mM KCl and 35mM CaCl

2and adjusted to pH

75 using sim2mL 1M Tris-HCl (pH 75) per litre MGM con-taining 12 18 or 23 (wv) total salts and HVD (halo-virus diluent) were prepared from the concentrated stock aspreviously described [46] Bacto-agar (Difco Laboratories)was added to MGM for solid (15 gL) or top-layer (7 gL)media

Table 1 lists the haloarchaea used in this study Allhaloarchaea were grown aerobically at 37∘C in either 18 or23 (wv)MGM(depending on the strain) andwith agitation(except for ldquoHar sinaiiensisrdquo)The plasmids used were pBlue-script II KS+ (Stratagene Cloning Systems) pUBP2 [47]and pWL102 [48] pUBP2 and pWL102 were first passagedthrough E coli JM110 [49] to prevent dam-methylationof DNA which has been shown to reduce transformationefficiency in some strains [45]

43 Negative-Stain Electron Microscopy The method fornegative-stain TEM was adapted from that of V Tarasovdescribed in the online resource The Halohandbook (httpwwwhaloarchaeacomresourceshalohandbookindexhtml)A 20120583L drop of the sample was placed on a clean surfaceand the virus particles were allowed to adsorb to a Formvarfilm 400-mesh copper grid (ProSciTech) for 15ndash2min Theywere then negatively stained with a 20120583L drop of 2 (wv)uranyl acetate for 1-2min Excess liquid was absorbed withfilter paper and the grid was allowed to air dry Grids wereexamined either on a Philips CM 120 BioTwin transmissionelectron microscope (Royal Philips Electronics) operating atan accelerating voltage of 120 kV or on a Siemens Elmiskop102 transmission electron microscope (Siemens AG) operat-ing at an accelerating voltage of 120 kV

44 Virus Host Range Twelve haloarchaeal strains from thegenera Haloarcula Halobacterium Haloferax HalorubrumandNatrialba (Table 1)Thirteen naturalHalorubrum isolates(H Camakaris unpublished data) and five uncharacterizedhaloarchaeal isolates fromLakeHardy Pink Lake (inWesternAustralia) and Serpentine Lake (D Walker and M K Seahunpublished data) were screened for PH1 susceptibilityLysates from PH1-infected Har hispanica cultures (1 times1011 PFUmL) were spotted onto lawns of each strain (usingmedia with 12 or 18 (wv)MGM depending on the strain)and incubated for 2ndash5 days at 30 and at 37∘C

45 Large Scale Virus Growth and Purification Liquid cul-tures of PH1 were grown by the infection (MOI 005) of anearly exponential Har hispanica culture in 18 (wv) MGMCultures were incubated aerobically at 37∘C with agitationfor 3 days Clearing (ie complete cell lysis) of PH1-infectedcultures did not occur and consequently cultures were harv-ested when the absorbance at 550 nm reached the minimumand the titre (determined by plaque assay) was at the maxi-mum usually sim1010ndash1011 PFUmL Virus was purified usingthe method described previously for SH1 [8 31] except thatafter the initial low speed spin (Sorvall GSA 6000 rpm30min 10∘C) virus was concentrated from the infected cul-ture by centrifugation at 26000 rpm (13 hr 10∘C) onto acushion of 30 (wv) sucrose in HVDThe pellet was resus-pended in a small volume of HVD and loaded onto a pre-formed linear 5ndash70 (wv) sucrose gradient followed byisopycnic centrifugation in 13 gmL CsCl (Beckman 70Ti60000 rpm 20 hr 10∘C) The white virus band occurredat a density of 129 gmL and was collected and diluted inhalovirus diluent (HVD see Section 4) and the virus pelleted(Beckman SW55 35000 rpm 75min 10∘C) and resuspended

14 Archaea

in a small volume of HVD and stored at 4∘C Virus recoveryat the major stages of a typical purification is given insupplementary Table 1 The specific infectivity of pure virussolutions was determined as the ratio of the PFUmL to theabsorbance at 260 nm

46 PH1 Single-Step Growth Curve An early exponentialphase culture of Har hispanica grown in 18 (wv) MGMwas infected with PH1 (MOI 50) Under these conditions thepercentage of infected cells was approximately 30 After anadsorption period of 1 hr at 37∘C the cells were washed threetimes with 18 (wv) MGM (at room temperature) resus-pended in 100mL 18 (wv) MGM (these methods ensuredthe removal of all residual virus) and incubated at 37∘Cwith shaking (100 rpm) Samples were removed at hourlyintervals for measurements of absorbance at 550 nm andthe number of infective centres Immediately after samplingtitres were determined by plaque assay with an indicator lawnof Har hispanica Each experiment was performed in tripli-cate

47 Halovirus Stability After various treatments sampleswere removed and diluted in HVD and virus titres weredetermined by plaque assay on Har hispanica Each experi-ment was performed in triplicate and representative data areshown Chloroform sensitivity was examined by the exposureof PH1 lysates in 18 (wv) MGM to chloroform in a volumeratio of 1 4 (chloroform to lysate) Incubation was at roomtemperature with constant agitation At appropriate timepoints the mix was allowed to settle and a sample wasremoved from the upper layer and diluted in HVDThe effectof a low ionic environment was examined by diluting PH1lysates (in 18 (wv) MGM) into double distilled H

2O in a

volume ratio of 1 1000 (lysate to double distilledH2O) Incu-

bation was at room temperature with constant agitationSamples were removed at various times and diluted in HVDThe pH stability of PH1 was determined by dilution of PH1lysates in 18 (wv)MGMin the appropriate pHbuffer (HVDbuffered with appropriate Tris-HCl) in a volume ratio of1 100 (lysate to buffer) Incubation was at room temperaturewith constant agitation After 30min samples were removedand diluted in HVD Thermal stability of PH1 was examinedby a 1 hr incubation of a virus lysate in 18 (wv) MGM atdifferent temperatures after which they were brought quicklyto room temperature diluted in HVD and titrated

48 Protein Procedures To remove salts purified virus prepa-rations were mixed with trichloroacetic acid (10 (vv) finalconcentration) and incubated on ice for 15min to allowthe proteins to precipitate After centrifugation (16000 g15min room temperature) the precipitate was washed threetimes in acetone dried and resuspended in double distilledH2O Proteins were dissolved in Laemmli sample buffer with

48mM 120573-mercaptoethanol [50] heated in boiling water for5min then separated on 12 (wv) NuPAGE Novex Bis-Tris Gels using MES-SDS running buffer according to themanufacturerrsquos directions (Invitrogen) After electrophoresisgels were rinsed in double distilled H

2O and stained with

01 (wv) Brilliant Blue G in 40 (vv) methanol and

10 (vv) acetic acid Gels were destained with severalchanges of 40 (vv) methanol and 10 (vv) acetic acidAlternatively gels were stained with GelCode GlycoproteinStaining Kit Pro-Q Emerald 488 Glycoprotein Gel and BlotStain Kit SYPRO Ruby Protein Gel Stain according tothe manufacturerrsquos directions (Invitrogen Molecular ProbesPierce Biotechnology)

Protein bands were cut from the gels and sent to theAustralian Proteome Analysis Facility (Macquarie Univer-sity) for trypsin digestion and analysis by matrix assistedlaser desorption ionisation time of flight mass spectrometry(MALDI-TOF MS) on an Applied Biosystems 4700 Pro-teomics Analyser (Applied Biosystems)

49 DNA Procedures Proteinase K-treated and nonpro-teinaseK-treatedDNApreparationsweremade frompurifiedPH1 using the methods described previously for SH1 [8]DNA was separated on 1 (wv) agarose gels in Tris-acetate-EDTA electrophoresis buffer and was stained with ethidiumbromide (Sigma-Aldrich)120582 DNA DNase I (RNase-free) exonuclease III mung

bean nuclease nuclease BAL-31 T4 DNA polymerase T4DNA ligase T7 exonuclease and type II restriction endonu-cleases were purchased from New England Biolabs RNase Aand yeast transfer RNA were purchased from Sigma-AldrichProteinase K was purchased from Promega Oligonucleotideprimers were purchased from Geneworks Australia

To clone fragments of the PH1 genome purified virusDNA was digested either with AccI Eco0109I and MseI orSmaI restriction endonucleases blunted with DNA Poly-merase I Large (Klenow) Fragment where appropriate andligated into AccI- Eco0109I- or SmaI-digested pBluescript IIKS+ using T4 DNA ligase according to the manufacturerrsquosinstructions (New England Biolabs) The DNA was intro-duced intoE coliXL1-Blue and transformants grown onLuriaagar containing 15 120583gmL tetracycline and 100 120583gmL ampi-cillinThe resulting cloneswere sequenced and the sequenceswere used to design specific oligonucleotide primers for PCRamplification andor primer walking using the virus genome(or specific restriction fragments) as templates

To amplify PH1 nucleic acid approximately 10 ng virusDNA was combined with 500 nM primers 100120583M of eachdNTP 1U Deep Vent DNA Polymerase (New England Bio-labs) and 1 timesThermoPol buffer (New England Biolabs) ina total reaction volume of 50 120583L Template was denatured at95∘C for 10min followed by 30 cycles of denaturation at 95∘Cfor 30 sec annealing at 56∘C for 30 sec extension at 75∘C for2min and a final extension at 75∘C for 10min PCR reactionswere performed on a PxE02ThermoCycler (Thermo ElectroCorporation) Sequencing reactions using 32 pmol primerwere performed by the dideoxy chain termination methodusing ABI PRISM Big Dye Terminator Mix version 31 onan ABI 3100 capillary sequencer (Applied Biosystems) bythe Applied Genetic Diagnostics Sequencing Service at theDepartment of Pathology (The University of Melbourne)

To obtain terminal genomic fragments PH1 DNA wasdigested by the enzyme Pas1 and the head (sim700 bp) andthe tail (sim1300 bp) fragments were purified by the QIAEXII gel extraction kits and treated with 05M piperidine for

Archaea 15

2 hr at 37∘C to remove protein residues The piperidine-treated fragments were sequenced by the Genomics BiotechCompany (Taipei Taiwan) using two primers TGACCA-ATTAATTAGGCCGGTTCGCC (PH1 head-R) and GTGC-CATACTGCTACAATTCT (PH1 tail-F)

Four hundred fifty-four whole genome sequencing PH1DNA samples (sim5 120583g) were sequenced using parallel pyrose-quencing on a Roche 454 Genome Sequencer System atMission Biotech (Taipei Taiwan) The largest contig was27399with 8803 reads (Newbler version 27) with 27387 posi-tions being of Q40 quality The average depth of coverage foreach base was 829TheGenBank accession for the entire PH1sequence is KC252997

Acknowledgments

The authors thank Carolyn Bath Mei Kwei (Joan) Seahand Danielle Walker for their early work in PH1 characteri-zation Simon Crawford Jocelyn Carpenter and Anna Fried-huber for their assistance with the transmission electronmicroscopy and Tiffany Cowie and Voula Kanellakis fortheir sequencing assistance This paper has been facilitatedby access to the Australian Proteome Analysis Facility estab-lished under the Australian Governmentrsquos Major NationalResearch Facilities program SLT was supported by a grantfrom the Biodiversity Research Center Academia SinicaTaiwan

References

[1] M Krupovic M F White P Forterre and D PrangishvilildquoPostcards from the edge structural genomics of archaealvirusesrdquo Advances in Virus Research vol 82 pp 33ndash62 2012

[2] P Stolt andW Zillig ldquoGene regulation in halophage 120601Hmdashmorethan promotersrdquo Systematic and Applied Microbiology vol 16no 4 pp 591ndash596 1994

[3] R Klein U Baranyi N Rossler B Greineder H Scholz and AWitte ldquoNatrialba magadii virus 120593Ch1 first complete nucleotidesequence and functional organization of a virus infecting ahaloalkaliphilic archaeonrdquo Molecular Microbiology vol 45 no3 pp 851ndash863 2002

[4] E Pagaling R D Haigh W D Grant et al ldquoSequence analysisof an Archaeal virus isolated from a hypersaline lake in InnerMongolia Chinardquo BMC Genomics vol 8 article 410 2007

[5] S L Tang SNuttall andMDyall-Smith ldquoHalovirusesHF1 andHF2 evidence for a recent and large recombination eventrdquo Jour-nal of Bacteriology vol 186 no 9 pp 2810ndash2817 2004

[6] C Bath and M L Dyall-smith ldquoHis1 an archaeal virus of theFuselloviridae family that infectsHaloarcula hispanicardquo Journalof Virology vol 72 no 11 pp 9392ndash9395 1998

[7] C Bath T Cukalac K Porter andM L Dyall-Smith ldquoHis1 andHis2 are distantly related spindle-shaped haloviruses belongingto the novel virus group Salterprovirusrdquo Virology vol 350 no1 pp 228ndash239 2006

[8] K Porter P Kukkaro J K H Bamford et al ldquoSH1 a novelspherical halovirus isolated from an Australian hypersalinelakerdquo Virology vol 335 no 1 pp 22ndash33 2005

[9] M Dyall-Smith S L Tang and C Bath ldquoHaloarchaeal viruseshow diverse are theyrdquo Research in Microbiology vol 154 no 4pp 309ndash313 2003

[10] A Oren G Bratbak and M Heldal ldquoOccurrence of virus-likeparticles in the Dead Seardquo Extremophiles vol 1 no 3 pp 143ndash149 1997

[11] T Sime-Ngando S Lucas A Robin et al et al ldquoDiversityof virus-host systems in hypersaline Lake Retba SenegalrdquoEnvironmental Microbiology vol 13 no 8 pp 1956ndash1972 2010

[12] N Guixa-Boixareu J I Calderon-Paz M Heldal G Bratbakand C Pedros-Alio ldquoViral lysis and bacterivory as prokaryoticloss factors along a salinity gradientrdquoAquaticMicrobial Ecologyvol 11 no 3 pp 215ndash227 1996

[13] S T Jaakkola R K Penttinen S T Vilen et al ldquoClosely relatedarchaeal Haloarcula hispanica icosahedral viruses HHIV-2 andSH1 have nonhomologous genes encoding host recognitionfunctionsrdquo Journal of Virology vol 86 no 9 pp 4734ndash47422012

[14] E Roine P Kukkaro L Paulin et al ldquoNew closely related halo-archaeal viral elements with different nucleic acid typesrdquo Jour-nal of Virology vol 84 no 7 pp 3682ndash3689 2010

[15] A Sencilo L Paulin S KellnerMHelm and E Roine ldquoRelatedhaloarchaeal pleomorphic viruses contain different genometypesrdquo Nucleic Acids Research vol 40 no 12 pp 5523ndash55342012

[16] H T Jaalinoja E Roine P Laurinmaki H M Kivela D HBamford and S J Butcher ldquoStructure and host-cell interactionof SH1 a membrane-containing halophilic euryarchaeal virusrdquoProceedings of the National Academy of Sciences of the UnitedStates of America vol 105 no 23 pp 8008ndash8013 2008

[17] M K Pietila N S Atanasova VManole et al ldquoVirion architec-ture unifies globally distributed pleolipoviruses infecting halo-philic archaeardquo Journal of Virology vol 86 no 9 pp 5067ndash50792012

[18] M L Holmes and M L Dyall-Smith ldquoA plasmid vector witha selectable marker for halophilic archaebacteriardquo Journal ofBacteriology vol 172 no 2 pp 756ndash761 1990

[19] M L Holmes F Pfeifer andM L Dyall-Smith ldquoAnalysis of thehalobacterial plasmid pHK2 minimal repliconrdquo Gene vol 153no 1 pp 117ndash121 1995

[20] M L Dyall-Smith F Pfeiffer K Klee et al ldquoHaloquadratumwalsbyi limited diversity in a Global Pondrdquo PLoS ONE vol 6no 6 Article ID e20968 2011

[21] S Casjens ldquoProphages and bacterial genomics what have welearned so farrdquoMolecular Microbiology vol 49 no 2 pp 277ndash300 2003

[22] R W Hendrix M C M Smith R N Burns M E Ford andG F Hatfull ldquoEvolutionary relationships among diverse bacte-riophages and prophages all the worldrsquos a phagerdquo Proceedings ofthe National Academy of Sciences of the United States of Americavol 96 no 5 pp 2192ndash2197 1999

[23] D Botstein ldquoA theory ofmodular evolution for bacteriophagesrdquoAnnals of the New York Academy of Sciences vol 354 pp 484ndash491 1980

[24] M Krupovic S Gribaldo D H Bamford and P Forterre ldquoTheevolutionary history of archaeal MCM helicases a case study ofvertical evolution combinedwithHitchhiking ofmobile geneticelementsrdquo Molecular Biology and Evolution vol 27 no 12 pp2716ndash2732 2010

[25] M Krupovic P Forterre and D H Bamford ldquoComparativeanalysis of the mosaic genomes of tailed archaeal viruses andproviruses suggests common themes for virion architecture andassembly with tailed viruses of bacteriardquo Journal of MolecularBiology vol 397 no 1 pp 144ndash160 2010

16 Archaea

[26] Z Zhang Y Liu S Wang et al et al ldquoTemperate membrane-containing halophilic archaeal virus SNJ1 has a circular dsDNAgenome identical to that of plasmid pHH205rdquoVirology vol 434no 2 pp 233ndash241 2012

[27] D H Bamford J J Ravantti G Ronnholm et al ldquoConstituentsof SH1 a novel lipid-containing virus infecting the halophiliceuryarchaeonHaloarcula hispanicardquo Journal of Virology vol 79no 14 pp 9097ndash9107 2005

[28] H M Kivela E Roine P Kukkaro S Laurinavicius P Somer-harju andDH Bamford ldquoQuantitative dissociation of archaealvirus SH1 reveals distinct capsid proteins and a lipid corerdquoVirology vol 356 no 1-2 pp 4ndash11 2006

[29] H Jaalinoja Electron Cryo-Microscopy Studies of Bacteriophage8 and Archaeal Virus SH1 University of Helsinki HelsinkiFinland 2007

[30] K Porter B E Russ J Yang andM L Dyall-Smith ldquoThe trans-cription programme of the protein-primed halovirus SH1rdquoMi-crobiology vol 154 no 11 pp 3599ndash3608 2008

[31] K Porter and M L Dyall-Smith ldquoTransfection of haloarchaeaby the DNAs of spindle and round haloviruses and the useof transposon mutagenesis to identify non-essential regionsrdquoMolecular Microbiology vol 70 no 5 pp 1236ndash1245 2008

[32] S E Luria General Virology John Wiley and Sons New YorkNY USA 1953

[33] C A Thomas K Saigo E McLeod and J Ito ldquoThe separationofDNA segments attached to proteinsrdquoAnalytical Biochemistryvol 93 pp 158ndash166 1979

[34] A L Delcher D Harmon S Kasif OWhite and S L SalzbergldquoImproved microbial gene identification with GLIMMERrdquoNucleic Acids Research vol 27 no 23 pp 4636ndash4641 1999

[35] R Zhang Y Yang P Fang et al ldquoDiversity of telomere palin-dromic sequences and replication genes among Streptomyceslinear plasmidsrdquo Applied and Environmental Microbiology vol72 no 9 pp 5728ndash5733 2006

[36] X Wang H S Lee F J Sugar F E Jenney M W W Adamsand J H Prestegard ldquoPF0610 a novel winged helix-turn-helix variant possessing a rubredoxin-like Zn ribbon motiffrom the hyperthermophilic archaeon Pyrococcus furiosusrdquoBiochemistry vol 46 no 3 pp 752ndash761 2007

[37] S W Cline L C Schalkwyk and W F Doolittle ldquoTransfor-mation of the archaebacterium Halobacterium volcanii withgenomic DNArdquo Journal of Bacteriology vol 171 no 9 pp 4987ndash4991 1989

[38] S Kamtekar A J Berman J Wang et al ldquoThe 12059329 DNA poly-merase protein-primer structure suggests a model for the ini-tiation to elongation transitionrdquo EMBO Journal vol 25 no 6pp 1335ndash1343 2006

[39] X Peng T Basta M Haring R A Garrett and D PrangishvilildquoGenome of theAcidianus bottle-shaped virus and insights intothe replication and packaging mechanismsrdquo Virology vol 364no 1 pp 237ndash243 2007

[40] C C Yang C H Huang C Y Li Y G Tsay S C Lee and CW Chen ldquoThe terminal proteins of linear Streptomyces chrom-osomes and plasmids a novel class of replication primingproteinsrdquo Molecular Microbiology vol 43 no 2 pp 297ndash3052002

[41] C C Yang Y H Chen H H Tsai C H Huang T W Huangand C W Chen ldquoIn vitro deoxynucleotidylation of the term-inal protein of Streptomyces linear chromosomesrdquo Applied andEnvironmental Microbiology vol 72 no 12 pp 7959ndash79612006

[42] P C Chang and S N Cohen ldquoBidirectional replication froman internal origin in a linear Streptomyces plasmidrdquo Science vol265 no 5174 pp 952ndash954 1994

[43] S W Wilhelm W H Jeffrey C A Suttle and D L MitchellldquoEstimation of biologically damaging UV levels in marinesurface waters with DNA and viral dosimetersrdquo Photochemistryand Photobiology vol 76 no 3 pp 268ndash273 2002

[44] T Allers and M Mevarech ldquoArchaeal geneticsmdashthe third wayrdquoNature Reviews Genetics vol 6 no 1 pp 58ndash73 2005

[45] M L Holmes S D Nuttall and M L Dyall-Smith ldquoConstruc-tion and use of halobacterial shuttle vectors and further studieson Haloferax DNA gyraserdquo Journal of Bacteriology vol 173 no12 pp 3807ndash3813 1991

[46] S D Nuttall and M L Dyall-Smith ldquoHF1 and HF2 novelbacteriophages of halophilic archaeardquo Virology vol 197 no 2pp 678ndash684 1993

[47] U Blaseio and F Pfeifer ldquoTransformation of Halobacteriumhalobium development of vectors and investigation of gas vesi-cle synthesisrdquo Proceedings of the National Academy of Sciences ofthe United States of America vol 87 no 17 pp 6772ndash6776 1990

[48] W L Lam and W F Doolittle ldquoShuttle vectors for the archae-bacterium Halobacterium volcaniirdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 86 no14 pp 5478ndash5482 1989

[49] C Yanisch-Perron J Vieira and J Messing ldquoImproved M13phage cloning vectors and host strains nucleotide sequences ofthe M13mp18 and pUC19 vectorsrdquo Gene vol 33 no 1 pp 103ndash119 1985

[50] U K Laemmli ldquoCleavage of structural proteins during theassembly of the head of bacteriophage T4rdquo Nature vol 227 no5259 pp 680ndash685 1970

[51] D G Burns H M Camakaris P H Janssen and M L Dyall-Smith ldquoCombined use of cultivation-dependent and culti-vation-independent methods indicates that members of mosthaloarchaeal groups in anAustralian crystallizer pond are culti-vablerdquo Applied and Environmental Microbiology vol 70 no 9pp 5258ndash5265 2004

[52] G Juez F Rodriguez-Valera A Ventosa and D J KushnerldquoHaloarcula hispanica spec nov and Haloferax gibbonsii specnov two new species of extremely halophilic archaebacteriardquoSystematic and Applied Microbiology vol 8 pp 75ndash79 1986

[53] A Oren M Ginzburg B Z Ginzburg L I Hochstein and BE Volcani ldquoHaloarcula marismortui (Volcani) sp nov nomrev an extremely halophilic bacterium from the Dead SeardquoInternational Journal of Systematic Bacteriology vol 40 no 2pp 209ndash210 1990

[54] M Torreblanca F Rodriguez-Valera G Juez A Ventosa MKamekura and M Kates ldquoClassification of non-alkaliphilichalobacteria based on numerical taxonomy and polar lipidcomposition and description of Haloarcula gen nov andHaloferax gen novrdquo Systematic and Applied Microbiology vol8 pp 89ndash99 1986

[55] A Ventosa and A Oren ldquoHalobacterium salinarum nom cor-rig a name to replace Halobacterium salinarium (Elazari-Vol-cani) and to include Halobacterium halobium and Halobac-terium cutirubrumrdquo International Journal of Systematic Bacte-riology vol 46 no 1 p 347 1996

[56] M C Gutierrez M Kamekura M L Holmes M L Dyall-Smith and A Ventosa ldquoTaxonomic characterization of Halo-ferax sp (ldquoH alicanteirdquo) strain Aa 22 description of Haloferaxlucentensis sp novrdquo Extremophiles vol 6 no 6 pp 479ndash4832002

Archaea 17

[57] M F Mullakhanbhai and H Larsen ldquoHalobacterium volcaniispec nov a Dead Sea halobacterium with a moderate saltrequirementrdquo Archives of Microbiology vol 104 no 1 pp 207ndash214 1975

[58] S D Nuttall and M L Dyall-Smith ldquoCh2 a novel halophilicarchaeon from an Australian solar salternrdquo International Jour-nal of Systematic Bacteriology vol 43 no 4 pp 729ndash734 1993

[59] P D Franzman E Stackebrandt K Sanderson et al ldquoHalobac-terium lacusprofundi sp nov a halophilic bacterium isolatedfrom Deep Lake Antarcticardquo Systematic and Applied Microbi-ology vol 11 pp 20ndash27 1988

[60] G A Tomlinson and L I Hochstein ldquoHalobacterium saccharo-vorum sp nov a carbohydrate metabolizing extremely halo-philic bacteriumrdquoCanadian Journal ofMicrobiology vol 22 no4 pp 587ndash591 1976

[61] A Ventosa M C Gutierrez M Kamekura and M L Dyall-Smith ldquoProposal to transfer Halococcus turkmenicus Halobac-terium trapanicum JCM9743 and strainGSL-11 toHaloterrigenaturkmenica gen nov comb novrdquo International Journal ofSystematic Bacteriology vol 49 no 1 pp 131ndash136 1999

[62] M Kamekura and M L Dyall-Smith ldquoTaxonomy of the familyHalobacteriaceae and the description of two new genera Halo-rubrobacterium and Natrialbardquo Journal of General and AppliedMicrobiology vol 41 no 4 pp 333ndash350 1995

[63] T J Carver KM RutherfordM BerrimanM A RajandreamB G Barrell and J Parkhill ldquoACT the Artemis comparisontoolrdquo Bioinformatics vol 21 no 16 pp 3422ndash3423 2005

[64] S W Cline W L Lam R L Charlebois L C Schalkwyk andW F Doolittle ldquoTransformationmethods for halophilic archae-bacteriardquo Canadian Journal of Microbiology vol 35 no 1 pp148ndash152 1989

Archaea 3

Table 1 Strains used in this study

Speciesisolate Strain ReferenceCSW 2094 Original isolate [51]Haloarcula hispanica ATCC 33960 [52]Haloarcula marismortui ATCC 43044 [53]ldquoHaloarcula sinaiiensisrdquo ATCC 33800 [54]Halobacterium salinarum NCIMB 763 [55]Haloferax gibbonsii ATCC 33959 [52]Haloferax lucentense NCIMB 13854 [56]Haloferax mediterranei ATCC 33500 [54]Haloferax volcanii ATCC 29605 [57]Halorubrum coriense ACAM 3911 [58]Halorubrum lacusprofundi ACAM 34 [59]Halorubrum saccharovorum NCIMB 2081 [60]Haloterrigena turkmenica NCIMB 784 [61]Natrialba asiatica JCM 9576 [62]

until well after this usually between 14 and 24 hr pi Arepresentative example of a single-step growth curve for PH1is shown in Figure 2 where the rise begins at sim6 hr pi andvisible cell lysis begins at sim14 hr pi This figure shows that asecond round of growth occurs at around 18 hr pi reflectingthe initial infection of only 30 of cells and it is during thissecond round of virus release that the cell density decreasesmost rapidly reaching a value of about 01 A

550 which is close

to the initial cell density In this example the average burstsize for the first growth step was 95 PFUcell

24 Virus Stability The stability of PH1 virus was testedunder several conditions (Figures 3(a)ndash3(d))When stored inHVD at 4∘C the infectivity of PH1 remained unaltered forseveral months (data not shown) A thermal stability curveis presented in Figure 3(a) and shows that PH1 is stable upto 56∘C above in which it rapidly loses titre Particles weresensitive to a reduced salt environment (Figure 3(b)) and tochloroform (Figure 3(d)) PH1 was most stable between pH8 and pH 9 (Figure 3(c)) In general the stability of PH1 wassimilar to that described previously for SH1 [8]

25 PH1 Structural Proteins and Protein Complexes Theproteins of purified PH1 virus were separated by SDS-polyacrylamide gel electrophoresis alongside the proteins ofSH1 virus (Figure 4) Nine PH1 protein bands were detectedwith molecular weights from 7 to 185 kDa (Figure 4(a)) andthese were designatedwith the prefixVP and a number corre-sponding to the homologous protein of SH1 [8 27] (see later)This nomenclature is consistent with that of the recentlydescribed HHIV-2 virus a member of the same virus group[13]The protein profile of PH1 was very similar to that of SH1[8] and with similar relative masses of the protein bands Onenotable difference was that PH1 proteins VP9 and VP10 ranclosely together (calculatedMWs are 165 and 167 kDa resp)compared to their SH1 homologs (165 and 169 kDa resp)It is probable given the strong sequence similarity to SH1(see later) that at least six additional PH1 structural proteinswere unable to be visualized using our staining techniques

0 5 10 15 200

01

02

Cell

den

sity

(A55

0)

Time (hr)

Viru

s titr

e (PF

Um

L)

1010

109

108

107

106

105

Figure 2 Growth curve of PH1 An early exponential culture ofHar hispanica in 18 (wv) MGM was infected with virus (MOI50) washed to remove unbound virus and incubated at 37∘C Atregular intervals samples were removed the absorbance at 550 nmwere measured (circles) and the number of infectious centres weredetermined by plaque assay (squares) Error bars represent onestandard deviation of the average titre

The potential glycosylation of virus proteins was examinedby staining similar protein gels either with a periodate-acid-Schiff (PAS) stain (GelCode Glycoprotein Staining Kit PierceBiotechnology USA) or the more sensitive fluorescent stain(Pro-Q Emerald 488 Glycoprotein Gel and Blot Stain KitMolecular Probes USA) No glycoproteins were detected

The major protein bands of PH1 (asterisked in Figure 4)were excised from gels digestedwith trypsin and analysed byMALDI-TOFMS and the results are summarized in Table 2From these data the structural proteins VP 1ndash4 7 9 10 and12 were found to be specified by ORFs 12 24 28 21 20 27 26and 19 respectively (see later)

26 Characteristics of the PH1 Genome Nucleic acid wasextracted from purified PH1 virus preparations treated withproteinase K and incubated with various nucleases to deter-mine the characteristics of the genome (Table 3) The PH1genome was sensitive to dsDNA endo- and exonucleasesbut not to ssDNA nuclease (mung bean) or RNase A Thisindicated that the PH1 genome is linear dsDNA with free(ie not covalently closed) termini Restriction endonucleasedigestions of the PH1 genome gave a length of approximately29 kb (data not shown)

The presence of terminal bound proteins was examinedusing a silica binding assay [33] Nonproteinase K-treatedPH1 DNA was restricted with AseI and the four resultingfragments passed through GFC filters under conditionswhere proteins bind firmly to the glass As shown in Figure 5the two internal AseI fragments (23 and 76 kb) passedthrough the filter but the right terminal fragment (174 kb)was bound indicating it carried an attached protein Theleft terminal fragment was too small (062 kb) and faintly

4 Archaea

0 20 40 60 80 100

Titre

(PFU

mL)

Temperature1011

1010

109

108

107

106

Temperature (∘C)

le105

(a)

0 5 10 15

0

Time (min)

Salt concentration

Titre

(PFU

mL)

107

106

105

104

minus5

(b)

4 5 6 7 8 9 10pH

pH

Titre

(PFU

mL)

108

107

(c)

Exposure time (min)0 5 10 15 20

Chloroform

Titre

(PFU

mL)

108

107

(d)

Figure 3 Stability of PH1 virus to various treatments and conditions Virus preparations (infected-cell supernatants in 18 (wv)MGM)wereexposed to various conditions after which the virus titre was determined (in duplicate) onHar hispanica cells (a) The effect of temperatureVirus was incubated for 1 hr with constant agitation at temperatures between 4 and 100∘C (b) The effect of lowered salt concentration Viruswas diluted 1 1000 in double-distilled H

2

O and incubated at room temperature with constant agitation Samples were removed at regularintervals (c)The effect of pHViruswas diluted 1 100 inTris-HCl buffers at the different pHs and incubatedwith constant agitation for 30min(d)The effect of chloroformChloroformwasmixedwith virus (1 4 ratio) and incubated at room temperaturewith constant agitation Sampleswere removed at regular intervals Open square symbols indicate where virus titres were undetectable Error bars represent one standarddeviation of the average titre

Archaea 5

Table 2 PH1 virus proteins identified by mass spectroscopy of tryptic peptides

Proteina Locus tag ORF MW (kDa) Matching peptide massesbObserved (calculated)

VP1 HhPH1 gp12 12 185 (158) 16VP2 HhPH1 gp24 24 100 (78) 15VP3 HhPH1 gp28 28 40 (38) 5VP4 HhPH1 gp21 21 35 (26) 5VP7 HhPH1 gp20 20 24 (20) 5VP9 HhPH1 gp27 27 16 (17) 3VP10 HhPH1 gp26 26 15 (17) 4VP12 HhPH1 gp19 19 7 (99) 4aVirus capsid proteins are numbered according to their similarity with SH1 capsid proteinsbThenumber of peptidemasses between 7500 and 35130mz identified byMALDI-TOFMS that correspond to the theoretical tryptic peptides of the predictedvirus protein

212 -

(kD

a)

PH1

158 -97 -66 -56 -42 -35 -

27 -

20 -

14 -

65 -

lowast

lowast

lowast

lowast

lowast

lowast

lowast

VP1VP2

VP3VP4

VP7

VP9

VP10

VP12

(a)

SH1

(b)

Figure 4 Structural proteins of halovirus PH1 Viral structuralproteinswere separated by SDS-PAGEon a 12 (wv) acrylamide geland stained with Brilliant Blue G (a) They were run in parallel withthe proteins of purified SH1 virus (b) The sizes of protein standardmarkers are indicated on the left side (in kDa) Asterisks denoteproteins bands of PH1 that were identified by mass spectroscopyThe numbering of proteins of PH1 (VP1ndashVP12) follows that of theSH1 homologs seen in (b) (and explained also in the text)

staining to be detected Further evidence was obtained by anuclease protection assay as previously used for SH1 DNA[31] As shown in Table 3 exonuclease III (a 31015840 exonuclease)was able to digest nonproteinase K-treated PH1 DNA butT7 exonuclease (a 51015840 exonuclease) and Bal31 (51015840 and 31015840exonucleases) were unable to digest PH1 DNA unless it hadbeen previously treated with proteinase KThis indicated thatboth 51015840 termini of the genome had protein attached

27 Sequence of the PH1 Genome The complete PH1 genomesequence was determined (accession KC252997) using a

1 2 3

-23

-76

-17410 -

6 -

3 -

2 -

15 -

05 -

4 -

(kB)

Figure 5 Detection of proteins bound to the termini of PH1 DNADNA was extracted from purified PH1 particles without proteinaseK digested with AseI then passed through a silica filter (GFC)under conditionswhere proteinswould bind to the filter Lane 1AseIfragments that did not bind and were eluted Lane 2 AseI fragmentsthat bound and were eluted only after protease treatment Lane 3AseI digest of PH1 DNA The positions of DNA size standards areindicated on the left side in kb The calculated sizes of the AseIfragments of PH1 DNA in lane 3 are indicated on the right side (alsoin kb) The predicted 062 kb PH1 AseI fragment was not detected

combination of cloned fragments PCR primer walkingon virus DNA and 454 whole genome sequencing (seeSection 4) It was found to be 28072 bp in length 676G + C with inverted terminal repeat sequences (ITRs) of337 bp Using GLIMMER [34] manual BLAST searching atthe GenBank database and comparison to related virusesthe PH1 genome sequence was predicted to contain 49 ORFs(Table 4) Most ORFs were closely spaced or overlappinggiving a gene density of 174 geneskb (average of 572 ntgene)The cumulative AT-skew plot of the PH1 genome shown atthe bottom of Figure 6 shows inflection points (circled) thatare consistent with the changes in transcription direction

6 Archaea

6 8 10 12 14 16 18 20 22 24 kb

1 2 4 6 8 10 12 14 162 kb

VP1

VP1-like fragments

A

A

VP2

VP2

VP7

VP1

2

20 21 22

VP4

VP1

3

VP5 p2

2

p9p4 p11

p12

p13

NP5

356A

Hmuk

p26

VP3

VP1

7

1728Hmuk

0477

Locus tag ZOD2009 19038

Locus tag HlacAJ 010100019872 NZ AGFZ01000123 (reversed)

NZ AEMG01000027 contig29

TransposaseHlacAJ 010100019767

ZOD2009 19153

Hbf lacisalsi

Hap paucihalophilus

(a)

VP18VP2VP1 A

VP1

VP18VP1

VP1

2

VP1

VP7

VP4

VP1

3

VP5

VP2

VP9

VP1

0

VP3

VP1

6

VP1

7V

P6 VP18

ATPa

se

kb1 2 4 6 8 10 12 14 16 18 20 22 24 26 28

VP2

VP2

VP18A

AA

2211 1312

94 26

HHIV-2

SH1

PH1

0

2

4

6

30 578 nt

30 889 nt

28 072 nt

(times102

)Cu

m A+

T sk

ew

Cumulative A + T skew of PH1

(b)

Figure 6 Genome alignments of haloviruses PH1 SH1 and HHIV-2 along with two related genomic loci (a) Genomic loci of Happaucihalophilus andHbf lacisalsi that contain genes related to halosphaeroviruses SH1 PH1 andHHIV-2The names andGenBank accessionsfor these contigs are given on the far right andORFs are coloured and labeled to indicate the relationships of theseORFs to those of the virusesbelowThe locus tag numbers for the first and last ORFs shown in each locus are given nearby their respectiveORFs In addition grey colouredORFs represent sequences that do not match any of the haloviruses and green coloured ORFs represent protein sequences that are closelyrelated to ORFs found within or very close to previously described virusplasmid loci so called ViPREs [20] The scale bars shown aboveeach contig show the position of the described region within the respective contig (b) The three virus genomes are labeled at the left withscale markers below (in kb) and the total length indicated at the far right At the bottom is a cumulative AT-skew plot of the PH1 genome(httpmolbiol-toolscaJie Zheng) with inflection points circledThe grey shaded bands between the genome diagrams indicate significantnucleotide similarity (usingACT [63]) AnnotatedORFs are represented by arrows with colours indicating structural proteins (red or brown)nonstructural proteins (yellow or orange) or the packaging ATPase (blue) The names of structural protein ORFs are indicated either withinthe arrow (eg VP1) or in text nearby The numbered orange coloured ORFs of HHIV-2 are homologous to ORFs found in the genomic loci(probably proviruses or provirus remnants) pictured in (a)

indicated by the annotated ORFs as has been seen in otherhaloviruses [7] There are no GATC motifs in the genomeand the inverse sequence CTAG is strongly underrepre-sented with just 2 motifs (68 expected)

When compared to the related viruses SH1 and HHIV-2(Figure 6) the PH1 genome is seen to be of similar length andis much more closely related to SH1 than to HHIV-2 (74and 54 nt identity resp) Like PH1 the genomes of SH1and HHIV-2 lack GATC motifs and CTAG is either absent(SH1) or underrepresented (HHIV-2) The ITRs of PH1 andSH1 share 785nucleotide identity but the PH1 ITR is longerthan that of SH1 (337 and 309 nt resp) partly because SH1has replaced an 18 bp sequence (gtcgtgcggtttcggcgg) found atthe internal end of its left-hand ITR with a sequence at thecorresponding position of its right-hand ITR that shows littleinverted sequence similarity In the PH1 ITR this sequenceis retained at both ends Whether this difference represents arecombination event or a mistake in replication of the ITRs

is unclear An alignment of the ITRs of all three virusesidentified a number of highly conserved regions (labeled 1ndash9 in Supplementary Figure 2 see Supplementary Materialavailable online at httpdxdoiorg1011552013456318) A16 bp sequence at the termini (region 1) is conserved con-sistent with the conservation seen at the termini of linearStreptomyces plasmids [35] A 16 bpGC-rich sequence aroundthe middle of the ITR (region 3) is also conserved Shortermotifs of either C-rich (region 8) or AT-rich sequence(regions 7 and 9) are found near the internal end of the ITRs

The grey shading between the schematic virus genomesin Figure 6(b) indicates regions of high nucleotide similarityrevealing that the differences between PH1 and SH1 are notevenly distributed but vary considerably along the lengthof the aligned genomes A comparison between SH1 andHHIV-2 has been published recently [13] and since the PH1and SH1 genomes are so similar we will focus the followingdescription on the differences between PH1 and SH1

Archaea 7

Table 3 Digestion of the PH1 genome by nucleasesa

Nuclease (amount) Proteinase K-treated UntreatedControl PH1 DNA Control PH1 DNA

DNase I (RNase-free)(100U) +b + +b +

Exonuclease III(100U) +b + +b +

Mung bean nuclease(10U) +c

minus +cminus

Nuclease BAL-31(1 U) +b + +b

minus

RNase A(5 120583gmL) +d

minus +dminus

T7 exonuclease(10U) +b + +b

minus

aExtracted nucleic acid from purified PH1 virus was treated with variousnucleases and then analysed by gel electrophoresis to detect whether thenuclease digested (+) or did not digest (minus) the genome Control nucleicacids used to confirm activity of the nucleases were b

120582 DNA ca DNAoligonucleotide and dyeast transfer RNA

Within corresponding ORFs there are blocks of codingsequence with low (or no) similarity These may represent insitu divergence or short recombination events (eg indels)There are also cases where an entire ORF in one virus hasno corresponding homolog in the other because of an inser-tiondeletion event (indel) Lastly there are replacementswhere the sequences within the corresponding regions showlow or no similarity but are flanked by sequence with highsimilarity The largest visible differences seen in Figure 6 aredue to sequence changes within or near the long genes encod-ing capsid proteins VP1 and VP18 A summary of the mainsequence differences between SH1 and PH1 and their loca-tions is given in Table 5 where these divergent regions (DV)are numbered from DV1 to DV18

The sequences of the two viruses are close enough that inmany cases the mechanism for the observed differences canbe inferred For example DV2 is likely the result of a deletionevent that has removed the SH1 ORF10 gene homolog fromthe PH1 genome At either end of the SH1 ORF are almostperfect direct repeats (cggcctgaccggcatgac) that would allowrepeat-mediated deletion to occur removing the interveningsequence Small direct repeats leading to deletions have beendescribed previously in the comparative analysis of Halo-quadratumwalsbyi genomes [20] DV5 appears to be an indelwhere SH1 ORFs 14ndash16 are absent in PH1 and an invertedrepeat (AGCCATG) found at each end of the SH1 divergentregion may be significant in the history of this changeThe proteins specified by SH1 ORFs 14ndash16 are presumablydispensable for PH1 (or their functions supplied by otherproteins) but a homolog of SH1 ORF14 is found also inHHIV-2 (ORF 6) and a homolog of SH1 ORF15 occurs inhalovirus His1 (ORF13) Replacement regions are also com-mon and can occur within ORFs (eg DV3 in capsid proteinVP1) or provide additional or alternative genes (eg DV12and DV13) Several divergent regions occur within the genefor capsid protein VP2 a hot spot for change because of the

repetitive nature of the coding sequence which specifies aprotein with runs of glycines andmany heptapeptide repeatsas described previously for SH1 VP2 [27] DV18 is a largereplacement that significantly alters the predicted length ofthe minor capsid protein VP18 in the two viruses (865519aa for SH1PH1 resp) and also provides SH1 with 3 ORFsthat are different or absent in PH1 ORFs 52 53 and 54 Forexample a homolog of SH1 ORF52 is not present in PH1 butis present in HHIV-2 (putative protein 40) While ORF48 ofPH1 shows no aa sequence homology to SH1 ORFs 53 and 54all of these predicted proteins carry CxxC motifs suggestiveof related functions such as DNA binding activity [36]

The genes of the PH1 genome are syntenic with those ofSH1 and are found in similarly oriented blocks of closelyspaced or overlapping ORFs that suggest that transcription isorganised into operons The programme of transcription ofthe SH1 genome has been reported previously [30] and thesequences in PH1 corresponding to the six promoter regions(P1ndashP6) determined in SH1 showed that five of these (P1ndashP5)are well conserved Only the region corresponding to SH1 P6was poorly conserved (data not shown) but this promoter isstrongly regulated in SH1 and only switches on late in infec-tion [30]

28 Structural Protein Genes The genes coding for the majorvirus structural proteins of PH1 (VP 1ndash4 7 9 10 and 12)were identified byMALDI-TOF (see above) and the genes forthese proteins are found in the same order and approximatepositions as in the SH1 genome (Figure 6(b) red ORFs)Genes coding for minor structural proteins can be deducedby sequence comparison with SH1 so that VP5 and VP6 arelikely to be encoded by PH1ORFs 25 and 29 respectively andVP13 and VP18 by ORFs 23 and 49 respectively This wouldgive PH1 a total of 11 structural proteins the same as SH1Of these VP7 is the most conserved between the two viru-ses (98 aa identity) followed by VP4 VP9 VP3 and VP12(91ndash94 identity) VP5 VP6 and VP13 (82ndash88 iden-tity) VP2 (77) VP1 (65) and lastly the least conservedprotein was VP18 (31)

29 Genomic Loci Related to PH1 and Other Halosphaeroviru-ses Clusters of halosphaerovirus-related genes are present inthe genomes of two recently sequenced haloarchaea Happaucihalophilus and Hbf lacisalsi (Figure 6(a)) Neither ofthese regions appears to represent complete virus genomebut they retain a number of genes that share a similar seq-uence and synteny with the virus genomes depicted belowthem (Figure 6(a)) Both loci show a mixed pattern ofrelatedness to SH1PH1 and HHIV-2 as indicated by ORFsof matching colour and gene name If these genomic locirepresent provirus integrants that have decayed over timethen the relationships they show suggest not only that halo-sphaeroviruses are a diverse virus group but also that asignificant level of recombination occurs between them lead-ing to mosaic gene combinations The two ORFs colouredgreen in theHap paucihalophilus locus (Figure 6) are relatedto ORFs found within or very close to genomic loci ofvirusplasmid genes found in other haloarchaea suggestingadditional links between (pro)viruses [20]

8 Archaea

Table4ORF

anno

tatio

nsof

theh

aloviru

sPH1g

enom

e

ORF

aPo

sitionb

Locustag

Leng

thc(aa)

pKi

Similaritycharacteris

tics

144

1ndash605

HhP

H1gp01

5043

76aa

similarityto

SH1O

RF2(YP271859)

2602ndash802

HhP

H1gp02

66109

80aa

similarityto

SH1O

RF3(YP271860)

3802ndash1059

HhP

H1gp03

8539

85aa

similarityto

SH1O

RF4(YP271861)44

sim

ilarityto

HHIV-2

protein1

41056ndash1283

HhP

H1gp04

7539

85aa

identityto

SH1O

RF5(YP271862)Predictedcoiled-coilregion

51276ndash1686

HhP

H1gp05

136

49

80aa

similarityto

SH1O

RF6(YP271863)Tw

otransm

embraned

omainsA

lsorelated

toHHIV-2

protein2(38

AFD

02283)

andto

Halobiform

alacisalsiORF

ZP09950018

(35

)6

1683ndash1829

HhP

H1gp06

4861

81aa

similarityto

SH1O

RF7(YP271864

)Predictedsig

nalsequence(Sign

alP)

71822ndash2130

HhP

H1gp07

102

58

91aa

similarityto

SH1O

RF8(YP271865)a

nd86to

HHIV-2

putativ

eprotein

3(A

FD02284)O

ther

homologsinclude

Nocardiaproteinpn

f2110(YP1220601)H

rrlacusprofund

iHlac0751

(YP0025654211)ORF

sinactin

ophage

VWB

(AAR2

9707)Frankia(EAN116

57)andgp40

ofMycobacteriu

mph

ageD

ori(AER

47690)

82123ndash2383

HhP

H1gp08

8642

90aa

similarityto

SH1O

RF9(YP271866)

92380ndash2714

HhP

H1gp09

111

52

88aa

similarityto

SH1O

RF11(YP271868)and

also

similarityto

putativ

eprotein

4of

HHIV-2

(AFD

02285)

HlacA

J19877of

Hbflacisa

lsiAJ

5(ZP09950016)

102861ndash300

4HhP

H1gp10

47104

Weaksim

ilarityto

SH1O

RF12

(YP271869)o

nlyover

theN

-term

inal18

resid

ues

113029ndash3100

HhP

H1gp11

23123

Not

presentinSH

1orH

HIV-2

123134ndash7453

HhP

H1gp12

1439

42

Capsid

protein

VP172sim

ilarityto

SH1V

P1(O

RF13Y

P271870)HHIV-2

VP1

(AFD

02286)H

lacA

J19872Hbflacisa

lsiAJ

5Predictedhelix-tu

rn-helixandRu

vA-like

domain(In

terproScan)

137505ndash8227

HhP

H1gp13

240

54

Predicted

P-loop

ATPa

sedo

main(C

OG0433)92aa

similarityto

SH1O

RF17

(YP271874)a

ndalso

similarityto

putativ

eAT

Pase

ofHHIV-2

(AFD

02288)A

TPaseo

fHaladapatus

paucihalophilusD

X253

(ZP0804

6180)HlacA

J19867of

Hbf

lacisalsi(ZP09950014)

148419ndash8228c

HhP

H1gp14

6351

43aa

similarityto

SH1O

RF18

(YP271875)Alanine-richCentraltransm

embraned

omain(Pho

bius)

158856ndash8

416c

HhP

H1gp15

146

41

92aa

similarityto

SH1O

RF19

(YP271876)a

ndalso

toHHIV-2

putativ

eprotein

9(A

FD02290)H

lacA

J19857of

Hbf

lacisalsiAJ

5ZO

D2009

19093of

Happau

cihalophilus

168853ndash9

500c

HhP

H1gp16

215

39

65aa

similarityto

SH1O

RF20

(YP271877)a

ndalso

topu

tativ

eprotein

11of

HHIV-2

(AFD

02292)Z

OD2009

19098of

Hap

paucihalophilusHlacA

J19852Hbflacisa

lsiAJ

5C-

term

inaltransm

embraned

omain(Pho

bius)

179500ndash9

919c

HhP

H1gp17

139

41

79aa

similarityto

SH1O

RF21

(YP271878)a

ndalso

topu

tativ

eprotein

12of

HHIV-2

(AFD

02293)hypotheticalprotein

HlacA

J19847of

Hbflacisa

lsi(ZP09950010)Transm

embraned

omain(Pho

bius)

189923ndash10594c

HhP

H1gp18

223

43

83aa

similarityto

SH1O

RF22

(YP271879)a

ndalso

topu

tativ

eprotein

13of

HHIV-2

(AFD

02294)Z

OD2009

19108of

Hap

paucihalophilusHlacA

J19842of

Hbflacisa

lsiP

redicted

signalsequence(sig

nalP)Con

tains4

CxxC

motifs

1910659ndash10943

HhP

H1gp19

94104

Capsid

protein

VP12

(ORF

19)91aa

similarityto

SH1V

P12(O

RF23Y

P271880)a

ndalso

toVP12of

HHIV-2

(AFD

02295)

andHlacA

J19837of

Hbflacisa

lsiZ

OD2009

19113of

Happau

cihalophilusTw

otransm

embraned

omains

(Pho

bius)

2010960ndash

11517

HhP

H1gp20

185

44

Capsid

protein

VP7(O

RF20)98aa

similarityto

SH1V

P7(O

RF24Y

P271881)a

ndalso

toVP7

ofHHIV-2

(AFD

02296)

ZOD2009

19118

ofHappau

cihalophilusHlacA

J19832of

Hbflacisa

lsiAJ

5

2111519ndash

12217

HhP

H1gp21

232

41

Capsid

protein

VP4(O

RF21)94aa

similarityto

SH1O

RF25

(YP271882)a

ndalso

toVP4

ofHHIV-2

(AFD

02297)

ZOD2009

19123of

Happau

cihalophilusHlacA

J19827of

Hbflacisa

lsiAJ

522

12233ndash12454

HhP

H1gp22

7339

81aa

similarityto

SH1O

RF26

(YP271883)a

ndalso

topu

tativ

eprotein

17of

HHIV-2

(AFD

02298)

2312458ndash12697

HhP

H1gp23

7948

78aa

similarityto

SH1capsid

protein

VP13

(ORF

27Y

P271884)a

ndalso

toVP13of

HHIV-2

(AFD

02299)P

redicted

coil-coildo

main

PredictedC-

term

inaltransm

embraned

omain(Pho

bius)

Archaea 9

Table4Con

tinued

ORF

aPo

sitionb

Locustag

Leng

thc(aa)

pKi

Similaritycharacteris

tics

2412701ndash1506

4HhP

H1gp24

787

39

Capsid

protein

VP2(O

RF24)75aa

similarityto

SH1V

P2(O

RF28Y

P271885)a

ndto

VP2

ofHHIV-2

(AFD

02300)

ZOD2009

19133of

Happau

cihalophilus(ZP

0804

6189)

2515065ndash15961

HhP

H1gp25

298

45

Putativec

apsid

protein

VP5(O

RF25)81aa

similarityto

SH1V

P5(O

RF29Y

P271886)a

ndto

HHIV-2

VP5

(AFD

02301)

HlacA

J19807of

HbflacisalsiAJ

5ZO

D2009

19133of

Happau

cihalophilus

261596

4ndash1644

6HhP

H1gp26

160

46

73aa

similarityto

SH1capsid

protein

VP10

(ORF

30Y

P271887)a

ndto

VP10of

HHIV-2

(AFD

02302)Z

OD2009

19138of

Happau

cihalophilusHlacA

J19802of

Hbflacisa

lsiP

redicted

C-term

inaltransm

embraned

omain(TMHMM)

271644

6ndash16895

HhP

H1gp27

149

42

Capsid

protein

VP9(O

RF27)94aa

similarityto

SH1V

P9(O

RF31Y

P271888)a

ndto

ZOD2009

19143of

Hap

paucihalophilus(ZP

0804

6191)

2816908ndash17921

HhP

H1gp28

337

43

Capsid

protein

VP3(O

RF28)91aa

similarityto

SH1V

P3(O

RF32Y

P271889)a

ndto

NJ7G

2365

ofNa

trinemaspJ7-2

2917928ndash18617

HhP

H1gp29

229

42

Capsidprotein

VP6(O

RF29)83aa

similarityto

SH1V

P6(O

RF33Y

P271890)Also

toNJ7G

3194

ofNa

trinemaspJ7-2and

Hmuk

0476

ofHalom

icrobium

mukohataeiPredictedcarboxypeptid

aser

egulatory-lik

edom

ain(C

arbo

xypepD

reg

pfam

13620)

3018614ndash

18943

HhP

H1gp30

109

49

71aa

similarityto

SH1O

RF34

(YP271891)a

ndto

putativ

eprotein

27of

HHIV-2

(AFD

02308)P

redicted

signalsequence

(signalP)

3119054ndash

19176c

HhP

H1gp31

4050

TwoCx

xCmotifsN

oho

molog

inSH

1orH

HIV-2

3219173ndash19289c

HhP

H1gp32

3841

68aa

similarityto

SH1O

RF37

(YP271894)

3319286ndash

19552c

HhP

H1gp33

8853

58aa

similarityto

SH1O

RF39

(YP271896)a

ndto

HHIV-2

putativ

eprotein

29(A

FD02310)Fou

rCxxCmotifs

3419549ndash

19731c

HhP

H1gp34

6070

Con

tainsC

xxCmotif

Noho

molog

inSH

1orH

HIV-2

3519728ndash20219c

HhP

H1gp35

163

56

93aa

similarityto

SH1O

RF41

(YP271898)a

ndto

putativ

eprotein

30of

HHIV-2

(AFD

02311)

3620216ndash

21415c

HhP

H1gp36

399

50

95aa

similarityto

SH1O

RF42

(YP271899)a

ndto

putativ

eprotein

31of

HHIV-2

(AFD

02312)

3721419ndash

21778c

HhP

H1gp37

11941

91aa

similarityto

SH1O

RF43

(YP271900)a

ndto

putativ

eprotein

32of

HHIV-2

(AFD

02313)

3821762ndash2300

6cHhP

H1gp38

414

42

83aa

similarityto

SH1O

RF44

(YP271901)a

ndto

putativ

eprotein

33of

HHIV-2

(AFD

02314)P

redicted

coil-coiland

helix-tu

rn-helixdo

mains

3923003ndash23158c

HhP

H1gp39

5165

69aa

similarityto

SH1O

RF45

(YP271902)

4023161ndash23370c

HhP

H1gp40

6935

74aa

similarityto

SH1O

RF46

(YP271903)and55sim

ilarityto

putativ

eprotein

35of

HHIV-2

(AFD

02316)

4123422ndash23586c

HhP

H1gp41

5476

86aa

similarityto

SH1O

RF47

(YP271904)Con

tains2

CxxC

motifsand

show

ssim

ilarityto

proteindo

mainfamily

PF14206Arginine-ric

h

4223583ndash24023c

HhP

H1gp42

48

77aa

similarityto

SH1O

RF48

(YP271905)a

ndto

putativ

eprotein

36of

HHIV-2

(AFD

02317)H

ham114540of

Hcc

hamelinensis(ZP11271999)Ph

iCh1p72of

Natrialba

phageP

hiCh

1(NP665989)DUF4

326(pfam14216)

family

domain

4324020ndash

24538c

HhP

H1gp43

172

47

79aa

similarityto

SH1O

RF49

(YP271906)and56sim

ilarityto

putativ

eprotein

37of

HHIV-2

(AFD

02318)

4424535ndash25053c

HhP

H1gp44

172

43

92aa

similarityto

SH1O

RF50

(YP271907)and72sim

ilarityto

putativ

eprotein

38of

HHIV-2

(AFD

02319)

4525050ndash

25277c

HhP

H1gp45

7543

Noho

molog

inSH

1orH

HIV-2

4625274ndash

25387c

HhP

H1gp46

3748

Noho

molog

inSH

1orH

HIV-2

4725533ndash25904

HhP

H1gp47

123

43

61aa

similarityto

SH1O

RF51

(YP271908)66sim

ilarityto

putativ

eprotein

39of

HHIV-2

(AFD

02320)P

redicted

COG1342

domain(D

NAbind

inghelix-tu

rn-helix)

4825901ndash26092c

HhP

H1gp48

6349

Con

tainsC

xxCmotif

Noho

molog

inSH

1orH

HIV-2

4926089ndash

2764

8cHhP

H1gp49

506

40

81aa

similarityto

SH1O

RF55

(YP271912)(bu

tonlyin

theN

-term

inalhalf)H

omolog

ofvirusstru

cturalprotein

VP18

ofHHIV-2

(AFD

02323)

a ORF

swerep

redicted

either

byGLIMMER

orby

manualsearching

forh

omologsintheG

enBa

nkdatabase

b Startandendpo

sitions

ofORF

sare

give

inbp

numbera

ccording

totheP

H1sequenced

epositedatGenBa

nk(KC2

52997)O

RFso

nthec

omplem

entary

strandared

enoted

bythes

uffixc

c Lengthof

thep

redicted

ORF

innu

mbero

faminoacids

10 Archaea

Table 5 Main differences (divergent regions) between the SH1 and PH1 genomes

RegionaSH1 startb SH1 stop Length (bp) PH1 start PH1 stop Length (bp) Comment

DV1 527 650 123 538 604 66

Replacement Both regions have direct repeats at their bordersPH1 has two sets GACCCGGC and CGCTGC while SH1 hasCCCGAC In SH1 this replacement region includes theC-terminal region of ORF2 and the N-terminal region ofORF3 In PH1 it covers only the C-terminal region of ORF1

DV2 2433 2588 155 2386 2387 mdash RMDc from PH1 SH1 repeat at border is TGACCGThisremoves the homolog of SH1 ORF10 from PH1

DV3 3324 3598 274 3137 3416 279 Replacement at the beginning of SH1 ORF13PH1 ORF12(capsid protein VP1)

DV4 6656 7128 472 6478 7085 607 Replacement near the C-terminus of SH1 ORF13PH1 ORF12(capsid protein VP1)

DV5 7491 8341 850 7446 7447 mdash

Indel that results in ORFs14-16 of SH1 not being present inPH1 SH1 ORF14 has a close homolog in HHIV-2 (ORF6) in asimilar position just after the VP1 homolog SH1 ORF15 is aconserved protein in haloarchaea (eg NP 2552A andHmuk 2978) and halovirus His1 (ORF13) Possible invertedrepeat (AGCCATG) at border of SH1 region

DV6 13705 13706 mdash 12818 12851 33RMD from SH1 PH1 repeat at border is CAGCGG(gt)G Thisremoves a part of capsid protein VP2 sequence from the SH1protein

DV7 13789 13863 74 12934 12941 7 Replacement This occurs within VP2 gene of both viruses

DV8 15142 15186 44 14220 14221 mdash RMD from PH1 SH1 repeat at border is TG(tc)CCGACGAand occurs within capsid protein VP2 gene of both viruses

DV9 15303 15317 14 14337 14338 mdash RMD from PH1 SH1 repeat at border is GCCGACGA andoccurs within capsid protein VP2 gene of both viruses

DV10 15504 15527 23 14529 14530 mdash RMD from PH1 SH1 repeat at border is GACGA and occurswithin capsid protein VP2 of both viruses

DV11 20026 20405 379 19019 19173 154Replacement beginning at the start of SH1 ORF35PH1 ORF31This region is longer in SH1 where it includes ORF36 an ORFthat has no homolog present in PH1

DV12 20440 20661 221 19208 19286 78 Replacement This is unequal and includes an ORF in SH1(ORF38) that is not present in PH1

DV13 20943 21085 142 19562 19728 166 Replacement This covers SH1 ORF40PH1 ORF34 Thepredicted proteins are not homologous

DV14 23541 23542 mdash 22171 22197 27 Probable RMD in SH1 PH1 repeat at border is CGTCTCGGand occurs in SH1 ORF44PH1 ORF38

DV15 24506 24521 15 23152 23153 mdash Probable RMD in PH1 SH1 repeat at border is CTCGGT andoccurs near the end of SH1 ORF45PH1 ORF39

DV16 24951 24964 13 23579 23580 mdash Probable RMD in PH1 SH1 repeat at border is CGGTC andoccurs in SH1 ORF47PH1 ORF41

DV17 26449 26450 mdash 25050 25274 224Probable RMD in SH1 PH1 repeat at border is TCATGCG andoccurs near the start of SH1 ORF50PH1 ORF44 and providesan extra ORF for PH1 (ORF45)

DV18 27074 29764 690 25895 26933 1038

Replacement Left border at start of SH1 ORF51PH1 ORF47and extends rightwards into SH1 ORF55PH1 ORF49 (VP18gene) It is an unequal replacement and SH1 has two moreORFs in this region than PH1

aDV Divergent regions between the genomes of SH1 and PH1bStart and stop positions refer to the GenBank sequences of the two viruses SH1 NC 0072171 PH1 KC252997cRMD repeat-mediated deletion event as described in [20]

The recently described halovirus SNJ1 carries manyORFsthat have homologs adjacent to a previously described ViPREofHmc mukohataei (from here on denoted by ViPREHmuk1)a locus that contains genes related to His2 (and other ple-olipoviruses) as well as toHaloferax plasmid pHK2 (probablya provirus) and to two small plasmids ofHqr walsbyi (sim6 kbpL6A and pL6B)The latter relationship provides wider linksto other viruses because one of the PL6 genes (Hqrw 6002)

has homologs in some pleolipoviruses (HPRV-3 and HGPV-1) while another gene (Hqrw 6005) is related to ORF16 ofthe spindle-shaped halovirus His1 [15] After adding the SNJ1gene homologs to ViPREHmuk1 it is extended significantlyand a close inspection of the flanking regions revealed atRNA-ala gene at one end and a partial copy of the sametRNA-ala gene (next to a phage integrase gene) at the otherend (supplementary Figure 1) Genes beyond these points

Archaea 11

Table 6 Transfection of haloarchaea by PH1 DNA

Speciesa Efficiency of transfectiontransformation withPH1 DNAb (PFU120583g of DNA) pUBP2 DNAbc (CFU120583g of DNA)

Har hispanica 53 plusmn 05 times 103 14 plusmn 08 times 104

Har marismortui 45 plusmn 07 times 103 18 plusmn 01 times 104

ldquoHar sinaiiensisrdquo 32 plusmn 06 times 102 mdashHalobacterium salinarum mdash 54 plusmn 42 times 103

Haloferax gibbonsii 28 plusmn 08 times 103 40 plusmn 26 times 103

Haloferax lucentense mdash 57 plusmn 11 times 103

Hfx volcanii mdash 21 plusmn 07 times 105

Hrr lacusprofundi 70 plusmn 22 times 102 95 plusmn 15 times 103

Haloterrigena turkmenica 16 plusmn 17 times 102 24 plusmn 06 times 103

Natrialba asiatica 35 plusmn 39 times 102 31 plusmn 10 times 104aOnly those species positive for transfection andor plasmid transformation are shownbRates of transfection by (nonprotease treated) PH1 DNA or transformation by plasmid pUB2 are averages of three independent experiments each performedin duplicate (plusmn standard deviation)cTransformants were selected on plates with 2 4 or 6120583gmL simvastatin depending on the strainmdash no plaques or colonies observed

appear to be related to cellular metabolism ViPREHmuk1is now 39379 nt in length flanked by a 59 nt direct repeat(potential att sites) and includes 54 ORFs many of which arerelated to known haloviruses are virus-like (eg integrasesmethyltransferases transcriptional regulators DNA methyl-transferase and DNA glycosylase) are homologs in or nearother knownViPREs or show little similarity to other knownproteins This locus also includes an ORC1CDC6 homolog(Hmuk 0446) which could provide a replication function

The mixture of very different virus and plasmid genesseen within ViPREHmuk1 is remarkable but it also revealshomologs found in or adjacent to genomic loci of otherspecies such as the gene homologs of Har marismortuiseen in the left end of ViPREHmuk1 (supplementary Figure1) When these genes are added to the previously describedHaloarcula locus (ViPREHmuk1) it is also extended signifi-cantly It becomes 25550 bp in length encompasses genesfrom Hmar 2382 to Hmar 2404 and is flanked by a fulltRNA-ala gene at one end and a partial copy at the other Itcarries an ORC1CDC6 homolog an integrase and a phiH-like repressor as well as pleolipovirus homologs

210 Transfection of Haloarchaea by PH1 DNA PH1 DNAwas introduced into Har hispanica cells using the PEGmethod [37] and the cells were screened for virus productionby plaque assay Figure 7 shows the increase in transfectedcells with input (nonproteinase K-treated) virus DNA Theestimated efficiencywas 53times 103 PFUper120583gDNA PH1DNAthat had been treated with proteinase K or with DNAase I(RNase-free) did not produce plaques (data not shown) PH1DNA was found to transfect six species of haloarchaea otherthanHar hispanica includingmembers of the generaHaloar-cula Haloferax Halorubrum Haloterrigena and Natrialba(Table 6) For these experiments the virus production fromtransfected cells was detected by plaque assay on indicatorlawns of Har hispanica

0

0

200 400 600 800 1000

Titre

(PFU

mL)

104

103

102

PH1 DNA (ng)

Figure 7 Transfection ofHar hispanica cells with PH1 DNA Vary-ing amounts of nonproteinase K-treated PH1 DNAwere introducedinto cells of Har hispanica using the PEG method [64] Cells werethen screened for infective centres by plaque assay Data shown is theaverage of three independent experiments performed in duplicateError bars represent one standard deviation of themean If protease-treated DNA was used no transfectant plaques were observed

3 Discussion

PH1 is very similar in particlemorphology genome structureand sequence to the previously described halovirus SH1 [816] Like SH1 it has long inverted terminal repeat sequencesand terminal proteins indicative of protein-primed replica-tion Purified PH1 particles have a relatively low buoyant

12 Archaea

density and are chloroform sensitive and show a layeredcapsid structure consistent with the likely presence of aninternalmembrane layer as has been shown for SH1HHIV-2and SNJ1 [13 26]The particle stability of PH1 to temperaturepH and reduced salt was similar to SH1 [8] The structuralproteins of PH1 are very similar in sequence and relativeabundance to those of SH1 so the two viruses are likely toshare the same particle geometry [16]

Viruses that use protein-primed replication such as bac-teriophageΦ29 carry a viral type BDNApolymerase that caninteract specifically with the proteins attached to the genomictermini and initiate strand synthesis [38] ArchaeovirusesHis1 [7] and Acidianus bottle-shaped virus [39] probablyuse this mode of replication However a polymerase genecannot be found in the genomes of PH1 SH1 or HHIV-2DNA polymerases are large enzymes and the only ORF ofSH1 that was long enough to encode such an enzyme andhad not previously been assigned as a structural protein wasORF55 which specified an 865 aa protein that contained noconserved domains indicative of polymerases [27]The recentstudy of HHIV-2 showed that the corresponding ORF in thisvirus (gene 42) is a structural protein of the virion and byinference this is likely also for the corresponding proteinsof SH1 and PH1 (no homolog is present in SNJ1) Withouta polymerase gene these viruses must use a host enzymebut searches for a viral type B DNA polymerase in thegenome of the hostHar hispanica or in the genomes of othersequenced haloarchaea did not find anymatchesThis arguesstrongly for a replication mechanism that is different to thatexemplified byΦ29 An attractive alternative is that displayedby Streptomyces linear plasmids which have 51015840-terminal pro-teins and use a cellular polymerase for replication [40] Theterminal proteins are not used for primary replication but forend patching [41] and it has been shown that if the term-inal repeat sequences are removed from these plasmids andthe ends ligated they can replicate as circular plasmids [42]This provides a testable hypothesis for the replication of halo-sphaeroviruses and also offers a pathway for switchingbetween linear (eg PH1) and circular (eg SNJ1) forms Useof a host polymerase would also fit with the ability of SH1 andPH1 DNA to transfect many different haloarchaeal species[31] as these enzymes and theirmode of action are highly con-served

The growth characteristics of PH1 inHar hispanica bothby single-cell burst and single-step growth experiments gavevalues of 50ndash100 virusescell significantly less than that ofSH1 (200 virusescell) [8] and HHIV-2 (180 virusescell) [13]The latter virus is lytic whereas SH1 PH1 and SNJ1 start pro-ducing extracellular viruswell before any decrease in cell den-sity indicating that virus release can occur without lysisThishas been most clearly shown with SH1 where almost 100of cells can be infected [8] Infection ofHar hispanica by PH1was less efficient (sim30 of cells) but the kinetics of virus pro-duction in single-step growth curves of SH1 and PH1 aresimilar with virus production occurring over several hoursrather than at a clearly defined time after infection The twoviruses are also very closely related and infect the same hostspecies so it is likely that they use the same scheme for cellexit For both viruses the cell density eventually decreases in

single-step cultures showing that virus infection does resultin cell death This mode of exit appears to maximise theproduction of the virus over time as the host cells survive foran extended period In natural hypersaline waters cell num-bers are usually high but growth rates are low [12] and thehigh incident UV (on often shallow ponds) would damagevirus DNA so reducing the half-life of released virus part-icles [43] These factors may have favoured the exit strategydisplayed by these viruses improving their chance of trans-mission to a new host

The motif GATC was absent in the genomes of PH1 SH1andHHIV-2 and themotif CTAGwas either absent or greatlyunderrepresented All three viruses share the same host butGATC and CTAG motifs are plentiful in the Har hispanicagenome (accessions CP002921 and CP002923) Avoidance ofthesemotifs in halovirusesHis1 His2 HF1 andHF2 has beenreported previously [7] Such purifying selection commonlyresults from host restriction enzymes and inHfx volcanii itis exactly these two motifs that are targeted [44] One Hfxvolcanii enzyme recognises A-methylated GATC sites [45]and another recognises un-methylated CTAG sites (whichare protected by methylation in the genome) In the currentstudy this could explain the negative transfection results forHfx volcanii as the PH1 genome contains two CTAGmotifsBy comparison the SH1 genome contains no CTAG motifsand is able to transfect Hfx volcanii [31]

Comparison of the PH1 and SH1 genomes allowed adetailed picture of the natural variation occurring betweenclosely related viruses revealing likely deletion events medi-ated by small repeats a process described previously in astudy of Hqr walsbyi and termed repeat-mediated deletion[20] There are also many replacements including blocks ofsequence that occur within long open reading frames (suchas the capsid protein genes VP1 and VP18) VP3 and VP6 areknown to form the large spikes on the external surface of theSH1 virion and presumably one or both interact with host cellreceptors [16] PH1 and SH1 have identical host ranges con-sistent with the high sequence similarity shown between theircorresponding VP3 and VP6 proteins

A previous study of SH1 could not detect transcriptsacross annotated ORFs 1ndash3 or 55-56 [30] ORFs 1 and 56occur in the terminal inverted repeat sequence and in thepresent study it was found that there were no ORFs cor-responding to these in the PH1 genome Given the close simi-larity of the two genomes the comparative data are in agree-ment with the transcriptional data indicating that these SH1ORFs are not used SH1 ORFs 2 and 3 are more problematicas good homologs of these are also present in PH1 The con-flict between the transcriptional data and the comparativegenomic evidence requires further experimental work toresolveThe status of SH1ORF55 has recently been confirmedby studies of the related virus HHIV-2 (discussed above)

The pleolipovirus group has expanded dramatically inthe last few years and it now comprises a diverse group ofviruseswith different genome types and replication strategiesWhat is evenmore remarkable is that a similar expansion cannow be seen with SH1-related viruses which includes viruseswith at least two replication modes and genome types (linearand circular) Evidence from haloarchaeal genome sequences

Archaea 13

show cases of not only provirus integrants or plasmids (egpKH2 and pHH205) but also genomic loci (ViPREs) thatcontain cassettes of virus genes from different sources andat least two cases (described in this study) have likely attsites that indicate circularisation and mobility While thereis a clear relationship between viruses with circular ds- andssDNA genomes that replicate via the rolling circle method(ie the ssDNA form is a replication intermediate) it ismore difficult to explain how capsid gene cassettes can movebetween these viruses and those with linear genomes andterminal proteins Within the pleolipoviruses His2 has alinear genome that contains a viral type B DNA polymeraseand terminal proteins while all the other described membershave circular genomes that contain a rep homolog and prob-ably replicate via the rolling circle method [14 19] Similarlyamong the halosphaeroviruses PH1 SH1 and HHIV-2 havelinear dsDNA with terminal proteins and probably replicatein the same way but the related SNJ1 virus has a circulardsDNA genome The clear relationships shown by the capsidgenes of viruses within each group plus the connectionsshown to haloviruses outside each group (eg the DNA poly-merases of His1 and His2) all speak of a vigorous meansof recombination one that can readily switch capsid genesbetween viruses with radically different replication strategiesHow is the process most likely to operate One possibility issuggested by ViPREs which appear to be mobile collectionsof capsid and replication genes from different sources Theyoffer fixed locations for recombination to occur providegene cassettes that can be reassorted to produce novel virusgenomes and some can probably recombine out as circularforms For example ViPREHmuk1 carries genes related topleolipoviruses halosphaeroviruses and other haloviruses Itis yet to be determined if any of the genes carried in theseloci are expressed in the cell (such as the genes for capsidproteins) or if ViPREs provide any selective advantage to thehost

4 Materials and Methods

41 Water Sample A water sample was collected in 1998from Pink Lake (32∘001015840 S and 115∘301015840 E) a hypersalinelake on Rottnest Island Western Australia Australia It wasscreened in the same year for haloviruses using the methodsdescribed in [8] A single plaque on a Har hispanica lawnplate was picked and replaque purified The novel haloviruswas designated PH1

42 Media Strains and Plasmids The media used in thisstudy are described in the online resource The Halohand-book (httpwwwhaloarchaeacomresourceshalohandbookindexhtml) Artificial salt water containing 30 (wv) totalsalts comprised of 4M NaCl 150mM MgCl

2 150mM

MgSO4 90mM KCl and 35mM CaCl

2and adjusted to pH

75 using sim2mL 1M Tris-HCl (pH 75) per litre MGM con-taining 12 18 or 23 (wv) total salts and HVD (halo-virus diluent) were prepared from the concentrated stock aspreviously described [46] Bacto-agar (Difco Laboratories)was added to MGM for solid (15 gL) or top-layer (7 gL)media

Table 1 lists the haloarchaea used in this study Allhaloarchaea were grown aerobically at 37∘C in either 18 or23 (wv)MGM(depending on the strain) andwith agitation(except for ldquoHar sinaiiensisrdquo)The plasmids used were pBlue-script II KS+ (Stratagene Cloning Systems) pUBP2 [47]and pWL102 [48] pUBP2 and pWL102 were first passagedthrough E coli JM110 [49] to prevent dam-methylationof DNA which has been shown to reduce transformationefficiency in some strains [45]

43 Negative-Stain Electron Microscopy The method fornegative-stain TEM was adapted from that of V Tarasovdescribed in the online resource The Halohandbook (httpwwwhaloarchaeacomresourceshalohandbookindexhtml)A 20120583L drop of the sample was placed on a clean surfaceand the virus particles were allowed to adsorb to a Formvarfilm 400-mesh copper grid (ProSciTech) for 15ndash2min Theywere then negatively stained with a 20120583L drop of 2 (wv)uranyl acetate for 1-2min Excess liquid was absorbed withfilter paper and the grid was allowed to air dry Grids wereexamined either on a Philips CM 120 BioTwin transmissionelectron microscope (Royal Philips Electronics) operating atan accelerating voltage of 120 kV or on a Siemens Elmiskop102 transmission electron microscope (Siemens AG) operat-ing at an accelerating voltage of 120 kV

44 Virus Host Range Twelve haloarchaeal strains from thegenera Haloarcula Halobacterium Haloferax HalorubrumandNatrialba (Table 1)Thirteen naturalHalorubrum isolates(H Camakaris unpublished data) and five uncharacterizedhaloarchaeal isolates fromLakeHardy Pink Lake (inWesternAustralia) and Serpentine Lake (D Walker and M K Seahunpublished data) were screened for PH1 susceptibilityLysates from PH1-infected Har hispanica cultures (1 times1011 PFUmL) were spotted onto lawns of each strain (usingmedia with 12 or 18 (wv)MGM depending on the strain)and incubated for 2ndash5 days at 30 and at 37∘C

45 Large Scale Virus Growth and Purification Liquid cul-tures of PH1 were grown by the infection (MOI 005) of anearly exponential Har hispanica culture in 18 (wv) MGMCultures were incubated aerobically at 37∘C with agitationfor 3 days Clearing (ie complete cell lysis) of PH1-infectedcultures did not occur and consequently cultures were harv-ested when the absorbance at 550 nm reached the minimumand the titre (determined by plaque assay) was at the maxi-mum usually sim1010ndash1011 PFUmL Virus was purified usingthe method described previously for SH1 [8 31] except thatafter the initial low speed spin (Sorvall GSA 6000 rpm30min 10∘C) virus was concentrated from the infected cul-ture by centrifugation at 26000 rpm (13 hr 10∘C) onto acushion of 30 (wv) sucrose in HVDThe pellet was resus-pended in a small volume of HVD and loaded onto a pre-formed linear 5ndash70 (wv) sucrose gradient followed byisopycnic centrifugation in 13 gmL CsCl (Beckman 70Ti60000 rpm 20 hr 10∘C) The white virus band occurredat a density of 129 gmL and was collected and diluted inhalovirus diluent (HVD see Section 4) and the virus pelleted(Beckman SW55 35000 rpm 75min 10∘C) and resuspended

14 Archaea

in a small volume of HVD and stored at 4∘C Virus recoveryat the major stages of a typical purification is given insupplementary Table 1 The specific infectivity of pure virussolutions was determined as the ratio of the PFUmL to theabsorbance at 260 nm

46 PH1 Single-Step Growth Curve An early exponentialphase culture of Har hispanica grown in 18 (wv) MGMwas infected with PH1 (MOI 50) Under these conditions thepercentage of infected cells was approximately 30 After anadsorption period of 1 hr at 37∘C the cells were washed threetimes with 18 (wv) MGM (at room temperature) resus-pended in 100mL 18 (wv) MGM (these methods ensuredthe removal of all residual virus) and incubated at 37∘Cwith shaking (100 rpm) Samples were removed at hourlyintervals for measurements of absorbance at 550 nm andthe number of infective centres Immediately after samplingtitres were determined by plaque assay with an indicator lawnof Har hispanica Each experiment was performed in tripli-cate

47 Halovirus Stability After various treatments sampleswere removed and diluted in HVD and virus titres weredetermined by plaque assay on Har hispanica Each experi-ment was performed in triplicate and representative data areshown Chloroform sensitivity was examined by the exposureof PH1 lysates in 18 (wv) MGM to chloroform in a volumeratio of 1 4 (chloroform to lysate) Incubation was at roomtemperature with constant agitation At appropriate timepoints the mix was allowed to settle and a sample wasremoved from the upper layer and diluted in HVDThe effectof a low ionic environment was examined by diluting PH1lysates (in 18 (wv) MGM) into double distilled H

2O in a

volume ratio of 1 1000 (lysate to double distilledH2O) Incu-

bation was at room temperature with constant agitationSamples were removed at various times and diluted in HVDThe pH stability of PH1 was determined by dilution of PH1lysates in 18 (wv)MGMin the appropriate pHbuffer (HVDbuffered with appropriate Tris-HCl) in a volume ratio of1 100 (lysate to buffer) Incubation was at room temperaturewith constant agitation After 30min samples were removedand diluted in HVD Thermal stability of PH1 was examinedby a 1 hr incubation of a virus lysate in 18 (wv) MGM atdifferent temperatures after which they were brought quicklyto room temperature diluted in HVD and titrated

48 Protein Procedures To remove salts purified virus prepa-rations were mixed with trichloroacetic acid (10 (vv) finalconcentration) and incubated on ice for 15min to allowthe proteins to precipitate After centrifugation (16000 g15min room temperature) the precipitate was washed threetimes in acetone dried and resuspended in double distilledH2O Proteins were dissolved in Laemmli sample buffer with

48mM 120573-mercaptoethanol [50] heated in boiling water for5min then separated on 12 (wv) NuPAGE Novex Bis-Tris Gels using MES-SDS running buffer according to themanufacturerrsquos directions (Invitrogen) After electrophoresisgels were rinsed in double distilled H

2O and stained with

01 (wv) Brilliant Blue G in 40 (vv) methanol and

10 (vv) acetic acid Gels were destained with severalchanges of 40 (vv) methanol and 10 (vv) acetic acidAlternatively gels were stained with GelCode GlycoproteinStaining Kit Pro-Q Emerald 488 Glycoprotein Gel and BlotStain Kit SYPRO Ruby Protein Gel Stain according tothe manufacturerrsquos directions (Invitrogen Molecular ProbesPierce Biotechnology)

Protein bands were cut from the gels and sent to theAustralian Proteome Analysis Facility (Macquarie Univer-sity) for trypsin digestion and analysis by matrix assistedlaser desorption ionisation time of flight mass spectrometry(MALDI-TOF MS) on an Applied Biosystems 4700 Pro-teomics Analyser (Applied Biosystems)

49 DNA Procedures Proteinase K-treated and nonpro-teinaseK-treatedDNApreparationsweremade frompurifiedPH1 using the methods described previously for SH1 [8]DNA was separated on 1 (wv) agarose gels in Tris-acetate-EDTA electrophoresis buffer and was stained with ethidiumbromide (Sigma-Aldrich)120582 DNA DNase I (RNase-free) exonuclease III mung

bean nuclease nuclease BAL-31 T4 DNA polymerase T4DNA ligase T7 exonuclease and type II restriction endonu-cleases were purchased from New England Biolabs RNase Aand yeast transfer RNA were purchased from Sigma-AldrichProteinase K was purchased from Promega Oligonucleotideprimers were purchased from Geneworks Australia

