dna sequences fiber genes of bacteriophage p2: evidencefor ... · p2 stocks were grown as described...

JOURNAL OF BACTERIOLOGY, Mar. 1992, p. 1462-1477 Vol. 174, No. 50021-9193/92/051462-16$02.00/0Copyright © 1992, American Society for Microbiology

DNA Sequences of the Tail Fiber Genes of Bacteriophage P2:Evidence for Horizontal Transfer of Tail Fiber Genes

among Unrelated BacteriophagesELISABETH HAGGARD-LJUNGQUIST,1* CONRAD HALLING,2 AND RICHARD CALENDAR3

Department of Microbial Genetics, Karolinska Institutet, Box 60400, S-104 01 Stockholm, Sweden'; Department ofMolecular Genetics and Cell Biology, University of Chicago, Chicago, Illinois 606372; and Department of

Molecular and Cell Biology, University of Califomia, Berkeley, California 947203

Received 3 October 1991/Accepted 28 December 1991

We have determined the DNA sequence of the bacteriophage P2 tail genes G and H, which code forpolypeptides of 175 and 669 residues, respectively. GeneH probably codes for the distal part of the P2 tail fiber,since the deduced sequence of its product contains regions similar to tail fiber proteins from phages Mu, P1,A, K3, and T2. The similarities of the carboxy-terminal portions of the P2, Mu, and P1 tail fiber proteins mayexplain the observation that these phages in general have the same host range. The P2H gene product is similarto the products ofboth A, open reading frame (ORF) 401 (sff, side tail fiber) and its downstream ORF, ORF 314.If 1 bp is inserted near the end ofORF 401, this reading frame becomes fused with ORF 314, creating an ORFthat may represent the complete sgf gene that encodes a 774-amino-acid-long side tail fiber protein. Thus, aframeshift mutation seems to be present in the common laboratory strain of A. Gene G of P2 probably codesfor a protein required for assembly of the tail fibers of the virion. The entire G gene product is very similar tothe products of genes U and U' of phage Mu; a region of these proteins is also found in the tail fiber assemblyproteins of phages Tula, TuIb, T4, and A. The similarities in the tail fiber genes of phages of different familiesprovide evidence that illegitimate recombination occurs at previously unappreciated levels and that phages aretaking advantage of the gene pool available to them to alter their host ranges under selective pressures.

The P2 virion has a straight tail 135 nm in length with acontractile sheath. (For a review of P2, see reference 10.)The structure of the P2 tail has been studied by high-resolution electron microscopy (55). The tail is joined to thefilled head at the neck region, where two disklike structuresare located. The upper disk is apparently located inside thehead, since it is seen only in virions in which the head isbroken open. The lower disk, or collar, is thinner than theupper disk and is located outside at the base of the head atthe top of the neck. The main part of the tail consists of aninner tube surrounded by a contractile sheath. At the distalend of the tail is located a poorly resolved base plate, towhich are attached a single tail spike 10 nm in length and sixtail fibers. The tail fibers have a length of 55 nm and arethicker at their distal ends.A large body of experimental results has shown that the

host range of tailed phages such as T2, T4, X, and Mu isdetermined by the tail fibers. During adsorption of the virionto the host, the tail fiber recognizes and binds reversibly to areceptor on the surface of the bacterial cell. This initialbinding is an absolute prerequisite to the subsequent steps ofirreversible binding and injection of phage DNA into the cell.It is assumed that the tail fibers of P2 behave similarly duringadsorption and infection.Antiserum prepared against virions has been shown for

T2, T4, X, and Mu to inactivate the virions mainly by bindingto the tail fibers, preventing their interaction with the cellreceptor (15, 29, 31, 34). In general, when two phages areserologically related, it is assumed that their tail fibers aresimilar in structure, since only similar or identical epitopes

* Corresponding author.

will be recognized by the same antibodies. Phages P2 and P1,which were isolated from the same bacterial strain, areserologically related (4, 9). Subsequent work has shown thatP2 and P1 are unrelated genetically, but the serologicalrelatedness of these two phages suggests that they may sharesimilar tail fiber genes (9, 98).The tail fibers of P2 may also be similar to the tail fibers of

phage Mu. Although Mu is unrelated to P1 (and to P2), thetail fiber genes of Mu are similar to those of P1 (19, 44, 50,91) and therefore possibly to those of P2. Further indirectevidence in support of the hypothesis that the tail fiber genesof P2 are similar to those of P1 and Mu is supplied by thefollowing observations: (i) the receptors for these threephages are located in the core structure of the lipopolysac-charide on the outer surface of the bacterial cell (9, 11, 51,59, 81), and (ii) bacterial strains selected for resistance to P2are usually resistant also to P1 and Mu but not to a widevariety of other phages (6, 34).Where in the P2 genome might the tail fiber genes be

located? To date, 12 tail genes and six tail proteins of P2have been identified (55, 57, 58, 88, 90). Tail genes R and Sare located in the distal portion of the ONMLKRS transcrip-tion unit (Fig. 1), which also contains head genes (0, N, M,and L) and the lysis gene (K). The other tail genes are locatedin the VWJIHG and FIFIIETUD transcription units (Fig. 1).The functions of only a few of the tail genes are known, sinceamber mutants in most tail genes produce virtually norecognizable tail structures (55). P2 R and P2 S mutantsproduce apparently complete tails that nevertheless cannotbe connected to full heads; these mutants also produce tailswith abnormally long tube structures (P2 R mutants) orsheath structures (P2 S mutants). Hence, R and S appear tocontrol tail length (55). On the basis of the abundance and

1462

on Novem

ber 1, 2020 by guesthttp://jb.asm

.org/D

ownloaded from

http://jb.asm.org/

HORIZONTAL TRANSFER OF TAIL FIBER GENES AMONG PHAGES

4-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~f.

PRGQ P O N M LK RSV WJ I H GZfun FlFl ETUD-intC° B A old

HgiAl AatulSinai Bgl Stul

I Y I

\-11, 100

Bgil Sspl Dral Sspl Hpal Avail SsplHaell Dral Asel Sspl / Dral BamHI Haell Dral

Accl Asulli SspI pralp I Mstl Sspl ApaLI Avail DraispiII I IYI IIII I IIII

37 38 39 40 41 42 43 44 45 46 47

aene H geneG

-4 *- - - -*4--*-- -k

4- 4- 4-- 4-

4- 4-- 4--p 4 4

FIG. 1. Strategy for sequencing the H and G genes of bacteriophage P2. The genetic map of bacteriophage P2 (37) is shown at the top. Thearrows indicate the direction and extent of the four late transcription units. Below the genetic map is a physical map of the P2 genome atpositions 36 to 47%. The arrows at the bottom indicate the direction and extent of each sequencing reaction for each strand of the DNA. Thearrows for the top strand (the sequence of which is presented in Fig. 2) are located above the solid line, and the arrows for the bottom strandare located below the solid line. Arrows oriented in the 5'-to-3' direction indicate fragments labelled at the 5' end and sequenced by themethod of Maxam and Gilbert (62) or fragments sequenced by the method of Sanger et al. (83) (Materials and Methods). Arrows oriented inthe 3'-to-5' direction indicate fragments labelled at the 3' end and sequenced by the method of Maxam and Gilbert.

sizes of their gene products in P2 phage tails, FI is believedto code for the sheath protein and FlI is believed to code forthe tube protein (55, 90). Hybrids of P2 and the P2-like phage186 that were selected for P2 host range and 186 immunitylack the P2 E, T, U, and D genes, demonstrating that thesegenes do not play a role in host specificity (43). Eliminationof these genes leads to the deduction that the tail fiber genesare located in the VWJIHG transcription unit. This suppo-sition is bolstered by the fact that 32P-labelled DNA of P1,which, as noted above, may have tail fibers similar to thoseof P2, hybridizes to the P2 BalI (29.9%)-to-EcoRI (47.5%)restriction fragment containing the VWJIHG transcriptionunit (36) (Fig. 1). Genes V, J, H, and G are essential genes

required for tail synthesis (55). Genes Wand I are two newlyidentified genes in the tail gene cluster discovered by DNAsequence analysis; W and I are essential genes, but the roleof these genes in tail synthesis has not yet been investigated(36). Of these genes, only the product of H has beenidentified; it is present at 14 to 18 copies in the virion, but itsfunction and location in the tail are unknown (55).

In our effort to complete the DNA sequence of the P2genome, we have begun the sequencing of the P2 VWJIHGtranscription unit. Here we report the DNA sequence of P2genes H and G. We have found that portions of these genes

are highly similar to portions of the tail fiber genes of Mu, P1,and other phages. On the basis of these similarities, we

propose that gene H encodes the tail fiber protein of P2 andthat gene G plays a role in the assembly of tail fibers. In our

computer analysis of the tail fiber proteins of a wide varietyof phages that infect Escherichia coli, we found considerableevidence for the hypothesis that horizontal transfer of tailfiber genes has occurred repeatedly among otherwise unre-

lated phages.

MATERIALS AND METHODS

Bacteria, phages, plasmids, and synthetic oligonucleotides.Organisms and plasmids used are listed in Table 1. The M13forward and reverse primers were obtained from Promega(Madison, Wis.). P2-specific oligonucleotides, which were

obtained from the Department of Biology at the Universityof Oslo, Scandinavian Gene Synthesis AB (Koping, Swe-den), or KABIGENE (Stockholm, Sweden), are listed inTable 2.

Preparation of phage stocks and marker rescue analysis. P2stocks were grown as described previously (8). Rescue ofgenetic markers from cloned P2 DNA fragments was per-formed as described elsewhere (60).Recombinant DNA techniques. P2 DNA was isolated from

phage as described in reference 96. For amplification of P2DNA, DNA from about 107 to 108 PFU of each P2 strain wassubjected to amplification with the GeneAmp kit from Per-kin-Elmer (Norwalk, Conn.). The reaction mixture, includ-ing 20 pmol of the respective primer, was subjected to 25cycles of 1 min at 94°C, 2 min at 42°C, and 3 min at 72'C,with a Perkin-Elmer Cetus thermocycler model PCR1000.When enzymatically amplified DNA was used for cloning,several isolates were kept for further analysis.

Plasmid constructions were performed according to stan-dard procedures (61) and are summarized in Table 1. E. coliC-la was used as the recipient for all transformations.DNA sequence analysis. For DNA sequenced by the

method of Maxam and Gilbert (62), purified phage DNA was

cleaved with the appropriate restriction enzymes. 3' terminiwere labeled with [t-3 P]ddATP (Amersham, Buckingham-shire, England) and terminal deoxynucleotidyl transferase(Amersham); 5' termini were labeled with [y-32P]ATP (NewEngland Nuclear, Boston, Mass.) and T4 polynucleotide

0

.4-4-

VOL. 174, 1992 1463

on Novem


.org/D

ownloaded from

http://jb.asm.org/

1464 HAGGARD-LJUNGQUIST ET AL.

TABLE 1. Bacteria, phages, and plasmids

Strain, phage, or plasmid Characteristic(s) Reference

E. coliC-la Prototrophic 84C-1757 rpsL (Smr) supD 88

PhagesP2 lg Wild type 7P2 del2 Ig A(Z-fun) 5P2 Gaml virl Deficient in tail synthesis 57P2 Ham59 virl Deficient in tail synthesis 88P2 Ham72 vir2O Deficient in tail synthesis 88

PlasmidspEE484 AsuII (41.4%)-to-BamHI (45.3%) fragment of P2 Ig ligated to pUC18 AccI-BamHI This workpEE485 pEE484 digested with HpaI and SmaI and religated; carries AsuII (41.4%)-to-HpaI This work

(43.0%) fragment of P2 lgpEE489 pEE484 digested with AvaIl and HindII, treated with Klenow fragment, and ligated; This work

carries AvaIl (44.0%)-to-BamHI (45.3%) fragment of P2 IgpEE601 Sall (36.6%)-to-BamHI (45.3%) fragment of P2 Ig ligated to pUC18 Sall-BamHI This workpEE616 pEE601 digested with AsuII and BaimHI, treated with Kienow fragment, and ligated; This work

carries Sall (36.6%)-to-Asull (41.4%) fragment of P2 IgpUC18 Apr lacZ'; vector for nucleotide sequence analysis 69

kinase (Promega). After cleavage with a second restrictionenzyme, the fragments were separated by gel electrophore-sis and isolated and chemical sequencing was performed.For DNA sequenced by the method of Sanger et al. (83), theDNA was subcloned into pUC18 (69), and 5'-[a-35S]thio-dATP and T7 DNA polymerase (Pharmacia, Uppsala, Swe-den) were used for the sequencing reaction, using theforward or reverse primers of M13 (Promega) or P2-specificprimers (Table 2). The products of the sequencing reactionswere separated on urea-acrylamide sequencing gels. Thegels were subsequently dried and autoradiographed.Computer analysis of DNA and protein sequences. DNA

and protein sequences were compared with sequences in theGenBank and EMBL DNA sequence data bases and theGenPept and SWISS-PROT protein sequence data bases byusing the FASTA program (71) using the PAM-250 matrix(24) through the electronic mail server at GenBank (12).Further analysis of the sequences was performed by usingFASTA and LFASTA (71) on Apple Macintosh computers.DNA and protein sequences and the results of the analyseswere organized by using HyperSeq, a HyperCard 2.0 stackfor Apple Macintosh computers (38).

