isolation, characterization, and complementation of a motility … · isolation, characterization,...

JOURNAL OF BACTERIOLOGY,0021-9193/97/$04.0010

Aug. 1997, p. 4802–4810 Vol. 179, No. 15

Copyright © 1997, American Society for Microbiology

Isolation, Characterization, and Complementation of a MotilityMutant of Spiroplasma citri

CHRISTOPHE JACOB, FRANCOIS NOUZIERES, SYBILLE DURET, JOSEPH-MARIE BOVE,AND JOEL RENAUDIN*

Laboratoire de Biologie Cellulaire et Moleculaire, Institut National de la Recherche Agronomique and Universite VictorSegalen Bordeaux 2, Domaine de la Grande Ferrade, B.P. 81, 33883 Villenave d’Ornon Cedex, France

Received 3 March 1997/Accepted 21 May 1997

The helical mollicute Spiroplasma citri, when growing on low-agar medium, forms fuzzy colonies withoccasional surrounding satellite colonies due to the ability of the spiroplasmal cells to move through the agarmatrix. In liquid medium, these helical organisms flex, twist, and rotate rapidly. By using Tn4001 insertionmutagenesis, a motility mutant was isolated on the basis of its nondiffuse, sharp-edged colonies. Dark-fieldmicroscopy observations revealed that the organism flexed at a low frequency and had lost the ability to rotateabout the helix axis. In this mutant, the transposon was shown to be inserted into an open reading frameencoding a putative polypeptide of 409 amino acids for which no significant homology with known proteins wasfound. The corresponding gene, named scm1, was recovered from the wild-type strain and introduced into themotility mutant by using the S. citri oriC plasmid pBOT1 as the vector. The appearance of fuzzy colonies andthe observation that spiroplasma cells displayed rotatory and flexional movements showed the motile pheno-type to be restored in the spiroplasmal transformants. The functional complementation of the motility mutantproves the scm1 gene product to be involved in the motility mechanism of S. citri.

Spiroplasmas are helical, wall-free bacteria belonging to theclass Mollicutes, a group of microorganisms phylogeneticallyrelated to gram-positive bacteria with low G1C contents (48).Besides helical morphology, motility is the most distinctiveproperty of spiroplasmas, especially since these organisms pos-sess no flagella, no periplasmic fibrils, and no axial filaments.Motility of spiroplasmas was first described by Davis and Wor-ley (12) for the corn stunt spiroplasma in extracts of infectedplants and was confirmed by Cole and coworkers (9) withcultured cells of Spiroplasma citri, the causal agent of citrussubborn disease. Spiroplasma cells display both rotatory andflexional movements which do not produce translational mo-tility in liquid media of low viscosity. Translational motilitycould be observed only when viscosity was increased by incor-porating agar or methylcellulose in the medium (11). As aresult of the ability of spiroplasma cells to move through theagar matrix, spiroplasmas, and in particular S. citri, form dif-fuse colonies when plated on low-agar medium. Aspects ofmotility, chemotaxis, and viscotaxis in spiroplasmas have beenfurther documented by Daniels and coworkers (11). On thebasis of their work, these authors suggest that translationalmotility is driven by rotation about the helix axis, whereasflexing of the cell body causes changes of direction. Also fromthese studies, it is known that motility of spiroplasmas is energydependent. However, the mechanism of motility has not beenelucidated, and the cellular elements involved in motility arestill unknown. In order to identify such cellular components,Cohen and coworkers (8) tried to obtain nonmotile mutants ofSpiroplasma melliferum BC3. In those studies, nitrous acid wasused as the mutagen and chemotaxis was used to enrich formotility mutants. Nonmotile mutants were selected on the