To clone fragments of the PH1 genome purified virusDNA was digested either with AccI Eco0109I and MseI orSmaI restriction endonucleases blunted with DNA Poly-merase I Large (Klenow) Fragment where appropriate andligated into AccI- Eco0109I- or SmaI-digested pBluescript IIKS+ using T4 DNA ligase according to the manufacturerrsquosinstructions (New England Biolabs) The DNA was intro-duced intoE coliXL1-Blue and transformants grown onLuriaagar containing 15 120583gmL tetracycline and 100 120583gmL ampi-cillinThe resulting cloneswere sequenced and the sequenceswere used to design specific oligonucleotide primers for PCRamplification andor primer walking using the virus genome(or specific restriction fragments) as templates

To amplify PH1 nucleic acid approximately 10 ng virusDNA was combined with 500 nM primers 100120583M of eachdNTP 1U Deep Vent DNA Polymerase (New England Bio-labs) and 1 timesThermoPol buffer (New England Biolabs) ina total reaction volume of 50 120583L Template was denatured at95∘C for 10min followed by 30 cycles of denaturation at 95∘Cfor 30 sec annealing at 56∘C for 30 sec extension at 75∘C for2min and a final extension at 75∘C for 10min PCR reactionswere performed on a PxE02ThermoCycler (Thermo ElectroCorporation) Sequencing reactions using 32 pmol primerwere performed by the dideoxy chain termination methodusing ABI PRISM Big Dye Terminator Mix version 31 onan ABI 3100 capillary sequencer (Applied Biosystems) bythe Applied Genetic Diagnostics Sequencing Service at theDepartment of Pathology (The University of Melbourne)

To obtain terminal genomic fragments PH1 DNA wasdigested by the enzyme Pas1 and the head (sim700 bp) andthe tail (sim1300 bp) fragments were purified by the QIAEXII gel extraction kits and treated with 05M piperidine for

Archaea 15

2 hr at 37∘C to remove protein residues The piperidine-treated fragments were sequenced by the Genomics BiotechCompany (Taipei Taiwan) using two primers TGACCA-ATTAATTAGGCCGGTTCGCC (PH1 head-R) and GTGC-CATACTGCTACAATTCT (PH1 tail-F)

Four hundred fifty-four whole genome sequencing PH1DNA samples (sim5 120583g) were sequenced using parallel pyrose-quencing on a Roche 454 Genome Sequencer System atMission Biotech (Taipei Taiwan) The largest contig was27399with 8803 reads (Newbler version 27) with 27387 posi-tions being of Q40 quality The average depth of coverage foreach base was 829TheGenBank accession for the entire PH1sequence is KC252997

Acknowledgments

The authors thank Carolyn Bath Mei Kwei (Joan) Seahand Danielle Walker for their early work in PH1 characteri-zation Simon Crawford Jocelyn Carpenter and Anna Fried-huber for their assistance with the transmission electronmicroscopy and Tiffany Cowie and Voula Kanellakis fortheir sequencing assistance This paper has been facilitatedby access to the Australian Proteome Analysis Facility estab-lished under the Australian Governmentrsquos Major NationalResearch Facilities program SLT was supported by a grantfrom the Biodiversity Research Center Academia SinicaTaiwan

References

[1] M Krupovic M F White P Forterre and D PrangishvilildquoPostcards from the edge structural genomics of archaealvirusesrdquo Advances in Virus Research vol 82 pp 33ndash62 2012

[2] P Stolt andW Zillig ldquoGene regulation in halophage 120601Hmdashmorethan promotersrdquo Systematic and Applied Microbiology vol 16no 4 pp 591ndash596 1994

[3] R Klein U Baranyi N Rossler B Greineder H Scholz and AWitte ldquoNatrialba magadii virus 120593Ch1 first complete nucleotidesequence and functional organization of a virus infecting ahaloalkaliphilic archaeonrdquo Molecular Microbiology vol 45 no3 pp 851ndash863 2002

[4] E Pagaling R D Haigh W D Grant et al ldquoSequence analysisof an Archaeal virus isolated from a hypersaline lake in InnerMongolia Chinardquo BMC Genomics vol 8 article 410 2007

[5] S L Tang SNuttall andMDyall-Smith ldquoHalovirusesHF1 andHF2 evidence for a recent and large recombination eventrdquo Jour-nal of Bacteriology vol 186 no 9 pp 2810ndash2817 2004

[6] C Bath and M L Dyall-smith ldquoHis1 an archaeal virus of theFuselloviridae family that infectsHaloarcula hispanicardquo Journalof Virology vol 72 no 11 pp 9392ndash9395 1998

[7] C Bath T Cukalac K Porter andM L Dyall-Smith ldquoHis1 andHis2 are distantly related spindle-shaped haloviruses belongingto the novel virus group Salterprovirusrdquo Virology vol 350 no1 pp 228ndash239 2006

[8] K Porter P Kukkaro J K H Bamford et al ldquoSH1 a novelspherical halovirus isolated from an Australian hypersalinelakerdquo Virology vol 335 no 1 pp 22ndash33 2005

[9] M Dyall-Smith S L Tang and C Bath ldquoHaloarchaeal viruseshow diverse are theyrdquo Research in Microbiology vol 154 no 4pp 309ndash313 2003

[10] A Oren G Bratbak and M Heldal ldquoOccurrence of virus-likeparticles in the Dead Seardquo Extremophiles vol 1 no 3 pp 143ndash149 1997

[11] T Sime-Ngando S Lucas A Robin et al et al ldquoDiversityof virus-host systems in hypersaline Lake Retba SenegalrdquoEnvironmental Microbiology vol 13 no 8 pp 1956ndash1972 2010

[12] N Guixa-Boixareu J I Calderon-Paz M Heldal G Bratbakand C Pedros-Alio ldquoViral lysis and bacterivory as prokaryoticloss factors along a salinity gradientrdquoAquaticMicrobial Ecologyvol 11 no 3 pp 215ndash227 1996

[13] S T Jaakkola R K Penttinen S T Vilen et al ldquoClosely relatedarchaeal Haloarcula hispanica icosahedral viruses HHIV-2 andSH1 have nonhomologous genes encoding host recognitionfunctionsrdquo Journal of Virology vol 86 no 9 pp 4734ndash47422012

[14] E Roine P Kukkaro L Paulin et al ldquoNew closely related halo-archaeal viral elements with different nucleic acid typesrdquo Jour-nal of Virology vol 84 no 7 pp 3682ndash3689 2010

[15] A Sencilo L Paulin S KellnerMHelm and E Roine ldquoRelatedhaloarchaeal pleomorphic viruses contain different genometypesrdquo Nucleic Acids Research vol 40 no 12 pp 5523ndash55342012

[16] H T Jaalinoja E Roine P Laurinmaki H M Kivela D HBamford and S J Butcher ldquoStructure and host-cell interactionof SH1 a membrane-containing halophilic euryarchaeal virusrdquoProceedings of the National Academy of Sciences of the UnitedStates of America vol 105 no 23 pp 8008ndash8013 2008

[17] M K Pietila N S Atanasova VManole et al ldquoVirion architec-ture unifies globally distributed pleolipoviruses infecting halo-philic archaeardquo Journal of Virology vol 86 no 9 pp 5067ndash50792012

[18] M L Holmes and M L Dyall-Smith ldquoA plasmid vector witha selectable marker for halophilic archaebacteriardquo Journal ofBacteriology vol 172 no 2 pp 756ndash761 1990

[19] M L Holmes F Pfeifer andM L Dyall-Smith ldquoAnalysis of thehalobacterial plasmid pHK2 minimal repliconrdquo Gene vol 153no 1 pp 117ndash121 1995

[20] M L Dyall-Smith F Pfeiffer K Klee et al ldquoHaloquadratumwalsbyi limited diversity in a Global Pondrdquo PLoS ONE vol 6no 6 Article ID e20968 2011

[21] S Casjens ldquoProphages and bacterial genomics what have welearned so farrdquoMolecular Microbiology vol 49 no 2 pp 277ndash300 2003

[22] R W Hendrix M C M Smith R N Burns M E Ford andG F Hatfull ldquoEvolutionary relationships among diverse bacte-riophages and prophages all the worldrsquos a phagerdquo Proceedings ofthe National Academy of Sciences of the United States of Americavol 96 no 5 pp 2192ndash2197 1999

[23] D Botstein ldquoA theory ofmodular evolution for bacteriophagesrdquoAnnals of the New York Academy of Sciences vol 354 pp 484ndash491 1980

[24] M Krupovic S Gribaldo D H Bamford and P Forterre ldquoTheevolutionary history of archaeal MCM helicases a case study ofvertical evolution combinedwithHitchhiking ofmobile geneticelementsrdquo Molecular Biology and Evolution vol 27 no 12 pp2716ndash2732 2010

[25] M Krupovic P Forterre and D H Bamford ldquoComparativeanalysis of the mosaic genomes of tailed archaeal viruses andproviruses suggests common themes for virion architecture andassembly with tailed viruses of bacteriardquo Journal of MolecularBiology vol 397 no 1 pp 144ndash160 2010

16 Archaea

[26] Z Zhang Y Liu S Wang et al et al ldquoTemperate membrane-containing halophilic archaeal virus SNJ1 has a circular dsDNAgenome identical to that of plasmid pHH205rdquoVirology vol 434no 2 pp 233ndash241 2012

[27] D H Bamford J J Ravantti G Ronnholm et al ldquoConstituentsof SH1 a novel lipid-containing virus infecting the halophiliceuryarchaeonHaloarcula hispanicardquo Journal of Virology vol 79no 14 pp 9097ndash9107 2005

[28] H M Kivela E Roine P Kukkaro S Laurinavicius P Somer-harju andDH Bamford ldquoQuantitative dissociation of archaealvirus SH1 reveals distinct capsid proteins and a lipid corerdquoVirology vol 356 no 1-2 pp 4ndash11 2006

[29] H Jaalinoja Electron Cryo-Microscopy Studies of Bacteriophage8 and Archaeal Virus SH1 University of Helsinki HelsinkiFinland 2007

[30] K Porter B E Russ J Yang andM L Dyall-Smith ldquoThe trans-cription programme of the protein-primed halovirus SH1rdquoMi-crobiology vol 154 no 11 pp 3599ndash3608 2008

[31] K Porter and M L Dyall-Smith ldquoTransfection of haloarchaeaby the DNAs of spindle and round haloviruses and the useof transposon mutagenesis to identify non-essential regionsrdquoMolecular Microbiology vol 70 no 5 pp 1236ndash1245 2008

[32] S E Luria General Virology John Wiley and Sons New YorkNY USA 1953

[33] C A Thomas K Saigo E McLeod and J Ito ldquoThe separationofDNA segments attached to proteinsrdquoAnalytical Biochemistryvol 93 pp 158ndash166 1979

[34] A L Delcher D Harmon S Kasif OWhite and S L SalzbergldquoImproved microbial gene identification with GLIMMERrdquoNucleic Acids Research vol 27 no 23 pp 4636ndash4641 1999

[35] R Zhang Y Yang P Fang et al ldquoDiversity of telomere palin-dromic sequences and replication genes among Streptomyceslinear plasmidsrdquo Applied and Environmental Microbiology vol72 no 9 pp 5728ndash5733 2006

[36] X Wang H S Lee F J Sugar F E Jenney M W W Adamsand J H Prestegard ldquoPF0610 a novel winged helix-turn-helix variant possessing a rubredoxin-like Zn ribbon motiffrom the hyperthermophilic archaeon Pyrococcus furiosusrdquoBiochemistry vol 46 no 3 pp 752ndash761 2007

[37] S W Cline L C Schalkwyk and W F Doolittle ldquoTransfor-mation of the archaebacterium Halobacterium volcanii withgenomic DNArdquo Journal of Bacteriology vol 171 no 9 pp 4987ndash4991 1989

[38] S Kamtekar A J Berman J Wang et al ldquoThe 12059329 DNA poly-merase protein-primer structure suggests a model for the ini-tiation to elongation transitionrdquo EMBO Journal vol 25 no 6pp 1335ndash1343 2006

[39] X Peng T Basta M Haring R A Garrett and D PrangishvilildquoGenome of theAcidianus bottle-shaped virus and insights intothe replication and packaging mechanismsrdquo Virology vol 364no 1 pp 237ndash243 2007

[40] C C Yang C H Huang C Y Li Y G Tsay S C Lee and CW Chen ldquoThe terminal proteins of linear Streptomyces chrom-osomes and plasmids a novel class of replication primingproteinsrdquo Molecular Microbiology vol 43 no 2 pp 297ndash3052002

[41] C C Yang Y H Chen H H Tsai C H Huang T W Huangand C W Chen ldquoIn vitro deoxynucleotidylation of the term-inal protein of Streptomyces linear chromosomesrdquo Applied andEnvironmental Microbiology vol 72 no 12 pp 7959ndash79612006

[42] P C Chang and S N Cohen ldquoBidirectional replication froman internal origin in a linear Streptomyces plasmidrdquo Science vol265 no 5174 pp 952ndash954 1994

[43] S W Wilhelm W H Jeffrey C A Suttle and D L MitchellldquoEstimation of biologically damaging UV levels in marinesurface waters with DNA and viral dosimetersrdquo Photochemistryand Photobiology vol 76 no 3 pp 268ndash273 2002

[44] T Allers and M Mevarech ldquoArchaeal geneticsmdashthe third wayrdquoNature Reviews Genetics vol 6 no 1 pp 58ndash73 2005

[45] M L Holmes S D Nuttall and M L Dyall-Smith ldquoConstruc-tion and use of halobacterial shuttle vectors and further studieson Haloferax DNA gyraserdquo Journal of Bacteriology vol 173 no12 pp 3807ndash3813 1991

[46] S D Nuttall and M L Dyall-Smith ldquoHF1 and HF2 novelbacteriophages of halophilic archaeardquo Virology vol 197 no 2pp 678ndash684 1993

[47] U Blaseio and F Pfeifer ldquoTransformation of Halobacteriumhalobium development of vectors and investigation of gas vesi-cle synthesisrdquo Proceedings of the National Academy of Sciences ofthe United States of America vol 87 no 17 pp 6772ndash6776 1990

[48] W L Lam and W F Doolittle ldquoShuttle vectors for the archae-bacterium Halobacterium volcaniirdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 86 no14 pp 5478ndash5482 1989

[49] C Yanisch-Perron J Vieira and J Messing ldquoImproved M13phage cloning vectors and host strains nucleotide sequences ofthe M13mp18 and pUC19 vectorsrdquo Gene vol 33 no 1 pp 103ndash119 1985

[50] U K Laemmli ldquoCleavage of structural proteins during theassembly of the head of bacteriophage T4rdquo Nature vol 227 no5259 pp 680ndash685 1970

[51] D G Burns H M Camakaris P H Janssen and M L Dyall-Smith ldquoCombined use of cultivation-dependent and culti-vation-independent methods indicates that members of mosthaloarchaeal groups in anAustralian crystallizer pond are culti-vablerdquo Applied and Environmental Microbiology vol 70 no 9pp 5258ndash5265 2004

[52] G Juez F Rodriguez-Valera A Ventosa and D J KushnerldquoHaloarcula hispanica spec nov and Haloferax gibbonsii specnov two new species of extremely halophilic archaebacteriardquoSystematic and Applied Microbiology vol 8 pp 75ndash79 1986

[53] A Oren M Ginzburg B Z Ginzburg L I Hochstein and BE Volcani ldquoHaloarcula marismortui (Volcani) sp nov nomrev an extremely halophilic bacterium from the Dead SeardquoInternational Journal of Systematic Bacteriology vol 40 no 2pp 209ndash210 1990

[54] M Torreblanca F Rodriguez-Valera G Juez A Ventosa MKamekura and M Kates ldquoClassification of non-alkaliphilichalobacteria based on numerical taxonomy and polar lipidcomposition and description of Haloarcula gen nov andHaloferax gen novrdquo Systematic and Applied Microbiology vol8 pp 89ndash99 1986

[55] A Ventosa and A Oren ldquoHalobacterium salinarum nom cor-rig a name to replace Halobacterium salinarium (Elazari-Vol-cani) and to include Halobacterium halobium and Halobac-terium cutirubrumrdquo International Journal of Systematic Bacte-riology vol 46 no 1 p 347 1996

[56] M C Gutierrez M Kamekura M L Holmes M L Dyall-Smith and A Ventosa ldquoTaxonomic characterization of Halo-ferax sp (ldquoH alicanteirdquo) strain Aa 22 description of Haloferaxlucentensis sp novrdquo Extremophiles vol 6 no 6 pp 479ndash4832002

Archaea 17

[57] M F Mullakhanbhai and H Larsen ldquoHalobacterium volcaniispec nov a Dead Sea halobacterium with a moderate saltrequirementrdquo Archives of Microbiology vol 104 no 1 pp 207ndash214 1975

[58] S D Nuttall and M L Dyall-Smith ldquoCh2 a novel halophilicarchaeon from an Australian solar salternrdquo International Jour-nal of Systematic Bacteriology vol 43 no 4 pp 729ndash734 1993

[59] P D Franzman E Stackebrandt K Sanderson et al ldquoHalobac-terium lacusprofundi sp nov a halophilic bacterium isolatedfrom Deep Lake Antarcticardquo Systematic and Applied Microbi-ology vol 11 pp 20ndash27 1988

[60] G A Tomlinson and L I Hochstein ldquoHalobacterium saccharo-vorum sp nov a carbohydrate metabolizing extremely halo-philic bacteriumrdquoCanadian Journal ofMicrobiology vol 22 no4 pp 587ndash591 1976

[61] A Ventosa M C Gutierrez M Kamekura and M L Dyall-Smith ldquoProposal to transfer Halococcus turkmenicus Halobac-terium trapanicum JCM9743 and strainGSL-11 toHaloterrigenaturkmenica gen nov comb novrdquo International Journal ofSystematic Bacteriology vol 49 no 1 pp 131ndash136 1999

[62] M Kamekura and M L Dyall-Smith ldquoTaxonomy of the familyHalobacteriaceae and the description of two new genera Halo-rubrobacterium and Natrialbardquo Journal of General and AppliedMicrobiology vol 41 no 4 pp 333ndash350 1995

[63] T J Carver KM RutherfordM BerrimanM A RajandreamB G Barrell and J Parkhill ldquoACT the Artemis comparisontoolrdquo Bioinformatics vol 21 no 16 pp 3422ndash3423 2005

[64] S W Cline W L Lam R L Charlebois L C Schalkwyk andW F Doolittle ldquoTransformationmethods for halophilic archae-bacteriardquo Canadian Journal of Microbiology vol 35 no 1 pp148ndash152 1989

4 Archaea

0 20 40 60 80 100

Titre

(PFU

mL)

Temperature1011

1010

109

108

107

106

Temperature (∘C)

le105

(a)

0 5 10 15

0

Time (min)

Salt concentration

Titre

(PFU

mL)

107

106

105

104

minus5

(b)

4 5 6 7 8 9 10pH

pH

Titre

(PFU

mL)

108

107

(c)

Exposure time (min)0 5 10 15 20

Chloroform

Titre

(PFU

mL)

108

107

(d)

Figure 3 Stability of PH1 virus to various treatments and conditions Virus preparations (infected-cell supernatants in 18 (wv)MGM)wereexposed to various conditions after which the virus titre was determined (in duplicate) onHar hispanica cells (a) The effect of temperatureVirus was incubated for 1 hr with constant agitation at temperatures between 4 and 100∘C (b) The effect of lowered salt concentration Viruswas diluted 1 1000 in double-distilled H

2

O and incubated at room temperature with constant agitation Samples were removed at regularintervals (c)The effect of pHViruswas diluted 1 100 inTris-HCl buffers at the different pHs and incubatedwith constant agitation for 30min(d)The effect of chloroformChloroformwasmixedwith virus (1 4 ratio) and incubated at room temperaturewith constant agitation Sampleswere removed at regular intervals Open square symbols indicate where virus titres were undetectable Error bars represent one standarddeviation of the average titre

Archaea 5

Table 2 PH1 virus proteins identified by mass spectroscopy of tryptic peptides

Proteina Locus tag ORF MW (kDa) Matching peptide massesbObserved (calculated)

VP1 HhPH1 gp12 12 185 (158) 16VP2 HhPH1 gp24 24 100 (78) 15VP3 HhPH1 gp28 28 40 (38) 5VP4 HhPH1 gp21 21 35 (26) 5VP7 HhPH1 gp20 20 24 (20) 5VP9 HhPH1 gp27 27 16 (17) 3VP10 HhPH1 gp26 26 15 (17) 4VP12 HhPH1 gp19 19 7 (99) 4aVirus capsid proteins are numbered according to their similarity with SH1 capsid proteinsbThenumber of peptidemasses between 7500 and 35130mz identified byMALDI-TOFMS that correspond to the theoretical tryptic peptides of the predictedvirus protein

212 -

(kD

a)

PH1

158 -97 -66 -56 -42 -35 -

27 -

20 -

14 -

65 -

lowast

lowast

lowast

lowast

lowast

lowast

lowast

VP1VP2

VP3VP4

VP7

VP9

VP10

VP12

(a)

SH1

(b)

Figure 4 Structural proteins of halovirus PH1 Viral structuralproteinswere separated by SDS-PAGEon a 12 (wv) acrylamide geland stained with Brilliant Blue G (a) They were run in parallel withthe proteins of purified SH1 virus (b) The sizes of protein standardmarkers are indicated on the left side (in kDa) Asterisks denoteproteins bands of PH1 that were identified by mass spectroscopyThe numbering of proteins of PH1 (VP1ndashVP12) follows that of theSH1 homologs seen in (b) (and explained also in the text)

staining to be detected Further evidence was obtained by anuclease protection assay as previously used for SH1 DNA[31] As shown in Table 3 exonuclease III (a 31015840 exonuclease)was able to digest nonproteinase K-treated PH1 DNA butT7 exonuclease (a 51015840 exonuclease) and Bal31 (51015840 and 31015840exonucleases) were unable to digest PH1 DNA unless it hadbeen previously treated with proteinase KThis indicated thatboth 51015840 termini of the genome had protein attached

27 Sequence of the PH1 Genome The complete PH1 genomesequence was determined (accession KC252997) using a

1 2 3

-23

-76

-17410 -

6 -

3 -

2 -

15 -

05 -

4 -

(kB)

Figure 5 Detection of proteins bound to the termini of PH1 DNADNA was extracted from purified PH1 particles without proteinaseK digested with AseI then passed through a silica filter (GFC)under conditionswhere proteinswould bind to the filter Lane 1AseIfragments that did not bind and were eluted Lane 2 AseI fragmentsthat bound and were eluted only after protease treatment Lane 3AseI digest of PH1 DNA The positions of DNA size standards areindicated on the left side in kb The calculated sizes of the AseIfragments of PH1 DNA in lane 3 are indicated on the right side (alsoin kb) The predicted 062 kb PH1 AseI fragment was not detected

combination of cloned fragments PCR primer walkingon virus DNA and 454 whole genome sequencing (seeSection 4) It was found to be 28072 bp in length 676G + C with inverted terminal repeat sequences (ITRs) of337 bp Using GLIMMER [34] manual BLAST searching atthe GenBank database and comparison to related virusesthe PH1 genome sequence was predicted to contain 49 ORFs(Table 4) Most ORFs were closely spaced or overlappinggiving a gene density of 174 geneskb (average of 572 ntgene)The cumulative AT-skew plot of the PH1 genome shown atthe bottom of Figure 6 shows inflection points (circled) thatare consistent with the changes in transcription direction

6 Archaea

6 8 10 12 14 16 18 20 22 24 kb

1 2 4 6 8 10 12 14 162 kb

VP1

VP1-like fragments

A

A

VP2

VP2

VP7

VP1

2

20 21 22

VP4

VP1

3

VP5 p2

2

p9p4 p11

p12

p13

NP5

356A

Hmuk

p26

VP3

VP1

7

1728Hmuk

0477

Locus tag ZOD2009 19038

Locus tag HlacAJ 010100019872 NZ AGFZ01000123 (reversed)

NZ AEMG01000027 contig29

TransposaseHlacAJ 010100019767

ZOD2009 19153

Hbf lacisalsi

Hap paucihalophilus

(a)

VP18VP2VP1 A

VP1

VP18VP1

VP1

2

VP1

VP7

VP4

VP1

3

VP5

VP2

VP9

VP1

0

VP3

VP1

6

VP1

7V

P6 VP18

ATPa

se

kb1 2 4 6 8 10 12 14 16 18 20 22 24 26 28

VP2

VP2

VP18A

AA

2211 1312

94 26

HHIV-2

SH1

PH1

0

2

4

6

30 578 nt

30 889 nt

28 072 nt

(times102

)Cu

m A+

T sk

ew

Cumulative A + T skew of PH1

(b)

Figure 6 Genome alignments of haloviruses PH1 SH1 and HHIV-2 along with two related genomic loci (a) Genomic loci of Happaucihalophilus andHbf lacisalsi that contain genes related to halosphaeroviruses SH1 PH1 andHHIV-2The names andGenBank accessionsfor these contigs are given on the far right andORFs are coloured and labeled to indicate the relationships of theseORFs to those of the virusesbelowThe locus tag numbers for the first and last ORFs shown in each locus are given nearby their respectiveORFs In addition grey colouredORFs represent sequences that do not match any of the haloviruses and green coloured ORFs represent protein sequences that are closelyrelated to ORFs found within or very close to previously described virusplasmid loci so called ViPREs [20] The scale bars shown aboveeach contig show the position of the described region within the respective contig (b) The three virus genomes are labeled at the left withscale markers below (in kb) and the total length indicated at the far right At the bottom is a cumulative AT-skew plot of the PH1 genome(httpmolbiol-toolscaJie Zheng) with inflection points circledThe grey shaded bands between the genome diagrams indicate significantnucleotide similarity (usingACT [63]) AnnotatedORFs are represented by arrows with colours indicating structural proteins (red or brown)nonstructural proteins (yellow or orange) or the packaging ATPase (blue) The names of structural protein ORFs are indicated either withinthe arrow (eg VP1) or in text nearby The numbered orange coloured ORFs of HHIV-2 are homologous to ORFs found in the genomic loci(probably proviruses or provirus remnants) pictured in (a)

indicated by the annotated ORFs as has been seen in otherhaloviruses [7] There are no GATC motifs in the genomeand the inverse sequence CTAG is strongly underrepre-sented with just 2 motifs (68 expected)

When compared to the related viruses SH1 and HHIV-2(Figure 6) the PH1 genome is seen to be of similar length andis much more closely related to SH1 than to HHIV-2 (74and 54 nt identity resp) Like PH1 the genomes of SH1and HHIV-2 lack GATC motifs and CTAG is either absent(SH1) or underrepresented (HHIV-2) The ITRs of PH1 andSH1 share 785nucleotide identity but the PH1 ITR is longerthan that of SH1 (337 and 309 nt resp) partly because SH1has replaced an 18 bp sequence (gtcgtgcggtttcggcgg) found atthe internal end of its left-hand ITR with a sequence at thecorresponding position of its right-hand ITR that shows littleinverted sequence similarity In the PH1 ITR this sequenceis retained at both ends Whether this difference represents arecombination event or a mistake in replication of the ITRs

is unclear An alignment of the ITRs of all three virusesidentified a number of highly conserved regions (labeled 1ndash9 in Supplementary Figure 2 see Supplementary Materialavailable online at httpdxdoiorg1011552013456318) A16 bp sequence at the termini (region 1) is conserved con-sistent with the conservation seen at the termini of linearStreptomyces plasmids [35] A 16 bpGC-rich sequence aroundthe middle of the ITR (region 3) is also conserved Shortermotifs of either C-rich (region 8) or AT-rich sequence(regions 7 and 9) are found near the internal end of the ITRs

The grey shading between the schematic virus genomesin Figure 6(b) indicates regions of high nucleotide similarityrevealing that the differences between PH1 and SH1 are notevenly distributed but vary considerably along the lengthof the aligned genomes A comparison between SH1 andHHIV-2 has been published recently [13] and since the PH1and SH1 genomes are so similar we will focus the followingdescription on the differences between PH1 and SH1

Archaea 7

Table 3 Digestion of the PH1 genome by nucleasesa

Nuclease (amount) Proteinase K-treated UntreatedControl PH1 DNA Control PH1 DNA

DNase I (RNase-free)(100U) +b + +b +

Exonuclease III(100U) +b + +b +

Mung bean nuclease(10U) +c

minus +cminus

Nuclease BAL-31(1 U) +b + +b

minus

RNase A(5 120583gmL) +d

minus +dminus

T7 exonuclease(10U) +b + +b

minus

aExtracted nucleic acid from purified PH1 virus was treated with variousnucleases and then analysed by gel electrophoresis to detect whether thenuclease digested (+) or did not digest (minus) the genome Control nucleicacids used to confirm activity of the nucleases were b

120582 DNA ca DNAoligonucleotide and dyeast transfer RNA

Within corresponding ORFs there are blocks of codingsequence with low (or no) similarity These may represent insitu divergence or short recombination events (eg indels)There are also cases where an entire ORF in one virus hasno corresponding homolog in the other because of an inser-tiondeletion event (indel) Lastly there are replacementswhere the sequences within the corresponding regions showlow or no similarity but are flanked by sequence with highsimilarity The largest visible differences seen in Figure 6 aredue to sequence changes within or near the long genes encod-ing capsid proteins VP1 and VP18 A summary of the mainsequence differences between SH1 and PH1 and their loca-tions is given in Table 5 where these divergent regions (DV)are numbered from DV1 to DV18

The sequences of the two viruses are close enough that inmany cases the mechanism for the observed differences canbe inferred For example DV2 is likely the result of a deletionevent that has removed the SH1 ORF10 gene homolog fromthe PH1 genome At either end of the SH1 ORF are almostperfect direct repeats (cggcctgaccggcatgac) that would allowrepeat-mediated deletion to occur removing the interveningsequence Small direct repeats leading to deletions have beendescribed previously in the comparative analysis of Halo-quadratumwalsbyi genomes [20] DV5 appears to be an indelwhere SH1 ORFs 14ndash16 are absent in PH1 and an invertedrepeat (AGCCATG) found at each end of the SH1 divergentregion may be significant in the history of this changeThe proteins specified by SH1 ORFs 14ndash16 are presumablydispensable for PH1 (or their functions supplied by otherproteins) but a homolog of SH1 ORF14 is found also inHHIV-2 (ORF 6) and a homolog of SH1 ORF15 occurs inhalovirus His1 (ORF13) Replacement regions are also com-mon and can occur within ORFs (eg DV3 in capsid proteinVP1) or provide additional or alternative genes (eg DV12and DV13) Several divergent regions occur within the genefor capsid protein VP2 a hot spot for change because of the

repetitive nature of the coding sequence which specifies aprotein with runs of glycines andmany heptapeptide repeatsas described previously for SH1 VP2 [27] DV18 is a largereplacement that significantly alters the predicted length ofthe minor capsid protein VP18 in the two viruses (865519aa for SH1PH1 resp) and also provides SH1 with 3 ORFsthat are different or absent in PH1 ORFs 52 53 and 54 Forexample a homolog of SH1 ORF52 is not present in PH1 butis present in HHIV-2 (putative protein 40) While ORF48 ofPH1 shows no aa sequence homology to SH1 ORFs 53 and 54all of these predicted proteins carry CxxC motifs suggestiveof related functions such as DNA binding activity [36]

The genes of the PH1 genome are syntenic with those ofSH1 and are found in similarly oriented blocks of closelyspaced or overlapping ORFs that suggest that transcription isorganised into operons The programme of transcription ofthe SH1 genome has been reported previously [30] and thesequences in PH1 corresponding to the six promoter regions(P1ndashP6) determined in SH1 showed that five of these (P1ndashP5)are well conserved Only the region corresponding to SH1 P6was poorly conserved (data not shown) but this promoter isstrongly regulated in SH1 and only switches on late in infec-tion [30]

28 Structural Protein Genes The genes coding for the majorvirus structural proteins of PH1 (VP 1ndash4 7 9 10 and 12)were identified byMALDI-TOF (see above) and the genes forthese proteins are found in the same order and approximatepositions as in the SH1 genome (Figure 6(b) red ORFs)Genes coding for minor structural proteins can be deducedby sequence comparison with SH1 so that VP5 and VP6 arelikely to be encoded by PH1ORFs 25 and 29 respectively andVP13 and VP18 by ORFs 23 and 49 respectively This wouldgive PH1 a total of 11 structural proteins the same as SH1Of these VP7 is the most conserved between the two viru-ses (98 aa identity) followed by VP4 VP9 VP3 and VP12(91ndash94 identity) VP5 VP6 and VP13 (82ndash88 iden-tity) VP2 (77) VP1 (65) and lastly the least conservedprotein was VP18 (31)

29 Genomic Loci Related to PH1 and Other Halosphaeroviru-ses Clusters of halosphaerovirus-related genes are present inthe genomes of two recently sequenced haloarchaea Happaucihalophilus and Hbf lacisalsi (Figure 6(a)) Neither ofthese regions appears to represent complete virus genomebut they retain a number of genes that share a similar seq-uence and synteny with the virus genomes depicted belowthem (Figure 6(a)) Both loci show a mixed pattern ofrelatedness to SH1PH1 and HHIV-2 as indicated by ORFsof matching colour and gene name If these genomic locirepresent provirus integrants that have decayed over timethen the relationships they show suggest not only that halo-sphaeroviruses are a diverse virus group but also that asignificant level of recombination occurs between them lead-ing to mosaic gene combinations The two ORFs colouredgreen in theHap paucihalophilus locus (Figure 6) are relatedto ORFs found within or very close to genomic loci ofvirusplasmid genes found in other haloarchaea suggestingadditional links between (pro)viruses [20]

8 Archaea

Table4ORF

anno

tatio

nsof

theh

aloviru

sPH1g

enom

e

ORF

aPo

sitionb

Locustag

Leng

thc(aa)

pKi

Similaritycharacteris

tics

144

1ndash605

HhP

H1gp01

5043

76aa

similarityto

SH1O

RF2(YP271859)

2602ndash802

HhP

H1gp02

66109

80aa

similarityto

SH1O

RF3(YP271860)

3802ndash1059

HhP

H1gp03

8539

85aa

similarityto

SH1O

RF4(YP271861)44

sim

ilarityto

HHIV-2

protein1

41056ndash1283

HhP

H1gp04

7539

85aa

identityto

SH1O

RF5(YP271862)Predictedcoiled-coilregion

51276ndash1686

HhP

H1gp05

136

49

80aa

similarityto

SH1O

RF6(YP271863)Tw

otransm

embraned

omainsA

lsorelated

toHHIV-2

protein2(38

AFD

02283)

andto

Halobiform

alacisalsiORF

ZP09950018

(35

)6

1683ndash1829

HhP

H1gp06

4861

81aa

similarityto

SH1O

RF7(YP271864

)Predictedsig

nalsequence(Sign

alP)

71822ndash2130

HhP

H1gp07

102

58

91aa

similarityto

SH1O

RF8(YP271865)a

nd86to

HHIV-2

putativ

eprotein

3(A

FD02284)O

ther

homologsinclude

Nocardiaproteinpn

f2110(YP1220601)H

rrlacusprofund

iHlac0751

(YP0025654211)ORF

sinactin

ophage

VWB

(AAR2

9707)Frankia(EAN116

57)andgp40

ofMycobacteriu

mph

ageD

ori(AER

47690)

82123ndash2383

HhP

H1gp08

8642

90aa

similarityto

SH1O

RF9(YP271866)

92380ndash2714

HhP

H1gp09

111

52

88aa

similarityto

SH1O

RF11(YP271868)and

also

similarityto

putativ

eprotein

4of

HHIV-2

(AFD

02285)

HlacA

J19877of

Hbflacisa

lsiAJ

5(ZP09950016)

102861ndash300

4HhP

H1gp10

47104

Weaksim

ilarityto

SH1O

RF12

(YP271869)o

nlyover

theN

-term

inal18

resid

ues

113029ndash3100

HhP

H1gp11

23123

Not

presentinSH

1orH

HIV-2

123134ndash7453

HhP

H1gp12

1439

42

Capsid

protein

VP172sim

ilarityto

SH1V

P1(O

RF13Y

P271870)HHIV-2

VP1

(AFD

02286)H

lacA

J19872Hbflacisa

lsiAJ

5Predictedhelix-tu

rn-helixandRu

vA-like

domain(In

terproScan)

137505ndash8227

HhP

H1gp13

240

54

Predicted

P-loop

ATPa

sedo

main(C

OG0433)92aa

similarityto

SH1O

RF17

(YP271874)a

ndalso

similarityto

putativ

eAT

Pase

ofHHIV-2

(AFD

02288)A

TPaseo

fHaladapatus

paucihalophilusD

X253

(ZP0804

6180)HlacA

J19867of

Hbf

lacisalsi(ZP09950014)

148419ndash8228c

HhP

H1gp14

6351

43aa

similarityto

SH1O

RF18

(YP271875)Alanine-richCentraltransm

embraned

omain(Pho

bius)

158856ndash8

416c

HhP

H1gp15

146

41

92aa

similarityto

SH1O

RF19

(YP271876)a

ndalso

toHHIV-2

putativ

eprotein

9(A

FD02290)H

lacA

J19857of

Hbf

lacisalsiAJ

5ZO

D2009

19093of

Happau

cihalophilus

168853ndash9

500c

HhP

H1gp16

215

39

65aa

similarityto

SH1O

RF20

(YP271877)a

ndalso

topu

tativ

eprotein

11of

HHIV-2

(AFD

02292)Z

OD2009

19098of

Hap

paucihalophilusHlacA

J19852Hbflacisa

lsiAJ

5C-

term

inaltransm

embraned

omain(Pho

bius)

179500ndash9

919c

HhP

H1gp17

139

41

79aa

similarityto

SH1O

RF21

(YP271878)a

ndalso

topu

tativ

eprotein

12of

HHIV-2

(AFD

02293)hypotheticalprotein

HlacA

J19847of

Hbflacisa

lsi(ZP09950010)Transm

embraned

omain(Pho

bius)

189923ndash10594c

HhP

H1gp18

223

43

83aa

similarityto

SH1O

RF22

(YP271879)a

ndalso

topu

tativ

eprotein

13of

HHIV-2

(AFD

02294)Z

OD2009

19108of

Hap

paucihalophilusHlacA

J19842of

Hbflacisa

lsiP

redicted

signalsequence(sig

nalP)Con

tains4

CxxC

motifs

1910659ndash10943

HhP

H1gp19

94104

Capsid

protein

VP12

(ORF

19)91aa

similarityto

SH1V

P12(O

RF23Y

P271880)a

ndalso

toVP12of

HHIV-2

(AFD

02295)

andHlacA

J19837of

Hbflacisa

lsiZ

OD2009

19113of

Happau

cihalophilusTw

otransm

embraned

omains

(Pho

bius)

2010960ndash

11517

HhP

H1gp20

185

44

Capsid

protein

VP7(O

RF20)98aa

similarityto

SH1V

P7(O

RF24Y

P271881)a

ndalso

toVP7

ofHHIV-2

(AFD

02296)

ZOD2009

19118

ofHappau

cihalophilusHlacA

J19832of

Hbflacisa

lsiAJ

5

2111519ndash

12217

HhP

H1gp21

232

41

Capsid

protein

VP4(O

RF21)94aa

similarityto

SH1O

RF25

(YP271882)a

ndalso

toVP4

ofHHIV-2

(AFD

02297)

ZOD2009

19123of

Happau

cihalophilusHlacA

J19827of

Hbflacisa

lsiAJ

522

12233ndash12454

HhP

H1gp22

7339

81aa

similarityto

SH1O

RF26

(YP271883)a

ndalso

topu

tativ

eprotein

17of

HHIV-2

(AFD

02298)

2312458ndash12697

HhP

H1gp23

7948

78aa

similarityto

SH1capsid

protein

VP13

(ORF

27Y

P271884)a

ndalso

toVP13of

HHIV-2

(AFD

02299)P

redicted

coil-coildo

main

PredictedC-

term

inaltransm

embraned

omain(Pho

bius)

Archaea 9

Table4Con

tinued

ORF

aPo

sitionb

Locustag

Leng

thc(aa)

pKi

Similaritycharacteris

tics

2412701ndash1506

4HhP

H1gp24

787

39

Capsid

protein

VP2(O

RF24)75aa

similarityto

SH1V

P2(O

RF28Y

P271885)a

ndto

VP2

ofHHIV-2

(AFD

02300)

ZOD2009

19133of

Happau

cihalophilus(ZP

0804

6189)

2515065ndash15961

HhP

H1gp25

298

45

Putativec

apsid

protein

VP5(O

RF25)81aa

similarityto

SH1V

P5(O

RF29Y

P271886)a

ndto

HHIV-2

VP5

(AFD

02301)

HlacA

J19807of

HbflacisalsiAJ

5ZO

D2009

19133of

Happau

cihalophilus

261596

4ndash1644

6HhP

H1gp26

160

46

73aa

similarityto

SH1capsid

protein

VP10

(ORF

30Y

P271887)a

ndto

VP10of

HHIV-2

(AFD

02302)Z

OD2009

19138of

Happau

cihalophilusHlacA

J19802of

Hbflacisa

lsiP

redicted

C-term

inaltransm

embraned

omain(TMHMM)

271644

6ndash16895

HhP

H1gp27

149

42

Capsid

protein

VP9(O

RF27)94aa

similarityto

SH1V

P9(O

RF31Y

P271888)a

ndto

ZOD2009

19143of

Hap

paucihalophilus(ZP

0804

6191)

2816908ndash17921

HhP

H1gp28

337

43

Capsid

protein

VP3(O

RF28)91aa

similarityto

SH1V

P3(O

RF32Y

P271889)a

ndto

NJ7G

2365

ofNa

trinemaspJ7-2

2917928ndash18617

HhP

H1gp29

229

42

Capsidprotein

VP6(O

RF29)83aa

similarityto

SH1V

P6(O

RF33Y

P271890)Also

toNJ7G

3194

ofNa

trinemaspJ7-2and

Hmuk

0476

ofHalom

icrobium

mukohataeiPredictedcarboxypeptid

aser

egulatory-lik

edom

ain(C

arbo

xypepD

reg

pfam

13620)

3018614ndash

18943

HhP

H1gp30

109

49

71aa

similarityto

SH1O

RF34

(YP271891)a

ndto

putativ

eprotein

27of

HHIV-2

(AFD

02308)P

redicted

signalsequence

(signalP)

3119054ndash

19176c

HhP

H1gp31

4050

TwoCx

xCmotifsN

oho

molog

inSH

1orH

HIV-2

3219173ndash19289c

HhP

H1gp32

3841

68aa

similarityto

SH1O

RF37

(YP271894)

3319286ndash

19552c

HhP

H1gp33

8853

58aa

similarityto

SH1O

RF39

(YP271896)a

ndto

HHIV-2

putativ

eprotein

29(A

FD02310)Fou

rCxxCmotifs

3419549ndash

19731c

HhP

H1gp34

6070

Con

tainsC

xxCmotif

Noho

molog

inSH

1orH

HIV-2

3519728ndash20219c

HhP

H1gp35

163

56

93aa

similarityto

SH1O

RF41

(YP271898)a

ndto

putativ

eprotein

30of

HHIV-2

(AFD

02311)

3620216ndash

21415c

HhP

H1gp36

399

50

95aa

similarityto

SH1O

RF42

(YP271899)a

ndto

putativ

eprotein

31of

HHIV-2

(AFD

02312)

3721419ndash

21778c

HhP

H1gp37

11941

91aa

similarityto

SH1O

RF43

(YP271900)a

ndto

putativ

eprotein

32of

HHIV-2

(AFD

02313)

3821762ndash2300

6cHhP

H1gp38

414

42

83aa

similarityto

SH1O

RF44

(YP271901)a

ndto

putativ

eprotein

33of

HHIV-2

(AFD

02314)P

redicted

coil-coiland

helix-tu

rn-helixdo

mains

3923003ndash23158c

HhP

H1gp39

5165

69aa

similarityto

SH1O

RF45

(YP271902)

4023161ndash23370c

HhP

H1gp40

6935

74aa

similarityto

SH1O

RF46

(YP271903)and55sim

ilarityto

putativ

eprotein

35of

HHIV-2

(AFD

02316)

4123422ndash23586c

HhP

H1gp41

5476

86aa

similarityto

SH1O

RF47

(YP271904)Con

tains2

CxxC

motifsand

show

ssim

ilarityto

proteindo

mainfamily

PF14206Arginine-ric

h

4223583ndash24023c

HhP

H1gp42

48

77aa

similarityto

SH1O

RF48

(YP271905)a

ndto

putativ

eprotein

36of

HHIV-2

(AFD

02317)H

ham114540of

Hcc

hamelinensis(ZP11271999)Ph

iCh1p72of

Natrialba

phageP

hiCh

1(NP665989)DUF4

326(pfam14216)

family

domain

4324020ndash

24538c

HhP

H1gp43

172

47

79aa

similarityto

SH1O

RF49

(YP271906)and56sim

ilarityto

putativ

eprotein

37of

HHIV-2

(AFD

02318)

4424535ndash25053c

HhP

H1gp44

172

43

92aa

similarityto

SH1O

RF50

(YP271907)and72sim

ilarityto

putativ

eprotein

38of

HHIV-2

(AFD

02319)

4525050ndash

25277c

HhP

H1gp45

7543

Noho

molog

inSH

1orH

HIV-2

4625274ndash

25387c

HhP

H1gp46

3748

Noho

molog

inSH

1orH

HIV-2

4725533ndash25904

HhP

H1gp47

123

43

61aa

similarityto

SH1O

RF51

(YP271908)66sim

ilarityto

putativ

eprotein

39of

HHIV-2

(AFD

02320)P

redicted

COG1342

domain(D

NAbind

inghelix-tu

rn-helix)

4825901ndash26092c

HhP

H1gp48

6349

Con

tainsC

xxCmotif

Noho

molog

inSH

1orH

HIV-2

4926089ndash

2764

8cHhP

H1gp49

506

40

81aa

similarityto

SH1O

RF55

(YP271912)(bu

tonlyin

theN

-term

inalhalf)H

omolog

ofvirusstru

cturalprotein

VP18

ofHHIV-2

(AFD

02323)

a ORF

swerep

redicted

either

byGLIMMER

orby

manualsearching

forh

omologsintheG

enBa

nkdatabase

b Startandendpo

sitions

ofORF

sare

give

inbp

numbera

ccording

totheP

H1sequenced

epositedatGenBa

nk(KC2

52997)O

RFso

nthec

omplem

entary

strandared

enoted

bythes

uffixc

c Lengthof

thep

redicted

ORF

innu

mbero

faminoacids

10 Archaea

Table 5 Main differences (divergent regions) between the SH1 and PH1 genomes

RegionaSH1 startb SH1 stop Length (bp) PH1 start PH1 stop Length (bp) Comment

DV1 527 650 123 538 604 66

Replacement Both regions have direct repeats at their bordersPH1 has two sets GACCCGGC and CGCTGC while SH1 hasCCCGAC In SH1 this replacement region includes theC-terminal region of ORF2 and the N-terminal region ofORF3 In PH1 it covers only the C-terminal region of ORF1

DV2 2433 2588 155 2386 2387 mdash RMDc from PH1 SH1 repeat at border is TGACCGThisremoves the homolog of SH1 ORF10 from PH1

DV3 3324 3598 274 3137 3416 279 Replacement at the beginning of SH1 ORF13PH1 ORF12(capsid protein VP1)

DV4 6656 7128 472 6478 7085 607 Replacement near the C-terminus of SH1 ORF13PH1 ORF12(capsid protein VP1)

DV5 7491 8341 850 7446 7447 mdash

Indel that results in ORFs14-16 of SH1 not being present inPH1 SH1 ORF14 has a close homolog in HHIV-2 (ORF6) in asimilar position just after the VP1 homolog SH1 ORF15 is aconserved protein in haloarchaea (eg NP 2552A andHmuk 2978) and halovirus His1 (ORF13) Possible invertedrepeat (AGCCATG) at border of SH1 region

DV6 13705 13706 mdash 12818 12851 33RMD from SH1 PH1 repeat at border is CAGCGG(gt)G Thisremoves a part of capsid protein VP2 sequence from the SH1protein

DV7 13789 13863 74 12934 12941 7 Replacement This occurs within VP2 gene of both viruses

DV8 15142 15186 44 14220 14221 mdash RMD from PH1 SH1 repeat at border is TG(tc)CCGACGAand occurs within capsid protein VP2 gene of both viruses

DV9 15303 15317 14 14337 14338 mdash RMD from PH1 SH1 repeat at border is GCCGACGA andoccurs within capsid protein VP2 gene of both viruses

DV10 15504 15527 23 14529 14530 mdash RMD from PH1 SH1 repeat at border is GACGA and occurswithin capsid protein VP2 of both viruses

DV11 20026 20405 379 19019 19173 154Replacement beginning at the start of SH1 ORF35PH1 ORF31This region is longer in SH1 where it includes ORF36 an ORFthat has no homolog present in PH1

DV12 20440 20661 221 19208 19286 78 Replacement This is unequal and includes an ORF in SH1(ORF38) that is not present in PH1

DV13 20943 21085 142 19562 19728 166 Replacement This covers SH1 ORF40PH1 ORF34 Thepredicted proteins are not homologous

DV14 23541 23542 mdash 22171 22197 27 Probable RMD in SH1 PH1 repeat at border is CGTCTCGGand occurs in SH1 ORF44PH1 ORF38

DV15 24506 24521 15 23152 23153 mdash Probable RMD in PH1 SH1 repeat at border is CTCGGT andoccurs near the end of SH1 ORF45PH1 ORF39

DV16 24951 24964 13 23579 23580 mdash Probable RMD in PH1 SH1 repeat at border is CGGTC andoccurs in SH1 ORF47PH1 ORF41

DV17 26449 26450 mdash 25050 25274 224Probable RMD in SH1 PH1 repeat at border is TCATGCG andoccurs near the start of SH1 ORF50PH1 ORF44 and providesan extra ORF for PH1 (ORF45)

DV18 27074 29764 690 25895 26933 1038

Replacement Left border at start of SH1 ORF51PH1 ORF47and extends rightwards into SH1 ORF55PH1 ORF49 (VP18gene) It is an unequal replacement and SH1 has two moreORFs in this region than PH1

aDV Divergent regions between the genomes of SH1 and PH1bStart and stop positions refer to the GenBank sequences of the two viruses SH1 NC 0072171 PH1 KC252997cRMD repeat-mediated deletion event as described in [20]

The recently described halovirus SNJ1 carries manyORFsthat have homologs adjacent to a previously described ViPREofHmc mukohataei (from here on denoted by ViPREHmuk1)a locus that contains genes related to His2 (and other ple-olipoviruses) as well as toHaloferax plasmid pHK2 (probablya provirus) and to two small plasmids ofHqr walsbyi (sim6 kbpL6A and pL6B)The latter relationship provides wider linksto other viruses because one of the PL6 genes (Hqrw 6002)

has homologs in some pleolipoviruses (HPRV-3 and HGPV-1) while another gene (Hqrw 6005) is related to ORF16 ofthe spindle-shaped halovirus His1 [15] After adding the SNJ1gene homologs to ViPREHmuk1 it is extended significantlyand a close inspection of the flanking regions revealed atRNA-ala gene at one end and a partial copy of the sametRNA-ala gene (next to a phage integrase gene) at the otherend (supplementary Figure 1) Genes beyond these points

Archaea 11

Table 6 Transfection of haloarchaea by PH1 DNA

Speciesa Efficiency of transfectiontransformation withPH1 DNAb (PFU120583g of DNA) pUBP2 DNAbc (CFU120583g of DNA)

Har hispanica 53 plusmn 05 times 103 14 plusmn 08 times 104

Har marismortui 45 plusmn 07 times 103 18 plusmn 01 times 104

ldquoHar sinaiiensisrdquo 32 plusmn 06 times 102 mdashHalobacterium salinarum mdash 54 plusmn 42 times 103

Haloferax gibbonsii 28 plusmn 08 times 103 40 plusmn 26 times 103

Haloferax lucentense mdash 57 plusmn 11 times 103

Hfx volcanii mdash 21 plusmn 07 times 105

Hrr lacusprofundi 70 plusmn 22 times 102 95 plusmn 15 times 103

Haloterrigena turkmenica 16 plusmn 17 times 102 24 plusmn 06 times 103

Natrialba asiatica 35 plusmn 39 times 102 31 plusmn 10 times 104aOnly those species positive for transfection andor plasmid transformation are shownbRates of transfection by (nonprotease treated) PH1 DNA or transformation by plasmid pUB2 are averages of three independent experiments each performedin duplicate (plusmn standard deviation)cTransformants were selected on plates with 2 4 or 6120583gmL simvastatin depending on the strainmdash no plaques or colonies observed

appear to be related to cellular metabolism ViPREHmuk1is now 39379 nt in length flanked by a 59 nt direct repeat(potential att sites) and includes 54 ORFs many of which arerelated to known haloviruses are virus-like (eg integrasesmethyltransferases transcriptional regulators DNA methyl-transferase and DNA glycosylase) are homologs in or nearother knownViPREs or show little similarity to other knownproteins This locus also includes an ORC1CDC6 homolog(Hmuk 0446) which could provide a replication function

The mixture of very different virus and plasmid genesseen within ViPREHmuk1 is remarkable but it also revealshomologs found in or adjacent to genomic loci of otherspecies such as the gene homologs of Har marismortuiseen in the left end of ViPREHmuk1 (supplementary Figure1) When these genes are added to the previously describedHaloarcula locus (ViPREHmuk1) it is also extended signifi-cantly It becomes 25550 bp in length encompasses genesfrom Hmar 2382 to Hmar 2404 and is flanked by a fulltRNA-ala gene at one end and a partial copy at the other Itcarries an ORC1CDC6 homolog an integrase and a phiH-like repressor as well as pleolipovirus homologs

210 Transfection of Haloarchaea by PH1 DNA PH1 DNAwas introduced into Har hispanica cells using the PEGmethod [37] and the cells were screened for virus productionby plaque assay Figure 7 shows the increase in transfectedcells with input (nonproteinase K-treated) virus DNA Theestimated efficiencywas 53times 103 PFUper120583gDNA PH1DNAthat had been treated with proteinase K or with DNAase I(RNase-free) did not produce plaques (data not shown) PH1DNA was found to transfect six species of haloarchaea otherthanHar hispanica includingmembers of the generaHaloar-cula Haloferax Halorubrum Haloterrigena and Natrialba(Table 6) For these experiments the virus production fromtransfected cells was detected by plaque assay on indicatorlawns of Har hispanica

0

0

200 400 600 800 1000

Titre

(PFU

mL)

104

103

102

PH1 DNA (ng)

Figure 7 Transfection ofHar hispanica cells with PH1 DNA Vary-ing amounts of nonproteinase K-treated PH1 DNAwere introducedinto cells of Har hispanica using the PEG method [64] Cells werethen screened for infective centres by plaque assay Data shown is theaverage of three independent experiments performed in duplicateError bars represent one standard deviation of themean If protease-treated DNA was used no transfectant plaques were observed

3 Discussion

PH1 is very similar in particlemorphology genome structureand sequence to the previously described halovirus SH1 [816] Like SH1 it has long inverted terminal repeat sequencesand terminal proteins indicative of protein-primed replica-tion Purified PH1 particles have a relatively low buoyant

12 Archaea

density and are chloroform sensitive and show a layeredcapsid structure consistent with the likely presence of aninternalmembrane layer as has been shown for SH1HHIV-2and SNJ1 [13 26]The particle stability of PH1 to temperaturepH and reduced salt was similar to SH1 [8] The structuralproteins of PH1 are very similar in sequence and relativeabundance to those of SH1 so the two viruses are likely toshare the same particle geometry [16]

Viruses that use protein-primed replication such as bac-teriophageΦ29 carry a viral type BDNApolymerase that caninteract specifically with the proteins attached to the genomictermini and initiate strand synthesis [38] ArchaeovirusesHis1 [7] and Acidianus bottle-shaped virus [39] probablyuse this mode of replication However a polymerase genecannot be found in the genomes of PH1 SH1 or HHIV-2DNA polymerases are large enzymes and the only ORF ofSH1 that was long enough to encode such an enzyme andhad not previously been assigned as a structural protein wasORF55 which specified an 865 aa protein that contained noconserved domains indicative of polymerases [27]The recentstudy of HHIV-2 showed that the corresponding ORF in thisvirus (gene 42) is a structural protein of the virion and byinference this is likely also for the corresponding proteinsof SH1 and PH1 (no homolog is present in SNJ1) Withouta polymerase gene these viruses must use a host enzymebut searches for a viral type B DNA polymerase in thegenome of the hostHar hispanica or in the genomes of othersequenced haloarchaea did not find anymatchesThis arguesstrongly for a replication mechanism that is different to thatexemplified byΦ29 An attractive alternative is that displayedby Streptomyces linear plasmids which have 51015840-terminal pro-teins and use a cellular polymerase for replication [40] Theterminal proteins are not used for primary replication but forend patching [41] and it has been shown that if the term-inal repeat sequences are removed from these plasmids andthe ends ligated they can replicate as circular plasmids [42]This provides a testable hypothesis for the replication of halo-sphaeroviruses and also offers a pathway for switchingbetween linear (eg PH1) and circular (eg SNJ1) forms Useof a host polymerase would also fit with the ability of SH1 andPH1 DNA to transfect many different haloarchaeal species[31] as these enzymes and theirmode of action are highly con-served

The growth characteristics of PH1 inHar hispanica bothby single-cell burst and single-step growth experiments gavevalues of 50ndash100 virusescell significantly less than that ofSH1 (200 virusescell) [8] and HHIV-2 (180 virusescell) [13]The latter virus is lytic whereas SH1 PH1 and SNJ1 start pro-ducing extracellular viruswell before any decrease in cell den-sity indicating that virus release can occur without lysisThishas been most clearly shown with SH1 where almost 100of cells can be infected [8] Infection ofHar hispanica by PH1was less efficient (sim30 of cells) but the kinetics of virus pro-duction in single-step growth curves of SH1 and PH1 aresimilar with virus production occurring over several hoursrather than at a clearly defined time after infection The twoviruses are also very closely related and infect the same hostspecies so it is likely that they use the same scheme for cellexit For both viruses the cell density eventually decreases in

single-step cultures showing that virus infection does resultin cell death This mode of exit appears to maximise theproduction of the virus over time as the host cells survive foran extended period In natural hypersaline waters cell num-bers are usually high but growth rates are low [12] and thehigh incident UV (on often shallow ponds) would damagevirus DNA so reducing the half-life of released virus part-icles [43] These factors may have favoured the exit strategydisplayed by these viruses improving their chance of trans-mission to a new host

The motif GATC was absent in the genomes of PH1 SH1andHHIV-2 and themotif CTAGwas either absent or greatlyunderrepresented All three viruses share the same host butGATC and CTAG motifs are plentiful in the Har hispanicagenome (accessions CP002921 and CP002923) Avoidance ofthesemotifs in halovirusesHis1 His2 HF1 andHF2 has beenreported previously [7] Such purifying selection commonlyresults from host restriction enzymes and inHfx volcanii itis exactly these two motifs that are targeted [44] One Hfxvolcanii enzyme recognises A-methylated GATC sites [45]and another recognises un-methylated CTAG sites (whichare protected by methylation in the genome) In the currentstudy this could explain the negative transfection results forHfx volcanii as the PH1 genome contains two CTAGmotifsBy comparison the SH1 genome contains no CTAG motifsand is able to transfect Hfx volcanii [31]

Comparison of the PH1 and SH1 genomes allowed adetailed picture of the natural variation occurring betweenclosely related viruses revealing likely deletion events medi-ated by small repeats a process described previously in astudy of Hqr walsbyi and termed repeat-mediated deletion[20] There are also many replacements including blocks ofsequence that occur within long open reading frames (suchas the capsid protein genes VP1 and VP18) VP3 and VP6 areknown to form the large spikes on the external surface of theSH1 virion and presumably one or both interact with host cellreceptors [16] PH1 and SH1 have identical host ranges con-sistent with the high sequence similarity shown between theircorresponding VP3 and VP6 proteins

A previous study of SH1 could not detect transcriptsacross annotated ORFs 1ndash3 or 55-56 [30] ORFs 1 and 56occur in the terminal inverted repeat sequence and in thepresent study it was found that there were no ORFs cor-responding to these in the PH1 genome Given the close simi-larity of the two genomes the comparative data are in agree-ment with the transcriptional data indicating that these SH1ORFs are not used SH1 ORFs 2 and 3 are more problematicas good homologs of these are also present in PH1 The con-flict between the transcriptional data and the comparativegenomic evidence requires further experimental work toresolveThe status of SH1ORF55 has recently been confirmedby studies of the related virus HHIV-2 (discussed above)

The pleolipovirus group has expanded dramatically inthe last few years and it now comprises a diverse group ofviruseswith different genome types and replication strategiesWhat is evenmore remarkable is that a similar expansion cannow be seen with SH1-related viruses which includes viruseswith at least two replication modes and genome types (linearand circular) Evidence from haloarchaeal genome sequences

Archaea 13

show cases of not only provirus integrants or plasmids (egpKH2 and pHH205) but also genomic loci (ViPREs) thatcontain cassettes of virus genes from different sources andat least two cases (described in this study) have likely attsites that indicate circularisation and mobility While thereis a clear relationship between viruses with circular ds- andssDNA genomes that replicate via the rolling circle method(ie the ssDNA form is a replication intermediate) it ismore difficult to explain how capsid gene cassettes can movebetween these viruses and those with linear genomes andterminal proteins Within the pleolipoviruses His2 has alinear genome that contains a viral type B DNA polymeraseand terminal proteins while all the other described membershave circular genomes that contain a rep homolog and prob-ably replicate via the rolling circle method [14 19] Similarlyamong the halosphaeroviruses PH1 SH1 and HHIV-2 havelinear dsDNA with terminal proteins and probably replicatein the same way but the related SNJ1 virus has a circulardsDNA genome The clear relationships shown by the capsidgenes of viruses within each group plus the connectionsshown to haloviruses outside each group (eg the DNA poly-merases of His1 and His2) all speak of a vigorous meansof recombination one that can readily switch capsid genesbetween viruses with radically different replication strategiesHow is the process most likely to operate One possibility issuggested by ViPREs which appear to be mobile collectionsof capsid and replication genes from different sources Theyoffer fixed locations for recombination to occur providegene cassettes that can be reassorted to produce novel virusgenomes and some can probably recombine out as circularforms For example ViPREHmuk1 carries genes related topleolipoviruses halosphaeroviruses and other haloviruses Itis yet to be determined if any of the genes carried in theseloci are expressed in the cell (such as the genes for capsidproteins) or if ViPREs provide any selective advantage to thehost

4 Materials and Methods

41 Water Sample A water sample was collected in 1998from Pink Lake (32∘001015840 S and 115∘301015840 E) a hypersalinelake on Rottnest Island Western Australia Australia It wasscreened in the same year for haloviruses using the methodsdescribed in [8] A single plaque on a Har hispanica lawnplate was picked and replaque purified The novel haloviruswas designated PH1

42 Media Strains and Plasmids The media used in thisstudy are described in the online resource The Halohand-book (httpwwwhaloarchaeacomresourceshalohandbookindexhtml) Artificial salt water containing 30 (wv) totalsalts comprised of 4M NaCl 150mM MgCl

2 150mM

MgSO4 90mM KCl and 35mM CaCl

2and adjusted to pH

75 using sim2mL 1M Tris-HCl (pH 75) per litre MGM con-taining 12 18 or 23 (wv) total salts and HVD (halo-virus diluent) were prepared from the concentrated stock aspreviously described [46] Bacto-agar (Difco Laboratories)was added to MGM for solid (15 gL) or top-layer (7 gL)media

Table 1 lists the haloarchaea used in this study Allhaloarchaea were grown aerobically at 37∘C in either 18 or23 (wv)MGM(depending on the strain) andwith agitation(except for ldquoHar sinaiiensisrdquo)The plasmids used were pBlue-script II KS+ (Stratagene Cloning Systems) pUBP2 [47]and pWL102 [48] pUBP2 and pWL102 were first passagedthrough E coli JM110 [49] to prevent dam-methylationof DNA which has been shown to reduce transformationefficiency in some strains [45]

43 Negative-Stain Electron Microscopy The method fornegative-stain TEM was adapted from that of V Tarasovdescribed in the online resource The Halohandbook (httpwwwhaloarchaeacomresourceshalohandbookindexhtml)A 20120583L drop of the sample was placed on a clean surfaceand the virus particles were allowed to adsorb to a Formvarfilm 400-mesh copper grid (ProSciTech) for 15ndash2min Theywere then negatively stained with a 20120583L drop of 2 (wv)uranyl acetate for 1-2min Excess liquid was absorbed withfilter paper and the grid was allowed to air dry Grids wereexamined either on a Philips CM 120 BioTwin transmissionelectron microscope (Royal Philips Electronics) operating atan accelerating voltage of 120 kV or on a Siemens Elmiskop102 transmission electron microscope (Siemens AG) operat-ing at an accelerating voltage of 120 kV

44 Virus Host Range Twelve haloarchaeal strains from thegenera Haloarcula Halobacterium Haloferax HalorubrumandNatrialba (Table 1)Thirteen naturalHalorubrum isolates(H Camakaris unpublished data) and five uncharacterizedhaloarchaeal isolates fromLakeHardy Pink Lake (inWesternAustralia) and Serpentine Lake (D Walker and M K Seahunpublished data) were screened for PH1 susceptibilityLysates from PH1-infected Har hispanica cultures (1 times1011 PFUmL) were spotted onto lawns of each strain (usingmedia with 12 or 18 (wv)MGM depending on the strain)and incubated for 2ndash5 days at 30 and at 37∘C

45 Large Scale Virus Growth and Purification Liquid cul-tures of PH1 were grown by the infection (MOI 005) of anearly exponential Har hispanica culture in 18 (wv) MGMCultures were incubated aerobically at 37∘C with agitationfor 3 days Clearing (ie complete cell lysis) of PH1-infectedcultures did not occur and consequently cultures were harv-ested when the absorbance at 550 nm reached the minimumand the titre (determined by plaque assay) was at the maxi-mum usually sim1010ndash1011 PFUmL Virus was purified usingthe method described previously for SH1 [8 31] except thatafter the initial low speed spin (Sorvall GSA 6000 rpm30min 10∘C) virus was concentrated from the infected cul-ture by centrifugation at 26000 rpm (13 hr 10∘C) onto acushion of 30 (wv) sucrose in HVDThe pellet was resus-pended in a small volume of HVD and loaded onto a pre-formed linear 5ndash70 (wv) sucrose gradient followed byisopycnic centrifugation in 13 gmL CsCl (Beckman 70Ti60000 rpm 20 hr 10∘C) The white virus band occurredat a density of 129 gmL and was collected and diluted inhalovirus diluent (HVD see Section 4) and the virus pelleted(Beckman SW55 35000 rpm 75min 10∘C) and resuspended

14 Archaea

in a small volume of HVD and stored at 4∘C Virus recoveryat the major stages of a typical purification is given insupplementary Table 1 The specific infectivity of pure virussolutions was determined as the ratio of the PFUmL to theabsorbance at 260 nm

46 PH1 Single-Step Growth Curve An early exponentialphase culture of Har hispanica grown in 18 (wv) MGMwas infected with PH1 (MOI 50) Under these conditions thepercentage of infected cells was approximately 30 After anadsorption period of 1 hr at 37∘C the cells were washed threetimes with 18 (wv) MGM (at room temperature) resus-pended in 100mL 18 (wv) MGM (these methods ensuredthe removal of all residual virus) and incubated at 37∘Cwith shaking (100 rpm) Samples were removed at hourlyintervals for measurements of absorbance at 550 nm andthe number of infective centres Immediately after samplingtitres were determined by plaque assay with an indicator lawnof Har hispanica Each experiment was performed in tripli-cate

47 Halovirus Stability After various treatments sampleswere removed and diluted in HVD and virus titres weredetermined by plaque assay on Har hispanica Each experi-ment was performed in triplicate and representative data areshown Chloroform sensitivity was examined by the exposureof PH1 lysates in 18 (wv) MGM to chloroform in a volumeratio of 1 4 (chloroform to lysate) Incubation was at roomtemperature with constant agitation At appropriate timepoints the mix was allowed to settle and a sample wasremoved from the upper layer and diluted in HVDThe effectof a low ionic environment was examined by diluting PH1lysates (in 18 (wv) MGM) into double distilled H

2O in a

volume ratio of 1 1000 (lysate to double distilledH2O) Incu-

bation was at room temperature with constant agitationSamples were removed at various times and diluted in HVDThe pH stability of PH1 was determined by dilution of PH1lysates in 18 (wv)MGMin the appropriate pHbuffer (HVDbuffered with appropriate Tris-HCl) in a volume ratio of1 100 (lysate to buffer) Incubation was at room temperaturewith constant agitation After 30min samples were removedand diluted in HVD Thermal stability of PH1 was examinedby a 1 hr incubation of a virus lysate in 18 (wv) MGM atdifferent temperatures after which they were brought quicklyto room temperature diluted in HVD and titrated

48 Protein Procedures To remove salts purified virus prepa-rations were mixed with trichloroacetic acid (10 (vv) finalconcentration) and incubated on ice for 15min to allowthe proteins to precipitate After centrifugation (16000 g15min room temperature) the precipitate was washed threetimes in acetone dried and resuspended in double distilledH2O Proteins were dissolved in Laemmli sample buffer with

48mM 120573-mercaptoethanol [50] heated in boiling water for5min then separated on 12 (wv) NuPAGE Novex Bis-Tris Gels using MES-SDS running buffer according to themanufacturerrsquos directions (Invitrogen) After electrophoresisgels were rinsed in double distilled H

2O and stained with

01 (wv) Brilliant Blue G in 40 (vv) methanol and

10 (vv) acetic acid Gels were destained with severalchanges of 40 (vv) methanol and 10 (vv) acetic acidAlternatively gels were stained with GelCode GlycoproteinStaining Kit Pro-Q Emerald 488 Glycoprotein Gel and BlotStain Kit SYPRO Ruby Protein Gel Stain according tothe manufacturerrsquos directions (Invitrogen Molecular ProbesPierce Biotechnology)

Protein bands were cut from the gels and sent to theAustralian Proteome Analysis Facility (Macquarie Univer-sity) for trypsin digestion and analysis by matrix assistedlaser desorption ionisation time of flight mass spectrometry(MALDI-TOF MS) on an Applied Biosystems 4700 Pro-teomics Analyser (Applied Biosystems)

49 DNA Procedures Proteinase K-treated and nonpro-teinaseK-treatedDNApreparationsweremade frompurifiedPH1 using the methods described previously for SH1 [8]DNA was separated on 1 (wv) agarose gels in Tris-acetate-EDTA electrophoresis buffer and was stained with ethidiumbromide (Sigma-Aldrich)120582 DNA DNase I (RNase-free) exonuclease III mung

bean nuclease nuclease BAL-31 T4 DNA polymerase T4DNA ligase T7 exonuclease and type II restriction endonu-cleases were purchased from New England Biolabs RNase Aand yeast transfer RNA were purchased from Sigma-AldrichProteinase K was purchased from Promega Oligonucleotideprimers were purchased from Geneworks Australia

To clone fragments of the PH1 genome purified virusDNA was digested either with AccI Eco0109I and MseI orSmaI restriction endonucleases blunted with DNA Poly-merase I Large (Klenow) Fragment where appropriate andligated into AccI- Eco0109I- or SmaI-digested pBluescript IIKS+ using T4 DNA ligase according to the manufacturerrsquosinstructions (New England Biolabs) The DNA was intro-duced intoE coliXL1-Blue and transformants grown onLuriaagar containing 15 120583gmL tetracycline and 100 120583gmL ampi-cillinThe resulting cloneswere sequenced and the sequenceswere used to design specific oligonucleotide primers for PCRamplification andor primer walking using the virus genome(or specific restriction fragments) as templates

To amplify PH1 nucleic acid approximately 10 ng virusDNA was combined with 500 nM primers 100120583M of eachdNTP 1U Deep Vent DNA Polymerase (New England Bio-labs) and 1 timesThermoPol buffer (New England Biolabs) ina total reaction volume of 50 120583L Template was denatured at95∘C for 10min followed by 30 cycles of denaturation at 95∘Cfor 30 sec annealing at 56∘C for 30 sec extension at 75∘C for2min and a final extension at 75∘C for 10min PCR reactionswere performed on a PxE02ThermoCycler (Thermo ElectroCorporation) Sequencing reactions using 32 pmol primerwere performed by the dideoxy chain termination methodusing ABI PRISM Big Dye Terminator Mix version 31 onan ABI 3100 capillary sequencer (Applied Biosystems) bythe Applied Genetic Diagnostics Sequencing Service at theDepartment of Pathology (The University of Melbourne)

To obtain terminal genomic fragments PH1 DNA wasdigested by the enzyme Pas1 and the head (sim700 bp) andthe tail (sim1300 bp) fragments were purified by the QIAEXII gel extraction kits and treated with 05M piperidine for

Archaea 15

2 hr at 37∘C to remove protein residues The piperidine-treated fragments were sequenced by the Genomics BiotechCompany (Taipei Taiwan) using two primers TGACCA-ATTAATTAGGCCGGTTCGCC (PH1 head-R) and GTGC-CATACTGCTACAATTCT (PH1 tail-F)

Four hundred fifty-four whole genome sequencing PH1DNA samples (sim5 120583g) were sequenced using parallel pyrose-quencing on a Roche 454 Genome Sequencer System atMission Biotech (Taipei Taiwan) The largest contig was27399with 8803 reads (Newbler version 27) with 27387 posi-tions being of Q40 quality The average depth of coverage foreach base was 829TheGenBank accession for the entire PH1sequence is KC252997

Acknowledgments

The authors thank Carolyn Bath Mei Kwei (Joan) Seahand Danielle Walker for their early work in PH1 characteri-zation Simon Crawford Jocelyn Carpenter and Anna Fried-huber for their assistance with the transmission electronmicroscopy and Tiffany Cowie and Voula Kanellakis fortheir sequencing assistance This paper has been facilitatedby access to the Australian Proteome Analysis Facility estab-lished under the Australian Governmentrsquos Major NationalResearch Facilities program SLT was supported by a grantfrom the Biodiversity Research Center Academia SinicaTaiwan

References

[1] M Krupovic M F White P Forterre and D PrangishvilildquoPostcards from the edge structural genomics of archaealvirusesrdquo Advances in Virus Research vol 82 pp 33ndash62 2012

[2] P Stolt andW Zillig ldquoGene regulation in halophage 120601Hmdashmorethan promotersrdquo Systematic and Applied Microbiology vol 16no 4 pp 591ndash596 1994

[3] R Klein U Baranyi N Rossler B Greineder H Scholz and AWitte ldquoNatrialba magadii virus 120593Ch1 first complete nucleotidesequence and functional organization of a virus infecting ahaloalkaliphilic archaeonrdquo Molecular Microbiology vol 45 no3 pp 851ndash863 2002

[4] E Pagaling R D Haigh W D Grant et al ldquoSequence analysisof an Archaeal virus isolated from a hypersaline lake in InnerMongolia Chinardquo BMC Genomics vol 8 article 410 2007

[5] S L Tang SNuttall andMDyall-Smith ldquoHalovirusesHF1 andHF2 evidence for a recent and large recombination eventrdquo Jour-nal of Bacteriology vol 186 no 9 pp 2810ndash2817 2004

[6] C Bath and M L Dyall-smith ldquoHis1 an archaeal virus of theFuselloviridae family that infectsHaloarcula hispanicardquo Journalof Virology vol 72 no 11 pp 9392ndash9395 1998

[7] C Bath T Cukalac K Porter andM L Dyall-Smith ldquoHis1 andHis2 are distantly related spindle-shaped haloviruses belongingto the novel virus group Salterprovirusrdquo Virology vol 350 no1 pp 228ndash239 2006

[8] K Porter P Kukkaro J K H Bamford et al ldquoSH1 a novelspherical halovirus isolated from an Australian hypersalinelakerdquo Virology vol 335 no 1 pp 22ndash33 2005

[9] M Dyall-Smith S L Tang and C Bath ldquoHaloarchaeal viruseshow diverse are theyrdquo Research in Microbiology vol 154 no 4pp 309ndash313 2003

[10] A Oren G Bratbak and M Heldal ldquoOccurrence of virus-likeparticles in the Dead Seardquo Extremophiles vol 1 no 3 pp 143ndash149 1997

[11] T Sime-Ngando S Lucas A Robin et al et al ldquoDiversityof virus-host systems in hypersaline Lake Retba SenegalrdquoEnvironmental Microbiology vol 13 no 8 pp 1956ndash1972 2010

[12] N Guixa-Boixareu J I Calderon-Paz M Heldal G Bratbakand C Pedros-Alio ldquoViral lysis and bacterivory as prokaryoticloss factors along a salinity gradientrdquoAquaticMicrobial Ecologyvol 11 no 3 pp 215ndash227 1996

[13] S T Jaakkola R K Penttinen S T Vilen et al ldquoClosely relatedarchaeal Haloarcula hispanica icosahedral viruses HHIV-2 andSH1 have nonhomologous genes encoding host recognitionfunctionsrdquo Journal of Virology vol 86 no 9 pp 4734ndash47422012

[14] E Roine P Kukkaro L Paulin et al ldquoNew closely related halo-archaeal viral elements with different nucleic acid typesrdquo Jour-nal of Virology vol 84 no 7 pp 3682ndash3689 2010

[15] A Sencilo L Paulin S KellnerMHelm and E Roine ldquoRelatedhaloarchaeal pleomorphic viruses contain different genometypesrdquo Nucleic Acids Research vol 40 no 12 pp 5523ndash55342012

[16] H T Jaalinoja E Roine P Laurinmaki H M Kivela D HBamford and S J Butcher ldquoStructure and host-cell interactionof SH1 a membrane-containing halophilic euryarchaeal virusrdquoProceedings of the National Academy of Sciences of the UnitedStates of America vol 105 no 23 pp 8008ndash8013 2008

[17] M K Pietila N S Atanasova VManole et al ldquoVirion architec-ture unifies globally distributed pleolipoviruses infecting halo-philic archaeardquo Journal of Virology vol 86 no 9 pp 5067ndash50792012

[18] M L Holmes and M L Dyall-Smith ldquoA plasmid vector witha selectable marker for halophilic archaebacteriardquo Journal ofBacteriology vol 172 no 2 pp 756ndash761 1990

[19] M L Holmes F Pfeifer andM L Dyall-Smith ldquoAnalysis of thehalobacterial plasmid pHK2 minimal repliconrdquo Gene vol 153no 1 pp 117ndash121 1995

[20] M L Dyall-Smith F Pfeiffer K Klee et al ldquoHaloquadratumwalsbyi limited diversity in a Global Pondrdquo PLoS ONE vol 6no 6 Article ID e20968 2011

[21] S Casjens ldquoProphages and bacterial genomics what have welearned so farrdquoMolecular Microbiology vol 49 no 2 pp 277ndash300 2003

[22] R W Hendrix M C M Smith R N Burns M E Ford andG F Hatfull ldquoEvolutionary relationships among diverse bacte-riophages and prophages all the worldrsquos a phagerdquo Proceedings ofthe National Academy of Sciences of the United States of Americavol 96 no 5 pp 2192ndash2197 1999

[23] D Botstein ldquoA theory ofmodular evolution for bacteriophagesrdquoAnnals of the New York Academy of Sciences vol 354 pp 484ndash491 1980

[24] M Krupovic S Gribaldo D H Bamford and P Forterre ldquoTheevolutionary history of archaeal MCM helicases a case study ofvertical evolution combinedwithHitchhiking ofmobile geneticelementsrdquo Molecular Biology and Evolution vol 27 no 12 pp2716ndash2732 2010

[25] M Krupovic P Forterre and D H Bamford ldquoComparativeanalysis of the mosaic genomes of tailed archaeal viruses andproviruses suggests common themes for virion architecture andassembly with tailed viruses of bacteriardquo Journal of MolecularBiology vol 397 no 1 pp 144ndash160 2010

16 Archaea

[26] Z Zhang Y Liu S Wang et al et al ldquoTemperate membrane-containing halophilic archaeal virus SNJ1 has a circular dsDNAgenome identical to that of plasmid pHH205rdquoVirology vol 434no 2 pp 233ndash241 2012

[27] D H Bamford J J Ravantti G Ronnholm et al ldquoConstituentsof SH1 a novel lipid-containing virus infecting the halophiliceuryarchaeonHaloarcula hispanicardquo Journal of Virology vol 79no 14 pp 9097ndash9107 2005

[28] H M Kivela E Roine P Kukkaro S Laurinavicius P Somer-harju andDH Bamford ldquoQuantitative dissociation of archaealvirus SH1 reveals distinct capsid proteins and a lipid corerdquoVirology vol 356 no 1-2 pp 4ndash11 2006

[29] H Jaalinoja Electron Cryo-Microscopy Studies of Bacteriophage8 and Archaeal Virus SH1 University of Helsinki HelsinkiFinland 2007

[30] K Porter B E Russ J Yang andM L Dyall-Smith ldquoThe trans-cription programme of the protein-primed halovirus SH1rdquoMi-crobiology vol 154 no 11 pp 3599ndash3608 2008

[31] K Porter and M L Dyall-Smith ldquoTransfection of haloarchaeaby the DNAs of spindle and round haloviruses and the useof transposon mutagenesis to identify non-essential regionsrdquoMolecular Microbiology vol 70 no 5 pp 1236ndash1245 2008

[32] S E Luria General Virology John Wiley and Sons New YorkNY USA 1953

[33] C A Thomas K Saigo E McLeod and J Ito ldquoThe separationofDNA segments attached to proteinsrdquoAnalytical Biochemistryvol 93 pp 158ndash166 1979

[34] A L Delcher D Harmon S Kasif OWhite and S L SalzbergldquoImproved microbial gene identification with GLIMMERrdquoNucleic Acids Research vol 27 no 23 pp 4636ndash4641 1999

[35] R Zhang Y Yang P Fang et al ldquoDiversity of telomere palin-dromic sequences and replication genes among Streptomyceslinear plasmidsrdquo Applied and Environmental Microbiology vol72 no 9 pp 5728ndash5733 2006

[36] X Wang H S Lee F J Sugar F E Jenney M W W Adamsand J H Prestegard ldquoPF0610 a novel winged helix-turn-helix variant possessing a rubredoxin-like Zn ribbon motiffrom the hyperthermophilic archaeon Pyrococcus furiosusrdquoBiochemistry vol 46 no 3 pp 752ndash761 2007

[37] S W Cline L C Schalkwyk and W F Doolittle ldquoTransfor-mation of the archaebacterium Halobacterium volcanii withgenomic DNArdquo Journal of Bacteriology vol 171 no 9 pp 4987ndash4991 1989

[38] S Kamtekar A J Berman J Wang et al ldquoThe 12059329 DNA poly-merase protein-primer structure suggests a model for the ini-tiation to elongation transitionrdquo EMBO Journal vol 25 no 6pp 1335ndash1343 2006

[39] X Peng T Basta M Haring R A Garrett and D PrangishvilildquoGenome of theAcidianus bottle-shaped virus and insights intothe replication and packaging mechanismsrdquo Virology vol 364no 1 pp 237ndash243 2007

[40] C C Yang C H Huang C Y Li Y G Tsay S C Lee and CW Chen ldquoThe terminal proteins of linear Streptomyces chrom-osomes and plasmids a novel class of replication primingproteinsrdquo Molecular Microbiology vol 43 no 2 pp 297ndash3052002

[41] C C Yang Y H Chen H H Tsai C H Huang T W Huangand C W Chen ldquoIn vitro deoxynucleotidylation of the term-inal protein of Streptomyces linear chromosomesrdquo Applied andEnvironmental Microbiology vol 72 no 12 pp 7959ndash79612006

[42] P C Chang and S N Cohen ldquoBidirectional replication froman internal origin in a linear Streptomyces plasmidrdquo Science vol265 no 5174 pp 952ndash954 1994

[43] S W Wilhelm W H Jeffrey C A Suttle and D L MitchellldquoEstimation of biologically damaging UV levels in marinesurface waters with DNA and viral dosimetersrdquo Photochemistryand Photobiology vol 76 no 3 pp 268ndash273 2002