Nucleotide sequence accession numbers. GenBank and

TABLE 2. P2-specific oligonucleotides

Oligonucleotide Sequence Locationa

37.8R GGTGCAGCAAAGCTGGCA 98-11538.0L ATCCCCGACGGCCATCGT 172-15538.9R TCAGGACGTTGGCAGACCT 416-43438.9L GACGATGATGACCATGCGA 454-43640.3R GCCCTTGTAGACTCGTCGCC 926-94540.7L GGGTAGCATCTTTCGGTTGC 1055-103641.5R GACTGGGCGAAGATTGGATT 1313-133241.6L TCAGTGTCACCATCGGCATC 1359-134043.6R CATCAATGGCACATGGTATA 2035-205443.8L GGATTACCTGACTTAATGT 2109-209144.5R ACCTCCAACCGACGTGCT 2321-233844.6L CATCCATTTACCTGAGTC 2359-234245.5L ACTCTGTGCCATAGATATGTA 2647-2627

a Location in DNA sequence of Fig. 2.

EMBL accession numbers and sources of DNA sequencesare presented in Table 3.

RESULTS AND DISCUSSION

Sequencing strategy.H and G are essential genes located inthe distal portion of the P2 VWJIHG transcription unit(56-58, 88). This transcription unit is separated from theF,F,JETUD transcription unit, which also contains tail

TABLE 3. Nucleotide sequences

Gene(s) Accesson no. Reference

E. coli CRF 86a M24530 1e14 (-) F Pn, ORF 293 K03521 72e14 (+) ORF PPus, ORF 183b K03521 72K3 37 X04747 74K3 38 X05560 75X tfa, stf J02459 23Ml 37, 38 X05676 66MuG(+)S, U None 46Mu G()S', Ud None 46Ox2 37, 38 X05675 66P1 19c M25470 35P1 cixR (19 or 19') K03173 42P1 cixL (19 or 19') X01828 42P1 19ve None 50P2 H, G M64677 This workT237 X04442 74T2 38 X05312 75T3 tf M14784 97T4 37 J02509 70T4 38 X05677 66T7 17 J02518 28Tula 37, 38 X55190 64TuIb 37, 38 X55191 64

a CRF 86 is located at positions 124 to 384; position 309 was changed fromA to G to alter the ochre (TGA) codon to a Trp codon (TGG).

b From inversion of the P region. ORF 183 begins at the TTG start codon atposition 397.

G inserted between positions 20833 and 20834 to join ORF 401 with ORF314, creating stf.

d From inversion of the G region.e Protein sequence only.

J. BACTERIOL.

on Novem


.org/D

ownloaded from

http://jb.asm.org/

HORIZONTAL TRANSFER OF TAIL FIBER GENES AMONG PHAGES 1465

genes, by a nonessential region containing genes Z and fun(Fig. 1). The transcriptional start points of the Vand F1 geneshave been located at 30.5 and 52.6% from the left end of the33-kb P2 genome, respectively (20). In P2 del2, a deletionextends from 45.5 to 51.2% and removes the region contain-ing Z and fun (36, 96). We sequenced genes H and G usingthe strategy shown in Fig. 1. The DNA sequence of thesegenes is presented in Fig. 2.

Identification of genes H and G. The DNA sequencecontains two significant open reading frames (ORFs). Tocorrelate the ORFs with genes H and G, we cloned restric-tion fragments of P2 DNA in plasmid pUC18 and carried outmarker rescue experiments in strains containing the recom-binant plasmids. As shown in Fig. 3, the region from 41.4 to45.3% rescues P2 Gam1, P2 Ham72, and P2 HamS9. Theright ORF is gene G, since cloned wild-type DNA from thisORF (pEE489) rescues P2 Gaml. The left ORF is gene H,since a cloned fragment of the wild-type ORF (pEE485)rescues P2 Ham72 and P2 HamS9.We confirmed the marker rescue results by sequencing

DNA fragments cloned from P2 Gaml, P2 Ham72, and P2HamS9 to identify the positions of the mutations. Themarker rescue data indicated that Gaml is located betweenpositions 44.0 and 45.3% (Fig. 3). Using primers 43.6R and45.5L (Table 2), we amplified a region of P2 Gaml virl from43.6 to 45.5% using the polymerase chain reaction andcloned it into pUC18. The entire DNA fragment was se-quenced, and the Gaml mutation was found to be a C-to-Tmutation at position 2207 that generated an amber (TAG)codon (Fig. 2). The first in-frame ATG codon of G is locatedat position 2075, and as it is preceded by an apparently goodribosome binding site centered 9 bp upstream (87), this is thelikely start codon. G ends at position 2602 with a TAA stopcodon and codes for a polypeptide of 175 amino acidresidues with a predicted Mr of 20,264.The HamS9 and Ham72 mutations were mapped between

positions 41.4 and 43.0% (Fig. 3). We amplified DNA frompositions 40.3 to 43.8% of P2 Ham59 and P2 Ham72 usingprimers 40.3R and 43.8L and cloned and sequenced thesefragments. Ham59 is a C-to-T mutation at position 1664 thatgives an amber codon; Ham72 is an A-to-T mutation atposition 1322 that also gives an amber codon (Fig. 2). Thefirst in-frame ATG codon is located at position 62, and sinceit is preceded by a potential ribosome binding site centered 6bp upstream, it is the likely start codon. H ends at position2071 with a TAA stop codon; there is a 3-bp gap between thestop codon of H and the start codon of G. H potentiallycodes for a protein of 669 residues with a predicted Mr of71,366, which is in accord with the observed relative molec-ular weight of 71,000 for the H gene product during sodiumdodecyl sulfate-polyacrylamide gel electrophoresis (55).

Potential transcriptional termination signal for the VWJIHGtranscription unit. Since G is the last gene of the VWJIHGtranscription unit, it is expected to be followed by a tran-scriptional termination signal. There is a region of hyphen-ated dyad symmetry downstream of G at positions 2968 to2994 that is followed by a T-rich region (Fig. 2). Thesefeatures are characteristic of a Rho-independent transcrip-tion terminator (78). The region between G and this putativeterminator contains stop codons in all reading frames andthus does not seem to contain any coding capacity.

Left end of the del2 deletion. The first base affected by thedel2 deletion, which removes the nonessential region be-tween the VWJIHG and FJF,,ETUD transcription units, waslocated by DNA sequencing to position 2622, only 20 bp tothe right of the stop codon of gene G (Fig. 2). The putative

Rho-independent terminator downstream of G would beremoved by the del2 deletion.

Analysis of protein sequences. The deduced amino acidsequences of the P2 H and G gene products (gpH and gpG)were compared with sequences in the GenPept and SWISS-PROT protein sequence data bases. P2 gpH contains regionshighly similar to regions of the tail fiber structural genes ofphages K3, T2, X, Mu, and P1. P2 gpG contains regionshighly similar to regions of the tail fiber assembly genes ofphages Mu and P1; less similarity was seen with the tail fiberassembly proteins of phages X, T4, TuIa, and TuIb. Weextended the analysis of tail fiber structural and assemblyproteins and found additional regions of similarity amongphages other than P2. Detailed alignments of some of thesimilar regions are presented in Fig. 4. The boundaries of theregions of similarity and the percentage of identical residuesin these regions are given in Tables 4 and 5. A summary ofthe similarities of tail fiber proteins is presented in Fig. 5. Itshould be noted that in this representation only identicalamino acids are shown, and as there are several cases ofvery conservative substitutions (I-L-V; D-E; and R-K), thehomologies are even higher than the figure indicates. Wedescribe each comparison in detail below.P2 gpH, Mu gpS and gpS', and Pl gpl9 and gpl9'. P2 gpH

is most similar to the Mu gpS and P1 gpl9 tail fiber structuralproteins. The tail fiber genes of phage Mu are located on theinvertible G segment, which acts as a genetic switch thatdetermines the host range of the phage (53) (Fig. 4). Whenthe G segment is in the plus orientation, genes S and U areexpressed and Mu phage can infect E. coli K-12 (16, 49, 89).When the segment is in the minus orientation, genes S' andU' are expressed and the phage cannot infect E. coli K-12(16, 49, 89) but can infect Citrobacter freundii, Shigellasonnei, E. coli C, Enterobacter cloacae, Serratia marces-cens, and Erwinia spp. (30, 47, 80, 92). S and U are requiredfor the production of tail fibers since in a Su- host, Mu G (+)amber mutants in S or U but in no other late genes producevirions lacking tail fibers (34). Mapping of S, U, S', and U'and DNA sequence analysis of the right end of Mu haverevealed that S and 5' have a 531-bp segment in common,termed Sc; hence, the first 177 amino acid residues of gpSand gpS' are identical (33, 46).The tail fiber genes of phage P1 are organized similarly to

those of Mu (Fig. 4) (48). P1 contains a 4.2-kb invertibleregion, the C segment, which also acts as a genetic switch toencode two host range specificities (44, 45, 91). Although P1and Mu are otherwise unrelated, the central 3-kb region ofthe P1 C segment is homologous to the entire Mu G segment(19). The homologous region of the C segment is flanked by0.62-kb inverted repeats (19, 42, 45), and the recombinationsites for inversion are located on the outer ends of theinverted repeats (45). When the C segment is in the plusorientation, genes 19 and U are expressed and P1 can infectE. coli K-12 and E. coli C (44). When the C segment is in theminus orientation, genes 19' and U' are expressed and P1 isunable to infect E. coli K-12 or wild-type E. coli C but is ableto infect a mutant strain of E. coli C (44). Host range mutantsof P1 have been isolated by point mutation of either set of P1tail fiber genes (44, 91). Gene 19 has been shown throughserological studies and an analysis of P1 mutants by electronmicroscopy to be required for the production of tail fibers(95). Since genes 19 and 19' share a long constant region,19c, the amino-terminal residues of the respective proteinsare identical (35, 50).The DNA sequence of the tail fiber genes of phage Mu has

been published (46), permitting the analysis of the deduced

VOL. 174, 1992

on Novem


.org/D

ownloaded from

http://jb.asm.org/

gene I gene HA G G V AY DOG D V I T VY P G * M S I K F R T V I T T AG

1 GCCGGTGGTGTGGCTTACGACGGCGATGTAATTACGGTTTACCCCGGATAAGTGAATAATGAGCATAAATTCAGMCCGTTATCACCACTGCCGGTRBS

A A K L A A A T A P G R R K V G I T T M A V G D G G G K L P V P D A

BglI 37.9

G Q T G L I H E V W R H A L N X I S Q D X R N S N Y I I A E L V I201 CCGGACAGACCGGGCTTATCCATGAAGTCTGGCGACATGCGCTGAACAAAATCAGCCAGGACAAACGAAACAGTAATTATATTATCGCCGAGCTGGTTAT