basis of their smooth, fried-egg-shaped (umbonate) colonies.However, because nitrous acid treatment induces essentiallypoint mutations which cannot be easily mapped and because atthat time no gene transfer system was available to complementthe mutants, the mutated genes could not be identified. Toovercome such difficulties, gene transfer in mollicutes has beenthe subject of much research (3, 15). With S. citri, severalapproaches for the development of cloning vectors were ex-amined. The combination of the S. citri plasmid pMH1 and thecat gene was found to be quite unstable in S. citri (40). In ourlaboratory, we have used the replicative form of the S. citrivirus SpV1 as a vector to express foreign genes in S. citri R8A2(28, 33, 43). However, upon passaging, the recombinant viralDNA has proved to be unstable (29). Therefore, artificial plas-mids were constructed by combining the oriC region of the S.citri chromosome with the tetM gene (35, 52). By using theseoriC plasmids as cloning vectors, foreign genes could be intro-duced into S. citri cells, in which they were transcribed, trans-lated into proteins, and stably maintained for generations (14).

Transposons Tn916 and Tn4001, from gram-positive bacte-ria, have been transformed into several mollicutes (15), dem-onstrating the capacity for transpositional insertion. In partic-ular, Tn4001, coding for gentamicin resistance, was used togenerate cytadherence-deficient mutants of Mycoplasma pneu-moniae and Mycoplasma genitalium (23, 32). Recently, Tn4001mutagenesis has also been applied to S. citri to generate mu-tants affected in plant pathogenicity (18, 19). The advantagesof gene inactivation by insertion of a transposable elementinclude production of null alleles, minimal damage to othergenes, and easy cloning of the interrupted locus. In the presentstudies, we have used Tn4001 mutagenesis to produce motilitymutants of S. citri. One motility mutant of 938 transformantswas obtained and was characterized.

MATERIALS AND METHODS

Bacterial strains and plasmids. Escherichia coli TG1 {supE Dhsd-5 thi D(lac-proAB) F9 [traD36 proAB1 lacIq lacZDM15]}, an EcoK2 derivative of JM101,was used as the host strain for subcloning experiments and for propagation of

* Corresponding author. Mailing address: Laboratoire de BiologieCellulaire et Moleculaire, Institut National de la RechercheAgronomique and Universite Victor Segalen Bordeaux 2, Domaine dela Grande Ferrade, B.P. 81, 33883 Villenave d’Ornon Cedex, France.Phone: (33) 05.56.84.31.51. Fax: (33) 05.56.84.31.59. E-mail: [email protected].

4802

on May 28, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

plasmids and M13 bacteriophages. For transformation with plasmid DNA, com-petent E. coli cells were prepared according to the Hanahan procedure (22).Plasmid pMUT containing Tn4001 has been previously described (18). It wasused as the transposon delivery system for insertional mutagenesis of S. citri.Plasmids pCJ1 and pCJ2 (this study) were obtained from a HindIII genomicDNA library of the S. citri motility mutant. They contain the spiroplasmalsequences adjacent to the IS256 ends of Tn4001 as HindIII DNA inserts of 2.5kbp (pCJ1) and 8.5 kbp (pCJ2). Plasmid pCJ3 (this study) was obtained from anNsiI-HindIII genomic DNA library of S. citri GII-3 constructed in E. coli. Itcontains a 2-kbp NsiI-HindIII fragment inserted into the pBluescript II KS1

vector (Stratagene Cloning Systems, La Jolla, Calif.). Plasmid pCJ6 (this study)was obtained by subcloning the 1.55-kbp Sau3AI fragment of pCJ3 at the BamHIsite of plasmid pBOT1. Plasmid pBOT1 carrying the S. citri oriC region has beendescribed elsewhere (35). The S. citri GII-3 wild-type strain was originally iso-lated from its leafhopper vector, Circulifer haematoceps, captured in Morocco(47). S. citri ASP1 has been previously described as a nonhelical or poorly helicalstrain of S. citri (46). Spiroplasmas were grown at 32°C in SP4 medium (49) fromwhich fresh yeast extract was omitted.

Microscopy observation. To examine live cells, drops of exponentially growingspiroplasma cultures were placed on slides with coverslips, and the edges of thecoverslips were sealed to prevent evaporation and consequent rapid streaming ofspiroplasmas across the slide. Observations were carried out by using a dark-fieldoptical system with a 1003 oil immersion objective.