[44] T Allers and M Mevarech ldquoArchaeal geneticsmdashthe third wayrdquoNature Reviews Genetics vol 6 no 1 pp 58ndash73 2005

[45] M L Holmes S D Nuttall and M L Dyall-Smith ldquoConstruc-tion and use of halobacterial shuttle vectors and further studieson Haloferax DNA gyraserdquo Journal of Bacteriology vol 173 no12 pp 3807ndash3813 1991

[46] S D Nuttall and M L Dyall-Smith ldquoHF1 and HF2 novelbacteriophages of halophilic archaeardquo Virology vol 197 no 2pp 678ndash684 1993

[47] U Blaseio and F Pfeifer ldquoTransformation of Halobacteriumhalobium development of vectors and investigation of gas vesi-cle synthesisrdquo Proceedings of the National Academy of Sciences ofthe United States of America vol 87 no 17 pp 6772ndash6776 1990

[48] W L Lam and W F Doolittle ldquoShuttle vectors for the archae-bacterium Halobacterium volcaniirdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 86 no14 pp 5478ndash5482 1989

[49] C Yanisch-Perron J Vieira and J Messing ldquoImproved M13phage cloning vectors and host strains nucleotide sequences ofthe M13mp18 and pUC19 vectorsrdquo Gene vol 33 no 1 pp 103ndash119 1985

[50] U K Laemmli ldquoCleavage of structural proteins during theassembly of the head of bacteriophage T4rdquo Nature vol 227 no5259 pp 680ndash685 1970

[51] D G Burns H M Camakaris P H Janssen and M L Dyall-Smith ldquoCombined use of cultivation-dependent and culti-vation-independent methods indicates that members of mosthaloarchaeal groups in anAustralian crystallizer pond are culti-vablerdquo Applied and Environmental Microbiology vol 70 no 9pp 5258ndash5265 2004

[52] G Juez F Rodriguez-Valera A Ventosa and D J KushnerldquoHaloarcula hispanica spec nov and Haloferax gibbonsii specnov two new species of extremely halophilic archaebacteriardquoSystematic and Applied Microbiology vol 8 pp 75ndash79 1986

[53] A Oren M Ginzburg B Z Ginzburg L I Hochstein and BE Volcani ldquoHaloarcula marismortui (Volcani) sp nov nomrev an extremely halophilic bacterium from the Dead SeardquoInternational Journal of Systematic Bacteriology vol 40 no 2pp 209ndash210 1990

[54] M Torreblanca F Rodriguez-Valera G Juez A Ventosa MKamekura and M Kates ldquoClassification of non-alkaliphilichalobacteria based on numerical taxonomy and polar lipidcomposition and description of Haloarcula gen nov andHaloferax gen novrdquo Systematic and Applied Microbiology vol8 pp 89ndash99 1986

[55] A Ventosa and A Oren ldquoHalobacterium salinarum nom cor-rig a name to replace Halobacterium salinarium (Elazari-Vol-cani) and to include Halobacterium halobium and Halobac-terium cutirubrumrdquo International Journal of Systematic Bacte-riology vol 46 no 1 p 347 1996

[56] M C Gutierrez M Kamekura M L Holmes M L Dyall-Smith and A Ventosa ldquoTaxonomic characterization of Halo-ferax sp (ldquoH alicanteirdquo) strain Aa 22 description of Haloferaxlucentensis sp novrdquo Extremophiles vol 6 no 6 pp 479ndash4832002

Archaea 17

[57] M F Mullakhanbhai and H Larsen ldquoHalobacterium volcaniispec nov a Dead Sea halobacterium with a moderate saltrequirementrdquo Archives of Microbiology vol 104 no 1 pp 207ndash214 1975

[58] S D Nuttall and M L Dyall-Smith ldquoCh2 a novel halophilicarchaeon from an Australian solar salternrdquo International Jour-nal of Systematic Bacteriology vol 43 no 4 pp 729ndash734 1993

[59] P D Franzman E Stackebrandt K Sanderson et al ldquoHalobac-terium lacusprofundi sp nov a halophilic bacterium isolatedfrom Deep Lake Antarcticardquo Systematic and Applied Microbi-ology vol 11 pp 20ndash27 1988

[60] G A Tomlinson and L I Hochstein ldquoHalobacterium saccharo-vorum sp nov a carbohydrate metabolizing extremely halo-philic bacteriumrdquoCanadian Journal ofMicrobiology vol 22 no4 pp 587ndash591 1976

[61] A Ventosa M C Gutierrez M Kamekura and M L Dyall-Smith ldquoProposal to transfer Halococcus turkmenicus Halobac-terium trapanicum JCM9743 and strainGSL-11 toHaloterrigenaturkmenica gen nov comb novrdquo International Journal ofSystematic Bacteriology vol 49 no 1 pp 131ndash136 1999

[62] M Kamekura and M L Dyall-Smith ldquoTaxonomy of the familyHalobacteriaceae and the description of two new genera Halo-rubrobacterium and Natrialbardquo Journal of General and AppliedMicrobiology vol 41 no 4 pp 333ndash350 1995

[63] T J Carver KM RutherfordM BerrimanM A RajandreamB G Barrell and J Parkhill ldquoACT the Artemis comparisontoolrdquo Bioinformatics vol 21 no 16 pp 3422ndash3423 2005

[64] S W Cline W L Lam R L Charlebois L C Schalkwyk andW F Doolittle ldquoTransformationmethods for halophilic archae-bacteriardquo Canadian Journal of Microbiology vol 35 no 1 pp148ndash152 1989

Archaea 5

Table 2 PH1 virus proteins identified by mass spectroscopy of tryptic peptides

Proteina Locus tag ORF MW (kDa) Matching peptide massesbObserved (calculated)

VP1 HhPH1 gp12 12 185 (158) 16VP2 HhPH1 gp24 24 100 (78) 15VP3 HhPH1 gp28 28 40 (38) 5VP4 HhPH1 gp21 21 35 (26) 5VP7 HhPH1 gp20 20 24 (20) 5VP9 HhPH1 gp27 27 16 (17) 3VP10 HhPH1 gp26 26 15 (17) 4VP12 HhPH1 gp19 19 7 (99) 4aVirus capsid proteins are numbered according to their similarity with SH1 capsid proteinsbThenumber of peptidemasses between 7500 and 35130mz identified byMALDI-TOFMS that correspond to the theoretical tryptic peptides of the predictedvirus protein

212 -

(kD

a)

PH1

158 -97 -66 -56 -42 -35 -

27 -

20 -

14 -

65 -

lowast

lowast

lowast

lowast

lowast

lowast

lowast

VP1VP2

VP3VP4

VP7

VP9

VP10

VP12

(a)

SH1

(b)

Figure 4 Structural proteins of halovirus PH1 Viral structuralproteinswere separated by SDS-PAGEon a 12 (wv) acrylamide geland stained with Brilliant Blue G (a) They were run in parallel withthe proteins of purified SH1 virus (b) The sizes of protein standardmarkers are indicated on the left side (in kDa) Asterisks denoteproteins bands of PH1 that were identified by mass spectroscopyThe numbering of proteins of PH1 (VP1ndashVP12) follows that of theSH1 homologs seen in (b) (and explained also in the text)

staining to be detected Further evidence was obtained by anuclease protection assay as previously used for SH1 DNA[31] As shown in Table 3 exonuclease III (a 31015840 exonuclease)was able to digest nonproteinase K-treated PH1 DNA butT7 exonuclease (a 51015840 exonuclease) and Bal31 (51015840 and 31015840exonucleases) were unable to digest PH1 DNA unless it hadbeen previously treated with proteinase KThis indicated thatboth 51015840 termini of the genome had protein attached

27 Sequence of the PH1 Genome The complete PH1 genomesequence was determined (accession KC252997) using a

1 2 3

-23

-76

-17410 -

6 -

3 -

2 -

15 -

05 -

4 -

(kB)

Figure 5 Detection of proteins bound to the termini of PH1 DNADNA was extracted from purified PH1 particles without proteinaseK digested with AseI then passed through a silica filter (GFC)under conditionswhere proteinswould bind to the filter Lane 1AseIfragments that did not bind and were eluted Lane 2 AseI fragmentsthat bound and were eluted only after protease treatment Lane 3AseI digest of PH1 DNA The positions of DNA size standards areindicated on the left side in kb The calculated sizes of the AseIfragments of PH1 DNA in lane 3 are indicated on the right side (alsoin kb) The predicted 062 kb PH1 AseI fragment was not detected

combination of cloned fragments PCR primer walkingon virus DNA and 454 whole genome sequencing (seeSection 4) It was found to be 28072 bp in length 676G + C with inverted terminal repeat sequences (ITRs) of337 bp Using GLIMMER [34] manual BLAST searching atthe GenBank database and comparison to related virusesthe PH1 genome sequence was predicted to contain 49 ORFs(Table 4) Most ORFs were closely spaced or overlappinggiving a gene density of 174 geneskb (average of 572 ntgene)The cumulative AT-skew plot of the PH1 genome shown atthe bottom of Figure 6 shows inflection points (circled) thatare consistent with the changes in transcription direction

6 Archaea

6 8 10 12 14 16 18 20 22 24 kb

1 2 4 6 8 10 12 14 162 kb

VP1

VP1-like fragments

A

A

VP2

VP2

VP7

VP1

2

20 21 22

VP4

VP1

3

VP5 p2

2

p9p4 p11

p12

p13

NP5

356A

Hmuk

p26

VP3

VP1

7

1728Hmuk

0477

Locus tag ZOD2009 19038

Locus tag HlacAJ 010100019872 NZ AGFZ01000123 (reversed)

NZ AEMG01000027 contig29

TransposaseHlacAJ 010100019767

ZOD2009 19153

Hbf lacisalsi

Hap paucihalophilus

(a)

VP18VP2VP1 A

VP1

VP18VP1

VP1

2

VP1

VP7

VP4

VP1

3

VP5

VP2

VP9

VP1

0

VP3

VP1

6

VP1

7V

P6 VP18

ATPa

se

kb1 2 4 6 8 10 12 14 16 18 20 22 24 26 28

VP2

VP2

VP18A

AA

2211 1312

94 26

HHIV-2

SH1

PH1

0

2

4

6

30 578 nt

30 889 nt

28 072 nt

(times102

)Cu

m A+

T sk

ew

Cumulative A + T skew of PH1

(b)

Figure 6 Genome alignments of haloviruses PH1 SH1 and HHIV-2 along with two related genomic loci (a) Genomic loci of Happaucihalophilus andHbf lacisalsi that contain genes related to halosphaeroviruses SH1 PH1 andHHIV-2The names andGenBank accessionsfor these contigs are given on the far right andORFs are coloured and labeled to indicate the relationships of theseORFs to those of the virusesbelowThe locus tag numbers for the first and last ORFs shown in each locus are given nearby their respectiveORFs In addition grey colouredORFs represent sequences that do not match any of the haloviruses and green coloured ORFs represent protein sequences that are closelyrelated to ORFs found within or very close to previously described virusplasmid loci so called ViPREs [20] The scale bars shown aboveeach contig show the position of the described region within the respective contig (b) The three virus genomes are labeled at the left withscale markers below (in kb) and the total length indicated at the far right At the bottom is a cumulative AT-skew plot of the PH1 genome(httpmolbiol-toolscaJie Zheng) with inflection points circledThe grey shaded bands between the genome diagrams indicate significantnucleotide similarity (usingACT [63]) AnnotatedORFs are represented by arrows with colours indicating structural proteins (red or brown)nonstructural proteins (yellow or orange) or the packaging ATPase (blue) The names of structural protein ORFs are indicated either withinthe arrow (eg VP1) or in text nearby The numbered orange coloured ORFs of HHIV-2 are homologous to ORFs found in the genomic loci(probably proviruses or provirus remnants) pictured in (a)

indicated by the annotated ORFs as has been seen in otherhaloviruses [7] There are no GATC motifs in the genomeand the inverse sequence CTAG is strongly underrepre-sented with just 2 motifs (68 expected)

When compared to the related viruses SH1 and HHIV-2(Figure 6) the PH1 genome is seen to be of similar length andis much more closely related to SH1 than to HHIV-2 (74and 54 nt identity resp) Like PH1 the genomes of SH1and HHIV-2 lack GATC motifs and CTAG is either absent(SH1) or underrepresented (HHIV-2) The ITRs of PH1 andSH1 share 785nucleotide identity but the PH1 ITR is longerthan that of SH1 (337 and 309 nt resp) partly because SH1has replaced an 18 bp sequence (gtcgtgcggtttcggcgg) found atthe internal end of its left-hand ITR with a sequence at thecorresponding position of its right-hand ITR that shows littleinverted sequence similarity In the PH1 ITR this sequenceis retained at both ends Whether this difference represents arecombination event or a mistake in replication of the ITRs

is unclear An alignment of the ITRs of all three virusesidentified a number of highly conserved regions (labeled 1ndash9 in Supplementary Figure 2 see Supplementary Materialavailable online at httpdxdoiorg1011552013456318) A16 bp sequence at the termini (region 1) is conserved con-sistent with the conservation seen at the termini of linearStreptomyces plasmids [35] A 16 bpGC-rich sequence aroundthe middle of the ITR (region 3) is also conserved Shortermotifs of either C-rich (region 8) or AT-rich sequence(regions 7 and 9) are found near the internal end of the ITRs

The grey shading between the schematic virus genomesin Figure 6(b) indicates regions of high nucleotide similarityrevealing that the differences between PH1 and SH1 are notevenly distributed but vary considerably along the lengthof the aligned genomes A comparison between SH1 andHHIV-2 has been published recently [13] and since the PH1and SH1 genomes are so similar we will focus the followingdescription on the differences between PH1 and SH1

Archaea 7

Table 3 Digestion of the PH1 genome by nucleasesa

Nuclease (amount) Proteinase K-treated UntreatedControl PH1 DNA Control PH1 DNA

DNase I (RNase-free)(100U) +b + +b +

Exonuclease III(100U) +b + +b +

Mung bean nuclease(10U) +c

minus +cminus

Nuclease BAL-31(1 U) +b + +b

minus

RNase A(5 120583gmL) +d

minus +dminus

T7 exonuclease(10U) +b + +b

minus

aExtracted nucleic acid from purified PH1 virus was treated with variousnucleases and then analysed by gel electrophoresis to detect whether thenuclease digested (+) or did not digest (minus) the genome Control nucleicacids used to confirm activity of the nucleases were b

120582 DNA ca DNAoligonucleotide and dyeast transfer RNA

Within corresponding ORFs there are blocks of codingsequence with low (or no) similarity These may represent insitu divergence or short recombination events (eg indels)There are also cases where an entire ORF in one virus hasno corresponding homolog in the other because of an inser-tiondeletion event (indel) Lastly there are replacementswhere the sequences within the corresponding regions showlow or no similarity but are flanked by sequence with highsimilarity The largest visible differences seen in Figure 6 aredue to sequence changes within or near the long genes encod-ing capsid proteins VP1 and VP18 A summary of the mainsequence differences between SH1 and PH1 and their loca-tions is given in Table 5 where these divergent regions (DV)are numbered from DV1 to DV18

The sequences of the two viruses are close enough that inmany cases the mechanism for the observed differences canbe inferred For example DV2 is likely the result of a deletionevent that has removed the SH1 ORF10 gene homolog fromthe PH1 genome At either end of the SH1 ORF are almostperfect direct repeats (cggcctgaccggcatgac) that would allowrepeat-mediated deletion to occur removing the interveningsequence Small direct repeats leading to deletions have beendescribed previously in the comparative analysis of Halo-quadratumwalsbyi genomes [20] DV5 appears to be an indelwhere SH1 ORFs 14ndash16 are absent in PH1 and an invertedrepeat (AGCCATG) found at each end of the SH1 divergentregion may be significant in the history of this changeThe proteins specified by SH1 ORFs 14ndash16 are presumablydispensable for PH1 (or their functions supplied by otherproteins) but a homolog of SH1 ORF14 is found also inHHIV-2 (ORF 6) and a homolog of SH1 ORF15 occurs inhalovirus His1 (ORF13) Replacement regions are also com-mon and can occur within ORFs (eg DV3 in capsid proteinVP1) or provide additional or alternative genes (eg DV12and DV13) Several divergent regions occur within the genefor capsid protein VP2 a hot spot for change because of the

repetitive nature of the coding sequence which specifies aprotein with runs of glycines andmany heptapeptide repeatsas described previously for SH1 VP2 [27] DV18 is a largereplacement that significantly alters the predicted length ofthe minor capsid protein VP18 in the two viruses (865519aa for SH1PH1 resp) and also provides SH1 with 3 ORFsthat are different or absent in PH1 ORFs 52 53 and 54 Forexample a homolog of SH1 ORF52 is not present in PH1 butis present in HHIV-2 (putative protein 40) While ORF48 ofPH1 shows no aa sequence homology to SH1 ORFs 53 and 54all of these predicted proteins carry CxxC motifs suggestiveof related functions such as DNA binding activity [36]

The genes of the PH1 genome are syntenic with those ofSH1 and are found in similarly oriented blocks of closelyspaced or overlapping ORFs that suggest that transcription isorganised into operons The programme of transcription ofthe SH1 genome has been reported previously [30] and thesequences in PH1 corresponding to the six promoter regions(P1ndashP6) determined in SH1 showed that five of these (P1ndashP5)are well conserved Only the region corresponding to SH1 P6was poorly conserved (data not shown) but this promoter isstrongly regulated in SH1 and only switches on late in infec-tion [30]

28 Structural Protein Genes The genes coding for the majorvirus structural proteins of PH1 (VP 1ndash4 7 9 10 and 12)were identified byMALDI-TOF (see above) and the genes forthese proteins are found in the same order and approximatepositions as in the SH1 genome (Figure 6(b) red ORFs)Genes coding for minor structural proteins can be deducedby sequence comparison with SH1 so that VP5 and VP6 arelikely to be encoded by PH1ORFs 25 and 29 respectively andVP13 and VP18 by ORFs 23 and 49 respectively This wouldgive PH1 a total of 11 structural proteins the same as SH1Of these VP7 is the most conserved between the two viru-ses (98 aa identity) followed by VP4 VP9 VP3 and VP12(91ndash94 identity) VP5 VP6 and VP13 (82ndash88 iden-tity) VP2 (77) VP1 (65) and lastly the least conservedprotein was VP18 (31)

29 Genomic Loci Related to PH1 and Other Halosphaeroviru-ses Clusters of halosphaerovirus-related genes are present inthe genomes of two recently sequenced haloarchaea Happaucihalophilus and Hbf lacisalsi (Figure 6(a)) Neither ofthese regions appears to represent complete virus genomebut they retain a number of genes that share a similar seq-uence and synteny with the virus genomes depicted belowthem (Figure 6(a)) Both loci show a mixed pattern ofrelatedness to SH1PH1 and HHIV-2 as indicated by ORFsof matching colour and gene name If these genomic locirepresent provirus integrants that have decayed over timethen the relationships they show suggest not only that halo-sphaeroviruses are a diverse virus group but also that asignificant level of recombination occurs between them lead-ing to mosaic gene combinations The two ORFs colouredgreen in theHap paucihalophilus locus (Figure 6) are relatedto ORFs found within or very close to genomic loci ofvirusplasmid genes found in other haloarchaea suggestingadditional links between (pro)viruses [20]

8 Archaea

Table4ORF

anno

tatio

nsof

theh

aloviru

sPH1g

enom

e

ORF

aPo

sitionb

Locustag

Leng

thc(aa)

pKi

Similaritycharacteris

tics

144

1ndash605

HhP

H1gp01

5043

76aa

similarityto

SH1O

RF2(YP271859)

2602ndash802

HhP

H1gp02

66109

80aa

similarityto

SH1O

RF3(YP271860)

3802ndash1059

HhP

H1gp03

8539

85aa

similarityto

SH1O

RF4(YP271861)44

sim

ilarityto

HHIV-2

protein1

41056ndash1283

HhP

H1gp04

7539

85aa

identityto

SH1O

RF5(YP271862)Predictedcoiled-coilregion

51276ndash1686

HhP

H1gp05

136

49

80aa

similarityto

SH1O

RF6(YP271863)Tw

otransm

embraned

omainsA

lsorelated

toHHIV-2

protein2(38

AFD

02283)

andto

Halobiform

alacisalsiORF

ZP09950018

(35

)6

1683ndash1829

HhP

H1gp06

4861

81aa

similarityto

SH1O

RF7(YP271864

)Predictedsig

nalsequence(Sign

alP)

71822ndash2130

HhP

H1gp07

102

58

91aa

similarityto

SH1O

RF8(YP271865)a

nd86to

HHIV-2

putativ

eprotein

3(A

FD02284)O

ther

homologsinclude

Nocardiaproteinpn

f2110(YP1220601)H

rrlacusprofund

iHlac0751

(YP0025654211)ORF

sinactin

ophage

VWB

(AAR2

9707)Frankia(EAN116

57)andgp40

ofMycobacteriu

mph

ageD

ori(AER

47690)

82123ndash2383

HhP

H1gp08

8642

90aa

similarityto

SH1O

RF9(YP271866)

92380ndash2714

HhP

H1gp09

111

52

88aa

similarityto

SH1O

RF11(YP271868)and

also

similarityto

putativ

eprotein

4of

HHIV-2

(AFD

02285)

HlacA

J19877of

Hbflacisa

lsiAJ

5(ZP09950016)

102861ndash300

4HhP

H1gp10

47104

Weaksim

ilarityto

SH1O

RF12

(YP271869)o

nlyover

theN

-term

inal18

resid

ues

113029ndash3100

HhP

H1gp11

23123

Not

presentinSH

1orH

HIV-2

123134ndash7453

HhP

H1gp12

1439

42

Capsid

protein

VP172sim

ilarityto

SH1V

P1(O

RF13Y

P271870)HHIV-2

VP1

(AFD

02286)H

lacA

J19872Hbflacisa

lsiAJ

5Predictedhelix-tu

rn-helixandRu

vA-like

domain(In

terproScan)

137505ndash8227

HhP

H1gp13

240

54

Predicted

P-loop

ATPa

sedo

main(C

OG0433)92aa

similarityto

SH1O

RF17

(YP271874)a

ndalso

similarityto

putativ

eAT

Pase

ofHHIV-2

(AFD

02288)A

TPaseo

fHaladapatus

paucihalophilusD

X253

(ZP0804

6180)HlacA

J19867of

Hbf

lacisalsi(ZP09950014)

148419ndash8228c

HhP

H1gp14

6351

43aa

similarityto

SH1O

RF18

(YP271875)Alanine-richCentraltransm

embraned

omain(Pho

bius)

158856ndash8

416c

HhP

H1gp15

146

41

92aa

similarityto

SH1O

RF19

(YP271876)a

ndalso

toHHIV-2

putativ

eprotein

9(A

FD02290)H

lacA

J19857of

Hbf

lacisalsiAJ

5ZO

D2009

19093of

Happau

cihalophilus

168853ndash9

500c

HhP

H1gp16

215

39

65aa

similarityto

SH1O

RF20

(YP271877)a

ndalso

topu

tativ

eprotein

11of

HHIV-2

(AFD

02292)Z

OD2009

19098of

Hap

paucihalophilusHlacA

J19852Hbflacisa

lsiAJ

5C-

term

inaltransm

embraned

omain(Pho

bius)

179500ndash9

919c

HhP

H1gp17

139

41

79aa

similarityto

SH1O

RF21

(YP271878)a

ndalso

topu

tativ

eprotein

12of

HHIV-2

(AFD

02293)hypotheticalprotein

HlacA

J19847of

Hbflacisa

lsi(ZP09950010)Transm

embraned

omain(Pho

bius)

189923ndash10594c

HhP

H1gp18

223

43

83aa

similarityto

SH1O

RF22

(YP271879)a

ndalso

topu

tativ

eprotein

13of

HHIV-2

(AFD

02294)Z

OD2009

19108of

Hap

paucihalophilusHlacA

J19842of

Hbflacisa

lsiP

redicted

signalsequence(sig

nalP)Con

tains4

CxxC

motifs

1910659ndash10943

HhP

H1gp19

94104

Capsid

protein

VP12

(ORF

19)91aa

similarityto

SH1V

P12(O

RF23Y

P271880)a

ndalso

toVP12of

HHIV-2

(AFD

02295)

andHlacA

J19837of

Hbflacisa

lsiZ

OD2009

19113of

Happau

cihalophilusTw

otransm

embraned

omains

(Pho

bius)

2010960ndash

11517

HhP

H1gp20

185

44

Capsid

protein

VP7(O

RF20)98aa

similarityto

SH1V

P7(O

RF24Y

P271881)a

ndalso

toVP7

ofHHIV-2

(AFD

02296)

ZOD2009

19118

ofHappau

cihalophilusHlacA

J19832of

Hbflacisa

lsiAJ

5

2111519ndash

12217

HhP

H1gp21

232

41

Capsid

protein

VP4(O

RF21)94aa

similarityto

SH1O

RF25

(YP271882)a

ndalso

toVP4

ofHHIV-2

(AFD

02297)

ZOD2009

19123of

Happau

cihalophilusHlacA

J19827of

Hbflacisa

lsiAJ

522

12233ndash12454

HhP

H1gp22

7339

81aa

similarityto

SH1O

RF26

(YP271883)a

ndalso

topu

tativ

eprotein

17of

HHIV-2

(AFD

02298)

2312458ndash12697

HhP

H1gp23

7948

78aa

similarityto

SH1capsid

protein

VP13

(ORF

27Y

P271884)a

ndalso

toVP13of

HHIV-2

(AFD

02299)P

redicted

coil-coildo

main

PredictedC-

term

inaltransm

embraned

omain(Pho

bius)

Archaea 9

Table4Con

tinued

ORF

aPo

sitionb

Locustag

Leng

thc(aa)

pKi

Similaritycharacteris

tics

2412701ndash1506

4HhP

H1gp24

787

39

Capsid

protein

VP2(O

RF24)75aa

similarityto

SH1V

P2(O

RF28Y

P271885)a

ndto

VP2

ofHHIV-2

(AFD

02300)

ZOD2009

19133of

Happau

cihalophilus(ZP

0804

6189)

2515065ndash15961

HhP

H1gp25

298

45

Putativec

apsid

protein

VP5(O

RF25)81aa

similarityto

SH1V

P5(O

RF29Y

P271886)a

ndto

HHIV-2

VP5

(AFD

02301)

HlacA

J19807of

HbflacisalsiAJ

5ZO

D2009

19133of

Happau

cihalophilus

261596

4ndash1644

6HhP

H1gp26

160

46

73aa

similarityto

SH1capsid

protein

VP10

(ORF

30Y

P271887)a

ndto

VP10of

HHIV-2

(AFD

02302)Z

OD2009

19138of

Happau

cihalophilusHlacA

J19802of

Hbflacisa

lsiP

redicted

C-term

inaltransm

embraned

omain(TMHMM)

271644

6ndash16895

HhP

H1gp27

149

42

Capsid

protein

VP9(O

RF27)94aa

similarityto

SH1V

P9(O

RF31Y

P271888)a

ndto

ZOD2009

19143of

Hap

paucihalophilus(ZP

0804

6191)

2816908ndash17921

HhP

H1gp28

337

43

Capsid

protein

VP3(O

RF28)91aa

similarityto

SH1V

P3(O

RF32Y

P271889)a

ndto

NJ7G

2365

ofNa

trinemaspJ7-2

2917928ndash18617

HhP

H1gp29

229

42

Capsidprotein

VP6(O

RF29)83aa

similarityto

SH1V

P6(O

RF33Y

P271890)Also

toNJ7G

3194

ofNa

trinemaspJ7-2and

Hmuk

0476

ofHalom

icrobium

mukohataeiPredictedcarboxypeptid

aser

egulatory-lik

edom

ain(C

arbo

xypepD

reg

pfam

13620)

3018614ndash

18943

HhP

H1gp30

109

49

71aa

similarityto

SH1O

RF34

(YP271891)a

ndto

putativ

eprotein

27of

HHIV-2

(AFD

02308)P

redicted

signalsequence

(signalP)

3119054ndash

19176c

HhP

H1gp31

4050

TwoCx

xCmotifsN

oho

molog

inSH

1orH

HIV-2

3219173ndash19289c

HhP

H1gp32

3841

68aa

similarityto

SH1O

RF37

(YP271894)

3319286ndash

19552c

HhP

H1gp33

8853

58aa

similarityto

SH1O

RF39

(YP271896)a

ndto

HHIV-2

putativ

eprotein

29(A

FD02310)Fou

rCxxCmotifs

3419549ndash

19731c

HhP

H1gp34

6070

Con

tainsC

xxCmotif

Noho

molog

inSH

1orH

HIV-2

3519728ndash20219c

HhP

H1gp35

163

56

93aa

similarityto

SH1O

RF41

(YP271898)a

ndto

putativ

eprotein

30of

HHIV-2

(AFD

02311)

3620216ndash

21415c

HhP

H1gp36

399

50

95aa

similarityto

SH1O

RF42

(YP271899)a

ndto

putativ

eprotein

31of

HHIV-2

(AFD

02312)

3721419ndash

21778c

HhP

H1gp37

11941

91aa

similarityto

SH1O

RF43

(YP271900)a

ndto

putativ

eprotein

32of

HHIV-2

(AFD

02313)

3821762ndash2300

6cHhP

H1gp38

414

42

83aa

similarityto

SH1O

RF44

(YP271901)a

ndto

putativ

eprotein

33of

HHIV-2

(AFD

02314)P

redicted

coil-coiland

helix-tu

rn-helixdo

mains

3923003ndash23158c

HhP

H1gp39

5165

69aa

similarityto

SH1O

RF45

(YP271902)

4023161ndash23370c

HhP

H1gp40

6935

74aa

similarityto

SH1O

RF46

(YP271903)and55sim

ilarityto

putativ

eprotein

35of

HHIV-2

(AFD

02316)

4123422ndash23586c

HhP

H1gp41

5476

86aa

similarityto

SH1O

RF47

(YP271904)Con

tains2

CxxC

motifsand

show

ssim

ilarityto

proteindo

mainfamily

PF14206Arginine-ric

h

4223583ndash24023c

HhP

H1gp42

48

77aa

similarityto

SH1O

RF48

(YP271905)a

ndto

putativ

eprotein

36of

HHIV-2

(AFD

02317)H

ham114540of

Hcc

hamelinensis(ZP11271999)Ph

iCh1p72of

Natrialba

phageP

hiCh

1(NP665989)DUF4

326(pfam14216)

family

domain

4324020ndash

24538c

HhP

H1gp43

172

47

79aa

similarityto

SH1O

RF49

(YP271906)and56sim

ilarityto

putativ

eprotein

37of

HHIV-2

(AFD

02318)

4424535ndash25053c

HhP

H1gp44

172

43

92aa

similarityto

SH1O

RF50

(YP271907)and72sim

ilarityto

putativ

eprotein

38of

HHIV-2

(AFD

02319)

4525050ndash

25277c

HhP

H1gp45

7543

Noho

molog

inSH

1orH

HIV-2

4625274ndash

25387c

HhP

H1gp46

3748

Noho

molog

inSH

1orH

HIV-2

4725533ndash25904

HhP

H1gp47

123

43

61aa

similarityto

SH1O

RF51

(YP271908)66sim

ilarityto

putativ

eprotein

39of

HHIV-2

(AFD

02320)P

redicted

COG1342

domain(D

NAbind

inghelix-tu

rn-helix)

4825901ndash26092c

HhP

H1gp48

6349

Con

tainsC

xxCmotif

Noho

molog

inSH

1orH

HIV-2

4926089ndash

2764

8cHhP

H1gp49

506

40

81aa

similarityto

SH1O

RF55

(YP271912)(bu

tonlyin

theN

-term

inalhalf)H

omolog

ofvirusstru

cturalprotein

VP18

ofHHIV-2

(AFD

02323)

a ORF

swerep

redicted

either

byGLIMMER

orby

manualsearching

forh

omologsintheG

enBa

nkdatabase

b Startandendpo

sitions

ofORF

sare

give

inbp

numbera

ccording

totheP

H1sequenced

epositedatGenBa

nk(KC2

52997)O

RFso

nthec

omplem

entary

strandared

enoted

bythes

uffixc

c Lengthof

thep

redicted

ORF

innu

mbero

faminoacids

10 Archaea

Table 5 Main differences (divergent regions) between the SH1 and PH1 genomes

RegionaSH1 startb SH1 stop Length (bp) PH1 start PH1 stop Length (bp) Comment

DV1 527 650 123 538 604 66

Replacement Both regions have direct repeats at their bordersPH1 has two sets GACCCGGC and CGCTGC while SH1 hasCCCGAC In SH1 this replacement region includes theC-terminal region of ORF2 and the N-terminal region ofORF3 In PH1 it covers only the C-terminal region of ORF1

DV2 2433 2588 155 2386 2387 mdash RMDc from PH1 SH1 repeat at border is TGACCGThisremoves the homolog of SH1 ORF10 from PH1

DV3 3324 3598 274 3137 3416 279 Replacement at the beginning of SH1 ORF13PH1 ORF12(capsid protein VP1)

DV4 6656 7128 472 6478 7085 607 Replacement near the C-terminus of SH1 ORF13PH1 ORF12(capsid protein VP1)

DV5 7491 8341 850 7446 7447 mdash

Indel that results in ORFs14-16 of SH1 not being present inPH1 SH1 ORF14 has a close homolog in HHIV-2 (ORF6) in asimilar position just after the VP1 homolog SH1 ORF15 is aconserved protein in haloarchaea (eg NP 2552A andHmuk 2978) and halovirus His1 (ORF13) Possible invertedrepeat (AGCCATG) at border of SH1 region

DV6 13705 13706 mdash 12818 12851 33RMD from SH1 PH1 repeat at border is CAGCGG(gt)G Thisremoves a part of capsid protein VP2 sequence from the SH1protein

DV7 13789 13863 74 12934 12941 7 Replacement This occurs within VP2 gene of both viruses

DV8 15142 15186 44 14220 14221 mdash RMD from PH1 SH1 repeat at border is TG(tc)CCGACGAand occurs within capsid protein VP2 gene of both viruses

DV9 15303 15317 14 14337 14338 mdash RMD from PH1 SH1 repeat at border is GCCGACGA andoccurs within capsid protein VP2 gene of both viruses

DV10 15504 15527 23 14529 14530 mdash RMD from PH1 SH1 repeat at border is GACGA and occurswithin capsid protein VP2 of both viruses

DV11 20026 20405 379 19019 19173 154Replacement beginning at the start of SH1 ORF35PH1 ORF31This region is longer in SH1 where it includes ORF36 an ORFthat has no homolog present in PH1

DV12 20440 20661 221 19208 19286 78 Replacement This is unequal and includes an ORF in SH1(ORF38) that is not present in PH1

DV13 20943 21085 142 19562 19728 166 Replacement This covers SH1 ORF40PH1 ORF34 Thepredicted proteins are not homologous

DV14 23541 23542 mdash 22171 22197 27 Probable RMD in SH1 PH1 repeat at border is CGTCTCGGand occurs in SH1 ORF44PH1 ORF38

DV15 24506 24521 15 23152 23153 mdash Probable RMD in PH1 SH1 repeat at border is CTCGGT andoccurs near the end of SH1 ORF45PH1 ORF39

DV16 24951 24964 13 23579 23580 mdash Probable RMD in PH1 SH1 repeat at border is CGGTC andoccurs in SH1 ORF47PH1 ORF41

DV17 26449 26450 mdash 25050 25274 224Probable RMD in SH1 PH1 repeat at border is TCATGCG andoccurs near the start of SH1 ORF50PH1 ORF44 and providesan extra ORF for PH1 (ORF45)

DV18 27074 29764 690 25895 26933 1038

Replacement Left border at start of SH1 ORF51PH1 ORF47and extends rightwards into SH1 ORF55PH1 ORF49 (VP18gene) It is an unequal replacement and SH1 has two moreORFs in this region than PH1

aDV Divergent regions between the genomes of SH1 and PH1bStart and stop positions refer to the GenBank sequences of the two viruses SH1 NC 0072171 PH1 KC252997cRMD repeat-mediated deletion event as described in [20]

The recently described halovirus SNJ1 carries manyORFsthat have homologs adjacent to a previously described ViPREofHmc mukohataei (from here on denoted by ViPREHmuk1)a locus that contains genes related to His2 (and other ple-olipoviruses) as well as toHaloferax plasmid pHK2 (probablya provirus) and to two small plasmids ofHqr walsbyi (sim6 kbpL6A and pL6B)The latter relationship provides wider linksto other viruses because one of the PL6 genes (Hqrw 6002)

has homologs in some pleolipoviruses (HPRV-3 and HGPV-1) while another gene (Hqrw 6005) is related to ORF16 ofthe spindle-shaped halovirus His1 [15] After adding the SNJ1gene homologs to ViPREHmuk1 it is extended significantlyand a close inspection of the flanking regions revealed atRNA-ala gene at one end and a partial copy of the sametRNA-ala gene (next to a phage integrase gene) at the otherend (supplementary Figure 1) Genes beyond these points

Archaea 11

Table 6 Transfection of haloarchaea by PH1 DNA

Speciesa Efficiency of transfectiontransformation withPH1 DNAb (PFU120583g of DNA) pUBP2 DNAbc (CFU120583g of DNA)

Har hispanica 53 plusmn 05 times 103 14 plusmn 08 times 104

Har marismortui 45 plusmn 07 times 103 18 plusmn 01 times 104

ldquoHar sinaiiensisrdquo 32 plusmn 06 times 102 mdashHalobacterium salinarum mdash 54 plusmn 42 times 103

Haloferax gibbonsii 28 plusmn 08 times 103 40 plusmn 26 times 103

Haloferax lucentense mdash 57 plusmn 11 times 103

Hfx volcanii mdash 21 plusmn 07 times 105

Hrr lacusprofundi 70 plusmn 22 times 102 95 plusmn 15 times 103

Haloterrigena turkmenica 16 plusmn 17 times 102 24 plusmn 06 times 103

Natrialba asiatica 35 plusmn 39 times 102 31 plusmn 10 times 104aOnly those species positive for transfection andor plasmid transformation are shownbRates of transfection by (nonprotease treated) PH1 DNA or transformation by plasmid pUB2 are averages of three independent experiments each performedin duplicate (plusmn standard deviation)cTransformants were selected on plates with 2 4 or 6120583gmL simvastatin depending on the strainmdash no plaques or colonies observed

appear to be related to cellular metabolism ViPREHmuk1is now 39379 nt in length flanked by a 59 nt direct repeat(potential att sites) and includes 54 ORFs many of which arerelated to known haloviruses are virus-like (eg integrasesmethyltransferases transcriptional regulators DNA methyl-transferase and DNA glycosylase) are homologs in or nearother knownViPREs or show little similarity to other knownproteins This locus also includes an ORC1CDC6 homolog(Hmuk 0446) which could provide a replication function

The mixture of very different virus and plasmid genesseen within ViPREHmuk1 is remarkable but it also revealshomologs found in or adjacent to genomic loci of otherspecies such as the gene homologs of Har marismortuiseen in the left end of ViPREHmuk1 (supplementary Figure1) When these genes are added to the previously describedHaloarcula locus (ViPREHmuk1) it is also extended signifi-cantly It becomes 25550 bp in length encompasses genesfrom Hmar 2382 to Hmar 2404 and is flanked by a fulltRNA-ala gene at one end and a partial copy at the other Itcarries an ORC1CDC6 homolog an integrase and a phiH-like repressor as well as pleolipovirus homologs

210 Transfection of Haloarchaea by PH1 DNA PH1 DNAwas introduced into Har hispanica cells using the PEGmethod [37] and the cells were screened for virus productionby plaque assay Figure 7 shows the increase in transfectedcells with input (nonproteinase K-treated) virus DNA Theestimated efficiencywas 53times 103 PFUper120583gDNA PH1DNAthat had been treated with proteinase K or with DNAase I(RNase-free) did not produce plaques (data not shown) PH1DNA was found to transfect six species of haloarchaea otherthanHar hispanica includingmembers of the generaHaloar-cula Haloferax Halorubrum Haloterrigena and Natrialba(Table 6) For these experiments the virus production fromtransfected cells was detected by plaque assay on indicatorlawns of Har hispanica

0

0

200 400 600 800 1000

Titre

(PFU

mL)

104

103

102

PH1 DNA (ng)

Figure 7 Transfection ofHar hispanica cells with PH1 DNA Vary-ing amounts of nonproteinase K-treated PH1 DNAwere introducedinto cells of Har hispanica using the PEG method [64] Cells werethen screened for infective centres by plaque assay Data shown is theaverage of three independent experiments performed in duplicateError bars represent one standard deviation of themean If protease-treated DNA was used no transfectant plaques were observed

3 Discussion

PH1 is very similar in particlemorphology genome structureand sequence to the previously described halovirus SH1 [816] Like SH1 it has long inverted terminal repeat sequencesand terminal proteins indicative of protein-primed replica-tion Purified PH1 particles have a relatively low buoyant

12 Archaea

density and are chloroform sensitive and show a layeredcapsid structure consistent with the likely presence of aninternalmembrane layer as has been shown for SH1HHIV-2and SNJ1 [13 26]The particle stability of PH1 to temperaturepH and reduced salt was similar to SH1 [8] The structuralproteins of PH1 are very similar in sequence and relativeabundance to those of SH1 so the two viruses are likely toshare the same particle geometry [16]

Viruses that use protein-primed replication such as bac-teriophageΦ29 carry a viral type BDNApolymerase that caninteract specifically with the proteins attached to the genomictermini and initiate strand synthesis [38] ArchaeovirusesHis1 [7] and Acidianus bottle-shaped virus [39] probablyuse this mode of replication However a polymerase genecannot be found in the genomes of PH1 SH1 or HHIV-2DNA polymerases are large enzymes and the only ORF ofSH1 that was long enough to encode such an enzyme andhad not previously been assigned as a structural protein wasORF55 which specified an 865 aa protein that contained noconserved domains indicative of polymerases [27]The recentstudy of HHIV-2 showed that the corresponding ORF in thisvirus (gene 42) is a structural protein of the virion and byinference this is likely also for the corresponding proteinsof SH1 and PH1 (no homolog is present in SNJ1) Withouta polymerase gene these viruses must use a host enzymebut searches for a viral type B DNA polymerase in thegenome of the hostHar hispanica or in the genomes of othersequenced haloarchaea did not find anymatchesThis arguesstrongly for a replication mechanism that is different to thatexemplified byΦ29 An attractive alternative is that displayedby Streptomyces linear plasmids which have 51015840-terminal pro-teins and use a cellular polymerase for replication [40] Theterminal proteins are not used for primary replication but forend patching [41] and it has been shown that if the term-inal repeat sequences are removed from these plasmids andthe ends ligated they can replicate as circular plasmids [42]This provides a testable hypothesis for the replication of halo-sphaeroviruses and also offers a pathway for switchingbetween linear (eg PH1) and circular (eg SNJ1) forms Useof a host polymerase would also fit with the ability of SH1 andPH1 DNA to transfect many different haloarchaeal species[31] as these enzymes and theirmode of action are highly con-served

The growth characteristics of PH1 inHar hispanica bothby single-cell burst and single-step growth experiments gavevalues of 50ndash100 virusescell significantly less than that ofSH1 (200 virusescell) [8] and HHIV-2 (180 virusescell) [13]The latter virus is lytic whereas SH1 PH1 and SNJ1 start pro-ducing extracellular viruswell before any decrease in cell den-sity indicating that virus release can occur without lysisThishas been most clearly shown with SH1 where almost 100of cells can be infected [8] Infection ofHar hispanica by PH1was less efficient (sim30 of cells) but the kinetics of virus pro-duction in single-step growth curves of SH1 and PH1 aresimilar with virus production occurring over several hoursrather than at a clearly defined time after infection The twoviruses are also very closely related and infect the same hostspecies so it is likely that they use the same scheme for cellexit For both viruses the cell density eventually decreases in

single-step cultures showing that virus infection does resultin cell death This mode of exit appears to maximise theproduction of the virus over time as the host cells survive foran extended period In natural hypersaline waters cell num-bers are usually high but growth rates are low [12] and thehigh incident UV (on often shallow ponds) would damagevirus DNA so reducing the half-life of released virus part-icles [43] These factors may have favoured the exit strategydisplayed by these viruses improving their chance of trans-mission to a new host

The motif GATC was absent in the genomes of PH1 SH1andHHIV-2 and themotif CTAGwas either absent or greatlyunderrepresented All three viruses share the same host butGATC and CTAG motifs are plentiful in the Har hispanicagenome (accessions CP002921 and CP002923) Avoidance ofthesemotifs in halovirusesHis1 His2 HF1 andHF2 has beenreported previously [7] Such purifying selection commonlyresults from host restriction enzymes and inHfx volcanii itis exactly these two motifs that are targeted [44] One Hfxvolcanii enzyme recognises A-methylated GATC sites [45]and another recognises un-methylated CTAG sites (whichare protected by methylation in the genome) In the currentstudy this could explain the negative transfection results forHfx volcanii as the PH1 genome contains two CTAGmotifsBy comparison the SH1 genome contains no CTAG motifsand is able to transfect Hfx volcanii [31]

Comparison of the PH1 and SH1 genomes allowed adetailed picture of the natural variation occurring betweenclosely related viruses revealing likely deletion events medi-ated by small repeats a process described previously in astudy of Hqr walsbyi and termed repeat-mediated deletion[20] There are also many replacements including blocks ofsequence that occur within long open reading frames (suchas the capsid protein genes VP1 and VP18) VP3 and VP6 areknown to form the large spikes on the external surface of theSH1 virion and presumably one or both interact with host cellreceptors [16] PH1 and SH1 have identical host ranges con-sistent with the high sequence similarity shown between theircorresponding VP3 and VP6 proteins

A previous study of SH1 could not detect transcriptsacross annotated ORFs 1ndash3 or 55-56 [30] ORFs 1 and 56occur in the terminal inverted repeat sequence and in thepresent study it was found that there were no ORFs cor-responding to these in the PH1 genome Given the close simi-larity of the two genomes the comparative data are in agree-ment with the transcriptional data indicating that these SH1ORFs are not used SH1 ORFs 2 and 3 are more problematicas good homologs of these are also present in PH1 The con-flict between the transcriptional data and the comparativegenomic evidence requires further experimental work toresolveThe status of SH1ORF55 has recently been confirmedby studies of the related virus HHIV-2 (discussed above)

The pleolipovirus group has expanded dramatically inthe last few years and it now comprises a diverse group ofviruseswith different genome types and replication strategiesWhat is evenmore remarkable is that a similar expansion cannow be seen with SH1-related viruses which includes viruseswith at least two replication modes and genome types (linearand circular) Evidence from haloarchaeal genome sequences

Archaea 13

show cases of not only provirus integrants or plasmids (egpKH2 and pHH205) but also genomic loci (ViPREs) thatcontain cassettes of virus genes from different sources andat least two cases (described in this study) have likely attsites that indicate circularisation and mobility While thereis a clear relationship between viruses with circular ds- andssDNA genomes that replicate via the rolling circle method(ie the ssDNA form is a replication intermediate) it ismore difficult to explain how capsid gene cassettes can movebetween these viruses and those with linear genomes andterminal proteins Within the pleolipoviruses His2 has alinear genome that contains a viral type B DNA polymeraseand terminal proteins while all the other described membershave circular genomes that contain a rep homolog and prob-ably replicate via the rolling circle method [14 19] Similarlyamong the halosphaeroviruses PH1 SH1 and HHIV-2 havelinear dsDNA with terminal proteins and probably replicatein the same way but the related SNJ1 virus has a circulardsDNA genome The clear relationships shown by the capsidgenes of viruses within each group plus the connectionsshown to haloviruses outside each group (eg the DNA poly-merases of His1 and His2) all speak of a vigorous meansof recombination one that can readily switch capsid genesbetween viruses with radically different replication strategiesHow is the process most likely to operate One possibility issuggested by ViPREs which appear to be mobile collectionsof capsid and replication genes from different sources Theyoffer fixed locations for recombination to occur providegene cassettes that can be reassorted to produce novel virusgenomes and some can probably recombine out as circularforms For example ViPREHmuk1 carries genes related topleolipoviruses halosphaeroviruses and other haloviruses Itis yet to be determined if any of the genes carried in theseloci are expressed in the cell (such as the genes for capsidproteins) or if ViPREs provide any selective advantage to thehost

4 Materials and Methods

41 Water Sample A water sample was collected in 1998from Pink Lake (32∘001015840 S and 115∘301015840 E) a hypersalinelake on Rottnest Island Western Australia Australia It wasscreened in the same year for haloviruses using the methodsdescribed in [8] A single plaque on a Har hispanica lawnplate was picked and replaque purified The novel haloviruswas designated PH1

42 Media Strains and Plasmids The media used in thisstudy are described in the online resource The Halohand-book (httpwwwhaloarchaeacomresourceshalohandbookindexhtml) Artificial salt water containing 30 (wv) totalsalts comprised of 4M NaCl 150mM MgCl

2 150mM

MgSO4 90mM KCl and 35mM CaCl

2and adjusted to pH

75 using sim2mL 1M Tris-HCl (pH 75) per litre MGM con-taining 12 18 or 23 (wv) total salts and HVD (halo-virus diluent) were prepared from the concentrated stock aspreviously described [46] Bacto-agar (Difco Laboratories)was added to MGM for solid (15 gL) or top-layer (7 gL)media

Table 1 lists the haloarchaea used in this study Allhaloarchaea were grown aerobically at 37∘C in either 18 or23 (wv)MGM(depending on the strain) andwith agitation(except for ldquoHar sinaiiensisrdquo)The plasmids used were pBlue-script II KS+ (Stratagene Cloning Systems) pUBP2 [47]and pWL102 [48] pUBP2 and pWL102 were first passagedthrough E coli JM110 [49] to prevent dam-methylationof DNA which has been shown to reduce transformationefficiency in some strains [45]

43 Negative-Stain Electron Microscopy The method fornegative-stain TEM was adapted from that of V Tarasovdescribed in the online resource The Halohandbook (httpwwwhaloarchaeacomresourceshalohandbookindexhtml)A 20120583L drop of the sample was placed on a clean surfaceand the virus particles were allowed to adsorb to a Formvarfilm 400-mesh copper grid (ProSciTech) for 15ndash2min Theywere then negatively stained with a 20120583L drop of 2 (wv)uranyl acetate for 1-2min Excess liquid was absorbed withfilter paper and the grid was allowed to air dry Grids wereexamined either on a Philips CM 120 BioTwin transmissionelectron microscope (Royal Philips Electronics) operating atan accelerating voltage of 120 kV or on a Siemens Elmiskop102 transmission electron microscope (Siemens AG) operat-ing at an accelerating voltage of 120 kV

44 Virus Host Range Twelve haloarchaeal strains from thegenera Haloarcula Halobacterium Haloferax HalorubrumandNatrialba (Table 1)Thirteen naturalHalorubrum isolates(H Camakaris unpublished data) and five uncharacterizedhaloarchaeal isolates fromLakeHardy Pink Lake (inWesternAustralia) and Serpentine Lake (D Walker and M K Seahunpublished data) were screened for PH1 susceptibilityLysates from PH1-infected Har hispanica cultures (1 times1011 PFUmL) were spotted onto lawns of each strain (usingmedia with 12 or 18 (wv)MGM depending on the strain)and incubated for 2ndash5 days at 30 and at 37∘C

45 Large Scale Virus Growth and Purification Liquid cul-tures of PH1 were grown by the infection (MOI 005) of anearly exponential Har hispanica culture in 18 (wv) MGMCultures were incubated aerobically at 37∘C with agitationfor 3 days Clearing (ie complete cell lysis) of PH1-infectedcultures did not occur and consequently cultures were harv-ested when the absorbance at 550 nm reached the minimumand the titre (determined by plaque assay) was at the maxi-mum usually sim1010ndash1011 PFUmL Virus was purified usingthe method described previously for SH1 [8 31] except thatafter the initial low speed spin (Sorvall GSA 6000 rpm30min 10∘C) virus was concentrated from the infected cul-ture by centrifugation at 26000 rpm (13 hr 10∘C) onto acushion of 30 (wv) sucrose in HVDThe pellet was resus-pended in a small volume of HVD and loaded onto a pre-formed linear 5ndash70 (wv) sucrose gradient followed byisopycnic centrifugation in 13 gmL CsCl (Beckman 70Ti60000 rpm 20 hr 10∘C) The white virus band occurredat a density of 129 gmL and was collected and diluted inhalovirus diluent (HVD see Section 4) and the virus pelleted(Beckman SW55 35000 rpm 75min 10∘C) and resuspended

14 Archaea

in a small volume of HVD and stored at 4∘C Virus recoveryat the major stages of a typical purification is given insupplementary Table 1 The specific infectivity of pure virussolutions was determined as the ratio of the PFUmL to theabsorbance at 260 nm

46 PH1 Single-Step Growth Curve An early exponentialphase culture of Har hispanica grown in 18 (wv) MGMwas infected with PH1 (MOI 50) Under these conditions thepercentage of infected cells was approximately 30 After anadsorption period of 1 hr at 37∘C the cells were washed threetimes with 18 (wv) MGM (at room temperature) resus-pended in 100mL 18 (wv) MGM (these methods ensuredthe removal of all residual virus) and incubated at 37∘Cwith shaking (100 rpm) Samples were removed at hourlyintervals for measurements of absorbance at 550 nm andthe number of infective centres Immediately after samplingtitres were determined by plaque assay with an indicator lawnof Har hispanica Each experiment was performed in tripli-cate

47 Halovirus Stability After various treatments sampleswere removed and diluted in HVD and virus titres weredetermined by plaque assay on Har hispanica Each experi-ment was performed in triplicate and representative data areshown Chloroform sensitivity was examined by the exposureof PH1 lysates in 18 (wv) MGM to chloroform in a volumeratio of 1 4 (chloroform to lysate) Incubation was at roomtemperature with constant agitation At appropriate timepoints the mix was allowed to settle and a sample wasremoved from the upper layer and diluted in HVDThe effectof a low ionic environment was examined by diluting PH1lysates (in 18 (wv) MGM) into double distilled H

2O in a

volume ratio of 1 1000 (lysate to double distilledH2O) Incu-

bation was at room temperature with constant agitationSamples were removed at various times and diluted in HVDThe pH stability of PH1 was determined by dilution of PH1lysates in 18 (wv)MGMin the appropriate pHbuffer (HVDbuffered with appropriate Tris-HCl) in a volume ratio of1 100 (lysate to buffer) Incubation was at room temperaturewith constant agitation After 30min samples were removedand diluted in HVD Thermal stability of PH1 was examinedby a 1 hr incubation of a virus lysate in 18 (wv) MGM atdifferent temperatures after which they were brought quicklyto room temperature diluted in HVD and titrated

48 Protein Procedures To remove salts purified virus prepa-rations were mixed with trichloroacetic acid (10 (vv) finalconcentration) and incubated on ice for 15min to allowthe proteins to precipitate After centrifugation (16000 g15min room temperature) the precipitate was washed threetimes in acetone dried and resuspended in double distilledH2O Proteins were dissolved in Laemmli sample buffer with

48mM 120573-mercaptoethanol [50] heated in boiling water for5min then separated on 12 (wv) NuPAGE Novex Bis-Tris Gels using MES-SDS running buffer according to themanufacturerrsquos directions (Invitrogen) After electrophoresisgels were rinsed in double distilled H

2O and stained with

01 (wv) Brilliant Blue G in 40 (vv) methanol and

10 (vv) acetic acid Gels were destained with severalchanges of 40 (vv) methanol and 10 (vv) acetic acidAlternatively gels were stained with GelCode GlycoproteinStaining Kit Pro-Q Emerald 488 Glycoprotein Gel and BlotStain Kit SYPRO Ruby Protein Gel Stain according tothe manufacturerrsquos directions (Invitrogen Molecular ProbesPierce Biotechnology)

Protein bands were cut from the gels and sent to theAustralian Proteome Analysis Facility (Macquarie Univer-sity) for trypsin digestion and analysis by matrix assistedlaser desorption ionisation time of flight mass spectrometry(MALDI-TOF MS) on an Applied Biosystems 4700 Pro-teomics Analyser (Applied Biosystems)

49 DNA Procedures Proteinase K-treated and nonpro-teinaseK-treatedDNApreparationsweremade frompurifiedPH1 using the methods described previously for SH1 [8]DNA was separated on 1 (wv) agarose gels in Tris-acetate-EDTA electrophoresis buffer and was stained with ethidiumbromide (Sigma-Aldrich)120582 DNA DNase I (RNase-free) exonuclease III mung

bean nuclease nuclease BAL-31 T4 DNA polymerase T4DNA ligase T7 exonuclease and type II restriction endonu-cleases were purchased from New England Biolabs RNase Aand yeast transfer RNA were purchased from Sigma-AldrichProteinase K was purchased from Promega Oligonucleotideprimers were purchased from Geneworks Australia

To clone fragments of the PH1 genome purified virusDNA was digested either with AccI Eco0109I and MseI orSmaI restriction endonucleases blunted with DNA Poly-merase I Large (Klenow) Fragment where appropriate andligated into AccI- Eco0109I- or SmaI-digested pBluescript IIKS+ using T4 DNA ligase according to the manufacturerrsquosinstructions (New England Biolabs) The DNA was intro-duced intoE coliXL1-Blue and transformants grown onLuriaagar containing 15 120583gmL tetracycline and 100 120583gmL ampi-cillinThe resulting cloneswere sequenced and the sequenceswere used to design specific oligonucleotide primers for PCRamplification andor primer walking using the virus genome(or specific restriction fragments) as templates

To amplify PH1 nucleic acid approximately 10 ng virusDNA was combined with 500 nM primers 100120583M of eachdNTP 1U Deep Vent DNA Polymerase (New England Bio-labs) and 1 timesThermoPol buffer (New England Biolabs) ina total reaction volume of 50 120583L Template was denatured at95∘C for 10min followed by 30 cycles of denaturation at 95∘Cfor 30 sec annealing at 56∘C for 30 sec extension at 75∘C for2min and a final extension at 75∘C for 10min PCR reactionswere performed on a PxE02ThermoCycler (Thermo ElectroCorporation) Sequencing reactions using 32 pmol primerwere performed by the dideoxy chain termination methodusing ABI PRISM Big Dye Terminator Mix version 31 onan ABI 3100 capillary sequencer (Applied Biosystems) bythe Applied Genetic Diagnostics Sequencing Service at theDepartment of Pathology (The University of Melbourne)

To obtain terminal genomic fragments PH1 DNA wasdigested by the enzyme Pas1 and the head (sim700 bp) andthe tail (sim1300 bp) fragments were purified by the QIAEXII gel extraction kits and treated with 05M piperidine for

Archaea 15

2 hr at 37∘C to remove protein residues The piperidine-treated fragments were sequenced by the Genomics BiotechCompany (Taipei Taiwan) using two primers TGACCA-ATTAATTAGGCCGGTTCGCC (PH1 head-R) and GTGC-CATACTGCTACAATTCT (PH1 tail-F)

Four hundred fifty-four whole genome sequencing PH1DNA samples (sim5 120583g) were sequenced using parallel pyrose-quencing on a Roche 454 Genome Sequencer System atMission Biotech (Taipei Taiwan) The largest contig was27399with 8803 reads (Newbler version 27) with 27387 posi-tions being of Q40 quality The average depth of coverage foreach base was 829TheGenBank accession for the entire PH1sequence is KC252997

Acknowledgments

The authors thank Carolyn Bath Mei Kwei (Joan) Seahand Danielle Walker for their early work in PH1 characteri-zation Simon Crawford Jocelyn Carpenter and Anna Fried-huber for their assistance with the transmission electronmicroscopy and Tiffany Cowie and Voula Kanellakis fortheir sequencing assistance This paper has been facilitatedby access to the Australian Proteome Analysis Facility estab-lished under the Australian Governmentrsquos Major NationalResearch Facilities program SLT was supported by a grantfrom the Biodiversity Research Center Academia SinicaTaiwan

References

[1] M Krupovic M F White P Forterre and D PrangishvilildquoPostcards from the edge structural genomics of archaealvirusesrdquo Advances in Virus Research vol 82 pp 33ndash62 2012

[2] P Stolt andW Zillig ldquoGene regulation in halophage 120601Hmdashmorethan promotersrdquo Systematic and Applied Microbiology vol 16no 4 pp 591ndash596 1994

[3] R Klein U Baranyi N Rossler B Greineder H Scholz and AWitte ldquoNatrialba magadii virus 120593Ch1 first complete nucleotidesequence and functional organization of a virus infecting ahaloalkaliphilic archaeonrdquo Molecular Microbiology vol 45 no3 pp 851ndash863 2002

[4] E Pagaling R D Haigh W D Grant et al ldquoSequence analysisof an Archaeal virus isolated from a hypersaline lake in InnerMongolia Chinardquo BMC Genomics vol 8 article 410 2007

[5] S L Tang SNuttall andMDyall-Smith ldquoHalovirusesHF1 andHF2 evidence for a recent and large recombination eventrdquo Jour-nal of Bacteriology vol 186 no 9 pp 2810ndash2817 2004

[6] C Bath and M L Dyall-smith ldquoHis1 an archaeal virus of theFuselloviridae family that infectsHaloarcula hispanicardquo Journalof Virology vol 72 no 11 pp 9392ndash9395 1998

[7] C Bath T Cukalac K Porter andM L Dyall-Smith ldquoHis1 andHis2 are distantly related spindle-shaped haloviruses belongingto the novel virus group Salterprovirusrdquo Virology vol 350 no1 pp 228ndash239 2006

[8] K Porter P Kukkaro J K H Bamford et al ldquoSH1 a novelspherical halovirus isolated from an Australian hypersalinelakerdquo Virology vol 335 no 1 pp 22ndash33 2005

[9] M Dyall-Smith S L Tang and C Bath ldquoHaloarchaeal viruseshow diverse are theyrdquo Research in Microbiology vol 154 no 4pp 309ndash313 2003

[10] A Oren G Bratbak and M Heldal ldquoOccurrence of virus-likeparticles in the Dead Seardquo Extremophiles vol 1 no 3 pp 143ndash149 1997

[11] T Sime-Ngando S Lucas A Robin et al et al ldquoDiversityof virus-host systems in hypersaline Lake Retba SenegalrdquoEnvironmental Microbiology vol 13 no 8 pp 1956ndash1972 2010

[12] N Guixa-Boixareu J I Calderon-Paz M Heldal G Bratbakand C Pedros-Alio ldquoViral lysis and bacterivory as prokaryoticloss factors along a salinity gradientrdquoAquaticMicrobial Ecologyvol 11 no 3 pp 215ndash227 1996

[13] S T Jaakkola R K Penttinen S T Vilen et al ldquoClosely relatedarchaeal Haloarcula hispanica icosahedral viruses HHIV-2 andSH1 have nonhomologous genes encoding host recognitionfunctionsrdquo Journal of Virology vol 86 no 9 pp 4734ndash47422012

[14] E Roine P Kukkaro L Paulin et al ldquoNew closely related halo-archaeal viral elements with different nucleic acid typesrdquo Jour-nal of Virology vol 84 no 7 pp 3682ndash3689 2010

[15] A Sencilo L Paulin S KellnerMHelm and E Roine ldquoRelatedhaloarchaeal pleomorphic viruses contain different genometypesrdquo Nucleic Acids Research vol 40 no 12 pp 5523ndash55342012

[16] H T Jaalinoja E Roine P Laurinmaki H M Kivela D HBamford and S J Butcher ldquoStructure and host-cell interactionof SH1 a membrane-containing halophilic euryarchaeal virusrdquoProceedings of the National Academy of Sciences of the UnitedStates of America vol 105 no 23 pp 8008ndash8013 2008

[17] M K Pietila N S Atanasova VManole et al ldquoVirion architec-ture unifies globally distributed pleolipoviruses infecting halo-philic archaeardquo Journal of Virology vol 86 no 9 pp 5067ndash50792012

[18] M L Holmes and M L Dyall-Smith ldquoA plasmid vector witha selectable marker for halophilic archaebacteriardquo Journal ofBacteriology vol 172 no 2 pp 756ndash761 1990

[19] M L Holmes F Pfeifer andM L Dyall-Smith ldquoAnalysis of thehalobacterial plasmid pHK2 minimal repliconrdquo Gene vol 153no 1 pp 117ndash121 1995

[20] M L Dyall-Smith F Pfeiffer K Klee et al ldquoHaloquadratumwalsbyi limited diversity in a Global Pondrdquo PLoS ONE vol 6no 6 Article ID e20968 2011

[21] S Casjens ldquoProphages and bacterial genomics what have welearned so farrdquoMolecular Microbiology vol 49 no 2 pp 277ndash300 2003

[22] R W Hendrix M C M Smith R N Burns M E Ford andG F Hatfull ldquoEvolutionary relationships among diverse bacte-riophages and prophages all the worldrsquos a phagerdquo Proceedings ofthe National Academy of Sciences of the United States of Americavol 96 no 5 pp 2192ndash2197 1999

[23] D Botstein ldquoA theory ofmodular evolution for bacteriophagesrdquoAnnals of the New York Academy of Sciences vol 354 pp 484ndash491 1980

[24] M Krupovic S Gribaldo D H Bamford and P Forterre ldquoTheevolutionary history of archaeal MCM helicases a case study ofvertical evolution combinedwithHitchhiking ofmobile geneticelementsrdquo Molecular Biology and Evolution vol 27 no 12 pp2716ndash2732 2010

[25] M Krupovic P Forterre and D H Bamford ldquoComparativeanalysis of the mosaic genomes of tailed archaeal viruses andproviruses suggests common themes for virion architecture andassembly with tailed viruses of bacteriardquo Journal of MolecularBiology vol 397 no 1 pp 144ndash160 2010

16 Archaea

[26] Z Zhang Y Liu S Wang et al et al ldquoTemperate membrane-containing halophilic archaeal virus SNJ1 has a circular dsDNAgenome identical to that of plasmid pHH205rdquoVirology vol 434no 2 pp 233ndash241 2012

[27] D H Bamford J J Ravantti G Ronnholm et al ldquoConstituentsof SH1 a novel lipid-containing virus infecting the halophiliceuryarchaeonHaloarcula hispanicardquo Journal of Virology vol 79no 14 pp 9097ndash9107 2005

[28] H M Kivela E Roine P Kukkaro S Laurinavicius P Somer-harju andDH Bamford ldquoQuantitative dissociation of archaealvirus SH1 reveals distinct capsid proteins and a lipid corerdquoVirology vol 356 no 1-2 pp 4ndash11 2006

[29] H Jaalinoja Electron Cryo-Microscopy Studies of Bacteriophage8 and Archaeal Virus SH1 University of Helsinki HelsinkiFinland 2007

[30] K Porter B E Russ J Yang andM L Dyall-Smith ldquoThe trans-cription programme of the protein-primed halovirus SH1rdquoMi-crobiology vol 154 no 11 pp 3599ndash3608 2008

[31] K Porter and M L Dyall-Smith ldquoTransfection of haloarchaeaby the DNAs of spindle and round haloviruses and the useof transposon mutagenesis to identify non-essential regionsrdquoMolecular Microbiology vol 70 no 5 pp 1236ndash1245 2008

[32] S E Luria General Virology John Wiley and Sons New YorkNY USA 1953

[33] C A Thomas K Saigo E McLeod and J Ito ldquoThe separationofDNA segments attached to proteinsrdquoAnalytical Biochemistryvol 93 pp 158ndash166 1979

[34] A L Delcher D Harmon S Kasif OWhite and S L SalzbergldquoImproved microbial gene identification with GLIMMERrdquoNucleic Acids Research vol 27 no 23 pp 4636ndash4641 1999

[35] R Zhang Y Yang P Fang et al ldquoDiversity of telomere palin-dromic sequences and replication genes among Streptomyceslinear plasmidsrdquo Applied and Environmental Microbiology vol72 no 9 pp 5728ndash5733 2006

[36] X Wang H S Lee F J Sugar F E Jenney M W W Adamsand J H Prestegard ldquoPF0610 a novel winged helix-turn-helix variant possessing a rubredoxin-like Zn ribbon motiffrom the hyperthermophilic archaeon Pyrococcus furiosusrdquoBiochemistry vol 46 no 3 pp 752ndash761 2007

[37] S W Cline L C Schalkwyk and W F Doolittle ldquoTransfor-mation of the archaebacterium Halobacterium volcanii withgenomic DNArdquo Journal of Bacteriology vol 171 no 9 pp 4987ndash4991 1989

[38] S Kamtekar A J Berman J Wang et al ldquoThe 12059329 DNA poly-merase protein-primer structure suggests a model for the ini-tiation to elongation transitionrdquo EMBO Journal vol 25 no 6pp 1335ndash1343 2006

[39] X Peng T Basta M Haring R A Garrett and D PrangishvilildquoGenome of theAcidianus bottle-shaped virus and insights intothe replication and packaging mechanismsrdquo Virology vol 364no 1 pp 237ndash243 2007

[40] C C Yang C H Huang C Y Li Y G Tsay S C Lee and CW Chen ldquoThe terminal proteins of linear Streptomyces chrom-osomes and plasmids a novel class of replication primingproteinsrdquo Molecular Microbiology vol 43 no 2 pp 297ndash3052002

[41] C C Yang Y H Chen H H Tsai C H Huang T W Huangand C W Chen ldquoIn vitro deoxynucleotidylation of the term-inal protein of Streptomyces linear chromosomesrdquo Applied andEnvironmental Microbiology vol 72 no 12 pp 7959ndash79612006

[42] P C Chang and S N Cohen ldquoBidirectional replication froman internal origin in a linear Streptomyces plasmidrdquo Science vol265 no 5174 pp 952ndash954 1994

[43] S W Wilhelm W H Jeffrey C A Suttle and D L MitchellldquoEstimation of biologically damaging UV levels in marinesurface waters with DNA and viral dosimetersrdquo Photochemistryand Photobiology vol 76 no 3 pp 268ndash273 2002

[44] T Allers and M Mevarech ldquoArchaeal geneticsmdashthe third wayrdquoNature Reviews Genetics vol 6 no 1 pp 58ndash73 2005

[45] M L Holmes S D Nuttall and M L Dyall-Smith ldquoConstruc-tion and use of halobacterial shuttle vectors and further studieson Haloferax DNA gyraserdquo Journal of Bacteriology vol 173 no12 pp 3807ndash3813 1991

[46] S D Nuttall and M L Dyall-Smith ldquoHF1 and HF2 novelbacteriophages of halophilic archaeardquo Virology vol 197 no 2pp 678ndash684 1993

[47] U Blaseio and F Pfeifer ldquoTransformation of Halobacteriumhalobium development of vectors and investigation of gas vesi-cle synthesisrdquo Proceedings of the National Academy of Sciences ofthe United States of America vol 87 no 17 pp 6772ndash6776 1990

[48] W L Lam and W F Doolittle ldquoShuttle vectors for the archae-bacterium Halobacterium volcaniirdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 86 no14 pp 5478ndash5482 1989

[49] C Yanisch-Perron J Vieira and J Messing ldquoImproved M13phage cloning vectors and host strains nucleotide sequences ofthe M13mp18 and pUC19 vectorsrdquo Gene vol 33 no 1 pp 103ndash119 1985

[50] U K Laemmli ldquoCleavage of structural proteins during theassembly of the head of bacteriophage T4rdquo Nature vol 227 no5259 pp 680ndash685 1970

[51] D G Burns H M Camakaris P H Janssen and M L Dyall-Smith ldquoCombined use of cultivation-dependent and culti-vation-independent methods indicates that members of mosthaloarchaeal groups in anAustralian crystallizer pond are culti-vablerdquo Applied and Environmental Microbiology vol 70 no 9pp 5258ndash5265 2004

[52] G Juez F Rodriguez-Valera A Ventosa and D J KushnerldquoHaloarcula hispanica spec nov and Haloferax gibbonsii specnov two new species of extremely halophilic archaebacteriardquoSystematic and Applied Microbiology vol 8 pp 75ndash79 1986

[53] A Oren M Ginzburg B Z Ginzburg L I Hochstein and BE Volcani ldquoHaloarcula marismortui (Volcani) sp nov nomrev an extremely halophilic bacterium from the Dead SeardquoInternational Journal of Systematic Bacteriology vol 40 no 2pp 209ndash210 1990

[54] M Torreblanca F Rodriguez-Valera G Juez A Ventosa MKamekura and M Kates ldquoClassification of non-alkaliphilichalobacteria based on numerical taxonomy and polar lipidcomposition and description of Haloarcula gen nov andHaloferax gen novrdquo Systematic and Applied Microbiology vol8 pp 89ndash99 1986

[55] A Ventosa and A Oren ldquoHalobacterium salinarum nom cor-rig a name to replace Halobacterium salinarium (Elazari-Vol-cani) and to include Halobacterium halobium and Halobac-terium cutirubrumrdquo International Journal of Systematic Bacte-riology vol 46 no 1 p 347 1996

[56] M C Gutierrez M Kamekura M L Holmes M L Dyall-Smith and A Ventosa ldquoTaxonomic characterization of Halo-ferax sp (ldquoH alicanteirdquo) strain Aa 22 description of Haloferaxlucentensis sp novrdquo Extremophiles vol 6 no 6 pp 479ndash4832002

Archaea 17

[57] M F Mullakhanbhai and H Larsen ldquoHalobacterium volcaniispec nov a Dead Sea halobacterium with a moderate saltrequirementrdquo Archives of Microbiology vol 104 no 1 pp 207ndash214 1975

[58] S D Nuttall and M L Dyall-Smith ldquoCh2 a novel halophilicarchaeon from an Australian solar salternrdquo International Jour-nal of Systematic Bacteriology vol 43 no 4 pp 729ndash734 1993

[59] P D Franzman E Stackebrandt K Sanderson et al ldquoHalobac-terium lacusprofundi sp nov a halophilic bacterium isolatedfrom Deep Lake Antarcticardquo Systematic and Applied Microbi-ology vol 11 pp 20ndash27 1988

[60] G A Tomlinson and L I Hochstein ldquoHalobacterium saccharo-vorum sp nov a carbohydrate metabolizing extremely halo-philic bacteriumrdquoCanadian Journal ofMicrobiology vol 22 no4 pp 587ndash591 1976

[61] A Ventosa M C Gutierrez M Kamekura and M L Dyall-Smith ldquoProposal to transfer Halococcus turkmenicus Halobac-terium trapanicum JCM9743 and strainGSL-11 toHaloterrigenaturkmenica gen nov comb novrdquo International Journal ofSystematic Bacteriology vol 49 no 1 pp 131ndash136 1999

[62] M Kamekura and M L Dyall-Smith ldquoTaxonomy of the familyHalobacteriaceae and the description of two new genera Halo-rubrobacterium and Natrialbardquo Journal of General and AppliedMicrobiology vol 41 no 4 pp 333ndash350 1995

[63] T J Carver KM RutherfordM BerrimanM A RajandreamB G Barrell and J Parkhill ldquoACT the Artemis comparisontoolrdquo Bioinformatics vol 21 no 16 pp 3422ndash3423 2005

[64] S W Cline W L Lam R L Charlebois L C Schalkwyk andW F Doolittle ldquoTransformationmethods for halophilic archae-bacteriardquo Canadian Journal of Microbiology vol 35 no 1 pp148ndash152 1989

6 Archaea

6 8 10 12 14 16 18 20 22 24 kb

1 2 4 6 8 10 12 14 162 kb

VP1

VP1-like fragments

A

A

VP2

VP2

VP7

VP1

2

20 21 22

VP4

VP1

3

VP5 p2

2

p9p4 p11

p12

p13

NP5

356A

Hmuk

p26

VP3

VP1

7

1728Hmuk

0477

Locus tag ZOD2009 19038

Locus tag HlacAJ 010100019872 NZ AGFZ01000123 (reversed)

NZ AEMG01000027 contig29

TransposaseHlacAJ 010100019767

ZOD2009 19153

Hbf lacisalsi

Hap paucihalophilus

(a)

VP18VP2VP1 A

VP1

VP18VP1

VP1

2

VP1

VP7

VP4

VP1

3

VP5

VP2

VP9

VP1

0

VP3

VP1

6

VP1

7V

P6 VP18

ATPa

se

kb1 2 4 6 8 10 12 14 16 18 20 22 24 26 28

VP2

VP2

VP18A

AA

2211 1312

94 26

HHIV-2

SH1

PH1

0

2

4

6

30 578 nt

30 889 nt

28 072 nt

(times102

)Cu

m A+

T sk

ew

Cumulative A + T skew of PH1

(b)

Figure 6 Genome alignments of haloviruses PH1 SH1 and HHIV-2 along with two related genomic loci (a) Genomic loci of Happaucihalophilus andHbf lacisalsi that contain genes related to halosphaeroviruses SH1 PH1 andHHIV-2The names andGenBank accessionsfor these contigs are given on the far right andORFs are coloured and labeled to indicate the relationships of theseORFs to those of the virusesbelowThe locus tag numbers for the first and last ORFs shown in each locus are given nearby their respectiveORFs In addition grey colouredORFs represent sequences that do not match any of the haloviruses and green coloured ORFs represent protein sequences that are closelyrelated to ORFs found within or very close to previously described virusplasmid loci so called ViPREs [20] The scale bars shown aboveeach contig show the position of the described region within the respective contig (b) The three virus genomes are labeled at the left withscale markers below (in kb) and the total length indicated at the far right At the bottom is a cumulative AT-skew plot of the PH1 genome(httpmolbiol-toolscaJie Zheng) with inflection points circledThe grey shaded bands between the genome diagrams indicate significantnucleotide similarity (usingACT [63]) AnnotatedORFs are represented by arrows with colours indicating structural proteins (red or brown)nonstructural proteins (yellow or orange) or the packaging ATPase (blue) The names of structural protein ORFs are indicated either withinthe arrow (eg VP1) or in text nearby The numbered orange coloured ORFs of HHIV-2 are homologous to ORFs found in the genomic loci(probably proviruses or provirus remnants) pictured in (a)

indicated by the annotated ORFs as has been seen in otherhaloviruses [7] There are no GATC motifs in the genomeand the inverse sequence CTAG is strongly underrepre-sented with just 2 motifs (68 expected)

When compared to the related viruses SH1 and HHIV-2(Figure 6) the PH1 genome is seen to be of similar length andis much more closely related to SH1 than to HHIV-2 (74and 54 nt identity resp) Like PH1 the genomes of SH1and HHIV-2 lack GATC motifs and CTAG is either absent(SH1) or underrepresented (HHIV-2) The ITRs of PH1 andSH1 share 785nucleotide identity but the PH1 ITR is longerthan that of SH1 (337 and 309 nt resp) partly because SH1has replaced an 18 bp sequence (gtcgtgcggtttcggcgg) found atthe internal end of its left-hand ITR with a sequence at thecorresponding position of its right-hand ITR that shows littleinverted sequence similarity In the PH1 ITR this sequenceis retained at both ends Whether this difference represents arecombination event or a mistake in replication of the ITRs

is unclear An alignment of the ITRs of all three virusesidentified a number of highly conserved regions (labeled 1ndash9 in Supplementary Figure 2 see Supplementary Materialavailable online at httpdxdoiorg1011552013456318) A16 bp sequence at the termini (region 1) is conserved con-sistent with the conservation seen at the termini of linearStreptomyces plasmids [35] A 16 bpGC-rich sequence aroundthe middle of the ITR (region 3) is also conserved Shortermotifs of either C-rich (region 8) or AT-rich sequence(regions 7 and 9) are found near the internal end of the ITRs

The grey shading between the schematic virus genomesin Figure 6(b) indicates regions of high nucleotide similarityrevealing that the differences between PH1 and SH1 are notevenly distributed but vary considerably along the lengthof the aligned genomes A comparison between SH1 andHHIV-2 has been published recently [13] and since the PH1and SH1 genomes are so similar we will focus the followingdescription on the differences between PH1 and SH1

Archaea 7

Table 3 Digestion of the PH1 genome by nucleasesa

Nuclease (amount) Proteinase K-treated UntreatedControl PH1 DNA Control PH1 DNA

DNase I (RNase-free)(100U) +b + +b +

Exonuclease III(100U) +b + +b +

Mung bean nuclease(10U) +c

minus +cminus

Nuclease BAL-31(1 U) +b + +b

minus

RNase A(5 120583gmL) +d

minus +dminus

T7 exonuclease(10U) +b + +b

minus

aExtracted nucleic acid from purified PH1 virus was treated with variousnucleases and then analysed by gel electrophoresis to detect whether thenuclease digested (+) or did not digest (minus) the genome Control nucleicacids used to confirm activity of the nucleases were b

120582 DNA ca DNAoligonucleotide and dyeast transfer RNA

Within corresponding ORFs there are blocks of codingsequence with low (or no) similarity These may represent insitu divergence or short recombination events (eg indels)There are also cases where an entire ORF in one virus hasno corresponding homolog in the other because of an inser-tiondeletion event (indel) Lastly there are replacementswhere the sequences within the corresponding regions showlow or no similarity but are flanked by sequence with highsimilarity The largest visible differences seen in Figure 6 aredue to sequence changes within or near the long genes encod-ing capsid proteins VP1 and VP18 A summary of the mainsequence differences between SH1 and PH1 and their loca-tions is given in Table 5 where these divergent regions (DV)are numbered from DV1 to DV18

The sequences of the two viruses are close enough that inmany cases the mechanism for the observed differences canbe inferred For example DV2 is likely the result of a deletionevent that has removed the SH1 ORF10 gene homolog fromthe PH1 genome At either end of the SH1 ORF are almostperfect direct repeats (cggcctgaccggcatgac) that would allowrepeat-mediated deletion to occur removing the interveningsequence Small direct repeats leading to deletions have beendescribed previously in the comparative analysis of Halo-quadratumwalsbyi genomes [20] DV5 appears to be an indelwhere SH1 ORFs 14ndash16 are absent in PH1 and an invertedrepeat (AGCCATG) found at each end of the SH1 divergentregion may be significant in the history of this changeThe proteins specified by SH1 ORFs 14ndash16 are presumablydispensable for PH1 (or their functions supplied by otherproteins) but a homolog of SH1 ORF14 is found also inHHIV-2 (ORF 6) and a homolog of SH1 ORF15 occurs inhalovirus His1 (ORF13) Replacement regions are also com-mon and can occur within ORFs (eg DV3 in capsid proteinVP1) or provide additional or alternative genes (eg DV12and DV13) Several divergent regions occur within the genefor capsid protein VP2 a hot spot for change because of the

repetitive nature of the coding sequence which specifies aprotein with runs of glycines andmany heptapeptide repeatsas described previously for SH1 VP2 [27] DV18 is a largereplacement that significantly alters the predicted length ofthe minor capsid protein VP18 in the two viruses (865519aa for SH1PH1 resp) and also provides SH1 with 3 ORFsthat are different or absent in PH1 ORFs 52 53 and 54 Forexample a homolog of SH1 ORF52 is not present in PH1 butis present in HHIV-2 (putative protein 40) While ORF48 ofPH1 shows no aa sequence homology to SH1 ORFs 53 and 54all of these predicted proteins carry CxxC motifs suggestiveof related functions such as DNA binding activity [36]

The genes of the PH1 genome are syntenic with those ofSH1 and are found in similarly oriented blocks of closelyspaced or overlapping ORFs that suggest that transcription isorganised into operons The programme of transcription ofthe SH1 genome has been reported previously [30] and thesequences in PH1 corresponding to the six promoter regions(P1ndashP6) determined in SH1 showed that five of these (P1ndashP5)are well conserved Only the region corresponding to SH1 P6was poorly conserved (data not shown) but this promoter isstrongly regulated in SH1 and only switches on late in infec-tion [30]