P P E V G G F W M R E L G L Y D D A G T L I A V A M M A E S Y K P301 TCCGCCGGAGGTGGGCGGTTTCTGGATGCGTGAGCTTGGCCTGTACGATGATGCGGGAACGTTAATTGCCGTGGCGAACATGGCCGAAAGCTATAAGCCA

A L A E G S G R W Q T C R M V I I V S S V A S V E L T I D T T T V M401 GCCCTTGCCGAAGGCTCAGGACGTTGGCAGACCTGTCGCATGGTCATCATCGTCAGCAGTGTGGCCTCAGTGGAGCTGACCATTGACACCACAMCGGTGA

A T Q D Y V D D K I A E H E Q S R R H P D A S L T A K G F T Q L S501 TGGCGACGCAGGATTACGTTGATGACAAAATTGCAGGCAGTCACZACCCCGGACGCCTCGCTGACAGCAAAAGGTTTTACTCAGTTAAG

HgIAI 3 9. 2 AatII 39.2

S A T N S T S E T L A A T P K A V X A A Y D L A N G K Y T A Q D A60 1 CAGTGCGACCAACAGCACGTCTGAAACACTGGCCGCAACGCCGAAAGCGGTAAAGGCCGCGTATGACCTGGCTAACGGGAAATATACCGCACAGGACGCC

T T A R 1K G L V Q L S S A T N S T S E T L A A T P X A V K T V M D E701 ACCACAGCGCGAAAaTGTCCAGCTTAGTAGCGCCACCAACAGCACGTCTGAAACGCTCGCCGCAACACCAAAAhGCCGTTAAGACGGTAATGGATG

StuI 39.7

T N K X A P L N S P A L T G T P T T P T A R Q G T N N T Q I A N T801 AAACGAACAAAAAAGCGCCATTAAACAGCCCTGCACTGACCGGAACGCCAACGACGCCAACTCCGCGACAGGGAACGAATAATACTCAGATCCCAAACAC

A F V M A A I A A L V D S S P D A L N T L N E L A A A L G N D P N901 GGCTTTCGTTATGGCCGCGATTGCCGCCCTTGTaACTCGTCGCCTGACGCACTGAATACGCTGAACGAGCTGGCGGCGGCGCTGGGCAATGACCCGAAT

Accl 40.3

F A T T M T N A L A G K Q P X D A T L T A L A G L A T A A D R F P Y1 001 TTTGCTACCACCATGACTAATGCGCTTGCGGGTAAGCAACCGAAAGATGCTACCCTGAC;:C=GGCGGGGCTTGCTACTGCGGCAGACAGGTTTCCGT

.HaellI 40. 7

F T G N D V A S L A T L T K V G R D I L A K S T V A A V I E Y L G110 1 ATTTTACGGGGAATGATGTTrC-GCCT=GGACCCTGACAAAAGTCGGGCGGGATATTCTGGCTAAATCGACCGTTGCCGCCGTTATCGAATATCTCGG

Bgrll 40.9

L Q E T V N R A G N A V Q X N G D T L S, G G L T F E N D S I L A W120 1 TTTACAGGAAACGGTAAACCGAGCCGGGAACGCCGTGCAAAhATGGCGATACCTTGTCCGCTCCACTTACTTTTGAAAACGACTCAATCCTTGCCTGG

I R N T D W A K I G F K N D A D G D T D S Y M W F E T G D N G N E Y1301 ArXTACTGACTGGGCGAAGATTGGA TGATGCCGATGGTGACACTGATTCATACATGTGGTTTGAAACGGGGGATAACGGCAATGaa£

AsuIlI 41 .4 T, am 72 Dral 41.5S SspI 41. 7

F K W R S R Q S T T T IC D L M T L K W D A L N I L V N A V I N G C1401 T=CAAATGGAGAAGCCGCCAGAGTACCACAACAAAAGACCTGATGACGTTGAAATGGGATGCACTAU=CTTGTTAATGCCTCCTckT=xGGCTG

SspI 41.9 Asel 42.0

F G V G T T N A L G G S S I V L G D N D T G F KC Q N G D G I L D V1 501 TTTTGGAGTTGGTACGACGAATGu;TAGSmGTAGCTCTATTGTTCTTGGTGATAATGATACCGGAZTACAGAATGGAGACGGTATTCTTGATGTT

Dralll 42.1 Dral 42.2

Y A N S Q R V F R F Q N G V A I A F K N I Q A G D S KXK F S L S S S1601 TATGCTAACAGTCAGCGTGTATTCCGTTTTCAGAATGGAGTGGCTATTGC TTTaAaA xT^fACAGGCAGG;TGATAGTAAAkAGTTCTCGCTATCCAGCT

Dral 42.5S T, amS9Sspl 4 2.5S

N T S T K N I T F N L W G A S T R P V V A E L G D E A G W H F Y S1701 CTAATACATCCACGAGAGTT2CCTTTAATTTATGGGGTGCTTCCACCCGTCCAG;TGGTTGCAGAGTTAGGCGATGAGGCCGGATGGCATTTCTATAG

Sspl 4 2. 7

Q R N T D N S V I F- A V N G Q M 0 P S N W G N F D S R Y V X D V R1 801 CCAGCGAAATACAGATAMCTCGGTA&TGCTLZ;&GGTCAGATGCAACCCAGCAACTGGGGAAATTTTGATTCCCGCTATGTGAAAGATGTTCGC

Sspl 43. 0 Rpal 43.0°

L G T R V V Q L M A R G G R Y E X A G H T I T G L R I I G E V D G D1 901 CTGGGTACGCGAGTTGTTCAATTGATGGCGCGAGGTGGTCGTTATGAAAAAGCCGGACACACGATTACCGGATTAAGAATCATTGGTGAhGTAGATGGCG

gene GD E A I F R P I Q K Y I N G T W Y N V A Q. V * M Q H L K N I K S

2 001 ATGhTGAAGCCATCTTCAGGCCGATACAAAAAxTACATCAATGGCACATGGTATAAC Tr.eoGr-CkGSTAAGTTATGCAGCA=AbaGAACATTAAGTCtEtI 43 .7 RBS DraI

G N P K T K E Q Y Q L T 1K N F D V I W L W S E, D G K N W Y E E V X21 01 AGGTAATCCAAXUCAAAAGAGCAATATCAGCTAACAAAGAATTTTGATGTTATCTGGTTAT:iGAAGACGCAAAAsCTGGTATGAGGAAGTGAAG

Avall 4 4.0

N F Q P D T I K I V Y D E N N I I V A I T R D A S T L N P E G F S V2 201 AACTTTCAGCCAGACACAATAAAGATTGTTTACGATGAAAAT^AT>TTATTGTCGCTATCACCAGAGATGCTTCAACGCTTAATCCTGAAGGTTTTAGCG

T, aml SspI 4 4.3

V E V P D I T S N R R A D D S G X W M F 1K D G A V V Kt R I Y T A D2301 TTGTTGAGGTTCCTGATATTACCTCCAACCGACGTGCTGACGACTCAGGTAAATGGATGTTTAAGGATGGTGCTGTGGTTAAACGGATTTATACGGCAGA

E Q Q Q Q A E S Q JC A A L L S E A E N V I Q P L E R A V R L N M A2 401 TGAACAGCAACAACAGGCAGAATCACAAAAG.GCCGCGTTACTTTCCGAAGCGGAAAACGTTATTCACCCACTGGAACGCGCTGTCAGGCTGAATATC.GCG

T D E E R A R -L E S W E R Y S V L V S R V D P A N P E W P E M P Q25S01 ACGGATGAGGAAC;TZGACTGGAGTCATGGGAhCGTTACAGCGTTCTGGTCAGCCG;TGTeGaTGCAAATCCTGAATGGCCGGAhATGCCGCAAT

ApaLI 4 S . 1 BamnHI 45S.3

2 601 AAGTTGTATGACCTCTGTTGTGA&TACATATCTATGGCACAGAGTAAAGCCTAATCTGAAG;TCS:TCTECCAAAAG.C:aTTATAAACAAGAAI ---> del2 Draill 45.6 AvaIl 45. 6

2701 CAATGGTCATATCAG;TGTGTTTTGATTGCATAGCTAACGTGCGTCTTCCTGTACAGAATCATAAGATGATAGGGCATAGGAGAT GATTA1TTATCGCGTGT

2 801 ITAAWTAGCTTTCTGCATCAAGTGTCACTTGCGAAGAGGTATTGGCGTATGATTGGCATCACTCAGTTCAGATA.ATATTTo AMTATTTGGTAAATDral 4 6.0 NaelI 4 6.1 SspI Sspl 46. 2

Dral

2 901 GTTCAGATTACACAATTCGGTCTGCGTCCCGTTAAAATTATCTACTATTAGTTATTACTATGAGGTGAATGGCAAGTGTTTTCCACTCGCCC^TTTTTGTT_____ __ > < --- ------

3001 TTTTATTTTTTT

1466

on Novem


.org/D

ownloaded from

http://jb.asm.org/


36 37 38 39 40 41 42 43 44 45

I gem H I I IG InEE601

pEE616

pEE489

I pEE485 I

Marker rescue

Ham72 Ham59 Gaml

+ + +

+ + +

_ _ +

+ + -

FIG. 3. Marker rescue mapping of amber mutations in genes G and H. At the left is a physical map of the region of the P2 genomecontaining the H and G genes. The DNA fragments of this region contained in plasmids used for marker rescue experiments are indicatedunderneath. The ability of each plasmid to rescue each mutation is shown in the table to the right. +, rescue observed; -, rescue notobserved.

amino acid sequences of the S, U, S', and U' gene products.In the case of P1, the DNA sequence of the tail fiber geneshas been determined, but only fragments have been pub-lished (35, 42, 50). Most of the deduced amino acid sequenceof gpl9 is available, but the sequences of genes U, 19', andU' are unpublished, although it has been reported that thesegenes are highly similar to their Mu counterparts (50). In anearlier analysis, Mu gpS and P1 gpl9 were shown to containa large region of similarity (regions C, G, H, and P in Fig. 4)which corresponds to the entire variable portion of Mu S(50).Three regions of P2 gpH, regions A, B, and C, are highly

similar to regions of Mu gpS and gpS' and P1 gpl9 and gpl9',with similarities ranging from 65 to 92% identity (Fig. 4 and5; Table 4). Region A is found in two nearly tandem copiesin gpH; the second copy is separated by four residues fromthe first (Fig. 5A). Region A is also found in the constantregions of Mu gpS and gpS' and P1 gpl9 and gpl9' (Fig. SB).Region B is contiguous to region A in gpH. At least a portionof region B is found in the constant part of gpl9 and gpl9';the gap in the available sequence may contain the rest ofregion B. Region B is absent from gpS and gpS'. Regions Aand B are also found in the product of the side tail fiber gene,stf, of phage X; this protein is discussed below. Region C iscontiguous to region B in gpH. Region C is also present ingpl9, but it has been separated from region B by residuesresulting from the translation of one of the inverted repeatsof the P1 C segment, which are not present in the P2 or Mugenomes. In gpS, a short gap separates region A from regionC. Region C is not present in gpS' nor, presumably, in gpl9'.A fourth region, P, is found at the carboxy termini of gpH,

gpS, gpS', and gpl9 (Fig. 5A). This region of gpH and gpl9is 80% identical over 76 residues; similarity is less with gpS(49% identity) and gpS' (31% identity). There is conclusiveevidence that this region determines the host range of phagesMu and P1 (50). Recombination between Mu and P1 canyield hybrid Mu phage with the host range of P1 (91); DNAsequencing has shown that the crossover occurs in region P(50). In addition, DNA sequencing of host range mutants ofMu showed that the mutations occur in region P (50). Since

the carboxy-terminal portions of the Mu and P1 tail fiberproteins determine host range and since they are similar tothe carboxy-terminal portion of P2 gpH, we conclude that P2H encodes at least the terminal portion of the tail fiber of P2virion and that H is the gene that determines the P2 hostrange. The similarity of the carboxy termini of the tail fibersof Mu, P1, and P2 provides a simple explanation for theserological cross-reactivity of these phages and for theirsimilar host ranges.P2 gpH and X Stf. The DNA sequence of phage X reveals

four ORFs, lom (lambda outer membrane protein), ORF 401,ORF 314, and ORF 194, in the nonessential b region to theright of the late head and tail genes (23, 82). ORF 401, ORF314, and ORF 194 could be cotranscribed with the late genesfrom the PR' promoter (22, 41). Two of these genes showsimilarities to the tail fiber genes of T4 (32, 63): ORF 314 issimilar to the 3' portion of T4 gene 37, which encodes thedistal portion of the T4 tail fiber (2, 3, 70), and ORF 194 issimilar to T4 gene 38, which encodes a protein involved inthe assembly of the tail fibers (66, 85). Occasionally, side tailfibers are observed in electron micrographs of A virions (52);because of the similarity of ORF 314 to T4 37, it has beensuggested that ORF 314 protein is incorporated into theseside tail fibers (63).