Transformation of S. citri. Electrotransformation of S. citri with the replicativeoriC plasmids pBOT1 and pCJ6 was by a method previously described (34, 43).The S. citri transformants were selected by plating on solid SP4 medium supple-mented with 2 mg of tetracycline per ml. Transformation of S. citri with thenonreplicative plasmid pMUT was by the same procedure except that approxi-mately 1010 spiroplasma cells and 50 mg of plasmid DNA per assay were used.The transformants were selected in the presence of 100 mg of gentamicin per ml.

DNA isolation. Large-scale and small-scale preparations of plasmids and M13DNA amplified in E. coli were carried out according to standard procedures (37).Isolation of spiroplasmal genomic DNA has been described elsewhere (51).

Construction and screening of the S. citri genomic DNA libraries. GenomicDNA of the S. citri motility mutant G540 was digested with HindIII, and thefragments were ligated to the HindIII-linearized pBS1 vector (Stratagene Clon-ing Systems). The recombinant clones containing sequences adjacent to theIS256 ends of Tn4001 were selected by in situ hybridization of colonies with the32P-labeled IS256 probe according to standard procedures (37). The IS256 probewas generated by PCR amplification of the gel-purified 1.3-kbp XbaI-HindIIIfragment of plasmid pMUT with primer pair IS1-IS3. Genomic DNA of thewild-type strain, S. citri GII-3, was digested to completion with endonucleasesNsiI and HindIII, and the restriction fragments were ligated to the pBluescript IIKS1 vector digested with PstI and HindIII. The bacterial clones containing S. citriopen reading frame 2 (ORF 2) were selected by in situ hybridization of colonieswith the 32P-labeled CJ6-CJ7 probe. The CJ6-CJ7 probe was obtained by PCRamplification of the gel-purified 2.5-kbp HindIII fragment of plasmid pCJ1 withprimer pair CJ6-CJ7.

Primers and PCR amplification. Primers IS1 (59-ATATTCTGTAAAGGATGACG-39), IS3 (59-CAGAGAAGCTGTTAAAG-39), and IS4 (59-ATGTAAAAGTCCTCCTGGG-39) correspond, respectively, to nucleotides 155 to 174, 923to 939, and 85 to 103 of the nucleotide sequence of IS256 (4). Primers CJ2(59-AAGTTTCTAATGGATTAGGG-39), CJ6 (59-CAATTACCAACCATGTTAGC-39), CJ7 (59-TTAATTGCAGTTGGAACAGC-39), CJ8 (59-GGTTAGTAATGCTGATCGC-39), and CJ17 (59-CTTTACAGGGAGATAGTGC-39) cor-respond, respectively, to nucleotides 2661 to 2680, 3576 to 3595, 2818 to 2837,2328 to 2346, and 2678 to 2696 of the sequence reported in this paper. PrimerEV4 (59-GATAGCACATTATTTTCCACG-39) corresponds to nucleotides 1601to 1622 of the nucleotide sequence of the S. citri oriC region (52). Amplificationwas carried out in a 50-ml reaction mixture containing 1 to 10 ng of target DNA(or approximately 105 spiroplasma cells), 50 mM Tris-HCl (pH 8.8), 2 mMMgCl2, 200 mg of bovine serum albumin per ml, 0.05% W1 detergent, 0.2 mMdeoxynucleoside triphosphates, 1 mM each primer, and 2.5 U of Taq DNApolymerase (GIBCO/BRL Life Technologies, Inc., Gaithersburg, Md.). The re-action was performed in a thermal cycler (Perkin-Elmer Cetus Corp., Norwalk,Conn.) as described previously (29) except that the annealing temperature wasset to 57°C for primer pair IS1-IS3 and 56°C for primer pairs CJ6-CJ7, CJ6-CJ17,IS4-CJ8, and EV4-CJ2.