28 Structural Protein Genes The genes coding for the majorvirus structural proteins of PH1 (VP 1ndash4 7 9 10 and 12)were identified byMALDI-TOF (see above) and the genes forthese proteins are found in the same order and approximatepositions as in the SH1 genome (Figure 6(b) red ORFs)Genes coding for minor structural proteins can be deducedby sequence comparison with SH1 so that VP5 and VP6 arelikely to be encoded by PH1ORFs 25 and 29 respectively andVP13 and VP18 by ORFs 23 and 49 respectively This wouldgive PH1 a total of 11 structural proteins the same as SH1Of these VP7 is the most conserved between the two viru-ses (98 aa identity) followed by VP4 VP9 VP3 and VP12(91ndash94 identity) VP5 VP6 and VP13 (82ndash88 iden-tity) VP2 (77) VP1 (65) and lastly the least conservedprotein was VP18 (31)

29 Genomic Loci Related to PH1 and Other Halosphaeroviru-ses Clusters of halosphaerovirus-related genes are present inthe genomes of two recently sequenced haloarchaea Happaucihalophilus and Hbf lacisalsi (Figure 6(a)) Neither ofthese regions appears to represent complete virus genomebut they retain a number of genes that share a similar seq-uence and synteny with the virus genomes depicted belowthem (Figure 6(a)) Both loci show a mixed pattern ofrelatedness to SH1PH1 and HHIV-2 as indicated by ORFsof matching colour and gene name If these genomic locirepresent provirus integrants that have decayed over timethen the relationships they show suggest not only that halo-sphaeroviruses are a diverse virus group but also that asignificant level of recombination occurs between them lead-ing to mosaic gene combinations The two ORFs colouredgreen in theHap paucihalophilus locus (Figure 6) are relatedto ORFs found within or very close to genomic loci ofvirusplasmid genes found in other haloarchaea suggestingadditional links between (pro)viruses [20]

8 Archaea

Table4ORF

anno

tatio

nsof

theh

aloviru

sPH1g

enom

e

ORF

aPo

sitionb

Locustag

Leng

thc(aa)

pKi

Similaritycharacteris

tics

144

1ndash605

HhP

H1gp01

5043

76aa

similarityto

SH1O

RF2(YP271859)

2602ndash802

HhP

H1gp02

66109

80aa

similarityto

SH1O

RF3(YP271860)

3802ndash1059

HhP

H1gp03

8539

85aa

similarityto

SH1O

RF4(YP271861)44

sim

ilarityto

HHIV-2

protein1

41056ndash1283

HhP

H1gp04

7539

85aa

identityto

SH1O

RF5(YP271862)Predictedcoiled-coilregion

51276ndash1686

HhP

H1gp05

136

49

80aa

similarityto

SH1O

RF6(YP271863)Tw

otransm

embraned

omainsA

lsorelated

toHHIV-2

protein2(38

AFD

02283)

andto

Halobiform

alacisalsiORF

ZP09950018

(35

)6

1683ndash1829

HhP

H1gp06

4861

81aa

similarityto

SH1O

RF7(YP271864

)Predictedsig

nalsequence(Sign

alP)

71822ndash2130

HhP

H1gp07

102

58

91aa

similarityto

SH1O

RF8(YP271865)a

nd86to

HHIV-2

putativ

eprotein

3(A

FD02284)O

ther

homologsinclude

Nocardiaproteinpn

f2110(YP1220601)H

rrlacusprofund

iHlac0751

(YP0025654211)ORF

sinactin

ophage

VWB

(AAR2

9707)Frankia(EAN116

57)andgp40

ofMycobacteriu

mph

ageD

ori(AER

47690)

82123ndash2383

HhP

H1gp08

8642

90aa

similarityto

SH1O

RF9(YP271866)

92380ndash2714

HhP

H1gp09

111

52

88aa

similarityto

SH1O

RF11(YP271868)and

also

similarityto

putativ

eprotein

4of

HHIV-2

(AFD

02285)

HlacA

J19877of

Hbflacisa

lsiAJ

5(ZP09950016)

102861ndash300

4HhP

H1gp10

47104

Weaksim

ilarityto

SH1O

RF12

(YP271869)o

nlyover

theN

-term

inal18

resid

ues

113029ndash3100

HhP

H1gp11

23123

Not

presentinSH

1orH

HIV-2

123134ndash7453

HhP

H1gp12

1439

42

Capsid

protein

VP172sim

ilarityto

SH1V

P1(O

RF13Y

P271870)HHIV-2

VP1

(AFD

02286)H

lacA

J19872Hbflacisa

lsiAJ

5Predictedhelix-tu

rn-helixandRu

vA-like

domain(In

terproScan)

137505ndash8227

HhP

H1gp13

240

54

Predicted

P-loop

ATPa

sedo

main(C

OG0433)92aa

similarityto

SH1O

RF17

(YP271874)a

ndalso

similarityto

putativ

eAT

Pase

ofHHIV-2

(AFD

02288)A

TPaseo

fHaladapatus

paucihalophilusD

X253

(ZP0804

6180)HlacA

J19867of

Hbf

lacisalsi(ZP09950014)

148419ndash8228c

HhP

H1gp14

6351

43aa

similarityto

SH1O

RF18

(YP271875)Alanine-richCentraltransm

embraned

omain(Pho

bius)

158856ndash8

416c

HhP

H1gp15

146

41

92aa

similarityto

SH1O

RF19

(YP271876)a

ndalso

toHHIV-2

putativ

eprotein

9(A

FD02290)H

lacA

J19857of

Hbf

lacisalsiAJ

5ZO

D2009

19093of

Happau

cihalophilus

168853ndash9

500c

HhP

H1gp16

215

39

65aa

similarityto

SH1O

RF20

(YP271877)a

ndalso

topu

tativ

eprotein

11of

HHIV-2

(AFD

02292)Z

OD2009

19098of

Hap

paucihalophilusHlacA

J19852Hbflacisa

lsiAJ

5C-

term

inaltransm

embraned

omain(Pho

bius)

179500ndash9

919c

HhP

H1gp17

139

41

79aa

similarityto

SH1O

RF21

(YP271878)a

ndalso

topu

tativ

eprotein

12of

HHIV-2

(AFD

02293)hypotheticalprotein

HlacA

J19847of

Hbflacisa

lsi(ZP09950010)Transm

embraned

omain(Pho

bius)

189923ndash10594c

HhP

H1gp18

223

43

83aa

similarityto

SH1O

RF22

(YP271879)a

ndalso

topu

tativ

eprotein

13of

HHIV-2

(AFD

02294)Z

OD2009

19108of

Hap

paucihalophilusHlacA

J19842of

Hbflacisa

lsiP

redicted

signalsequence(sig

nalP)Con

tains4

CxxC

motifs

1910659ndash10943

HhP

H1gp19

94104

Capsid

protein

VP12

(ORF

19)91aa

similarityto

SH1V

P12(O

RF23Y

P271880)a

ndalso

toVP12of

HHIV-2

(AFD

02295)

andHlacA

J19837of

Hbflacisa

lsiZ

OD2009

19113of

Happau

cihalophilusTw

otransm

embraned

omains

(Pho

bius)

2010960ndash

11517

HhP

H1gp20

185

44

Capsid

protein

VP7(O

RF20)98aa

similarityto

SH1V

P7(O

RF24Y

P271881)a

ndalso

toVP7

ofHHIV-2

(AFD

02296)

ZOD2009

19118

ofHappau

cihalophilusHlacA

J19832of

Hbflacisa

lsiAJ

5

2111519ndash

12217

HhP

H1gp21

232

41

Capsid

protein

VP4(O

RF21)94aa

similarityto

SH1O

RF25

(YP271882)a

ndalso

toVP4

ofHHIV-2

(AFD

02297)

ZOD2009

19123of

Happau

cihalophilusHlacA

J19827of

Hbflacisa

lsiAJ

522

12233ndash12454

HhP

H1gp22

7339

81aa

similarityto

SH1O

RF26

(YP271883)a

ndalso

topu

tativ

eprotein

17of

HHIV-2

(AFD

02298)

2312458ndash12697

HhP

H1gp23

7948

78aa

similarityto

SH1capsid

protein

VP13

(ORF

27Y

P271884)a

ndalso

toVP13of

HHIV-2

(AFD

02299)P

redicted

coil-coildo

main

PredictedC-

term

inaltransm

embraned

omain(Pho

bius)

Archaea 9

Table4Con

tinued

ORF

aPo

sitionb

Locustag

Leng

thc(aa)

pKi

Similaritycharacteris

tics

2412701ndash1506

4HhP

H1gp24

787

39

Capsid

protein

VP2(O

RF24)75aa

similarityto

SH1V

P2(O

RF28Y

P271885)a

ndto

VP2

ofHHIV-2

(AFD

02300)

ZOD2009

19133of

Happau

cihalophilus(ZP

0804

6189)

2515065ndash15961

HhP

H1gp25

298

45

Putativec

apsid

protein

VP5(O

RF25)81aa

similarityto

SH1V

P5(O

RF29Y

P271886)a

ndto

HHIV-2

VP5

(AFD

02301)

HlacA

J19807of

HbflacisalsiAJ

5ZO

D2009

19133of

Happau

cihalophilus

261596

4ndash1644

6HhP

H1gp26

160

46

73aa

similarityto

SH1capsid

protein

VP10

(ORF

30Y

P271887)a

ndto

VP10of

HHIV-2

(AFD

02302)Z

OD2009

19138of

Happau

cihalophilusHlacA

J19802of

Hbflacisa

lsiP

redicted

C-term

inaltransm

embraned

omain(TMHMM)

271644

6ndash16895

HhP

H1gp27

149

42

Capsid

protein

VP9(O

RF27)94aa

similarityto

SH1V

P9(O

RF31Y

P271888)a

ndto

ZOD2009

19143of

Hap

paucihalophilus(ZP

0804

6191)

2816908ndash17921

HhP

H1gp28

337

43

Capsid

protein

VP3(O

RF28)91aa

similarityto

SH1V

P3(O

RF32Y

P271889)a

ndto

NJ7G

2365

ofNa

trinemaspJ7-2

2917928ndash18617

HhP

H1gp29

229

42

Capsidprotein

VP6(O

RF29)83aa

similarityto

SH1V

P6(O

RF33Y

P271890)Also

toNJ7G

3194

ofNa

trinemaspJ7-2and

Hmuk

0476

ofHalom

icrobium

mukohataeiPredictedcarboxypeptid

aser

egulatory-lik

edom

ain(C

arbo

xypepD

reg

pfam

13620)

3018614ndash

18943

HhP

H1gp30

109

49

71aa

similarityto

SH1O

RF34

(YP271891)a

ndto

putativ

eprotein

27of

HHIV-2

(AFD

02308)P

redicted

signalsequence

(signalP)

3119054ndash

19176c

HhP

H1gp31

4050

TwoCx

xCmotifsN

oho

molog

inSH

1orH

HIV-2

3219173ndash19289c

HhP

H1gp32

3841

68aa

similarityto

SH1O

RF37

(YP271894)

3319286ndash

19552c

HhP

H1gp33

8853

58aa

similarityto

SH1O

RF39

(YP271896)a

ndto

HHIV-2

putativ

eprotein

29(A

FD02310)Fou

rCxxCmotifs

3419549ndash

19731c

HhP

H1gp34

6070

Con

tainsC

xxCmotif

Noho

molog

inSH

1orH

HIV-2

3519728ndash20219c

HhP

H1gp35

163

56

93aa

similarityto

SH1O

RF41

(YP271898)a

ndto

putativ

eprotein

30of

HHIV-2

(AFD

02311)

3620216ndash

21415c

HhP

H1gp36

399

50

95aa

similarityto

SH1O

RF42

(YP271899)a

ndto

putativ

eprotein

31of

HHIV-2

(AFD

02312)

3721419ndash

21778c

HhP

H1gp37

11941

91aa

similarityto

SH1O

RF43

(YP271900)a

ndto

putativ

eprotein

32of

HHIV-2

(AFD

02313)

3821762ndash2300

6cHhP

H1gp38

414

42

83aa

similarityto

SH1O

RF44

(YP271901)a

ndto

putativ

eprotein

33of

HHIV-2

(AFD

02314)P

redicted

coil-coiland

helix-tu

rn-helixdo

mains

3923003ndash23158c

HhP

H1gp39

5165

69aa

similarityto

SH1O

RF45

(YP271902)

4023161ndash23370c

HhP

H1gp40

6935

74aa

similarityto

SH1O

RF46

(YP271903)and55sim

ilarityto

putativ

eprotein

35of

HHIV-2

(AFD

02316)

4123422ndash23586c

HhP

H1gp41

5476

86aa

similarityto

SH1O

RF47

(YP271904)Con

tains2

CxxC

motifsand

show

ssim

ilarityto

proteindo

mainfamily

PF14206Arginine-ric

h

4223583ndash24023c

HhP

H1gp42

48

77aa

similarityto

SH1O

RF48

(YP271905)a

ndto

putativ

eprotein

36of

HHIV-2

(AFD

02317)H

ham114540of

Hcc

hamelinensis(ZP11271999)Ph

iCh1p72of

Natrialba

phageP

hiCh

1(NP665989)DUF4

326(pfam14216)

family

domain

4324020ndash

24538c

HhP

H1gp43

172

47

79aa

similarityto

SH1O

RF49

(YP271906)and56sim

ilarityto

putativ

eprotein

37of

HHIV-2

(AFD

02318)

4424535ndash25053c

HhP

H1gp44

172

43

92aa

similarityto

SH1O

RF50

(YP271907)and72sim

ilarityto

putativ

eprotein

38of

HHIV-2

(AFD

02319)

4525050ndash

25277c

HhP

H1gp45

7543

Noho

molog

inSH

1orH

HIV-2

4625274ndash

25387c

HhP

H1gp46

3748

Noho

molog

inSH

1orH

HIV-2

4725533ndash25904

HhP

H1gp47

123

43

61aa

similarityto

SH1O

RF51

(YP271908)66sim

ilarityto

putativ

eprotein

39of

HHIV-2

(AFD

02320)P

redicted

COG1342

domain(D

NAbind

inghelix-tu

rn-helix)

4825901ndash26092c

HhP

H1gp48

6349

Con

tainsC

xxCmotif

Noho

molog

inSH

1orH

HIV-2

4926089ndash

2764

8cHhP

H1gp49

506

40

81aa

similarityto

SH1O

RF55

(YP271912)(bu

tonlyin

theN

-term

inalhalf)H

omolog

ofvirusstru

cturalprotein

VP18

ofHHIV-2

(AFD

02323)

a ORF

swerep

redicted

either

byGLIMMER

orby

manualsearching

forh

omologsintheG

enBa

nkdatabase

b Startandendpo

sitions

ofORF

sare

give

inbp

numbera

ccording

totheP

H1sequenced

epositedatGenBa

nk(KC2

52997)O

RFso

nthec

omplem

entary

strandared

enoted

bythes

uffixc

c Lengthof

thep

redicted

ORF

innu

mbero

faminoacids

10 Archaea

Table 5 Main differences (divergent regions) between the SH1 and PH1 genomes

RegionaSH1 startb SH1 stop Length (bp) PH1 start PH1 stop Length (bp) Comment

DV1 527 650 123 538 604 66

Replacement Both regions have direct repeats at their bordersPH1 has two sets GACCCGGC and CGCTGC while SH1 hasCCCGAC In SH1 this replacement region includes theC-terminal region of ORF2 and the N-terminal region ofORF3 In PH1 it covers only the C-terminal region of ORF1

DV2 2433 2588 155 2386 2387 mdash RMDc from PH1 SH1 repeat at border is TGACCGThisremoves the homolog of SH1 ORF10 from PH1

DV3 3324 3598 274 3137 3416 279 Replacement at the beginning of SH1 ORF13PH1 ORF12(capsid protein VP1)

DV4 6656 7128 472 6478 7085 607 Replacement near the C-terminus of SH1 ORF13PH1 ORF12(capsid protein VP1)

DV5 7491 8341 850 7446 7447 mdash

Indel that results in ORFs14-16 of SH1 not being present inPH1 SH1 ORF14 has a close homolog in HHIV-2 (ORF6) in asimilar position just after the VP1 homolog SH1 ORF15 is aconserved protein in haloarchaea (eg NP 2552A andHmuk 2978) and halovirus His1 (ORF13) Possible invertedrepeat (AGCCATG) at border of SH1 region

DV6 13705 13706 mdash 12818 12851 33RMD from SH1 PH1 repeat at border is CAGCGG(gt)G Thisremoves a part of capsid protein VP2 sequence from the SH1protein

DV7 13789 13863 74 12934 12941 7 Replacement This occurs within VP2 gene of both viruses

DV8 15142 15186 44 14220 14221 mdash RMD from PH1 SH1 repeat at border is TG(tc)CCGACGAand occurs within capsid protein VP2 gene of both viruses

DV9 15303 15317 14 14337 14338 mdash RMD from PH1 SH1 repeat at border is GCCGACGA andoccurs within capsid protein VP2 gene of both viruses

DV10 15504 15527 23 14529 14530 mdash RMD from PH1 SH1 repeat at border is GACGA and occurswithin capsid protein VP2 of both viruses

DV11 20026 20405 379 19019 19173 154Replacement beginning at the start of SH1 ORF35PH1 ORF31This region is longer in SH1 where it includes ORF36 an ORFthat has no homolog present in PH1

DV12 20440 20661 221 19208 19286 78 Replacement This is unequal and includes an ORF in SH1(ORF38) that is not present in PH1

DV13 20943 21085 142 19562 19728 166 Replacement This covers SH1 ORF40PH1 ORF34 Thepredicted proteins are not homologous

DV14 23541 23542 mdash 22171 22197 27 Probable RMD in SH1 PH1 repeat at border is CGTCTCGGand occurs in SH1 ORF44PH1 ORF38

DV15 24506 24521 15 23152 23153 mdash Probable RMD in PH1 SH1 repeat at border is CTCGGT andoccurs near the end of SH1 ORF45PH1 ORF39

DV16 24951 24964 13 23579 23580 mdash Probable RMD in PH1 SH1 repeat at border is CGGTC andoccurs in SH1 ORF47PH1 ORF41

DV17 26449 26450 mdash 25050 25274 224Probable RMD in SH1 PH1 repeat at border is TCATGCG andoccurs near the start of SH1 ORF50PH1 ORF44 and providesan extra ORF for PH1 (ORF45)

DV18 27074 29764 690 25895 26933 1038

Replacement Left border at start of SH1 ORF51PH1 ORF47and extends rightwards into SH1 ORF55PH1 ORF49 (VP18gene) It is an unequal replacement and SH1 has two moreORFs in this region than PH1

aDV Divergent regions between the genomes of SH1 and PH1bStart and stop positions refer to the GenBank sequences of the two viruses SH1 NC 0072171 PH1 KC252997cRMD repeat-mediated deletion event as described in [20]

The recently described halovirus SNJ1 carries manyORFsthat have homologs adjacent to a previously described ViPREofHmc mukohataei (from here on denoted by ViPREHmuk1)a locus that contains genes related to His2 (and other ple-olipoviruses) as well as toHaloferax plasmid pHK2 (probablya provirus) and to two small plasmids ofHqr walsbyi (sim6 kbpL6A and pL6B)The latter relationship provides wider linksto other viruses because one of the PL6 genes (Hqrw 6002)

has homologs in some pleolipoviruses (HPRV-3 and HGPV-1) while another gene (Hqrw 6005) is related to ORF16 ofthe spindle-shaped halovirus His1 [15] After adding the SNJ1gene homologs to ViPREHmuk1 it is extended significantlyand a close inspection of the flanking regions revealed atRNA-ala gene at one end and a partial copy of the sametRNA-ala gene (next to a phage integrase gene) at the otherend (supplementary Figure 1) Genes beyond these points

Archaea 11

Table 6 Transfection of haloarchaea by PH1 DNA

Speciesa Efficiency of transfectiontransformation withPH1 DNAb (PFU120583g of DNA) pUBP2 DNAbc (CFU120583g of DNA)

Har hispanica 53 plusmn 05 times 103 14 plusmn 08 times 104

Har marismortui 45 plusmn 07 times 103 18 plusmn 01 times 104

ldquoHar sinaiiensisrdquo 32 plusmn 06 times 102 mdashHalobacterium salinarum mdash 54 plusmn 42 times 103

Haloferax gibbonsii 28 plusmn 08 times 103 40 plusmn 26 times 103

Haloferax lucentense mdash 57 plusmn 11 times 103

Hfx volcanii mdash 21 plusmn 07 times 105

Hrr lacusprofundi 70 plusmn 22 times 102 95 plusmn 15 times 103

Haloterrigena turkmenica 16 plusmn 17 times 102 24 plusmn 06 times 103

Natrialba asiatica 35 plusmn 39 times 102 31 plusmn 10 times 104aOnly those species positive for transfection andor plasmid transformation are shownbRates of transfection by (nonprotease treated) PH1 DNA or transformation by plasmid pUB2 are averages of three independent experiments each performedin duplicate (plusmn standard deviation)cTransformants were selected on plates with 2 4 or 6120583gmL simvastatin depending on the strainmdash no plaques or colonies observed

appear to be related to cellular metabolism ViPREHmuk1is now 39379 nt in length flanked by a 59 nt direct repeat(potential att sites) and includes 54 ORFs many of which arerelated to known haloviruses are virus-like (eg integrasesmethyltransferases transcriptional regulators DNA methyl-transferase and DNA glycosylase) are homologs in or nearother knownViPREs or show little similarity to other knownproteins This locus also includes an ORC1CDC6 homolog(Hmuk 0446) which could provide a replication function

The mixture of very different virus and plasmid genesseen within ViPREHmuk1 is remarkable but it also revealshomologs found in or adjacent to genomic loci of otherspecies such as the gene homologs of Har marismortuiseen in the left end of ViPREHmuk1 (supplementary Figure1) When these genes are added to the previously describedHaloarcula locus (ViPREHmuk1) it is also extended signifi-cantly It becomes 25550 bp in length encompasses genesfrom Hmar 2382 to Hmar 2404 and is flanked by a fulltRNA-ala gene at one end and a partial copy at the other Itcarries an ORC1CDC6 homolog an integrase and a phiH-like repressor as well as pleolipovirus homologs

210 Transfection of Haloarchaea by PH1 DNA PH1 DNAwas introduced into Har hispanica cells using the PEGmethod [37] and the cells were screened for virus productionby plaque assay Figure 7 shows the increase in transfectedcells with input (nonproteinase K-treated) virus DNA Theestimated efficiencywas 53times 103 PFUper120583gDNA PH1DNAthat had been treated with proteinase K or with DNAase I(RNase-free) did not produce plaques (data not shown) PH1DNA was found to transfect six species of haloarchaea otherthanHar hispanica includingmembers of the generaHaloar-cula Haloferax Halorubrum Haloterrigena and Natrialba(Table 6) For these experiments the virus production fromtransfected cells was detected by plaque assay on indicatorlawns of Har hispanica

0

0

200 400 600 800 1000

Titre

(PFU

mL)

104

103

102

PH1 DNA (ng)

Figure 7 Transfection ofHar hispanica cells with PH1 DNA Vary-ing amounts of nonproteinase K-treated PH1 DNAwere introducedinto cells of Har hispanica using the PEG method [64] Cells werethen screened for infective centres by plaque assay Data shown is theaverage of three independent experiments performed in duplicateError bars represent one standard deviation of themean If protease-treated DNA was used no transfectant plaques were observed

3 Discussion

PH1 is very similar in particlemorphology genome structureand sequence to the previously described halovirus SH1 [816] Like SH1 it has long inverted terminal repeat sequencesand terminal proteins indicative of protein-primed replica-tion Purified PH1 particles have a relatively low buoyant

12 Archaea

density and are chloroform sensitive and show a layeredcapsid structure consistent with the likely presence of aninternalmembrane layer as has been shown for SH1HHIV-2and SNJ1 [13 26]The particle stability of PH1 to temperaturepH and reduced salt was similar to SH1 [8] The structuralproteins of PH1 are very similar in sequence and relativeabundance to those of SH1 so the two viruses are likely toshare the same particle geometry [16]

Viruses that use protein-primed replication such as bac-teriophageΦ29 carry a viral type BDNApolymerase that caninteract specifically with the proteins attached to the genomictermini and initiate strand synthesis [38] ArchaeovirusesHis1 [7] and Acidianus bottle-shaped virus [39] probablyuse this mode of replication However a polymerase genecannot be found in the genomes of PH1 SH1 or HHIV-2DNA polymerases are large enzymes and the only ORF ofSH1 that was long enough to encode such an enzyme andhad not previously been assigned as a structural protein wasORF55 which specified an 865 aa protein that contained noconserved domains indicative of polymerases [27]The recentstudy of HHIV-2 showed that the corresponding ORF in thisvirus (gene 42) is a structural protein of the virion and byinference this is likely also for the corresponding proteinsof SH1 and PH1 (no homolog is present in SNJ1) Withouta polymerase gene these viruses must use a host enzymebut searches for a viral type B DNA polymerase in thegenome of the hostHar hispanica or in the genomes of othersequenced haloarchaea did not find anymatchesThis arguesstrongly for a replication mechanism that is different to thatexemplified byΦ29 An attractive alternative is that displayedby Streptomyces linear plasmids which have 51015840-terminal pro-teins and use a cellular polymerase for replication [40] Theterminal proteins are not used for primary replication but forend patching [41] and it has been shown that if the term-inal repeat sequences are removed from these plasmids andthe ends ligated they can replicate as circular plasmids [42]This provides a testable hypothesis for the replication of halo-sphaeroviruses and also offers a pathway for switchingbetween linear (eg PH1) and circular (eg SNJ1) forms Useof a host polymerase would also fit with the ability of SH1 andPH1 DNA to transfect many different haloarchaeal species[31] as these enzymes and theirmode of action are highly con-served

The growth characteristics of PH1 inHar hispanica bothby single-cell burst and single-step growth experiments gavevalues of 50ndash100 virusescell significantly less than that ofSH1 (200 virusescell) [8] and HHIV-2 (180 virusescell) [13]The latter virus is lytic whereas SH1 PH1 and SNJ1 start pro-ducing extracellular viruswell before any decrease in cell den-sity indicating that virus release can occur without lysisThishas been most clearly shown with SH1 where almost 100of cells can be infected [8] Infection ofHar hispanica by PH1was less efficient (sim30 of cells) but the kinetics of virus pro-duction in single-step growth curves of SH1 and PH1 aresimilar with virus production occurring over several hoursrather than at a clearly defined time after infection The twoviruses are also very closely related and infect the same hostspecies so it is likely that they use the same scheme for cellexit For both viruses the cell density eventually decreases in

single-step cultures showing that virus infection does resultin cell death This mode of exit appears to maximise theproduction of the virus over time as the host cells survive foran extended period In natural hypersaline waters cell num-bers are usually high but growth rates are low [12] and thehigh incident UV (on often shallow ponds) would damagevirus DNA so reducing the half-life of released virus part-icles [43] These factors may have favoured the exit strategydisplayed by these viruses improving their chance of trans-mission to a new host

The motif GATC was absent in the genomes of PH1 SH1andHHIV-2 and themotif CTAGwas either absent or greatlyunderrepresented All three viruses share the same host butGATC and CTAG motifs are plentiful in the Har hispanicagenome (accessions CP002921 and CP002923) Avoidance ofthesemotifs in halovirusesHis1 His2 HF1 andHF2 has beenreported previously [7] Such purifying selection commonlyresults from host restriction enzymes and inHfx volcanii itis exactly these two motifs that are targeted [44] One Hfxvolcanii enzyme recognises A-methylated GATC sites [45]and another recognises un-methylated CTAG sites (whichare protected by methylation in the genome) In the currentstudy this could explain the negative transfection results forHfx volcanii as the PH1 genome contains two CTAGmotifsBy comparison the SH1 genome contains no CTAG motifsand is able to transfect Hfx volcanii [31]

Comparison of the PH1 and SH1 genomes allowed adetailed picture of the natural variation occurring betweenclosely related viruses revealing likely deletion events medi-ated by small repeats a process described previously in astudy of Hqr walsbyi and termed repeat-mediated deletion[20] There are also many replacements including blocks ofsequence that occur within long open reading frames (suchas the capsid protein genes VP1 and VP18) VP3 and VP6 areknown to form the large spikes on the external surface of theSH1 virion and presumably one or both interact with host cellreceptors [16] PH1 and SH1 have identical host ranges con-sistent with the high sequence similarity shown between theircorresponding VP3 and VP6 proteins

A previous study of SH1 could not detect transcriptsacross annotated ORFs 1ndash3 or 55-56 [30] ORFs 1 and 56occur in the terminal inverted repeat sequence and in thepresent study it was found that there were no ORFs cor-responding to these in the PH1 genome Given the close simi-larity of the two genomes the comparative data are in agree-ment with the transcriptional data indicating that these SH1ORFs are not used SH1 ORFs 2 and 3 are more problematicas good homologs of these are also present in PH1 The con-flict between the transcriptional data and the comparativegenomic evidence requires further experimental work toresolveThe status of SH1ORF55 has recently been confirmedby studies of the related virus HHIV-2 (discussed above)

The pleolipovirus group has expanded dramatically inthe last few years and it now comprises a diverse group ofviruseswith different genome types and replication strategiesWhat is evenmore remarkable is that a similar expansion cannow be seen with SH1-related viruses which includes viruseswith at least two replication modes and genome types (linearand circular) Evidence from haloarchaeal genome sequences

Archaea 13

show cases of not only provirus integrants or plasmids (egpKH2 and pHH205) but also genomic loci (ViPREs) thatcontain cassettes of virus genes from different sources andat least two cases (described in this study) have likely attsites that indicate circularisation and mobility While thereis a clear relationship between viruses with circular ds- andssDNA genomes that replicate via the rolling circle method(ie the ssDNA form is a replication intermediate) it ismore difficult to explain how capsid gene cassettes can movebetween these viruses and those with linear genomes andterminal proteins Within the pleolipoviruses His2 has alinear genome that contains a viral type B DNA polymeraseand terminal proteins while all the other described membershave circular genomes that contain a rep homolog and prob-ably replicate via the rolling circle method [14 19] Similarlyamong the halosphaeroviruses PH1 SH1 and HHIV-2 havelinear dsDNA with terminal proteins and probably replicatein the same way but the related SNJ1 virus has a circulardsDNA genome The clear relationships shown by the capsidgenes of viruses within each group plus the connectionsshown to haloviruses outside each group (eg the DNA poly-merases of His1 and His2) all speak of a vigorous meansof recombination one that can readily switch capsid genesbetween viruses with radically different replication strategiesHow is the process most likely to operate One possibility issuggested by ViPREs which appear to be mobile collectionsof capsid and replication genes from different sources Theyoffer fixed locations for recombination to occur providegene cassettes that can be reassorted to produce novel virusgenomes and some can probably recombine out as circularforms For example ViPREHmuk1 carries genes related topleolipoviruses halosphaeroviruses and other haloviruses Itis yet to be determined if any of the genes carried in theseloci are expressed in the cell (such as the genes for capsidproteins) or if ViPREs provide any selective advantage to thehost

4 Materials and Methods

41 Water Sample A water sample was collected in 1998from Pink Lake (32∘001015840 S and 115∘301015840 E) a hypersalinelake on Rottnest Island Western Australia Australia It wasscreened in the same year for haloviruses using the methodsdescribed in [8] A single plaque on a Har hispanica lawnplate was picked and replaque purified The novel haloviruswas designated PH1

42 Media Strains and Plasmids The media used in thisstudy are described in the online resource The Halohand-book (httpwwwhaloarchaeacomresourceshalohandbookindexhtml) Artificial salt water containing 30 (wv) totalsalts comprised of 4M NaCl 150mM MgCl

2 150mM

MgSO4 90mM KCl and 35mM CaCl

2and adjusted to pH

75 using sim2mL 1M Tris-HCl (pH 75) per litre MGM con-taining 12 18 or 23 (wv) total salts and HVD (halo-virus diluent) were prepared from the concentrated stock aspreviously described [46] Bacto-agar (Difco Laboratories)was added to MGM for solid (15 gL) or top-layer (7 gL)media

Table 1 lists the haloarchaea used in this study Allhaloarchaea were grown aerobically at 37∘C in either 18 or23 (wv)MGM(depending on the strain) andwith agitation(except for ldquoHar sinaiiensisrdquo)The plasmids used were pBlue-script II KS+ (Stratagene Cloning Systems) pUBP2 [47]and pWL102 [48] pUBP2 and pWL102 were first passagedthrough E coli JM110 [49] to prevent dam-methylationof DNA which has been shown to reduce transformationefficiency in some strains [45]

43 Negative-Stain Electron Microscopy The method fornegative-stain TEM was adapted from that of V Tarasovdescribed in the online resource The Halohandbook (httpwwwhaloarchaeacomresourceshalohandbookindexhtml)A 20120583L drop of the sample was placed on a clean surfaceand the virus particles were allowed to adsorb to a Formvarfilm 400-mesh copper grid (ProSciTech) for 15ndash2min Theywere then negatively stained with a 20120583L drop of 2 (wv)uranyl acetate for 1-2min Excess liquid was absorbed withfilter paper and the grid was allowed to air dry Grids wereexamined either on a Philips CM 120 BioTwin transmissionelectron microscope (Royal Philips Electronics) operating atan accelerating voltage of 120 kV or on a Siemens Elmiskop102 transmission electron microscope (Siemens AG) operat-ing at an accelerating voltage of 120 kV

44 Virus Host Range Twelve haloarchaeal strains from thegenera Haloarcula Halobacterium Haloferax HalorubrumandNatrialba (Table 1)Thirteen naturalHalorubrum isolates(H Camakaris unpublished data) and five uncharacterizedhaloarchaeal isolates fromLakeHardy Pink Lake (inWesternAustralia) and Serpentine Lake (D Walker and M K Seahunpublished data) were screened for PH1 susceptibilityLysates from PH1-infected Har hispanica cultures (1 times1011 PFUmL) were spotted onto lawns of each strain (usingmedia with 12 or 18 (wv)MGM depending on the strain)and incubated for 2ndash5 days at 30 and at 37∘C

45 Large Scale Virus Growth and Purification Liquid cul-tures of PH1 were grown by the infection (MOI 005) of anearly exponential Har hispanica culture in 18 (wv) MGMCultures were incubated aerobically at 37∘C with agitationfor 3 days Clearing (ie complete cell lysis) of PH1-infectedcultures did not occur and consequently cultures were harv-ested when the absorbance at 550 nm reached the minimumand the titre (determined by plaque assay) was at the maxi-mum usually sim1010ndash1011 PFUmL Virus was purified usingthe method described previously for SH1 [8 31] except thatafter the initial low speed spin (Sorvall GSA 6000 rpm30min 10∘C) virus was concentrated from the infected cul-ture by centrifugation at 26000 rpm (13 hr 10∘C) onto acushion of 30 (wv) sucrose in HVDThe pellet was resus-pended in a small volume of HVD and loaded onto a pre-formed linear 5ndash70 (wv) sucrose gradient followed byisopycnic centrifugation in 13 gmL CsCl (Beckman 70Ti60000 rpm 20 hr 10∘C) The white virus band occurredat a density of 129 gmL and was collected and diluted inhalovirus diluent (HVD see Section 4) and the virus pelleted(Beckman SW55 35000 rpm 75min 10∘C) and resuspended

14 Archaea

in a small volume of HVD and stored at 4∘C Virus recoveryat the major stages of a typical purification is given insupplementary Table 1 The specific infectivity of pure virussolutions was determined as the ratio of the PFUmL to theabsorbance at 260 nm

46 PH1 Single-Step Growth Curve An early exponentialphase culture of Har hispanica grown in 18 (wv) MGMwas infected with PH1 (MOI 50) Under these conditions thepercentage of infected cells was approximately 30 After anadsorption period of 1 hr at 37∘C the cells were washed threetimes with 18 (wv) MGM (at room temperature) resus-pended in 100mL 18 (wv) MGM (these methods ensuredthe removal of all residual virus) and incubated at 37∘Cwith shaking (100 rpm) Samples were removed at hourlyintervals for measurements of absorbance at 550 nm andthe number of infective centres Immediately after samplingtitres were determined by plaque assay with an indicator lawnof Har hispanica Each experiment was performed in tripli-cate

47 Halovirus Stability After various treatments sampleswere removed and diluted in HVD and virus titres weredetermined by plaque assay on Har hispanica Each experi-ment was performed in triplicate and representative data areshown Chloroform sensitivity was examined by the exposureof PH1 lysates in 18 (wv) MGM to chloroform in a volumeratio of 1 4 (chloroform to lysate) Incubation was at roomtemperature with constant agitation At appropriate timepoints the mix was allowed to settle and a sample wasremoved from the upper layer and diluted in HVDThe effectof a low ionic environment was examined by diluting PH1lysates (in 18 (wv) MGM) into double distilled H

2O in a

volume ratio of 1 1000 (lysate to double distilledH2O) Incu-

bation was at room temperature with constant agitationSamples were removed at various times and diluted in HVDThe pH stability of PH1 was determined by dilution of PH1lysates in 18 (wv)MGMin the appropriate pHbuffer (HVDbuffered with appropriate Tris-HCl) in a volume ratio of1 100 (lysate to buffer) Incubation was at room temperaturewith constant agitation After 30min samples were removedand diluted in HVD Thermal stability of PH1 was examinedby a 1 hr incubation of a virus lysate in 18 (wv) MGM atdifferent temperatures after which they were brought quicklyto room temperature diluted in HVD and titrated

48 Protein Procedures To remove salts purified virus prepa-rations were mixed with trichloroacetic acid (10 (vv) finalconcentration) and incubated on ice for 15min to allowthe proteins to precipitate After centrifugation (16000 g15min room temperature) the precipitate was washed threetimes in acetone dried and resuspended in double distilledH2O Proteins were dissolved in Laemmli sample buffer with

48mM 120573-mercaptoethanol [50] heated in boiling water for5min then separated on 12 (wv) NuPAGE Novex Bis-Tris Gels using MES-SDS running buffer according to themanufacturerrsquos directions (Invitrogen) After electrophoresisgels were rinsed in double distilled H

2O and stained with

01 (wv) Brilliant Blue G in 40 (vv) methanol and

10 (vv) acetic acid Gels were destained with severalchanges of 40 (vv) methanol and 10 (vv) acetic acidAlternatively gels were stained with GelCode GlycoproteinStaining Kit Pro-Q Emerald 488 Glycoprotein Gel and BlotStain Kit SYPRO Ruby Protein Gel Stain according tothe manufacturerrsquos directions (Invitrogen Molecular ProbesPierce Biotechnology)

Protein bands were cut from the gels and sent to theAustralian Proteome Analysis Facility (Macquarie Univer-sity) for trypsin digestion and analysis by matrix assistedlaser desorption ionisation time of flight mass spectrometry(MALDI-TOF MS) on an Applied Biosystems 4700 Pro-teomics Analyser (Applied Biosystems)

49 DNA Procedures Proteinase K-treated and nonpro-teinaseK-treatedDNApreparationsweremade frompurifiedPH1 using the methods described previously for SH1 [8]DNA was separated on 1 (wv) agarose gels in Tris-acetate-EDTA electrophoresis buffer and was stained with ethidiumbromide (Sigma-Aldrich)120582 DNA DNase I (RNase-free) exonuclease III mung

bean nuclease nuclease BAL-31 T4 DNA polymerase T4DNA ligase T7 exonuclease and type II restriction endonu-cleases were purchased from New England Biolabs RNase Aand yeast transfer RNA were purchased from Sigma-AldrichProteinase K was purchased from Promega Oligonucleotideprimers were purchased from Geneworks Australia

To clone fragments of the PH1 genome purified virusDNA was digested either with AccI Eco0109I and MseI orSmaI restriction endonucleases blunted with DNA Poly-merase I Large (Klenow) Fragment where appropriate andligated into AccI- Eco0109I- or SmaI-digested pBluescript IIKS+ using T4 DNA ligase according to the manufacturerrsquosinstructions (New England Biolabs) The DNA was intro-duced intoE coliXL1-Blue and transformants grown onLuriaagar containing 15 120583gmL tetracycline and 100 120583gmL ampi-cillinThe resulting cloneswere sequenced and the sequenceswere used to design specific oligonucleotide primers for PCRamplification andor primer walking using the virus genome(or specific restriction fragments) as templates

To amplify PH1 nucleic acid approximately 10 ng virusDNA was combined with 500 nM primers 100120583M of eachdNTP 1U Deep Vent DNA Polymerase (New England Bio-labs) and 1 timesThermoPol buffer (New England Biolabs) ina total reaction volume of 50 120583L Template was denatured at95∘C for 10min followed by 30 cycles of denaturation at 95∘Cfor 30 sec annealing at 56∘C for 30 sec extension at 75∘C for2min and a final extension at 75∘C for 10min PCR reactionswere performed on a PxE02ThermoCycler (Thermo ElectroCorporation) Sequencing reactions using 32 pmol primerwere performed by the dideoxy chain termination methodusing ABI PRISM Big Dye Terminator Mix version 31 onan ABI 3100 capillary sequencer (Applied Biosystems) bythe Applied Genetic Diagnostics Sequencing Service at theDepartment of Pathology (The University of Melbourne)

To obtain terminal genomic fragments PH1 DNA wasdigested by the enzyme Pas1 and the head (sim700 bp) andthe tail (sim1300 bp) fragments were purified by the QIAEXII gel extraction kits and treated with 05M piperidine for

Archaea 15

2 hr at 37∘C to remove protein residues The piperidine-treated fragments were sequenced by the Genomics BiotechCompany (Taipei Taiwan) using two primers TGACCA-ATTAATTAGGCCGGTTCGCC (PH1 head-R) and GTGC-CATACTGCTACAATTCT (PH1 tail-F)

Four hundred fifty-four whole genome sequencing PH1DNA samples (sim5 120583g) were sequenced using parallel pyrose-quencing on a Roche 454 Genome Sequencer System atMission Biotech (Taipei Taiwan) The largest contig was27399with 8803 reads (Newbler version 27) with 27387 posi-tions being of Q40 quality The average depth of coverage foreach base was 829TheGenBank accession for the entire PH1sequence is KC252997

Acknowledgments

The authors thank Carolyn Bath Mei Kwei (Joan) Seahand Danielle Walker for their early work in PH1 characteri-zation Simon Crawford Jocelyn Carpenter and Anna Fried-huber for their assistance with the transmission electronmicroscopy and Tiffany Cowie and Voula Kanellakis fortheir sequencing assistance This paper has been facilitatedby access to the Australian Proteome Analysis Facility estab-lished under the Australian Governmentrsquos Major NationalResearch Facilities program SLT was supported by a grantfrom the Biodiversity Research Center Academia SinicaTaiwan

References

[1] M Krupovic M F White P Forterre and D PrangishvilildquoPostcards from the edge structural genomics of archaealvirusesrdquo Advances in Virus Research vol 82 pp 33ndash62 2012

[2] P Stolt andW Zillig ldquoGene regulation in halophage 120601Hmdashmorethan promotersrdquo Systematic and Applied Microbiology vol 16no 4 pp 591ndash596 1994

[3] R Klein U Baranyi N Rossler B Greineder H Scholz and AWitte ldquoNatrialba magadii virus 120593Ch1 first complete nucleotidesequence and functional organization of a virus infecting ahaloalkaliphilic archaeonrdquo Molecular Microbiology vol 45 no3 pp 851ndash863 2002

[4] E Pagaling R D Haigh W D Grant et al ldquoSequence analysisof an Archaeal virus isolated from a hypersaline lake in InnerMongolia Chinardquo BMC Genomics vol 8 article 410 2007

[5] S L Tang SNuttall andMDyall-Smith ldquoHalovirusesHF1 andHF2 evidence for a recent and large recombination eventrdquo Jour-nal of Bacteriology vol 186 no 9 pp 2810ndash2817 2004

[6] C Bath and M L Dyall-smith ldquoHis1 an archaeal virus of theFuselloviridae family that infectsHaloarcula hispanicardquo Journalof Virology vol 72 no 11 pp 9392ndash9395 1998

[7] C Bath T Cukalac K Porter andM L Dyall-Smith ldquoHis1 andHis2 are distantly related spindle-shaped haloviruses belongingto the novel virus group Salterprovirusrdquo Virology vol 350 no1 pp 228ndash239 2006

[8] K Porter P Kukkaro J K H Bamford et al ldquoSH1 a novelspherical halovirus isolated from an Australian hypersalinelakerdquo Virology vol 335 no 1 pp 22ndash33 2005

[9] M Dyall-Smith S L Tang and C Bath ldquoHaloarchaeal viruseshow diverse are theyrdquo Research in Microbiology vol 154 no 4pp 309ndash313 2003

[10] A Oren G Bratbak and M Heldal ldquoOccurrence of virus-likeparticles in the Dead Seardquo Extremophiles vol 1 no 3 pp 143ndash149 1997

[11] T Sime-Ngando S Lucas A Robin et al et al ldquoDiversityof virus-host systems in hypersaline Lake Retba SenegalrdquoEnvironmental Microbiology vol 13 no 8 pp 1956ndash1972 2010

[12] N Guixa-Boixareu J I Calderon-Paz M Heldal G Bratbakand C Pedros-Alio ldquoViral lysis and bacterivory as prokaryoticloss factors along a salinity gradientrdquoAquaticMicrobial Ecologyvol 11 no 3 pp 215ndash227 1996

[13] S T Jaakkola R K Penttinen S T Vilen et al ldquoClosely relatedarchaeal Haloarcula hispanica icosahedral viruses HHIV-2 andSH1 have nonhomologous genes encoding host recognitionfunctionsrdquo Journal of Virology vol 86 no 9 pp 4734ndash47422012

[14] E Roine P Kukkaro L Paulin et al ldquoNew closely related halo-archaeal viral elements with different nucleic acid typesrdquo Jour-nal of Virology vol 84 no 7 pp 3682ndash3689 2010

[15] A Sencilo L Paulin S KellnerMHelm and E Roine ldquoRelatedhaloarchaeal pleomorphic viruses contain different genometypesrdquo Nucleic Acids Research vol 40 no 12 pp 5523ndash55342012

[16] H T Jaalinoja E Roine P Laurinmaki H M Kivela D HBamford and S J Butcher ldquoStructure and host-cell interactionof SH1 a membrane-containing halophilic euryarchaeal virusrdquoProceedings of the National Academy of Sciences of the UnitedStates of America vol 105 no 23 pp 8008ndash8013 2008

[17] M K Pietila N S Atanasova VManole et al ldquoVirion architec-ture unifies globally distributed pleolipoviruses infecting halo-philic archaeardquo Journal of Virology vol 86 no 9 pp 5067ndash50792012

[18] M L Holmes and M L Dyall-Smith ldquoA plasmid vector witha selectable marker for halophilic archaebacteriardquo Journal ofBacteriology vol 172 no 2 pp 756ndash761 1990

[19] M L Holmes F Pfeifer andM L Dyall-Smith ldquoAnalysis of thehalobacterial plasmid pHK2 minimal repliconrdquo Gene vol 153no 1 pp 117ndash121 1995

[20] M L Dyall-Smith F Pfeiffer K Klee et al ldquoHaloquadratumwalsbyi limited diversity in a Global Pondrdquo PLoS ONE vol 6no 6 Article ID e20968 2011

[21] S Casjens ldquoProphages and bacterial genomics what have welearned so farrdquoMolecular Microbiology vol 49 no 2 pp 277ndash300 2003

[22] R W Hendrix M C M Smith R N Burns M E Ford andG F Hatfull ldquoEvolutionary relationships among diverse bacte-riophages and prophages all the worldrsquos a phagerdquo Proceedings ofthe National Academy of Sciences of the United States of Americavol 96 no 5 pp 2192ndash2197 1999

[23] D Botstein ldquoA theory ofmodular evolution for bacteriophagesrdquoAnnals of the New York Academy of Sciences vol 354 pp 484ndash491 1980

[24] M Krupovic S Gribaldo D H Bamford and P Forterre ldquoTheevolutionary history of archaeal MCM helicases a case study ofvertical evolution combinedwithHitchhiking ofmobile geneticelementsrdquo Molecular Biology and Evolution vol 27 no 12 pp2716ndash2732 2010

[25] M Krupovic P Forterre and D H Bamford ldquoComparativeanalysis of the mosaic genomes of tailed archaeal viruses andproviruses suggests common themes for virion architecture andassembly with tailed viruses of bacteriardquo Journal of MolecularBiology vol 397 no 1 pp 144ndash160 2010

16 Archaea

[26] Z Zhang Y Liu S Wang et al et al ldquoTemperate membrane-containing halophilic archaeal virus SNJ1 has a circular dsDNAgenome identical to that of plasmid pHH205rdquoVirology vol 434no 2 pp 233ndash241 2012

[27] D H Bamford J J Ravantti G Ronnholm et al ldquoConstituentsof SH1 a novel lipid-containing virus infecting the halophiliceuryarchaeonHaloarcula hispanicardquo Journal of Virology vol 79no 14 pp 9097ndash9107 2005

[28] H M Kivela E Roine P Kukkaro S Laurinavicius P Somer-harju andDH Bamford ldquoQuantitative dissociation of archaealvirus SH1 reveals distinct capsid proteins and a lipid corerdquoVirology vol 356 no 1-2 pp 4ndash11 2006

[29] H Jaalinoja Electron Cryo-Microscopy Studies of Bacteriophage8 and Archaeal Virus SH1 University of Helsinki HelsinkiFinland 2007

[30] K Porter B E Russ J Yang andM L Dyall-Smith ldquoThe trans-cription programme of the protein-primed halovirus SH1rdquoMi-crobiology vol 154 no 11 pp 3599ndash3608 2008

[31] K Porter and M L Dyall-Smith ldquoTransfection of haloarchaeaby the DNAs of spindle and round haloviruses and the useof transposon mutagenesis to identify non-essential regionsrdquoMolecular Microbiology vol 70 no 5 pp 1236ndash1245 2008

[32] S E Luria General Virology John Wiley and Sons New YorkNY USA 1953

[33] C A Thomas K Saigo E McLeod and J Ito ldquoThe separationofDNA segments attached to proteinsrdquoAnalytical Biochemistryvol 93 pp 158ndash166 1979

[34] A L Delcher D Harmon S Kasif OWhite and S L SalzbergldquoImproved microbial gene identification with GLIMMERrdquoNucleic Acids Research vol 27 no 23 pp 4636ndash4641 1999

[35] R Zhang Y Yang P Fang et al ldquoDiversity of telomere palin-dromic sequences and replication genes among Streptomyceslinear plasmidsrdquo Applied and Environmental Microbiology vol72 no 9 pp 5728ndash5733 2006

[36] X Wang H S Lee F J Sugar F E Jenney M W W Adamsand J H Prestegard ldquoPF0610 a novel winged helix-turn-helix variant possessing a rubredoxin-like Zn ribbon motiffrom the hyperthermophilic archaeon Pyrococcus furiosusrdquoBiochemistry vol 46 no 3 pp 752ndash761 2007

[37] S W Cline L C Schalkwyk and W F Doolittle ldquoTransfor-mation of the archaebacterium Halobacterium volcanii withgenomic DNArdquo Journal of Bacteriology vol 171 no 9 pp 4987ndash4991 1989

[38] S Kamtekar A J Berman J Wang et al ldquoThe 12059329 DNA poly-merase protein-primer structure suggests a model for the ini-tiation to elongation transitionrdquo EMBO Journal vol 25 no 6pp 1335ndash1343 2006

[39] X Peng T Basta M Haring R A Garrett and D PrangishvilildquoGenome of theAcidianus bottle-shaped virus and insights intothe replication and packaging mechanismsrdquo Virology vol 364no 1 pp 237ndash243 2007

[40] C C Yang C H Huang C Y Li Y G Tsay S C Lee and CW Chen ldquoThe terminal proteins of linear Streptomyces chrom-osomes and plasmids a novel class of replication primingproteinsrdquo Molecular Microbiology vol 43 no 2 pp 297ndash3052002

[41] C C Yang Y H Chen H H Tsai C H Huang T W Huangand C W Chen ldquoIn vitro deoxynucleotidylation of the term-inal protein of Streptomyces linear chromosomesrdquo Applied andEnvironmental Microbiology vol 72 no 12 pp 7959ndash79612006

[42] P C Chang and S N Cohen ldquoBidirectional replication froman internal origin in a linear Streptomyces plasmidrdquo Science vol265 no 5174 pp 952ndash954 1994

[43] S W Wilhelm W H Jeffrey C A Suttle and D L MitchellldquoEstimation of biologically damaging UV levels in marinesurface waters with DNA and viral dosimetersrdquo Photochemistryand Photobiology vol 76 no 3 pp 268ndash273 2002

[44] T Allers and M Mevarech ldquoArchaeal geneticsmdashthe third wayrdquoNature Reviews Genetics vol 6 no 1 pp 58ndash73 2005

[45] M L Holmes S D Nuttall and M L Dyall-Smith ldquoConstruc-tion and use of halobacterial shuttle vectors and further studieson Haloferax DNA gyraserdquo Journal of Bacteriology vol 173 no12 pp 3807ndash3813 1991

[46] S D Nuttall and M L Dyall-Smith ldquoHF1 and HF2 novelbacteriophages of halophilic archaeardquo Virology vol 197 no 2pp 678ndash684 1993

[47] U Blaseio and F Pfeifer ldquoTransformation of Halobacteriumhalobium development of vectors and investigation of gas vesi-cle synthesisrdquo Proceedings of the National Academy of Sciences ofthe United States of America vol 87 no 17 pp 6772ndash6776 1990

[48] W L Lam and W F Doolittle ldquoShuttle vectors for the archae-bacterium Halobacterium volcaniirdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 86 no14 pp 5478ndash5482 1989

[49] C Yanisch-Perron J Vieira and J Messing ldquoImproved M13phage cloning vectors and host strains nucleotide sequences ofthe M13mp18 and pUC19 vectorsrdquo Gene vol 33 no 1 pp 103ndash119 1985

[50] U K Laemmli ldquoCleavage of structural proteins during theassembly of the head of bacteriophage T4rdquo Nature vol 227 no5259 pp 680ndash685 1970

[51] D G Burns H M Camakaris P H Janssen and M L Dyall-Smith ldquoCombined use of cultivation-dependent and culti-vation-independent methods indicates that members of mosthaloarchaeal groups in anAustralian crystallizer pond are culti-vablerdquo Applied and Environmental Microbiology vol 70 no 9pp 5258ndash5265 2004

[52] G Juez F Rodriguez-Valera A Ventosa and D J KushnerldquoHaloarcula hispanica spec nov and Haloferax gibbonsii specnov two new species of extremely halophilic archaebacteriardquoSystematic and Applied Microbiology vol 8 pp 75ndash79 1986

[53] A Oren M Ginzburg B Z Ginzburg L I Hochstein and BE Volcani ldquoHaloarcula marismortui (Volcani) sp nov nomrev an extremely halophilic bacterium from the Dead SeardquoInternational Journal of Systematic Bacteriology vol 40 no 2pp 209ndash210 1990

[54] M Torreblanca F Rodriguez-Valera G Juez A Ventosa MKamekura and M Kates ldquoClassification of non-alkaliphilichalobacteria based on numerical taxonomy and polar lipidcomposition and description of Haloarcula gen nov andHaloferax gen novrdquo Systematic and Applied Microbiology vol8 pp 89ndash99 1986

[55] A Ventosa and A Oren ldquoHalobacterium salinarum nom cor-rig a name to replace Halobacterium salinarium (Elazari-Vol-cani) and to include Halobacterium halobium and Halobac-terium cutirubrumrdquo International Journal of Systematic Bacte-riology vol 46 no 1 p 347 1996

[56] M C Gutierrez M Kamekura M L Holmes M L Dyall-Smith and A Ventosa ldquoTaxonomic characterization of Halo-ferax sp (ldquoH alicanteirdquo) strain Aa 22 description of Haloferaxlucentensis sp novrdquo Extremophiles vol 6 no 6 pp 479ndash4832002

Archaea 17

[57] M F Mullakhanbhai and H Larsen ldquoHalobacterium volcaniispec nov a Dead Sea halobacterium with a moderate saltrequirementrdquo Archives of Microbiology vol 104 no 1 pp 207ndash214 1975

[58] S D Nuttall and M L Dyall-Smith ldquoCh2 a novel halophilicarchaeon from an Australian solar salternrdquo International Jour-nal of Systematic Bacteriology vol 43 no 4 pp 729ndash734 1993

[59] P D Franzman E Stackebrandt K Sanderson et al ldquoHalobac-terium lacusprofundi sp nov a halophilic bacterium isolatedfrom Deep Lake Antarcticardquo Systematic and Applied Microbi-ology vol 11 pp 20ndash27 1988

[60] G A Tomlinson and L I Hochstein ldquoHalobacterium saccharo-vorum sp nov a carbohydrate metabolizing extremely halo-philic bacteriumrdquoCanadian Journal ofMicrobiology vol 22 no4 pp 587ndash591 1976

[61] A Ventosa M C Gutierrez M Kamekura and M L Dyall-Smith ldquoProposal to transfer Halococcus turkmenicus Halobac-terium trapanicum JCM9743 and strainGSL-11 toHaloterrigenaturkmenica gen nov comb novrdquo International Journal ofSystematic Bacteriology vol 49 no 1 pp 131ndash136 1999

[62] M Kamekura and M L Dyall-Smith ldquoTaxonomy of the familyHalobacteriaceae and the description of two new genera Halo-rubrobacterium and Natrialbardquo Journal of General and AppliedMicrobiology vol 41 no 4 pp 333ndash350 1995

[63] T J Carver KM RutherfordM BerrimanM A RajandreamB G Barrell and J Parkhill ldquoACT the Artemis comparisontoolrdquo Bioinformatics vol 21 no 16 pp 3422ndash3423 2005

[64] S W Cline W L Lam R L Charlebois L C Schalkwyk andW F Doolittle ldquoTransformationmethods for halophilic archae-bacteriardquo Canadian Journal of Microbiology vol 35 no 1 pp148ndash152 1989

Archaea 7

Table 3 Digestion of the PH1 genome by nucleasesa

Nuclease (amount) Proteinase K-treated UntreatedControl PH1 DNA Control PH1 DNA

DNase I (RNase-free)(100U) +b + +b +

Exonuclease III(100U) +b + +b +

Mung bean nuclease(10U) +c

minus +cminus

Nuclease BAL-31(1 U) +b + +b

minus

RNase A(5 120583gmL) +d

minus +dminus

T7 exonuclease(10U) +b + +b

minus

aExtracted nucleic acid from purified PH1 virus was treated with variousnucleases and then analysed by gel electrophoresis to detect whether thenuclease digested (+) or did not digest (minus) the genome Control nucleicacids used to confirm activity of the nucleases were b

120582 DNA ca DNAoligonucleotide and dyeast transfer RNA

Within corresponding ORFs there are blocks of codingsequence with low (or no) similarity These may represent insitu divergence or short recombination events (eg indels)There are also cases where an entire ORF in one virus hasno corresponding homolog in the other because of an inser-tiondeletion event (indel) Lastly there are replacementswhere the sequences within the corresponding regions showlow or no similarity but are flanked by sequence with highsimilarity The largest visible differences seen in Figure 6 aredue to sequence changes within or near the long genes encod-ing capsid proteins VP1 and VP18 A summary of the mainsequence differences between SH1 and PH1 and their loca-tions is given in Table 5 where these divergent regions (DV)are numbered from DV1 to DV18

The sequences of the two viruses are close enough that inmany cases the mechanism for the observed differences canbe inferred For example DV2 is likely the result of a deletionevent that has removed the SH1 ORF10 gene homolog fromthe PH1 genome At either end of the SH1 ORF are almostperfect direct repeats (cggcctgaccggcatgac) that would allowrepeat-mediated deletion to occur removing the interveningsequence Small direct repeats leading to deletions have beendescribed previously in the comparative analysis of Halo-quadratumwalsbyi genomes [20] DV5 appears to be an indelwhere SH1 ORFs 14ndash16 are absent in PH1 and an invertedrepeat (AGCCATG) found at each end of the SH1 divergentregion may be significant in the history of this changeThe proteins specified by SH1 ORFs 14ndash16 are presumablydispensable for PH1 (or their functions supplied by otherproteins) but a homolog of SH1 ORF14 is found also inHHIV-2 (ORF 6) and a homolog of SH1 ORF15 occurs inhalovirus His1 (ORF13) Replacement regions are also com-mon and can occur within ORFs (eg DV3 in capsid proteinVP1) or provide additional or alternative genes (eg DV12and DV13) Several divergent regions occur within the genefor capsid protein VP2 a hot spot for change because of the

repetitive nature of the coding sequence which specifies aprotein with runs of glycines andmany heptapeptide repeatsas described previously for SH1 VP2 [27] DV18 is a largereplacement that significantly alters the predicted length ofthe minor capsid protein VP18 in the two viruses (865519aa for SH1PH1 resp) and also provides SH1 with 3 ORFsthat are different or absent in PH1 ORFs 52 53 and 54 Forexample a homolog of SH1 ORF52 is not present in PH1 butis present in HHIV-2 (putative protein 40) While ORF48 ofPH1 shows no aa sequence homology to SH1 ORFs 53 and 54all of these predicted proteins carry CxxC motifs suggestiveof related functions such as DNA binding activity [36]

The genes of the PH1 genome are syntenic with those ofSH1 and are found in similarly oriented blocks of closelyspaced or overlapping ORFs that suggest that transcription isorganised into operons The programme of transcription ofthe SH1 genome has been reported previously [30] and thesequences in PH1 corresponding to the six promoter regions(P1ndashP6) determined in SH1 showed that five of these (P1ndashP5)are well conserved Only the region corresponding to SH1 P6was poorly conserved (data not shown) but this promoter isstrongly regulated in SH1 and only switches on late in infec-tion [30]

28 Structural Protein Genes The genes coding for the majorvirus structural proteins of PH1 (VP 1ndash4 7 9 10 and 12)were identified byMALDI-TOF (see above) and the genes forthese proteins are found in the same order and approximatepositions as in the SH1 genome (Figure 6(b) red ORFs)Genes coding for minor structural proteins can be deducedby sequence comparison with SH1 so that VP5 and VP6 arelikely to be encoded by PH1ORFs 25 and 29 respectively andVP13 and VP18 by ORFs 23 and 49 respectively This wouldgive PH1 a total of 11 structural proteins the same as SH1Of these VP7 is the most conserved between the two viru-ses (98 aa identity) followed by VP4 VP9 VP3 and VP12(91ndash94 identity) VP5 VP6 and VP13 (82ndash88 iden-tity) VP2 (77) VP1 (65) and lastly the least conservedprotein was VP18 (31)

29 Genomic Loci Related to PH1 and Other Halosphaeroviru-ses Clusters of halosphaerovirus-related genes are present inthe genomes of two recently sequenced haloarchaea Happaucihalophilus and Hbf lacisalsi (Figure 6(a)) Neither ofthese regions appears to represent complete virus genomebut they retain a number of genes that share a similar seq-uence and synteny with the virus genomes depicted belowthem (Figure 6(a)) Both loci show a mixed pattern ofrelatedness to SH1PH1 and HHIV-2 as indicated by ORFsof matching colour and gene name If these genomic locirepresent provirus integrants that have decayed over timethen the relationships they show suggest not only that halo-sphaeroviruses are a diverse virus group but also that asignificant level of recombination occurs between them lead-ing to mosaic gene combinations The two ORFs colouredgreen in theHap paucihalophilus locus (Figure 6) are relatedto ORFs found within or very close to genomic loci ofvirusplasmid genes found in other haloarchaea suggestingadditional links between (pro)viruses [20]

8 Archaea

Table4ORF

anno

tatio

nsof

theh

aloviru

sPH1g

enom

e

ORF

aPo

sitionb

Locustag

Leng

thc(aa)

pKi

Similaritycharacteris

tics

144

1ndash605

HhP

H1gp01

5043

76aa

similarityto

SH1O

RF2(YP271859)

2602ndash802

HhP

H1gp02

66109

80aa

similarityto

SH1O

RF3(YP271860)

3802ndash1059

HhP

H1gp03

8539

85aa

similarityto

SH1O

RF4(YP271861)44

sim

ilarityto

HHIV-2

protein1

41056ndash1283

HhP

H1gp04

7539

85aa

identityto

SH1O

RF5(YP271862)Predictedcoiled-coilregion

51276ndash1686

HhP

H1gp05

136

49

80aa

similarityto

SH1O

RF6(YP271863)Tw

otransm

embraned

omainsA

lsorelated

toHHIV-2

protein2(38

AFD

02283)

andto

Halobiform

alacisalsiORF

ZP09950018

(35

)6

1683ndash1829

HhP

H1gp06

4861

81aa

similarityto

SH1O

RF7(YP271864

)Predictedsig

nalsequence(Sign

alP)

71822ndash2130

HhP

H1gp07

102

58

91aa

similarityto

SH1O

RF8(YP271865)a

nd86to

HHIV-2

putativ

eprotein

3(A

FD02284)O

ther

homologsinclude

Nocardiaproteinpn

f2110(YP1220601)H

rrlacusprofund

iHlac0751

(YP0025654211)ORF

sinactin

ophage

VWB

(AAR2

9707)Frankia(EAN116

57)andgp40

ofMycobacteriu

mph

ageD

ori(AER

47690)

82123ndash2383

HhP

H1gp08

8642

90aa

similarityto

SH1O

RF9(YP271866)

92380ndash2714

HhP

H1gp09

111

52

88aa

similarityto

SH1O

RF11(YP271868)and

also

similarityto

putativ

eprotein

4of

HHIV-2

(AFD

02285)

HlacA

J19877of

Hbflacisa

lsiAJ

5(ZP09950016)

102861ndash300

4HhP

H1gp10

47104

Weaksim

ilarityto

SH1O

RF12

(YP271869)o

nlyover

theN

-term

inal18

resid

ues

113029ndash3100

HhP

H1gp11

23123

Not

presentinSH

1orH

HIV-2

123134ndash7453

HhP

H1gp12

1439

42

Capsid

protein

VP172sim

ilarityto

SH1V

P1(O

RF13Y

P271870)HHIV-2

VP1

(AFD

02286)H

lacA

J19872Hbflacisa

lsiAJ

5Predictedhelix-tu

rn-helixandRu

vA-like

domain(In

terproScan)

137505ndash8227

HhP

H1gp13

240

54

Predicted

P-loop

ATPa

sedo

main(C

OG0433)92aa

similarityto

SH1O

RF17

(YP271874)a

ndalso

similarityto

putativ

eAT

Pase

ofHHIV-2

(AFD

02288)A

TPaseo

fHaladapatus

paucihalophilusD

X253

(ZP0804

6180)HlacA

J19867of

Hbf

lacisalsi(ZP09950014)

148419ndash8228c

HhP

H1gp14

6351

43aa

similarityto

SH1O

RF18

(YP271875)Alanine-richCentraltransm

embraned

omain(Pho

bius)

158856ndash8

416c

HhP

H1gp15

146

41

92aa

similarityto

SH1O

RF19

(YP271876)a

ndalso

toHHIV-2

putativ

eprotein

9(A

FD02290)H

lacA

J19857of

Hbf

lacisalsiAJ

5ZO

D2009

19093of

Happau

cihalophilus

168853ndash9

500c

HhP

H1gp16

215

39

65aa

similarityto

SH1O

RF20

(YP271877)a

ndalso

topu

tativ

eprotein

11of

HHIV-2

(AFD

02292)Z

OD2009

19098of

Hap

paucihalophilusHlacA

J19852Hbflacisa

lsiAJ

5C-

term

inaltransm

embraned

omain(Pho

bius)

179500ndash9

919c

HhP

H1gp17

139

41

79aa

similarityto

SH1O

RF21

(YP271878)a

ndalso

topu

tativ

eprotein

12of

HHIV-2

(AFD

02293)hypotheticalprotein

HlacA

J19847of

Hbflacisa

lsi(ZP09950010)Transm

embraned

omain(Pho

bius)

189923ndash10594c

HhP

H1gp18

223

43

83aa

similarityto

SH1O

RF22

(YP271879)a

ndalso

topu

tativ

eprotein

13of

HHIV-2

(AFD

02294)Z

OD2009

19108of

Hap

paucihalophilusHlacA

J19842of

Hbflacisa

lsiP

redicted

signalsequence(sig

nalP)Con

tains4

CxxC

motifs

1910659ndash10943

HhP

H1gp19

94104

Capsid

protein

VP12

(ORF

19)91aa

similarityto

SH1V

P12(O

RF23Y

P271880)a

ndalso

toVP12of

HHIV-2

(AFD

02295)

andHlacA

J19837of

Hbflacisa

lsiZ

OD2009

19113of

Happau

cihalophilusTw

otransm

embraned

omains

(Pho

bius)

2010960ndash

11517

HhP

H1gp20

185

44

Capsid

protein

VP7(O

RF20)98aa

similarityto

SH1V

P7(O

RF24Y

P271881)a

ndalso

toVP7

ofHHIV-2

(AFD

02296)

ZOD2009

19118

ofHappau

cihalophilusHlacA

J19832of

Hbflacisa

lsiAJ

5

2111519ndash

12217

HhP

H1gp21

232

41

Capsid

protein

VP4(O

RF21)94aa

similarityto

SH1O

RF25

(YP271882)a

ndalso

toVP4

ofHHIV-2

(AFD

02297)

ZOD2009

19123of

Happau

cihalophilusHlacA

J19827of

Hbflacisa

lsiAJ

522

12233ndash12454

HhP

H1gp22

7339

81aa

similarityto

SH1O

RF26

(YP271883)a

ndalso

topu

tativ

eprotein

17of

HHIV-2

(AFD

02298)

2312458ndash12697

HhP

H1gp23

7948

78aa

similarityto

SH1capsid

protein

VP13

(ORF

27Y

P271884)a

ndalso

toVP13of

HHIV-2

(AFD

02299)P

redicted

coil-coildo

main

PredictedC-

term

inaltransm

embraned

omain(Pho

bius)

Archaea 9

Table4Con

tinued

ORF

aPo

sitionb

Locustag

Leng

thc(aa)

pKi

Similaritycharacteris

tics

2412701ndash1506

4HhP

H1gp24

787

39

Capsid

protein

VP2(O

RF24)75aa

similarityto

SH1V

P2(O

RF28Y

P271885)a

ndto

VP2

ofHHIV-2

(AFD

02300)

ZOD2009

19133of

Happau

cihalophilus(ZP

0804

6189)

2515065ndash15961

HhP

H1gp25

298

45

Putativec

apsid

protein

VP5(O

RF25)81aa

similarityto

SH1V

P5(O

RF29Y

P271886)a

ndto

HHIV-2

VP5

(AFD

02301)

HlacA

J19807of

HbflacisalsiAJ

5ZO

D2009

19133of

Happau

cihalophilus

261596

4ndash1644

6HhP

H1gp26

160

46

73aa

similarityto

SH1capsid

protein

VP10

(ORF

30Y

P271887)a

ndto

VP10of

HHIV-2

(AFD

02302)Z

OD2009

19138of

Happau

cihalophilusHlacA

J19802of

Hbflacisa

lsiP

redicted

C-term

inaltransm

embraned

omain(TMHMM)

271644

6ndash16895

HhP

H1gp27

149

42

Capsid

protein

VP9(O

RF27)94aa

similarityto

SH1V

P9(O

RF31Y

P271888)a

ndto

ZOD2009

19143of

Hap

paucihalophilus(ZP

0804

6191)

2816908ndash17921

HhP

H1gp28

337

43

Capsid

protein

VP3(O

RF28)91aa

similarityto

SH1V

P3(O

RF32Y

P271889)a

ndto

NJ7G

2365

ofNa

trinemaspJ7-2

2917928ndash18617

HhP

H1gp29

229

42

Capsidprotein

VP6(O

RF29)83aa

similarityto

SH1V

P6(O

RF33Y

P271890)Also

toNJ7G

3194

ofNa

trinemaspJ7-2and

Hmuk

0476

ofHalom

icrobium

mukohataeiPredictedcarboxypeptid

aser

egulatory-lik

edom

ain(C

arbo

xypepD

reg

pfam

13620)

3018614ndash

18943

HhP

H1gp30

109

49

71aa

similarityto

SH1O

RF34

(YP271891)a

ndto

putativ

eprotein

27of

HHIV-2

(AFD

02308)P

redicted

signalsequence

(signalP)

3119054ndash

19176c

HhP

H1gp31

4050

TwoCx

xCmotifsN

oho

molog

inSH

1orH

HIV-2

3219173ndash19289c

HhP

H1gp32

3841

68aa

similarityto

SH1O

RF37

(YP271894)

3319286ndash

19552c

HhP

H1gp33

8853

58aa

similarityto

SH1O

RF39

(YP271896)a

ndto

HHIV-2

putativ

eprotein

29(A

FD02310)Fou

rCxxCmotifs

3419549ndash

19731c

HhP

H1gp34

6070

Con

tainsC

xxCmotif

Noho

molog

inSH

1orH

HIV-2

3519728ndash20219c

HhP

H1gp35

163

56

93aa

similarityto

SH1O

RF41

(YP271898)a

ndto

putativ

eprotein

30of

HHIV-2

(AFD

02311)

3620216ndash

21415c

HhP

H1gp36

399

50

95aa

similarityto

SH1O

RF42

(YP271899)a

ndto

putativ

eprotein

31of

HHIV-2

(AFD

02312)

3721419ndash

21778c

HhP

H1gp37

11941

91aa

similarityto

SH1O

RF43

(YP271900)a

ndto

putativ

eprotein

32of

HHIV-2

(AFD

02313)

3821762ndash2300

6cHhP

H1gp38

414

42

83aa

similarityto

SH1O

RF44

(YP271901)a

ndto

putativ

eprotein

33of

HHIV-2

(AFD

02314)P

redicted

coil-coiland

helix-tu

rn-helixdo

mains

3923003ndash23158c

HhP

H1gp39

5165

69aa

similarityto

SH1O

RF45

(YP271902)

4023161ndash23370c

HhP

H1gp40

6935

74aa

similarityto

SH1O

RF46

(YP271903)and55sim

ilarityto

putativ

eprotein

35of

HHIV-2

(AFD

02316)

4123422ndash23586c

HhP

H1gp41

5476

86aa

similarityto

SH1O

RF47

(YP271904)Con

tains2

CxxC

motifsand

show

ssim

ilarityto

proteindo

mainfamily

PF14206Arginine-ric

h

4223583ndash24023c

HhP

H1gp42

48

77aa

similarityto

SH1O

RF48

(YP271905)a

ndto

putativ

eprotein

36of

HHIV-2

(AFD

02317)H

ham114540of

Hcc

hamelinensis(ZP11271999)Ph

iCh1p72of

Natrialba

phageP

hiCh

1(NP665989)DUF4

326(pfam14216)

family

domain

4324020ndash

24538c

HhP

H1gp43

172

47

79aa

similarityto

SH1O

RF49

(YP271906)and56sim

ilarityto

putativ

eprotein

37of

HHIV-2

(AFD

02318)

4424535ndash25053c

HhP

H1gp44

172

43

92aa

similarityto

SH1O

RF50

(YP271907)and72sim

ilarityto

putativ

eprotein

38of

HHIV-2

(AFD

02319)

4525050ndash

25277c

HhP

H1gp45

7543

Noho

molog

inSH

1orH

HIV-2

4625274ndash

25387c

HhP

H1gp46

3748

Noho

molog

inSH

1orH

HIV-2

4725533ndash25904

HhP

H1gp47

123

43

61aa

similarityto

SH1O

RF51

(YP271908)66sim

ilarityto

putativ

eprotein

39of

HHIV-2

(AFD

02320)P

redicted

COG1342

domain(D

NAbind

inghelix-tu

rn-helix)

4825901ndash26092c

HhP

H1gp48

6349

Con

tainsC

xxCmotif

Noho

molog

inSH

1orH

HIV-2

4926089ndash

2764

8cHhP

H1gp49

506

40

81aa

similarityto

SH1O

RF55

(YP271912)(bu

tonlyin

theN

-term

inalhalf)H

omolog

ofvirusstru

cturalprotein

VP18

ofHHIV-2

(AFD

02323)

a ORF

swerep

redicted

either

byGLIMMER

orby

manualsearching

forh

omologsintheG

enBa

nkdatabase

b Startandendpo

sitions

ofORF

sare

give

inbp

numbera

ccording

totheP

H1sequenced

epositedatGenBa

nk(KC2

52997)O

RFso

nthec

omplem

entary

strandared

enoted

bythes

uffixc

c Lengthof

thep

redicted

ORF

innu

mbero

faminoacids

10 Archaea

Table 5 Main differences (divergent regions) between the SH1 and PH1 genomes

RegionaSH1 startb SH1 stop Length (bp) PH1 start PH1 stop Length (bp) Comment

DV1 527 650 123 538 604 66

Replacement Both regions have direct repeats at their bordersPH1 has two sets GACCCGGC and CGCTGC while SH1 hasCCCGAC In SH1 this replacement region includes theC-terminal region of ORF2 and the N-terminal region ofORF3 In PH1 it covers only the C-terminal region of ORF1

DV2 2433 2588 155 2386 2387 mdash RMDc from PH1 SH1 repeat at border is TGACCGThisremoves the homolog of SH1 ORF10 from PH1

DV3 3324 3598 274 3137 3416 279 Replacement at the beginning of SH1 ORF13PH1 ORF12(capsid protein VP1)

DV4 6656 7128 472 6478 7085 607 Replacement near the C-terminus of SH1 ORF13PH1 ORF12(capsid protein VP1)

DV5 7491 8341 850 7446 7447 mdash

Indel that results in ORFs14-16 of SH1 not being present inPH1 SH1 ORF14 has a close homolog in HHIV-2 (ORF6) in asimilar position just after the VP1 homolog SH1 ORF15 is aconserved protein in haloarchaea (eg NP 2552A andHmuk 2978) and halovirus His1 (ORF13) Possible invertedrepeat (AGCCATG) at border of SH1 region

DV6 13705 13706 mdash 12818 12851 33RMD from SH1 PH1 repeat at border is CAGCGG(gt)G Thisremoves a part of capsid protein VP2 sequence from the SH1protein

DV7 13789 13863 74 12934 12941 7 Replacement This occurs within VP2 gene of both viruses

DV8 15142 15186 44 14220 14221 mdash RMD from PH1 SH1 repeat at border is TG(tc)CCGACGAand occurs within capsid protein VP2 gene of both viruses

DV9 15303 15317 14 14337 14338 mdash RMD from PH1 SH1 repeat at border is GCCGACGA andoccurs within capsid protein VP2 gene of both viruses

DV10 15504 15527 23 14529 14530 mdash RMD from PH1 SH1 repeat at border is GACGA and occurswithin capsid protein VP2 of both viruses

DV11 20026 20405 379 19019 19173 154Replacement beginning at the start of SH1 ORF35PH1 ORF31This region is longer in SH1 where it includes ORF36 an ORFthat has no homolog present in PH1

DV12 20440 20661 221 19208 19286 78 Replacement This is unequal and includes an ORF in SH1(ORF38) that is not present in PH1

DV13 20943 21085 142 19562 19728 166 Replacement This covers SH1 ORF40PH1 ORF34 Thepredicted proteins are not homologous

DV14 23541 23542 mdash 22171 22197 27 Probable RMD in SH1 PH1 repeat at border is CGTCTCGGand occurs in SH1 ORF44PH1 ORF38

DV15 24506 24521 15 23152 23153 mdash Probable RMD in PH1 SH1 repeat at border is CTCGGT andoccurs near the end of SH1 ORF45PH1 ORF39

DV16 24951 24964 13 23579 23580 mdash Probable RMD in PH1 SH1 repeat at border is CGGTC andoccurs in SH1 ORF47PH1 ORF41

DV17 26449 26450 mdash 25050 25274 224Probable RMD in SH1 PH1 repeat at border is TCATGCG andoccurs near the start of SH1 ORF50PH1 ORF44 and providesan extra ORF for PH1 (ORF45)

DV18 27074 29764 690 25895 26933 1038

Replacement Left border at start of SH1 ORF51PH1 ORF47and extends rightwards into SH1 ORF55PH1 ORF49 (VP18gene) It is an unequal replacement and SH1 has two moreORFs in this region than PH1

aDV Divergent regions between the genomes of SH1 and PH1bStart and stop positions refer to the GenBank sequences of the two viruses SH1 NC 0072171 PH1 KC252997cRMD repeat-mediated deletion event as described in [20]

The recently described halovirus SNJ1 carries manyORFsthat have homologs adjacent to a previously described ViPREofHmc mukohataei (from here on denoted by ViPREHmuk1)a locus that contains genes related to His2 (and other ple-olipoviruses) as well as toHaloferax plasmid pHK2 (probablya provirus) and to two small plasmids ofHqr walsbyi (sim6 kbpL6A and pL6B)The latter relationship provides wider linksto other viruses because one of the PL6 genes (Hqrw 6002)

has homologs in some pleolipoviruses (HPRV-3 and HGPV-1) while another gene (Hqrw 6005) is related to ORF16 ofthe spindle-shaped halovirus His1 [15] After adding the SNJ1gene homologs to ViPREHmuk1 it is extended significantlyand a close inspection of the flanking regions revealed atRNA-ala gene at one end and a partial copy of the sametRNA-ala gene (next to a phage integrase gene) at the otherend (supplementary Figure 1) Genes beyond these points

Archaea 11

Table 6 Transfection of haloarchaea by PH1 DNA

Speciesa Efficiency of transfectiontransformation withPH1 DNAb (PFU120583g of DNA) pUBP2 DNAbc (CFU120583g of DNA)

Har hispanica 53 plusmn 05 times 103 14 plusmn 08 times 104

Har marismortui 45 plusmn 07 times 103 18 plusmn 01 times 104

ldquoHar sinaiiensisrdquo 32 plusmn 06 times 102 mdashHalobacterium salinarum mdash 54 plusmn 42 times 103

Haloferax gibbonsii 28 plusmn 08 times 103 40 plusmn 26 times 103

Haloferax lucentense mdash 57 plusmn 11 times 103

Hfx volcanii mdash 21 plusmn 07 times 105

Hrr lacusprofundi 70 plusmn 22 times 102 95 plusmn 15 times 103

Haloterrigena turkmenica 16 plusmn 17 times 102 24 plusmn 06 times 103

Natrialba asiatica 35 plusmn 39 times 102 31 plusmn 10 times 104aOnly those species positive for transfection andor plasmid transformation are shownbRates of transfection by (nonprotease treated) PH1 DNA or transformation by plasmid pUB2 are averages of three independent experiments each performedin duplicate (plusmn standard deviation)cTransformants were selected on plates with 2 4 or 6120583gmL simvastatin depending on the strainmdash no plaques or colonies observed

appear to be related to cellular metabolism ViPREHmuk1is now 39379 nt in length flanked by a 59 nt direct repeat(potential att sites) and includes 54 ORFs many of which arerelated to known haloviruses are virus-like (eg integrasesmethyltransferases transcriptional regulators DNA methyl-transferase and DNA glycosylase) are homologs in or nearother knownViPREs or show little similarity to other knownproteins This locus also includes an ORC1CDC6 homolog(Hmuk 0446) which could provide a replication function

The mixture of very different virus and plasmid genesseen within ViPREHmuk1 is remarkable but it also revealshomologs found in or adjacent to genomic loci of otherspecies such as the gene homologs of Har marismortuiseen in the left end of ViPREHmuk1 (supplementary Figure1) When these genes are added to the previously describedHaloarcula locus (ViPREHmuk1) it is also extended signifi-cantly It becomes 25550 bp in length encompasses genesfrom Hmar 2382 to Hmar 2404 and is flanked by a fulltRNA-ala gene at one end and a partial copy at the other Itcarries an ORC1CDC6 homolog an integrase and a phiH-like repressor as well as pleolipovirus homologs

210 Transfection of Haloarchaea by PH1 DNA PH1 DNAwas introduced into Har hispanica cells using the PEGmethod [37] and the cells were screened for virus productionby plaque assay Figure 7 shows the increase in transfectedcells with input (nonproteinase K-treated) virus DNA Theestimated efficiencywas 53times 103 PFUper120583gDNA PH1DNAthat had been treated with proteinase K or with DNAase I(RNase-free) did not produce plaques (data not shown) PH1DNA was found to transfect six species of haloarchaea otherthanHar hispanica includingmembers of the generaHaloar-cula Haloferax Halorubrum Haloterrigena and Natrialba(Table 6) For these experiments the virus production fromtransfected cells was detected by plaque assay on indicatorlawns of Har hispanica