In a recent remarkable paper, Montag et al. demonstratedconclusively that ORF 314 and ORF 194 encode tail fiberproteins (67). They made the following observations. (i)Virions produced by wild-type X have side tail fibers, but noside tail fibers are made by AplacS, in which lac DNA issubstituted for the region including ORF 401, ORF 314, andORF 194. Curiously, the production of side tail fibers isdependent on the bacterial host. (ii) Synthesis of the productof ORF 314 from an expression vector was not detectableuntil ORF 314 was fused to the 5' portion of lacZ, suggestingthat ORF 314 lacks good translational signals. (The DNAsequence of ORF 314 reveals a poor or nonexistent ribosomebinding site [23, 82].) (iii) When the 3' portion of T4 37 isreplaced by most of ORF 314 in a gene fusion and T4 38 isreplaced by ORF 194, the resultant T4 hybrid phage acquiresa host range different from those of both wild-type T4 and X.

FIG. 2. DNA sequence of the H and G tail genes of bacteriophage P2. The DNA sequence is shown in the same orientation as the mapsin Fig. 1. The deduced amino acid sequences of the H and G gene products are shown above the DNA sequence. Restriction sites andpotential ribosome binding sites (RBS) are underlined. The sequence of each mutation is given below the DNA sequence. The dashed arrowsnear the end of the DNA sequence indicate the hyphenated dyad of a potential Rho-independent transcription terminator.

VOL. 174, 1992

on Novem


.org/D

ownloaded from

http://jb.asm.org/


|- ~~ a

17 J: K

17

aO, R '| U |37 38

aO R ,| U |

8 G : J M N I

37

37GL ,8 N

37

38

@ R" U338

37 38

Tula

Tulb

T4 I 3D' E1 F I

37

e14 P(+)

e14 P(-)

E. cofi

li1LIP2

Mu G(+)

Mu G(-)

L 9IJ3L M S T

37 38

M- S, T VIW

37 38

G L 11V S T lVI38

SiTlPLZnt

ORF P ORF 183plus

s s pi

},s I wl I IORF Pm ORF 293 pin

CRF 86 appY

stf tfa

H G

LLsU Li

Sc St U' gin

C segmert

5_ G_ __U:ZII _UI_l9c l9v ~~~U

------------- n---- -i- ------------------

I I

19c 19V cmn

T3

17

Ox2

Ml

K3

T2

I 1

P1 C(+)

P1 C(-)

dn

I I

D I I FI I

J. BACTERIOL.

.w

6

on Novem


.org/D

ownloaded from

http://jb.asm.org/


(iv) Antibodies raised against the lacZ-ORF 314 fusionprotein inactivate T4 and T4-related Tula and TuIb phageparticles, and the antibodies can be observed bound to thetail fibers of these phages and of the T4-A hybrid phage.Surprisingly, however, the antibodies do not bind the sidetail fibers of X. On the basis of these and previous results,Montag et al. proposed that ORF 314 be named stf (side tailfiber) and that ORF 194 be named tfa (tail fiber assembly)(65, 67).

In our analysis, we found that adjacent regions of P2 gpHare similar to regions of the X gpORF 401 and gpORF 314(Fig. 6A). Examination of the DNA sequences revealed thatthe region of the P2H gene comprising bp 692 to 1201 is 78%identical to a region of X comprising bp 20709 to 21217 (Fig.6B). This region of X includes 147 bp at the 3' end of ORF401, the 173-bp spacer between ORF 401 and ORF 314, and189 bp at the 5' end of ORF 314. Optimal alignment of thetwo DNA sequences required the insertion of a 1-bp gapafter position 20833 of the X sequence, within ORF 401 (Fig.6B). If the GC base pair present at this position in the P2sequence is inserted into the gap in the X sequence, ORF 401is fused to ORF 314 to create an ORF of 774 amino acidcodons (Fig. 6A). The frameshift is not the artificial result ofan error in the sequencing of X: this region of X wassequenced nine times, both strands were sequenced, andsequencing was performed with inosine reactions and form-amide gels to minimize compressions of the bands in thesequencing gels (21, 82). Hence, we hypothesize (i) that thelonger ORF represents the true stfgene that encodes the sidetail fiber protein and (ii) that a frameshift mutation is presentin the common laboratory strain of X that truncates stf toORF 401 and creates an apparent second ORF, ORF 314.These hypotheses are consistent with and provide an

explanation for the following results of Montag et al. (67). (i)Synthesis of gpORF 314 from an expression vector is notobserved because ORF 314 is a gene fragment missing its 5'end and hence the signals for efficient translation. The ORF314 fragment of stfcan be expressed only when it is fused toanother gene, e.g., lacZ on the expression plasmid or gene37 in the hybrid T4 phage, that supplies the translationinitiation signals. (ii) The observed X side tail fibers are thedefective products of the truncated ORF 401 portion of thestf gene. Since nearly half of stf is missing because of the1-bp deletion, its product may be unstable and partially orcompletely degraded in the cell. Thus, the host dependencefor side tail fiber synthesis could be explained by host-dependent proteolytic activity. (iii) Antibodies raised againstthe lacZ-ORF 314 fusion product do not bind the side tailfibers of X because the ORF 314 portion of the tail fiberprotein is not present in the phage virions. This assumeseither that the ORF 314 fragment of stf is not translated orthat its product cannot associate with the defective ORF 401portion of the tail fibers.The similarities between P2 gpH and X Stf encompass

regions A and B. Both regions also occur in the constantportion of P1 gpl9 and gpl9', and region A appears in theconstant portion of Mu gpS and gpS'. The similarities arehighly significant, with identity ranging from 63 to 88%.

Since this work was performed, Edouard Kellenberger hasprovided the original strain of phage X, and Roger Hendrixhas shown that it exhibits side tail fibers and contains anextra protein of 80 to 85 kDa, as predicted by our computeranalysis (40a).

Proteins with regions similar to the T-even phage tail fiberproteins. The T-even phages are closely related, but they canbe divided into two groups based on the structure of their tailfibers. In phages similar to T4, including TuIa and TuIb, thedistal part of the tail fiber consists of a parallel dimer of theproduct of gene 37 (2, 3, 86). Assembly of the tail fiberrequires the product of gene 38, which acts as an accessoryprotein and is not incorporated into the virion (25, 66, 85,94). The host range of the T4-like phages is determined bythe carboxy-terminal portion of gp37 (2, 3, 64).The second group of T-even phages includes T2, Ox2, Ml,

and K3. For the T2-like phages, the distal portion of the tailfiber also consists of the product of gene 37. The amino-terminal two-thirds of the gene 37 products of the T4-like andthe T2-like phages (regions D, E, F, G, L, and M in Fig. 4)are homologous. However, the carboxy-terminal portion(region R of Fig. 4) of gp37 of the T2-like phages isproteolytically modified through the removal of approxi-mately 120 terminal residues, and the product of gene 38 isattached (27, 76). Furthermore, and in contrast to the T4-likephages, the host range of the T2-like phages is determined bygene 38 (27, 76). Genes 38 of the T2-like phages are homol-ogous (region U of Fig. 4) but unrelated to genes 38 of theT4-like phages (regions V, W, and Y of Fig. 4) (27, 76),reflecting the different functions of the corresponding pro-teins.

In our analysis, we identified the following similaritiesbetween regions of the tail fiber genes of the T-even phagesand other phages. (i) P2 gpH, Mu gpS, and P1 gp19 haveregions C and P in common, but between these regions is anarea in which gpH differs from the other two proteins (Fig.4). Part of this area of gpH, region 0, is also found in gp37 ofphages T2 and K3 but not in gp37 of the other T2-likephages, Ox2 and Ml (Fig. 5B). The similarity is highlysignificant (66% identity between gpH and either K3 or T2gp37 over a stretch of 56 residues). (ii) In Mu gpS and P1gpl9, regions G and H are located between the regions C andP also found in P2 gpH (Fig. 4). Region G is also found ingp37 of phages K3, T2, and T4. An alignment ofMu gpS withT2 gp37 is presented in Fig. 5C. The similarity of region G inphages Mu and P1 to the T-even phages ranges from 33 to38% identity over a stretch of 62 residues. (iii) As notedpreviously, the carboxy-terminal portion of the product ofthe side tail fiber gene, stf, of A is similar to and cansubstitute for the carboxy-terminal portion of the product ofT4 gene 37, thereby conferring a new host range on the T4hybrid (32, 63, 67). The similar regions of these two proteinsare regions S and T (Fig. 4); these regions are also found inthe appropriate positions in gp37 of the T4-like phages TuIaand TuIb. Similarity of these proteins ranges from 72 to 91%identity for region S (a stretch of about 100 residues; Fig. 5E)and from 41 to 84% identity for region T (about 154 residuesin length). (iv) Phages T3 and T7 are closely related, and the

FIG. 4. Summary of alignments of tail fiber genes. The name of each gene is given underneath the box designating that gene. A jagged edgeat the 5' or 3' end of a gene indicates that the sequence of the gene is not available beyond the region shown. Gene boxes drawn with dashedlines indicate genes for which the positions are known but the sequences have not been published. The letters inside the gene boxes indicatesimilar regions, and the boundaries between regions are indicated with dashed lines. Regions in black are similar to regions of the P2 H andG genes. The lines between genes indicate intergenic sequences. The black inverted triangles indicate the sites of recombination for inversionof the e14 P segment, the Mu G segment, and the P1 C segment. The figure is drawn to scale.