Southern blot hybridization. Restricted DNA was blotted to positivelycharged nylon membranes by the alkali transfer procedure and hybridized withthe 32P-labeled probe under standard stringent conditions (37). The DNA probeswere labeled by the random priming procedure (16, 17) with [a-32P]dATP (110TBq/mmol) as the labeled nucleotide.

RNA isolation and Northern blot hybridization. Total RNA from S. citri cellswas isolated by the guanidinium thiocyanate-cesium chloride method (6) andanalyzed by Northern blot hybridization according to the procedure outlined byStamburski et al. (42).

DNA sequencing and nucleotide sequence analyses. Double-stranded or sin-gle-stranded DNA was sequenced by the dideoxy chain termination technique(38) with the T7 sequencing kit from Pharmacia Biotech (Uppsala, Sweden).Nucleotide and amino sequences used for comparison were imported fromGenBank (Los Alamos, N.Mex.) with the retrieve electronic mail server of the

National Center for Biotechnology Information at the National Library of Med-icine, National Institutes of Health, Bethesda, Md. Sequence analyses wereperformed with the Seqaid II codon bias program (41), Blast program (1), andGenetics Computer Group software package (13).

Nucleotide sequence accession number. The nucleotide sequence data re-ported in this paper will appear in the GenBank nucleotide sequence databaseswith the accession number U89875.

RESULTS

Construction and screening of gentamicin-resistant inser-tional mutants of S. citri GII-3. S. citri GII-3 cells were elec-trotransformed with the suicide plasmid pMUT containing theTn4001 transposon, and the transformants were selected byplating the transformation mixture on 1% agar SP4 mediumcontaining 100 mg of gentamicin per ml. Among gentamicin-resistant transformants, motility mutants were screened on thebasis of their nondiffuse, nonsatellited colonies. FollowingTn4001 mutagenesis, two mutants with dense colonies, of atotal of 938 transformants, were selected. The two S. citrimutants were named G540 and G547.

Colony morphology of S. citri mutants G540 and G547. Theshape of spiroplasmal colonies is known to be dependent onthe agar content of the medium. As the concentration of agaris lowered, the colonies formed by the wild-type cells becomemore fuzzy. For that reason, the two S. citri mutants, G540 andG547, were plated on various agar concentrations (0.4, 0.6, 0.8,1, and 1.2%), and the colony morphology of these mutants wascompared to that of the wild-type strain GII-3 as well as to thatof strain ASP1, which was previously described as a naturalnonhelical nonmotile mutant of S. citri. The photographs inFig. 1 show colonies of GII-3 (a and b), G540 (c and d), ASP1(e and f), and G547 (g and h) grown on 0.6% (a, c, e, and g) or1% (b, d, f, and h) agar plates. When the concentration of agarwas lowered from 1 to 0.6%, the colonies formed by the GII-3cells became more diffuse, and the rounded centers of thecolonies were no longer visible (Fig. 1a and b). In contrast tothe wild-type strain, which formed fuzzy colonies, the G540mutant (Fig. 1c and d), like strain ASP1 (Fig. 1e and f), formedfried-egg-shaped colonies with no surrounding satellite colo-nies, even at the low agar concentration. Figure 1 also showsthat the colonies formed by G547 (g and h) differ from those ofGII-3 (a and b). However, the presence of large numbers ofsatellite colonies, particularly when plated on 0.6% agar, sug-gested that this mutant, in contrast to mutant G540, is still ableto move, especially in low-agar medium. When cultured inbroth medium containing gentamicin, the doubling time ofG540 was similar to that of GII-3 in gentamicin-free SP4 me-dium (approximately 6 h), and it reached about the same titer(108 to 109 CFU/ml). In contrast, G547 grew at a lower rate,with a doubling time of about 9 h, and reached a 10-fold-lowertiter than the wild-type strain.