0

0

200 400 600 800 1000

Titre

(PFU

mL)

104

103

102

PH1 DNA (ng)

Figure 7 Transfection ofHar hispanica cells with PH1 DNA Vary-ing amounts of nonproteinase K-treated PH1 DNAwere introducedinto cells of Har hispanica using the PEG method [64] Cells werethen screened for infective centres by plaque assay Data shown is theaverage of three independent experiments performed in duplicateError bars represent one standard deviation of themean If protease-treated DNA was used no transfectant plaques were observed

3 Discussion

PH1 is very similar in particlemorphology genome structureand sequence to the previously described halovirus SH1 [816] Like SH1 it has long inverted terminal repeat sequencesand terminal proteins indicative of protein-primed replica-tion Purified PH1 particles have a relatively low buoyant

12 Archaea

density and are chloroform sensitive and show a layeredcapsid structure consistent with the likely presence of aninternalmembrane layer as has been shown for SH1HHIV-2and SNJ1 [13 26]The particle stability of PH1 to temperaturepH and reduced salt was similar to SH1 [8] The structuralproteins of PH1 are very similar in sequence and relativeabundance to those of SH1 so the two viruses are likely toshare the same particle geometry [16]

Viruses that use protein-primed replication such as bac-teriophageΦ29 carry a viral type BDNApolymerase that caninteract specifically with the proteins attached to the genomictermini and initiate strand synthesis [38] ArchaeovirusesHis1 [7] and Acidianus bottle-shaped virus [39] probablyuse this mode of replication However a polymerase genecannot be found in the genomes of PH1 SH1 or HHIV-2DNA polymerases are large enzymes and the only ORF ofSH1 that was long enough to encode such an enzyme andhad not previously been assigned as a structural protein wasORF55 which specified an 865 aa protein that contained noconserved domains indicative of polymerases [27]The recentstudy of HHIV-2 showed that the corresponding ORF in thisvirus (gene 42) is a structural protein of the virion and byinference this is likely also for the corresponding proteinsof SH1 and PH1 (no homolog is present in SNJ1) Withouta polymerase gene these viruses must use a host enzymebut searches for a viral type B DNA polymerase in thegenome of the hostHar hispanica or in the genomes of othersequenced haloarchaea did not find anymatchesThis arguesstrongly for a replication mechanism that is different to thatexemplified byΦ29 An attractive alternative is that displayedby Streptomyces linear plasmids which have 51015840-terminal pro-teins and use a cellular polymerase for replication [40] Theterminal proteins are not used for primary replication but forend patching [41] and it has been shown that if the term-inal repeat sequences are removed from these plasmids andthe ends ligated they can replicate as circular plasmids [42]This provides a testable hypothesis for the replication of halo-sphaeroviruses and also offers a pathway for switchingbetween linear (eg PH1) and circular (eg SNJ1) forms Useof a host polymerase would also fit with the ability of SH1 andPH1 DNA to transfect many different haloarchaeal species[31] as these enzymes and theirmode of action are highly con-served

The growth characteristics of PH1 inHar hispanica bothby single-cell burst and single-step growth experiments gavevalues of 50ndash100 virusescell significantly less than that ofSH1 (200 virusescell) [8] and HHIV-2 (180 virusescell) [13]The latter virus is lytic whereas SH1 PH1 and SNJ1 start pro-ducing extracellular viruswell before any decrease in cell den-sity indicating that virus release can occur without lysisThishas been most clearly shown with SH1 where almost 100of cells can be infected [8] Infection ofHar hispanica by PH1was less efficient (sim30 of cells) but the kinetics of virus pro-duction in single-step growth curves of SH1 and PH1 aresimilar with virus production occurring over several hoursrather than at a clearly defined time after infection The twoviruses are also very closely related and infect the same hostspecies so it is likely that they use the same scheme for cellexit For both viruses the cell density eventually decreases in

single-step cultures showing that virus infection does resultin cell death This mode of exit appears to maximise theproduction of the virus over time as the host cells survive foran extended period In natural hypersaline waters cell num-bers are usually high but growth rates are low [12] and thehigh incident UV (on often shallow ponds) would damagevirus DNA so reducing the half-life of released virus part-icles [43] These factors may have favoured the exit strategydisplayed by these viruses improving their chance of trans-mission to a new host

The motif GATC was absent in the genomes of PH1 SH1andHHIV-2 and themotif CTAGwas either absent or greatlyunderrepresented All three viruses share the same host butGATC and CTAG motifs are plentiful in the Har hispanicagenome (accessions CP002921 and CP002923) Avoidance ofthesemotifs in halovirusesHis1 His2 HF1 andHF2 has beenreported previously [7] Such purifying selection commonlyresults from host restriction enzymes and inHfx volcanii itis exactly these two motifs that are targeted [44] One Hfxvolcanii enzyme recognises A-methylated GATC sites [45]and another recognises un-methylated CTAG sites (whichare protected by methylation in the genome) In the currentstudy this could explain the negative transfection results forHfx volcanii as the PH1 genome contains two CTAGmotifsBy comparison the SH1 genome contains no CTAG motifsand is able to transfect Hfx volcanii [31]

Comparison of the PH1 and SH1 genomes allowed adetailed picture of the natural variation occurring betweenclosely related viruses revealing likely deletion events medi-ated by small repeats a process described previously in astudy of Hqr walsbyi and termed repeat-mediated deletion[20] There are also many replacements including blocks ofsequence that occur within long open reading frames (suchas the capsid protein genes VP1 and VP18) VP3 and VP6 areknown to form the large spikes on the external surface of theSH1 virion and presumably one or both interact with host cellreceptors [16] PH1 and SH1 have identical host ranges con-sistent with the high sequence similarity shown between theircorresponding VP3 and VP6 proteins

A previous study of SH1 could not detect transcriptsacross annotated ORFs 1ndash3 or 55-56 [30] ORFs 1 and 56occur in the terminal inverted repeat sequence and in thepresent study it was found that there were no ORFs cor-responding to these in the PH1 genome Given the close simi-larity of the two genomes the comparative data are in agree-ment with the transcriptional data indicating that these SH1ORFs are not used SH1 ORFs 2 and 3 are more problematicas good homologs of these are also present in PH1 The con-flict between the transcriptional data and the comparativegenomic evidence requires further experimental work toresolveThe status of SH1ORF55 has recently been confirmedby studies of the related virus HHIV-2 (discussed above)

The pleolipovirus group has expanded dramatically inthe last few years and it now comprises a diverse group ofviruseswith different genome types and replication strategiesWhat is evenmore remarkable is that a similar expansion cannow be seen with SH1-related viruses which includes viruseswith at least two replication modes and genome types (linearand circular) Evidence from haloarchaeal genome sequences

Archaea 13

show cases of not only provirus integrants or plasmids (egpKH2 and pHH205) but also genomic loci (ViPREs) thatcontain cassettes of virus genes from different sources andat least two cases (described in this study) have likely attsites that indicate circularisation and mobility While thereis a clear relationship between viruses with circular ds- andssDNA genomes that replicate via the rolling circle method(ie the ssDNA form is a replication intermediate) it ismore difficult to explain how capsid gene cassettes can movebetween these viruses and those with linear genomes andterminal proteins Within the pleolipoviruses His2 has alinear genome that contains a viral type B DNA polymeraseand terminal proteins while all the other described membershave circular genomes that contain a rep homolog and prob-ably replicate via the rolling circle method [14 19] Similarlyamong the halosphaeroviruses PH1 SH1 and HHIV-2 havelinear dsDNA with terminal proteins and probably replicatein the same way but the related SNJ1 virus has a circulardsDNA genome The clear relationships shown by the capsidgenes of viruses within each group plus the connectionsshown to haloviruses outside each group (eg the DNA poly-merases of His1 and His2) all speak of a vigorous meansof recombination one that can readily switch capsid genesbetween viruses with radically different replication strategiesHow is the process most likely to operate One possibility issuggested by ViPREs which appear to be mobile collectionsof capsid and replication genes from different sources Theyoffer fixed locations for recombination to occur providegene cassettes that can be reassorted to produce novel virusgenomes and some can probably recombine out as circularforms For example ViPREHmuk1 carries genes related topleolipoviruses halosphaeroviruses and other haloviruses Itis yet to be determined if any of the genes carried in theseloci are expressed in the cell (such as the genes for capsidproteins) or if ViPREs provide any selective advantage to thehost

4 Materials and Methods

41 Water Sample A water sample was collected in 1998from Pink Lake (32∘001015840 S and 115∘301015840 E) a hypersalinelake on Rottnest Island Western Australia Australia It wasscreened in the same year for haloviruses using the methodsdescribed in [8] A single plaque on a Har hispanica lawnplate was picked and replaque purified The novel haloviruswas designated PH1

42 Media Strains and Plasmids The media used in thisstudy are described in the online resource The Halohand-book (httpwwwhaloarchaeacomresourceshalohandbookindexhtml) Artificial salt water containing 30 (wv) totalsalts comprised of 4M NaCl 150mM MgCl

2 150mM

MgSO4 90mM KCl and 35mM CaCl

2and adjusted to pH

75 using sim2mL 1M Tris-HCl (pH 75) per litre MGM con-taining 12 18 or 23 (wv) total salts and HVD (halo-virus diluent) were prepared from the concentrated stock aspreviously described [46] Bacto-agar (Difco Laboratories)was added to MGM for solid (15 gL) or top-layer (7 gL)media

Table 1 lists the haloarchaea used in this study Allhaloarchaea were grown aerobically at 37∘C in either 18 or23 (wv)MGM(depending on the strain) andwith agitation(except for ldquoHar sinaiiensisrdquo)The plasmids used were pBlue-script II KS+ (Stratagene Cloning Systems) pUBP2 [47]and pWL102 [48] pUBP2 and pWL102 were first passagedthrough E coli JM110 [49] to prevent dam-methylationof DNA which has been shown to reduce transformationefficiency in some strains [45]

43 Negative-Stain Electron Microscopy The method fornegative-stain TEM was adapted from that of V Tarasovdescribed in the online resource The Halohandbook (httpwwwhaloarchaeacomresourceshalohandbookindexhtml)A 20120583L drop of the sample was placed on a clean surfaceand the virus particles were allowed to adsorb to a Formvarfilm 400-mesh copper grid (ProSciTech) for 15ndash2min Theywere then negatively stained with a 20120583L drop of 2 (wv)uranyl acetate for 1-2min Excess liquid was absorbed withfilter paper and the grid was allowed to air dry Grids wereexamined either on a Philips CM 120 BioTwin transmissionelectron microscope (Royal Philips Electronics) operating atan accelerating voltage of 120 kV or on a Siemens Elmiskop102 transmission electron microscope (Siemens AG) operat-ing at an accelerating voltage of 120 kV

44 Virus Host Range Twelve haloarchaeal strains from thegenera Haloarcula Halobacterium Haloferax HalorubrumandNatrialba (Table 1)Thirteen naturalHalorubrum isolates(H Camakaris unpublished data) and five uncharacterizedhaloarchaeal isolates fromLakeHardy Pink Lake (inWesternAustralia) and Serpentine Lake (D Walker and M K Seahunpublished data) were screened for PH1 susceptibilityLysates from PH1-infected Har hispanica cultures (1 times1011 PFUmL) were spotted onto lawns of each strain (usingmedia with 12 or 18 (wv)MGM depending on the strain)and incubated for 2ndash5 days at 30 and at 37∘C

45 Large Scale Virus Growth and Purification Liquid cul-tures of PH1 were grown by the infection (MOI 005) of anearly exponential Har hispanica culture in 18 (wv) MGMCultures were incubated aerobically at 37∘C with agitationfor 3 days Clearing (ie complete cell lysis) of PH1-infectedcultures did not occur and consequently cultures were harv-ested when the absorbance at 550 nm reached the minimumand the titre (determined by plaque assay) was at the maxi-mum usually sim1010ndash1011 PFUmL Virus was purified usingthe method described previously for SH1 [8 31] except thatafter the initial low speed spin (Sorvall GSA 6000 rpm30min 10∘C) virus was concentrated from the infected cul-ture by centrifugation at 26000 rpm (13 hr 10∘C) onto acushion of 30 (wv) sucrose in HVDThe pellet was resus-pended in a small volume of HVD and loaded onto a pre-formed linear 5ndash70 (wv) sucrose gradient followed byisopycnic centrifugation in 13 gmL CsCl (Beckman 70Ti60000 rpm 20 hr 10∘C) The white virus band occurredat a density of 129 gmL and was collected and diluted inhalovirus diluent (HVD see Section 4) and the virus pelleted(Beckman SW55 35000 rpm 75min 10∘C) and resuspended

14 Archaea

in a small volume of HVD and stored at 4∘C Virus recoveryat the major stages of a typical purification is given insupplementary Table 1 The specific infectivity of pure virussolutions was determined as the ratio of the PFUmL to theabsorbance at 260 nm

46 PH1 Single-Step Growth Curve An early exponentialphase culture of Har hispanica grown in 18 (wv) MGMwas infected with PH1 (MOI 50) Under these conditions thepercentage of infected cells was approximately 30 After anadsorption period of 1 hr at 37∘C the cells were washed threetimes with 18 (wv) MGM (at room temperature) resus-pended in 100mL 18 (wv) MGM (these methods ensuredthe removal of all residual virus) and incubated at 37∘Cwith shaking (100 rpm) Samples were removed at hourlyintervals for measurements of absorbance at 550 nm andthe number of infective centres Immediately after samplingtitres were determined by plaque assay with an indicator lawnof Har hispanica Each experiment was performed in tripli-cate

47 Halovirus Stability After various treatments sampleswere removed and diluted in HVD and virus titres weredetermined by plaque assay on Har hispanica Each experi-ment was performed in triplicate and representative data areshown Chloroform sensitivity was examined by the exposureof PH1 lysates in 18 (wv) MGM to chloroform in a volumeratio of 1 4 (chloroform to lysate) Incubation was at roomtemperature with constant agitation At appropriate timepoints the mix was allowed to settle and a sample wasremoved from the upper layer and diluted in HVDThe effectof a low ionic environment was examined by diluting PH1lysates (in 18 (wv) MGM) into double distilled H

2O in a

volume ratio of 1 1000 (lysate to double distilledH2O) Incu-

bation was at room temperature with constant agitationSamples were removed at various times and diluted in HVDThe pH stability of PH1 was determined by dilution of PH1lysates in 18 (wv)MGMin the appropriate pHbuffer (HVDbuffered with appropriate Tris-HCl) in a volume ratio of1 100 (lysate to buffer) Incubation was at room temperaturewith constant agitation After 30min samples were removedand diluted in HVD Thermal stability of PH1 was examinedby a 1 hr incubation of a virus lysate in 18 (wv) MGM atdifferent temperatures after which they were brought quicklyto room temperature diluted in HVD and titrated

48 Protein Procedures To remove salts purified virus prepa-rations were mixed with trichloroacetic acid (10 (vv) finalconcentration) and incubated on ice for 15min to allowthe proteins to precipitate After centrifugation (16000 g15min room temperature) the precipitate was washed threetimes in acetone dried and resuspended in double distilledH2O Proteins were dissolved in Laemmli sample buffer with

48mM 120573-mercaptoethanol [50] heated in boiling water for5min then separated on 12 (wv) NuPAGE Novex Bis-Tris Gels using MES-SDS running buffer according to themanufacturerrsquos directions (Invitrogen) After electrophoresisgels were rinsed in double distilled H

2O and stained with

01 (wv) Brilliant Blue G in 40 (vv) methanol and

10 (vv) acetic acid Gels were destained with severalchanges of 40 (vv) methanol and 10 (vv) acetic acidAlternatively gels were stained with GelCode GlycoproteinStaining Kit Pro-Q Emerald 488 Glycoprotein Gel and BlotStain Kit SYPRO Ruby Protein Gel Stain according tothe manufacturerrsquos directions (Invitrogen Molecular ProbesPierce Biotechnology)

Protein bands were cut from the gels and sent to theAustralian Proteome Analysis Facility (Macquarie Univer-sity) for trypsin digestion and analysis by matrix assistedlaser desorption ionisation time of flight mass spectrometry(MALDI-TOF MS) on an Applied Biosystems 4700 Pro-teomics Analyser (Applied Biosystems)

49 DNA Procedures Proteinase K-treated and nonpro-teinaseK-treatedDNApreparationsweremade frompurifiedPH1 using the methods described previously for SH1 [8]DNA was separated on 1 (wv) agarose gels in Tris-acetate-EDTA electrophoresis buffer and was stained with ethidiumbromide (Sigma-Aldrich)120582 DNA DNase I (RNase-free) exonuclease III mung

bean nuclease nuclease BAL-31 T4 DNA polymerase T4DNA ligase T7 exonuclease and type II restriction endonu-cleases were purchased from New England Biolabs RNase Aand yeast transfer RNA were purchased from Sigma-AldrichProteinase K was purchased from Promega Oligonucleotideprimers were purchased from Geneworks Australia

To clone fragments of the PH1 genome purified virusDNA was digested either with AccI Eco0109I and MseI orSmaI restriction endonucleases blunted with DNA Poly-merase I Large (Klenow) Fragment where appropriate andligated into AccI- Eco0109I- or SmaI-digested pBluescript IIKS+ using T4 DNA ligase according to the manufacturerrsquosinstructions (New England Biolabs) The DNA was intro-duced intoE coliXL1-Blue and transformants grown onLuriaagar containing 15 120583gmL tetracycline and 100 120583gmL ampi-cillinThe resulting cloneswere sequenced and the sequenceswere used to design specific oligonucleotide primers for PCRamplification andor primer walking using the virus genome(or specific restriction fragments) as templates

To amplify PH1 nucleic acid approximately 10 ng virusDNA was combined with 500 nM primers 100120583M of eachdNTP 1U Deep Vent DNA Polymerase (New England Bio-labs) and 1 timesThermoPol buffer (New England Biolabs) ina total reaction volume of 50 120583L Template was denatured at95∘C for 10min followed by 30 cycles of denaturation at 95∘Cfor 30 sec annealing at 56∘C for 30 sec extension at 75∘C for2min and a final extension at 75∘C for 10min PCR reactionswere performed on a PxE02ThermoCycler (Thermo ElectroCorporation) Sequencing reactions using 32 pmol primerwere performed by the dideoxy chain termination methodusing ABI PRISM Big Dye Terminator Mix version 31 onan ABI 3100 capillary sequencer (Applied Biosystems) bythe Applied Genetic Diagnostics Sequencing Service at theDepartment of Pathology (The University of Melbourne)

To obtain terminal genomic fragments PH1 DNA wasdigested by the enzyme Pas1 and the head (sim700 bp) andthe tail (sim1300 bp) fragments were purified by the QIAEXII gel extraction kits and treated with 05M piperidine for

Archaea 15

2 hr at 37∘C to remove protein residues The piperidine-treated fragments were sequenced by the Genomics BiotechCompany (Taipei Taiwan) using two primers TGACCA-ATTAATTAGGCCGGTTCGCC (PH1 head-R) and GTGC-CATACTGCTACAATTCT (PH1 tail-F)

Four hundred fifty-four whole genome sequencing PH1DNA samples (sim5 120583g) were sequenced using parallel pyrose-quencing on a Roche 454 Genome Sequencer System atMission Biotech (Taipei Taiwan) The largest contig was27399with 8803 reads (Newbler version 27) with 27387 posi-tions being of Q40 quality The average depth of coverage foreach base was 829TheGenBank accession for the entire PH1sequence is KC252997

Acknowledgments

The authors thank Carolyn Bath Mei Kwei (Joan) Seahand Danielle Walker for their early work in PH1 characteri-zation Simon Crawford Jocelyn Carpenter and Anna Fried-huber for their assistance with the transmission electronmicroscopy and Tiffany Cowie and Voula Kanellakis fortheir sequencing assistance This paper has been facilitatedby access to the Australian Proteome Analysis Facility estab-lished under the Australian Governmentrsquos Major NationalResearch Facilities program SLT was supported by a grantfrom the Biodiversity Research Center Academia SinicaTaiwan

References

[1] M Krupovic M F White P Forterre and D PrangishvilildquoPostcards from the edge structural genomics of archaealvirusesrdquo Advances in Virus Research vol 82 pp 33ndash62 2012

[2] P Stolt andW Zillig ldquoGene regulation in halophage 120601Hmdashmorethan promotersrdquo Systematic and Applied Microbiology vol 16no 4 pp 591ndash596 1994

[3] R Klein U Baranyi N Rossler B Greineder H Scholz and AWitte ldquoNatrialba magadii virus 120593Ch1 first complete nucleotidesequence and functional organization of a virus infecting ahaloalkaliphilic archaeonrdquo Molecular Microbiology vol 45 no3 pp 851ndash863 2002

[4] E Pagaling R D Haigh W D Grant et al ldquoSequence analysisof an Archaeal virus isolated from a hypersaline lake in InnerMongolia Chinardquo BMC Genomics vol 8 article 410 2007

[5] S L Tang SNuttall andMDyall-Smith ldquoHalovirusesHF1 andHF2 evidence for a recent and large recombination eventrdquo Jour-nal of Bacteriology vol 186 no 9 pp 2810ndash2817 2004

[6] C Bath and M L Dyall-smith ldquoHis1 an archaeal virus of theFuselloviridae family that infectsHaloarcula hispanicardquo Journalof Virology vol 72 no 11 pp 9392ndash9395 1998

[7] C Bath T Cukalac K Porter andM L Dyall-Smith ldquoHis1 andHis2 are distantly related spindle-shaped haloviruses belongingto the novel virus group Salterprovirusrdquo Virology vol 350 no1 pp 228ndash239 2006

[8] K Porter P Kukkaro J K H Bamford et al ldquoSH1 a novelspherical halovirus isolated from an Australian hypersalinelakerdquo Virology vol 335 no 1 pp 22ndash33 2005

[9] M Dyall-Smith S L Tang and C Bath ldquoHaloarchaeal viruseshow diverse are theyrdquo Research in Microbiology vol 154 no 4pp 309ndash313 2003

[10] A Oren G Bratbak and M Heldal ldquoOccurrence of virus-likeparticles in the Dead Seardquo Extremophiles vol 1 no 3 pp 143ndash149 1997

[11] T Sime-Ngando S Lucas A Robin et al et al ldquoDiversityof virus-host systems in hypersaline Lake Retba SenegalrdquoEnvironmental Microbiology vol 13 no 8 pp 1956ndash1972 2010

[12] N Guixa-Boixareu J I Calderon-Paz M Heldal G Bratbakand C Pedros-Alio ldquoViral lysis and bacterivory as prokaryoticloss factors along a salinity gradientrdquoAquaticMicrobial Ecologyvol 11 no 3 pp 215ndash227 1996

[13] S T Jaakkola R K Penttinen S T Vilen et al ldquoClosely relatedarchaeal Haloarcula hispanica icosahedral viruses HHIV-2 andSH1 have nonhomologous genes encoding host recognitionfunctionsrdquo Journal of Virology vol 86 no 9 pp 4734ndash47422012

[14] E Roine P Kukkaro L Paulin et al ldquoNew closely related halo-archaeal viral elements with different nucleic acid typesrdquo Jour-nal of Virology vol 84 no 7 pp 3682ndash3689 2010

[15] A Sencilo L Paulin S KellnerMHelm and E Roine ldquoRelatedhaloarchaeal pleomorphic viruses contain different genometypesrdquo Nucleic Acids Research vol 40 no 12 pp 5523ndash55342012

[16] H T Jaalinoja E Roine P Laurinmaki H M Kivela D HBamford and S J Butcher ldquoStructure and host-cell interactionof SH1 a membrane-containing halophilic euryarchaeal virusrdquoProceedings of the National Academy of Sciences of the UnitedStates of America vol 105 no 23 pp 8008ndash8013 2008

[17] M K Pietila N S Atanasova VManole et al ldquoVirion architec-ture unifies globally distributed pleolipoviruses infecting halo-philic archaeardquo Journal of Virology vol 86 no 9 pp 5067ndash50792012

[18] M L Holmes and M L Dyall-Smith ldquoA plasmid vector witha selectable marker for halophilic archaebacteriardquo Journal ofBacteriology vol 172 no 2 pp 756ndash761 1990

[19] M L Holmes F Pfeifer andM L Dyall-Smith ldquoAnalysis of thehalobacterial plasmid pHK2 minimal repliconrdquo Gene vol 153no 1 pp 117ndash121 1995

[20] M L Dyall-Smith F Pfeiffer K Klee et al ldquoHaloquadratumwalsbyi limited diversity in a Global Pondrdquo PLoS ONE vol 6no 6 Article ID e20968 2011

[21] S Casjens ldquoProphages and bacterial genomics what have welearned so farrdquoMolecular Microbiology vol 49 no 2 pp 277ndash300 2003

[22] R W Hendrix M C M Smith R N Burns M E Ford andG F Hatfull ldquoEvolutionary relationships among diverse bacte-riophages and prophages all the worldrsquos a phagerdquo Proceedings ofthe National Academy of Sciences of the United States of Americavol 96 no 5 pp 2192ndash2197 1999

[23] D Botstein ldquoA theory ofmodular evolution for bacteriophagesrdquoAnnals of the New York Academy of Sciences vol 354 pp 484ndash491 1980

[24] M Krupovic S Gribaldo D H Bamford and P Forterre ldquoTheevolutionary history of archaeal MCM helicases a case study ofvertical evolution combinedwithHitchhiking ofmobile geneticelementsrdquo Molecular Biology and Evolution vol 27 no 12 pp2716ndash2732 2010

[25] M Krupovic P Forterre and D H Bamford ldquoComparativeanalysis of the mosaic genomes of tailed archaeal viruses andproviruses suggests common themes for virion architecture andassembly with tailed viruses of bacteriardquo Journal of MolecularBiology vol 397 no 1 pp 144ndash160 2010

16 Archaea

[26] Z Zhang Y Liu S Wang et al et al ldquoTemperate membrane-containing halophilic archaeal virus SNJ1 has a circular dsDNAgenome identical to that of plasmid pHH205rdquoVirology vol 434no 2 pp 233ndash241 2012

[27] D H Bamford J J Ravantti G Ronnholm et al ldquoConstituentsof SH1 a novel lipid-containing virus infecting the halophiliceuryarchaeonHaloarcula hispanicardquo Journal of Virology vol 79no 14 pp 9097ndash9107 2005

[28] H M Kivela E Roine P Kukkaro S Laurinavicius P Somer-harju andDH Bamford ldquoQuantitative dissociation of archaealvirus SH1 reveals distinct capsid proteins and a lipid corerdquoVirology vol 356 no 1-2 pp 4ndash11 2006

[29] H Jaalinoja Electron Cryo-Microscopy Studies of Bacteriophage8 and Archaeal Virus SH1 University of Helsinki HelsinkiFinland 2007

[30] K Porter B E Russ J Yang andM L Dyall-Smith ldquoThe trans-cription programme of the protein-primed halovirus SH1rdquoMi-crobiology vol 154 no 11 pp 3599ndash3608 2008

[31] K Porter and M L Dyall-Smith ldquoTransfection of haloarchaeaby the DNAs of spindle and round haloviruses and the useof transposon mutagenesis to identify non-essential regionsrdquoMolecular Microbiology vol 70 no 5 pp 1236ndash1245 2008

[32] S E Luria General Virology John Wiley and Sons New YorkNY USA 1953

[33] C A Thomas K Saigo E McLeod and J Ito ldquoThe separationofDNA segments attached to proteinsrdquoAnalytical Biochemistryvol 93 pp 158ndash166 1979

[34] A L Delcher D Harmon S Kasif OWhite and S L SalzbergldquoImproved microbial gene identification with GLIMMERrdquoNucleic Acids Research vol 27 no 23 pp 4636ndash4641 1999

[35] R Zhang Y Yang P Fang et al ldquoDiversity of telomere palin-dromic sequences and replication genes among Streptomyceslinear plasmidsrdquo Applied and Environmental Microbiology vol72 no 9 pp 5728ndash5733 2006

[36] X Wang H S Lee F J Sugar F E Jenney M W W Adamsand J H Prestegard ldquoPF0610 a novel winged helix-turn-helix variant possessing a rubredoxin-like Zn ribbon motiffrom the hyperthermophilic archaeon Pyrococcus furiosusrdquoBiochemistry vol 46 no 3 pp 752ndash761 2007

[37] S W Cline L C Schalkwyk and W F Doolittle ldquoTransfor-mation of the archaebacterium Halobacterium volcanii withgenomic DNArdquo Journal of Bacteriology vol 171 no 9 pp 4987ndash4991 1989

[38] S Kamtekar A J Berman J Wang et al ldquoThe 12059329 DNA poly-merase protein-primer structure suggests a model for the ini-tiation to elongation transitionrdquo EMBO Journal vol 25 no 6pp 1335ndash1343 2006

[39] X Peng T Basta M Haring R A Garrett and D PrangishvilildquoGenome of theAcidianus bottle-shaped virus and insights intothe replication and packaging mechanismsrdquo Virology vol 364no 1 pp 237ndash243 2007

[40] C C Yang C H Huang C Y Li Y G Tsay S C Lee and CW Chen ldquoThe terminal proteins of linear Streptomyces chrom-osomes and plasmids a novel class of replication primingproteinsrdquo Molecular Microbiology vol 43 no 2 pp 297ndash3052002

[41] C C Yang Y H Chen H H Tsai C H Huang T W Huangand C W Chen ldquoIn vitro deoxynucleotidylation of the term-inal protein of Streptomyces linear chromosomesrdquo Applied andEnvironmental Microbiology vol 72 no 12 pp 7959ndash79612006

[42] P C Chang and S N Cohen ldquoBidirectional replication froman internal origin in a linear Streptomyces plasmidrdquo Science vol265 no 5174 pp 952ndash954 1994

[43] S W Wilhelm W H Jeffrey C A Suttle and D L MitchellldquoEstimation of biologically damaging UV levels in marinesurface waters with DNA and viral dosimetersrdquo Photochemistryand Photobiology vol 76 no 3 pp 268ndash273 2002

[44] T Allers and M Mevarech ldquoArchaeal geneticsmdashthe third wayrdquoNature Reviews Genetics vol 6 no 1 pp 58ndash73 2005

[45] M L Holmes S D Nuttall and M L Dyall-Smith ldquoConstruc-tion and use of halobacterial shuttle vectors and further studieson Haloferax DNA gyraserdquo Journal of Bacteriology vol 173 no12 pp 3807ndash3813 1991

[46] S D Nuttall and M L Dyall-Smith ldquoHF1 and HF2 novelbacteriophages of halophilic archaeardquo Virology vol 197 no 2pp 678ndash684 1993

[47] U Blaseio and F Pfeifer ldquoTransformation of Halobacteriumhalobium development of vectors and investigation of gas vesi-cle synthesisrdquo Proceedings of the National Academy of Sciences ofthe United States of America vol 87 no 17 pp 6772ndash6776 1990

[48] W L Lam and W F Doolittle ldquoShuttle vectors for the archae-bacterium Halobacterium volcaniirdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 86 no14 pp 5478ndash5482 1989

[49] C Yanisch-Perron J Vieira and J Messing ldquoImproved M13phage cloning vectors and host strains nucleotide sequences ofthe M13mp18 and pUC19 vectorsrdquo Gene vol 33 no 1 pp 103ndash119 1985

[50] U K Laemmli ldquoCleavage of structural proteins during theassembly of the head of bacteriophage T4rdquo Nature vol 227 no5259 pp 680ndash685 1970

[51] D G Burns H M Camakaris P H Janssen and M L Dyall-Smith ldquoCombined use of cultivation-dependent and culti-vation-independent methods indicates that members of mosthaloarchaeal groups in anAustralian crystallizer pond are culti-vablerdquo Applied and Environmental Microbiology vol 70 no 9pp 5258ndash5265 2004

[52] G Juez F Rodriguez-Valera A Ventosa and D J KushnerldquoHaloarcula hispanica spec nov and Haloferax gibbonsii specnov two new species of extremely halophilic archaebacteriardquoSystematic and Applied Microbiology vol 8 pp 75ndash79 1986

[53] A Oren M Ginzburg B Z Ginzburg L I Hochstein and BE Volcani ldquoHaloarcula marismortui (Volcani) sp nov nomrev an extremely halophilic bacterium from the Dead SeardquoInternational Journal of Systematic Bacteriology vol 40 no 2pp 209ndash210 1990

[54] M Torreblanca F Rodriguez-Valera G Juez A Ventosa MKamekura and M Kates ldquoClassification of non-alkaliphilichalobacteria based on numerical taxonomy and polar lipidcomposition and description of Haloarcula gen nov andHaloferax gen novrdquo Systematic and Applied Microbiology vol8 pp 89ndash99 1986

[55] A Ventosa and A Oren ldquoHalobacterium salinarum nom cor-rig a name to replace Halobacterium salinarium (Elazari-Vol-cani) and to include Halobacterium halobium and Halobac-terium cutirubrumrdquo International Journal of Systematic Bacte-riology vol 46 no 1 p 347 1996

[56] M C Gutierrez M Kamekura M L Holmes M L Dyall-Smith and A Ventosa ldquoTaxonomic characterization of Halo-ferax sp (ldquoH alicanteirdquo) strain Aa 22 description of Haloferaxlucentensis sp novrdquo Extremophiles vol 6 no 6 pp 479ndash4832002

Archaea 17

[57] M F Mullakhanbhai and H Larsen ldquoHalobacterium volcaniispec nov a Dead Sea halobacterium with a moderate saltrequirementrdquo Archives of Microbiology vol 104 no 1 pp 207ndash214 1975

[58] S D Nuttall and M L Dyall-Smith ldquoCh2 a novel halophilicarchaeon from an Australian solar salternrdquo International Jour-nal of Systematic Bacteriology vol 43 no 4 pp 729ndash734 1993

[59] P D Franzman E Stackebrandt K Sanderson et al ldquoHalobac-terium lacusprofundi sp nov a halophilic bacterium isolatedfrom Deep Lake Antarcticardquo Systematic and Applied Microbi-ology vol 11 pp 20ndash27 1988

[60] G A Tomlinson and L I Hochstein ldquoHalobacterium saccharo-vorum sp nov a carbohydrate metabolizing extremely halo-philic bacteriumrdquoCanadian Journal ofMicrobiology vol 22 no4 pp 587ndash591 1976

[61] A Ventosa M C Gutierrez M Kamekura and M L Dyall-Smith ldquoProposal to transfer Halococcus turkmenicus Halobac-terium trapanicum JCM9743 and strainGSL-11 toHaloterrigenaturkmenica gen nov comb novrdquo International Journal ofSystematic Bacteriology vol 49 no 1 pp 131ndash136 1999

[62] M Kamekura and M L Dyall-Smith ldquoTaxonomy of the familyHalobacteriaceae and the description of two new genera Halo-rubrobacterium and Natrialbardquo Journal of General and AppliedMicrobiology vol 41 no 4 pp 333ndash350 1995

[63] T J Carver KM RutherfordM BerrimanM A RajandreamB G Barrell and J Parkhill ldquoACT the Artemis comparisontoolrdquo Bioinformatics vol 21 no 16 pp 3422ndash3423 2005

[64] S W Cline W L Lam R L Charlebois L C Schalkwyk andW F Doolittle ldquoTransformationmethods for halophilic archae-bacteriardquo Canadian Journal of Microbiology vol 35 no 1 pp148ndash152 1989

8 Archaea

Table4ORF

anno

tatio

nsof

theh

aloviru

sPH1g

enom

e

ORF

aPo

sitionb

Locustag

Leng

thc(aa)

pKi

Similaritycharacteris

tics

144

1ndash605

HhP

H1gp01

5043

76aa

similarityto

SH1O

RF2(YP271859)

2602ndash802

HhP

H1gp02

66109

80aa

similarityto

SH1O

RF3(YP271860)

3802ndash1059

HhP

H1gp03

8539

85aa

similarityto

SH1O

RF4(YP271861)44

sim

ilarityto

HHIV-2

protein1

41056ndash1283

HhP

H1gp04

7539

85aa

identityto

SH1O

RF5(YP271862)Predictedcoiled-coilregion

51276ndash1686

HhP

H1gp05

136

49

80aa

similarityto

SH1O

RF6(YP271863)Tw

otransm

embraned

omainsA

lsorelated

toHHIV-2

protein2(38

AFD

02283)

andto

Halobiform

alacisalsiORF

ZP09950018

(35

)6

1683ndash1829

HhP

H1gp06

4861

81aa

similarityto

SH1O

RF7(YP271864

)Predictedsig

nalsequence(Sign

alP)

71822ndash2130

HhP

H1gp07

102

58

91aa

similarityto

SH1O

RF8(YP271865)a

nd86to

HHIV-2

putativ

eprotein

3(A

FD02284)O

ther

homologsinclude

Nocardiaproteinpn

f2110(YP1220601)H

rrlacusprofund

iHlac0751

(YP0025654211)ORF

sinactin

ophage

VWB

(AAR2

9707)Frankia(EAN116

57)andgp40

ofMycobacteriu

mph

ageD

ori(AER

47690)

82123ndash2383

HhP

H1gp08

8642

90aa

similarityto

SH1O

RF9(YP271866)

92380ndash2714

HhP

H1gp09

111

52

88aa

similarityto

SH1O

RF11(YP271868)and

also

similarityto

putativ

eprotein

4of

HHIV-2

(AFD

02285)

HlacA

J19877of

Hbflacisa

lsiAJ

5(ZP09950016)

102861ndash300

4HhP

H1gp10

47104

Weaksim

ilarityto

SH1O

RF12

(YP271869)o

nlyover

theN

-term

inal18

resid

ues

113029ndash3100

HhP

H1gp11

23123

Not

presentinSH

1orH

HIV-2

123134ndash7453

HhP

H1gp12

1439

42

Capsid

protein

VP172sim

ilarityto

SH1V

P1(O

RF13Y

P271870)HHIV-2

VP1

(AFD

02286)H

lacA

J19872Hbflacisa

lsiAJ

5Predictedhelix-tu

rn-helixandRu

vA-like

domain(In

terproScan)

137505ndash8227

HhP

H1gp13

240

54

Predicted

P-loop

ATPa

sedo

main(C

OG0433)92aa

similarityto

SH1O

RF17

(YP271874)a

ndalso

similarityto

putativ

eAT

Pase

ofHHIV-2

(AFD

02288)A

TPaseo

fHaladapatus

paucihalophilusD

X253

(ZP0804

6180)HlacA

J19867of

Hbf

lacisalsi(ZP09950014)

148419ndash8228c

HhP

H1gp14

6351

43aa

similarityto

SH1O

RF18

(YP271875)Alanine-richCentraltransm

embraned

omain(Pho

bius)

158856ndash8

416c

HhP

H1gp15

146

41

92aa

similarityto

SH1O

RF19

(YP271876)a

ndalso

toHHIV-2

putativ

eprotein

9(A

FD02290)H

lacA

J19857of

Hbf

lacisalsiAJ

5ZO

D2009

19093of

Happau

cihalophilus

168853ndash9

500c

HhP

H1gp16

215

39

65aa

similarityto

SH1O

RF20

(YP271877)a

ndalso

topu

tativ

eprotein

11of

HHIV-2

(AFD

02292)Z

OD2009

19098of

Hap

paucihalophilusHlacA

J19852Hbflacisa

lsiAJ

5C-

term

inaltransm

embraned

omain(Pho

bius)

179500ndash9

919c

HhP

H1gp17

139

41

79aa

similarityto

SH1O

RF21

(YP271878)a

ndalso

topu

tativ

eprotein

12of

HHIV-2

(AFD

02293)hypotheticalprotein

HlacA

J19847of

Hbflacisa

lsi(ZP09950010)Transm

embraned

omain(Pho

bius)

189923ndash10594c

HhP

H1gp18

223

43

83aa

similarityto

SH1O

RF22

(YP271879)a

ndalso

topu

tativ

eprotein

13of

HHIV-2

(AFD

02294)Z

OD2009

19108of

Hap

paucihalophilusHlacA

J19842of

Hbflacisa

lsiP

redicted

signalsequence(sig

nalP)Con

tains4

CxxC

motifs

1910659ndash10943

HhP

H1gp19

94104

Capsid

protein

VP12

(ORF

19)91aa

similarityto

SH1V

P12(O

RF23Y

P271880)a

ndalso

toVP12of

HHIV-2

(AFD

02295)

andHlacA

J19837of

Hbflacisa

lsiZ

OD2009

19113of

Happau

cihalophilusTw

otransm

embraned

omains

(Pho

bius)

2010960ndash

11517

HhP

H1gp20

185

44

Capsid

protein

VP7(O

RF20)98aa

similarityto

SH1V

P7(O

RF24Y

P271881)a

ndalso

toVP7

ofHHIV-2

(AFD

02296)

ZOD2009

19118

ofHappau

cihalophilusHlacA

J19832of

Hbflacisa

lsiAJ

5

2111519ndash

12217

HhP

H1gp21

232

41

Capsid

protein

VP4(O

RF21)94aa

similarityto

SH1O

RF25

(YP271882)a

ndalso

toVP4

ofHHIV-2

(AFD

02297)

ZOD2009

19123of

Happau

cihalophilusHlacA

J19827of

Hbflacisa

lsiAJ

522

12233ndash12454

HhP

H1gp22

7339

81aa

similarityto

SH1O

RF26

(YP271883)a

ndalso

topu

tativ

eprotein

17of

HHIV-2

(AFD

02298)

2312458ndash12697

HhP

H1gp23

7948

78aa

similarityto

SH1capsid

protein

VP13

(ORF

27Y

P271884)a

ndalso

toVP13of

HHIV-2

(AFD

02299)P

redicted

coil-coildo

main

PredictedC-

term

inaltransm

embraned

omain(Pho

bius)

Archaea 9

Table4Con

tinued

ORF

aPo

sitionb

Locustag

Leng

thc(aa)

pKi

Similaritycharacteris

tics

2412701ndash1506

4HhP

H1gp24

787

39

Capsid

protein

VP2(O

RF24)75aa

similarityto

SH1V

P2(O

RF28Y

P271885)a

ndto

VP2

ofHHIV-2

(AFD

02300)

ZOD2009

19133of

Happau

cihalophilus(ZP

0804

6189)

2515065ndash15961

HhP

H1gp25

298

45

Putativec

apsid

protein

VP5(O

RF25)81aa

similarityto

SH1V

P5(O

RF29Y

P271886)a

ndto

HHIV-2

VP5

(AFD

02301)

HlacA

J19807of

HbflacisalsiAJ

5ZO

D2009

19133of

Happau

cihalophilus

261596

4ndash1644

6HhP

H1gp26

160

46

73aa

similarityto

SH1capsid

protein

VP10

(ORF

30Y

P271887)a

ndto

VP10of

HHIV-2

(AFD

02302)Z

OD2009

19138of

Happau

cihalophilusHlacA

J19802of

Hbflacisa

lsiP

redicted

C-term

inaltransm

embraned

omain(TMHMM)

271644

6ndash16895

HhP

H1gp27

149

42

Capsid

protein

VP9(O

RF27)94aa

similarityto

SH1V

P9(O

RF31Y

P271888)a

ndto

ZOD2009

19143of

Hap

paucihalophilus(ZP

0804

6191)

2816908ndash17921

HhP

H1gp28

337

43

Capsid

protein

VP3(O

RF28)91aa

similarityto

SH1V

P3(O

RF32Y

P271889)a

ndto

NJ7G

2365

ofNa

trinemaspJ7-2

2917928ndash18617

HhP

H1gp29

229

42

Capsidprotein

VP6(O

RF29)83aa

similarityto

SH1V

P6(O

RF33Y

P271890)Also

toNJ7G

3194

ofNa

trinemaspJ7-2and

Hmuk

0476

ofHalom

icrobium

mukohataeiPredictedcarboxypeptid

aser

egulatory-lik

edom

ain(C

arbo

xypepD

reg

pfam

13620)

3018614ndash

18943

HhP

H1gp30

109

49

71aa

similarityto

SH1O

RF34

(YP271891)a

ndto

putativ

eprotein

27of

HHIV-2

(AFD

02308)P

redicted

signalsequence

(signalP)

3119054ndash

19176c

HhP

H1gp31

4050

TwoCx

xCmotifsN

oho

molog

inSH

1orH

HIV-2

3219173ndash19289c

HhP

H1gp32

3841

68aa

similarityto

SH1O

RF37

(YP271894)

3319286ndash

19552c

HhP

H1gp33

8853

58aa

similarityto

SH1O

RF39

(YP271896)a

ndto

HHIV-2

putativ

eprotein

29(A

FD02310)Fou

rCxxCmotifs

3419549ndash

19731c

HhP

H1gp34

6070

Con

tainsC

xxCmotif

Noho

molog

inSH

1orH

HIV-2

3519728ndash20219c

HhP

H1gp35

163

56

93aa

similarityto

SH1O

RF41

(YP271898)a

ndto

putativ

eprotein

30of

HHIV-2

(AFD

02311)

3620216ndash

21415c

HhP

H1gp36

399

50

95aa

similarityto

SH1O

RF42

(YP271899)a

ndto

putativ

eprotein

31of

HHIV-2

(AFD

02312)

3721419ndash

21778c

HhP

H1gp37

11941

91aa

similarityto

SH1O

RF43

(YP271900)a

ndto

putativ

eprotein

32of

HHIV-2

(AFD

02313)

3821762ndash2300

6cHhP

H1gp38

414

42

83aa

similarityto

SH1O

RF44

(YP271901)a

ndto

putativ

eprotein

33of

HHIV-2

(AFD

02314)P

redicted

coil-coiland

helix-tu

rn-helixdo

mains

3923003ndash23158c

HhP

H1gp39

5165

69aa

similarityto

SH1O

RF45

(YP271902)

4023161ndash23370c

HhP

H1gp40

6935

74aa

similarityto

SH1O

RF46

(YP271903)and55sim

ilarityto

putativ

eprotein

35of

HHIV-2

(AFD

02316)

4123422ndash23586c

HhP

H1gp41

5476

86aa

similarityto

SH1O

RF47

(YP271904)Con

tains2

CxxC

motifsand

show

ssim

ilarityto

proteindo

mainfamily

PF14206Arginine-ric

h

4223583ndash24023c

HhP

H1gp42

48

77aa

similarityto

SH1O

RF48

(YP271905)a

ndto

putativ

eprotein

36of

HHIV-2

(AFD

02317)H

ham114540of

Hcc

hamelinensis(ZP11271999)Ph

iCh1p72of

Natrialba

phageP

hiCh

1(NP665989)DUF4

326(pfam14216)

family

domain

4324020ndash

24538c

HhP

H1gp43

172

47

79aa

similarityto

SH1O

RF49

(YP271906)and56sim

ilarityto

putativ

eprotein

37of

HHIV-2

(AFD

02318)

4424535ndash25053c

HhP

H1gp44

172

43

92aa

similarityto

SH1O

RF50

(YP271907)and72sim

ilarityto

putativ

eprotein

38of

HHIV-2

(AFD

02319)

4525050ndash

25277c

HhP

H1gp45

7543

Noho

molog

inSH

1orH

HIV-2

4625274ndash

25387c

HhP

H1gp46

3748

Noho

molog

inSH

1orH

HIV-2

4725533ndash25904

HhP

H1gp47

123

43

61aa

similarityto

SH1O

RF51

(YP271908)66sim

ilarityto

putativ

eprotein

39of

HHIV-2

(AFD

02320)P

redicted

COG1342

domain(D

NAbind

inghelix-tu

rn-helix)

4825901ndash26092c

HhP

H1gp48

6349

Con

tainsC

xxCmotif

Noho

molog

inSH

1orH

HIV-2

4926089ndash

2764

8cHhP

H1gp49

506

40

81aa

similarityto

SH1O

RF55

(YP271912)(bu

tonlyin

theN

-term

inalhalf)H

omolog

ofvirusstru

cturalprotein

VP18

ofHHIV-2

(AFD

02323)

a ORF

swerep

redicted

either

byGLIMMER

orby

manualsearching

forh

omologsintheG

enBa

nkdatabase

b Startandendpo

sitions

ofORF

sare

give

inbp

numbera

ccording

totheP

H1sequenced

epositedatGenBa

nk(KC2

52997)O

RFso

nthec

omplem

entary

strandared

enoted

bythes

uffixc

c Lengthof

thep

redicted

ORF

innu

mbero

faminoacids

10 Archaea

Table 5 Main differences (divergent regions) between the SH1 and PH1 genomes

RegionaSH1 startb SH1 stop Length (bp) PH1 start PH1 stop Length (bp) Comment

DV1 527 650 123 538 604 66

Replacement Both regions have direct repeats at their bordersPH1 has two sets GACCCGGC and CGCTGC while SH1 hasCCCGAC In SH1 this replacement region includes theC-terminal region of ORF2 and the N-terminal region ofORF3 In PH1 it covers only the C-terminal region of ORF1

DV2 2433 2588 155 2386 2387 mdash RMDc from PH1 SH1 repeat at border is TGACCGThisremoves the homolog of SH1 ORF10 from PH1

DV3 3324 3598 274 3137 3416 279 Replacement at the beginning of SH1 ORF13PH1 ORF12(capsid protein VP1)

DV4 6656 7128 472 6478 7085 607 Replacement near the C-terminus of SH1 ORF13PH1 ORF12(capsid protein VP1)

DV5 7491 8341 850 7446 7447 mdash

Indel that results in ORFs14-16 of SH1 not being present inPH1 SH1 ORF14 has a close homolog in HHIV-2 (ORF6) in asimilar position just after the VP1 homolog SH1 ORF15 is aconserved protein in haloarchaea (eg NP 2552A andHmuk 2978) and halovirus His1 (ORF13) Possible invertedrepeat (AGCCATG) at border of SH1 region

DV6 13705 13706 mdash 12818 12851 33RMD from SH1 PH1 repeat at border is CAGCGG(gt)G Thisremoves a part of capsid protein VP2 sequence from the SH1protein

DV7 13789 13863 74 12934 12941 7 Replacement This occurs within VP2 gene of both viruses

DV8 15142 15186 44 14220 14221 mdash RMD from PH1 SH1 repeat at border is TG(tc)CCGACGAand occurs within capsid protein VP2 gene of both viruses

DV9 15303 15317 14 14337 14338 mdash RMD from PH1 SH1 repeat at border is GCCGACGA andoccurs within capsid protein VP2 gene of both viruses

DV10 15504 15527 23 14529 14530 mdash RMD from PH1 SH1 repeat at border is GACGA and occurswithin capsid protein VP2 of both viruses

DV11 20026 20405 379 19019 19173 154Replacement beginning at the start of SH1 ORF35PH1 ORF31This region is longer in SH1 where it includes ORF36 an ORFthat has no homolog present in PH1

DV12 20440 20661 221 19208 19286 78 Replacement This is unequal and includes an ORF in SH1(ORF38) that is not present in PH1

DV13 20943 21085 142 19562 19728 166 Replacement This covers SH1 ORF40PH1 ORF34 Thepredicted proteins are not homologous

DV14 23541 23542 mdash 22171 22197 27 Probable RMD in SH1 PH1 repeat at border is CGTCTCGGand occurs in SH1 ORF44PH1 ORF38

DV15 24506 24521 15 23152 23153 mdash Probable RMD in PH1 SH1 repeat at border is CTCGGT andoccurs near the end of SH1 ORF45PH1 ORF39

DV16 24951 24964 13 23579 23580 mdash Probable RMD in PH1 SH1 repeat at border is CGGTC andoccurs in SH1 ORF47PH1 ORF41

DV17 26449 26450 mdash 25050 25274 224Probable RMD in SH1 PH1 repeat at border is TCATGCG andoccurs near the start of SH1 ORF50PH1 ORF44 and providesan extra ORF for PH1 (ORF45)

DV18 27074 29764 690 25895 26933 1038

Replacement Left border at start of SH1 ORF51PH1 ORF47and extends rightwards into SH1 ORF55PH1 ORF49 (VP18gene) It is an unequal replacement and SH1 has two moreORFs in this region than PH1

aDV Divergent regions between the genomes of SH1 and PH1bStart and stop positions refer to the GenBank sequences of the two viruses SH1 NC 0072171 PH1 KC252997cRMD repeat-mediated deletion event as described in [20]

The recently described halovirus SNJ1 carries manyORFsthat have homologs adjacent to a previously described ViPREofHmc mukohataei (from here on denoted by ViPREHmuk1)a locus that contains genes related to His2 (and other ple-olipoviruses) as well as toHaloferax plasmid pHK2 (probablya provirus) and to two small plasmids ofHqr walsbyi (sim6 kbpL6A and pL6B)The latter relationship provides wider linksto other viruses because one of the PL6 genes (Hqrw 6002)

has homologs in some pleolipoviruses (HPRV-3 and HGPV-1) while another gene (Hqrw 6005) is related to ORF16 ofthe spindle-shaped halovirus His1 [15] After adding the SNJ1gene homologs to ViPREHmuk1 it is extended significantlyand a close inspection of the flanking regions revealed atRNA-ala gene at one end and a partial copy of the sametRNA-ala gene (next to a phage integrase gene) at the otherend (supplementary Figure 1) Genes beyond these points

Archaea 11

Table 6 Transfection of haloarchaea by PH1 DNA

Speciesa Efficiency of transfectiontransformation withPH1 DNAb (PFU120583g of DNA) pUBP2 DNAbc (CFU120583g of DNA)

Har hispanica 53 plusmn 05 times 103 14 plusmn 08 times 104

Har marismortui 45 plusmn 07 times 103 18 plusmn 01 times 104

ldquoHar sinaiiensisrdquo 32 plusmn 06 times 102 mdashHalobacterium salinarum mdash 54 plusmn 42 times 103

Haloferax gibbonsii 28 plusmn 08 times 103 40 plusmn 26 times 103

Haloferax lucentense mdash 57 plusmn 11 times 103

Hfx volcanii mdash 21 plusmn 07 times 105

Hrr lacusprofundi 70 plusmn 22 times 102 95 plusmn 15 times 103

Haloterrigena turkmenica 16 plusmn 17 times 102 24 plusmn 06 times 103

Natrialba asiatica 35 plusmn 39 times 102 31 plusmn 10 times 104aOnly those species positive for transfection andor plasmid transformation are shownbRates of transfection by (nonprotease treated) PH1 DNA or transformation by plasmid pUB2 are averages of three independent experiments each performedin duplicate (plusmn standard deviation)cTransformants were selected on plates with 2 4 or 6120583gmL simvastatin depending on the strainmdash no plaques or colonies observed

appear to be related to cellular metabolism ViPREHmuk1is now 39379 nt in length flanked by a 59 nt direct repeat(potential att sites) and includes 54 ORFs many of which arerelated to known haloviruses are virus-like (eg integrasesmethyltransferases transcriptional regulators DNA methyl-transferase and DNA glycosylase) are homologs in or nearother knownViPREs or show little similarity to other knownproteins This locus also includes an ORC1CDC6 homolog(Hmuk 0446) which could provide a replication function

The mixture of very different virus and plasmid genesseen within ViPREHmuk1 is remarkable but it also revealshomologs found in or adjacent to genomic loci of otherspecies such as the gene homologs of Har marismortuiseen in the left end of ViPREHmuk1 (supplementary Figure1) When these genes are added to the previously describedHaloarcula locus (ViPREHmuk1) it is also extended signifi-cantly It becomes 25550 bp in length encompasses genesfrom Hmar 2382 to Hmar 2404 and is flanked by a fulltRNA-ala gene at one end and a partial copy at the other Itcarries an ORC1CDC6 homolog an integrase and a phiH-like repressor as well as pleolipovirus homologs

210 Transfection of Haloarchaea by PH1 DNA PH1 DNAwas introduced into Har hispanica cells using the PEGmethod [37] and the cells were screened for virus productionby plaque assay Figure 7 shows the increase in transfectedcells with input (nonproteinase K-treated) virus DNA Theestimated efficiencywas 53times 103 PFUper120583gDNA PH1DNAthat had been treated with proteinase K or with DNAase I(RNase-free) did not produce plaques (data not shown) PH1DNA was found to transfect six species of haloarchaea otherthanHar hispanica includingmembers of the generaHaloar-cula Haloferax Halorubrum Haloterrigena and Natrialba(Table 6) For these experiments the virus production fromtransfected cells was detected by plaque assay on indicatorlawns of Har hispanica

0

0

200 400 600 800 1000

Titre

(PFU

mL)

104

103

102

PH1 DNA (ng)

Figure 7 Transfection ofHar hispanica cells with PH1 DNA Vary-ing amounts of nonproteinase K-treated PH1 DNAwere introducedinto cells of Har hispanica using the PEG method [64] Cells werethen screened for infective centres by plaque assay Data shown is theaverage of three independent experiments performed in duplicateError bars represent one standard deviation of themean If protease-treated DNA was used no transfectant plaques were observed

3 Discussion

PH1 is very similar in particlemorphology genome structureand sequence to the previously described halovirus SH1 [816] Like SH1 it has long inverted terminal repeat sequencesand terminal proteins indicative of protein-primed replica-tion Purified PH1 particles have a relatively low buoyant

12 Archaea

density and are chloroform sensitive and show a layeredcapsid structure consistent with the likely presence of aninternalmembrane layer as has been shown for SH1HHIV-2and SNJ1 [13 26]The particle stability of PH1 to temperaturepH and reduced salt was similar to SH1 [8] The structuralproteins of PH1 are very similar in sequence and relativeabundance to those of SH1 so the two viruses are likely toshare the same particle geometry [16]

Viruses that use protein-primed replication such as bac-teriophageΦ29 carry a viral type BDNApolymerase that caninteract specifically with the proteins attached to the genomictermini and initiate strand synthesis [38] ArchaeovirusesHis1 [7] and Acidianus bottle-shaped virus [39] probablyuse this mode of replication However a polymerase genecannot be found in the genomes of PH1 SH1 or HHIV-2DNA polymerases are large enzymes and the only ORF ofSH1 that was long enough to encode such an enzyme andhad not previously been assigned as a structural protein wasORF55 which specified an 865 aa protein that contained noconserved domains indicative of polymerases [27]The recentstudy of HHIV-2 showed that the corresponding ORF in thisvirus (gene 42) is a structural protein of the virion and byinference this is likely also for the corresponding proteinsof SH1 and PH1 (no homolog is present in SNJ1) Withouta polymerase gene these viruses must use a host enzymebut searches for a viral type B DNA polymerase in thegenome of the hostHar hispanica or in the genomes of othersequenced haloarchaea did not find anymatchesThis arguesstrongly for a replication mechanism that is different to thatexemplified byΦ29 An attractive alternative is that displayedby Streptomyces linear plasmids which have 51015840-terminal pro-teins and use a cellular polymerase for replication [40] Theterminal proteins are not used for primary replication but forend patching [41] and it has been shown that if the term-inal repeat sequences are removed from these plasmids andthe ends ligated they can replicate as circular plasmids [42]This provides a testable hypothesis for the replication of halo-sphaeroviruses and also offers a pathway for switchingbetween linear (eg PH1) and circular (eg SNJ1) forms Useof a host polymerase would also fit with the ability of SH1 andPH1 DNA to transfect many different haloarchaeal species[31] as these enzymes and theirmode of action are highly con-served

The growth characteristics of PH1 inHar hispanica bothby single-cell burst and single-step growth experiments gavevalues of 50ndash100 virusescell significantly less than that ofSH1 (200 virusescell) [8] and HHIV-2 (180 virusescell) [13]The latter virus is lytic whereas SH1 PH1 and SNJ1 start pro-ducing extracellular viruswell before any decrease in cell den-sity indicating that virus release can occur without lysisThishas been most clearly shown with SH1 where almost 100of cells can be infected [8] Infection ofHar hispanica by PH1was less efficient (sim30 of cells) but the kinetics of virus pro-duction in single-step growth curves of SH1 and PH1 aresimilar with virus production occurring over several hoursrather than at a clearly defined time after infection The twoviruses are also very closely related and infect the same hostspecies so it is likely that they use the same scheme for cellexit For both viruses the cell density eventually decreases in

single-step cultures showing that virus infection does resultin cell death This mode of exit appears to maximise theproduction of the virus over time as the host cells survive foran extended period In natural hypersaline waters cell num-bers are usually high but growth rates are low [12] and thehigh incident UV (on often shallow ponds) would damagevirus DNA so reducing the half-life of released virus part-icles [43] These factors may have favoured the exit strategydisplayed by these viruses improving their chance of trans-mission to a new host

The motif GATC was absent in the genomes of PH1 SH1andHHIV-2 and themotif CTAGwas either absent or greatlyunderrepresented All three viruses share the same host butGATC and CTAG motifs are plentiful in the Har hispanicagenome (accessions CP002921 and CP002923) Avoidance ofthesemotifs in halovirusesHis1 His2 HF1 andHF2 has beenreported previously [7] Such purifying selection commonlyresults from host restriction enzymes and inHfx volcanii itis exactly these two motifs that are targeted [44] One Hfxvolcanii enzyme recognises A-methylated GATC sites [45]and another recognises un-methylated CTAG sites (whichare protected by methylation in the genome) In the currentstudy this could explain the negative transfection results forHfx volcanii as the PH1 genome contains two CTAGmotifsBy comparison the SH1 genome contains no CTAG motifsand is able to transfect Hfx volcanii [31]

Comparison of the PH1 and SH1 genomes allowed adetailed picture of the natural variation occurring betweenclosely related viruses revealing likely deletion events medi-ated by small repeats a process described previously in astudy of Hqr walsbyi and termed repeat-mediated deletion[20] There are also many replacements including blocks ofsequence that occur within long open reading frames (suchas the capsid protein genes VP1 and VP18) VP3 and VP6 areknown to form the large spikes on the external surface of theSH1 virion and presumably one or both interact with host cellreceptors [16] PH1 and SH1 have identical host ranges con-sistent with the high sequence similarity shown between theircorresponding VP3 and VP6 proteins

A previous study of SH1 could not detect transcriptsacross annotated ORFs 1ndash3 or 55-56 [30] ORFs 1 and 56occur in the terminal inverted repeat sequence and in thepresent study it was found that there were no ORFs cor-responding to these in the PH1 genome Given the close simi-larity of the two genomes the comparative data are in agree-ment with the transcriptional data indicating that these SH1ORFs are not used SH1 ORFs 2 and 3 are more problematicas good homologs of these are also present in PH1 The con-flict between the transcriptional data and the comparativegenomic evidence requires further experimental work toresolveThe status of SH1ORF55 has recently been confirmedby studies of the related virus HHIV-2 (discussed above)

The pleolipovirus group has expanded dramatically inthe last few years and it now comprises a diverse group ofviruseswith different genome types and replication strategiesWhat is evenmore remarkable is that a similar expansion cannow be seen with SH1-related viruses which includes viruseswith at least two replication modes and genome types (linearand circular) Evidence from haloarchaeal genome sequences

Archaea 13

show cases of not only provirus integrants or plasmids (egpKH2 and pHH205) but also genomic loci (ViPREs) thatcontain cassettes of virus genes from different sources andat least two cases (described in this study) have likely attsites that indicate circularisation and mobility While thereis a clear relationship between viruses with circular ds- andssDNA genomes that replicate via the rolling circle method(ie the ssDNA form is a replication intermediate) it ismore difficult to explain how capsid gene cassettes can movebetween these viruses and those with linear genomes andterminal proteins Within the pleolipoviruses His2 has alinear genome that contains a viral type B DNA polymeraseand terminal proteins while all the other described membershave circular genomes that contain a rep homolog and prob-ably replicate via the rolling circle method [14 19] Similarlyamong the halosphaeroviruses PH1 SH1 and HHIV-2 havelinear dsDNA with terminal proteins and probably replicatein the same way but the related SNJ1 virus has a circulardsDNA genome The clear relationships shown by the capsidgenes of viruses within each group plus the connectionsshown to haloviruses outside each group (eg the DNA poly-merases of His1 and His2) all speak of a vigorous meansof recombination one that can readily switch capsid genesbetween viruses with radically different replication strategiesHow is the process most likely to operate One possibility issuggested by ViPREs which appear to be mobile collectionsof capsid and replication genes from different sources Theyoffer fixed locations for recombination to occur providegene cassettes that can be reassorted to produce novel virusgenomes and some can probably recombine out as circularforms For example ViPREHmuk1 carries genes related topleolipoviruses halosphaeroviruses and other haloviruses Itis yet to be determined if any of the genes carried in theseloci are expressed in the cell (such as the genes for capsidproteins) or if ViPREs provide any selective advantage to thehost

4 Materials and Methods

41 Water Sample A water sample was collected in 1998from Pink Lake (32∘001015840 S and 115∘301015840 E) a hypersalinelake on Rottnest Island Western Australia Australia It wasscreened in the same year for haloviruses using the methodsdescribed in [8] A single plaque on a Har hispanica lawnplate was picked and replaque purified The novel haloviruswas designated PH1

42 Media Strains and Plasmids The media used in thisstudy are described in the online resource The Halohand-book (httpwwwhaloarchaeacomresourceshalohandbookindexhtml) Artificial salt water containing 30 (wv) totalsalts comprised of 4M NaCl 150mM MgCl

2 150mM

MgSO4 90mM KCl and 35mM CaCl

2and adjusted to pH

75 using sim2mL 1M Tris-HCl (pH 75) per litre MGM con-taining 12 18 or 23 (wv) total salts and HVD (halo-virus diluent) were prepared from the concentrated stock aspreviously described [46] Bacto-agar (Difco Laboratories)was added to MGM for solid (15 gL) or top-layer (7 gL)media

Table 1 lists the haloarchaea used in this study Allhaloarchaea were grown aerobically at 37∘C in either 18 or23 (wv)MGM(depending on the strain) andwith agitation(except for ldquoHar sinaiiensisrdquo)The plasmids used were pBlue-script II KS+ (Stratagene Cloning Systems) pUBP2 [47]and pWL102 [48] pUBP2 and pWL102 were first passagedthrough E coli JM110 [49] to prevent dam-methylationof DNA which has been shown to reduce transformationefficiency in some strains [45]

43 Negative-Stain Electron Microscopy The method fornegative-stain TEM was adapted from that of V Tarasovdescribed in the online resource The Halohandbook (httpwwwhaloarchaeacomresourceshalohandbookindexhtml)A 20120583L drop of the sample was placed on a clean surfaceand the virus particles were allowed to adsorb to a Formvarfilm 400-mesh copper grid (ProSciTech) for 15ndash2min Theywere then negatively stained with a 20120583L drop of 2 (wv)uranyl acetate for 1-2min Excess liquid was absorbed withfilter paper and the grid was allowed to air dry Grids wereexamined either on a Philips CM 120 BioTwin transmissionelectron microscope (Royal Philips Electronics) operating atan accelerating voltage of 120 kV or on a Siemens Elmiskop102 transmission electron microscope (Siemens AG) operat-ing at an accelerating voltage of 120 kV

44 Virus Host Range Twelve haloarchaeal strains from thegenera Haloarcula Halobacterium Haloferax HalorubrumandNatrialba (Table 1)Thirteen naturalHalorubrum isolates(H Camakaris unpublished data) and five uncharacterizedhaloarchaeal isolates fromLakeHardy Pink Lake (inWesternAustralia) and Serpentine Lake (D Walker and M K Seahunpublished data) were screened for PH1 susceptibilityLysates from PH1-infected Har hispanica cultures (1 times1011 PFUmL) were spotted onto lawns of each strain (usingmedia with 12 or 18 (wv)MGM depending on the strain)and incubated for 2ndash5 days at 30 and at 37∘C

45 Large Scale Virus Growth and Purification Liquid cul-tures of PH1 were grown by the infection (MOI 005) of anearly exponential Har hispanica culture in 18 (wv) MGMCultures were incubated aerobically at 37∘C with agitationfor 3 days Clearing (ie complete cell lysis) of PH1-infectedcultures did not occur and consequently cultures were harv-ested when the absorbance at 550 nm reached the minimumand the titre (determined by plaque assay) was at the maxi-mum usually sim1010ndash1011 PFUmL Virus was purified usingthe method described previously for SH1 [8 31] except thatafter the initial low speed spin (Sorvall GSA 6000 rpm30min 10∘C) virus was concentrated from the infected cul-ture by centrifugation at 26000 rpm (13 hr 10∘C) onto acushion of 30 (wv) sucrose in HVDThe pellet was resus-pended in a small volume of HVD and loaded onto a pre-formed linear 5ndash70 (wv) sucrose gradient followed byisopycnic centrifugation in 13 gmL CsCl (Beckman 70Ti60000 rpm 20 hr 10∘C) The white virus band occurredat a density of 129 gmL and was collected and diluted inhalovirus diluent (HVD see Section 4) and the virus pelleted(Beckman SW55 35000 rpm 75min 10∘C) and resuspended

14 Archaea

in a small volume of HVD and stored at 4∘C Virus recoveryat the major stages of a typical purification is given insupplementary Table 1 The specific infectivity of pure virussolutions was determined as the ratio of the PFUmL to theabsorbance at 260 nm

46 PH1 Single-Step Growth Curve An early exponentialphase culture of Har hispanica grown in 18 (wv) MGMwas infected with PH1 (MOI 50) Under these conditions thepercentage of infected cells was approximately 30 After anadsorption period of 1 hr at 37∘C the cells were washed threetimes with 18 (wv) MGM (at room temperature) resus-pended in 100mL 18 (wv) MGM (these methods ensuredthe removal of all residual virus) and incubated at 37∘Cwith shaking (100 rpm) Samples were removed at hourlyintervals for measurements of absorbance at 550 nm andthe number of infective centres Immediately after samplingtitres were determined by plaque assay with an indicator lawnof Har hispanica Each experiment was performed in tripli-cate

47 Halovirus Stability After various treatments sampleswere removed and diluted in HVD and virus titres weredetermined by plaque assay on Har hispanica Each experi-ment was performed in triplicate and representative data areshown Chloroform sensitivity was examined by the exposureof PH1 lysates in 18 (wv) MGM to chloroform in a volumeratio of 1 4 (chloroform to lysate) Incubation was at roomtemperature with constant agitation At appropriate timepoints the mix was allowed to settle and a sample wasremoved from the upper layer and diluted in HVDThe effectof a low ionic environment was examined by diluting PH1lysates (in 18 (wv) MGM) into double distilled H

2O in a

volume ratio of 1 1000 (lysate to double distilledH2O) Incu-

bation was at room temperature with constant agitationSamples were removed at various times and diluted in HVDThe pH stability of PH1 was determined by dilution of PH1lysates in 18 (wv)MGMin the appropriate pHbuffer (HVDbuffered with appropriate Tris-HCl) in a volume ratio of1 100 (lysate to buffer) Incubation was at room temperaturewith constant agitation After 30min samples were removedand diluted in HVD Thermal stability of PH1 was examinedby a 1 hr incubation of a virus lysate in 18 (wv) MGM atdifferent temperatures after which they were brought quicklyto room temperature diluted in HVD and titrated

48 Protein Procedures To remove salts purified virus prepa-rations were mixed with trichloroacetic acid (10 (vv) finalconcentration) and incubated on ice for 15min to allowthe proteins to precipitate After centrifugation (16000 g15min room temperature) the precipitate was washed threetimes in acetone dried and resuspended in double distilledH2O Proteins were dissolved in Laemmli sample buffer with

48mM 120573-mercaptoethanol [50] heated in boiling water for5min then separated on 12 (wv) NuPAGE Novex Bis-Tris Gels using MES-SDS running buffer according to themanufacturerrsquos directions (Invitrogen) After electrophoresisgels were rinsed in double distilled H

2O and stained with

01 (wv) Brilliant Blue G in 40 (vv) methanol and

10 (vv) acetic acid Gels were destained with severalchanges of 40 (vv) methanol and 10 (vv) acetic acidAlternatively gels were stained with GelCode GlycoproteinStaining Kit Pro-Q Emerald 488 Glycoprotein Gel and BlotStain Kit SYPRO Ruby Protein Gel Stain according tothe manufacturerrsquos directions (Invitrogen Molecular ProbesPierce Biotechnology)

Protein bands were cut from the gels and sent to theAustralian Proteome Analysis Facility (Macquarie Univer-sity) for trypsin digestion and analysis by matrix assistedlaser desorption ionisation time of flight mass spectrometry(MALDI-TOF MS) on an Applied Biosystems 4700 Pro-teomics Analyser (Applied Biosystems)

49 DNA Procedures Proteinase K-treated and nonpro-teinaseK-treatedDNApreparationsweremade frompurifiedPH1 using the methods described previously for SH1 [8]DNA was separated on 1 (wv) agarose gels in Tris-acetate-EDTA electrophoresis buffer and was stained with ethidiumbromide (Sigma-Aldrich)120582 DNA DNase I (RNase-free) exonuclease III mung

bean nuclease nuclease BAL-31 T4 DNA polymerase T4DNA ligase T7 exonuclease and type II restriction endonu-cleases were purchased from New England Biolabs RNase Aand yeast transfer RNA were purchased from Sigma-AldrichProteinase K was purchased from Promega Oligonucleotideprimers were purchased from Geneworks Australia

To clone fragments of the PH1 genome purified virusDNA was digested either with AccI Eco0109I and MseI orSmaI restriction endonucleases blunted with DNA Poly-merase I Large (Klenow) Fragment where appropriate andligated into AccI- Eco0109I- or SmaI-digested pBluescript IIKS+ using T4 DNA ligase according to the manufacturerrsquosinstructions (New England Biolabs) The DNA was intro-duced intoE coliXL1-Blue and transformants grown onLuriaagar containing 15 120583gmL tetracycline and 100 120583gmL ampi-cillinThe resulting cloneswere sequenced and the sequenceswere used to design specific oligonucleotide primers for PCRamplification andor primer walking using the virus genome(or specific restriction fragments) as templates

To amplify PH1 nucleic acid approximately 10 ng virusDNA was combined with 500 nM primers 100120583M of eachdNTP 1U Deep Vent DNA Polymerase (New England Bio-labs) and 1 timesThermoPol buffer (New England Biolabs) ina total reaction volume of 50 120583L Template was denatured at95∘C for 10min followed by 30 cycles of denaturation at 95∘Cfor 30 sec annealing at 56∘C for 30 sec extension at 75∘C for2min and a final extension at 75∘C for 10min PCR reactionswere performed on a PxE02ThermoCycler (Thermo ElectroCorporation) Sequencing reactions using 32 pmol primerwere performed by the dideoxy chain termination methodusing ABI PRISM Big Dye Terminator Mix version 31 onan ABI 3100 capillary sequencer (Applied Biosystems) bythe Applied Genetic Diagnostics Sequencing Service at theDepartment of Pathology (The University of Melbourne)

To obtain terminal genomic fragments PH1 DNA wasdigested by the enzyme Pas1 and the head (sim700 bp) andthe tail (sim1300 bp) fragments were purified by the QIAEXII gel extraction kits and treated with 05M piperidine for

Archaea 15

2 hr at 37∘C to remove protein residues The piperidine-treated fragments were sequenced by the Genomics BiotechCompany (Taipei Taiwan) using two primers TGACCA-ATTAATTAGGCCGGTTCGCC (PH1 head-R) and GTGC-CATACTGCTACAATTCT (PH1 tail-F)

Four hundred fifty-four whole genome sequencing PH1DNA samples (sim5 120583g) were sequenced using parallel pyrose-quencing on a Roche 454 Genome Sequencer System atMission Biotech (Taipei Taiwan) The largest contig was27399with 8803 reads (Newbler version 27) with 27387 posi-tions being of Q40 quality The average depth of coverage foreach base was 829TheGenBank accession for the entire PH1sequence is KC252997

Acknowledgments

The authors thank Carolyn Bath Mei Kwei (Joan) Seahand Danielle Walker for their early work in PH1 characteri-zation Simon Crawford Jocelyn Carpenter and Anna Fried-huber for their assistance with the transmission electronmicroscopy and Tiffany Cowie and Voula Kanellakis fortheir sequencing assistance This paper has been facilitatedby access to the Australian Proteome Analysis Facility estab-lished under the Australian Governmentrsquos Major NationalResearch Facilities program SLT was supported by a grantfrom the Biodiversity Research Center Academia SinicaTaiwan

References

[1] M Krupovic M F White P Forterre and D PrangishvilildquoPostcards from the edge structural genomics of archaealvirusesrdquo Advances in Virus Research vol 82 pp 33ndash62 2012

[2] P Stolt andW Zillig ldquoGene regulation in halophage 120601Hmdashmorethan promotersrdquo Systematic and Applied Microbiology vol 16no 4 pp 591ndash596 1994

[3] R Klein U Baranyi N Rossler B Greineder H Scholz and AWitte ldquoNatrialba magadii virus 120593Ch1 first complete nucleotidesequence and functional organization of a virus infecting ahaloalkaliphilic archaeonrdquo Molecular Microbiology vol 45 no3 pp 851ndash863 2002

[4] E Pagaling R D Haigh W D Grant et al ldquoSequence analysisof an Archaeal virus isolated from a hypersaline lake in InnerMongolia Chinardquo BMC Genomics vol 8 article 410 2007

[5] S L Tang SNuttall andMDyall-Smith ldquoHalovirusesHF1 andHF2 evidence for a recent and large recombination eventrdquo Jour-nal of Bacteriology vol 186 no 9 pp 2810ndash2817 2004

[6] C Bath and M L Dyall-smith ldquoHis1 an archaeal virus of theFuselloviridae family that infectsHaloarcula hispanicardquo Journalof Virology vol 72 no 11 pp 9392ndash9395 1998

[7] C Bath T Cukalac K Porter andM L Dyall-Smith ldquoHis1 andHis2 are distantly related spindle-shaped haloviruses belongingto the novel virus group Salterprovirusrdquo Virology vol 350 no1 pp 228ndash239 2006

[8] K Porter P Kukkaro J K H Bamford et al ldquoSH1 a novelspherical halovirus isolated from an Australian hypersalinelakerdquo Virology vol 335 no 1 pp 22ndash33 2005

[9] M Dyall-Smith S L Tang and C Bath ldquoHaloarchaeal viruseshow diverse are theyrdquo Research in Microbiology vol 154 no 4pp 309ndash313 2003

[10] A Oren G Bratbak and M Heldal ldquoOccurrence of virus-likeparticles in the Dead Seardquo Extremophiles vol 1 no 3 pp 143ndash149 1997

[11] T Sime-Ngando S Lucas A Robin et al et al ldquoDiversityof virus-host systems in hypersaline Lake Retba SenegalrdquoEnvironmental Microbiology vol 13 no 8 pp 1956ndash1972 2010

[12] N Guixa-Boixareu J I Calderon-Paz M Heldal G Bratbakand C Pedros-Alio ldquoViral lysis and bacterivory as prokaryoticloss factors along a salinity gradientrdquoAquaticMicrobial Ecologyvol 11 no 3 pp 215ndash227 1996

[13] S T Jaakkola R K Penttinen S T Vilen et al ldquoClosely relatedarchaeal Haloarcula hispanica icosahedral viruses HHIV-2 andSH1 have nonhomologous genes encoding host recognitionfunctionsrdquo Journal of Virology vol 86 no 9 pp 4734ndash47422012

[14] E Roine P Kukkaro L Paulin et al ldquoNew closely related halo-archaeal viral elements with different nucleic acid typesrdquo Jour-nal of Virology vol 84 no 7 pp 3682ndash3689 2010

[15] A Sencilo L Paulin S KellnerMHelm and E Roine ldquoRelatedhaloarchaeal pleomorphic viruses contain different genometypesrdquo Nucleic Acids Research vol 40 no 12 pp 5523ndash55342012

[16] H T Jaalinoja E Roine P Laurinmaki H M Kivela D HBamford and S J Butcher ldquoStructure and host-cell interactionof SH1 a membrane-containing halophilic euryarchaeal virusrdquoProceedings of the National Academy of Sciences of the UnitedStates of America vol 105 no 23 pp 8008ndash8013 2008

[17] M K Pietila N S Atanasova VManole et al ldquoVirion architec-ture unifies globally distributed pleolipoviruses infecting halo-philic archaeardquo Journal of Virology vol 86 no 9 pp 5067ndash50792012

[18] M L Holmes and M L Dyall-Smith ldquoA plasmid vector witha selectable marker for halophilic archaebacteriardquo Journal ofBacteriology vol 172 no 2 pp 756ndash761 1990

[19] M L Holmes F Pfeifer andM L Dyall-Smith ldquoAnalysis of thehalobacterial plasmid pHK2 minimal repliconrdquo Gene vol 153no 1 pp 117ndash121 1995

[20] M L Dyall-Smith F Pfeiffer K Klee et al ldquoHaloquadratumwalsbyi limited diversity in a Global Pondrdquo PLoS ONE vol 6no 6 Article ID e20968 2011

[21] S Casjens ldquoProphages and bacterial genomics what have welearned so farrdquoMolecular Microbiology vol 49 no 2 pp 277ndash300 2003

[22] R W Hendrix M C M Smith R N Burns M E Ford andG F Hatfull ldquoEvolutionary relationships among diverse bacte-riophages and prophages all the worldrsquos a phagerdquo Proceedings ofthe National Academy of Sciences of the United States of Americavol 96 no 5 pp 2192ndash2197 1999

[23] D Botstein ldquoA theory ofmodular evolution for bacteriophagesrdquoAnnals of the New York Academy of Sciences vol 354 pp 484ndash491 1980

[24] M Krupovic S Gribaldo D H Bamford and P Forterre ldquoTheevolutionary history of archaeal MCM helicases a case study ofvertical evolution combinedwithHitchhiking ofmobile geneticelementsrdquo Molecular Biology and Evolution vol 27 no 12 pp2716ndash2732 2010

[25] M Krupovic P Forterre and D H Bamford ldquoComparativeanalysis of the mosaic genomes of tailed archaeal viruses andproviruses suggests common themes for virion architecture andassembly with tailed viruses of bacteriardquo Journal of MolecularBiology vol 397 no 1 pp 144ndash160 2010

16 Archaea

[26] Z Zhang Y Liu S Wang et al et al ldquoTemperate membrane-containing halophilic archaeal virus SNJ1 has a circular dsDNAgenome identical to that of plasmid pHH205rdquoVirology vol 434no 2 pp 233ndash241 2012

[27] D H Bamford J J Ravantti G Ronnholm et al ldquoConstituentsof SH1 a novel lipid-containing virus infecting the halophiliceuryarchaeonHaloarcula hispanicardquo Journal of Virology vol 79no 14 pp 9097ndash9107 2005

[28] H M Kivela E Roine P Kukkaro S Laurinavicius P Somer-harju andDH Bamford ldquoQuantitative dissociation of archaealvirus SH1 reveals distinct capsid proteins and a lipid corerdquoVirology vol 356 no 1-2 pp 4ndash11 2006

[29] H Jaalinoja Electron Cryo-Microscopy Studies of Bacteriophage8 and Archaeal Virus SH1 University of Helsinki HelsinkiFinland 2007

[30] K Porter B E Russ J Yang andM L Dyall-Smith ldquoThe trans-cription programme of the protein-primed halovirus SH1rdquoMi-crobiology vol 154 no 11 pp 3599ndash3608 2008

[31] K Porter and M L Dyall-Smith ldquoTransfection of haloarchaeaby the DNAs of spindle and round haloviruses and the useof transposon mutagenesis to identify non-essential regionsrdquoMolecular Microbiology vol 70 no 5 pp 1236ndash1245 2008

[32] S E Luria General Virology John Wiley and Sons New YorkNY USA 1953

[33] C A Thomas K Saigo E McLeod and J Ito ldquoThe separationofDNA segments attached to proteinsrdquoAnalytical Biochemistryvol 93 pp 158ndash166 1979

[34] A L Delcher D Harmon S Kasif OWhite and S L SalzbergldquoImproved microbial gene identification with GLIMMERrdquoNucleic Acids Research vol 27 no 23 pp 4636ndash4641 1999

[35] R Zhang Y Yang P Fang et al ldquoDiversity of telomere palin-dromic sequences and replication genes among Streptomyceslinear plasmidsrdquo Applied and Environmental Microbiology vol72 no 9 pp 5728ndash5733 2006

[36] X Wang H S Lee F J Sugar F E Jenney M W W Adamsand J H Prestegard ldquoPF0610 a novel winged helix-turn-helix variant possessing a rubredoxin-like Zn ribbon motiffrom the hyperthermophilic archaeon Pyrococcus furiosusrdquoBiochemistry vol 46 no 3 pp 752ndash761 2007