VOL. 174, 1992

on Novem


.org/D

ownloaded from

http://jb.asm.org/


TABLE 4. Regions of similarity of tail fiber structural proteins

Gene product Region Residues Gene products with similar regions [% identity]Stf [40], T4 37 [36], Tula 37 [36], TuIb 37 [36]

T2 37 [94], T4 37 [89]T4 37 [76]T4 37 [91], T2 37 [63]T4 37 [64], T2 37 [56], Mu Sv [33], P1 19v [33]T3 Tf [30], T7 17 [29]Tula 37 [69], T2 37 [63], T4 37 [54]T2 37 [47]T2 37 [93], P2 H [66]T2 37 [100], Ml 37 [45], Ox2 37 [45]T2 38 [63], Ml 38 [51], Ox2 38 [45]P2 H (2) [88], P1 19c [77], P2 H (1) [73], Mu Sc [63]P2 H [79], P1 19c [64]T4 37 [76], TuIb 37 [73], Tula 37 [72], e14 ORF Pmin [40]Tula 37 [58], TuIb 37 [56], T4 37 [41]Ox2 37 [98]Ox2 37 [98], K3 37 [45], T2 37 [45]Ox2 38 [69], K3 38 [51], T2 38 [50]P1 19c [69], P2 H (1) [68], P2 H (2) [65], ) Stf [63]P1 19v [95], P2 H [92]P1 19v [99], T2 37 [38], T4 37 [37], K3 37 [33]P1 19v [95]P1 19v [54], P2 H [49], Mu Sv' [29]Mu Sv [31], P1 19v [31], P2 H [31]Ml 37 [98]Ml 37 [98], K3 37 [45], T2 37 [45]Ml 38 [69], K3 38 [45], T2 38 [45]A Stf [77], P2 H (2) [77], P2 H (1) [74], Mu Sc [69]P2 H [69], A Stf [64]Mu S [95], P2 H [89]Mu Sv [99], T2 37 [39], T4 37 [36], K3 37 [33]Mu Sv [95]P2 H [80], Mu Sv [54], Mu Sv' [29]P1 19c [74], X Stf [73], P2 H (2) [70], Mu Sc [68]X Stf [88], P1 19c [77], P2 H (1) [70], Mu Sc [65]X Stf [79], P1 19c [69]Mu Sv [92], P1 19v [89]K3 37 [66], T2 37 [66]P1 19v [80], Mu Sv [49], Mu Sv' [31]K3 37 [94], T4 37 [84]T4 37 [64], K3 37 [63]T4 37 [87], K3 37 [56], P1 19v [39], Mu Sv [38]T4 37 [84], Tula 37 [61]Tula 37 [68], K3 37 [63], T4 37 [53]K3 37 [47]K3 37 [93], P2 H [66]K3 37 [100], Ml 37 [45], Ox2 37 [45]K3 38 [63], Ml 38 [50], Ox2 38 [45]T7 17 [87]T7 17 [90], K3 37 [30]T7 17 [39]K3 37 [89], T2 37 [84]K3 37 [76]K3 37 [91], T2 37 [64]T2 37 [87], K3 37 [64], Mu Sv [37], P1 19v [36]T2 37 [84], TuIa 37 [60]Tula 37 [62], K3 37 [54], T2 37 [53]TuIb 37 [91], Tula 37 [85], X Stf [76], e14 ORF Pmin [36]Tula 37 [47], TuIb 37 [47], X Stf [41]T3 Tf [87]T3 Tf [90], K3 37 [29]T3 Tf [39]T2 37 [61], T4 37 [60]K3 37 [69], T2 37 [68], T4 37 [62]TuIb 37 [88], T4 37 [85], X Stf [72], e14 ORF Pmin [36]TuIb 37 [84], X Stf [58], T4 37 [47]T4 37 [91], Tula 37 [88], X Stf [73], e14 ORF Pmin [36]Tula 37 [84], X Stf [56], T4 37 [47]

a The sequence is incomplete.

el4 ORF PminK3 37

SDEFGJMN0RUABSTQa

RUACGHppQaRUAaBa

CGHpA (1)A (2)BC0pDFGLMN0RUIJKDEFGLMSTIJK

MSTST

K3 38X Stf

Ml 37

Ml 38Mu ScMu Sv

Mu Sv'Ox2 37

Ox2 38P1 19c

P1 19v

P2 H

T2 37

T2 38T3 Tf

T4 37

T7 17

Tula 37

TuIb 37

12-1161-63

64-190191-565566-639664-750728-762763-10061083-11381139-1243

1-260355-394395-526520-620621-774

1-132133-237

1-262111-150175-270260-331332-425426-504412-488

1-132133-237

1-266213-247345-399602-697687-758759-852853-926168-207212-251252-383384-479481-536594-669

1-63206-568569-643644-772773-806807-10501181-12361237-1341

1-2621-361

362-447448-557

1-6364-189190-564565-639640-768769-802803-902903-1026

1-361362-447448-553

1-8788-121122-221222-382

7-108107-267

J. BACTERIOL.

on Novem


.org/D

ownloaded from

http://jb.asm.org/


TABLE 5. Regions of similarity of tail fiber assembly proteins

Gene product

e14 ORF 183

e14 ORF 293

E. coli CRF 86

A Tfa

Mu U

Mu U'

P2 G

T4 38

TuIa 38a

TuIb 38

Region

y

Vwy

y

Vwy

xy

xy

xy

Vwy

Vwy

Vwy

Position

46-153

18-7172-117118-224

1-86

12-6667-111112-194

1-9798-175

1-9798-177

1-9798-175

1-5556-9798-183

1-5556-9798-142

1-5556-9798-183

Gene products with similar regions [% identity]

e14 ORF 293 [31], X Tfa [28], E. cOli CRF 86 [23], P2 G [23], Mu U [22],Mu U' [18], T4 38 [11], TuIb 38 [11], Tula 38 [6]

T4 38 [31], Tula 38 [31], TuIb 38 [31], A Tfa [31]X Tfa [44], TuIa 38 [36], T4 38 [29], TuIb 38 [29]X Tfa [35], e14 ORF 183 [31], Mu U' [26], P2 G [22], Mu U [22],

E. coli CRF 86 [22], Tula 38 [13], T4 38 [10], TuIb 38 [10]X Tfa [56], P2 G [29], Mu U [28], e14 ORF 183 [23], e14 ORF 293 [22],Mu U' [18], T4 38 [10], Tula 38 [10] TuIb 38 [10]

T4 38 [36], Tula 38 [36], TuIb 38 [36], e14 ORF 293 [31]e14 ORF 293 [44], T4 38 [36], TuIa 38 [36], TuIb 38 [36]E. coli CRF 86 [56], P2 G [35], Mu U [35], e14 ORF 293 [35],

e14 ORF 183 [28], Tula 38 [28], T4 38 [25], TuIb 38 [25], Mu U' [22]P2 G [93], Mu U' [72]P2 G [94], Mu U' [39], X Tfa [35], E. cOli CRF 86 [28], e14 ORF 183 [22],

e14 ORF 293 [22], T4 38 [18], TuIb 38 [18], Tula 38 [13]P2 G [73], Mu U [72]Mu U [39], P2 G [38], e14 ORF 293 [26], X Tta [22], e14 ORF 183 [18],

E. coli CRF 86 [18], T4 38 [12], TuIb 38 [12], Tula 38 [8]Mu U [93], Mu U' [73]Mu U [94], Mu U' [38], A Tfa [35], E. cOli CRF 86 [29], e14 ORF 183 [23],

e14 ORF 293 [22], T4 38 [17], TuIb 38 [17], TuIa 38 [13]TuIa 38 [100], TuIb 38 [100], A Tfa [36], e14 ORF 293 [31]TuIb 38 [100], Tula 38 [60], A Tfa [36], e14 ORF 293 [29]TuIb 38 [99], Tula 38 [60], A Tfa [25], Mu U [18], P2 G [17], Mu U' [12],

e14 ORF 183 [11], e14 ORF 293 [10], E. cOli CRF 86 [10]T4 38 [100], TuIb 38 [100], A Tfa [36], e14 ORF 293 [31]T4 38 [60], TuIb 38 [60], A Tfa [36], e14 ORF 293 [36]T4 38 [60], Tula 38 [60], A Tfa [28], e14 ORF 293 [13], Mu U [13],P2 G [13], E. cOli CRF 86 [10], Mu U' [8], e14 ORF 183 [6]

T4 38 [100], TuIb 38 [100], K Tfa [36], e14 ORF 293 [31]T4 38 [100], TuIa 38 [60], K Tfa [36], e14 ORF 293 [29]T4 38 [99], TuIa 38 [60], K Tfa [25], Mu U [18], P2 G [17], Mu U' [12],

e14 ORF 183 [11], e14 ORF 293 [10], E. coli CRF 86 [10]a The sequence is incomplete.

products of their tail fiber genes, T3 tf and T7 17, are verysimilar. These proteins contain a region, J, that is 29 to 30%identical over a stretch of 87 residues to a region in gp37 ofthe T-even phage K3 (Fig. 5D). Region J does not occur inany of the other T-even phage tail fiber proteins.P2 gpG and Mu gpU and gpU'. Mu gpU or gpU' is essential

for the formation of functional tail fibers; however, it is notknown whether these proteins are incorporated into thevirion (34). Alignment of the sequences of P2 gpG with MugpU and gpU' revealed that the three proteins are remark-ably similar (Fig. 7A). P2 gpH and Mu gpU are 93% identicalover their entire lengths. There is less similarity at the DNAlevel since there are 40 silent point mutations. The Mu gpU'is 72 to 73% identical in region X to gpH and gpU but only 38to 39% identical in region Y (Table 5). The similarity of gpUand gpU' is reflected in the ability of gene U' to complementMu U mutants in Mugin+ phage (34). The P1 U and U' geneshave been shown through electron microscope heteroduplexstudies of the invertible G and C segments of Mu and P1 tobe very similar to the Mu U and U' genes (19). The similarityof U to U' was also apparent in these studies since someheteroduplex formation occurred between the regions con-taining U and U' (19). Although the DNA sequences of theP1 U and U' genes have not been published, it has beenreported that they are very similar to sequences of the Mu Uand U' genes (50).

Tail fiber assembly proteins. As noted earlier, T4 gp38 isnot incorporated into the virion but rather plays a catalyticrole in the assembly of dimers of gp37 (25, 66, 85, 94). The

product of the A tfa (ORF 194) gene is similar to T4 gp38(27% identity over the entire length of Tfa) (Fig. 7B) (32),and the product of tfa complements amber mutants of gene38 of T4 (65). Thus, gp38 and Tfa serve as tail fiber assemblyproteins. It has since been shown that genes 38 of the T4-likephages Tula and TuIb are homologous to T4 38 (64); TuIbgp38 is 99% identical to T4 gp38 over the lengths of theproteins, and Tula gp38 is 100% identical to T4 gp38 inregion V and 60% identical in regionsW and Y (Fig. 4; Table5).

In our analysis, we found suggestive similarities betweenTfa and P2 gpG (and hence Mu gpU). P2 gpG is 35% identicalto A Tfa in region Y at the carboxy termini of the proteins(Fig. 7B). As Fig. 7 indicates only the identical amino acidsand as there are several cases of very conservative substi-tutions, the homologies are even higher than the figureshows. The amino-terminal portions of the proteins (regionX of gpG and regions V and W of Tfa) have no similarity.The similarities in region Y suggest that gpG acts as the tailfiber assembly protein of P2. Similarity of gpG to T4 gp38 inregion Y is only 17% identity. It is possible that gpG, Tfa,and T4 gp38 are homologous proteins and that the similar-ities of the proteins reflect the similarities of the tail fiberproteins on which they act. The K side tail fiber protein, Stf,is similar in regions S and T to T4 gp37 and in regions A andB to P2 gpH, but T4 gp37 and P2 gpH have no similarregions.

IfgpG acts similarly to T4 gp38, then its function would beto act catalytically to aid the formation of dimers of P2 gpH.