[37] S W Cline L C Schalkwyk and W F Doolittle ldquoTransfor-mation of the archaebacterium Halobacterium volcanii withgenomic DNArdquo Journal of Bacteriology vol 171 no 9 pp 4987ndash4991 1989

[38] S Kamtekar A J Berman J Wang et al ldquoThe 12059329 DNA poly-merase protein-primer structure suggests a model for the ini-tiation to elongation transitionrdquo EMBO Journal vol 25 no 6pp 1335ndash1343 2006

[39] X Peng T Basta M Haring R A Garrett and D PrangishvilildquoGenome of theAcidianus bottle-shaped virus and insights intothe replication and packaging mechanismsrdquo Virology vol 364no 1 pp 237ndash243 2007

[40] C C Yang C H Huang C Y Li Y G Tsay S C Lee and CW Chen ldquoThe terminal proteins of linear Streptomyces chrom-osomes and plasmids a novel class of replication primingproteinsrdquo Molecular Microbiology vol 43 no 2 pp 297ndash3052002

[41] C C Yang Y H Chen H H Tsai C H Huang T W Huangand C W Chen ldquoIn vitro deoxynucleotidylation of the term-inal protein of Streptomyces linear chromosomesrdquo Applied andEnvironmental Microbiology vol 72 no 12 pp 7959ndash79612006

[42] P C Chang and S N Cohen ldquoBidirectional replication froman internal origin in a linear Streptomyces plasmidrdquo Science vol265 no 5174 pp 952ndash954 1994

[43] S W Wilhelm W H Jeffrey C A Suttle and D L MitchellldquoEstimation of biologically damaging UV levels in marinesurface waters with DNA and viral dosimetersrdquo Photochemistryand Photobiology vol 76 no 3 pp 268ndash273 2002

[44] T Allers and M Mevarech ldquoArchaeal geneticsmdashthe third wayrdquoNature Reviews Genetics vol 6 no 1 pp 58ndash73 2005

[45] M L Holmes S D Nuttall and M L Dyall-Smith ldquoConstruc-tion and use of halobacterial shuttle vectors and further studieson Haloferax DNA gyraserdquo Journal of Bacteriology vol 173 no12 pp 3807ndash3813 1991

[46] S D Nuttall and M L Dyall-Smith ldquoHF1 and HF2 novelbacteriophages of halophilic archaeardquo Virology vol 197 no 2pp 678ndash684 1993

[47] U Blaseio and F Pfeifer ldquoTransformation of Halobacteriumhalobium development of vectors and investigation of gas vesi-cle synthesisrdquo Proceedings of the National Academy of Sciences ofthe United States of America vol 87 no 17 pp 6772ndash6776 1990

[48] W L Lam and W F Doolittle ldquoShuttle vectors for the archae-bacterium Halobacterium volcaniirdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 86 no14 pp 5478ndash5482 1989

[49] C Yanisch-Perron J Vieira and J Messing ldquoImproved M13phage cloning vectors and host strains nucleotide sequences ofthe M13mp18 and pUC19 vectorsrdquo Gene vol 33 no 1 pp 103ndash119 1985

[50] U K Laemmli ldquoCleavage of structural proteins during theassembly of the head of bacteriophage T4rdquo Nature vol 227 no5259 pp 680ndash685 1970

[51] D G Burns H M Camakaris P H Janssen and M L Dyall-Smith ldquoCombined use of cultivation-dependent and culti-vation-independent methods indicates that members of mosthaloarchaeal groups in anAustralian crystallizer pond are culti-vablerdquo Applied and Environmental Microbiology vol 70 no 9pp 5258ndash5265 2004

[52] G Juez F Rodriguez-Valera A Ventosa and D J KushnerldquoHaloarcula hispanica spec nov and Haloferax gibbonsii specnov two new species of extremely halophilic archaebacteriardquoSystematic and Applied Microbiology vol 8 pp 75ndash79 1986

[53] A Oren M Ginzburg B Z Ginzburg L I Hochstein and BE Volcani ldquoHaloarcula marismortui (Volcani) sp nov nomrev an extremely halophilic bacterium from the Dead SeardquoInternational Journal of Systematic Bacteriology vol 40 no 2pp 209ndash210 1990

[54] M Torreblanca F Rodriguez-Valera G Juez A Ventosa MKamekura and M Kates ldquoClassification of non-alkaliphilichalobacteria based on numerical taxonomy and polar lipidcomposition and description of Haloarcula gen nov andHaloferax gen novrdquo Systematic and Applied Microbiology vol8 pp 89ndash99 1986

[55] A Ventosa and A Oren ldquoHalobacterium salinarum nom cor-rig a name to replace Halobacterium salinarium (Elazari-Vol-cani) and to include Halobacterium halobium and Halobac-terium cutirubrumrdquo International Journal of Systematic Bacte-riology vol 46 no 1 p 347 1996

[56] M C Gutierrez M Kamekura M L Holmes M L Dyall-Smith and A Ventosa ldquoTaxonomic characterization of Halo-ferax sp (ldquoH alicanteirdquo) strain Aa 22 description of Haloferaxlucentensis sp novrdquo Extremophiles vol 6 no 6 pp 479ndash4832002

Archaea 17

[57] M F Mullakhanbhai and H Larsen ldquoHalobacterium volcaniispec nov a Dead Sea halobacterium with a moderate saltrequirementrdquo Archives of Microbiology vol 104 no 1 pp 207ndash214 1975

[58] S D Nuttall and M L Dyall-Smith ldquoCh2 a novel halophilicarchaeon from an Australian solar salternrdquo International Jour-nal of Systematic Bacteriology vol 43 no 4 pp 729ndash734 1993

[59] P D Franzman E Stackebrandt K Sanderson et al ldquoHalobac-terium lacusprofundi sp nov a halophilic bacterium isolatedfrom Deep Lake Antarcticardquo Systematic and Applied Microbi-ology vol 11 pp 20ndash27 1988

[60] G A Tomlinson and L I Hochstein ldquoHalobacterium saccharo-vorum sp nov a carbohydrate metabolizing extremely halo-philic bacteriumrdquoCanadian Journal ofMicrobiology vol 22 no4 pp 587ndash591 1976

[61] A Ventosa M C Gutierrez M Kamekura and M L Dyall-Smith ldquoProposal to transfer Halococcus turkmenicus Halobac-terium trapanicum JCM9743 and strainGSL-11 toHaloterrigenaturkmenica gen nov comb novrdquo International Journal ofSystematic Bacteriology vol 49 no 1 pp 131ndash136 1999

[62] M Kamekura and M L Dyall-Smith ldquoTaxonomy of the familyHalobacteriaceae and the description of two new genera Halo-rubrobacterium and Natrialbardquo Journal of General and AppliedMicrobiology vol 41 no 4 pp 333ndash350 1995

[63] T J Carver KM RutherfordM BerrimanM A RajandreamB G Barrell and J Parkhill ldquoACT the Artemis comparisontoolrdquo Bioinformatics vol 21 no 16 pp 3422ndash3423 2005

[64] S W Cline W L Lam R L Charlebois L C Schalkwyk andW F Doolittle ldquoTransformationmethods for halophilic archae-bacteriardquo Canadian Journal of Microbiology vol 35 no 1 pp148ndash152 1989

Archaea 9

Table4Con

tinued

ORF

aPo

sitionb

Locustag

Leng

thc(aa)

pKi

Similaritycharacteris

tics

2412701ndash1506

4HhP

H1gp24

787

39

Capsid

protein

VP2(O

RF24)75aa

similarityto

SH1V

P2(O

RF28Y

P271885)a

ndto

VP2

ofHHIV-2

(AFD

02300)

ZOD2009

19133of

Happau

cihalophilus(ZP

0804

6189)

2515065ndash15961

HhP

H1gp25

298

45

Putativec

apsid

protein

VP5(O

RF25)81aa

similarityto

SH1V

P5(O

RF29Y

P271886)a

ndto

HHIV-2

VP5

(AFD

02301)

HlacA

J19807of

HbflacisalsiAJ

5ZO

D2009

19133of

Happau

cihalophilus

261596

4ndash1644

6HhP

H1gp26

160

46

73aa

similarityto

SH1capsid

protein

VP10

(ORF

30Y

P271887)a

ndto

VP10of

HHIV-2

(AFD

02302)Z

OD2009

19138of

Happau

cihalophilusHlacA

J19802of

Hbflacisa

lsiP

redicted

C-term

inaltransm

embraned

omain(TMHMM)

271644

6ndash16895

HhP

H1gp27

149

42

Capsid

protein

VP9(O

RF27)94aa

similarityto

SH1V

P9(O

RF31Y

P271888)a

ndto

ZOD2009

19143of

Hap

paucihalophilus(ZP

0804

6191)

2816908ndash17921

HhP

H1gp28

337

43

Capsid

protein

VP3(O

RF28)91aa

similarityto

SH1V

P3(O

RF32Y

P271889)a

ndto

NJ7G

2365

ofNa

trinemaspJ7-2

2917928ndash18617

HhP

H1gp29

229

42

Capsidprotein

VP6(O

RF29)83aa

similarityto

SH1V

P6(O

RF33Y

P271890)Also

toNJ7G

3194

ofNa

trinemaspJ7-2and

Hmuk

0476

ofHalom

icrobium

mukohataeiPredictedcarboxypeptid

aser

egulatory-lik

edom

ain(C

arbo

xypepD

reg

pfam

13620)

3018614ndash

18943

HhP

H1gp30

109

49

71aa

similarityto

SH1O

RF34

(YP271891)a

ndto

putativ

eprotein

27of

HHIV-2

(AFD

02308)P

redicted

signalsequence

(signalP)

3119054ndash

19176c

HhP

H1gp31

4050

TwoCx

xCmotifsN

oho

molog

inSH

1orH

HIV-2

3219173ndash19289c

HhP

H1gp32

3841

68aa

similarityto

SH1O

RF37

(YP271894)

3319286ndash

19552c

HhP

H1gp33

8853

58aa

similarityto

SH1O

RF39

(YP271896)a

ndto

HHIV-2

putativ

eprotein

29(A

FD02310)Fou

rCxxCmotifs

3419549ndash

19731c

HhP

H1gp34

6070

Con

tainsC

xxCmotif

Noho

molog

inSH

1orH

HIV-2

3519728ndash20219c

HhP

H1gp35

163

56

93aa

similarityto

SH1O

RF41

(YP271898)a

ndto

putativ

eprotein

30of

HHIV-2

(AFD

02311)

3620216ndash

21415c

HhP

H1gp36

399

50

95aa

similarityto

SH1O

RF42

(YP271899)a

ndto

putativ

eprotein

31of

HHIV-2

(AFD

02312)

3721419ndash

21778c

HhP

H1gp37

11941

91aa

similarityto

SH1O

RF43

(YP271900)a

ndto

putativ

eprotein

32of

HHIV-2

(AFD

02313)

3821762ndash2300

6cHhP

H1gp38

414

42

83aa

similarityto

SH1O

RF44

(YP271901)a

ndto

putativ

eprotein

33of

HHIV-2

(AFD

02314)P

redicted

coil-coiland

helix-tu

rn-helixdo

mains

3923003ndash23158c

HhP

H1gp39

5165

69aa

similarityto

SH1O

RF45

(YP271902)

4023161ndash23370c

HhP

H1gp40

6935

74aa

similarityto

SH1O

RF46

(YP271903)and55sim

ilarityto

putativ

eprotein

35of

HHIV-2

(AFD

02316)

4123422ndash23586c

HhP

H1gp41

5476

86aa

similarityto

SH1O

RF47

(YP271904)Con

tains2

CxxC

motifsand

show

ssim

ilarityto

proteindo

mainfamily

PF14206Arginine-ric

h

4223583ndash24023c

HhP

H1gp42

48

77aa

similarityto

SH1O

RF48

(YP271905)a

ndto

putativ

eprotein

36of

HHIV-2

(AFD

02317)H

ham114540of

Hcc

hamelinensis(ZP11271999)Ph

iCh1p72of

Natrialba

phageP

hiCh

1(NP665989)DUF4

326(pfam14216)

family

domain

4324020ndash

24538c

HhP

H1gp43

172

47

79aa

similarityto

SH1O

RF49

(YP271906)and56sim

ilarityto

putativ

eprotein

37of

HHIV-2

(AFD

02318)

4424535ndash25053c

HhP

H1gp44

172

43

92aa

similarityto

SH1O

RF50

(YP271907)and72sim

ilarityto

putativ

eprotein

38of

HHIV-2

(AFD

02319)

4525050ndash

25277c

HhP

H1gp45

7543

Noho

molog

inSH

1orH

HIV-2

4625274ndash

25387c

HhP

H1gp46

3748

Noho

molog

inSH

1orH

HIV-2

4725533ndash25904

HhP

H1gp47

123

43

61aa

similarityto

SH1O

RF51

(YP271908)66sim

ilarityto

putativ

eprotein

39of

HHIV-2

(AFD

02320)P

redicted

COG1342

domain(D

NAbind

inghelix-tu

rn-helix)

4825901ndash26092c

HhP

H1gp48

6349

Con

tainsC

xxCmotif

Noho

molog

inSH

1orH

HIV-2

4926089ndash

2764

8cHhP

H1gp49

506

40

81aa

similarityto

SH1O

RF55

(YP271912)(bu

tonlyin

theN

-term

inalhalf)H

omolog

ofvirusstru

cturalprotein

VP18

ofHHIV-2

(AFD

02323)

a ORF

swerep

redicted

either

byGLIMMER

orby

manualsearching

forh

omologsintheG

enBa

nkdatabase

b Startandendpo

sitions

ofORF

sare

give

inbp

numbera

ccording

totheP

H1sequenced

epositedatGenBa

nk(KC2

52997)O

RFso

nthec

omplem

entary

strandared

enoted

bythes

uffixc

c Lengthof

thep

redicted

ORF

innu

mbero

faminoacids

10 Archaea

Table 5 Main differences (divergent regions) between the SH1 and PH1 genomes

RegionaSH1 startb SH1 stop Length (bp) PH1 start PH1 stop Length (bp) Comment

DV1 527 650 123 538 604 66

Replacement Both regions have direct repeats at their bordersPH1 has two sets GACCCGGC and CGCTGC while SH1 hasCCCGAC In SH1 this replacement region includes theC-terminal region of ORF2 and the N-terminal region ofORF3 In PH1 it covers only the C-terminal region of ORF1

DV2 2433 2588 155 2386 2387 mdash RMDc from PH1 SH1 repeat at border is TGACCGThisremoves the homolog of SH1 ORF10 from PH1

DV3 3324 3598 274 3137 3416 279 Replacement at the beginning of SH1 ORF13PH1 ORF12(capsid protein VP1)

DV4 6656 7128 472 6478 7085 607 Replacement near the C-terminus of SH1 ORF13PH1 ORF12(capsid protein VP1)

DV5 7491 8341 850 7446 7447 mdash

Indel that results in ORFs14-16 of SH1 not being present inPH1 SH1 ORF14 has a close homolog in HHIV-2 (ORF6) in asimilar position just after the VP1 homolog SH1 ORF15 is aconserved protein in haloarchaea (eg NP 2552A andHmuk 2978) and halovirus His1 (ORF13) Possible invertedrepeat (AGCCATG) at border of SH1 region

DV6 13705 13706 mdash 12818 12851 33RMD from SH1 PH1 repeat at border is CAGCGG(gt)G Thisremoves a part of capsid protein VP2 sequence from the SH1protein

DV7 13789 13863 74 12934 12941 7 Replacement This occurs within VP2 gene of both viruses

DV8 15142 15186 44 14220 14221 mdash RMD from PH1 SH1 repeat at border is TG(tc)CCGACGAand occurs within capsid protein VP2 gene of both viruses

DV9 15303 15317 14 14337 14338 mdash RMD from PH1 SH1 repeat at border is GCCGACGA andoccurs within capsid protein VP2 gene of both viruses

DV10 15504 15527 23 14529 14530 mdash RMD from PH1 SH1 repeat at border is GACGA and occurswithin capsid protein VP2 of both viruses

DV11 20026 20405 379 19019 19173 154Replacement beginning at the start of SH1 ORF35PH1 ORF31This region is longer in SH1 where it includes ORF36 an ORFthat has no homolog present in PH1

DV12 20440 20661 221 19208 19286 78 Replacement This is unequal and includes an ORF in SH1(ORF38) that is not present in PH1

DV13 20943 21085 142 19562 19728 166 Replacement This covers SH1 ORF40PH1 ORF34 Thepredicted proteins are not homologous

DV14 23541 23542 mdash 22171 22197 27 Probable RMD in SH1 PH1 repeat at border is CGTCTCGGand occurs in SH1 ORF44PH1 ORF38

DV15 24506 24521 15 23152 23153 mdash Probable RMD in PH1 SH1 repeat at border is CTCGGT andoccurs near the end of SH1 ORF45PH1 ORF39

DV16 24951 24964 13 23579 23580 mdash Probable RMD in PH1 SH1 repeat at border is CGGTC andoccurs in SH1 ORF47PH1 ORF41

DV17 26449 26450 mdash 25050 25274 224Probable RMD in SH1 PH1 repeat at border is TCATGCG andoccurs near the start of SH1 ORF50PH1 ORF44 and providesan extra ORF for PH1 (ORF45)

DV18 27074 29764 690 25895 26933 1038

Replacement Left border at start of SH1 ORF51PH1 ORF47and extends rightwards into SH1 ORF55PH1 ORF49 (VP18gene) It is an unequal replacement and SH1 has two moreORFs in this region than PH1

aDV Divergent regions between the genomes of SH1 and PH1bStart and stop positions refer to the GenBank sequences of the two viruses SH1 NC 0072171 PH1 KC252997cRMD repeat-mediated deletion event as described in [20]

The recently described halovirus SNJ1 carries manyORFsthat have homologs adjacent to a previously described ViPREofHmc mukohataei (from here on denoted by ViPREHmuk1)a locus that contains genes related to His2 (and other ple-olipoviruses) as well as toHaloferax plasmid pHK2 (probablya provirus) and to two small plasmids ofHqr walsbyi (sim6 kbpL6A and pL6B)The latter relationship provides wider linksto other viruses because one of the PL6 genes (Hqrw 6002)

has homologs in some pleolipoviruses (HPRV-3 and HGPV-1) while another gene (Hqrw 6005) is related to ORF16 ofthe spindle-shaped halovirus His1 [15] After adding the SNJ1gene homologs to ViPREHmuk1 it is extended significantlyand a close inspection of the flanking regions revealed atRNA-ala gene at one end and a partial copy of the sametRNA-ala gene (next to a phage integrase gene) at the otherend (supplementary Figure 1) Genes beyond these points

Archaea 11

Table 6 Transfection of haloarchaea by PH1 DNA

Speciesa Efficiency of transfectiontransformation withPH1 DNAb (PFU120583g of DNA) pUBP2 DNAbc (CFU120583g of DNA)

Har hispanica 53 plusmn 05 times 103 14 plusmn 08 times 104

Har marismortui 45 plusmn 07 times 103 18 plusmn 01 times 104

ldquoHar sinaiiensisrdquo 32 plusmn 06 times 102 mdashHalobacterium salinarum mdash 54 plusmn 42 times 103

Haloferax gibbonsii 28 plusmn 08 times 103 40 plusmn 26 times 103

Haloferax lucentense mdash 57 plusmn 11 times 103

Hfx volcanii mdash 21 plusmn 07 times 105

Hrr lacusprofundi 70 plusmn 22 times 102 95 plusmn 15 times 103

Haloterrigena turkmenica 16 plusmn 17 times 102 24 plusmn 06 times 103

Natrialba asiatica 35 plusmn 39 times 102 31 plusmn 10 times 104aOnly those species positive for transfection andor plasmid transformation are shownbRates of transfection by (nonprotease treated) PH1 DNA or transformation by plasmid pUB2 are averages of three independent experiments each performedin duplicate (plusmn standard deviation)cTransformants were selected on plates with 2 4 or 6120583gmL simvastatin depending on the strainmdash no plaques or colonies observed

appear to be related to cellular metabolism ViPREHmuk1is now 39379 nt in length flanked by a 59 nt direct repeat(potential att sites) and includes 54 ORFs many of which arerelated to known haloviruses are virus-like (eg integrasesmethyltransferases transcriptional regulators DNA methyl-transferase and DNA glycosylase) are homologs in or nearother knownViPREs or show little similarity to other knownproteins This locus also includes an ORC1CDC6 homolog(Hmuk 0446) which could provide a replication function

The mixture of very different virus and plasmid genesseen within ViPREHmuk1 is remarkable but it also revealshomologs found in or adjacent to genomic loci of otherspecies such as the gene homologs of Har marismortuiseen in the left end of ViPREHmuk1 (supplementary Figure1) When these genes are added to the previously describedHaloarcula locus (ViPREHmuk1) it is also extended signifi-cantly It becomes 25550 bp in length encompasses genesfrom Hmar 2382 to Hmar 2404 and is flanked by a fulltRNA-ala gene at one end and a partial copy at the other Itcarries an ORC1CDC6 homolog an integrase and a phiH-like repressor as well as pleolipovirus homologs

210 Transfection of Haloarchaea by PH1 DNA PH1 DNAwas introduced into Har hispanica cells using the PEGmethod [37] and the cells were screened for virus productionby plaque assay Figure 7 shows the increase in transfectedcells with input (nonproteinase K-treated) virus DNA Theestimated efficiencywas 53times 103 PFUper120583gDNA PH1DNAthat had been treated with proteinase K or with DNAase I(RNase-free) did not produce plaques (data not shown) PH1DNA was found to transfect six species of haloarchaea otherthanHar hispanica includingmembers of the generaHaloar-cula Haloferax Halorubrum Haloterrigena and Natrialba(Table 6) For these experiments the virus production fromtransfected cells was detected by plaque assay on indicatorlawns of Har hispanica

0

0

200 400 600 800 1000

Titre

(PFU

mL)

104

103

102

PH1 DNA (ng)

Figure 7 Transfection ofHar hispanica cells with PH1 DNA Vary-ing amounts of nonproteinase K-treated PH1 DNAwere introducedinto cells of Har hispanica using the PEG method [64] Cells werethen screened for infective centres by plaque assay Data shown is theaverage of three independent experiments performed in duplicateError bars represent one standard deviation of themean If protease-treated DNA was used no transfectant plaques were observed

3 Discussion

PH1 is very similar in particlemorphology genome structureand sequence to the previously described halovirus SH1 [816] Like SH1 it has long inverted terminal repeat sequencesand terminal proteins indicative of protein-primed replica-tion Purified PH1 particles have a relatively low buoyant

12 Archaea

density and are chloroform sensitive and show a layeredcapsid structure consistent with the likely presence of aninternalmembrane layer as has been shown for SH1HHIV-2and SNJ1 [13 26]The particle stability of PH1 to temperaturepH and reduced salt was similar to SH1 [8] The structuralproteins of PH1 are very similar in sequence and relativeabundance to those of SH1 so the two viruses are likely toshare the same particle geometry [16]

Viruses that use protein-primed replication such as bac-teriophageΦ29 carry a viral type BDNApolymerase that caninteract specifically with the proteins attached to the genomictermini and initiate strand synthesis [38] ArchaeovirusesHis1 [7] and Acidianus bottle-shaped virus [39] probablyuse this mode of replication However a polymerase genecannot be found in the genomes of PH1 SH1 or HHIV-2DNA polymerases are large enzymes and the only ORF ofSH1 that was long enough to encode such an enzyme andhad not previously been assigned as a structural protein wasORF55 which specified an 865 aa protein that contained noconserved domains indicative of polymerases [27]The recentstudy of HHIV-2 showed that the corresponding ORF in thisvirus (gene 42) is a structural protein of the virion and byinference this is likely also for the corresponding proteinsof SH1 and PH1 (no homolog is present in SNJ1) Withouta polymerase gene these viruses must use a host enzymebut searches for a viral type B DNA polymerase in thegenome of the hostHar hispanica or in the genomes of othersequenced haloarchaea did not find anymatchesThis arguesstrongly for a replication mechanism that is different to thatexemplified byΦ29 An attractive alternative is that displayedby Streptomyces linear plasmids which have 51015840-terminal pro-teins and use a cellular polymerase for replication [40] Theterminal proteins are not used for primary replication but forend patching [41] and it has been shown that if the term-inal repeat sequences are removed from these plasmids andthe ends ligated they can replicate as circular plasmids [42]This provides a testable hypothesis for the replication of halo-sphaeroviruses and also offers a pathway for switchingbetween linear (eg PH1) and circular (eg SNJ1) forms Useof a host polymerase would also fit with the ability of SH1 andPH1 DNA to transfect many different haloarchaeal species[31] as these enzymes and theirmode of action are highly con-served

The growth characteristics of PH1 inHar hispanica bothby single-cell burst and single-step growth experiments gavevalues of 50ndash100 virusescell significantly less than that ofSH1 (200 virusescell) [8] and HHIV-2 (180 virusescell) [13]The latter virus is lytic whereas SH1 PH1 and SNJ1 start pro-ducing extracellular viruswell before any decrease in cell den-sity indicating that virus release can occur without lysisThishas been most clearly shown with SH1 where almost 100of cells can be infected [8] Infection ofHar hispanica by PH1was less efficient (sim30 of cells) but the kinetics of virus pro-duction in single-step growth curves of SH1 and PH1 aresimilar with virus production occurring over several hoursrather than at a clearly defined time after infection The twoviruses are also very closely related and infect the same hostspecies so it is likely that they use the same scheme for cellexit For both viruses the cell density eventually decreases in

single-step cultures showing that virus infection does resultin cell death This mode of exit appears to maximise theproduction of the virus over time as the host cells survive foran extended period In natural hypersaline waters cell num-bers are usually high but growth rates are low [12] and thehigh incident UV (on often shallow ponds) would damagevirus DNA so reducing the half-life of released virus part-icles [43] These factors may have favoured the exit strategydisplayed by these viruses improving their chance of trans-mission to a new host

The motif GATC was absent in the genomes of PH1 SH1andHHIV-2 and themotif CTAGwas either absent or greatlyunderrepresented All three viruses share the same host butGATC and CTAG motifs are plentiful in the Har hispanicagenome (accessions CP002921 and CP002923) Avoidance ofthesemotifs in halovirusesHis1 His2 HF1 andHF2 has beenreported previously [7] Such purifying selection commonlyresults from host restriction enzymes and inHfx volcanii itis exactly these two motifs that are targeted [44] One Hfxvolcanii enzyme recognises A-methylated GATC sites [45]and another recognises un-methylated CTAG sites (whichare protected by methylation in the genome) In the currentstudy this could explain the negative transfection results forHfx volcanii as the PH1 genome contains two CTAGmotifsBy comparison the SH1 genome contains no CTAG motifsand is able to transfect Hfx volcanii [31]

Comparison of the PH1 and SH1 genomes allowed adetailed picture of the natural variation occurring betweenclosely related viruses revealing likely deletion events medi-ated by small repeats a process described previously in astudy of Hqr walsbyi and termed repeat-mediated deletion[20] There are also many replacements including blocks ofsequence that occur within long open reading frames (suchas the capsid protein genes VP1 and VP18) VP3 and VP6 areknown to form the large spikes on the external surface of theSH1 virion and presumably one or both interact with host cellreceptors [16] PH1 and SH1 have identical host ranges con-sistent with the high sequence similarity shown between theircorresponding VP3 and VP6 proteins

A previous study of SH1 could not detect transcriptsacross annotated ORFs 1ndash3 or 55-56 [30] ORFs 1 and 56occur in the terminal inverted repeat sequence and in thepresent study it was found that there were no ORFs cor-responding to these in the PH1 genome Given the close simi-larity of the two genomes the comparative data are in agree-ment with the transcriptional data indicating that these SH1ORFs are not used SH1 ORFs 2 and 3 are more problematicas good homologs of these are also present in PH1 The con-flict between the transcriptional data and the comparativegenomic evidence requires further experimental work toresolveThe status of SH1ORF55 has recently been confirmedby studies of the related virus HHIV-2 (discussed above)

The pleolipovirus group has expanded dramatically inthe last few years and it now comprises a diverse group ofviruseswith different genome types and replication strategiesWhat is evenmore remarkable is that a similar expansion cannow be seen with SH1-related viruses which includes viruseswith at least two replication modes and genome types (linearand circular) Evidence from haloarchaeal genome sequences

Archaea 13

show cases of not only provirus integrants or plasmids (egpKH2 and pHH205) but also genomic loci (ViPREs) thatcontain cassettes of virus genes from different sources andat least two cases (described in this study) have likely attsites that indicate circularisation and mobility While thereis a clear relationship between viruses with circular ds- andssDNA genomes that replicate via the rolling circle method(ie the ssDNA form is a replication intermediate) it ismore difficult to explain how capsid gene cassettes can movebetween these viruses and those with linear genomes andterminal proteins Within the pleolipoviruses His2 has alinear genome that contains a viral type B DNA polymeraseand terminal proteins while all the other described membershave circular genomes that contain a rep homolog and prob-ably replicate via the rolling circle method [14 19] Similarlyamong the halosphaeroviruses PH1 SH1 and HHIV-2 havelinear dsDNA with terminal proteins and probably replicatein the same way but the related SNJ1 virus has a circulardsDNA genome The clear relationships shown by the capsidgenes of viruses within each group plus the connectionsshown to haloviruses outside each group (eg the DNA poly-merases of His1 and His2) all speak of a vigorous meansof recombination one that can readily switch capsid genesbetween viruses with radically different replication strategiesHow is the process most likely to operate One possibility issuggested by ViPREs which appear to be mobile collectionsof capsid and replication genes from different sources Theyoffer fixed locations for recombination to occur providegene cassettes that can be reassorted to produce novel virusgenomes and some can probably recombine out as circularforms For example ViPREHmuk1 carries genes related topleolipoviruses halosphaeroviruses and other haloviruses Itis yet to be determined if any of the genes carried in theseloci are expressed in the cell (such as the genes for capsidproteins) or if ViPREs provide any selective advantage to thehost

4 Materials and Methods

41 Water Sample A water sample was collected in 1998from Pink Lake (32∘001015840 S and 115∘301015840 E) a hypersalinelake on Rottnest Island Western Australia Australia It wasscreened in the same year for haloviruses using the methodsdescribed in [8] A single plaque on a Har hispanica lawnplate was picked and replaque purified The novel haloviruswas designated PH1

42 Media Strains and Plasmids The media used in thisstudy are described in the online resource The Halohand-book (httpwwwhaloarchaeacomresourceshalohandbookindexhtml) Artificial salt water containing 30 (wv) totalsalts comprised of 4M NaCl 150mM MgCl

2 150mM

MgSO4 90mM KCl and 35mM CaCl

2and adjusted to pH

75 using sim2mL 1M Tris-HCl (pH 75) per litre MGM con-taining 12 18 or 23 (wv) total salts and HVD (halo-virus diluent) were prepared from the concentrated stock aspreviously described [46] Bacto-agar (Difco Laboratories)was added to MGM for solid (15 gL) or top-layer (7 gL)media

Table 1 lists the haloarchaea used in this study Allhaloarchaea were grown aerobically at 37∘C in either 18 or23 (wv)MGM(depending on the strain) andwith agitation(except for ldquoHar sinaiiensisrdquo)The plasmids used were pBlue-script II KS+ (Stratagene Cloning Systems) pUBP2 [47]and pWL102 [48] pUBP2 and pWL102 were first passagedthrough E coli JM110 [49] to prevent dam-methylationof DNA which has been shown to reduce transformationefficiency in some strains [45]

43 Negative-Stain Electron Microscopy The method fornegative-stain TEM was adapted from that of V Tarasovdescribed in the online resource The Halohandbook (httpwwwhaloarchaeacomresourceshalohandbookindexhtml)A 20120583L drop of the sample was placed on a clean surfaceand the virus particles were allowed to adsorb to a Formvarfilm 400-mesh copper grid (ProSciTech) for 15ndash2min Theywere then negatively stained with a 20120583L drop of 2 (wv)uranyl acetate for 1-2min Excess liquid was absorbed withfilter paper and the grid was allowed to air dry Grids wereexamined either on a Philips CM 120 BioTwin transmissionelectron microscope (Royal Philips Electronics) operating atan accelerating voltage of 120 kV or on a Siemens Elmiskop102 transmission electron microscope (Siemens AG) operat-ing at an accelerating voltage of 120 kV

44 Virus Host Range Twelve haloarchaeal strains from thegenera Haloarcula Halobacterium Haloferax HalorubrumandNatrialba (Table 1)Thirteen naturalHalorubrum isolates(H Camakaris unpublished data) and five uncharacterizedhaloarchaeal isolates fromLakeHardy Pink Lake (inWesternAustralia) and Serpentine Lake (D Walker and M K Seahunpublished data) were screened for PH1 susceptibilityLysates from PH1-infected Har hispanica cultures (1 times1011 PFUmL) were spotted onto lawns of each strain (usingmedia with 12 or 18 (wv)MGM depending on the strain)and incubated for 2ndash5 days at 30 and at 37∘C

45 Large Scale Virus Growth and Purification Liquid cul-tures of PH1 were grown by the infection (MOI 005) of anearly exponential Har hispanica culture in 18 (wv) MGMCultures were incubated aerobically at 37∘C with agitationfor 3 days Clearing (ie complete cell lysis) of PH1-infectedcultures did not occur and consequently cultures were harv-ested when the absorbance at 550 nm reached the minimumand the titre (determined by plaque assay) was at the maxi-mum usually sim1010ndash1011 PFUmL Virus was purified usingthe method described previously for SH1 [8 31] except thatafter the initial low speed spin (Sorvall GSA 6000 rpm30min 10∘C) virus was concentrated from the infected cul-ture by centrifugation at 26000 rpm (13 hr 10∘C) onto acushion of 30 (wv) sucrose in HVDThe pellet was resus-pended in a small volume of HVD and loaded onto a pre-formed linear 5ndash70 (wv) sucrose gradient followed byisopycnic centrifugation in 13 gmL CsCl (Beckman 70Ti60000 rpm 20 hr 10∘C) The white virus band occurredat a density of 129 gmL and was collected and diluted inhalovirus diluent (HVD see Section 4) and the virus pelleted(Beckman SW55 35000 rpm 75min 10∘C) and resuspended

14 Archaea

in a small volume of HVD and stored at 4∘C Virus recoveryat the major stages of a typical purification is given insupplementary Table 1 The specific infectivity of pure virussolutions was determined as the ratio of the PFUmL to theabsorbance at 260 nm

46 PH1 Single-Step Growth Curve An early exponentialphase culture of Har hispanica grown in 18 (wv) MGMwas infected with PH1 (MOI 50) Under these conditions thepercentage of infected cells was approximately 30 After anadsorption period of 1 hr at 37∘C the cells were washed threetimes with 18 (wv) MGM (at room temperature) resus-pended in 100mL 18 (wv) MGM (these methods ensuredthe removal of all residual virus) and incubated at 37∘Cwith shaking (100 rpm) Samples were removed at hourlyintervals for measurements of absorbance at 550 nm andthe number of infective centres Immediately after samplingtitres were determined by plaque assay with an indicator lawnof Har hispanica Each experiment was performed in tripli-cate

47 Halovirus Stability After various treatments sampleswere removed and diluted in HVD and virus titres weredetermined by plaque assay on Har hispanica Each experi-ment was performed in triplicate and representative data areshown Chloroform sensitivity was examined by the exposureof PH1 lysates in 18 (wv) MGM to chloroform in a volumeratio of 1 4 (chloroform to lysate) Incubation was at roomtemperature with constant agitation At appropriate timepoints the mix was allowed to settle and a sample wasremoved from the upper layer and diluted in HVDThe effectof a low ionic environment was examined by diluting PH1lysates (in 18 (wv) MGM) into double distilled H

2O in a

volume ratio of 1 1000 (lysate to double distilledH2O) Incu-

bation was at room temperature with constant agitationSamples were removed at various times and diluted in HVDThe pH stability of PH1 was determined by dilution of PH1lysates in 18 (wv)MGMin the appropriate pHbuffer (HVDbuffered with appropriate Tris-HCl) in a volume ratio of1 100 (lysate to buffer) Incubation was at room temperaturewith constant agitation After 30min samples were removedand diluted in HVD Thermal stability of PH1 was examinedby a 1 hr incubation of a virus lysate in 18 (wv) MGM atdifferent temperatures after which they were brought quicklyto room temperature diluted in HVD and titrated

48 Protein Procedures To remove salts purified virus prepa-rations were mixed with trichloroacetic acid (10 (vv) finalconcentration) and incubated on ice for 15min to allowthe proteins to precipitate After centrifugation (16000 g15min room temperature) the precipitate was washed threetimes in acetone dried and resuspended in double distilledH2O Proteins were dissolved in Laemmli sample buffer with

48mM 120573-mercaptoethanol [50] heated in boiling water for5min then separated on 12 (wv) NuPAGE Novex Bis-Tris Gels using MES-SDS running buffer according to themanufacturerrsquos directions (Invitrogen) After electrophoresisgels were rinsed in double distilled H

2O and stained with

01 (wv) Brilliant Blue G in 40 (vv) methanol and

10 (vv) acetic acid Gels were destained with severalchanges of 40 (vv) methanol and 10 (vv) acetic acidAlternatively gels were stained with GelCode GlycoproteinStaining Kit Pro-Q Emerald 488 Glycoprotein Gel and BlotStain Kit SYPRO Ruby Protein Gel Stain according tothe manufacturerrsquos directions (Invitrogen Molecular ProbesPierce Biotechnology)

Protein bands were cut from the gels and sent to theAustralian Proteome Analysis Facility (Macquarie Univer-sity) for trypsin digestion and analysis by matrix assistedlaser desorption ionisation time of flight mass spectrometry(MALDI-TOF MS) on an Applied Biosystems 4700 Pro-teomics Analyser (Applied Biosystems)

49 DNA Procedures Proteinase K-treated and nonpro-teinaseK-treatedDNApreparationsweremade frompurifiedPH1 using the methods described previously for SH1 [8]DNA was separated on 1 (wv) agarose gels in Tris-acetate-EDTA electrophoresis buffer and was stained with ethidiumbromide (Sigma-Aldrich)120582 DNA DNase I (RNase-free) exonuclease III mung

bean nuclease nuclease BAL-31 T4 DNA polymerase T4DNA ligase T7 exonuclease and type II restriction endonu-cleases were purchased from New England Biolabs RNase Aand yeast transfer RNA were purchased from Sigma-AldrichProteinase K was purchased from Promega Oligonucleotideprimers were purchased from Geneworks Australia

To clone fragments of the PH1 genome purified virusDNA was digested either with AccI Eco0109I and MseI orSmaI restriction endonucleases blunted with DNA Poly-merase I Large (Klenow) Fragment where appropriate andligated into AccI- Eco0109I- or SmaI-digested pBluescript IIKS+ using T4 DNA ligase according to the manufacturerrsquosinstructions (New England Biolabs) The DNA was intro-duced intoE coliXL1-Blue and transformants grown onLuriaagar containing 15 120583gmL tetracycline and 100 120583gmL ampi-cillinThe resulting cloneswere sequenced and the sequenceswere used to design specific oligonucleotide primers for PCRamplification andor primer walking using the virus genome(or specific restriction fragments) as templates

To amplify PH1 nucleic acid approximately 10 ng virusDNA was combined with 500 nM primers 100120583M of eachdNTP 1U Deep Vent DNA Polymerase (New England Bio-labs) and 1 timesThermoPol buffer (New England Biolabs) ina total reaction volume of 50 120583L Template was denatured at95∘C for 10min followed by 30 cycles of denaturation at 95∘Cfor 30 sec annealing at 56∘C for 30 sec extension at 75∘C for2min and a final extension at 75∘C for 10min PCR reactionswere performed on a PxE02ThermoCycler (Thermo ElectroCorporation) Sequencing reactions using 32 pmol primerwere performed by the dideoxy chain termination methodusing ABI PRISM Big Dye Terminator Mix version 31 onan ABI 3100 capillary sequencer (Applied Biosystems) bythe Applied Genetic Diagnostics Sequencing Service at theDepartment of Pathology (The University of Melbourne)

To obtain terminal genomic fragments PH1 DNA wasdigested by the enzyme Pas1 and the head (sim700 bp) andthe tail (sim1300 bp) fragments were purified by the QIAEXII gel extraction kits and treated with 05M piperidine for

Archaea 15

2 hr at 37∘C to remove protein residues The piperidine-treated fragments were sequenced by the Genomics BiotechCompany (Taipei Taiwan) using two primers TGACCA-ATTAATTAGGCCGGTTCGCC (PH1 head-R) and GTGC-CATACTGCTACAATTCT (PH1 tail-F)

Four hundred fifty-four whole genome sequencing PH1DNA samples (sim5 120583g) were sequenced using parallel pyrose-quencing on a Roche 454 Genome Sequencer System atMission Biotech (Taipei Taiwan) The largest contig was27399with 8803 reads (Newbler version 27) with 27387 posi-tions being of Q40 quality The average depth of coverage foreach base was 829TheGenBank accession for the entire PH1sequence is KC252997

Acknowledgments

The authors thank Carolyn Bath Mei Kwei (Joan) Seahand Danielle Walker for their early work in PH1 characteri-zation Simon Crawford Jocelyn Carpenter and Anna Fried-huber for their assistance with the transmission electronmicroscopy and Tiffany Cowie and Voula Kanellakis fortheir sequencing assistance This paper has been facilitatedby access to the Australian Proteome Analysis Facility estab-lished under the Australian Governmentrsquos Major NationalResearch Facilities program SLT was supported by a grantfrom the Biodiversity Research Center Academia SinicaTaiwan

References

[1] M Krupovic M F White P Forterre and D PrangishvilildquoPostcards from the edge structural genomics of archaealvirusesrdquo Advances in Virus Research vol 82 pp 33ndash62 2012

[2] P Stolt andW Zillig ldquoGene regulation in halophage 120601Hmdashmorethan promotersrdquo Systematic and Applied Microbiology vol 16no 4 pp 591ndash596 1994

[3] R Klein U Baranyi N Rossler B Greineder H Scholz and AWitte ldquoNatrialba magadii virus 120593Ch1 first complete nucleotidesequence and functional organization of a virus infecting ahaloalkaliphilic archaeonrdquo Molecular Microbiology vol 45 no3 pp 851ndash863 2002

[4] E Pagaling R D Haigh W D Grant et al ldquoSequence analysisof an Archaeal virus isolated from a hypersaline lake in InnerMongolia Chinardquo BMC Genomics vol 8 article 410 2007

[5] S L Tang SNuttall andMDyall-Smith ldquoHalovirusesHF1 andHF2 evidence for a recent and large recombination eventrdquo Jour-nal of Bacteriology vol 186 no 9 pp 2810ndash2817 2004

[6] C Bath and M L Dyall-smith ldquoHis1 an archaeal virus of theFuselloviridae family that infectsHaloarcula hispanicardquo Journalof Virology vol 72 no 11 pp 9392ndash9395 1998

[7] C Bath T Cukalac K Porter andM L Dyall-Smith ldquoHis1 andHis2 are distantly related spindle-shaped haloviruses belongingto the novel virus group Salterprovirusrdquo Virology vol 350 no1 pp 228ndash239 2006

[8] K Porter P Kukkaro J K H Bamford et al ldquoSH1 a novelspherical halovirus isolated from an Australian hypersalinelakerdquo Virology vol 335 no 1 pp 22ndash33 2005

[9] M Dyall-Smith S L Tang and C Bath ldquoHaloarchaeal viruseshow diverse are theyrdquo Research in Microbiology vol 154 no 4pp 309ndash313 2003

[10] A Oren G Bratbak and M Heldal ldquoOccurrence of virus-likeparticles in the Dead Seardquo Extremophiles vol 1 no 3 pp 143ndash149 1997

[11] T Sime-Ngando S Lucas A Robin et al et al ldquoDiversityof virus-host systems in hypersaline Lake Retba SenegalrdquoEnvironmental Microbiology vol 13 no 8 pp 1956ndash1972 2010

[12] N Guixa-Boixareu J I Calderon-Paz M Heldal G Bratbakand C Pedros-Alio ldquoViral lysis and bacterivory as prokaryoticloss factors along a salinity gradientrdquoAquaticMicrobial Ecologyvol 11 no 3 pp 215ndash227 1996

[13] S T Jaakkola R K Penttinen S T Vilen et al ldquoClosely relatedarchaeal Haloarcula hispanica icosahedral viruses HHIV-2 andSH1 have nonhomologous genes encoding host recognitionfunctionsrdquo Journal of Virology vol 86 no 9 pp 4734ndash47422012

[14] E Roine P Kukkaro L Paulin et al ldquoNew closely related halo-archaeal viral elements with different nucleic acid typesrdquo Jour-nal of Virology vol 84 no 7 pp 3682ndash3689 2010

[15] A Sencilo L Paulin S KellnerMHelm and E Roine ldquoRelatedhaloarchaeal pleomorphic viruses contain different genometypesrdquo Nucleic Acids Research vol 40 no 12 pp 5523ndash55342012

[16] H T Jaalinoja E Roine P Laurinmaki H M Kivela D HBamford and S J Butcher ldquoStructure and host-cell interactionof SH1 a membrane-containing halophilic euryarchaeal virusrdquoProceedings of the National Academy of Sciences of the UnitedStates of America vol 105 no 23 pp 8008ndash8013 2008

[17] M K Pietila N S Atanasova VManole et al ldquoVirion architec-ture unifies globally distributed pleolipoviruses infecting halo-philic archaeardquo Journal of Virology vol 86 no 9 pp 5067ndash50792012

[18] M L Holmes and M L Dyall-Smith ldquoA plasmid vector witha selectable marker for halophilic archaebacteriardquo Journal ofBacteriology vol 172 no 2 pp 756ndash761 1990

[19] M L Holmes F Pfeifer andM L Dyall-Smith ldquoAnalysis of thehalobacterial plasmid pHK2 minimal repliconrdquo Gene vol 153no 1 pp 117ndash121 1995

[20] M L Dyall-Smith F Pfeiffer K Klee et al ldquoHaloquadratumwalsbyi limited diversity in a Global Pondrdquo PLoS ONE vol 6no 6 Article ID e20968 2011

[21] S Casjens ldquoProphages and bacterial genomics what have welearned so farrdquoMolecular Microbiology vol 49 no 2 pp 277ndash300 2003

[22] R W Hendrix M C M Smith R N Burns M E Ford andG F Hatfull ldquoEvolutionary relationships among diverse bacte-riophages and prophages all the worldrsquos a phagerdquo Proceedings ofthe National Academy of Sciences of the United States of Americavol 96 no 5 pp 2192ndash2197 1999

[23] D Botstein ldquoA theory ofmodular evolution for bacteriophagesrdquoAnnals of the New York Academy of Sciences vol 354 pp 484ndash491 1980

[24] M Krupovic S Gribaldo D H Bamford and P Forterre ldquoTheevolutionary history of archaeal MCM helicases a case study ofvertical evolution combinedwithHitchhiking ofmobile geneticelementsrdquo Molecular Biology and Evolution vol 27 no 12 pp2716ndash2732 2010

[25] M Krupovic P Forterre and D H Bamford ldquoComparativeanalysis of the mosaic genomes of tailed archaeal viruses andproviruses suggests common themes for virion architecture andassembly with tailed viruses of bacteriardquo Journal of MolecularBiology vol 397 no 1 pp 144ndash160 2010

16 Archaea

[26] Z Zhang Y Liu S Wang et al et al ldquoTemperate membrane-containing halophilic archaeal virus SNJ1 has a circular dsDNAgenome identical to that of plasmid pHH205rdquoVirology vol 434no 2 pp 233ndash241 2012

[27] D H Bamford J J Ravantti G Ronnholm et al ldquoConstituentsof SH1 a novel lipid-containing virus infecting the halophiliceuryarchaeonHaloarcula hispanicardquo Journal of Virology vol 79no 14 pp 9097ndash9107 2005

[28] H M Kivela E Roine P Kukkaro S Laurinavicius P Somer-harju andDH Bamford ldquoQuantitative dissociation of archaealvirus SH1 reveals distinct capsid proteins and a lipid corerdquoVirology vol 356 no 1-2 pp 4ndash11 2006

[29] H Jaalinoja Electron Cryo-Microscopy Studies of Bacteriophage8 and Archaeal Virus SH1 University of Helsinki HelsinkiFinland 2007

[30] K Porter B E Russ J Yang andM L Dyall-Smith ldquoThe trans-cription programme of the protein-primed halovirus SH1rdquoMi-crobiology vol 154 no 11 pp 3599ndash3608 2008

[31] K Porter and M L Dyall-Smith ldquoTransfection of haloarchaeaby the DNAs of spindle and round haloviruses and the useof transposon mutagenesis to identify non-essential regionsrdquoMolecular Microbiology vol 70 no 5 pp 1236ndash1245 2008

[32] S E Luria General Virology John Wiley and Sons New YorkNY USA 1953

[33] C A Thomas K Saigo E McLeod and J Ito ldquoThe separationofDNA segments attached to proteinsrdquoAnalytical Biochemistryvol 93 pp 158ndash166 1979

[34] A L Delcher D Harmon S Kasif OWhite and S L SalzbergldquoImproved microbial gene identification with GLIMMERrdquoNucleic Acids Research vol 27 no 23 pp 4636ndash4641 1999

[35] R Zhang Y Yang P Fang et al ldquoDiversity of telomere palin-dromic sequences and replication genes among Streptomyceslinear plasmidsrdquo Applied and Environmental Microbiology vol72 no 9 pp 5728ndash5733 2006

[36] X Wang H S Lee F J Sugar F E Jenney M W W Adamsand J H Prestegard ldquoPF0610 a novel winged helix-turn-helix variant possessing a rubredoxin-like Zn ribbon motiffrom the hyperthermophilic archaeon Pyrococcus furiosusrdquoBiochemistry vol 46 no 3 pp 752ndash761 2007

[37] S W Cline L C Schalkwyk and W F Doolittle ldquoTransfor-mation of the archaebacterium Halobacterium volcanii withgenomic DNArdquo Journal of Bacteriology vol 171 no 9 pp 4987ndash4991 1989

[38] S Kamtekar A J Berman J Wang et al ldquoThe 12059329 DNA poly-merase protein-primer structure suggests a model for the ini-tiation to elongation transitionrdquo EMBO Journal vol 25 no 6pp 1335ndash1343 2006

[39] X Peng T Basta M Haring R A Garrett and D PrangishvilildquoGenome of theAcidianus bottle-shaped virus and insights intothe replication and packaging mechanismsrdquo Virology vol 364no 1 pp 237ndash243 2007

[40] C C Yang C H Huang C Y Li Y G Tsay S C Lee and CW Chen ldquoThe terminal proteins of linear Streptomyces chrom-osomes and plasmids a novel class of replication primingproteinsrdquo Molecular Microbiology vol 43 no 2 pp 297ndash3052002

[41] C C Yang Y H Chen H H Tsai C H Huang T W Huangand C W Chen ldquoIn vitro deoxynucleotidylation of the term-inal protein of Streptomyces linear chromosomesrdquo Applied andEnvironmental Microbiology vol 72 no 12 pp 7959ndash79612006

[42] P C Chang and S N Cohen ldquoBidirectional replication froman internal origin in a linear Streptomyces plasmidrdquo Science vol265 no 5174 pp 952ndash954 1994

[43] S W Wilhelm W H Jeffrey C A Suttle and D L MitchellldquoEstimation of biologically damaging UV levels in marinesurface waters with DNA and viral dosimetersrdquo Photochemistryand Photobiology vol 76 no 3 pp 268ndash273 2002

[44] T Allers and M Mevarech ldquoArchaeal geneticsmdashthe third wayrdquoNature Reviews Genetics vol 6 no 1 pp 58ndash73 2005

[45] M L Holmes S D Nuttall and M L Dyall-Smith ldquoConstruc-tion and use of halobacterial shuttle vectors and further studieson Haloferax DNA gyraserdquo Journal of Bacteriology vol 173 no12 pp 3807ndash3813 1991

[46] S D Nuttall and M L Dyall-Smith ldquoHF1 and HF2 novelbacteriophages of halophilic archaeardquo Virology vol 197 no 2pp 678ndash684 1993

[47] U Blaseio and F Pfeifer ldquoTransformation of Halobacteriumhalobium development of vectors and investigation of gas vesi-cle synthesisrdquo Proceedings of the National Academy of Sciences ofthe United States of America vol 87 no 17 pp 6772ndash6776 1990

[48] W L Lam and W F Doolittle ldquoShuttle vectors for the archae-bacterium Halobacterium volcaniirdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 86 no14 pp 5478ndash5482 1989

[49] C Yanisch-Perron J Vieira and J Messing ldquoImproved M13phage cloning vectors and host strains nucleotide sequences ofthe M13mp18 and pUC19 vectorsrdquo Gene vol 33 no 1 pp 103ndash119 1985

[50] U K Laemmli ldquoCleavage of structural proteins during theassembly of the head of bacteriophage T4rdquo Nature vol 227 no5259 pp 680ndash685 1970

[51] D G Burns H M Camakaris P H Janssen and M L Dyall-Smith ldquoCombined use of cultivation-dependent and culti-vation-independent methods indicates that members of mosthaloarchaeal groups in anAustralian crystallizer pond are culti-vablerdquo Applied and Environmental Microbiology vol 70 no 9pp 5258ndash5265 2004

[52] G Juez F Rodriguez-Valera A Ventosa and D J KushnerldquoHaloarcula hispanica spec nov and Haloferax gibbonsii specnov two new species of extremely halophilic archaebacteriardquoSystematic and Applied Microbiology vol 8 pp 75ndash79 1986

[53] A Oren M Ginzburg B Z Ginzburg L I Hochstein and BE Volcani ldquoHaloarcula marismortui (Volcani) sp nov nomrev an extremely halophilic bacterium from the Dead SeardquoInternational Journal of Systematic Bacteriology vol 40 no 2pp 209ndash210 1990

[54] M Torreblanca F Rodriguez-Valera G Juez A Ventosa MKamekura and M Kates ldquoClassification of non-alkaliphilichalobacteria based on numerical taxonomy and polar lipidcomposition and description of Haloarcula gen nov andHaloferax gen novrdquo Systematic and Applied Microbiology vol8 pp 89ndash99 1986

[55] A Ventosa and A Oren ldquoHalobacterium salinarum nom cor-rig a name to replace Halobacterium salinarium (Elazari-Vol-cani) and to include Halobacterium halobium and Halobac-terium cutirubrumrdquo International Journal of Systematic Bacte-riology vol 46 no 1 p 347 1996

[56] M C Gutierrez M Kamekura M L Holmes M L Dyall-Smith and A Ventosa ldquoTaxonomic characterization of Halo-ferax sp (ldquoH alicanteirdquo) strain Aa 22 description of Haloferaxlucentensis sp novrdquo Extremophiles vol 6 no 6 pp 479ndash4832002

Archaea 17

[57] M F Mullakhanbhai and H Larsen ldquoHalobacterium volcaniispec nov a Dead Sea halobacterium with a moderate saltrequirementrdquo Archives of Microbiology vol 104 no 1 pp 207ndash214 1975

[58] S D Nuttall and M L Dyall-Smith ldquoCh2 a novel halophilicarchaeon from an Australian solar salternrdquo International Jour-nal of Systematic Bacteriology vol 43 no 4 pp 729ndash734 1993

[59] P D Franzman E Stackebrandt K Sanderson et al ldquoHalobac-terium lacusprofundi sp nov a halophilic bacterium isolatedfrom Deep Lake Antarcticardquo Systematic and Applied Microbi-ology vol 11 pp 20ndash27 1988

[60] G A Tomlinson and L I Hochstein ldquoHalobacterium saccharo-vorum sp nov a carbohydrate metabolizing extremely halo-philic bacteriumrdquoCanadian Journal ofMicrobiology vol 22 no4 pp 587ndash591 1976

[61] A Ventosa M C Gutierrez M Kamekura and M L Dyall-Smith ldquoProposal to transfer Halococcus turkmenicus Halobac-terium trapanicum JCM9743 and strainGSL-11 toHaloterrigenaturkmenica gen nov comb novrdquo International Journal ofSystematic Bacteriology vol 49 no 1 pp 131ndash136 1999

[62] M Kamekura and M L Dyall-Smith ldquoTaxonomy of the familyHalobacteriaceae and the description of two new genera Halo-rubrobacterium and Natrialbardquo Journal of General and AppliedMicrobiology vol 41 no 4 pp 333ndash350 1995

[63] T J Carver KM RutherfordM BerrimanM A RajandreamB G Barrell and J Parkhill ldquoACT the Artemis comparisontoolrdquo Bioinformatics vol 21 no 16 pp 3422ndash3423 2005

[64] S W Cline W L Lam R L Charlebois L C Schalkwyk andW F Doolittle ldquoTransformationmethods for halophilic archae-bacteriardquo Canadian Journal of Microbiology vol 35 no 1 pp148ndash152 1989

10 Archaea

Table 5 Main differences (divergent regions) between the SH1 and PH1 genomes

RegionaSH1 startb SH1 stop Length (bp) PH1 start PH1 stop Length (bp) Comment

DV1 527 650 123 538 604 66

Replacement Both regions have direct repeats at their bordersPH1 has two sets GACCCGGC and CGCTGC while SH1 hasCCCGAC In SH1 this replacement region includes theC-terminal region of ORF2 and the N-terminal region ofORF3 In PH1 it covers only the C-terminal region of ORF1

DV2 2433 2588 155 2386 2387 mdash RMDc from PH1 SH1 repeat at border is TGACCGThisremoves the homolog of SH1 ORF10 from PH1

DV3 3324 3598 274 3137 3416 279 Replacement at the beginning of SH1 ORF13PH1 ORF12(capsid protein VP1)

DV4 6656 7128 472 6478 7085 607 Replacement near the C-terminus of SH1 ORF13PH1 ORF12(capsid protein VP1)

DV5 7491 8341 850 7446 7447 mdash

Indel that results in ORFs14-16 of SH1 not being present inPH1 SH1 ORF14 has a close homolog in HHIV-2 (ORF6) in asimilar position just after the VP1 homolog SH1 ORF15 is aconserved protein in haloarchaea (eg NP 2552A andHmuk 2978) and halovirus His1 (ORF13) Possible invertedrepeat (AGCCATG) at border of SH1 region

DV6 13705 13706 mdash 12818 12851 33RMD from SH1 PH1 repeat at border is CAGCGG(gt)G Thisremoves a part of capsid protein VP2 sequence from the SH1protein

DV7 13789 13863 74 12934 12941 7 Replacement This occurs within VP2 gene of both viruses

DV8 15142 15186 44 14220 14221 mdash RMD from PH1 SH1 repeat at border is TG(tc)CCGACGAand occurs within capsid protein VP2 gene of both viruses

DV9 15303 15317 14 14337 14338 mdash RMD from PH1 SH1 repeat at border is GCCGACGA andoccurs within capsid protein VP2 gene of both viruses

DV10 15504 15527 23 14529 14530 mdash RMD from PH1 SH1 repeat at border is GACGA and occurswithin capsid protein VP2 of both viruses

DV11 20026 20405 379 19019 19173 154Replacement beginning at the start of SH1 ORF35PH1 ORF31This region is longer in SH1 where it includes ORF36 an ORFthat has no homolog present in PH1

DV12 20440 20661 221 19208 19286 78 Replacement This is unequal and includes an ORF in SH1(ORF38) that is not present in PH1

DV13 20943 21085 142 19562 19728 166 Replacement This covers SH1 ORF40PH1 ORF34 Thepredicted proteins are not homologous

DV14 23541 23542 mdash 22171 22197 27 Probable RMD in SH1 PH1 repeat at border is CGTCTCGGand occurs in SH1 ORF44PH1 ORF38

DV15 24506 24521 15 23152 23153 mdash Probable RMD in PH1 SH1 repeat at border is CTCGGT andoccurs near the end of SH1 ORF45PH1 ORF39

DV16 24951 24964 13 23579 23580 mdash Probable RMD in PH1 SH1 repeat at border is CGGTC andoccurs in SH1 ORF47PH1 ORF41

DV17 26449 26450 mdash 25050 25274 224Probable RMD in SH1 PH1 repeat at border is TCATGCG andoccurs near the start of SH1 ORF50PH1 ORF44 and providesan extra ORF for PH1 (ORF45)

DV18 27074 29764 690 25895 26933 1038

Replacement Left border at start of SH1 ORF51PH1 ORF47and extends rightwards into SH1 ORF55PH1 ORF49 (VP18gene) It is an unequal replacement and SH1 has two moreORFs in this region than PH1

aDV Divergent regions between the genomes of SH1 and PH1bStart and stop positions refer to the GenBank sequences of the two viruses SH1 NC 0072171 PH1 KC252997cRMD repeat-mediated deletion event as described in [20]

The recently described halovirus SNJ1 carries manyORFsthat have homologs adjacent to a previously described ViPREofHmc mukohataei (from here on denoted by ViPREHmuk1)a locus that contains genes related to His2 (and other ple-olipoviruses) as well as toHaloferax plasmid pHK2 (probablya provirus) and to two small plasmids ofHqr walsbyi (sim6 kbpL6A and pL6B)The latter relationship provides wider linksto other viruses because one of the PL6 genes (Hqrw 6002)

has homologs in some pleolipoviruses (HPRV-3 and HGPV-1) while another gene (Hqrw 6005) is related to ORF16 ofthe spindle-shaped halovirus His1 [15] After adding the SNJ1gene homologs to ViPREHmuk1 it is extended significantlyand a close inspection of the flanking regions revealed atRNA-ala gene at one end and a partial copy of the sametRNA-ala gene (next to a phage integrase gene) at the otherend (supplementary Figure 1) Genes beyond these points

Archaea 11

Table 6 Transfection of haloarchaea by PH1 DNA

Speciesa Efficiency of transfectiontransformation withPH1 DNAb (PFU120583g of DNA) pUBP2 DNAbc (CFU120583g of DNA)

Har hispanica 53 plusmn 05 times 103 14 plusmn 08 times 104

Har marismortui 45 plusmn 07 times 103 18 plusmn 01 times 104

ldquoHar sinaiiensisrdquo 32 plusmn 06 times 102 mdashHalobacterium salinarum mdash 54 plusmn 42 times 103

Haloferax gibbonsii 28 plusmn 08 times 103 40 plusmn 26 times 103

Haloferax lucentense mdash 57 plusmn 11 times 103

Hfx volcanii mdash 21 plusmn 07 times 105

Hrr lacusprofundi 70 plusmn 22 times 102 95 plusmn 15 times 103

Haloterrigena turkmenica 16 plusmn 17 times 102 24 plusmn 06 times 103

Natrialba asiatica 35 plusmn 39 times 102 31 plusmn 10 times 104aOnly those species positive for transfection andor plasmid transformation are shownbRates of transfection by (nonprotease treated) PH1 DNA or transformation by plasmid pUB2 are averages of three independent experiments each performedin duplicate (plusmn standard deviation)cTransformants were selected on plates with 2 4 or 6120583gmL simvastatin depending on the strainmdash no plaques or colonies observed

appear to be related to cellular metabolism ViPREHmuk1is now 39379 nt in length flanked by a 59 nt direct repeat(potential att sites) and includes 54 ORFs many of which arerelated to known haloviruses are virus-like (eg integrasesmethyltransferases transcriptional regulators DNA methyl-transferase and DNA glycosylase) are homologs in or nearother knownViPREs or show little similarity to other knownproteins This locus also includes an ORC1CDC6 homolog(Hmuk 0446) which could provide a replication function

The mixture of very different virus and plasmid genesseen within ViPREHmuk1 is remarkable but it also revealshomologs found in or adjacent to genomic loci of otherspecies such as the gene homologs of Har marismortuiseen in the left end of ViPREHmuk1 (supplementary Figure1) When these genes are added to the previously describedHaloarcula locus (ViPREHmuk1) it is also extended signifi-cantly It becomes 25550 bp in length encompasses genesfrom Hmar 2382 to Hmar 2404 and is flanked by a fulltRNA-ala gene at one end and a partial copy at the other Itcarries an ORC1CDC6 homolog an integrase and a phiH-like repressor as well as pleolipovirus homologs

210 Transfection of Haloarchaea by PH1 DNA PH1 DNAwas introduced into Har hispanica cells using the PEGmethod [37] and the cells were screened for virus productionby plaque assay Figure 7 shows the increase in transfectedcells with input (nonproteinase K-treated) virus DNA Theestimated efficiencywas 53times 103 PFUper120583gDNA PH1DNAthat had been treated with proteinase K or with DNAase I(RNase-free) did not produce plaques (data not shown) PH1DNA was found to transfect six species of haloarchaea otherthanHar hispanica includingmembers of the generaHaloar-cula Haloferax Halorubrum Haloterrigena and Natrialba(Table 6) For these experiments the virus production fromtransfected cells was detected by plaque assay on indicatorlawns of Har hispanica

0

0

200 400 600 800 1000

Titre

(PFU

mL)

104

103

102

PH1 DNA (ng)

Figure 7 Transfection ofHar hispanica cells with PH1 DNA Vary-ing amounts of nonproteinase K-treated PH1 DNAwere introducedinto cells of Har hispanica using the PEG method [64] Cells werethen screened for infective centres by plaque assay Data shown is theaverage of three independent experiments performed in duplicateError bars represent one standard deviation of themean If protease-treated DNA was used no transfectant plaques were observed

3 Discussion

PH1 is very similar in particlemorphology genome structureand sequence to the previously described halovirus SH1 [816] Like SH1 it has long inverted terminal repeat sequencesand terminal proteins indicative of protein-primed replica-tion Purified PH1 particles have a relatively low buoyant

12 Archaea

density and are chloroform sensitive and show a layeredcapsid structure consistent with the likely presence of aninternalmembrane layer as has been shown for SH1HHIV-2and SNJ1 [13 26]The particle stability of PH1 to temperaturepH and reduced salt was similar to SH1 [8] The structuralproteins of PH1 are very similar in sequence and relativeabundance to those of SH1 so the two viruses are likely toshare the same particle geometry [16]

Viruses that use protein-primed replication such as bac-teriophageΦ29 carry a viral type BDNApolymerase that caninteract specifically with the proteins attached to the genomictermini and initiate strand synthesis [38] ArchaeovirusesHis1 [7] and Acidianus bottle-shaped virus [39] probablyuse this mode of replication However a polymerase genecannot be found in the genomes of PH1 SH1 or HHIV-2DNA polymerases are large enzymes and the only ORF ofSH1 that was long enough to encode such an enzyme andhad not previously been assigned as a structural protein wasORF55 which specified an 865 aa protein that contained noconserved domains indicative of polymerases [27]The recentstudy of HHIV-2 showed that the corresponding ORF in thisvirus (gene 42) is a structural protein of the virion and byinference this is likely also for the corresponding proteinsof SH1 and PH1 (no homolog is present in SNJ1) Withouta polymerase gene these viruses must use a host enzymebut searches for a viral type B DNA polymerase in thegenome of the hostHar hispanica or in the genomes of othersequenced haloarchaea did not find anymatchesThis arguesstrongly for a replication mechanism that is different to thatexemplified byΦ29 An attractive alternative is that displayedby Streptomyces linear plasmids which have 51015840-terminal pro-teins and use a cellular polymerase for replication [40] Theterminal proteins are not used for primary replication but forend patching [41] and it has been shown that if the term-inal repeat sequences are removed from these plasmids andthe ends ligated they can replicate as circular plasmids [42]This provides a testable hypothesis for the replication of halo-sphaeroviruses and also offers a pathway for switchingbetween linear (eg PH1) and circular (eg SNJ1) forms Useof a host polymerase would also fit with the ability of SH1 andPH1 DNA to transfect many different haloarchaeal species[31] as these enzymes and theirmode of action are highly con-served

The growth characteristics of PH1 inHar hispanica bothby single-cell burst and single-step growth experiments gavevalues of 50ndash100 virusescell significantly less than that ofSH1 (200 virusescell) [8] and HHIV-2 (180 virusescell) [13]The latter virus is lytic whereas SH1 PH1 and SNJ1 start pro-ducing extracellular viruswell before any decrease in cell den-sity indicating that virus release can occur without lysisThishas been most clearly shown with SH1 where almost 100of cells can be infected [8] Infection ofHar hispanica by PH1was less efficient (sim30 of cells) but the kinetics of virus pro-duction in single-step growth curves of SH1 and PH1 aresimilar with virus production occurring over several hoursrather than at a clearly defined time after infection The twoviruses are also very closely related and infect the same hostspecies so it is likely that they use the same scheme for cellexit For both viruses the cell density eventually decreases in

single-step cultures showing that virus infection does resultin cell death This mode of exit appears to maximise theproduction of the virus over time as the host cells survive foran extended period In natural hypersaline waters cell num-bers are usually high but growth rates are low [12] and thehigh incident UV (on often shallow ponds) would damagevirus DNA so reducing the half-life of released virus part-icles [43] These factors may have favoured the exit strategydisplayed by these viruses improving their chance of trans-mission to a new host

The motif GATC was absent in the genomes of PH1 SH1andHHIV-2 and themotif CTAGwas either absent or greatlyunderrepresented All three viruses share the same host butGATC and CTAG motifs are plentiful in the Har hispanicagenome (accessions CP002921 and CP002923) Avoidance ofthesemotifs in halovirusesHis1 His2 HF1 andHF2 has beenreported previously [7] Such purifying selection commonlyresults from host restriction enzymes and inHfx volcanii itis exactly these two motifs that are targeted [44] One Hfxvolcanii enzyme recognises A-methylated GATC sites [45]and another recognises un-methylated CTAG sites (whichare protected by methylation in the genome) In the currentstudy this could explain the negative transfection results forHfx volcanii as the PH1 genome contains two CTAGmotifsBy comparison the SH1 genome contains no CTAG motifsand is able to transfect Hfx volcanii [31]

Comparison of the PH1 and SH1 genomes allowed adetailed picture of the natural variation occurring betweenclosely related viruses revealing likely deletion events medi-ated by small repeats a process described previously in astudy of Hqr walsbyi and termed repeat-mediated deletion[20] There are also many replacements including blocks ofsequence that occur within long open reading frames (suchas the capsid protein genes VP1 and VP18) VP3 and VP6 areknown to form the large spikes on the external surface of theSH1 virion and presumably one or both interact with host cellreceptors [16] PH1 and SH1 have identical host ranges con-sistent with the high sequence similarity shown between theircorresponding VP3 and VP6 proteins

A previous study of SH1 could not detect transcriptsacross annotated ORFs 1ndash3 or 55-56 [30] ORFs 1 and 56occur in the terminal inverted repeat sequence and in thepresent study it was found that there were no ORFs cor-responding to these in the PH1 genome Given the close simi-larity of the two genomes the comparative data are in agree-ment with the transcriptional data indicating that these SH1ORFs are not used SH1 ORFs 2 and 3 are more problematicas good homologs of these are also present in PH1 The con-flict between the transcriptional data and the comparativegenomic evidence requires further experimental work toresolveThe status of SH1ORF55 has recently been confirmedby studies of the related virus HHIV-2 (discussed above)

The pleolipovirus group has expanded dramatically inthe last few years and it now comprises a diverse group ofviruseswith different genome types and replication strategiesWhat is evenmore remarkable is that a similar expansion cannow be seen with SH1-related viruses which includes viruseswith at least two replication modes and genome types (linearand circular) Evidence from haloarchaeal genome sequences

Archaea 13

show cases of not only provirus integrants or plasmids (egpKH2 and pHH205) but also genomic loci (ViPREs) thatcontain cassettes of virus genes from different sources andat least two cases (described in this study) have likely attsites that indicate circularisation and mobility While thereis a clear relationship between viruses with circular ds- andssDNA genomes that replicate via the rolling circle method(ie the ssDNA form is a replication intermediate) it ismore difficult to explain how capsid gene cassettes can movebetween these viruses and those with linear genomes andterminal proteins Within the pleolipoviruses His2 has alinear genome that contains a viral type B DNA polymeraseand terminal proteins while all the other described membershave circular genomes that contain a rep homolog and prob-ably replicate via the rolling circle method [14 19] Similarlyamong the halosphaeroviruses PH1 SH1 and HHIV-2 havelinear dsDNA with terminal proteins and probably replicatein the same way but the related SNJ1 virus has a circulardsDNA genome The clear relationships shown by the capsidgenes of viruses within each group plus the connectionsshown to haloviruses outside each group (eg the DNA poly-merases of His1 and His2) all speak of a vigorous meansof recombination one that can readily switch capsid genesbetween viruses with radically different replication strategiesHow is the process most likely to operate One possibility issuggested by ViPREs which appear to be mobile collectionsof capsid and replication genes from different sources Theyoffer fixed locations for recombination to occur providegene cassettes that can be reassorted to produce novel virusgenomes and some can probably recombine out as circularforms For example ViPREHmuk1 carries genes related topleolipoviruses halosphaeroviruses and other haloviruses Itis yet to be determined if any of the genes carried in theseloci are expressed in the cell (such as the genes for capsidproteins) or if ViPREs provide any selective advantage to thehost

4 Materials and Methods

41 Water Sample A water sample was collected in 1998from Pink Lake (32∘001015840 S and 115∘301015840 E) a hypersalinelake on Rottnest Island Western Australia Australia It wasscreened in the same year for haloviruses using the methodsdescribed in [8] A single plaque on a Har hispanica lawnplate was picked and replaque purified The novel haloviruswas designated PH1

42 Media Strains and Plasmids The media used in thisstudy are described in the online resource The Halohand-book (httpwwwhaloarchaeacomresourceshalohandbookindexhtml) Artificial salt water containing 30 (wv) totalsalts comprised of 4M NaCl 150mM MgCl

2 150mM

MgSO4 90mM KCl and 35mM CaCl

2and adjusted to pH

75 using sim2mL 1M Tris-HCl (pH 75) per litre MGM con-taining 12 18 or 23 (wv) total salts and HVD (halo-virus diluent) were prepared from the concentrated stock aspreviously described [46] Bacto-agar (Difco Laboratories)was added to MGM for solid (15 gL) or top-layer (7 gL)media

Table 1 lists the haloarchaea used in this study Allhaloarchaea were grown aerobically at 37∘C in either 18 or23 (wv)MGM(depending on the strain) andwith agitation(except for ldquoHar sinaiiensisrdquo)The plasmids used were pBlue-script II KS+ (Stratagene Cloning Systems) pUBP2 [47]and pWL102 [48] pUBP2 and pWL102 were first passagedthrough E coli JM110 [49] to prevent dam-methylationof DNA which has been shown to reduce transformationefficiency in some strains [45]

43 Negative-Stain Electron Microscopy The method fornegative-stain TEM was adapted from that of V Tarasovdescribed in the online resource The Halohandbook (httpwwwhaloarchaeacomresourceshalohandbookindexhtml)A 20120583L drop of the sample was placed on a clean surfaceand the virus particles were allowed to adsorb to a Formvarfilm 400-mesh copper grid (ProSciTech) for 15ndash2min Theywere then negatively stained with a 20120583L drop of 2 (wv)uranyl acetate for 1-2min Excess liquid was absorbed withfilter paper and the grid was allowed to air dry Grids wereexamined either on a Philips CM 120 BioTwin transmissionelectron microscope (Royal Philips Electronics) operating atan accelerating voltage of 120 kV or on a Siemens Elmiskop102 transmission electron microscope (Siemens AG) operat-ing at an accelerating voltage of 120 kV

44 Virus Host Range Twelve haloarchaeal strains from thegenera Haloarcula Halobacterium Haloferax HalorubrumandNatrialba (Table 1)Thirteen naturalHalorubrum isolates(H Camakaris unpublished data) and five uncharacterizedhaloarchaeal isolates fromLakeHardy Pink Lake (inWesternAustralia) and Serpentine Lake (D Walker and M K Seahunpublished data) were screened for PH1 susceptibilityLysates from PH1-infected Har hispanica cultures (1 times1011 PFUmL) were spotted onto lawns of each strain (usingmedia with 12 or 18 (wv)MGM depending on the strain)and incubated for 2ndash5 days at 30 and at 37∘C

45 Large Scale Virus Growth and Purification Liquid cul-tures of PH1 were grown by the infection (MOI 005) of anearly exponential Har hispanica culture in 18 (wv) MGMCultures were incubated aerobically at 37∘C with agitationfor 3 days Clearing (ie complete cell lysis) of PH1-infectedcultures did not occur and consequently cultures were harv-ested when the absorbance at 550 nm reached the minimumand the titre (determined by plaque assay) was at the maxi-mum usually sim1010ndash1011 PFUmL Virus was purified usingthe method described previously for SH1 [8 31] except thatafter the initial low speed spin (Sorvall GSA 6000 rpm30min 10∘C) virus was concentrated from the infected cul-ture by centrifugation at 26000 rpm (13 hr 10∘C) onto acushion of 30 (wv) sucrose in HVDThe pellet was resus-pended in a small volume of HVD and loaded onto a pre-formed linear 5ndash70 (wv) sucrose gradient followed byisopycnic centrifugation in 13 gmL CsCl (Beckman 70Ti60000 rpm 20 hr 10∘C) The white virus band occurredat a density of 129 gmL and was collected and diluted inhalovirus diluent (HVD see Section 4) and the virus pelleted(Beckman SW55 35000 rpm 75min 10∘C) and resuspended

14 Archaea

in a small volume of HVD and stored at 4∘C Virus recoveryat the major stages of a typical purification is given insupplementary Table 1 The specific infectivity of pure virussolutions was determined as the ratio of the PFUmL to theabsorbance at 260 nm

46 PH1 Single-Step Growth Curve An early exponentialphase culture of Har hispanica grown in 18 (wv) MGMwas infected with PH1 (MOI 50) Under these conditions thepercentage of infected cells was approximately 30 After anadsorption period of 1 hr at 37∘C the cells were washed threetimes with 18 (wv) MGM (at room temperature) resus-pended in 100mL 18 (wv) MGM (these methods ensuredthe removal of all residual virus) and incubated at 37∘Cwith shaking (100 rpm) Samples were removed at hourlyintervals for measurements of absorbance at 550 nm andthe number of infective centres Immediately after samplingtitres were determined by plaque assay with an indicator lawnof Har hispanica Each experiment was performed in tripli-cate

47 Halovirus Stability After various treatments sampleswere removed and diluted in HVD and virus titres weredetermined by plaque assay on Har hispanica Each experi-ment was performed in triplicate and representative data areshown Chloroform sensitivity was examined by the exposureof PH1 lysates in 18 (wv) MGM to chloroform in a volumeratio of 1 4 (chloroform to lysate) Incubation was at roomtemperature with constant agitation At appropriate timepoints the mix was allowed to settle and a sample wasremoved from the upper layer and diluted in HVDThe effectof a low ionic environment was examined by diluting PH1lysates (in 18 (wv) MGM) into double distilled H

2O in a

volume ratio of 1 1000 (lysate to double distilledH2O) Incu-

bation was at room temperature with constant agitationSamples were removed at various times and diluted in HVDThe pH stability of PH1 was determined by dilution of PH1lysates in 18 (wv)MGMin the appropriate pHbuffer (HVDbuffered with appropriate Tris-HCl) in a volume ratio of1 100 (lysate to buffer) Incubation was at room temperaturewith constant agitation After 30min samples were removedand diluted in HVD Thermal stability of PH1 was examinedby a 1 hr incubation of a virus lysate in 18 (wv) MGM atdifferent temperatures after which they were brought quicklyto room temperature diluted in HVD and titrated

48 Protein Procedures To remove salts purified virus prepa-rations were mixed with trichloroacetic acid (10 (vv) finalconcentration) and incubated on ice for 15min to allowthe proteins to precipitate After centrifugation (16000 g15min room temperature) the precipitate was washed threetimes in acetone dried and resuspended in double distilledH2O Proteins were dissolved in Laemmli sample buffer with

48mM 120573-mercaptoethanol [50] heated in boiling water for5min then separated on 12 (wv) NuPAGE Novex Bis-Tris Gels using MES-SDS running buffer according to themanufacturerrsquos directions (Invitrogen) After electrophoresisgels were rinsed in double distilled H

2O and stained with

01 (wv) Brilliant Blue G in 40 (vv) methanol and

10 (vv) acetic acid Gels were destained with severalchanges of 40 (vv) methanol and 10 (vv) acetic acidAlternatively gels were stained with GelCode GlycoproteinStaining Kit Pro-Q Emerald 488 Glycoprotein Gel and BlotStain Kit SYPRO Ruby Protein Gel Stain according tothe manufacturerrsquos directions (Invitrogen Molecular ProbesPierce Biotechnology)

Protein bands were cut from the gels and sent to theAustralian Proteome Analysis Facility (Macquarie Univer-sity) for trypsin digestion and analysis by matrix assistedlaser desorption ionisation time of flight mass spectrometry(MALDI-TOF MS) on an Applied Biosystems 4700 Pro-teomics Analyser (Applied Biosystems)

49 DNA Procedures Proteinase K-treated and nonpro-teinaseK-treatedDNApreparationsweremade frompurifiedPH1 using the methods described previously for SH1 [8]DNA was separated on 1 (wv) agarose gels in Tris-acetate-EDTA electrophoresis buffer and was stained with ethidiumbromide (Sigma-Aldrich)120582 DNA DNase I (RNase-free) exonuclease III mung

bean nuclease nuclease BAL-31 T4 DNA polymerase T4DNA ligase T7 exonuclease and type II restriction endonu-cleases were purchased from New England Biolabs RNase Aand yeast transfer RNA were purchased from Sigma-AldrichProteinase K was purchased from Promega Oligonucleotideprimers were purchased from Geneworks Australia

To clone fragments of the PH1 genome purified virusDNA was digested either with AccI Eco0109I and MseI orSmaI restriction endonucleases blunted with DNA Poly-merase I Large (Klenow) Fragment where appropriate andligated into AccI- Eco0109I- or SmaI-digested pBluescript IIKS+ using T4 DNA ligase according to the manufacturerrsquosinstructions (New England Biolabs) The DNA was intro-duced intoE coliXL1-Blue and transformants grown onLuriaagar containing 15 120583gmL tetracycline and 100 120583gmL ampi-cillinThe resulting cloneswere sequenced and the sequenceswere used to design specific oligonucleotide primers for PCRamplification andor primer walking using the virus genome(or specific restriction fragments) as templates

To amplify PH1 nucleic acid approximately 10 ng virusDNA was combined with 500 nM primers 100120583M of eachdNTP 1U Deep Vent DNA Polymerase (New England Bio-labs) and 1 timesThermoPol buffer (New England Biolabs) ina total reaction volume of 50 120583L Template was denatured at95∘C for 10min followed by 30 cycles of denaturation at 95∘Cfor 30 sec annealing at 56∘C for 30 sec extension at 75∘C for2min and a final extension at 75∘C for 10min PCR reactionswere performed on a PxE02ThermoCycler (Thermo ElectroCorporation) Sequencing reactions using 32 pmol primerwere performed by the dideoxy chain termination methodusing ABI PRISM Big Dye Terminator Mix version 31 onan ABI 3100 capillary sequencer (Applied Biosystems) bythe Applied Genetic Diagnostics Sequencing Service at theDepartment of Pathology (The University of Melbourne)

To obtain terminal genomic fragments PH1 DNA wasdigested by the enzyme Pas1 and the head (sim700 bp) andthe tail (sim1300 bp) fragments were purified by the QIAEXII gel extraction kits and treated with 05M piperidine for

Archaea 15

2 hr at 37∘C to remove protein residues The piperidine-treated fragments were sequenced by the Genomics BiotechCompany (Taipei Taiwan) using two primers TGACCA-ATTAATTAGGCCGGTTCGCC (PH1 head-R) and GTGC-CATACTGCTACAATTCT (PH1 tail-F)

Four hundred fifty-four whole genome sequencing PH1DNA samples (sim5 120583g) were sequenced using parallel pyrose-quencing on a Roche 454 Genome Sequencer System atMission Biotech (Taipei Taiwan) The largest contig was27399with 8803 reads (Newbler version 27) with 27387 posi-tions being of Q40 quality The average depth of coverage foreach base was 829TheGenBank accession for the entire PH1sequence is KC252997

Acknowledgments

The authors thank Carolyn Bath Mei Kwei (Joan) Seahand Danielle Walker for their early work in PH1 characteri-zation Simon Crawford Jocelyn Carpenter and Anna Fried-huber for their assistance with the transmission electronmicroscopy and Tiffany Cowie and Voula Kanellakis fortheir sequencing assistance This paper has been facilitatedby access to the Australian Proteome Analysis Facility estab-lished under the Australian Governmentrsquos Major NationalResearch Facilities program SLT was supported by a grantfrom the Biodiversity Research Center Academia SinicaTaiwan

References

[1] M Krupovic M F White P Forterre and D PrangishvilildquoPostcards from the edge structural genomics of archaealvirusesrdquo Advances in Virus Research vol 82 pp 33ndash62 2012

[2] P Stolt andW Zillig ldquoGene regulation in halophage 120601Hmdashmorethan promotersrdquo Systematic and Applied Microbiology vol 16no 4 pp 591ndash596 1994

[3] R Klein U Baranyi N Rossler B Greineder H Scholz and AWitte ldquoNatrialba magadii virus 120593Ch1 first complete nucleotidesequence and functional organization of a virus infecting ahaloalkaliphilic archaeonrdquo Molecular Microbiology vol 45 no3 pp 851ndash863 2002

[4] E Pagaling R D Haigh W D Grant et al ldquoSequence analysisof an Archaeal virus isolated from a hypersaline lake in InnerMongolia Chinardquo BMC Genomics vol 8 article 410 2007

[5] S L Tang SNuttall andMDyall-Smith ldquoHalovirusesHF1 andHF2 evidence for a recent and large recombination eventrdquo Jour-nal of Bacteriology vol 186 no 9 pp 2810ndash2817 2004

[6] C Bath and M L Dyall-smith ldquoHis1 an archaeal virus of theFuselloviridae family that infectsHaloarcula hispanicardquo Journalof Virology vol 72 no 11 pp 9392ndash9395 1998

[7] C Bath T Cukalac K Porter andM L Dyall-Smith ldquoHis1 andHis2 are distantly related spindle-shaped haloviruses belongingto the novel virus group Salterprovirusrdquo Virology vol 350 no1 pp 228ndash239 2006

[8] K Porter P Kukkaro J K H Bamford et al ldquoSH1 a novelspherical halovirus isolated from an Australian hypersalinelakerdquo Virology vol 335 no 1 pp 22ndash33 2005

[9] M Dyall-Smith S L Tang and C Bath ldquoHaloarchaeal viruseshow diverse are theyrdquo Research in Microbiology vol 154 no 4pp 309ndash313 2003

[10] A Oren G Bratbak and M Heldal ldquoOccurrence of virus-likeparticles in the Dead Seardquo Extremophiles vol 1 no 3 pp 143ndash149 1997

[11] T Sime-Ngando S Lucas A Robin et al et al ldquoDiversityof virus-host systems in hypersaline Lake Retba SenegalrdquoEnvironmental Microbiology vol 13 no 8 pp 1956ndash1972 2010

[12] N Guixa-Boixareu J I Calderon-Paz M Heldal G Bratbakand C Pedros-Alio ldquoViral lysis and bacterivory as prokaryoticloss factors along a salinity gradientrdquoAquaticMicrobial Ecologyvol 11 no 3 pp 215ndash227 1996

[13] S T Jaakkola R K Penttinen S T Vilen et al ldquoClosely relatedarchaeal Haloarcula hispanica icosahedral viruses HHIV-2 andSH1 have nonhomologous genes encoding host recognitionfunctionsrdquo Journal of Virology vol 86 no 9 pp 4734ndash47422012

[14] E Roine P Kukkaro L Paulin et al ldquoNew closely related halo-archaeal viral elements with different nucleic acid typesrdquo Jour-nal of Virology vol 84 no 7 pp 3682ndash3689 2010

[15] A Sencilo L Paulin S KellnerMHelm and E Roine ldquoRelatedhaloarchaeal pleomorphic viruses contain different genometypesrdquo Nucleic Acids Research vol 40 no 12 pp 5523ndash55342012

[16] H T Jaalinoja E Roine P Laurinmaki H M Kivela D HBamford and S J Butcher ldquoStructure and host-cell interactionof SH1 a membrane-containing halophilic euryarchaeal virusrdquoProceedings of the National Academy of Sciences of the UnitedStates of America vol 105 no 23 pp 8008ndash8013 2008

[17] M K Pietila N S Atanasova VManole et al ldquoVirion architec-ture unifies globally distributed pleolipoviruses infecting halo-philic archaeardquo Journal of Virology vol 86 no 9 pp 5067ndash50792012

[18] M L Holmes and M L Dyall-Smith ldquoA plasmid vector witha selectable marker for halophilic archaebacteriardquo Journal ofBacteriology vol 172 no 2 pp 756ndash761 1990

[19] M L Holmes F Pfeifer andM L Dyall-Smith ldquoAnalysis of thehalobacterial plasmid pHK2 minimal repliconrdquo Gene vol 153no 1 pp 117ndash121 1995

[20] M L Dyall-Smith F Pfeiffer K Klee et al ldquoHaloquadratumwalsbyi limited diversity in a Global Pondrdquo PLoS ONE vol 6no 6 Article ID e20968 2011

[21] S Casjens ldquoProphages and bacterial genomics what have welearned so farrdquoMolecular Microbiology vol 49 no 2 pp 277ndash300 2003

[22] R W Hendrix M C M Smith R N Burns M E Ford andG F Hatfull ldquoEvolutionary relationships among diverse bacte-riophages and prophages all the worldrsquos a phagerdquo Proceedings ofthe National Academy of Sciences of the United States of Americavol 96 no 5 pp 2192ndash2197 1999

[23] D Botstein ldquoA theory ofmodular evolution for bacteriophagesrdquoAnnals of the New York Academy of Sciences vol 354 pp 484ndash491 1980

[24] M Krupovic S Gribaldo D H Bamford and P Forterre ldquoTheevolutionary history of archaeal MCM helicases a case study ofvertical evolution combinedwithHitchhiking ofmobile geneticelementsrdquo Molecular Biology and Evolution vol 27 no 12 pp2716ndash2732 2010

[25] M Krupovic P Forterre and D H Bamford ldquoComparativeanalysis of the mosaic genomes of tailed archaeal viruses andproviruses suggests common themes for virion architecture andassembly with tailed viruses of bacteriardquo Journal of MolecularBiology vol 397 no 1 pp 144ndash160 2010

16 Archaea

[26] Z Zhang Y Liu S Wang et al et al ldquoTemperate membrane-containing halophilic archaeal virus SNJ1 has a circular dsDNAgenome identical to that of plasmid pHH205rdquoVirology vol 434no 2 pp 233ndash241 2012

[27] D H Bamford J J Ravantti G Ronnholm et al ldquoConstituentsof SH1 a novel lipid-containing virus infecting the halophiliceuryarchaeonHaloarcula hispanicardquo Journal of Virology vol 79no 14 pp 9097ndash9107 2005

[28] H M Kivela E Roine P Kukkaro S Laurinavicius P Somer-harju andDH Bamford ldquoQuantitative dissociation of archaealvirus SH1 reveals distinct capsid proteins and a lipid corerdquoVirology vol 356 no 1-2 pp 4ndash11 2006

[29] H Jaalinoja Electron Cryo-Microscopy Studies of Bacteriophage8 and Archaeal Virus SH1 University of Helsinki HelsinkiFinland 2007

[30] K Porter B E Russ J Yang andM L Dyall-Smith ldquoThe trans-cription programme of the protein-primed halovirus SH1rdquoMi-crobiology vol 154 no 11 pp 3599ndash3608 2008

[31] K Porter and M L Dyall-Smith ldquoTransfection of haloarchaeaby the DNAs of spindle and round haloviruses and the useof transposon mutagenesis to identify non-essential regionsrdquoMolecular Microbiology vol 70 no 5 pp 1236ndash1245 2008

[32] S E Luria General Virology John Wiley and Sons New YorkNY USA 1953

[33] C A Thomas K Saigo E McLeod and J Ito ldquoThe separationofDNA segments attached to proteinsrdquoAnalytical Biochemistryvol 93 pp 158ndash166 1979

[34] A L Delcher D Harmon S Kasif OWhite and S L SalzbergldquoImproved microbial gene identification with GLIMMERrdquoNucleic Acids Research vol 27 no 23 pp 4636ndash4641 1999

[35] R Zhang Y Yang P Fang et al ldquoDiversity of telomere palin-dromic sequences and replication genes among Streptomyceslinear plasmidsrdquo Applied and Environmental Microbiology vol72 no 9 pp 5728ndash5733 2006

[36] X Wang H S Lee F J Sugar F E Jenney M W W Adamsand J H Prestegard ldquoPF0610 a novel winged helix-turn-helix variant possessing a rubredoxin-like Zn ribbon motiffrom the hyperthermophilic archaeon Pyrococcus furiosusrdquoBiochemistry vol 46 no 3 pp 752ndash761 2007

[37] S W Cline L C Schalkwyk and W F Doolittle ldquoTransfor-mation of the archaebacterium Halobacterium volcanii withgenomic DNArdquo Journal of Bacteriology vol 171 no 9 pp 4987ndash4991 1989

[38] S Kamtekar A J Berman J Wang et al ldquoThe 12059329 DNA poly-merase protein-primer structure suggests a model for the ini-tiation to elongation transitionrdquo EMBO Journal vol 25 no 6pp 1335ndash1343 2006

[39] X Peng T Basta M Haring R A Garrett and D PrangishvilildquoGenome of theAcidianus bottle-shaped virus and insights intothe replication and packaging mechanismsrdquo Virology vol 364no 1 pp 237ndash243 2007

[40] C C Yang C H Huang C Y Li Y G Tsay S C Lee and CW Chen ldquoThe terminal proteins of linear Streptomyces chrom-osomes and plasmids a novel class of replication primingproteinsrdquo Molecular Microbiology vol 43 no 2 pp 297ndash3052002

[41] C C Yang Y H Chen H H Tsai C H Huang T W Huangand C W Chen ldquoIn vitro deoxynucleotidylation of the term-inal protein of Streptomyces linear chromosomesrdquo Applied andEnvironmental Microbiology vol 72 no 12 pp 7959ndash79612006

[42] P C Chang and S N Cohen ldquoBidirectional replication froman internal origin in a linear Streptomyces plasmidrdquo Science vol265 no 5174 pp 952ndash954 1994

[43] S W Wilhelm W H Jeffrey C A Suttle and D L MitchellldquoEstimation of biologically damaging UV levels in marinesurface waters with DNA and viral dosimetersrdquo Photochemistryand Photobiology vol 76 no 3 pp 268ndash273 2002

[44] T Allers and M Mevarech ldquoArchaeal geneticsmdashthe third wayrdquoNature Reviews Genetics vol 6 no 1 pp 58ndash73 2005

[45] M L Holmes S D Nuttall and M L Dyall-Smith ldquoConstruc-tion and use of halobacterial shuttle vectors and further studieson Haloferax DNA gyraserdquo Journal of Bacteriology vol 173 no12 pp 3807ndash3813 1991

[46] S D Nuttall and M L Dyall-Smith ldquoHF1 and HF2 novelbacteriophages of halophilic archaeardquo Virology vol 197 no 2pp 678ndash684 1993

[47] U Blaseio and F Pfeifer ldquoTransformation of Halobacteriumhalobium development of vectors and investigation of gas vesi-cle synthesisrdquo Proceedings of the National Academy of Sciences ofthe United States of America vol 87 no 17 pp 6772ndash6776 1990

[48] W L Lam and W F Doolittle ldquoShuttle vectors for the archae-bacterium Halobacterium volcaniirdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 86 no14 pp 5478ndash5482 1989

[49] C Yanisch-Perron J Vieira and J Messing ldquoImproved M13phage cloning vectors and host strains nucleotide sequences ofthe M13mp18 and pUC19 vectorsrdquo Gene vol 33 no 1 pp 103ndash119 1985

[50] U K Laemmli ldquoCleavage of structural proteins during theassembly of the head of bacteriophage T4rdquo Nature vol 227 no5259 pp 680ndash685 1970

[51] D G Burns H M Camakaris P H Janssen and M L Dyall-Smith ldquoCombined use of cultivation-dependent and culti-vation-independent methods indicates that members of mosthaloarchaeal groups in anAustralian crystallizer pond are culti-vablerdquo Applied and Environmental Microbiology vol 70 no 9pp 5258ndash5265 2004

[52] G Juez F Rodriguez-Valera A Ventosa and D J KushnerldquoHaloarcula hispanica spec nov and Haloferax gibbonsii specnov two new species of extremely halophilic archaebacteriardquoSystematic and Applied Microbiology vol 8 pp 75ndash79 1986

[53] A Oren M Ginzburg B Z Ginzburg L I Hochstein and BE Volcani ldquoHaloarcula marismortui (Volcani) sp nov nomrev an extremely halophilic bacterium from the Dead SeardquoInternational Journal of Systematic Bacteriology vol 40 no 2pp 209ndash210 1990

[54] M Torreblanca F Rodriguez-Valera G Juez A Ventosa MKamekura and M Kates ldquoClassification of non-alkaliphilichalobacteria based on numerical taxonomy and polar lipidcomposition and description of Haloarcula gen nov andHaloferax gen novrdquo Systematic and Applied Microbiology vol8 pp 89ndash99 1986

[55] A Ventosa and A Oren ldquoHalobacterium salinarum nom cor-rig a name to replace Halobacterium salinarium (Elazari-Vol-cani) and to include Halobacterium halobium and Halobac-terium cutirubrumrdquo International Journal of Systematic Bacte-riology vol 46 no 1 p 347 1996

[56] M C Gutierrez M Kamekura M L Holmes M L Dyall-Smith and A Ventosa ldquoTaxonomic characterization of Halo-ferax sp (ldquoH alicanteirdquo) strain Aa 22 description of Haloferaxlucentensis sp novrdquo Extremophiles vol 6 no 6 pp 479ndash4832002

Archaea 17

[57] M F Mullakhanbhai and H Larsen ldquoHalobacterium volcaniispec nov a Dead Sea halobacterium with a moderate saltrequirementrdquo Archives of Microbiology vol 104 no 1 pp 207ndash214 1975

[58] S D Nuttall and M L Dyall-Smith ldquoCh2 a novel halophilicarchaeon from an Australian solar salternrdquo International Jour-nal of Systematic Bacteriology vol 43 no 4 pp 729ndash734 1993

[59] P D Franzman E Stackebrandt K Sanderson et al ldquoHalobac-terium lacusprofundi sp nov a halophilic bacterium isolatedfrom Deep Lake Antarcticardquo Systematic and Applied Microbi-ology vol 11 pp 20ndash27 1988

[60] G A Tomlinson and L I Hochstein ldquoHalobacterium saccharo-vorum sp nov a carbohydrate metabolizing extremely halo-philic bacteriumrdquoCanadian Journal ofMicrobiology vol 22 no4 pp 587ndash591 1976

[61] A Ventosa M C Gutierrez M Kamekura and M L Dyall-Smith ldquoProposal to transfer Halococcus turkmenicus Halobac-terium trapanicum JCM9743 and strainGSL-11 toHaloterrigenaturkmenica gen nov comb novrdquo International Journal ofSystematic Bacteriology vol 49 no 1 pp 131ndash136 1999

[62] M Kamekura and M L Dyall-Smith ldquoTaxonomy of the familyHalobacteriaceae and the description of two new genera Halo-rubrobacterium and Natrialbardquo Journal of General and AppliedMicrobiology vol 41 no 4 pp 333ndash350 1995

[63] T J Carver KM RutherfordM BerrimanM A RajandreamB G Barrell and J Parkhill ldquoACT the Artemis comparisontoolrdquo Bioinformatics vol 21 no 16 pp 3422ndash3423 2005

[64] S W Cline W L Lam R L Charlebois L C Schalkwyk andW F Doolittle ldquoTransformationmethods for halophilic archae-bacteriardquo Canadian Journal of Microbiology vol 35 no 1 pp148ndash152 1989

Archaea 11

Table 6 Transfection of haloarchaea by PH1 DNA

Speciesa Efficiency of transfectiontransformation withPH1 DNAb (PFU120583g of DNA) pUBP2 DNAbc (CFU120583g of DNA)

Har hispanica 53 plusmn 05 times 103 14 plusmn 08 times 104

Har marismortui 45 plusmn 07 times 103 18 plusmn 01 times 104

ldquoHar sinaiiensisrdquo 32 plusmn 06 times 102 mdashHalobacterium salinarum mdash 54 plusmn 42 times 103

Haloferax gibbonsii 28 plusmn 08 times 103 40 plusmn 26 times 103

Haloferax lucentense mdash 57 plusmn 11 times 103

Hfx volcanii mdash 21 plusmn 07 times 105

Hrr lacusprofundi 70 plusmn 22 times 102 95 plusmn 15 times 103

Haloterrigena turkmenica 16 plusmn 17 times 102 24 plusmn 06 times 103

Natrialba asiatica 35 plusmn 39 times 102 31 plusmn 10 times 104aOnly those species positive for transfection andor plasmid transformation are shownbRates of transfection by (nonprotease treated) PH1 DNA or transformation by plasmid pUB2 are averages of three independent experiments each performedin duplicate (plusmn standard deviation)cTransformants were selected on plates with 2 4 or 6120583gmL simvastatin depending on the strainmdash no plaques or colonies observed

appear to be related to cellular metabolism ViPREHmuk1is now 39379 nt in length flanked by a 59 nt direct repeat(potential att sites) and includes 54 ORFs many of which arerelated to known haloviruses are virus-like (eg integrasesmethyltransferases transcriptional regulators DNA methyl-transferase and DNA glycosylase) are homologs in or nearother knownViPREs or show little similarity to other knownproteins This locus also includes an ORC1CDC6 homolog(Hmuk 0446) which could provide a replication function

The mixture of very different virus and plasmid genesseen within ViPREHmuk1 is remarkable but it also revealshomologs found in or adjacent to genomic loci of otherspecies such as the gene homologs of Har marismortuiseen in the left end of ViPREHmuk1 (supplementary Figure1) When these genes are added to the previously describedHaloarcula locus (ViPREHmuk1) it is also extended signifi-cantly It becomes 25550 bp in length encompasses genesfrom Hmar 2382 to Hmar 2404 and is flanked by a fulltRNA-ala gene at one end and a partial copy at the other Itcarries an ORC1CDC6 homolog an integrase and a phiH-like repressor as well as pleolipovirus homologs

210 Transfection of Haloarchaea by PH1 DNA PH1 DNAwas introduced into Har hispanica cells using the PEGmethod [37] and the cells were screened for virus productionby plaque assay Figure 7 shows the increase in transfectedcells with input (nonproteinase K-treated) virus DNA Theestimated efficiencywas 53times 103 PFUper120583gDNA PH1DNAthat had been treated with proteinase K or with DNAase I(RNase-free) did not produce plaques (data not shown) PH1DNA was found to transfect six species of haloarchaea otherthanHar hispanica includingmembers of the generaHaloar-cula Haloferax Halorubrum Haloterrigena and Natrialba(Table 6) For these experiments the virus production fromtransfected cells was detected by plaque assay on indicatorlawns of Har hispanica

0

0

200 400 600 800 1000

Titre

(PFU

mL)

104

103

102

PH1 DNA (ng)

Figure 7 Transfection ofHar hispanica cells with PH1 DNA Vary-ing amounts of nonproteinase K-treated PH1 DNAwere introducedinto cells of Har hispanica using the PEG method [64] Cells werethen screened for infective centres by plaque assay Data shown is theaverage of three independent experiments performed in duplicateError bars represent one standard deviation of themean If protease-treated DNA was used no transfectant plaques were observed

3 Discussion

PH1 is very similar in particlemorphology genome structureand sequence to the previously described halovirus SH1 [816] Like SH1 it has long inverted terminal repeat sequencesand terminal proteins indicative of protein-primed replica-tion Purified PH1 particles have a relatively low buoyant

12 Archaea

density and are chloroform sensitive and show a layeredcapsid structure consistent with the likely presence of aninternalmembrane layer as has been shown for SH1HHIV-2and SNJ1 [13 26]The particle stability of PH1 to temperaturepH and reduced salt was similar to SH1 [8] The structuralproteins of PH1 are very similar in sequence and relativeabundance to those of SH1 so the two viruses are likely toshare the same particle geometry [16]

Viruses that use protein-primed replication such as bac-teriophageΦ29 carry a viral type BDNApolymerase that caninteract specifically with the proteins attached to the genomictermini and initiate strand synthesis [38] ArchaeovirusesHis1 [7] and Acidianus bottle-shaped virus [39] probablyuse this mode of replication However a polymerase genecannot be found in the genomes of PH1 SH1 or HHIV-2DNA polymerases are large enzymes and the only ORF ofSH1 that was long enough to encode such an enzyme andhad not previously been assigned as a structural protein wasORF55 which specified an 865 aa protein that contained noconserved domains indicative of polymerases [27]The recentstudy of HHIV-2 showed that the corresponding ORF in thisvirus (gene 42) is a structural protein of the virion and byinference this is likely also for the corresponding proteinsof SH1 and PH1 (no homolog is present in SNJ1) Withouta polymerase gene these viruses must use a host enzymebut searches for a viral type B DNA polymerase in thegenome of the hostHar hispanica or in the genomes of othersequenced haloarchaea did not find anymatchesThis arguesstrongly for a replication mechanism that is different to thatexemplified byΦ29 An attractive alternative is that displayedby Streptomyces linear plasmids which have 51015840-terminal pro-teins and use a cellular polymerase for replication [40] Theterminal proteins are not used for primary replication but forend patching [41] and it has been shown that if the term-inal repeat sequences are removed from these plasmids andthe ends ligated they can replicate as circular plasmids [42]This provides a testable hypothesis for the replication of halo-sphaeroviruses and also offers a pathway for switchingbetween linear (eg PH1) and circular (eg SNJ1) forms Useof a host polymerase would also fit with the ability of SH1 andPH1 DNA to transfect many different haloarchaeal species[31] as these enzymes and theirmode of action are highly con-served

The growth characteristics of PH1 inHar hispanica bothby single-cell burst and single-step growth experiments gavevalues of 50ndash100 virusescell significantly less than that ofSH1 (200 virusescell) [8] and HHIV-2 (180 virusescell) [13]The latter virus is lytic whereas SH1 PH1 and SNJ1 start pro-ducing extracellular viruswell before any decrease in cell den-sity indicating that virus release can occur without lysisThishas been most clearly shown with SH1 where almost 100of cells can be infected [8] Infection ofHar hispanica by PH1was less efficient (sim30 of cells) but the kinetics of virus pro-duction in single-step growth curves of SH1 and PH1 aresimilar with virus production occurring over several hoursrather than at a clearly defined time after infection The twoviruses are also very closely related and infect the same hostspecies so it is likely that they use the same scheme for cellexit For both viruses the cell density eventually decreases in

single-step cultures showing that virus infection does resultin cell death This mode of exit appears to maximise theproduction of the virus over time as the host cells survive foran extended period In natural hypersaline waters cell num-bers are usually high but growth rates are low [12] and thehigh incident UV (on often shallow ponds) would damagevirus DNA so reducing the half-life of released virus part-icles [43] These factors may have favoured the exit strategydisplayed by these viruses improving their chance of trans-mission to a new host

The motif GATC was absent in the genomes of PH1 SH1andHHIV-2 and themotif CTAGwas either absent or greatlyunderrepresented All three viruses share the same host butGATC and CTAG motifs are plentiful in the Har hispanicagenome (accessions CP002921 and CP002923) Avoidance ofthesemotifs in halovirusesHis1 His2 HF1 andHF2 has beenreported previously [7] Such purifying selection commonlyresults from host restriction enzymes and inHfx volcanii itis exactly these two motifs that are targeted [44] One Hfxvolcanii enzyme recognises A-methylated GATC sites [45]and another recognises un-methylated CTAG sites (whichare protected by methylation in the genome) In the currentstudy this could explain the negative transfection results forHfx volcanii as the PH1 genome contains two CTAGmotifsBy comparison the SH1 genome contains no CTAG motifsand is able to transfect Hfx volcanii [31]

Comparison of the PH1 and SH1 genomes allowed adetailed picture of the natural variation occurring betweenclosely related viruses revealing likely deletion events medi-ated by small repeats a process described previously in astudy of Hqr walsbyi and termed repeat-mediated deletion[20] There are also many replacements including blocks ofsequence that occur within long open reading frames (suchas the capsid protein genes VP1 and VP18) VP3 and VP6 areknown to form the large spikes on the external surface of theSH1 virion and presumably one or both interact with host cellreceptors [16] PH1 and SH1 have identical host ranges con-sistent with the high sequence similarity shown between theircorresponding VP3 and VP6 proteins

A previous study of SH1 could not detect transcriptsacross annotated ORFs 1ndash3 or 55-56 [30] ORFs 1 and 56occur in the terminal inverted repeat sequence and in thepresent study it was found that there were no ORFs cor-responding to these in the PH1 genome Given the close simi-larity of the two genomes the comparative data are in agree-ment with the transcriptional data indicating that these SH1ORFs are not used SH1 ORFs 2 and 3 are more problematicas good homologs of these are also present in PH1 The con-flict between the transcriptional data and the comparativegenomic evidence requires further experimental work toresolveThe status of SH1ORF55 has recently been confirmedby studies of the related virus HHIV-2 (discussed above)

The pleolipovirus group has expanded dramatically inthe last few years and it now comprises a diverse group ofviruseswith different genome types and replication strategiesWhat is evenmore remarkable is that a similar expansion cannow be seen with SH1-related viruses which includes viruseswith at least two replication modes and genome types (linearand circular) Evidence from haloarchaeal genome sequences

Archaea 13

show cases of not only provirus integrants or plasmids (egpKH2 and pHH205) but also genomic loci (ViPREs) thatcontain cassettes of virus genes from different sources andat least two cases (described in this study) have likely attsites that indicate circularisation and mobility While thereis a clear relationship between viruses with circular ds- andssDNA genomes that replicate via the rolling circle method(ie the ssDNA form is a replication intermediate) it ismore difficult to explain how capsid gene cassettes can movebetween these viruses and those with linear genomes andterminal proteins Within the pleolipoviruses His2 has alinear genome that contains a viral type B DNA polymeraseand terminal proteins while all the other described membershave circular genomes that contain a rep homolog and prob-ably replicate via the rolling circle method [14 19] Similarlyamong the halosphaeroviruses PH1 SH1 and HHIV-2 havelinear dsDNA with terminal proteins and probably replicatein the same way but the related SNJ1 virus has a circulardsDNA genome The clear relationships shown by the capsidgenes of viruses within each group plus the connectionsshown to haloviruses outside each group (eg the DNA poly-merases of His1 and His2) all speak of a vigorous meansof recombination one that can readily switch capsid genesbetween viruses with radically different replication strategiesHow is the process most likely to operate One possibility issuggested by ViPREs which appear to be mobile collectionsof capsid and replication genes from different sources Theyoffer fixed locations for recombination to occur providegene cassettes that can be reassorted to produce novel virusgenomes and some can probably recombine out as circularforms For example ViPREHmuk1 carries genes related topleolipoviruses halosphaeroviruses and other haloviruses Itis yet to be determined if any of the genes carried in theseloci are expressed in the cell (such as the genes for capsidproteins) or if ViPREs provide any selective advantage to thehost

4 Materials and Methods

41 Water Sample A water sample was collected in 1998from Pink Lake (32∘001015840 S and 115∘301015840 E) a hypersalinelake on Rottnest Island Western Australia Australia It wasscreened in the same year for haloviruses using the methodsdescribed in [8] A single plaque on a Har hispanica lawnplate was picked and replaque purified The novel haloviruswas designated PH1

42 Media Strains and Plasmids The media used in thisstudy are described in the online resource The Halohand-book (httpwwwhaloarchaeacomresourceshalohandbookindexhtml) Artificial salt water containing 30 (wv) totalsalts comprised of 4M NaCl 150mM MgCl

2 150mM

MgSO4 90mM KCl and 35mM CaCl

2and adjusted to pH

75 using sim2mL 1M Tris-HCl (pH 75) per litre MGM con-taining 12 18 or 23 (wv) total salts and HVD (halo-virus diluent) were prepared from the concentrated stock aspreviously described [46] Bacto-agar (Difco Laboratories)was added to MGM for solid (15 gL) or top-layer (7 gL)media

Table 1 lists the haloarchaea used in this study Allhaloarchaea were grown aerobically at 37∘C in either 18 or23 (wv)MGM(depending on the strain) andwith agitation(except for ldquoHar sinaiiensisrdquo)The plasmids used were pBlue-script II KS+ (Stratagene Cloning Systems) pUBP2 [47]and pWL102 [48] pUBP2 and pWL102 were first passagedthrough E coli JM110 [49] to prevent dam-methylationof DNA which has been shown to reduce transformationefficiency in some strains [45]

43 Negative-Stain Electron Microscopy The method fornegative-stain TEM was adapted from that of V Tarasovdescribed in the online resource The Halohandbook (httpwwwhaloarchaeacomresourceshalohandbookindexhtml)A 20120583L drop of the sample was placed on a clean surfaceand the virus particles were allowed to adsorb to a Formvarfilm 400-mesh copper grid (ProSciTech) for 15ndash2min Theywere then negatively stained with a 20120583L drop of 2 (wv)uranyl acetate for 1-2min Excess liquid was absorbed withfilter paper and the grid was allowed to air dry Grids wereexamined either on a Philips CM 120 BioTwin transmissionelectron microscope (Royal Philips Electronics) operating atan accelerating voltage of 120 kV or on a Siemens Elmiskop102 transmission electron microscope (Siemens AG) operat-ing at an accelerating voltage of 120 kV

44 Virus Host Range Twelve haloarchaeal strains from thegenera Haloarcula Halobacterium Haloferax HalorubrumandNatrialba (Table 1)Thirteen naturalHalorubrum isolates(H Camakaris unpublished data) and five uncharacterizedhaloarchaeal isolates fromLakeHardy Pink Lake (inWesternAustralia) and Serpentine Lake (D Walker and M K Seahunpublished data) were screened for PH1 susceptibilityLysates from PH1-infected Har hispanica cultures (1 times1011 PFUmL) were spotted onto lawns of each strain (usingmedia with 12 or 18 (wv)MGM depending on the strain)and incubated for 2ndash5 days at 30 and at 37∘C

45 Large Scale Virus Growth and Purification Liquid cul-tures of PH1 were grown by the infection (MOI 005) of anearly exponential Har hispanica culture in 18 (wv) MGMCultures were incubated aerobically at 37∘C with agitationfor 3 days Clearing (ie complete cell lysis) of PH1-infectedcultures did not occur and consequently cultures were harv-ested when the absorbance at 550 nm reached the minimumand the titre (determined by plaque assay) was at the maxi-mum usually sim1010ndash1011 PFUmL Virus was purified usingthe method described previously for SH1 [8 31] except thatafter the initial low speed spin (Sorvall GSA 6000 rpm30min 10∘C) virus was concentrated from the infected cul-ture by centrifugation at 26000 rpm (13 hr 10∘C) onto acushion of 30 (wv) sucrose in HVDThe pellet was resus-pended in a small volume of HVD and loaded onto a pre-formed linear 5ndash70 (wv) sucrose gradient followed byisopycnic centrifugation in 13 gmL CsCl (Beckman 70Ti60000 rpm 20 hr 10∘C) The white virus band occurredat a density of 129 gmL and was collected and diluted inhalovirus diluent (HVD see Section 4) and the virus pelleted(Beckman SW55 35000 rpm 75min 10∘C) and resuspended

14 Archaea

in a small volume of HVD and stored at 4∘C Virus recoveryat the major stages of a typical purification is given insupplementary Table 1 The specific infectivity of pure virussolutions was determined as the ratio of the PFUmL to theabsorbance at 260 nm

46 PH1 Single-Step Growth Curve An early exponentialphase culture of Har hispanica grown in 18 (wv) MGMwas infected with PH1 (MOI 50) Under these conditions thepercentage of infected cells was approximately 30 After anadsorption period of 1 hr at 37∘C the cells were washed threetimes with 18 (wv) MGM (at room temperature) resus-pended in 100mL 18 (wv) MGM (these methods ensuredthe removal of all residual virus) and incubated at 37∘Cwith shaking (100 rpm) Samples were removed at hourlyintervals for measurements of absorbance at 550 nm andthe number of infective centres Immediately after samplingtitres were determined by plaque assay with an indicator lawnof Har hispanica Each experiment was performed in tripli-cate

47 Halovirus Stability After various treatments sampleswere removed and diluted in HVD and virus titres weredetermined by plaque assay on Har hispanica Each experi-ment was performed in triplicate and representative data areshown Chloroform sensitivity was examined by the exposureof PH1 lysates in 18 (wv) MGM to chloroform in a volumeratio of 1 4 (chloroform to lysate) Incubation was at roomtemperature with constant agitation At appropriate timepoints the mix was allowed to settle and a sample wasremoved from the upper layer and diluted in HVDThe effectof a low ionic environment was examined by diluting PH1lysates (in 18 (wv) MGM) into double distilled H

2O in a

volume ratio of 1 1000 (lysate to double distilledH2O) Incu-

bation was at room temperature with constant agitationSamples were removed at various times and diluted in HVDThe pH stability of PH1 was determined by dilution of PH1lysates in 18 (wv)MGMin the appropriate pHbuffer (HVDbuffered with appropriate Tris-HCl) in a volume ratio of1 100 (lysate to buffer) Incubation was at room temperaturewith constant agitation After 30min samples were removedand diluted in HVD Thermal stability of PH1 was examinedby a 1 hr incubation of a virus lysate in 18 (wv) MGM atdifferent temperatures after which they were brought quicklyto room temperature diluted in HVD and titrated

48 Protein Procedures To remove salts purified virus prepa-rations were mixed with trichloroacetic acid (10 (vv) finalconcentration) and incubated on ice for 15min to allowthe proteins to precipitate After centrifugation (16000 g15min room temperature) the precipitate was washed threetimes in acetone dried and resuspended in double distilledH2O Proteins were dissolved in Laemmli sample buffer with

48mM 120573-mercaptoethanol [50] heated in boiling water for5min then separated on 12 (wv) NuPAGE Novex Bis-Tris Gels using MES-SDS running buffer according to themanufacturerrsquos directions (Invitrogen) After electrophoresisgels were rinsed in double distilled H

2O and stained with

01 (wv) Brilliant Blue G in 40 (vv) methanol and

10 (vv) acetic acid Gels were destained with severalchanges of 40 (vv) methanol and 10 (vv) acetic acidAlternatively gels were stained with GelCode GlycoproteinStaining Kit Pro-Q Emerald 488 Glycoprotein Gel and BlotStain Kit SYPRO Ruby Protein Gel Stain according tothe manufacturerrsquos directions (Invitrogen Molecular ProbesPierce Biotechnology)

Protein bands were cut from the gels and sent to theAustralian Proteome Analysis Facility (Macquarie Univer-sity) for trypsin digestion and analysis by matrix assistedlaser desorption ionisation time of flight mass spectrometry(MALDI-TOF MS) on an Applied Biosystems 4700 Pro-teomics Analyser (Applied Biosystems)

49 DNA Procedures Proteinase K-treated and nonpro-teinaseK-treatedDNApreparationsweremade frompurifiedPH1 using the methods described previously for SH1 [8]DNA was separated on 1 (wv) agarose gels in Tris-acetate-EDTA electrophoresis buffer and was stained with ethidiumbromide (Sigma-Aldrich)120582 DNA DNase I (RNase-free) exonuclease III mung

bean nuclease nuclease BAL-31 T4 DNA polymerase T4DNA ligase T7 exonuclease and type II restriction endonu-cleases were purchased from New England Biolabs RNase Aand yeast transfer RNA were purchased from Sigma-AldrichProteinase K was purchased from Promega Oligonucleotideprimers were purchased from Geneworks Australia

To clone fragments of the PH1 genome purified virusDNA was digested either with AccI Eco0109I and MseI orSmaI restriction endonucleases blunted with DNA Poly-merase I Large (Klenow) Fragment where appropriate andligated into AccI- Eco0109I- or SmaI-digested pBluescript IIKS+ using T4 DNA ligase according to the manufacturerrsquosinstructions (New England Biolabs) The DNA was intro-duced intoE coliXL1-Blue and transformants grown onLuriaagar containing 15 120583gmL tetracycline and 100 120583gmL ampi-cillinThe resulting cloneswere sequenced and the sequenceswere used to design specific oligonucleotide primers for PCRamplification andor primer walking using the virus genome(or specific restriction fragments) as templates

To amplify PH1 nucleic acid approximately 10 ng virusDNA was combined with 500 nM primers 100120583M of eachdNTP 1U Deep Vent DNA Polymerase (New England Bio-labs) and 1 timesThermoPol buffer (New England Biolabs) ina total reaction volume of 50 120583L Template was denatured at95∘C for 10min followed by 30 cycles of denaturation at 95∘Cfor 30 sec annealing at 56∘C for 30 sec extension at 75∘C for2min and a final extension at 75∘C for 10min PCR reactionswere performed on a PxE02ThermoCycler (Thermo ElectroCorporation) Sequencing reactions using 32 pmol primerwere performed by the dideoxy chain termination methodusing ABI PRISM Big Dye Terminator Mix version 31 onan ABI 3100 capillary sequencer (Applied Biosystems) bythe Applied Genetic Diagnostics Sequencing Service at theDepartment of Pathology (The University of Melbourne)

To obtain terminal genomic fragments PH1 DNA wasdigested by the enzyme Pas1 and the head (sim700 bp) andthe tail (sim1300 bp) fragments were purified by the QIAEXII gel extraction kits and treated with 05M piperidine for

Archaea 15

2 hr at 37∘C to remove protein residues The piperidine-treated fragments were sequenced by the Genomics BiotechCompany (Taipei Taiwan) using two primers TGACCA-ATTAATTAGGCCGGTTCGCC (PH1 head-R) and GTGC-CATACTGCTACAATTCT (PH1 tail-F)

Four hundred fifty-four whole genome sequencing PH1DNA samples (sim5 120583g) were sequenced using parallel pyrose-quencing on a Roche 454 Genome Sequencer System atMission Biotech (Taipei Taiwan) The largest contig was27399with 8803 reads (Newbler version 27) with 27387 posi-tions being of Q40 quality The average depth of coverage foreach base was 829TheGenBank accession for the entire PH1sequence is KC252997

Acknowledgments

The authors thank Carolyn Bath Mei Kwei (Joan) Seahand Danielle Walker for their early work in PH1 characteri-zation Simon Crawford Jocelyn Carpenter and Anna Fried-huber for their assistance with the transmission electronmicroscopy and Tiffany Cowie and Voula Kanellakis fortheir sequencing assistance This paper has been facilitatedby access to the Australian Proteome Analysis Facility estab-lished under the Australian Governmentrsquos Major NationalResearch Facilities program SLT was supported by a grantfrom the Biodiversity Research Center Academia SinicaTaiwan

References

[1] M Krupovic M F White P Forterre and D PrangishvilildquoPostcards from the edge structural genomics of archaealvirusesrdquo Advances in Virus Research vol 82 pp 33ndash62 2012

[2] P Stolt andW Zillig ldquoGene regulation in halophage 120601Hmdashmorethan promotersrdquo Systematic and Applied Microbiology vol 16no 4 pp 591ndash596 1994

[3] R Klein U Baranyi N Rossler B Greineder H Scholz and AWitte ldquoNatrialba magadii virus 120593Ch1 first complete nucleotidesequence and functional organization of a virus infecting ahaloalkaliphilic archaeonrdquo Molecular Microbiology vol 45 no3 pp 851ndash863 2002

[4] E Pagaling R D Haigh W D Grant et al ldquoSequence analysisof an Archaeal virus isolated from a hypersaline lake in InnerMongolia Chinardquo BMC Genomics vol 8 article 410 2007

[5] S L Tang SNuttall andMDyall-Smith ldquoHalovirusesHF1 andHF2 evidence for a recent and large recombination eventrdquo Jour-nal of Bacteriology vol 186 no 9 pp 2810ndash2817 2004

[6] C Bath and M L Dyall-smith ldquoHis1 an archaeal virus of theFuselloviridae family that infectsHaloarcula hispanicardquo Journalof Virology vol 72 no 11 pp 9392ndash9395 1998

[7] C Bath T Cukalac K Porter andM L Dyall-Smith ldquoHis1 andHis2 are distantly related spindle-shaped haloviruses belongingto the novel virus group Salterprovirusrdquo Virology vol 350 no1 pp 228ndash239 2006

[8] K Porter P Kukkaro J K H Bamford et al ldquoSH1 a novelspherical halovirus isolated from an Australian hypersalinelakerdquo Virology vol 335 no 1 pp 22ndash33 2005

[9] M Dyall-Smith S L Tang and C Bath ldquoHaloarchaeal viruseshow diverse are theyrdquo Research in Microbiology vol 154 no 4pp 309ndash313 2003

[10] A Oren G Bratbak and M Heldal ldquoOccurrence of virus-likeparticles in the Dead Seardquo Extremophiles vol 1 no 3 pp 143ndash149 1997

[11] T Sime-Ngando S Lucas A Robin et al et al ldquoDiversityof virus-host systems in hypersaline Lake Retba SenegalrdquoEnvironmental Microbiology vol 13 no 8 pp 1956ndash1972 2010

[12] N Guixa-Boixareu J I Calderon-Paz M Heldal G Bratbakand C Pedros-Alio ldquoViral lysis and bacterivory as prokaryoticloss factors along a salinity gradientrdquoAquaticMicrobial Ecologyvol 11 no 3 pp 215ndash227 1996

[13] S T Jaakkola R K Penttinen S T Vilen et al ldquoClosely relatedarchaeal Haloarcula hispanica icosahedral viruses HHIV-2 andSH1 have nonhomologous genes encoding host recognitionfunctionsrdquo Journal of Virology vol 86 no 9 pp 4734ndash47422012

[14] E Roine P Kukkaro L Paulin et al ldquoNew closely related halo-archaeal viral elements with different nucleic acid typesrdquo Jour-nal of Virology vol 84 no 7 pp 3682ndash3689 2010

[15] A Sencilo L Paulin S KellnerMHelm and E Roine ldquoRelatedhaloarchaeal pleomorphic viruses contain different genometypesrdquo Nucleic Acids Research vol 40 no 12 pp 5523ndash55342012

[16] H T Jaalinoja E Roine P Laurinmaki H M Kivela D HBamford and S J Butcher ldquoStructure and host-cell interactionof SH1 a membrane-containing halophilic euryarchaeal virusrdquoProceedings of the National Academy of Sciences of the UnitedStates of America vol 105 no 23 pp 8008ndash8013 2008

[17] M K Pietila N S Atanasova VManole et al ldquoVirion architec-ture unifies globally distributed pleolipoviruses infecting halo-philic archaeardquo Journal of Virology vol 86 no 9 pp 5067ndash50792012

[18] M L Holmes and M L Dyall-Smith ldquoA plasmid vector witha selectable marker for halophilic archaebacteriardquo Journal ofBacteriology vol 172 no 2 pp 756ndash761 1990

[19] M L Holmes F Pfeifer andM L Dyall-Smith ldquoAnalysis of thehalobacterial plasmid pHK2 minimal repliconrdquo Gene vol 153no 1 pp 117ndash121 1995

[20] M L Dyall-Smith F Pfeiffer K Klee et al ldquoHaloquadratumwalsbyi limited diversity in a Global Pondrdquo PLoS ONE vol 6no 6 Article ID e20968 2011

[21] S Casjens ldquoProphages and bacterial genomics what have welearned so farrdquoMolecular Microbiology vol 49 no 2 pp 277ndash300 2003

[22] R W Hendrix M C M Smith R N Burns M E Ford andG F Hatfull ldquoEvolutionary relationships among diverse bacte-riophages and prophages all the worldrsquos a phagerdquo Proceedings ofthe National Academy of Sciences of the United States of Americavol 96 no 5 pp 2192ndash2197 1999

[23] D Botstein ldquoA theory ofmodular evolution for bacteriophagesrdquoAnnals of the New York Academy of Sciences vol 354 pp 484ndash491 1980

[24] M Krupovic S Gribaldo D H Bamford and P Forterre ldquoTheevolutionary history of archaeal MCM helicases a case study ofvertical evolution combinedwithHitchhiking ofmobile geneticelementsrdquo Molecular Biology and Evolution vol 27 no 12 pp2716ndash2732 2010

[25] M Krupovic P Forterre and D H Bamford ldquoComparativeanalysis of the mosaic genomes of tailed archaeal viruses andproviruses suggests common themes for virion architecture andassembly with tailed viruses of bacteriardquo Journal of MolecularBiology vol 397 no 1 pp 144ndash160 2010

16 Archaea

[26] Z Zhang Y Liu S Wang et al et al ldquoTemperate membrane-containing halophilic archaeal virus SNJ1 has a circular dsDNAgenome identical to that of plasmid pHH205rdquoVirology vol 434no 2 pp 233ndash241 2012

[27] D H Bamford J J Ravantti G Ronnholm et al ldquoConstituentsof SH1 a novel lipid-containing virus infecting the halophiliceuryarchaeonHaloarcula hispanicardquo Journal of Virology vol 79no 14 pp 9097ndash9107 2005

[28] H M Kivela E Roine P Kukkaro S Laurinavicius P Somer-harju andDH Bamford ldquoQuantitative dissociation of archaealvirus SH1 reveals distinct capsid proteins and a lipid corerdquoVirology vol 356 no 1-2 pp 4ndash11 2006

[29] H Jaalinoja Electron Cryo-Microscopy Studies of Bacteriophage8 and Archaeal Virus SH1 University of Helsinki HelsinkiFinland 2007

[30] K Porter B E Russ J Yang andM L Dyall-Smith ldquoThe trans-cription programme of the protein-primed halovirus SH1rdquoMi-crobiology vol 154 no 11 pp 3599ndash3608 2008

[31] K Porter and M L Dyall-Smith ldquoTransfection of haloarchaeaby the DNAs of spindle and round haloviruses and the useof transposon mutagenesis to identify non-essential regionsrdquoMolecular Microbiology vol 70 no 5 pp 1236ndash1245 2008

[32] S E Luria General Virology John Wiley and Sons New YorkNY USA 1953

[33] C A Thomas K Saigo E McLeod and J Ito ldquoThe separationofDNA segments attached to proteinsrdquoAnalytical Biochemistryvol 93 pp 158ndash166 1979

[34] A L Delcher D Harmon S Kasif OWhite and S L SalzbergldquoImproved microbial gene identification with GLIMMERrdquoNucleic Acids Research vol 27 no 23 pp 4636ndash4641 1999

[35] R Zhang Y Yang P Fang et al ldquoDiversity of telomere palin-dromic sequences and replication genes among Streptomyceslinear plasmidsrdquo Applied and Environmental Microbiology vol72 no 9 pp 5728ndash5733 2006

[36] X Wang H S Lee F J Sugar F E Jenney M W W Adamsand J H Prestegard ldquoPF0610 a novel winged helix-turn-helix variant possessing a rubredoxin-like Zn ribbon motiffrom the hyperthermophilic archaeon Pyrococcus furiosusrdquoBiochemistry vol 46 no 3 pp 752ndash761 2007

[37] S W Cline L C Schalkwyk and W F Doolittle ldquoTransfor-mation of the archaebacterium Halobacterium volcanii withgenomic DNArdquo Journal of Bacteriology vol 171 no 9 pp 4987ndash4991 1989

[38] S Kamtekar A J Berman J Wang et al ldquoThe 12059329 DNA poly-merase protein-primer structure suggests a model for the ini-tiation to elongation transitionrdquo EMBO Journal vol 25 no 6pp 1335ndash1343 2006

[39] X Peng T Basta M Haring R A Garrett and D PrangishvilildquoGenome of theAcidianus bottle-shaped virus and insights intothe replication and packaging mechanismsrdquo Virology vol 364no 1 pp 237ndash243 2007

[40] C C Yang C H Huang C Y Li Y G Tsay S C Lee and CW Chen ldquoThe terminal proteins of linear Streptomyces chrom-osomes and plasmids a novel class of replication primingproteinsrdquo Molecular Microbiology vol 43 no 2 pp 297ndash3052002

[41] C C Yang Y H Chen H H Tsai C H Huang T W Huangand C W Chen ldquoIn vitro deoxynucleotidylation of the term-inal protein of Streptomyces linear chromosomesrdquo Applied andEnvironmental Microbiology vol 72 no 12 pp 7959ndash79612006

[42] P C Chang and S N Cohen ldquoBidirectional replication froman internal origin in a linear Streptomyces plasmidrdquo Science vol265 no 5174 pp 952ndash954 1994

[43] S W Wilhelm W H Jeffrey C A Suttle and D L MitchellldquoEstimation of biologically damaging UV levels in marinesurface waters with DNA and viral dosimetersrdquo Photochemistryand Photobiology vol 76 no 3 pp 268ndash273 2002

[44] T Allers and M Mevarech ldquoArchaeal geneticsmdashthe third wayrdquoNature Reviews Genetics vol 6 no 1 pp 58ndash73 2005

[45] M L Holmes S D Nuttall and M L Dyall-Smith ldquoConstruc-tion and use of halobacterial shuttle vectors and further studieson Haloferax DNA gyraserdquo Journal of Bacteriology vol 173 no12 pp 3807ndash3813 1991

[46] S D Nuttall and M L Dyall-Smith ldquoHF1 and HF2 novelbacteriophages of halophilic archaeardquo Virology vol 197 no 2pp 678ndash684 1993

[47] U Blaseio and F Pfeifer ldquoTransformation of Halobacteriumhalobium development of vectors and investigation of gas vesi-cle synthesisrdquo Proceedings of the National Academy of Sciences ofthe United States of America vol 87 no 17 pp 6772ndash6776 1990

[48] W L Lam and W F Doolittle ldquoShuttle vectors for the archae-bacterium Halobacterium volcaniirdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 86 no14 pp 5478ndash5482 1989

[49] C Yanisch-Perron J Vieira and J Messing ldquoImproved M13phage cloning vectors and host strains nucleotide sequences ofthe M13mp18 and pUC19 vectorsrdquo Gene vol 33 no 1 pp 103ndash119 1985

[50] U K Laemmli ldquoCleavage of structural proteins during theassembly of the head of bacteriophage T4rdquo Nature vol 227 no5259 pp 680ndash685 1970

[51] D G Burns H M Camakaris P H Janssen and M L Dyall-Smith ldquoCombined use of cultivation-dependent and culti-vation-independent methods indicates that members of mosthaloarchaeal groups in anAustralian crystallizer pond are culti-vablerdquo Applied and Environmental Microbiology vol 70 no 9pp 5258ndash5265 2004

[52] G Juez F Rodriguez-Valera A Ventosa and D J KushnerldquoHaloarcula hispanica spec nov and Haloferax gibbonsii specnov two new species of extremely halophilic archaebacteriardquoSystematic and Applied Microbiology vol 8 pp 75ndash79 1986

[53] A Oren M Ginzburg B Z Ginzburg L I Hochstein and BE Volcani ldquoHaloarcula marismortui (Volcani) sp nov nomrev an extremely halophilic bacterium from the Dead SeardquoInternational Journal of Systematic Bacteriology vol 40 no 2pp 209ndash210 1990

[54] M Torreblanca F Rodriguez-Valera G Juez A Ventosa MKamekura and M Kates ldquoClassification of non-alkaliphilichalobacteria based on numerical taxonomy and polar lipidcomposition and description of Haloarcula gen nov andHaloferax gen novrdquo Systematic and Applied Microbiology vol8 pp 89ndash99 1986

[55] A Ventosa and A Oren ldquoHalobacterium salinarum nom cor-rig a name to replace Halobacterium salinarium (Elazari-Vol-cani) and to include Halobacterium halobium and Halobac-terium cutirubrumrdquo International Journal of Systematic Bacte-riology vol 46 no 1 p 347 1996

[56] M C Gutierrez M Kamekura M L Holmes M L Dyall-Smith and A Ventosa ldquoTaxonomic characterization of Halo-ferax sp (ldquoH alicanteirdquo) strain Aa 22 description of Haloferaxlucentensis sp novrdquo Extremophiles vol 6 no 6 pp 479ndash4832002

Archaea 17

[57] M F Mullakhanbhai and H Larsen ldquoHalobacterium volcaniispec nov a Dead Sea halobacterium with a moderate saltrequirementrdquo Archives of Microbiology vol 104 no 1 pp 207ndash214 1975

[58] S D Nuttall and M L Dyall-Smith ldquoCh2 a novel halophilicarchaeon from an Australian solar salternrdquo International Jour-nal of Systematic Bacteriology vol 43 no 4 pp 729ndash734 1993

[59] P D Franzman E Stackebrandt K Sanderson et al ldquoHalobac-terium lacusprofundi sp nov a halophilic bacterium isolatedfrom Deep Lake Antarcticardquo Systematic and Applied Microbi-ology vol 11 pp 20ndash27 1988

[60] G A Tomlinson and L I Hochstein ldquoHalobacterium saccharo-vorum sp nov a carbohydrate metabolizing extremely halo-philic bacteriumrdquoCanadian Journal ofMicrobiology vol 22 no4 pp 587ndash591 1976

[61] A Ventosa M C Gutierrez M Kamekura and M L Dyall-Smith ldquoProposal to transfer Halococcus turkmenicus Halobac-terium trapanicum JCM9743 and strainGSL-11 toHaloterrigenaturkmenica gen nov comb novrdquo International Journal ofSystematic Bacteriology vol 49 no 1 pp 131ndash136 1999

[62] M Kamekura and M L Dyall-Smith ldquoTaxonomy of the familyHalobacteriaceae and the description of two new genera Halo-rubrobacterium and Natrialbardquo Journal of General and AppliedMicrobiology vol 41 no 4 pp 333ndash350 1995

[63] T J Carver KM RutherfordM BerrimanM A RajandreamB G Barrell and J Parkhill ldquoACT the Artemis comparisontoolrdquo Bioinformatics vol 21 no 16 pp 3422ndash3423 2005

[64] S W Cline W L Lam R L Charlebois L C Schalkwyk andW F Doolittle ldquoTransformationmethods for halophilic archae-bacteriardquo Canadian Journal of Microbiology vol 35 no 1 pp148ndash152 1989

12 Archaea

density and are chloroform sensitive and show a layeredcapsid structure consistent with the likely presence of aninternalmembrane layer as has been shown for SH1HHIV-2and SNJ1 [13 26]The particle stability of PH1 to temperaturepH and reduced salt was similar to SH1 [8] The structuralproteins of PH1 are very similar in sequence and relativeabundance to those of SH1 so the two viruses are likely toshare the same particle geometry [16]

Viruses that use protein-primed replication such as bac-teriophageΦ29 carry a viral type BDNApolymerase that caninteract specifically with the proteins attached to the genomictermini and initiate strand synthesis [38] ArchaeovirusesHis1 [7] and Acidianus bottle-shaped virus [39] probablyuse this mode of replication However a polymerase genecannot be found in the genomes of PH1 SH1 or HHIV-2DNA polymerases are large enzymes and the only ORF ofSH1 that was long enough to encode such an enzyme andhad not previously been assigned as a structural protein wasORF55 which specified an 865 aa protein that contained noconserved domains indicative of polymerases [27]The recentstudy of HHIV-2 showed that the corresponding ORF in thisvirus (gene 42) is a structural protein of the virion and byinference this is likely also for the corresponding proteinsof SH1 and PH1 (no homolog is present in SNJ1) Withouta polymerase gene these viruses must use a host enzymebut searches for a viral type B DNA polymerase in thegenome of the hostHar hispanica or in the genomes of othersequenced haloarchaea did not find anymatchesThis arguesstrongly for a replication mechanism that is different to thatexemplified byΦ29 An attractive alternative is that displayedby Streptomyces linear plasmids which have 51015840-terminal pro-teins and use a cellular polymerase for replication [40] Theterminal proteins are not used for primary replication but forend patching [41] and it has been shown that if the term-inal repeat sequences are removed from these plasmids andthe ends ligated they can replicate as circular plasmids [42]This provides a testable hypothesis for the replication of halo-sphaeroviruses and also offers a pathway for switchingbetween linear (eg PH1) and circular (eg SNJ1) forms Useof a host polymerase would also fit with the ability of SH1 andPH1 DNA to transfect many different haloarchaeal species[31] as these enzymes and theirmode of action are highly con-served

The growth characteristics of PH1 inHar hispanica bothby single-cell burst and single-step growth experiments gavevalues of 50ndash100 virusescell significantly less than that ofSH1 (200 virusescell) [8] and HHIV-2 (180 virusescell) [13]The latter virus is lytic whereas SH1 PH1 and SNJ1 start pro-ducing extracellular viruswell before any decrease in cell den-sity indicating that virus release can occur without lysisThishas been most clearly shown with SH1 where almost 100of cells can be infected [8] Infection ofHar hispanica by PH1was less efficient (sim30 of cells) but the kinetics of virus pro-duction in single-step growth curves of SH1 and PH1 aresimilar with virus production occurring over several hoursrather than at a clearly defined time after infection The twoviruses are also very closely related and infect the same hostspecies so it is likely that they use the same scheme for cellexit For both viruses the cell density eventually decreases in

single-step cultures showing that virus infection does resultin cell death This mode of exit appears to maximise theproduction of the virus over time as the host cells survive foran extended period In natural hypersaline waters cell num-bers are usually high but growth rates are low [12] and thehigh incident UV (on often shallow ponds) would damagevirus DNA so reducing the half-life of released virus part-icles [43] These factors may have favoured the exit strategydisplayed by these viruses improving their chance of trans-mission to a new host

The motif GATC was absent in the genomes of PH1 SH1andHHIV-2 and themotif CTAGwas either absent or greatlyunderrepresented All three viruses share the same host butGATC and CTAG motifs are plentiful in the Har hispanicagenome (accessions CP002921 and CP002923) Avoidance ofthesemotifs in halovirusesHis1 His2 HF1 andHF2 has beenreported previously [7] Such purifying selection commonlyresults from host restriction enzymes and inHfx volcanii itis exactly these two motifs that are targeted [44] One Hfxvolcanii enzyme recognises A-methylated GATC sites [45]and another recognises un-methylated CTAG sites (whichare protected by methylation in the genome) In the currentstudy this could explain the negative transfection results forHfx volcanii as the PH1 genome contains two CTAGmotifsBy comparison the SH1 genome contains no CTAG motifsand is able to transfect Hfx volcanii [31]

Comparison of the PH1 and SH1 genomes allowed adetailed picture of the natural variation occurring betweenclosely related viruses revealing likely deletion events medi-ated by small repeats a process described previously in astudy of Hqr walsbyi and termed repeat-mediated deletion[20] There are also many replacements including blocks ofsequence that occur within long open reading frames (suchas the capsid protein genes VP1 and VP18) VP3 and VP6 areknown to form the large spikes on the external surface of theSH1 virion and presumably one or both interact with host cellreceptors [16] PH1 and SH1 have identical host ranges con-sistent with the high sequence similarity shown between theircorresponding VP3 and VP6 proteins

A previous study of SH1 could not detect transcriptsacross annotated ORFs 1ndash3 or 55-56 [30] ORFs 1 and 56occur in the terminal inverted repeat sequence and in thepresent study it was found that there were no ORFs cor-responding to these in the PH1 genome Given the close simi-larity of the two genomes the comparative data are in agree-ment with the transcriptional data indicating that these SH1ORFs are not used SH1 ORFs 2 and 3 are more problematicas good homologs of these are also present in PH1 The con-flict between the transcriptional data and the comparativegenomic evidence requires further experimental work toresolveThe status of SH1ORF55 has recently been confirmedby studies of the related virus HHIV-2 (discussed above)

The pleolipovirus group has expanded dramatically inthe last few years and it now comprises a diverse group ofviruseswith different genome types and replication strategiesWhat is evenmore remarkable is that a similar expansion cannow be seen with SH1-related viruses which includes viruseswith at least two replication modes and genome types (linearand circular) Evidence from haloarchaeal genome sequences

Archaea 13

show cases of not only provirus integrants or plasmids (egpKH2 and pHH205) but also genomic loci (ViPREs) thatcontain cassettes of virus genes from different sources andat least two cases (described in this study) have likely attsites that indicate circularisation and mobility While thereis a clear relationship between viruses with circular ds- andssDNA genomes that replicate via the rolling circle method(ie the ssDNA form is a replication intermediate) it ismore difficult to explain how capsid gene cassettes can movebetween these viruses and those with linear genomes andterminal proteins Within the pleolipoviruses His2 has alinear genome that contains a viral type B DNA polymeraseand terminal proteins while all the other described membershave circular genomes that contain a rep homolog and prob-ably replicate via the rolling circle method [14 19] Similarlyamong the halosphaeroviruses PH1 SH1 and HHIV-2 havelinear dsDNA with terminal proteins and probably replicatein the same way but the related SNJ1 virus has a circulardsDNA genome The clear relationships shown by the capsidgenes of viruses within each group plus the connectionsshown to haloviruses outside each group (eg the DNA poly-merases of His1 and His2) all speak of a vigorous meansof recombination one that can readily switch capsid genesbetween viruses with radically different replication strategiesHow is the process most likely to operate One possibility issuggested by ViPREs which appear to be mobile collectionsof capsid and replication genes from different sources Theyoffer fixed locations for recombination to occur providegene cassettes that can be reassorted to produce novel virusgenomes and some can probably recombine out as circularforms For example ViPREHmuk1 carries genes related topleolipoviruses halosphaeroviruses and other haloviruses Itis yet to be determined if any of the genes carried in theseloci are expressed in the cell (such as the genes for capsidproteins) or if ViPREs provide any selective advantage to thehost

4 Materials and Methods

41 Water Sample A water sample was collected in 1998from Pink Lake (32∘001015840 S and 115∘301015840 E) a hypersalinelake on Rottnest Island Western Australia Australia It wasscreened in the same year for haloviruses using the methodsdescribed in [8] A single plaque on a Har hispanica lawnplate was picked and replaque purified The novel haloviruswas designated PH1

42 Media Strains and Plasmids The media used in thisstudy are described in the online resource The Halohand-book (httpwwwhaloarchaeacomresourceshalohandbookindexhtml) Artificial salt water containing 30 (wv) totalsalts comprised of 4M NaCl 150mM MgCl

2 150mM

MgSO4 90mM KCl and 35mM CaCl

2and adjusted to pH

75 using sim2mL 1M Tris-HCl (pH 75) per litre MGM con-taining 12 18 or 23 (wv) total salts and HVD (halo-virus diluent) were prepared from the concentrated stock aspreviously described [46] Bacto-agar (Difco Laboratories)was added to MGM for solid (15 gL) or top-layer (7 gL)media

Table 1 lists the haloarchaea used in this study Allhaloarchaea were grown aerobically at 37∘C in either 18 or23 (wv)MGM(depending on the strain) andwith agitation(except for ldquoHar sinaiiensisrdquo)The plasmids used were pBlue-script II KS+ (Stratagene Cloning Systems) pUBP2 [47]and pWL102 [48] pUBP2 and pWL102 were first passagedthrough E coli JM110 [49] to prevent dam-methylationof DNA which has been shown to reduce transformationefficiency in some strains [45]

43 Negative-Stain Electron Microscopy The method fornegative-stain TEM was adapted from that of V Tarasovdescribed in the online resource The Halohandbook (httpwwwhaloarchaeacomresourceshalohandbookindexhtml)A 20120583L drop of the sample was placed on a clean surfaceand the virus particles were allowed to adsorb to a Formvarfilm 400-mesh copper grid (ProSciTech) for 15ndash2min Theywere then negatively stained with a 20120583L drop of 2 (wv)uranyl acetate for 1-2min Excess liquid was absorbed withfilter paper and the grid was allowed to air dry Grids wereexamined either on a Philips CM 120 BioTwin transmissionelectron microscope (Royal Philips Electronics) operating atan accelerating voltage of 120 kV or on a Siemens Elmiskop102 transmission electron microscope (Siemens AG) operat-ing at an accelerating voltage of 120 kV

44 Virus Host Range Twelve haloarchaeal strains from thegenera Haloarcula Halobacterium Haloferax HalorubrumandNatrialba (Table 1)Thirteen naturalHalorubrum isolates(H Camakaris unpublished data) and five uncharacterizedhaloarchaeal isolates fromLakeHardy Pink Lake (inWesternAustralia) and Serpentine Lake (D Walker and M K Seahunpublished data) were screened for PH1 susceptibilityLysates from PH1-infected Har hispanica cultures (1 times1011 PFUmL) were spotted onto lawns of each strain (usingmedia with 12 or 18 (wv)MGM depending on the strain)and incubated for 2ndash5 days at 30 and at 37∘C

45 Large Scale Virus Growth and Purification Liquid cul-tures of PH1 were grown by the infection (MOI 005) of anearly exponential Har hispanica culture in 18 (wv) MGMCultures were incubated aerobically at 37∘C with agitationfor 3 days Clearing (ie complete cell lysis) of PH1-infectedcultures did not occur and consequently cultures were harv-ested when the absorbance at 550 nm reached the minimumand the titre (determined by plaque assay) was at the maxi-mum usually sim1010ndash1011 PFUmL Virus was purified usingthe method described previously for SH1 [8 31] except thatafter the initial low speed spin (Sorvall GSA 6000 rpm30min 10∘C) virus was concentrated from the infected cul-ture by centrifugation at 26000 rpm (13 hr 10∘C) onto acushion of 30 (wv) sucrose in HVDThe pellet was resus-pended in a small volume of HVD and loaded onto a pre-formed linear 5ndash70 (wv) sucrose gradient followed byisopycnic centrifugation in 13 gmL CsCl (Beckman 70Ti60000 rpm 20 hr 10∘C) The white virus band occurredat a density of 129 gmL and was collected and diluted inhalovirus diluent (HVD see Section 4) and the virus pelleted(Beckman SW55 35000 rpm 75min 10∘C) and resuspended

14 Archaea

in a small volume of HVD and stored at 4∘C Virus recoveryat the major stages of a typical purification is given insupplementary Table 1 The specific infectivity of pure virussolutions was determined as the ratio of the PFUmL to theabsorbance at 260 nm

46 PH1 Single-Step Growth Curve An early exponentialphase culture of Har hispanica grown in 18 (wv) MGMwas infected with PH1 (MOI 50) Under these conditions thepercentage of infected cells was approximately 30 After anadsorption period of 1 hr at 37∘C the cells were washed threetimes with 18 (wv) MGM (at room temperature) resus-pended in 100mL 18 (wv) MGM (these methods ensuredthe removal of all residual virus) and incubated at 37∘Cwith shaking (100 rpm) Samples were removed at hourlyintervals for measurements of absorbance at 550 nm andthe number of infective centres Immediately after samplingtitres were determined by plaque assay with an indicator lawnof Har hispanica Each experiment was performed in tripli-cate

47 Halovirus Stability After various treatments sampleswere removed and diluted in HVD and virus titres weredetermined by plaque assay on Har hispanica Each experi-ment was performed in triplicate and representative data areshown Chloroform sensitivity was examined by the exposureof PH1 lysates in 18 (wv) MGM to chloroform in a volumeratio of 1 4 (chloroform to lysate) Incubation was at roomtemperature with constant agitation At appropriate timepoints the mix was allowed to settle and a sample wasremoved from the upper layer and diluted in HVDThe effectof a low ionic environment was examined by diluting PH1lysates (in 18 (wv) MGM) into double distilled H

2O in a

volume ratio of 1 1000 (lysate to double distilledH2O) Incu-

bation was at room temperature with constant agitationSamples were removed at various times and diluted in HVDThe pH stability of PH1 was determined by dilution of PH1lysates in 18 (wv)MGMin the appropriate pHbuffer (HVDbuffered with appropriate Tris-HCl) in a volume ratio of1 100 (lysate to buffer) Incubation was at room temperaturewith constant agitation After 30min samples were removedand diluted in HVD Thermal stability of PH1 was examinedby a 1 hr incubation of a virus lysate in 18 (wv) MGM atdifferent temperatures after which they were brought quicklyto room temperature diluted in HVD and titrated

48 Protein Procedures To remove salts purified virus prepa-rations were mixed with trichloroacetic acid (10 (vv) finalconcentration) and incubated on ice for 15min to allowthe proteins to precipitate After centrifugation (16000 g15min room temperature) the precipitate was washed threetimes in acetone dried and resuspended in double distilledH2O Proteins were dissolved in Laemmli sample buffer with

48mM 120573-mercaptoethanol [50] heated in boiling water for5min then separated on 12 (wv) NuPAGE Novex Bis-Tris Gels using MES-SDS running buffer according to themanufacturerrsquos directions (Invitrogen) After electrophoresisgels were rinsed in double distilled H

2O and stained with

01 (wv) Brilliant Blue G in 40 (vv) methanol and

10 (vv) acetic acid Gels were destained with severalchanges of 40 (vv) methanol and 10 (vv) acetic acidAlternatively gels were stained with GelCode GlycoproteinStaining Kit Pro-Q Emerald 488 Glycoprotein Gel and BlotStain Kit SYPRO Ruby Protein Gel Stain according tothe manufacturerrsquos directions (Invitrogen Molecular ProbesPierce Biotechnology)

Protein bands were cut from the gels and sent to theAustralian Proteome Analysis Facility (Macquarie Univer-sity) for trypsin digestion and analysis by matrix assistedlaser desorption ionisation time of flight mass spectrometry(MALDI-TOF MS) on an Applied Biosystems 4700 Pro-teomics Analyser (Applied Biosystems)

49 DNA Procedures Proteinase K-treated and nonpro-teinaseK-treatedDNApreparationsweremade frompurifiedPH1 using the methods described previously for SH1 [8]DNA was separated on 1 (wv) agarose gels in Tris-acetate-EDTA electrophoresis buffer and was stained with ethidiumbromide (Sigma-Aldrich)120582 DNA DNase I (RNase-free) exonuclease III mung

bean nuclease nuclease BAL-31 T4 DNA polymerase T4DNA ligase T7 exonuclease and type II restriction endonu-cleases were purchased from New England Biolabs RNase Aand yeast transfer RNA were purchased from Sigma-AldrichProteinase K was purchased from Promega Oligonucleotideprimers were purchased from Geneworks Australia

To clone fragments of the PH1 genome purified virusDNA was digested either with AccI Eco0109I and MseI orSmaI restriction endonucleases blunted with DNA Poly-merase I Large (Klenow) Fragment where appropriate andligated into AccI- Eco0109I- or SmaI-digested pBluescript IIKS+ using T4 DNA ligase according to the manufacturerrsquosinstructions (New England Biolabs) The DNA was intro-duced intoE coliXL1-Blue and transformants grown onLuriaagar containing 15 120583gmL tetracycline and 100 120583gmL ampi-cillinThe resulting cloneswere sequenced and the sequenceswere used to design specific oligonucleotide primers for PCRamplification andor primer walking using the virus genome(or specific restriction fragments) as templates

To amplify PH1 nucleic acid approximately 10 ng virusDNA was combined with 500 nM primers 100120583M of eachdNTP 1U Deep Vent DNA Polymerase (New England Bio-labs) and 1 timesThermoPol buffer (New England Biolabs) ina total reaction volume of 50 120583L Template was denatured at95∘C for 10min followed by 30 cycles of denaturation at 95∘Cfor 30 sec annealing at 56∘C for 30 sec extension at 75∘C for2min and a final extension at 75∘C for 10min PCR reactionswere performed on a PxE02ThermoCycler (Thermo ElectroCorporation) Sequencing reactions using 32 pmol primerwere performed by the dideoxy chain termination methodusing ABI PRISM Big Dye Terminator Mix version 31 onan ABI 3100 capillary sequencer (Applied Biosystems) bythe Applied Genetic Diagnostics Sequencing Service at theDepartment of Pathology (The University of Melbourne)

To obtain terminal genomic fragments PH1 DNA wasdigested by the enzyme Pas1 and the head (sim700 bp) andthe tail (sim1300 bp) fragments were purified by the QIAEXII gel extraction kits and treated with 05M piperidine for

Archaea 15

2 hr at 37∘C to remove protein residues The piperidine-treated fragments were sequenced by the Genomics BiotechCompany (Taipei Taiwan) using two primers TGACCA-ATTAATTAGGCCGGTTCGCC (PH1 head-R) and GTGC-CATACTGCTACAATTCT (PH1 tail-F)

Four hundred fifty-four whole genome sequencing PH1DNA samples (sim5 120583g) were sequenced using parallel pyrose-quencing on a Roche 454 Genome Sequencer System atMission Biotech (Taipei Taiwan) The largest contig was27399with 8803 reads (Newbler version 27) with 27387 posi-tions being of Q40 quality The average depth of coverage foreach base was 829TheGenBank accession for the entire PH1sequence is KC252997

Acknowledgments

The authors thank Carolyn Bath Mei Kwei (Joan) Seahand Danielle Walker for their early work in PH1 characteri-zation Simon Crawford Jocelyn Carpenter and Anna Fried-huber for their assistance with the transmission electronmicroscopy and Tiffany Cowie and Voula Kanellakis fortheir sequencing assistance This paper has been facilitatedby access to the Australian Proteome Analysis Facility estab-lished under the Australian Governmentrsquos Major NationalResearch Facilities program SLT was supported by a grantfrom the Biodiversity Research Center Academia SinicaTaiwan

References

[1] M Krupovic M F White P Forterre and D PrangishvilildquoPostcards from the edge structural genomics of archaealvirusesrdquo Advances in Virus Research vol 82 pp 33ndash62 2012

[2] P Stolt andW Zillig ldquoGene regulation in halophage 120601Hmdashmorethan promotersrdquo Systematic and Applied Microbiology vol 16no 4 pp 591ndash596 1994

[3] R Klein U Baranyi N Rossler B Greineder H Scholz and AWitte ldquoNatrialba magadii virus 120593Ch1 first complete nucleotidesequence and functional organization of a virus infecting ahaloalkaliphilic archaeonrdquo Molecular Microbiology vol 45 no3 pp 851ndash863 2002

[4] E Pagaling R D Haigh W D Grant et al ldquoSequence analysisof an Archaeal virus isolated from a hypersaline lake in InnerMongolia Chinardquo BMC Genomics vol 8 article 410 2007

[5] S L Tang SNuttall andMDyall-Smith ldquoHalovirusesHF1 andHF2 evidence for a recent and large recombination eventrdquo Jour-nal of Bacteriology vol 186 no 9 pp 2810ndash2817 2004

[6] C Bath and M L Dyall-smith ldquoHis1 an archaeal virus of theFuselloviridae family that infectsHaloarcula hispanicardquo Journalof Virology vol 72 no 11 pp 9392ndash9395 1998

[7] C Bath T Cukalac K Porter andM L Dyall-Smith ldquoHis1 andHis2 are distantly related spindle-shaped haloviruses belongingto the novel virus group Salterprovirusrdquo Virology vol 350 no1 pp 228ndash239 2006

[8] K Porter P Kukkaro J K H Bamford et al ldquoSH1 a novelspherical halovirus isolated from an Australian hypersalinelakerdquo Virology vol 335 no 1 pp 22ndash33 2005

[9] M Dyall-Smith S L Tang and C Bath ldquoHaloarchaeal viruseshow diverse are theyrdquo Research in Microbiology vol 154 no 4pp 309ndash313 2003

[10] A Oren G Bratbak and M Heldal ldquoOccurrence of virus-likeparticles in the Dead Seardquo Extremophiles vol 1 no 3 pp 143ndash149 1997

[11] T Sime-Ngando S Lucas A Robin et al et al ldquoDiversityof virus-host systems in hypersaline Lake Retba SenegalrdquoEnvironmental Microbiology vol 13 no 8 pp 1956ndash1972 2010

[12] N Guixa-Boixareu J I Calderon-Paz M Heldal G Bratbakand C Pedros-Alio ldquoViral lysis and bacterivory as prokaryoticloss factors along a salinity gradientrdquoAquaticMicrobial Ecologyvol 11 no 3 pp 215ndash227 1996

[13] S T Jaakkola R K Penttinen S T Vilen et al ldquoClosely relatedarchaeal Haloarcula hispanica icosahedral viruses HHIV-2 andSH1 have nonhomologous genes encoding host recognitionfunctionsrdquo Journal of Virology vol 86 no 9 pp 4734ndash47422012

[14] E Roine P Kukkaro L Paulin et al ldquoNew closely related halo-archaeal viral elements with different nucleic acid typesrdquo Jour-nal of Virology vol 84 no 7 pp 3682ndash3689 2010

[15] A Sencilo L Paulin S KellnerMHelm and E Roine ldquoRelatedhaloarchaeal pleomorphic viruses contain different genometypesrdquo Nucleic Acids Research vol 40 no 12 pp 5523ndash55342012

[16] H T Jaalinoja E Roine P Laurinmaki H M Kivela D HBamford and S J Butcher ldquoStructure and host-cell interactionof SH1 a membrane-containing halophilic euryarchaeal virusrdquoProceedings of the National Academy of Sciences of the UnitedStates of America vol 105 no 23 pp 8008ndash8013 2008

[17] M K Pietila N S Atanasova VManole et al ldquoVirion architec-ture unifies globally distributed pleolipoviruses infecting halo-philic archaeardquo Journal of Virology vol 86 no 9 pp 5067ndash50792012

[18] M L Holmes and M L Dyall-Smith ldquoA plasmid vector witha selectable marker for halophilic archaebacteriardquo Journal ofBacteriology vol 172 no 2 pp 756ndash761 1990

[19] M L Holmes F Pfeifer andM L Dyall-Smith ldquoAnalysis of thehalobacterial plasmid pHK2 minimal repliconrdquo Gene vol 153no 1 pp 117ndash121 1995

[20] M L Dyall-Smith F Pfeiffer K Klee et al ldquoHaloquadratumwalsbyi limited diversity in a Global Pondrdquo PLoS ONE vol 6no 6 Article ID e20968 2011

[21] S Casjens ldquoProphages and bacterial genomics what have welearned so farrdquoMolecular Microbiology vol 49 no 2 pp 277ndash300 2003

[22] R W Hendrix M C M Smith R N Burns M E Ford andG F Hatfull ldquoEvolutionary relationships among diverse bacte-riophages and prophages all the worldrsquos a phagerdquo Proceedings ofthe National Academy of Sciences of the United States of Americavol 96 no 5 pp 2192ndash2197 1999

[23] D Botstein ldquoA theory ofmodular evolution for bacteriophagesrdquoAnnals of the New York Academy of Sciences vol 354 pp 484ndash491 1980

[24] M Krupovic S Gribaldo D H Bamford and P Forterre ldquoTheevolutionary history of archaeal MCM helicases a case study ofvertical evolution combinedwithHitchhiking ofmobile geneticelementsrdquo Molecular Biology and Evolution vol 27 no 12 pp2716ndash2732 2010

[25] M Krupovic P Forterre and D H Bamford ldquoComparativeanalysis of the mosaic genomes of tailed archaeal viruses andproviruses suggests common themes for virion architecture andassembly with tailed viruses of bacteriardquo Journal of MolecularBiology vol 397 no 1 pp 144ndash160 2010

16 Archaea

[26] Z Zhang Y Liu S Wang et al et al ldquoTemperate membrane-containing halophilic archaeal virus SNJ1 has a circular dsDNAgenome identical to that of plasmid pHH205rdquoVirology vol 434no 2 pp 233ndash241 2012

[27] D H Bamford J J Ravantti G Ronnholm et al ldquoConstituentsof SH1 a novel lipid-containing virus infecting the halophiliceuryarchaeonHaloarcula hispanicardquo Journal of Virology vol 79no 14 pp 9097ndash9107 2005

[28] H M Kivela E Roine P Kukkaro S Laurinavicius P Somer-harju andDH Bamford ldquoQuantitative dissociation of archaealvirus SH1 reveals distinct capsid proteins and a lipid corerdquoVirology vol 356 no 1-2 pp 4ndash11 2006

[29] H Jaalinoja Electron Cryo-Microscopy Studies of Bacteriophage8 and Archaeal Virus SH1 University of Helsinki HelsinkiFinland 2007

[30] K Porter B E Russ J Yang andM L Dyall-Smith ldquoThe trans-cription programme of the protein-primed halovirus SH1rdquoMi-crobiology vol 154 no 11 pp 3599ndash3608 2008

[31] K Porter and M L Dyall-Smith ldquoTransfection of haloarchaeaby the DNAs of spindle and round haloviruses and the useof transposon mutagenesis to identify non-essential regionsrdquoMolecular Microbiology vol 70 no 5 pp 1236ndash1245 2008

[32] S E Luria General Virology John Wiley and Sons New YorkNY USA 1953

[33] C A Thomas K Saigo E McLeod and J Ito ldquoThe separationofDNA segments attached to proteinsrdquoAnalytical Biochemistryvol 93 pp 158ndash166 1979

[34] A L Delcher D Harmon S Kasif OWhite and S L SalzbergldquoImproved microbial gene identification with GLIMMERrdquoNucleic Acids Research vol 27 no 23 pp 4636ndash4641 1999

[35] R Zhang Y Yang P Fang et al ldquoDiversity of telomere palin-dromic sequences and replication genes among Streptomyceslinear plasmidsrdquo Applied and Environmental Microbiology vol72 no 9 pp 5728ndash5733 2006

[36] X Wang H S Lee F J Sugar F E Jenney M W W Adamsand J H Prestegard ldquoPF0610 a novel winged helix-turn-helix variant possessing a rubredoxin-like Zn ribbon motiffrom the hyperthermophilic archaeon Pyrococcus furiosusrdquoBiochemistry vol 46 no 3 pp 752ndash761 2007

[37] S W Cline L C Schalkwyk and W F Doolittle ldquoTransfor-mation of the archaebacterium Halobacterium volcanii withgenomic DNArdquo Journal of Bacteriology vol 171 no 9 pp 4987ndash4991 1989

[38] S Kamtekar A J Berman J Wang et al ldquoThe 12059329 DNA poly-merase protein-primer structure suggests a model for the ini-tiation to elongation transitionrdquo EMBO Journal vol 25 no 6pp 1335ndash1343 2006

[39] X Peng T Basta M Haring R A Garrett and D PrangishvilildquoGenome of theAcidianus bottle-shaped virus and insights intothe replication and packaging mechanismsrdquo Virology vol 364no 1 pp 237ndash243 2007

[40] C C Yang C H Huang C Y Li Y G Tsay S C Lee and CW Chen ldquoThe terminal proteins of linear Streptomyces chrom-osomes and plasmids a novel class of replication primingproteinsrdquo Molecular Microbiology vol 43 no 2 pp 297ndash3052002

[41] C C Yang Y H Chen H H Tsai C H Huang T W Huangand C W Chen ldquoIn vitro deoxynucleotidylation of the term-inal protein of Streptomyces linear chromosomesrdquo Applied andEnvironmental Microbiology vol 72 no 12 pp 7959ndash79612006

[42] P C Chang and S N Cohen ldquoBidirectional replication froman internal origin in a linear Streptomyces plasmidrdquo Science vol265 no 5174 pp 952ndash954 1994

[43] S W Wilhelm W H Jeffrey C A Suttle and D L MitchellldquoEstimation of biologically damaging UV levels in marinesurface waters with DNA and viral dosimetersrdquo Photochemistryand Photobiology vol 76 no 3 pp 268ndash273 2002

[44] T Allers and M Mevarech ldquoArchaeal geneticsmdashthe third wayrdquoNature Reviews Genetics vol 6 no 1 pp 58ndash73 2005

[45] M L Holmes S D Nuttall and M L Dyall-Smith ldquoConstruc-tion and use of halobacterial shuttle vectors and further studieson Haloferax DNA gyraserdquo Journal of Bacteriology vol 173 no12 pp 3807ndash3813 1991

[46] S D Nuttall and M L Dyall-Smith ldquoHF1 and HF2 novelbacteriophages of halophilic archaeardquo Virology vol 197 no 2pp 678ndash684 1993

[47] U Blaseio and F Pfeifer ldquoTransformation of Halobacteriumhalobium development of vectors and investigation of gas vesi-cle synthesisrdquo Proceedings of the National Academy of Sciences ofthe United States of America vol 87 no 17 pp 6772ndash6776 1990