VOL. 174, 1992

on Novem


.org/D

ownloaded from

http://jb.asm.org/

A 12 112 1

B 12 R

164- 215 .PTS208- 235 ... LN<n

A

rss :se 2 " S * @ ^ . .s *o ;¢ 4,

,.Nm.SN £ I.. *V

1- 64

12 1 65- 174 01mLTDAGTLDAVAIaZ3T3AIAZGSQR?MsvZZv3V3LT11IMA1YDYVD1IAZ1ZQs3PJWDA5LTA1

A A12 1 175- 264 w1Q4jSAT4iSZTLT1<K=AMu So 106-155...NSWP1 19a 179- 247 ....D4DSVT<i=DVLt.. zX atf 350- 407 ..A A S

B

12 R 265- 354 ti E 1 :T".W ALvrla a aPI l9a 345- 370 TV@'* ^^"bSdiX atf 408- 497 A>ZL1135Z EI "LMDV1.6DWIM

121 355- 442 @cA1aboS-11-r.a a g * gogi g

Mu So 162- 177mu ST 178- 233 asMu ST 178- 234 _PI 190 371- 3 91 4MI->>>^:';#lp119- 392- 427 sauzrzs = - X11Ilo 590- 660 ...flLT a 55.a-4* aMl*a egl atE *498- 532 t IAA*14 ORU Pnna 1- 19 []XI^QA.14 011 Plinv 20- 24 SAV

P2 N 443- 532 ZZMM-;a sEvIM.x M. I8aZMmu tv 234- 275 u'* ...s a.Mu ST' 235- 251 y TAN.P1 19v 661- 702 :.r.esA;I;**_.______________s___

12 37 1176-1232 eas13 37 1076-1134

12 11 533- 621 m@DI111fLIlg WC$TlNLV0AJT1TPVVAZL4D AGWN1IQ3DIVWXa amu ST 421- 454 ...WKCmu IT' 407- 440 m.6I 1P1 19t 846- 880 *....c.12 37 1233-1248 E1NPxwnDZ11...13 37 1135-1150 MMPvUnDZ .=...

p

12 R 622- 669 Of -. *sa -a''mu ST 455- 504 L61yX- amu ST 44i- 486 W VaaW D u-"11 19- 861- 926

CMuST 255- 334 ... 712LAZV-ZI1IxdiAXs3, *|C 2 37 564- 646 .. .?M ST.,::

1< ~~~~~~~~~J,

n 3 tf ! 337- 439 . ..AQAT!-xL-Qp=M-Y".Zgl\fAM<! -x -1V 1C 37 6S9- 742 . . . JCSDCFk;vTn SCy PSC-Si SX< AoSLt< S MVSj :=Q

J~~~~~13 tf 440- 45213 37 743- 755 s aso0I :

i ~~~~~~~~AEs

37 790- $77 .....LJgf<................. __E t 515- 55M...

*14 031 Pua 7- 92...D

S 114 37 678- 907 EIAI). atE 596- 625 EYEd.14 O01 la4n 93- 121 MDM

GIl - AU !3'DII*

FIG. 5. Alignments of tail fiber structural proteins. (A) Alignment of the two copies of region A in P2 H protein; (B) alignment of P2 Hprotein with tail fiber structural proteins from phages Mu, P1, X, e14, T2, and K3; (C) alignment of the G regions of Mu Sv and T2 37 proteins;(D) alignment of the J regions of T3 tf and K3 37 proteins; (E) alignment of the S regions of T4 37, X stf, and e14 Pmin proteins.

1472

us xKrRTVXTT-h-n-h-h-xT--h-h6xTAPGRPJMXTTMAVGDGGGKLPVPDAOQTGLXRZVNRR%ZXXXSQDXRNSN=XULVXPPRV

1-0..rTPAY ZAIJ. AO

XTI;*WAPZX:::

on Novem


.org/D

ownloaded from

http://jb.asm.org/


AH

ORF 401 ORF 314

sif

B 1P2 60B )X 20697

P2 779X 20796

P2 879X 20895

P2 979X 20995

P2 1079X 21095

P2 1179X 21195

a*_sc; s_~~~~~~~~~~~~cccxcrcE_E s~~~~~~~~~~~~~~~~~~~~~

,r>.".x*fewsiNe..tie.zi..

a:S~~~~~~~~~~~~~~~~~~~~~~~_TCi;f= _G CPTCGNC.Tc s s s f f = u # j, ,s. X di.oc%2.wS.LEf>8ZG

FIG. 6. Alignment of similar regions of the P2 H gene and the X stf gene. (A) Alignments based on protein sequences. The solid regionsindicate similar protein sequences in the gene products of P2 H, A ORF 401 and ORF 314, and X stf. (B) Alignment of the DNA sequencesof P2 H and X stf. The box indicates the position of a 1-bp gap that has been introduced into the X sequence after position 20833.

This means there would be 12 copies of gpH per virion; thisis not far from the estimate of 14 to 18 copies (55). This alsomeans that gpG may not be incorporated into P2 virions. Noprotein of a size appropriate for gpG (predicted Mr 20,264)has been identified as a tail proteiti of P2 (55).A cryptic gene ofE. coli similar to X ffa. Comparison of the

DNA sequence of X tfa with the DNA seqUences in GenBankrevealed that there is a highly similar cryptic gene of E. coliupstream of appY (1). The DNA sequence of this crypticgene (1) is presented in Fig. 8. If the TGA ochre stop codonat positions 307 to 309 is altered to a tryptophan TGG codon,a cryptic reading frame of 86 amino acid codons (hence thename, CRF 86) results. In the CRF 86 gene product, astretch of 50 residues near the carboxy terminus has 92%identity with the same region of the X tfia gene product (Fig.7B). Flanking DNA sequences have no similarity to theDNA sequence of X. It is unknown whether CRF 86 isexpressed by E. coli.

Potential tail fiber genes in the invertible P segment of e14.e14 is a defective cryptic prophage located at 25 min on theE. coli K-12 chromosome. e14 contains a 1.8-kb invertibleregion, the P segment, the function of which is unknown (72,93). Inversion of the P segment is controlled by thepin gene,which is related to the gin and cin invertase genes of Mu andP1, respectively (93). Four significant ORFs can be identifiedin the P segment (Fig. 4) (72). The sequences of two of these,ORF Pmin and ORF Pl,u1, are incomplete since the constant5' end shared by these ORFs has not been sequenced.

Assuming that transcription enters the P segment from theleft (Fig. 4), ORF Pmin and ORF 293 would be transcribedwhen the segment is in the minus orientation and ORF Pplusand ORF 183 would be transcribed when the segment is inthe plus orientation. ORF 293 and ORF 183, which are onopposite strands of the DNA, overlap by 462 bp. Althoughthe el4pin gene can complement Mu gin, the e14 P segmentshows no relationship to the Mu G segment in tests involvingcomplementation, marker rescue, or hybridization (72).Our analysis suggests that the P segment serves as a

genetic switch that allows e14 to alternate tail fiber proteins.When the P segment is in the minus orientation, ORF Pminand ORF 293 could be expressed. The ORF Pmin geneproduct contains a region, S, that is also present in the gene37 tail fiber structural proteins of TuIa, TuIb, and T4 as wellas in the Stf side tail fiber protein of X (Fig. 4 and SE).However, ORF Pmin protein is missing region T, whichappears in the Tula, TuIb, T4, and X proteins. The geneimmediately downstream of ORF Pmin, ORF 293, encodes aprotein that contains regions V, W, and Y (Fig. 4 and 7B).Hence, ORF 293 is closely related to the tail fiber assemblygenes of Tula, TuIb, T4, and X and more distantly related tothe tail fiber assembly genes of P2, Mu, and presumably P1.ORF 293 is unusual in that it contains a region beyond regionY that is not present in the tail fiber assembly proteins of theother phages.When the P segment is in the plus orientation, ORF Pp,us

and ORF 183 could be expressed. ORF Pp,us protein is not

I .. I I

VOL. 174, 1992

I 1 I

I

on Novem


.org/D

ownloaded from

http://jb.asm.org/


A 12 0

Mu

Mu V

12 0

Mu v

B Xtf.54 38*14 293.14 01 183

X. coli CRY o6X tfaT4 38*14 OY 293*14 18312 a

@01l CR1 8sX tfaT4 38*14 O03 293*14 031 18312 a

1- 90

1- 901- 90

91- 17591- 17591- 177

1- 831- 721- 891-

1- 65

84- 170

73- 14990- 17719- 106

97- 156

66- 86

171- 194150- 183178- 222107- 151157- 175

mArR"zQw-zk!RwzwKT

y I1-tmosLLK8rzEv415r

K ;- _l..~,

iExa*-sX3--ba

FIG. 7. Alignments of tail fiber assembly proteins. (A) Alignment of P2 G protein with the Mu U and U' proteins; (B) alignment of P2 Gprotein with tail fiber assembly proteins from phages X, T4, and e14 and with CRF 86, a cryptic gene of E. coli. Identities are indicated withblack boxes at any position at which two proteins have identical residues.

similar to any other protein in the sequence data bases, butORF 183 protein contains regions W and Y, which arecharacteristic of the tail fiber assembly proteins (Fig. 4 and7B). (In order to maximize the similarity of ORF 183 proteinto the tail fiber assembly proteins, we assumed that transla-tion of ORF 183 begins at a TTG start codon at position 397of the sequence of Plasterk and van de Putte [72] with the Psegment reversed. Translation could begin alternatively at a

GTG codon at position 511 or an ATG codon at position601.) ORF 183 protein lacks region V, which precedes regionW in the TuIa, TuIb, T4, and X tail fiber assembly proteins.ORF 183 protein is like ORF 293 protein in that it contains a

region beyond region Y; these carboxy-terminal regions ofthe ORF 183 and ORF 293 proteins show some similarity inthe first 22 residues beyond region Y (Fig. 7B) but are

otherwise unrelated.Methods of extending host range. Both phage and bacterial

cells are under selective pressure, the cells to resist infectionand the phage to initiate infection. There are a number ofways in which a phage can alter or extend its host range.

(i) Point mutations in the tail fiber genes have been shownin a number of systems to alter host range. This suggests that

101

201

301

the tail fibers have evolved in such a way that minor changesin the tail fiber can counteract host range mutants of thebacterial hosts.

(ii) It is possible for a phage to acquire a new host range byobtaining a portion of a tail fiber gene from another phage.We suggest that the similarities in the tail fiber genes of suchunrelated phages as P2, P1, Mu, X, and T4, etc., providepersuasive evidence that phages have done this repeatedly.

(iii) Similarly, it is possible for a phage to alter its hostrange by picking up a fragment of genetic debris left in thebacterial chromosome by another phage. For example, frag-ments of genes that hybridize to genes 36 and 37 of T2 havebeen detected in the E. coli chromosome (77). As a secondexample, we have documented above the presence of a

cryptic gene in the E. coli chromosome, CRF 86, that issimilar to the X tail fiber assembly gene.

(iv) Phages such as Mu, P1, and probably e14 use invert-ible DNA to alternate between two mutually exclusive setsof tail fiber genes which confer different host ranges.

(v) Some phages may simultaneously use two separatesystems of tail fibers, each recognizing a different receptor.Phages X and T5 may provide examples of such a system.

CRF 8 6M L P Q H S D I E I A W Y A S I Q Q E P N G W K T V

GGCAATGAATTACAAGGGGTTAAATGCTGCCGCAGCATAGCGATATTGAAA TAGCCTGGTATGCTTCAATACAGCAGGAGCCGAATGGCTGGAAGACCGT

T T Q F Y I Q E F S E Y I A P L Q D A V D L E I A T E E E R S L L

CACCACACAGTTCTACATCCAGGAATTCAGTGAGTATATTGCGCCACTGCAGGATGCTGTAGATCTGGAAATCGCAACGGAGGAAGAAAGATCGTTGCTG

w

E A * K K Y R V L L N R V D T S V A P D I E W L I Q P * *

GAAGCCTGAAAAAAGTATCGGGTGCTGCTAAACCGTGTGGACACTTCCGTAGCACCAGATATCGAGTGGCTTATTCAACCATAATAAACAGTATGTATATTGG

FIG. 8. DNA sequence of the CRF 86 gene of E. coli and the deduced amino acid sequence of its product. The DNA sequence is fromreference 1 and is numbered identically. At position 309 we have extended the length of the ORF by replacing the TGA (ochre stop) codonwith a TGG (tryptophan) codon.