[48] W L Lam and W F Doolittle ldquoShuttle vectors for the archae-bacterium Halobacterium volcaniirdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 86 no14 pp 5478ndash5482 1989

[49] C Yanisch-Perron J Vieira and J Messing ldquoImproved M13phage cloning vectors and host strains nucleotide sequences ofthe M13mp18 and pUC19 vectorsrdquo Gene vol 33 no 1 pp 103ndash119 1985

[50] U K Laemmli ldquoCleavage of structural proteins during theassembly of the head of bacteriophage T4rdquo Nature vol 227 no5259 pp 680ndash685 1970

[51] D G Burns H M Camakaris P H Janssen and M L Dyall-Smith ldquoCombined use of cultivation-dependent and culti-vation-independent methods indicates that members of mosthaloarchaeal groups in anAustralian crystallizer pond are culti-vablerdquo Applied and Environmental Microbiology vol 70 no 9pp 5258ndash5265 2004

[52] G Juez F Rodriguez-Valera A Ventosa and D J KushnerldquoHaloarcula hispanica spec nov and Haloferax gibbonsii specnov two new species of extremely halophilic archaebacteriardquoSystematic and Applied Microbiology vol 8 pp 75ndash79 1986

[53] A Oren M Ginzburg B Z Ginzburg L I Hochstein and BE Volcani ldquoHaloarcula marismortui (Volcani) sp nov nomrev an extremely halophilic bacterium from the Dead SeardquoInternational Journal of Systematic Bacteriology vol 40 no 2pp 209ndash210 1990

[54] M Torreblanca F Rodriguez-Valera G Juez A Ventosa MKamekura and M Kates ldquoClassification of non-alkaliphilichalobacteria based on numerical taxonomy and polar lipidcomposition and description of Haloarcula gen nov andHaloferax gen novrdquo Systematic and Applied Microbiology vol8 pp 89ndash99 1986

[55] A Ventosa and A Oren ldquoHalobacterium salinarum nom cor-rig a name to replace Halobacterium salinarium (Elazari-Vol-cani) and to include Halobacterium halobium and Halobac-terium cutirubrumrdquo International Journal of Systematic Bacte-riology vol 46 no 1 p 347 1996

[56] M C Gutierrez M Kamekura M L Holmes M L Dyall-Smith and A Ventosa ldquoTaxonomic characterization of Halo-ferax sp (ldquoH alicanteirdquo) strain Aa 22 description of Haloferaxlucentensis sp novrdquo Extremophiles vol 6 no 6 pp 479ndash4832002

Archaea 17

[57] M F Mullakhanbhai and H Larsen ldquoHalobacterium volcaniispec nov a Dead Sea halobacterium with a moderate saltrequirementrdquo Archives of Microbiology vol 104 no 1 pp 207ndash214 1975

[58] S D Nuttall and M L Dyall-Smith ldquoCh2 a novel halophilicarchaeon from an Australian solar salternrdquo International Jour-nal of Systematic Bacteriology vol 43 no 4 pp 729ndash734 1993

[59] P D Franzman E Stackebrandt K Sanderson et al ldquoHalobac-terium lacusprofundi sp nov a halophilic bacterium isolatedfrom Deep Lake Antarcticardquo Systematic and Applied Microbi-ology vol 11 pp 20ndash27 1988

[60] G A Tomlinson and L I Hochstein ldquoHalobacterium saccharo-vorum sp nov a carbohydrate metabolizing extremely halo-philic bacteriumrdquoCanadian Journal ofMicrobiology vol 22 no4 pp 587ndash591 1976

[61] A Ventosa M C Gutierrez M Kamekura and M L Dyall-Smith ldquoProposal to transfer Halococcus turkmenicus Halobac-terium trapanicum JCM9743 and strainGSL-11 toHaloterrigenaturkmenica gen nov comb novrdquo International Journal ofSystematic Bacteriology vol 49 no 1 pp 131ndash136 1999

[62] M Kamekura and M L Dyall-Smith ldquoTaxonomy of the familyHalobacteriaceae and the description of two new genera Halo-rubrobacterium and Natrialbardquo Journal of General and AppliedMicrobiology vol 41 no 4 pp 333ndash350 1995

[63] T J Carver KM RutherfordM BerrimanM A RajandreamB G Barrell and J Parkhill ldquoACT the Artemis comparisontoolrdquo Bioinformatics vol 21 no 16 pp 3422ndash3423 2005

[64] S W Cline W L Lam R L Charlebois L C Schalkwyk andW F Doolittle ldquoTransformationmethods for halophilic archae-bacteriardquo Canadian Journal of Microbiology vol 35 no 1 pp148ndash152 1989

Archaea 13

show cases of not only provirus integrants or plasmids (egpKH2 and pHH205) but also genomic loci (ViPREs) thatcontain cassettes of virus genes from different sources andat least two cases (described in this study) have likely attsites that indicate circularisation and mobility While thereis a clear relationship between viruses with circular ds- andssDNA genomes that replicate via the rolling circle method(ie the ssDNA form is a replication intermediate) it ismore difficult to explain how capsid gene cassettes can movebetween these viruses and those with linear genomes andterminal proteins Within the pleolipoviruses His2 has alinear genome that contains a viral type B DNA polymeraseand terminal proteins while all the other described membershave circular genomes that contain a rep homolog and prob-ably replicate via the rolling circle method [14 19] Similarlyamong the halosphaeroviruses PH1 SH1 and HHIV-2 havelinear dsDNA with terminal proteins and probably replicatein the same way but the related SNJ1 virus has a circulardsDNA genome The clear relationships shown by the capsidgenes of viruses within each group plus the connectionsshown to haloviruses outside each group (eg the DNA poly-merases of His1 and His2) all speak of a vigorous meansof recombination one that can readily switch capsid genesbetween viruses with radically different replication strategiesHow is the process most likely to operate One possibility issuggested by ViPREs which appear to be mobile collectionsof capsid and replication genes from different sources Theyoffer fixed locations for recombination to occur providegene cassettes that can be reassorted to produce novel virusgenomes and some can probably recombine out as circularforms For example ViPREHmuk1 carries genes related topleolipoviruses halosphaeroviruses and other haloviruses Itis yet to be determined if any of the genes carried in theseloci are expressed in the cell (such as the genes for capsidproteins) or if ViPREs provide any selective advantage to thehost

4 Materials and Methods

41 Water Sample A water sample was collected in 1998from Pink Lake (32∘001015840 S and 115∘301015840 E) a hypersalinelake on Rottnest Island Western Australia Australia It wasscreened in the same year for haloviruses using the methodsdescribed in [8] A single plaque on a Har hispanica lawnplate was picked and replaque purified The novel haloviruswas designated PH1

42 Media Strains and Plasmids The media used in thisstudy are described in the online resource The Halohand-book (httpwwwhaloarchaeacomresourceshalohandbookindexhtml) Artificial salt water containing 30 (wv) totalsalts comprised of 4M NaCl 150mM MgCl

2 150mM

MgSO4 90mM KCl and 35mM CaCl

2and adjusted to pH

75 using sim2mL 1M Tris-HCl (pH 75) per litre MGM con-taining 12 18 or 23 (wv) total salts and HVD (halo-virus diluent) were prepared from the concentrated stock aspreviously described [46] Bacto-agar (Difco Laboratories)was added to MGM for solid (15 gL) or top-layer (7 gL)media

Table 1 lists the haloarchaea used in this study Allhaloarchaea were grown aerobically at 37∘C in either 18 or23 (wv)MGM(depending on the strain) andwith agitation(except for ldquoHar sinaiiensisrdquo)The plasmids used were pBlue-script II KS+ (Stratagene Cloning Systems) pUBP2 [47]and pWL102 [48] pUBP2 and pWL102 were first passagedthrough E coli JM110 [49] to prevent dam-methylationof DNA which has been shown to reduce transformationefficiency in some strains [45]

43 Negative-Stain Electron Microscopy The method fornegative-stain TEM was adapted from that of V Tarasovdescribed in the online resource The Halohandbook (httpwwwhaloarchaeacomresourceshalohandbookindexhtml)A 20120583L drop of the sample was placed on a clean surfaceand the virus particles were allowed to adsorb to a Formvarfilm 400-mesh copper grid (ProSciTech) for 15ndash2min Theywere then negatively stained with a 20120583L drop of 2 (wv)uranyl acetate for 1-2min Excess liquid was absorbed withfilter paper and the grid was allowed to air dry Grids wereexamined either on a Philips CM 120 BioTwin transmissionelectron microscope (Royal Philips Electronics) operating atan accelerating voltage of 120 kV or on a Siemens Elmiskop102 transmission electron microscope (Siemens AG) operat-ing at an accelerating voltage of 120 kV

44 Virus Host Range Twelve haloarchaeal strains from thegenera Haloarcula Halobacterium Haloferax HalorubrumandNatrialba (Table 1)Thirteen naturalHalorubrum isolates(H Camakaris unpublished data) and five uncharacterizedhaloarchaeal isolates fromLakeHardy Pink Lake (inWesternAustralia) and Serpentine Lake (D Walker and M K Seahunpublished data) were screened for PH1 susceptibilityLysates from PH1-infected Har hispanica cultures (1 times1011 PFUmL) were spotted onto lawns of each strain (usingmedia with 12 or 18 (wv)MGM depending on the strain)and incubated for 2ndash5 days at 30 and at 37∘C

45 Large Scale Virus Growth and Purification Liquid cul-tures of PH1 were grown by the infection (MOI 005) of anearly exponential Har hispanica culture in 18 (wv) MGMCultures were incubated aerobically at 37∘C with agitationfor 3 days Clearing (ie complete cell lysis) of PH1-infectedcultures did not occur and consequently cultures were harv-ested when the absorbance at 550 nm reached the minimumand the titre (determined by plaque assay) was at the maxi-mum usually sim1010ndash1011 PFUmL Virus was purified usingthe method described previously for SH1 [8 31] except thatafter the initial low speed spin (Sorvall GSA 6000 rpm30min 10∘C) virus was concentrated from the infected cul-ture by centrifugation at 26000 rpm (13 hr 10∘C) onto acushion of 30 (wv) sucrose in HVDThe pellet was resus-pended in a small volume of HVD and loaded onto a pre-formed linear 5ndash70 (wv) sucrose gradient followed byisopycnic centrifugation in 13 gmL CsCl (Beckman 70Ti60000 rpm 20 hr 10∘C) The white virus band occurredat a density of 129 gmL and was collected and diluted inhalovirus diluent (HVD see Section 4) and the virus pelleted(Beckman SW55 35000 rpm 75min 10∘C) and resuspended

14 Archaea

in a small volume of HVD and stored at 4∘C Virus recoveryat the major stages of a typical purification is given insupplementary Table 1 The specific infectivity of pure virussolutions was determined as the ratio of the PFUmL to theabsorbance at 260 nm

46 PH1 Single-Step Growth Curve An early exponentialphase culture of Har hispanica grown in 18 (wv) MGMwas infected with PH1 (MOI 50) Under these conditions thepercentage of infected cells was approximately 30 After anadsorption period of 1 hr at 37∘C the cells were washed threetimes with 18 (wv) MGM (at room temperature) resus-pended in 100mL 18 (wv) MGM (these methods ensuredthe removal of all residual virus) and incubated at 37∘Cwith shaking (100 rpm) Samples were removed at hourlyintervals for measurements of absorbance at 550 nm andthe number of infective centres Immediately after samplingtitres were determined by plaque assay with an indicator lawnof Har hispanica Each experiment was performed in tripli-cate

47 Halovirus Stability After various treatments sampleswere removed and diluted in HVD and virus titres weredetermined by plaque assay on Har hispanica Each experi-ment was performed in triplicate and representative data areshown Chloroform sensitivity was examined by the exposureof PH1 lysates in 18 (wv) MGM to chloroform in a volumeratio of 1 4 (chloroform to lysate) Incubation was at roomtemperature with constant agitation At appropriate timepoints the mix was allowed to settle and a sample wasremoved from the upper layer and diluted in HVDThe effectof a low ionic environment was examined by diluting PH1lysates (in 18 (wv) MGM) into double distilled H

2O in a

volume ratio of 1 1000 (lysate to double distilledH2O) Incu-

bation was at room temperature with constant agitationSamples were removed at various times and diluted in HVDThe pH stability of PH1 was determined by dilution of PH1lysates in 18 (wv)MGMin the appropriate pHbuffer (HVDbuffered with appropriate Tris-HCl) in a volume ratio of1 100 (lysate to buffer) Incubation was at room temperaturewith constant agitation After 30min samples were removedand diluted in HVD Thermal stability of PH1 was examinedby a 1 hr incubation of a virus lysate in 18 (wv) MGM atdifferent temperatures after which they were brought quicklyto room temperature diluted in HVD and titrated

48 Protein Procedures To remove salts purified virus prepa-rations were mixed with trichloroacetic acid (10 (vv) finalconcentration) and incubated on ice for 15min to allowthe proteins to precipitate After centrifugation (16000 g15min room temperature) the precipitate was washed threetimes in acetone dried and resuspended in double distilledH2O Proteins were dissolved in Laemmli sample buffer with

48mM 120573-mercaptoethanol [50] heated in boiling water for5min then separated on 12 (wv) NuPAGE Novex Bis-Tris Gels using MES-SDS running buffer according to themanufacturerrsquos directions (Invitrogen) After electrophoresisgels were rinsed in double distilled H

2O and stained with

01 (wv) Brilliant Blue G in 40 (vv) methanol and

10 (vv) acetic acid Gels were destained with severalchanges of 40 (vv) methanol and 10 (vv) acetic acidAlternatively gels were stained with GelCode GlycoproteinStaining Kit Pro-Q Emerald 488 Glycoprotein Gel and BlotStain Kit SYPRO Ruby Protein Gel Stain according tothe manufacturerrsquos directions (Invitrogen Molecular ProbesPierce Biotechnology)

Protein bands were cut from the gels and sent to theAustralian Proteome Analysis Facility (Macquarie Univer-sity) for trypsin digestion and analysis by matrix assistedlaser desorption ionisation time of flight mass spectrometry(MALDI-TOF MS) on an Applied Biosystems 4700 Pro-teomics Analyser (Applied Biosystems)

49 DNA Procedures Proteinase K-treated and nonpro-teinaseK-treatedDNApreparationsweremade frompurifiedPH1 using the methods described previously for SH1 [8]DNA was separated on 1 (wv) agarose gels in Tris-acetate-EDTA electrophoresis buffer and was stained with ethidiumbromide (Sigma-Aldrich)120582 DNA DNase I (RNase-free) exonuclease III mung

bean nuclease nuclease BAL-31 T4 DNA polymerase T4DNA ligase T7 exonuclease and type II restriction endonu-cleases were purchased from New England Biolabs RNase Aand yeast transfer RNA were purchased from Sigma-AldrichProteinase K was purchased from Promega Oligonucleotideprimers were purchased from Geneworks Australia

To clone fragments of the PH1 genome purified virusDNA was digested either with AccI Eco0109I and MseI orSmaI restriction endonucleases blunted with DNA Poly-merase I Large (Klenow) Fragment where appropriate andligated into AccI- Eco0109I- or SmaI-digested pBluescript IIKS+ using T4 DNA ligase according to the manufacturerrsquosinstructions (New England Biolabs) The DNA was intro-duced intoE coliXL1-Blue and transformants grown onLuriaagar containing 15 120583gmL tetracycline and 100 120583gmL ampi-cillinThe resulting cloneswere sequenced and the sequenceswere used to design specific oligonucleotide primers for PCRamplification andor primer walking using the virus genome(or specific restriction fragments) as templates

To amplify PH1 nucleic acid approximately 10 ng virusDNA was combined with 500 nM primers 100120583M of eachdNTP 1U Deep Vent DNA Polymerase (New England Bio-labs) and 1 timesThermoPol buffer (New England Biolabs) ina total reaction volume of 50 120583L Template was denatured at95∘C for 10min followed by 30 cycles of denaturation at 95∘Cfor 30 sec annealing at 56∘C for 30 sec extension at 75∘C for2min and a final extension at 75∘C for 10min PCR reactionswere performed on a PxE02ThermoCycler (Thermo ElectroCorporation) Sequencing reactions using 32 pmol primerwere performed by the dideoxy chain termination methodusing ABI PRISM Big Dye Terminator Mix version 31 onan ABI 3100 capillary sequencer (Applied Biosystems) bythe Applied Genetic Diagnostics Sequencing Service at theDepartment of Pathology (The University of Melbourne)

To obtain terminal genomic fragments PH1 DNA wasdigested by the enzyme Pas1 and the head (sim700 bp) andthe tail (sim1300 bp) fragments were purified by the QIAEXII gel extraction kits and treated with 05M piperidine for

Archaea 15

2 hr at 37∘C to remove protein residues The piperidine-treated fragments were sequenced by the Genomics BiotechCompany (Taipei Taiwan) using two primers TGACCA-ATTAATTAGGCCGGTTCGCC (PH1 head-R) and GTGC-CATACTGCTACAATTCT (PH1 tail-F)

Four hundred fifty-four whole genome sequencing PH1DNA samples (sim5 120583g) were sequenced using parallel pyrose-quencing on a Roche 454 Genome Sequencer System atMission Biotech (Taipei Taiwan) The largest contig was27399with 8803 reads (Newbler version 27) with 27387 posi-tions being of Q40 quality The average depth of coverage foreach base was 829TheGenBank accession for the entire PH1sequence is KC252997

Acknowledgments

The authors thank Carolyn Bath Mei Kwei (Joan) Seahand Danielle Walker for their early work in PH1 characteri-zation Simon Crawford Jocelyn Carpenter and Anna Fried-huber for their assistance with the transmission electronmicroscopy and Tiffany Cowie and Voula Kanellakis fortheir sequencing assistance This paper has been facilitatedby access to the Australian Proteome Analysis Facility estab-lished under the Australian Governmentrsquos Major NationalResearch Facilities program SLT was supported by a grantfrom the Biodiversity Research Center Academia SinicaTaiwan

References

[1] M Krupovic M F White P Forterre and D PrangishvilildquoPostcards from the edge structural genomics of archaealvirusesrdquo Advances in Virus Research vol 82 pp 33ndash62 2012

[2] P Stolt andW Zillig ldquoGene regulation in halophage 120601Hmdashmorethan promotersrdquo Systematic and Applied Microbiology vol 16no 4 pp 591ndash596 1994

[3] R Klein U Baranyi N Rossler B Greineder H Scholz and AWitte ldquoNatrialba magadii virus 120593Ch1 first complete nucleotidesequence and functional organization of a virus infecting ahaloalkaliphilic archaeonrdquo Molecular Microbiology vol 45 no3 pp 851ndash863 2002

[4] E Pagaling R D Haigh W D Grant et al ldquoSequence analysisof an Archaeal virus isolated from a hypersaline lake in InnerMongolia Chinardquo BMC Genomics vol 8 article 410 2007

[5] S L Tang SNuttall andMDyall-Smith ldquoHalovirusesHF1 andHF2 evidence for a recent and large recombination eventrdquo Jour-nal of Bacteriology vol 186 no 9 pp 2810ndash2817 2004

[6] C Bath and M L Dyall-smith ldquoHis1 an archaeal virus of theFuselloviridae family that infectsHaloarcula hispanicardquo Journalof Virology vol 72 no 11 pp 9392ndash9395 1998

[7] C Bath T Cukalac K Porter andM L Dyall-Smith ldquoHis1 andHis2 are distantly related spindle-shaped haloviruses belongingto the novel virus group Salterprovirusrdquo Virology vol 350 no1 pp 228ndash239 2006

[8] K Porter P Kukkaro J K H Bamford et al ldquoSH1 a novelspherical halovirus isolated from an Australian hypersalinelakerdquo Virology vol 335 no 1 pp 22ndash33 2005

[9] M Dyall-Smith S L Tang and C Bath ldquoHaloarchaeal viruseshow diverse are theyrdquo Research in Microbiology vol 154 no 4pp 309ndash313 2003

[10] A Oren G Bratbak and M Heldal ldquoOccurrence of virus-likeparticles in the Dead Seardquo Extremophiles vol 1 no 3 pp 143ndash149 1997

[11] T Sime-Ngando S Lucas A Robin et al et al ldquoDiversityof virus-host systems in hypersaline Lake Retba SenegalrdquoEnvironmental Microbiology vol 13 no 8 pp 1956ndash1972 2010

[12] N Guixa-Boixareu J I Calderon-Paz M Heldal G Bratbakand C Pedros-Alio ldquoViral lysis and bacterivory as prokaryoticloss factors along a salinity gradientrdquoAquaticMicrobial Ecologyvol 11 no 3 pp 215ndash227 1996

[13] S T Jaakkola R K Penttinen S T Vilen et al ldquoClosely relatedarchaeal Haloarcula hispanica icosahedral viruses HHIV-2 andSH1 have nonhomologous genes encoding host recognitionfunctionsrdquo Journal of Virology vol 86 no 9 pp 4734ndash47422012

[14] E Roine P Kukkaro L Paulin et al ldquoNew closely related halo-archaeal viral elements with different nucleic acid typesrdquo Jour-nal of Virology vol 84 no 7 pp 3682ndash3689 2010

[15] A Sencilo L Paulin S KellnerMHelm and E Roine ldquoRelatedhaloarchaeal pleomorphic viruses contain different genometypesrdquo Nucleic Acids Research vol 40 no 12 pp 5523ndash55342012

[16] H T Jaalinoja E Roine P Laurinmaki H M Kivela D HBamford and S J Butcher ldquoStructure and host-cell interactionof SH1 a membrane-containing halophilic euryarchaeal virusrdquoProceedings of the National Academy of Sciences of the UnitedStates of America vol 105 no 23 pp 8008ndash8013 2008

[17] M K Pietila N S Atanasova VManole et al ldquoVirion architec-ture unifies globally distributed pleolipoviruses infecting halo-philic archaeardquo Journal of Virology vol 86 no 9 pp 5067ndash50792012

[18] M L Holmes and M L Dyall-Smith ldquoA plasmid vector witha selectable marker for halophilic archaebacteriardquo Journal ofBacteriology vol 172 no 2 pp 756ndash761 1990

[19] M L Holmes F Pfeifer andM L Dyall-Smith ldquoAnalysis of thehalobacterial plasmid pHK2 minimal repliconrdquo Gene vol 153no 1 pp 117ndash121 1995

[20] M L Dyall-Smith F Pfeiffer K Klee et al ldquoHaloquadratumwalsbyi limited diversity in a Global Pondrdquo PLoS ONE vol 6no 6 Article ID e20968 2011

[21] S Casjens ldquoProphages and bacterial genomics what have welearned so farrdquoMolecular Microbiology vol 49 no 2 pp 277ndash300 2003

[22] R W Hendrix M C M Smith R N Burns M E Ford andG F Hatfull ldquoEvolutionary relationships among diverse bacte-riophages and prophages all the worldrsquos a phagerdquo Proceedings ofthe National Academy of Sciences of the United States of Americavol 96 no 5 pp 2192ndash2197 1999

[23] D Botstein ldquoA theory ofmodular evolution for bacteriophagesrdquoAnnals of the New York Academy of Sciences vol 354 pp 484ndash491 1980

[24] M Krupovic S Gribaldo D H Bamford and P Forterre ldquoTheevolutionary history of archaeal MCM helicases a case study ofvertical evolution combinedwithHitchhiking ofmobile geneticelementsrdquo Molecular Biology and Evolution vol 27 no 12 pp2716ndash2732 2010

[25] M Krupovic P Forterre and D H Bamford ldquoComparativeanalysis of the mosaic genomes of tailed archaeal viruses andproviruses suggests common themes for virion architecture andassembly with tailed viruses of bacteriardquo Journal of MolecularBiology vol 397 no 1 pp 144ndash160 2010

16 Archaea

[26] Z Zhang Y Liu S Wang et al et al ldquoTemperate membrane-containing halophilic archaeal virus SNJ1 has a circular dsDNAgenome identical to that of plasmid pHH205rdquoVirology vol 434no 2 pp 233ndash241 2012

[27] D H Bamford J J Ravantti G Ronnholm et al ldquoConstituentsof SH1 a novel lipid-containing virus infecting the halophiliceuryarchaeonHaloarcula hispanicardquo Journal of Virology vol 79no 14 pp 9097ndash9107 2005

[28] H M Kivela E Roine P Kukkaro S Laurinavicius P Somer-harju andDH Bamford ldquoQuantitative dissociation of archaealvirus SH1 reveals distinct capsid proteins and a lipid corerdquoVirology vol 356 no 1-2 pp 4ndash11 2006

[29] H Jaalinoja Electron Cryo-Microscopy Studies of Bacteriophage8 and Archaeal Virus SH1 University of Helsinki HelsinkiFinland 2007

[30] K Porter B E Russ J Yang andM L Dyall-Smith ldquoThe trans-cription programme of the protein-primed halovirus SH1rdquoMi-crobiology vol 154 no 11 pp 3599ndash3608 2008

[31] K Porter and M L Dyall-Smith ldquoTransfection of haloarchaeaby the DNAs of spindle and round haloviruses and the useof transposon mutagenesis to identify non-essential regionsrdquoMolecular Microbiology vol 70 no 5 pp 1236ndash1245 2008

[32] S E Luria General Virology John Wiley and Sons New YorkNY USA 1953

[33] C A Thomas K Saigo E McLeod and J Ito ldquoThe separationofDNA segments attached to proteinsrdquoAnalytical Biochemistryvol 93 pp 158ndash166 1979

[34] A L Delcher D Harmon S Kasif OWhite and S L SalzbergldquoImproved microbial gene identification with GLIMMERrdquoNucleic Acids Research vol 27 no 23 pp 4636ndash4641 1999

[35] R Zhang Y Yang P Fang et al ldquoDiversity of telomere palin-dromic sequences and replication genes among Streptomyceslinear plasmidsrdquo Applied and Environmental Microbiology vol72 no 9 pp 5728ndash5733 2006

[36] X Wang H S Lee F J Sugar F E Jenney M W W Adamsand J H Prestegard ldquoPF0610 a novel winged helix-turn-helix variant possessing a rubredoxin-like Zn ribbon motiffrom the hyperthermophilic archaeon Pyrococcus furiosusrdquoBiochemistry vol 46 no 3 pp 752ndash761 2007

[37] S W Cline L C Schalkwyk and W F Doolittle ldquoTransfor-mation of the archaebacterium Halobacterium volcanii withgenomic DNArdquo Journal of Bacteriology vol 171 no 9 pp 4987ndash4991 1989

[38] S Kamtekar A J Berman J Wang et al ldquoThe 12059329 DNA poly-merase protein-primer structure suggests a model for the ini-tiation to elongation transitionrdquo EMBO Journal vol 25 no 6pp 1335ndash1343 2006

[39] X Peng T Basta M Haring R A Garrett and D PrangishvilildquoGenome of theAcidianus bottle-shaped virus and insights intothe replication and packaging mechanismsrdquo Virology vol 364no 1 pp 237ndash243 2007

[40] C C Yang C H Huang C Y Li Y G Tsay S C Lee and CW Chen ldquoThe terminal proteins of linear Streptomyces chrom-osomes and plasmids a novel class of replication primingproteinsrdquo Molecular Microbiology vol 43 no 2 pp 297ndash3052002

[41] C C Yang Y H Chen H H Tsai C H Huang T W Huangand C W Chen ldquoIn vitro deoxynucleotidylation of the term-inal protein of Streptomyces linear chromosomesrdquo Applied andEnvironmental Microbiology vol 72 no 12 pp 7959ndash79612006

[42] P C Chang and S N Cohen ldquoBidirectional replication froman internal origin in a linear Streptomyces plasmidrdquo Science vol265 no 5174 pp 952ndash954 1994

[43] S W Wilhelm W H Jeffrey C A Suttle and D L MitchellldquoEstimation of biologically damaging UV levels in marinesurface waters with DNA and viral dosimetersrdquo Photochemistryand Photobiology vol 76 no 3 pp 268ndash273 2002

[44] T Allers and M Mevarech ldquoArchaeal geneticsmdashthe third wayrdquoNature Reviews Genetics vol 6 no 1 pp 58ndash73 2005

[45] M L Holmes S D Nuttall and M L Dyall-Smith ldquoConstruc-tion and use of halobacterial shuttle vectors and further studieson Haloferax DNA gyraserdquo Journal of Bacteriology vol 173 no12 pp 3807ndash3813 1991

[46] S D Nuttall and M L Dyall-Smith ldquoHF1 and HF2 novelbacteriophages of halophilic archaeardquo Virology vol 197 no 2pp 678ndash684 1993

[47] U Blaseio and F Pfeifer ldquoTransformation of Halobacteriumhalobium development of vectors and investigation of gas vesi-cle synthesisrdquo Proceedings of the National Academy of Sciences ofthe United States of America vol 87 no 17 pp 6772ndash6776 1990

[48] W L Lam and W F Doolittle ldquoShuttle vectors for the archae-bacterium Halobacterium volcaniirdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 86 no14 pp 5478ndash5482 1989

[49] C Yanisch-Perron J Vieira and J Messing ldquoImproved M13phage cloning vectors and host strains nucleotide sequences ofthe M13mp18 and pUC19 vectorsrdquo Gene vol 33 no 1 pp 103ndash119 1985

[50] U K Laemmli ldquoCleavage of structural proteins during theassembly of the head of bacteriophage T4rdquo Nature vol 227 no5259 pp 680ndash685 1970

[51] D G Burns H M Camakaris P H Janssen and M L Dyall-Smith ldquoCombined use of cultivation-dependent and culti-vation-independent methods indicates that members of mosthaloarchaeal groups in anAustralian crystallizer pond are culti-vablerdquo Applied and Environmental Microbiology vol 70 no 9pp 5258ndash5265 2004

[52] G Juez F Rodriguez-Valera A Ventosa and D J KushnerldquoHaloarcula hispanica spec nov and Haloferax gibbonsii specnov two new species of extremely halophilic archaebacteriardquoSystematic and Applied Microbiology vol 8 pp 75ndash79 1986

[53] A Oren M Ginzburg B Z Ginzburg L I Hochstein and BE Volcani ldquoHaloarcula marismortui (Volcani) sp nov nomrev an extremely halophilic bacterium from the Dead SeardquoInternational Journal of Systematic Bacteriology vol 40 no 2pp 209ndash210 1990

[54] M Torreblanca F Rodriguez-Valera G Juez A Ventosa MKamekura and M Kates ldquoClassification of non-alkaliphilichalobacteria based on numerical taxonomy and polar lipidcomposition and description of Haloarcula gen nov andHaloferax gen novrdquo Systematic and Applied Microbiology vol8 pp 89ndash99 1986

[55] A Ventosa and A Oren ldquoHalobacterium salinarum nom cor-rig a name to replace Halobacterium salinarium (Elazari-Vol-cani) and to include Halobacterium halobium and Halobac-terium cutirubrumrdquo International Journal of Systematic Bacte-riology vol 46 no 1 p 347 1996

[56] M C Gutierrez M Kamekura M L Holmes M L Dyall-Smith and A Ventosa ldquoTaxonomic characterization of Halo-ferax sp (ldquoH alicanteirdquo) strain Aa 22 description of Haloferaxlucentensis sp novrdquo Extremophiles vol 6 no 6 pp 479ndash4832002

Archaea 17

[57] M F Mullakhanbhai and H Larsen ldquoHalobacterium volcaniispec nov a Dead Sea halobacterium with a moderate saltrequirementrdquo Archives of Microbiology vol 104 no 1 pp 207ndash214 1975

[58] S D Nuttall and M L Dyall-Smith ldquoCh2 a novel halophilicarchaeon from an Australian solar salternrdquo International Jour-nal of Systematic Bacteriology vol 43 no 4 pp 729ndash734 1993

[59] P D Franzman E Stackebrandt K Sanderson et al ldquoHalobac-terium lacusprofundi sp nov a halophilic bacterium isolatedfrom Deep Lake Antarcticardquo Systematic and Applied Microbi-ology vol 11 pp 20ndash27 1988

[60] G A Tomlinson and L I Hochstein ldquoHalobacterium saccharo-vorum sp nov a carbohydrate metabolizing extremely halo-philic bacteriumrdquoCanadian Journal ofMicrobiology vol 22 no4 pp 587ndash591 1976

[61] A Ventosa M C Gutierrez M Kamekura and M L Dyall-Smith ldquoProposal to transfer Halococcus turkmenicus Halobac-terium trapanicum JCM9743 and strainGSL-11 toHaloterrigenaturkmenica gen nov comb novrdquo International Journal ofSystematic Bacteriology vol 49 no 1 pp 131ndash136 1999

[62] M Kamekura and M L Dyall-Smith ldquoTaxonomy of the familyHalobacteriaceae and the description of two new genera Halo-rubrobacterium and Natrialbardquo Journal of General and AppliedMicrobiology vol 41 no 4 pp 333ndash350 1995

[63] T J Carver KM RutherfordM BerrimanM A RajandreamB G Barrell and J Parkhill ldquoACT the Artemis comparisontoolrdquo Bioinformatics vol 21 no 16 pp 3422ndash3423 2005

[64] S W Cline W L Lam R L Charlebois L C Schalkwyk andW F Doolittle ldquoTransformationmethods for halophilic archae-bacteriardquo Canadian Journal of Microbiology vol 35 no 1 pp148ndash152 1989

14 Archaea

in a small volume of HVD and stored at 4∘C Virus recoveryat the major stages of a typical purification is given insupplementary Table 1 The specific infectivity of pure virussolutions was determined as the ratio of the PFUmL to theabsorbance at 260 nm

46 PH1 Single-Step Growth Curve An early exponentialphase culture of Har hispanica grown in 18 (wv) MGMwas infected with PH1 (MOI 50) Under these conditions thepercentage of infected cells was approximately 30 After anadsorption period of 1 hr at 37∘C the cells were washed threetimes with 18 (wv) MGM (at room temperature) resus-pended in 100mL 18 (wv) MGM (these methods ensuredthe removal of all residual virus) and incubated at 37∘Cwith shaking (100 rpm) Samples were removed at hourlyintervals for measurements of absorbance at 550 nm andthe number of infective centres Immediately after samplingtitres were determined by plaque assay with an indicator lawnof Har hispanica Each experiment was performed in tripli-cate

47 Halovirus Stability After various treatments sampleswere removed and diluted in HVD and virus titres weredetermined by plaque assay on Har hispanica Each experi-ment was performed in triplicate and representative data areshown Chloroform sensitivity was examined by the exposureof PH1 lysates in 18 (wv) MGM to chloroform in a volumeratio of 1 4 (chloroform to lysate) Incubation was at roomtemperature with constant agitation At appropriate timepoints the mix was allowed to settle and a sample wasremoved from the upper layer and diluted in HVDThe effectof a low ionic environment was examined by diluting PH1lysates (in 18 (wv) MGM) into double distilled H

2O in a

volume ratio of 1 1000 (lysate to double distilledH2O) Incu-

bation was at room temperature with constant agitationSamples were removed at various times and diluted in HVDThe pH stability of PH1 was determined by dilution of PH1lysates in 18 (wv)MGMin the appropriate pHbuffer (HVDbuffered with appropriate Tris-HCl) in a volume ratio of1 100 (lysate to buffer) Incubation was at room temperaturewith constant agitation After 30min samples were removedand diluted in HVD Thermal stability of PH1 was examinedby a 1 hr incubation of a virus lysate in 18 (wv) MGM atdifferent temperatures after which they were brought quicklyto room temperature diluted in HVD and titrated

48 Protein Procedures To remove salts purified virus prepa-rations were mixed with trichloroacetic acid (10 (vv) finalconcentration) and incubated on ice for 15min to allowthe proteins to precipitate After centrifugation (16000 g15min room temperature) the precipitate was washed threetimes in acetone dried and resuspended in double distilledH2O Proteins were dissolved in Laemmli sample buffer with

48mM 120573-mercaptoethanol [50] heated in boiling water for5min then separated on 12 (wv) NuPAGE Novex Bis-Tris Gels using MES-SDS running buffer according to themanufacturerrsquos directions (Invitrogen) After electrophoresisgels were rinsed in double distilled H

2O and stained with

01 (wv) Brilliant Blue G in 40 (vv) methanol and

10 (vv) acetic acid Gels were destained with severalchanges of 40 (vv) methanol and 10 (vv) acetic acidAlternatively gels were stained with GelCode GlycoproteinStaining Kit Pro-Q Emerald 488 Glycoprotein Gel and BlotStain Kit SYPRO Ruby Protein Gel Stain according tothe manufacturerrsquos directions (Invitrogen Molecular ProbesPierce Biotechnology)

Protein bands were cut from the gels and sent to theAustralian Proteome Analysis Facility (Macquarie Univer-sity) for trypsin digestion and analysis by matrix assistedlaser desorption ionisation time of flight mass spectrometry(MALDI-TOF MS) on an Applied Biosystems 4700 Pro-teomics Analyser (Applied Biosystems)

49 DNA Procedures Proteinase K-treated and nonpro-teinaseK-treatedDNApreparationsweremade frompurifiedPH1 using the methods described previously for SH1 [8]DNA was separated on 1 (wv) agarose gels in Tris-acetate-EDTA electrophoresis buffer and was stained with ethidiumbromide (Sigma-Aldrich)120582 DNA DNase I (RNase-free) exonuclease III mung

bean nuclease nuclease BAL-31 T4 DNA polymerase T4DNA ligase T7 exonuclease and type II restriction endonu-cleases were purchased from New England Biolabs RNase Aand yeast transfer RNA were purchased from Sigma-AldrichProteinase K was purchased from Promega Oligonucleotideprimers were purchased from Geneworks Australia

To clone fragments of the PH1 genome purified virusDNA was digested either with AccI Eco0109I and MseI orSmaI restriction endonucleases blunted with DNA Poly-merase I Large (Klenow) Fragment where appropriate andligated into AccI- Eco0109I- or SmaI-digested pBluescript IIKS+ using T4 DNA ligase according to the manufacturerrsquosinstructions (New England Biolabs) The DNA was intro-duced intoE coliXL1-Blue and transformants grown onLuriaagar containing 15 120583gmL tetracycline and 100 120583gmL ampi-cillinThe resulting cloneswere sequenced and the sequenceswere used to design specific oligonucleotide primers for PCRamplification andor primer walking using the virus genome(or specific restriction fragments) as templates

To amplify PH1 nucleic acid approximately 10 ng virusDNA was combined with 500 nM primers 100120583M of eachdNTP 1U Deep Vent DNA Polymerase (New England Bio-labs) and 1 timesThermoPol buffer (New England Biolabs) ina total reaction volume of 50 120583L Template was denatured at95∘C for 10min followed by 30 cycles of denaturation at 95∘Cfor 30 sec annealing at 56∘C for 30 sec extension at 75∘C for2min and a final extension at 75∘C for 10min PCR reactionswere performed on a PxE02ThermoCycler (Thermo ElectroCorporation) Sequencing reactions using 32 pmol primerwere performed by the dideoxy chain termination methodusing ABI PRISM Big Dye Terminator Mix version 31 onan ABI 3100 capillary sequencer (Applied Biosystems) bythe Applied Genetic Diagnostics Sequencing Service at theDepartment of Pathology (The University of Melbourne)

To obtain terminal genomic fragments PH1 DNA wasdigested by the enzyme Pas1 and the head (sim700 bp) andthe tail (sim1300 bp) fragments were purified by the QIAEXII gel extraction kits and treated with 05M piperidine for

Archaea 15

2 hr at 37∘C to remove protein residues The piperidine-treated fragments were sequenced by the Genomics BiotechCompany (Taipei Taiwan) using two primers TGACCA-ATTAATTAGGCCGGTTCGCC (PH1 head-R) and GTGC-CATACTGCTACAATTCT (PH1 tail-F)

Four hundred fifty-four whole genome sequencing PH1DNA samples (sim5 120583g) were sequenced using parallel pyrose-quencing on a Roche 454 Genome Sequencer System atMission Biotech (Taipei Taiwan) The largest contig was27399with 8803 reads (Newbler version 27) with 27387 posi-tions being of Q40 quality The average depth of coverage foreach base was 829TheGenBank accession for the entire PH1sequence is KC252997

Acknowledgments

The authors thank Carolyn Bath Mei Kwei (Joan) Seahand Danielle Walker for their early work in PH1 characteri-zation Simon Crawford Jocelyn Carpenter and Anna Fried-huber for their assistance with the transmission electronmicroscopy and Tiffany Cowie and Voula Kanellakis fortheir sequencing assistance This paper has been facilitatedby access to the Australian Proteome Analysis Facility estab-lished under the Australian Governmentrsquos Major NationalResearch Facilities program SLT was supported by a grantfrom the Biodiversity Research Center Academia SinicaTaiwan

References

[1] M Krupovic M F White P Forterre and D PrangishvilildquoPostcards from the edge structural genomics of archaealvirusesrdquo Advances in Virus Research vol 82 pp 33ndash62 2012

[2] P Stolt andW Zillig ldquoGene regulation in halophage 120601Hmdashmorethan promotersrdquo Systematic and Applied Microbiology vol 16no 4 pp 591ndash596 1994

[3] R Klein U Baranyi N Rossler B Greineder H Scholz and AWitte ldquoNatrialba magadii virus 120593Ch1 first complete nucleotidesequence and functional organization of a virus infecting ahaloalkaliphilic archaeonrdquo Molecular Microbiology vol 45 no3 pp 851ndash863 2002

[4] E Pagaling R D Haigh W D Grant et al ldquoSequence analysisof an Archaeal virus isolated from a hypersaline lake in InnerMongolia Chinardquo BMC Genomics vol 8 article 410 2007

[5] S L Tang SNuttall andMDyall-Smith ldquoHalovirusesHF1 andHF2 evidence for a recent and large recombination eventrdquo Jour-nal of Bacteriology vol 186 no 9 pp 2810ndash2817 2004

[6] C Bath and M L Dyall-smith ldquoHis1 an archaeal virus of theFuselloviridae family that infectsHaloarcula hispanicardquo Journalof Virology vol 72 no 11 pp 9392ndash9395 1998

[7] C Bath T Cukalac K Porter andM L Dyall-Smith ldquoHis1 andHis2 are distantly related spindle-shaped haloviruses belongingto the novel virus group Salterprovirusrdquo Virology vol 350 no1 pp 228ndash239 2006

[8] K Porter P Kukkaro J K H Bamford et al ldquoSH1 a novelspherical halovirus isolated from an Australian hypersalinelakerdquo Virology vol 335 no 1 pp 22ndash33 2005

[9] M Dyall-Smith S L Tang and C Bath ldquoHaloarchaeal viruseshow diverse are theyrdquo Research in Microbiology vol 154 no 4pp 309ndash313 2003

[10] A Oren G Bratbak and M Heldal ldquoOccurrence of virus-likeparticles in the Dead Seardquo Extremophiles vol 1 no 3 pp 143ndash149 1997

[11] T Sime-Ngando S Lucas A Robin et al et al ldquoDiversityof virus-host systems in hypersaline Lake Retba SenegalrdquoEnvironmental Microbiology vol 13 no 8 pp 1956ndash1972 2010

[12] N Guixa-Boixareu J I Calderon-Paz M Heldal G Bratbakand C Pedros-Alio ldquoViral lysis and bacterivory as prokaryoticloss factors along a salinity gradientrdquoAquaticMicrobial Ecologyvol 11 no 3 pp 215ndash227 1996

[13] S T Jaakkola R K Penttinen S T Vilen et al ldquoClosely relatedarchaeal Haloarcula hispanica icosahedral viruses HHIV-2 andSH1 have nonhomologous genes encoding host recognitionfunctionsrdquo Journal of Virology vol 86 no 9 pp 4734ndash47422012

[14] E Roine P Kukkaro L Paulin et al ldquoNew closely related halo-archaeal viral elements with different nucleic acid typesrdquo Jour-nal of Virology vol 84 no 7 pp 3682ndash3689 2010

[15] A Sencilo L Paulin S KellnerMHelm and E Roine ldquoRelatedhaloarchaeal pleomorphic viruses contain different genometypesrdquo Nucleic Acids Research vol 40 no 12 pp 5523ndash55342012

[16] H T Jaalinoja E Roine P Laurinmaki H M Kivela D HBamford and S J Butcher ldquoStructure and host-cell interactionof SH1 a membrane-containing halophilic euryarchaeal virusrdquoProceedings of the National Academy of Sciences of the UnitedStates of America vol 105 no 23 pp 8008ndash8013 2008

[17] M K Pietila N S Atanasova VManole et al ldquoVirion architec-ture unifies globally distributed pleolipoviruses infecting halo-philic archaeardquo Journal of Virology vol 86 no 9 pp 5067ndash50792012

[18] M L Holmes and M L Dyall-Smith ldquoA plasmid vector witha selectable marker for halophilic archaebacteriardquo Journal ofBacteriology vol 172 no 2 pp 756ndash761 1990

[19] M L Holmes F Pfeifer andM L Dyall-Smith ldquoAnalysis of thehalobacterial plasmid pHK2 minimal repliconrdquo Gene vol 153no 1 pp 117ndash121 1995

[20] M L Dyall-Smith F Pfeiffer K Klee et al ldquoHaloquadratumwalsbyi limited diversity in a Global Pondrdquo PLoS ONE vol 6no 6 Article ID e20968 2011

[21] S Casjens ldquoProphages and bacterial genomics what have welearned so farrdquoMolecular Microbiology vol 49 no 2 pp 277ndash300 2003

[22] R W Hendrix M C M Smith R N Burns M E Ford andG F Hatfull ldquoEvolutionary relationships among diverse bacte-riophages and prophages all the worldrsquos a phagerdquo Proceedings ofthe National Academy of Sciences of the United States of Americavol 96 no 5 pp 2192ndash2197 1999

[23] D Botstein ldquoA theory ofmodular evolution for bacteriophagesrdquoAnnals of the New York Academy of Sciences vol 354 pp 484ndash491 1980

[24] M Krupovic S Gribaldo D H Bamford and P Forterre ldquoTheevolutionary history of archaeal MCM helicases a case study ofvertical evolution combinedwithHitchhiking ofmobile geneticelementsrdquo Molecular Biology and Evolution vol 27 no 12 pp2716ndash2732 2010

[25] M Krupovic P Forterre and D H Bamford ldquoComparativeanalysis of the mosaic genomes of tailed archaeal viruses andproviruses suggests common themes for virion architecture andassembly with tailed viruses of bacteriardquo Journal of MolecularBiology vol 397 no 1 pp 144ndash160 2010

16 Archaea

[26] Z Zhang Y Liu S Wang et al et al ldquoTemperate membrane-containing halophilic archaeal virus SNJ1 has a circular dsDNAgenome identical to that of plasmid pHH205rdquoVirology vol 434no 2 pp 233ndash241 2012

[27] D H Bamford J J Ravantti G Ronnholm et al ldquoConstituentsof SH1 a novel lipid-containing virus infecting the halophiliceuryarchaeonHaloarcula hispanicardquo Journal of Virology vol 79no 14 pp 9097ndash9107 2005

[28] H M Kivela E Roine P Kukkaro S Laurinavicius P Somer-harju andDH Bamford ldquoQuantitative dissociation of archaealvirus SH1 reveals distinct capsid proteins and a lipid corerdquoVirology vol 356 no 1-2 pp 4ndash11 2006

[29] H Jaalinoja Electron Cryo-Microscopy Studies of Bacteriophage8 and Archaeal Virus SH1 University of Helsinki HelsinkiFinland 2007

[30] K Porter B E Russ J Yang andM L Dyall-Smith ldquoThe trans-cription programme of the protein-primed halovirus SH1rdquoMi-crobiology vol 154 no 11 pp 3599ndash3608 2008

[31] K Porter and M L Dyall-Smith ldquoTransfection of haloarchaeaby the DNAs of spindle and round haloviruses and the useof transposon mutagenesis to identify non-essential regionsrdquoMolecular Microbiology vol 70 no 5 pp 1236ndash1245 2008

[32] S E Luria General Virology John Wiley and Sons New YorkNY USA 1953

[33] C A Thomas K Saigo E McLeod and J Ito ldquoThe separationofDNA segments attached to proteinsrdquoAnalytical Biochemistryvol 93 pp 158ndash166 1979

[34] A L Delcher D Harmon S Kasif OWhite and S L SalzbergldquoImproved microbial gene identification with GLIMMERrdquoNucleic Acids Research vol 27 no 23 pp 4636ndash4641 1999

[35] R Zhang Y Yang P Fang et al ldquoDiversity of telomere palin-dromic sequences and replication genes among Streptomyceslinear plasmidsrdquo Applied and Environmental Microbiology vol72 no 9 pp 5728ndash5733 2006

[36] X Wang H S Lee F J Sugar F E Jenney M W W Adamsand J H Prestegard ldquoPF0610 a novel winged helix-turn-helix variant possessing a rubredoxin-like Zn ribbon motiffrom the hyperthermophilic archaeon Pyrococcus furiosusrdquoBiochemistry vol 46 no 3 pp 752ndash761 2007

[37] S W Cline L C Schalkwyk and W F Doolittle ldquoTransfor-mation of the archaebacterium Halobacterium volcanii withgenomic DNArdquo Journal of Bacteriology vol 171 no 9 pp 4987ndash4991 1989

[38] S Kamtekar A J Berman J Wang et al ldquoThe 12059329 DNA poly-merase protein-primer structure suggests a model for the ini-tiation to elongation transitionrdquo EMBO Journal vol 25 no 6pp 1335ndash1343 2006

[39] X Peng T Basta M Haring R A Garrett and D PrangishvilildquoGenome of theAcidianus bottle-shaped virus and insights intothe replication and packaging mechanismsrdquo Virology vol 364no 1 pp 237ndash243 2007

[40] C C Yang C H Huang C Y Li Y G Tsay S C Lee and CW Chen ldquoThe terminal proteins of linear Streptomyces chrom-osomes and plasmids a novel class of replication primingproteinsrdquo Molecular Microbiology vol 43 no 2 pp 297ndash3052002

[41] C C Yang Y H Chen H H Tsai C H Huang T W Huangand C W Chen ldquoIn vitro deoxynucleotidylation of the term-inal protein of Streptomyces linear chromosomesrdquo Applied andEnvironmental Microbiology vol 72 no 12 pp 7959ndash79612006

[42] P C Chang and S N Cohen ldquoBidirectional replication froman internal origin in a linear Streptomyces plasmidrdquo Science vol265 no 5174 pp 952ndash954 1994

[43] S W Wilhelm W H Jeffrey C A Suttle and D L MitchellldquoEstimation of biologically damaging UV levels in marinesurface waters with DNA and viral dosimetersrdquo Photochemistryand Photobiology vol 76 no 3 pp 268ndash273 2002

[44] T Allers and M Mevarech ldquoArchaeal geneticsmdashthe third wayrdquoNature Reviews Genetics vol 6 no 1 pp 58ndash73 2005

[45] M L Holmes S D Nuttall and M L Dyall-Smith ldquoConstruc-tion and use of halobacterial shuttle vectors and further studieson Haloferax DNA gyraserdquo Journal of Bacteriology vol 173 no12 pp 3807ndash3813 1991

[46] S D Nuttall and M L Dyall-Smith ldquoHF1 and HF2 novelbacteriophages of halophilic archaeardquo Virology vol 197 no 2pp 678ndash684 1993

[47] U Blaseio and F Pfeifer ldquoTransformation of Halobacteriumhalobium development of vectors and investigation of gas vesi-cle synthesisrdquo Proceedings of the National Academy of Sciences ofthe United States of America vol 87 no 17 pp 6772ndash6776 1990

[48] W L Lam and W F Doolittle ldquoShuttle vectors for the archae-bacterium Halobacterium volcaniirdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 86 no14 pp 5478ndash5482 1989

[49] C Yanisch-Perron J Vieira and J Messing ldquoImproved M13phage cloning vectors and host strains nucleotide sequences ofthe M13mp18 and pUC19 vectorsrdquo Gene vol 33 no 1 pp 103ndash119 1985

[50] U K Laemmli ldquoCleavage of structural proteins during theassembly of the head of bacteriophage T4rdquo Nature vol 227 no5259 pp 680ndash685 1970

[51] D G Burns H M Camakaris P H Janssen and M L Dyall-Smith ldquoCombined use of cultivation-dependent and culti-vation-independent methods indicates that members of mosthaloarchaeal groups in anAustralian crystallizer pond are culti-vablerdquo Applied and Environmental Microbiology vol 70 no 9pp 5258ndash5265 2004

[52] G Juez F Rodriguez-Valera A Ventosa and D J KushnerldquoHaloarcula hispanica spec nov and Haloferax gibbonsii specnov two new species of extremely halophilic archaebacteriardquoSystematic and Applied Microbiology vol 8 pp 75ndash79 1986

[53] A Oren M Ginzburg B Z Ginzburg L I Hochstein and BE Volcani ldquoHaloarcula marismortui (Volcani) sp nov nomrev an extremely halophilic bacterium from the Dead SeardquoInternational Journal of Systematic Bacteriology vol 40 no 2pp 209ndash210 1990

[54] M Torreblanca F Rodriguez-Valera G Juez A Ventosa MKamekura and M Kates ldquoClassification of non-alkaliphilichalobacteria based on numerical taxonomy and polar lipidcomposition and description of Haloarcula gen nov andHaloferax gen novrdquo Systematic and Applied Microbiology vol8 pp 89ndash99 1986

[55] A Ventosa and A Oren ldquoHalobacterium salinarum nom cor-rig a name to replace Halobacterium salinarium (Elazari-Vol-cani) and to include Halobacterium halobium and Halobac-terium cutirubrumrdquo International Journal of Systematic Bacte-riology vol 46 no 1 p 347 1996

[56] M C Gutierrez M Kamekura M L Holmes M L Dyall-Smith and A Ventosa ldquoTaxonomic characterization of Halo-ferax sp (ldquoH alicanteirdquo) strain Aa 22 description of Haloferaxlucentensis sp novrdquo Extremophiles vol 6 no 6 pp 479ndash4832002

Archaea 17

[57] M F Mullakhanbhai and H Larsen ldquoHalobacterium volcaniispec nov a Dead Sea halobacterium with a moderate saltrequirementrdquo Archives of Microbiology vol 104 no 1 pp 207ndash214 1975

[58] S D Nuttall and M L Dyall-Smith ldquoCh2 a novel halophilicarchaeon from an Australian solar salternrdquo International Jour-nal of Systematic Bacteriology vol 43 no 4 pp 729ndash734 1993

[59] P D Franzman E Stackebrandt K Sanderson et al ldquoHalobac-terium lacusprofundi sp nov a halophilic bacterium isolatedfrom Deep Lake Antarcticardquo Systematic and Applied Microbi-ology vol 11 pp 20ndash27 1988

[60] G A Tomlinson and L I Hochstein ldquoHalobacterium saccharo-vorum sp nov a carbohydrate metabolizing extremely halo-philic bacteriumrdquoCanadian Journal ofMicrobiology vol 22 no4 pp 587ndash591 1976

[61] A Ventosa M C Gutierrez M Kamekura and M L Dyall-Smith ldquoProposal to transfer Halococcus turkmenicus Halobac-terium trapanicum JCM9743 and strainGSL-11 toHaloterrigenaturkmenica gen nov comb novrdquo International Journal ofSystematic Bacteriology vol 49 no 1 pp 131ndash136 1999

[62] M Kamekura and M L Dyall-Smith ldquoTaxonomy of the familyHalobacteriaceae and the description of two new genera Halo-rubrobacterium and Natrialbardquo Journal of General and AppliedMicrobiology vol 41 no 4 pp 333ndash350 1995

[63] T J Carver KM RutherfordM BerrimanM A RajandreamB G Barrell and J Parkhill ldquoACT the Artemis comparisontoolrdquo Bioinformatics vol 21 no 16 pp 3422ndash3423 2005

[64] S W Cline W L Lam R L Charlebois L C Schalkwyk andW F Doolittle ldquoTransformationmethods for halophilic archae-bacteriardquo Canadian Journal of Microbiology vol 35 no 1 pp148ndash152 1989

Archaea 15

2 hr at 37∘C to remove protein residues The piperidine-treated fragments were sequenced by the Genomics BiotechCompany (Taipei Taiwan) using two primers TGACCA-ATTAATTAGGCCGGTTCGCC (PH1 head-R) and GTGC-CATACTGCTACAATTCT (PH1 tail-F)

Four hundred fifty-four whole genome sequencing PH1DNA samples (sim5 120583g) were sequenced using parallel pyrose-quencing on a Roche 454 Genome Sequencer System atMission Biotech (Taipei Taiwan) The largest contig was27399with 8803 reads (Newbler version 27) with 27387 posi-tions being of Q40 quality The average depth of coverage foreach base was 829TheGenBank accession for the entire PH1sequence is KC252997

Acknowledgments

The authors thank Carolyn Bath Mei Kwei (Joan) Seahand Danielle Walker for their early work in PH1 characteri-zation Simon Crawford Jocelyn Carpenter and Anna Fried-huber for their assistance with the transmission electronmicroscopy and Tiffany Cowie and Voula Kanellakis fortheir sequencing assistance This paper has been facilitatedby access to the Australian Proteome Analysis Facility estab-lished under the Australian Governmentrsquos Major NationalResearch Facilities program SLT was supported by a grantfrom the Biodiversity Research Center Academia SinicaTaiwan

References

[1] M Krupovic M F White P Forterre and D PrangishvilildquoPostcards from the edge structural genomics of archaealvirusesrdquo Advances in Virus Research vol 82 pp 33ndash62 2012

[2] P Stolt andW Zillig ldquoGene regulation in halophage 120601Hmdashmorethan promotersrdquo Systematic and Applied Microbiology vol 16no 4 pp 591ndash596 1994

[3] R Klein U Baranyi N Rossler B Greineder H Scholz and AWitte ldquoNatrialba magadii virus 120593Ch1 first complete nucleotidesequence and functional organization of a virus infecting ahaloalkaliphilic archaeonrdquo Molecular Microbiology vol 45 no3 pp 851ndash863 2002

[4] E Pagaling R D Haigh W D Grant et al ldquoSequence analysisof an Archaeal virus isolated from a hypersaline lake in InnerMongolia Chinardquo BMC Genomics vol 8 article 410 2007

[5] S L Tang SNuttall andMDyall-Smith ldquoHalovirusesHF1 andHF2 evidence for a recent and large recombination eventrdquo Jour-nal of Bacteriology vol 186 no 9 pp 2810ndash2817 2004

[6] C Bath and M L Dyall-smith ldquoHis1 an archaeal virus of theFuselloviridae family that infectsHaloarcula hispanicardquo Journalof Virology vol 72 no 11 pp 9392ndash9395 1998

[7] C Bath T Cukalac K Porter andM L Dyall-Smith ldquoHis1 andHis2 are distantly related spindle-shaped haloviruses belongingto the novel virus group Salterprovirusrdquo Virology vol 350 no1 pp 228ndash239 2006

[8] K Porter P Kukkaro J K H Bamford et al ldquoSH1 a novelspherical halovirus isolated from an Australian hypersalinelakerdquo Virology vol 335 no 1 pp 22ndash33 2005

[9] M Dyall-Smith S L Tang and C Bath ldquoHaloarchaeal viruseshow diverse are theyrdquo Research in Microbiology vol 154 no 4pp 309ndash313 2003

[10] A Oren G Bratbak and M Heldal ldquoOccurrence of virus-likeparticles in the Dead Seardquo Extremophiles vol 1 no 3 pp 143ndash149 1997

[11] T Sime-Ngando S Lucas A Robin et al et al ldquoDiversityof virus-host systems in hypersaline Lake Retba SenegalrdquoEnvironmental Microbiology vol 13 no 8 pp 1956ndash1972 2010

[12] N Guixa-Boixareu J I Calderon-Paz M Heldal G Bratbakand C Pedros-Alio ldquoViral lysis and bacterivory as prokaryoticloss factors along a salinity gradientrdquoAquaticMicrobial Ecologyvol 11 no 3 pp 215ndash227 1996

[13] S T Jaakkola R K Penttinen S T Vilen et al ldquoClosely relatedarchaeal Haloarcula hispanica icosahedral viruses HHIV-2 andSH1 have nonhomologous genes encoding host recognitionfunctionsrdquo Journal of Virology vol 86 no 9 pp 4734ndash47422012

[14] E Roine P Kukkaro L Paulin et al ldquoNew closely related halo-archaeal viral elements with different nucleic acid typesrdquo Jour-nal of Virology vol 84 no 7 pp 3682ndash3689 2010

[15] A Sencilo L Paulin S KellnerMHelm and E Roine ldquoRelatedhaloarchaeal pleomorphic viruses contain different genometypesrdquo Nucleic Acids Research vol 40 no 12 pp 5523ndash55342012

[16] H T Jaalinoja E Roine P Laurinmaki H M Kivela D HBamford and S J Butcher ldquoStructure and host-cell interactionof SH1 a membrane-containing halophilic euryarchaeal virusrdquoProceedings of the National Academy of Sciences of the UnitedStates of America vol 105 no 23 pp 8008ndash8013 2008

[17] M K Pietila N S Atanasova VManole et al ldquoVirion architec-ture unifies globally distributed pleolipoviruses infecting halo-philic archaeardquo Journal of Virology vol 86 no 9 pp 5067ndash50792012

[18] M L Holmes and M L Dyall-Smith ldquoA plasmid vector witha selectable marker for halophilic archaebacteriardquo Journal ofBacteriology vol 172 no 2 pp 756ndash761 1990

[19] M L Holmes F Pfeifer andM L Dyall-Smith ldquoAnalysis of thehalobacterial plasmid pHK2 minimal repliconrdquo Gene vol 153no 1 pp 117ndash121 1995

[20] M L Dyall-Smith F Pfeiffer K Klee et al ldquoHaloquadratumwalsbyi limited diversity in a Global Pondrdquo PLoS ONE vol 6no 6 Article ID e20968 2011

[21] S Casjens ldquoProphages and bacterial genomics what have welearned so farrdquoMolecular Microbiology vol 49 no 2 pp 277ndash300 2003

[22] R W Hendrix M C M Smith R N Burns M E Ford andG F Hatfull ldquoEvolutionary relationships among diverse bacte-riophages and prophages all the worldrsquos a phagerdquo Proceedings ofthe National Academy of Sciences of the United States of Americavol 96 no 5 pp 2192ndash2197 1999

[23] D Botstein ldquoA theory ofmodular evolution for bacteriophagesrdquoAnnals of the New York Academy of Sciences vol 354 pp 484ndash491 1980

[24] M Krupovic S Gribaldo D H Bamford and P Forterre ldquoTheevolutionary history of archaeal MCM helicases a case study ofvertical evolution combinedwithHitchhiking ofmobile geneticelementsrdquo Molecular Biology and Evolution vol 27 no 12 pp2716ndash2732 2010

[25] M Krupovic P Forterre and D H Bamford ldquoComparativeanalysis of the mosaic genomes of tailed archaeal viruses andproviruses suggests common themes for virion architecture andassembly with tailed viruses of bacteriardquo Journal of MolecularBiology vol 397 no 1 pp 144ndash160 2010

16 Archaea

[26] Z Zhang Y Liu S Wang et al et al ldquoTemperate membrane-containing halophilic archaeal virus SNJ1 has a circular dsDNAgenome identical to that of plasmid pHH205rdquoVirology vol 434no 2 pp 233ndash241 2012

[27] D H Bamford J J Ravantti G Ronnholm et al ldquoConstituentsof SH1 a novel lipid-containing virus infecting the halophiliceuryarchaeonHaloarcula hispanicardquo Journal of Virology vol 79no 14 pp 9097ndash9107 2005

[28] H M Kivela E Roine P Kukkaro S Laurinavicius P Somer-harju andDH Bamford ldquoQuantitative dissociation of archaealvirus SH1 reveals distinct capsid proteins and a lipid corerdquoVirology vol 356 no 1-2 pp 4ndash11 2006

[29] H Jaalinoja Electron Cryo-Microscopy Studies of Bacteriophage8 and Archaeal Virus SH1 University of Helsinki HelsinkiFinland 2007

[30] K Porter B E Russ J Yang andM L Dyall-Smith ldquoThe trans-cription programme of the protein-primed halovirus SH1rdquoMi-crobiology vol 154 no 11 pp 3599ndash3608 2008

[31] K Porter and M L Dyall-Smith ldquoTransfection of haloarchaeaby the DNAs of spindle and round haloviruses and the useof transposon mutagenesis to identify non-essential regionsrdquoMolecular Microbiology vol 70 no 5 pp 1236ndash1245 2008

[32] S E Luria General Virology John Wiley and Sons New YorkNY USA 1953

[33] C A Thomas K Saigo E McLeod and J Ito ldquoThe separationofDNA segments attached to proteinsrdquoAnalytical Biochemistryvol 93 pp 158ndash166 1979

[34] A L Delcher D Harmon S Kasif OWhite and S L SalzbergldquoImproved microbial gene identification with GLIMMERrdquoNucleic Acids Research vol 27 no 23 pp 4636ndash4641 1999

[35] R Zhang Y Yang P Fang et al ldquoDiversity of telomere palin-dromic sequences and replication genes among Streptomyceslinear plasmidsrdquo Applied and Environmental Microbiology vol72 no 9 pp 5728ndash5733 2006

[36] X Wang H S Lee F J Sugar F E Jenney M W W Adamsand J H Prestegard ldquoPF0610 a novel winged helix-turn-helix variant possessing a rubredoxin-like Zn ribbon motiffrom the hyperthermophilic archaeon Pyrococcus furiosusrdquoBiochemistry vol 46 no 3 pp 752ndash761 2007

[37] S W Cline L C Schalkwyk and W F Doolittle ldquoTransfor-mation of the archaebacterium Halobacterium volcanii withgenomic DNArdquo Journal of Bacteriology vol 171 no 9 pp 4987ndash4991 1989

[38] S Kamtekar A J Berman J Wang et al ldquoThe 12059329 DNA poly-merase protein-primer structure suggests a model for the ini-tiation to elongation transitionrdquo EMBO Journal vol 25 no 6pp 1335ndash1343 2006

[39] X Peng T Basta M Haring R A Garrett and D PrangishvilildquoGenome of theAcidianus bottle-shaped virus and insights intothe replication and packaging mechanismsrdquo Virology vol 364no 1 pp 237ndash243 2007

[40] C C Yang C H Huang C Y Li Y G Tsay S C Lee and CW Chen ldquoThe terminal proteins of linear Streptomyces chrom-osomes and plasmids a novel class of replication primingproteinsrdquo Molecular Microbiology vol 43 no 2 pp 297ndash3052002

[41] C C Yang Y H Chen H H Tsai C H Huang T W Huangand C W Chen ldquoIn vitro deoxynucleotidylation of the term-inal protein of Streptomyces linear chromosomesrdquo Applied andEnvironmental Microbiology vol 72 no 12 pp 7959ndash79612006

[42] P C Chang and S N Cohen ldquoBidirectional replication froman internal origin in a linear Streptomyces plasmidrdquo Science vol265 no 5174 pp 952ndash954 1994

[43] S W Wilhelm W H Jeffrey C A Suttle and D L MitchellldquoEstimation of biologically damaging UV levels in marinesurface waters with DNA and viral dosimetersrdquo Photochemistryand Photobiology vol 76 no 3 pp 268ndash273 2002

[44] T Allers and M Mevarech ldquoArchaeal geneticsmdashthe third wayrdquoNature Reviews Genetics vol 6 no 1 pp 58ndash73 2005

[45] M L Holmes S D Nuttall and M L Dyall-Smith ldquoConstruc-tion and use of halobacterial shuttle vectors and further studieson Haloferax DNA gyraserdquo Journal of Bacteriology vol 173 no12 pp 3807ndash3813 1991

[46] S D Nuttall and M L Dyall-Smith ldquoHF1 and HF2 novelbacteriophages of halophilic archaeardquo Virology vol 197 no 2pp 678ndash684 1993

[47] U Blaseio and F Pfeifer ldquoTransformation of Halobacteriumhalobium development of vectors and investigation of gas vesi-cle synthesisrdquo Proceedings of the National Academy of Sciences ofthe United States of America vol 87 no 17 pp 6772ndash6776 1990

[48] W L Lam and W F Doolittle ldquoShuttle vectors for the archae-bacterium Halobacterium volcaniirdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 86 no14 pp 5478ndash5482 1989

[49] C Yanisch-Perron J Vieira and J Messing ldquoImproved M13phage cloning vectors and host strains nucleotide sequences ofthe M13mp18 and pUC19 vectorsrdquo Gene vol 33 no 1 pp 103ndash119 1985

[50] U K Laemmli ldquoCleavage of structural proteins during theassembly of the head of bacteriophage T4rdquo Nature vol 227 no5259 pp 680ndash685 1970

[51] D G Burns H M Camakaris P H Janssen and M L Dyall-Smith ldquoCombined use of cultivation-dependent and culti-vation-independent methods indicates that members of mosthaloarchaeal groups in anAustralian crystallizer pond are culti-vablerdquo Applied and Environmental Microbiology vol 70 no 9pp 5258ndash5265 2004

[52] G Juez F Rodriguez-Valera A Ventosa and D J KushnerldquoHaloarcula hispanica spec nov and Haloferax gibbonsii specnov two new species of extremely halophilic archaebacteriardquoSystematic and Applied Microbiology vol 8 pp 75ndash79 1986

[53] A Oren M Ginzburg B Z Ginzburg L I Hochstein and BE Volcani ldquoHaloarcula marismortui (Volcani) sp nov nomrev an extremely halophilic bacterium from the Dead SeardquoInternational Journal of Systematic Bacteriology vol 40 no 2pp 209ndash210 1990

[54] M Torreblanca F Rodriguez-Valera G Juez A Ventosa MKamekura and M Kates ldquoClassification of non-alkaliphilichalobacteria based on numerical taxonomy and polar lipidcomposition and description of Haloarcula gen nov andHaloferax gen novrdquo Systematic and Applied Microbiology vol8 pp 89ndash99 1986

[55] A Ventosa and A Oren ldquoHalobacterium salinarum nom cor-rig a name to replace Halobacterium salinarium (Elazari-Vol-cani) and to include Halobacterium halobium and Halobac-terium cutirubrumrdquo International Journal of Systematic Bacte-riology vol 46 no 1 p 347 1996

[56] M C Gutierrez M Kamekura M L Holmes M L Dyall-Smith and A Ventosa ldquoTaxonomic characterization of Halo-ferax sp (ldquoH alicanteirdquo) strain Aa 22 description of Haloferaxlucentensis sp novrdquo Extremophiles vol 6 no 6 pp 479ndash4832002

Archaea 17

[57] M F Mullakhanbhai and H Larsen ldquoHalobacterium volcaniispec nov a Dead Sea halobacterium with a moderate saltrequirementrdquo Archives of Microbiology vol 104 no 1 pp 207ndash214 1975

[58] S D Nuttall and M L Dyall-Smith ldquoCh2 a novel halophilicarchaeon from an Australian solar salternrdquo International Jour-nal of Systematic Bacteriology vol 43 no 4 pp 729ndash734 1993

[59] P D Franzman E Stackebrandt K Sanderson et al ldquoHalobac-terium lacusprofundi sp nov a halophilic bacterium isolatedfrom Deep Lake Antarcticardquo Systematic and Applied Microbi-ology vol 11 pp 20ndash27 1988

[60] G A Tomlinson and L I Hochstein ldquoHalobacterium saccharo-vorum sp nov a carbohydrate metabolizing extremely halo-philic bacteriumrdquoCanadian Journal ofMicrobiology vol 22 no4 pp 587ndash591 1976

[61] A Ventosa M C Gutierrez M Kamekura and M L Dyall-Smith ldquoProposal to transfer Halococcus turkmenicus Halobac-terium trapanicum JCM9743 and strainGSL-11 toHaloterrigenaturkmenica gen nov comb novrdquo International Journal ofSystematic Bacteriology vol 49 no 1 pp 131ndash136 1999

[62] M Kamekura and M L Dyall-Smith ldquoTaxonomy of the familyHalobacteriaceae and the description of two new genera Halo-rubrobacterium and Natrialbardquo Journal of General and AppliedMicrobiology vol 41 no 4 pp 333ndash350 1995

[63] T J Carver KM RutherfordM BerrimanM A RajandreamB G Barrell and J Parkhill ldquoACT the Artemis comparisontoolrdquo Bioinformatics vol 21 no 16 pp 3422ndash3423 2005

[64] S W Cline W L Lam R L Charlebois L C Schalkwyk andW F Doolittle ldquoTransformationmethods for halophilic archae-bacteriardquo Canadian Journal of Microbiology vol 35 no 1 pp148ndash152 1989

16 Archaea

[26] Z Zhang Y Liu S Wang et al et al ldquoTemperate membrane-containing halophilic archaeal virus SNJ1 has a circular dsDNAgenome identical to that of plasmid pHH205rdquoVirology vol 434no 2 pp 233ndash241 2012

[27] D H Bamford J J Ravantti G Ronnholm et al ldquoConstituentsof SH1 a novel lipid-containing virus infecting the halophiliceuryarchaeonHaloarcula hispanicardquo Journal of Virology vol 79no 14 pp 9097ndash9107 2005

[28] H M Kivela E Roine P Kukkaro S Laurinavicius P Somer-harju andDH Bamford ldquoQuantitative dissociation of archaealvirus SH1 reveals distinct capsid proteins and a lipid corerdquoVirology vol 356 no 1-2 pp 4ndash11 2006

[29] H Jaalinoja Electron Cryo-Microscopy Studies of Bacteriophage8 and Archaeal Virus SH1 University of Helsinki HelsinkiFinland 2007

[30] K Porter B E Russ J Yang andM L Dyall-Smith ldquoThe trans-cription programme of the protein-primed halovirus SH1rdquoMi-crobiology vol 154 no 11 pp 3599ndash3608 2008

[31] K Porter and M L Dyall-Smith ldquoTransfection of haloarchaeaby the DNAs of spindle and round haloviruses and the useof transposon mutagenesis to identify non-essential regionsrdquoMolecular Microbiology vol 70 no 5 pp 1236ndash1245 2008

[32] S E Luria General Virology John Wiley and Sons New YorkNY USA 1953

[33] C A Thomas K Saigo E McLeod and J Ito ldquoThe separationofDNA segments attached to proteinsrdquoAnalytical Biochemistryvol 93 pp 158ndash166 1979

[34] A L Delcher D Harmon S Kasif OWhite and S L SalzbergldquoImproved microbial gene identification with GLIMMERrdquoNucleic Acids Research vol 27 no 23 pp 4636ndash4641 1999

[35] R Zhang Y Yang P Fang et al ldquoDiversity of telomere palin-dromic sequences and replication genes among Streptomyceslinear plasmidsrdquo Applied and Environmental Microbiology vol72 no 9 pp 5728ndash5733 2006

[36] X Wang H S Lee F J Sugar F E Jenney M W W Adamsand J H Prestegard ldquoPF0610 a novel winged helix-turn-helix variant possessing a rubredoxin-like Zn ribbon motiffrom the hyperthermophilic archaeon Pyrococcus furiosusrdquoBiochemistry vol 46 no 3 pp 752ndash761 2007

[37] S W Cline L C Schalkwyk and W F Doolittle ldquoTransfor-mation of the archaebacterium Halobacterium volcanii withgenomic DNArdquo Journal of Bacteriology vol 171 no 9 pp 4987ndash4991 1989

[38] S Kamtekar A J Berman J Wang et al ldquoThe 12059329 DNA poly-merase protein-primer structure suggests a model for the ini-tiation to elongation transitionrdquo EMBO Journal vol 25 no 6pp 1335ndash1343 2006

[39] X Peng T Basta M Haring R A Garrett and D PrangishvilildquoGenome of theAcidianus bottle-shaped virus and insights intothe replication and packaging mechanismsrdquo Virology vol 364no 1 pp 237ndash243 2007

[40] C C Yang C H Huang C Y Li Y G Tsay S C Lee and CW Chen ldquoThe terminal proteins of linear Streptomyces chrom-osomes and plasmids a novel class of replication primingproteinsrdquo Molecular Microbiology vol 43 no 2 pp 297ndash3052002

[41] C C Yang Y H Chen H H Tsai C H Huang T W Huangand C W Chen ldquoIn vitro deoxynucleotidylation of the term-inal protein of Streptomyces linear chromosomesrdquo Applied andEnvironmental Microbiology vol 72 no 12 pp 7959ndash79612006

[42] P C Chang and S N Cohen ldquoBidirectional replication froman internal origin in a linear Streptomyces plasmidrdquo Science vol265 no 5174 pp 952ndash954 1994

[43] S W Wilhelm W H Jeffrey C A Suttle and D L MitchellldquoEstimation of biologically damaging UV levels in marinesurface waters with DNA and viral dosimetersrdquo Photochemistryand Photobiology vol 76 no 3 pp 268ndash273 2002

[44] T Allers and M Mevarech ldquoArchaeal geneticsmdashthe third wayrdquoNature Reviews Genetics vol 6 no 1 pp 58ndash73 2005

[45] M L Holmes S D Nuttall and M L Dyall-Smith ldquoConstruc-tion and use of halobacterial shuttle vectors and further studieson Haloferax DNA gyraserdquo Journal of Bacteriology vol 173 no12 pp 3807ndash3813 1991

[46] S D Nuttall and M L Dyall-Smith ldquoHF1 and HF2 novelbacteriophages of halophilic archaeardquo Virology vol 197 no 2pp 678ndash684 1993

[47] U Blaseio and F Pfeifer ldquoTransformation of Halobacteriumhalobium development of vectors and investigation of gas vesi-cle synthesisrdquo Proceedings of the National Academy of Sciences ofthe United States of America vol 87 no 17 pp 6772ndash6776 1990

[48] W L Lam and W F Doolittle ldquoShuttle vectors for the archae-bacterium Halobacterium volcaniirdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 86 no14 pp 5478ndash5482 1989

[49] C Yanisch-Perron J Vieira and J Messing ldquoImproved M13phage cloning vectors and host strains nucleotide sequences ofthe M13mp18 and pUC19 vectorsrdquo Gene vol 33 no 1 pp 103ndash119 1985

[50] U K Laemmli ldquoCleavage of structural proteins during theassembly of the head of bacteriophage T4rdquo Nature vol 227 no5259 pp 680ndash685 1970

[51] D G Burns H M Camakaris P H Janssen and M L Dyall-Smith ldquoCombined use of cultivation-dependent and culti-vation-independent methods indicates that members of mosthaloarchaeal groups in anAustralian crystallizer pond are culti-vablerdquo Applied and Environmental Microbiology vol 70 no 9pp 5258ndash5265 2004

[52] G Juez F Rodriguez-Valera A Ventosa and D J KushnerldquoHaloarcula hispanica spec nov and Haloferax gibbonsii specnov two new species of extremely halophilic archaebacteriardquoSystematic and Applied Microbiology vol 8 pp 75ndash79 1986

[53] A Oren M Ginzburg B Z Ginzburg L I Hochstein and BE Volcani ldquoHaloarcula marismortui (Volcani) sp nov nomrev an extremely halophilic bacterium from the Dead SeardquoInternational Journal of Systematic Bacteriology vol 40 no 2pp 209ndash210 1990

[54] M Torreblanca F Rodriguez-Valera G Juez A Ventosa MKamekura and M Kates ldquoClassification of non-alkaliphilichalobacteria based on numerical taxonomy and polar lipidcomposition and description of Haloarcula gen nov andHaloferax gen novrdquo Systematic and Applied Microbiology vol8 pp 89ndash99 1986

[55] A Ventosa and A Oren ldquoHalobacterium salinarum nom cor-rig a name to replace Halobacterium salinarium (Elazari-Vol-cani) and to include Halobacterium halobium and Halobac-terium cutirubrumrdquo International Journal of Systematic Bacte-riology vol 46 no 1 p 347 1996

[56] M C Gutierrez M Kamekura M L Holmes M L Dyall-Smith and A Ventosa ldquoTaxonomic characterization of Halo-ferax sp (ldquoH alicanteirdquo) strain Aa 22 description of Haloferaxlucentensis sp novrdquo Extremophiles vol 6 no 6 pp 479ndash4832002

Archaea 17

[57] M F Mullakhanbhai and H Larsen ldquoHalobacterium volcaniispec nov a Dead Sea halobacterium with a moderate saltrequirementrdquo Archives of Microbiology vol 104 no 1 pp 207ndash214 1975

[58] S D Nuttall and M L Dyall-Smith ldquoCh2 a novel halophilicarchaeon from an Australian solar salternrdquo International Jour-nal of Systematic Bacteriology vol 43 no 4 pp 729ndash734 1993

[59] P D Franzman E Stackebrandt K Sanderson et al ldquoHalobac-terium lacusprofundi sp nov a halophilic bacterium isolatedfrom Deep Lake Antarcticardquo Systematic and Applied Microbi-ology vol 11 pp 20ndash27 1988

[60] G A Tomlinson and L I Hochstein ldquoHalobacterium saccharo-vorum sp nov a carbohydrate metabolizing extremely halo-philic bacteriumrdquoCanadian Journal ofMicrobiology vol 22 no4 pp 587ndash591 1976

[61] A Ventosa M C Gutierrez M Kamekura and M L Dyall-Smith ldquoProposal to transfer Halococcus turkmenicus Halobac-terium trapanicum JCM9743 and strainGSL-11 toHaloterrigenaturkmenica gen nov comb novrdquo International Journal ofSystematic Bacteriology vol 49 no 1 pp 131ndash136 1999

[62] M Kamekura and M L Dyall-Smith ldquoTaxonomy of the familyHalobacteriaceae and the description of two new genera Halo-rubrobacterium and Natrialbardquo Journal of General and AppliedMicrobiology vol 41 no 4 pp 333ndash350 1995

[63] T J Carver KM RutherfordM BerrimanM A RajandreamB G Barrell and J Parkhill ldquoACT the Artemis comparisontoolrdquo Bioinformatics vol 21 no 16 pp 3422ndash3423 2005

[64] S W Cline W L Lam R L Charlebois L C Schalkwyk andW F Doolittle ldquoTransformationmethods for halophilic archae-bacteriardquo Canadian Journal of Microbiology vol 35 no 1 pp148ndash152 1989

Archaea 17

[57] M F Mullakhanbhai and H Larsen ldquoHalobacterium volcaniispec nov a Dead Sea halobacterium with a moderate saltrequirementrdquo Archives of Microbiology vol 104 no 1 pp 207ndash214 1975

[58] S D Nuttall and M L Dyall-Smith ldquoCh2 a novel halophilicarchaeon from an Australian solar salternrdquo International Jour-nal of Systematic Bacteriology vol 43 no 4 pp 729ndash734 1993

[59] P D Franzman E Stackebrandt K Sanderson et al ldquoHalobac-terium lacusprofundi sp nov a halophilic bacterium isolatedfrom Deep Lake Antarcticardquo Systematic and Applied Microbi-ology vol 11 pp 20ndash27 1988

[60] G A Tomlinson and L I Hochstein ldquoHalobacterium saccharo-vorum sp nov a carbohydrate metabolizing extremely halo-philic bacteriumrdquoCanadian Journal ofMicrobiology vol 22 no4 pp 587ndash591 1976

[61] A Ventosa M C Gutierrez M Kamekura and M L Dyall-Smith ldquoProposal to transfer Halococcus turkmenicus Halobac-terium trapanicum JCM9743 and strainGSL-11 toHaloterrigenaturkmenica gen nov comb novrdquo International Journal ofSystematic Bacteriology vol 49 no 1 pp 131ndash136 1999

[62] M Kamekura and M L Dyall-Smith ldquoTaxonomy of the familyHalobacteriaceae and the description of two new genera Halo-rubrobacterium and Natrialbardquo Journal of General and AppliedMicrobiology vol 41 no 4 pp 333ndash350 1995

[63] T J Carver KM RutherfordM BerrimanM A RajandreamB G Barrell and J Parkhill ldquoACT the Artemis comparisontoolrdquo Bioinformatics vol 21 no 16 pp 3422ndash3423 2005

[64] S W Cline W L Lam R L Charlebois L C Schalkwyk andW F Doolittle ldquoTransformationmethods for halophilic archae-bacteriardquo Canadian Journal of Microbiology vol 35 no 1 pp148ndash152 1989