.xi4xp x9

-tp BOW

J. BACTERIOL.

on Novem


.org/D

ownloaded from

http://jb.asm.org/


The normal receptor for A is the lamB gene product (73), amaltose transport protein. Host recognition of phage lambdais determined by the product of gene J, which makes up thesingle terminal tail fiber (15, 26, 68). (The product of J isunrelated to any of the tail fiber genes examined and in factis unrelated to any of the genes in the GenBank and EMBLdata bases [data not shown].) However, the laboratory strainof A apparently carries defective genes, stf and tfa, whichencode the side tail fibers. T4 hybrids containing part of stand all of tfa use the outer membrane protein OmpC as areceptor (67). This suggests that a X phage with functional stfand tfa genes could recognize two receptors, LamB andOmpC. Since the defect in the side tail fiber genes of Xappears to be a 1-bp deletion in stf, it is possible thatcompensatory errors in DNA replication could lead topartially or fully functional side tail fiber genes. This mayexplain the observation that X can transduce strains with adeletion of the lamB gene at low efficiency (14). Phage T5 hastwo sets of tail fibers, the L-shaped tail fibers, the product ofthe ltf gene, which bind to the lipopolysaccharide, and asingle straight terminal tail fiber, the product of the oad gene,which binds to the FhuA receptor protein (13, 39, 40). TheL-shaped tail fibers accelerate adsorption, but neither theL-shaped tail fibers nor the lipopolysaccharide receptor areessential for infection (39, 40, 79). The oad gene, whichencodes the terminal receptor binding protein of T5, hasbeen sequenced (54); the oad gene product shows no simi-larity to any of the tail fiber proteins or to any other proteinin the data bases (data not shown).

(vi) A phage may have the ability to recognize twoseparate receptors using the same tail fiber. This has beendocumented in the case of T4, which can use its tail fibers,the product of gene 37, to recognize the outer membraneprotein OmpC or the lipopolysaccharide as a receptor withequivalent efficiency (64).

Evolution of tail fiber genes. The question of how the tailfiber genes have evolved is tied up in the question of theevolution of the phages themselves (17, 18). Campbell hasargued that the gene pool available to a particular phagepotentially includes, among other sources, other phages andthe genetic debris left behind by phages in bacterial chromo-somes (17). Certainly recombination can occur among thehomologous sequences carried by members of a particularfamily of phages. But the similarities in the tail fiber genes ofphages of different families, as described in this paper,provide evidence that illegitimate recombination occurs atpreviously unappreciated levels and that phages are takingadvantage of the gene pool available to them to alter theirhost ranges under selective pressures. The result is thescrambling of fragments of tail fiber genes through horizontaltransfer among unrelated phages, resulting in highly chi-meric genes such as the P2 H gene, which has elementsrelated to X, Mu, P1, K3, and T2.

It appears that under selective pressures, evolution of thetail fiber genes moves faster than evolution of other phagegenes. For example, the T-even phages (T4, Tula, TuIb, T2,K3, Ml, and Ox2) are known to be closely related. Theamino-terminal portions of their tail fiber proteins are highlyconserved, yet the carboxy-terminal portions exhibit re-markable divergences. This is apparently because the amino-terminal portions of the tail fiber proteins interact with otherphage proteins (which seem to be highly conserved within aphage family); the carboxy-terminal portions interact withthe host receptors and therefore experience greater selectivepressure for change (33). In the cases of P2, Mu, and P1, theamino-terminal portions of their respective tail fiber proteins

are very different, presumably reflecting the very differentproteins with which they must interact. Apparently thesephages have exchanged among themselves DNA fragmentsthat generated the similarities in the carboxy-terminal por-tions of their tail fiber proteins.

ACKNOWLEDGMENTS

This work was supported by grants 72 and 7078 from the SwedishMedical Research Council to E.H., an American Cancer Societypostdoctoral fellowship to C.H., and Public Health Service grantA108722 from the National Institutes of Health to R.C.We are especially grateful to Robert Haselkorn, in whose labora-

tory the computer analysis was carried out. The generous gift ofsynthetic oligonucleotides from KABI-GENE is gratefully acknowl-edged. We thank Edouard Kellenberger, Gail Christie, and RainerZiermann for helpful discussions; Erich Six for providing phage; andMel Sunshine, G. Bertani, Roger Hendrix, and Alan Coulson forcommunicating unpublished results.

REFERENCES1. Atlung, T., A. Nielsen, and F. G. Hansen. 1989. Isolation,

characterization, and nucleotide sequence of appY, a regulatorygene for growth-phase-dependent gene expression in Esche-richia coli. J. Bacteriol. 171:1683-1691.

2. Beckendorf, S. K. 1973. Structure of the distal half of thebacteriophage T4 tail fiber. J. Mol. Biol. 73:37-53.

3. Beckendorf, S. K., J. S. Kim, and I. Lielausis. 1973. Structure ofbacteriophage T4 genes 37 and 38. J. Mol. Biol. 73:17-35.

4. Bertani, G. 1951. Studies on lysogenesis. I. The mode of phageliberation by lysogenic Escherichia coli. J. Bacteriol. 62:293-300.

5. Bertani, G. 1975. Deletions in bacteriophage P2. Circularity ofthe genetic map and its orientation relative to the DNA dena-turation map. Mol. Gen. Genet. 136:107-137.

6. Bertani, G. (California Institute of Technology). 1991. Personalcommunication.

7. Bertani, G., E. Ljungquist, K. Jagusztyn-Krynicka, and S. Jupp.1978. Defective particle assembly in wild type P2 bacteriophageand its correction by the Ig mutation. J. Gen. Virol. 38:251-261.

8. Bertani, L. E., and G. Bertani. 1970. Preparation and charac-terization of temperate, noninducible bacteriophage P2. J. Gen.Virol. 6:201-212.

9. Bertani, L. E., and G. Bertani. 1971. Genetics of P2 and relatedphages. Adv. Genet. 16:199-237.

10. Bertani, L. E., and E. W. Six. 1988. The P2-like phages and theirparasite, P4, p. 73-143. In R. Calendar (ed.), The bacterio-phages, vol. 2. Plenum Publishing Corp., New York.

11. Beumer, J., M. P. Beumer-Jochmans, J. Dirkx, and D. Dekegel.1965. Etat actuel des connaissances concernant la nature et lalocalisation des recepteurs des bacteriophages dans la paroicellulaire des Shigella et des Escherichia. Bull. Acad. R. Med.Belg. 5:749-790.

12. Bilofsky, H. S., and C. Burks. 1988. The GenBank geneticsequence data bank. Nucleic Acids Res. 16:1861-1864.

13. Braun, V., K. Schaller, and H. Wolff. 1973. A common receptorprotein for phage T5 and colicin M in the outer membrane ofEscherichia coli B. Biochim. Biophys. Acta 323:87-97.

14. Braun-Breton, C., and M. Hofnung. 1978. Explanations ac-counting for transduction by bacteriophage X in maltose nega-tive bacteriophage X resistant mutants of Escherichia coli K-12.Mol. Gen. Genet. 159:143-149.

15. Buchwald, M., and L. Siminovitch. 1969. Production of serum-blocking material by mutants of the left arm of the X chromo-some. Virology 38:1-7.

16. Bukhari, A. I., and L. Ambrosio. 1978. The invertible segmentof bacteriophage Mu determines the adsorption properties ofMu particles. Nature (London) 271:575-577.

17. Campbell, A. 1988. Phage evolution and speciation, p. 1-14. InR. Calendar (ed.), The bacteriophages, vol. 1. Plenum Publish-ing Corp., New York.

18. Campbell, A., and D. Botstein. 1983. Evolution of the lambdoidphages, p. 365-380. In R. W. Hendrix, J. W. Roberts, F. W.

VOL. 174, 1992

on Novem


.org/D

ownloaded from

http://jb.asm.org/


Stahl, and R. A. Weisberg (ed.), Lambda II. Cold Spring HarborLaboratory, Cold Spring Harbor, N.Y.

19. Chow, L. T., and A. I. Bukhari. 1976. The invertible DNAsegments of coliphage Mu and P1 are identical. Virology 74:242-248.

20. Christie, G. E., and R. Calendar. 1985. Bacteriophage P2 latepromoters. II. Comparison of the four late promoter sequences.J. Mol. Biol. 181:373-382.

21. Coulson, A. (MRC Laboratory of Molecular Biology). 1990.Personal communication.

22. Court, D., and A. B. Oppenheim. 1983. Phage lambda's acces-sory genes, p. 251-277. In R. W. Hendrix, J. W. Roberts, F. W.Stahl, and R. A. Weisberg (ed.), Lambda II. Cold Spring HarborLaboratory, Cold Spring Harbor, N.Y.

23. Daniels, D. L., J. L. Schroeder, W. Szybalski, F. Sanger, A. R.Coulson, G. F. Hong, D. F. Hill, G. B. Petersen, and F. R.Blattner. 1983. Complete annotated lambda sequence, p. 519-676. In R. W. Hendrix, J. W. Roberts, F. W. Stahl, and R. A.Weisberg (ed.), Lambda II. Cold Spring Harbor Laboratory,Cold Spring Harbor, N.Y.

24. Dayhoff, M. O., R. M. Schwartz, and B. C. Orcutt. 1978. Amodel of evolutionary change in proteins, p. 345-352. In M.Dayhoff (ed.), Atlas of protein sequence and structure, vol. 5,suppl. 3. National Biomedical Research Foundation, SilverSpring, Md.

25. Dickson, R. C. 1973. Assembly of bacteriophage T4 tail fibers. J.Mol. Biol. 79:633-647.

26. Dove, W. F. 1966. Action of the lambda chromosome. I. Controlof functions late in bacteriophage development. J. Mol. Biol.19:187-201.

27. Drexler, K., I. Riede, D. Montag, M.-L. Eschbach, and U.Henning. 1989. Receptor specificity of the Escherichia coliT-even type phage Ox2. J. Mol. Biol. 207:797-803.

28. Dunn, J. J., and F. W. Studier. 1983. The complete nucleotidesequence of bacteriophage T7 DNA, and the locations of T7genetic elements. J. Mol. Biol. 166:477-535.

29. Edgar, R. S., and I. Liesausis. 1965. Serological studies withmutants of phage T4D defective in genes determining tail fiberstructure. Genetics 52:1187-1200.

30. Faelen, M., A. Toussaint, N. Lefvebvre, M. Mergeay, J. Braipson-Thiry, and G. Thiry. 1981. Certaines souches de Erwinia sontsensibles au bacteriophage Mu. Arch. Intern. Physiol. Biochim.89:55-56.

31. Franklin, N. C. 1961. Serological study of tail structure andfunction in coliphages T2 and T4. Virology 14:417-429.

32. George, D. G., L.-S. L. Yeh, and W. C. Barker. 1983. Unex-pected relationships between the bacteriophage X hypotheticalproteins and bacteriophage T4 tail-fiber proteins. Biochem.Biophys. Res. Commun. 115:1061-1068.

33. Giphart-Gassler, M., R. H. A. Plasterk, and P. van de Putte.1982. G inversion in bacteriophage Mu: a novel way of genesplicing. Nature (London) 297:339-342.

34. Grundy, F. J., and M. M. Howe. 1984. Involvement of theinvertible G segment in bacteriophage Mu tail fiber biosynthe-sis. Virology 134:296-317.

35. Guidolin, A., J.-M. Zingg, and W. Arber. 1989. Organization ofthe bacteriophage P1 tail-fibre operon. Gene 76:239-243.

36. HaggArd-Ljungquist, E. Unpublished data.37. Hagghrd-Ljungquist, E. 1990. Bacteriophage P2, p. 163-169. In

S. J. O'Brien (ed.), Genetic maps, 5th ed., book 1. Cold SpringHarbor Laboratory, Cold Spring Harbor, N.Y.

38. Halling, C., and R. Haselkorn. Unpublished data.39. Heller, K. J., and V. Braun. 1979. Accelerated adsorption of

bacteriophage T5 to Escherichia coli F, resulting from revers-

ible tail fiber-lipopolysaccharide binding. J. Bacteriol. 139:32-38.

40. Heller, K. J., and V. Braun. 1982. Polymannose 0 antigens ofEscherichia coli, the binding sites for the reversible adsorptionof bacteriophage T5+ via the L-shaped tail fibers. J. Virol.41:222-227.

40a.Hendrix, R. Personal communication.41. Hendrix, R. W. 1971. Identification of proteins coded in phage

lambda, p. 355-370. In A. D. Hershey (ed.), The bacteriophage

lambda. Cold Spring Harbor Laboratory, Cold Spring Harbor,N.Y.

42. Hiestand-Nauer, R., and S. lida. 1983. Sequence of the site-specific recombinase gene cin and of its substrates serving in theinversion of the C segment of bacteriophage P1. EMBO J.2:1733-1740.

43. Hocking, S. M., and J. B. Egan. 1982. Genetic characterizationof twelve P2-186 hybrid bacteriophages. Mol. Gen. Genet.187:174-176.

44. lida, S. 1984. Bacteriophage P1 carries two related sets of genesdetermining its host range in the invertible C segment of itsgenome. Virology 134:421-434.

45. lida, S., J. Meyer, K. E. Kennedy, and W. Arber. 1982. Asite-specific, conservative system carried by bacteriophage P1.Mapping the recombinase gene cin and the cross-over sites cixfor the inversion of the C segment. EMBO J. 1:1445-1453.

46. Kahmann, R., and D. Kamp. 1987. Sequence of the right end ofMu, p. 297-308. In N. Symonds, A. Toussaint, P. van de Putte,and M. M. Howe (ed.), Phage Mu. Cold Spring Harbor Labo-ratory, Cold Spring Harbor, N.Y.

47. Kamp, D. 1981. Invertible deoxyribonucleic acid: the G segmentof bacteriophage Mu, p. 73-76. In D. Schlessinger (ed.), Micro-biology-1981. American Society for Microbiology, Washing-ton, D.C.

48. Kamp, D. 1987. The evolution of Mu, p. 259-269. In N.Symonds, A. Toussaint, P. van de Putte, and M. M. Howe (ed.),Phage Mu. Cold Spring Harbor Laboratory, Cold Spring Har-bor, N.Y.

49. Kamp, D., R. Kahmann, D. Zipser, T. R. Broker, and L. T.Chow. 1978. Inversion of the G DNA segment of phage Mucontrols phage infectivity. Nature (London) 271:577-580.

50. Kamp, D., E. Kardas, Y. Ritthaler, R. Sandulache, R.Schmucker, and B. Stern. 1984. Comparative analysis of invert-ible DNA in phage genomes. Cold Spring Harbor Symp. Quant.Biol. 49:301-311.

51. Kamp, D., and R. Sandulache. 1983. Recognition of cell surfacereceptors is controlled by invertible DNA of phage Mu. FEMSMicrobiol. Lett. 16:131-135.

52. Katsura, I. 1983. Tail assembly and injection, p. 331-346. InR. W. Hendrix, J. W. Roberts, F. W. Stahl, and R. A. Weisberg(ed.), Lambda II. Cold Spring Harbor Laboratory, Cold SpringHarbor, N.Y.

53. Koch, C., G. Mertens, F. Rudt, R. Kahmann, R. Kanaar,R. H. A. Plasterk, P. van de Putte, R. Sandulache, and D. Kamp.1987. The invertible G segment, p. 75-91. In N. Symonds, A.Toussaint, P. van de Putte, and M. M. Howe (ed.), Phage Mu.Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.

54. Krauel, V., and K. J. Heller. 1991. Cloning, sequencing, andrecombinational analysis with bacteriophage BF23 of the bac-teriophage T5 oad gene encoding the receptor-binding protein.J. Bacteriol. 173:1287-1297.

55. Lengyel, J. A., R. N. Goldstein, M. Marsh, and R. Calendar.1974. Structure of the bacteriophage P2 tail. Virology 62:161-174.

56. Lindahl, G. 1969. Genetic map of bacteriophage P2. Virology39:839-860.

57. Lindahl, G. 1971. On the control of transcription in bacterio-phage P2. Virology 46:620-633.

58. Lindahl, G. 1974. Characterization of conditional lethal mutantsof bacteriophage P2. Mol. Gen. Genet. 128:249-260.

59. Lindberg, A. A. 1973. Bacteriophage receptors. Annu. Rev.Microbiol. 27:205-241.

60. Ljungquist, E., and L. E. Bertani. 1983. Properties and productsof the cloned int gene of bacteriophage P2. Mol. Gen. Genet.192:87-94.

61. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecularcloning: a laboratory manual. Cold Spring Harbor Laboratory,Cold Spring Harbor, N.Y.

62. Maxam, A., and W. Gilbert. 1980. Sequencing end-labeled DNAwith base specific cleavages. Methods Enzymol. 65:499-560.

63. Michel, C. J., B. Hacq, D. G. Argues, and T. A. Bickle. 1986. Aremarkable amino acid sequence homology between a phageT4 tail fiber protein and ORF314 of phage X located in the

J. BACTERIOL.

on Novem


.org/D

ownloaded from

http://jb.asm.org/


tail operon. Gene 44:147-150.64. Montag, D., S. Hashemolhosseini, and U. Henning. 1990. Recep-

tor-recognizing proteins of T-even type bacteriophages. Thereceptor-recognizing area of proteins 37 of phages T4, Tula andTuIb. J. Mol. Biol. 216:327-334.

65. Montag, D., and U. Henning. 1987. An open reading frame in theEscherichia coli bacteriophage X genome encodes a protein thatfunctions in assembly of the long tail fibers of bacteriophage T4.J. Bacteriol. 169:5884-5886.

66. Montag, D., I. Riede, M.-L. Eschbach, M. Degen, and U.Henning. 1987. Receptor recognizing proteins of T-even typebacteriophages. Constant and hypervariable regions and anunusual case of evolution. J. Mol. Biol. 196:165-174.

67. Montag, D., H. Schwarz, and U. Henning. 1989. A component ofthe side tail fiber of Escherichia coli bacteriophage X canfunctionally replace the receptor-recognizing part of a long tailfiber protein of the unrelated bacteriophage T4. J. Bacteriol.171:4378-4384.

68. Mount, D. W. A., A. W. Harris, C. R. Fuerst, and L. Simino-vitch. 1968. Mutations in bacteriophage lambda affecting parti-cle morphogenesis. Virology 35:134-149.

69. Norrander, J., T. Kempe, and J. Messing. 1983. Construction ofimproved M13 vectors using oligodeoxynucleotide-directed mu-tagenesis. Gene 26:101-106.

70. Oliver, D. B., and R. A. Crowther. 1981. DNA sequence of thetail fibre genes 36 and 37 of bacteriophage T4. J. Mol. Biol.153:545-568.

71. Pearson, W. R., and D. J. Lipman. 1988. Improved tools forbiological sequence comparison. Proc. Natl. Acad. Sci. USA85:2444-2448.

72. Plasterk, R. H. A., and P. van de Putte. 1985. The invertibleP-DNA segment in the chromosome of Escherichia coli. EMBOJ. 4:237-242.

73. Randall-Hazelbauer, L., and M. Schwartz. 1973. Isolation of thebacteriophage lambda receptor from Escherichia coli. J. Bacte-riol. 116:1436-1446.

74. Riede, I., K. Drexler, M.-L. Eschbach, and U. Henning. 1986.DNA sequence of the tail fiber genes 37, encoding the receptorrecognizing part of the fiber, of bacteriophages T2 and K3. J.Mol. Biol. 191:255-266.

75. Riede, I., K. Drexler, M.-L. Eschbach, and U. Henning. 1987.DNA sequences of genes 38 encoding a receptor-recognizingprotein of bacteriophage T2, K3, and of K3 host range mutants.J. Mol. Biol. 194:31-39.

76. Riede, I., K. Drexler, H. Schwarz, and U. Henning. 1987. T-eventype bacteriophages use an adhesin for recognition of cellularreceptors. J. Mol. Biol. 194:23-30.

77. Riede, I., M.-L. Eschbach, and U. Henning. 1985. Presence ofDNA, encoding parts of bacteriophage tail fiber genes, in thechromosome of Escherichia coli K-12. J. Bacteriol. 163:832-836.

78. Rosenberg, M., and D. Court. 1979. Regulatory sequencesinvolved in the promotion and termination of RNA transcrip-tion. Annu. Rev. Genet. 13:319-353.

79. Saigo, K. 1978. Isolation of high-density mutants and identifica-tion of non-essential structural proteins in bacteriophage T5:dispensability of L-shaped tail fibers and a secondary headprotein. Virology 85:422-433.

80. Sandulache, R., P. Prehm, D. Expert, A. Toussaint, and D.

Kamp. 1985. The cell wall receptor for bacteriophage Mu G(-)in Erwinia and Eschenchia coli C. FEMS Microbiol. Lett.28:307-310.

81. Sandulache, R., P. Prehm, and D. Kamp. 1984. Cell wallreceptor for bacteriophage Mu G(+). J. Bacteriol. 160:299-303.

82. Sanger, F., A. R. Coulson, G. F. Hong, D. F. Hill, and G. B.Petersen. 1982. Nucleotide sequence of bacteriophage A DNA.J. Mol. Biol. 162:729-773.

83. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequenc-ing with chain-terminating inhibitors. Proc. Natl. Acad. Sci.USA 74:5463-5467.

84. Sasaki, I., and G. Bertani. 1965. Growth abnormalities in Hfrderivatives in Eschenchia coli strain C. J. Gen. Microbiol.40:365-377.

85. Snustad, D. P. 1968. Dominance interactions in Escherichia colicells mixedly infected with bacteriophage T4D wild-type andamber mutants and their possible implications as to type ofgene-product function: catalytic vs. stoichiometric. Virology35:550-563.

86. Snyder, M., and W. B. Wood. 1989. Genetic definition of twofunctional elements in a bacteriophage T4 host-range "cas-sette." Genetics 122:471-479.

87. Steitz, J. A., and K. Jakes. 1975. How ribosomes select initiatorregions in mRNA: base pair formation between the 3'-terminusof 16S rRNA and the mRNA during initiation of protein synthe-sis in E. coli. Proc. Natl. Acad. Sci. USA 72:4734-4738.

88. Sunshine, M. G., M. Thorn, W. Gibbs, R. Calendar, and B.Kelly. 1971. P2 phage amber mutants: characterization by use ofa polarity suppressor. Virology 46:691-702.

89. Symonds, N., and A. Coelho. 1978. Role of the G segment in thegrowth of phage Mu. Nature (London) 271:573-574.

90. Temple, L. M., S. L. Forsburg, R. Calendar, and G. E. Christie.1991. Nucleotide sequence of the genes encoding the major tailsheath and tail tube proteins of bacteriophage P2. Virology181:353-358.

91. Toussaint, A., N. Lerebvre, J. Scott, J. A. Cowan, F. de Bruijn,and A. I. Bukhari. 1978. Relationships between temperatephages Mu and P1. Virology 89:146-161.

92. van de Putte, P., S. Cramer, and M. Giphart-Gassler. 1980.Invertible DNA determines host specificity of bacteriophageMu. Nature (London) 286:218-222.

93. van de Putte, P., R. Plasterk, and A. Kuipers. 1984. A Mu gincomplementing function and an invertible DNA region in Esch-erichia coli K-12 are situated on the genetic element e14. J.Bacteriol. 158:517-522.

94. Vanderslice, R. W., and C. D. Yegian. 1974. The identification oflate bacteriophage T4 proteins on sodium dodecyl sulfate poly-acrylamide gels. Virology 60:265-275.

95. Walker, J. T., and D. H. Walker, Jr. 1983. Coliphage P1morphogenesis: analysis of mutants by electron microscopy. J.Virol. 45:1118-1139.

96. Westoo, E., and E. Ljungquist. 1979. A restriction cleavage mapof bacteriophage P2. Mol. Gen. Genet. 171:91-102.

97. Yamada, M., H. Fujisawa, H. Kato, K. Hamada, and T. Mina-gawa. 1986. Cloning and sequencing of the genetic right end ofbacteriophage T3 DNA. Virology 151:350-361.

98. Yarmolinsky, M. B., and N. Sternberg. 1988. Bacteriophage P1,p. 291-438. In R. Calendar (ed.), The bacteriophages, vol. 1.Plenum Publishing Corp., New York.

VOL. 174, 1992

on Novem


.org/D

ownloaded from

http://jb.asm.org/

dna sequences fiber genes of bacteriophage p2: evidencefor ... · p2 stocks were grown as described...

Documents