Motility behavior of S. citri mutants G540 and G547. Inliquid media, spiroplasma cells, and in particular S. citri GII-3,display flexing and twisting of the cell body and rotation aboutthe helix axis. This screw motion is responsible for translationalmovement of spiroplasmas in viscous media. As previouslydescribed (46), cells of the nonhelical S. citri strain ASP1 pro-duced erratic twitching movements only. Examination of thetwo insertional mutants G540 and G547 revealed that, in con-trast to ASP1, both had retained their helical morphology.However, while G547 seemed to display the same motilitymovements as the wild-type strain, the frequencies of flexingand twisting of G540 were considerably reduced. In addition,this mutant apparently did not rotate about the helix axis. Thiswas further confirmed when the medium viscosity was in-creased by adding 0.1% gum xanthan. Under these conditions,

VOL. 179, 1997 SPIROPLASMA CITRI MOTILITY MUTANT 4803


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

the G540 cells were unable to exhibit translational motility, incontrast to the GII-3 cells, which swam through the solutionwith a screw motion. In the case of G540, the absence ofrotating activity, and hence of translational motility in viscousmedium, would explain the formation of dense colonies com-pared to the fuzzy colonies formed by the wild-type strainGII-3.

Site of Tn4001 insertion in mutant G540. Mutant G540,showing motility defects, was chosen for further study of themolecular events associated with the mutagenesis procedure.To determine the site where the transposon inserted into thechromosome, genomic DNA of G540 was digested with variousrestriction endonucleases and hybridized with two differentprobes. One probe represented the pMUT plasmid used forthe transformation of S. citri GII-3, and the other was the785-bp IS1-IS3 PCR fragment of IS256. Hybridizations withthe pMUT probe revealed that a single copy of Tn4001, lo-cated within an EcoRI fragment of approximately 15 kbp, waspresent in the genome of G540 (data not shown). In order tolocate spiroplasmal sequences adjacent to the transposon, re-striction fragments were also hybridized with the IS256-specificprobe (Fig. 2A). The results showed that the probe hybridizedwith two HindIII fragments with sizes of 8.5 and 2.5 kbp (Fig.2A, lane 1). In agreement with the presence of an HpaI siteand an ScaI site within the probe, restriction of genomic DNAwith these enzymes yielded four fragments, with sizes of 10.5,3.5, 2.3, and 1.3 kbp for HpaI (Fig. 2A, lane 6) and larger than13, 1.8, 1.5, and 0.9 kbp for ScaI (lane 11), all hybridizing with

the IS256 probe. Using combinations of single and doubledigestions of the G540 DNA with the enzymes HindIII, HincII,Sau3AI, HpaI, ScaI, and AluI, we established the restrictionmap of the chromosomal region where the insertion of Tn4001has occurred (Fig. 2B). According to this map, the spiroplas-mal sequences adjacent to the transposon were located withinthe 8.5-kbp HindIII fragment on the left end of the transposonand within the 2.5-kbp HindIII fragment on the right end of thetransposon. These two fragments were recovered from agenomic DNA library of G540. The nucleotide sequences ofthe 3.5-kbp HpaI fragment, obtained by subcloning the 8.5-kbpHindIII fragment, and of the 2.5-kbp HindIII fragment weredetermined. Sequence analyses revealed the presence of twodirectly repeated sequences of 8 nucleotides bordering thetransposon. These sequences, TATTTTTA, very probably rep-resent duplications of the target DNA at the transposon inser-tion site in the chromosome of G540. Indeed, sequencing ofthis region in the wild-type strain GII-3 showed that only onecopy of this sequence was present in the chromosomal DNA ofthe wild-type strain (see Fig. 4, nucleotides 2808 to 2815).

FIG. 1. Colonies of S. citri grown on 0.6 or 1% agar SP4 medium. Photo-graphs were taken 7 days after plating. (a and b) S. citri GII-3; (c and d) S. citrimutant G540; (e and f) S. citri ASP1; (g and h) S. citri mutant G547. Magnifi-cation, 315. FIG. 2. (A) Southern blot hybridization of genomic DNA of the S. citri

mutant G540 with the 32P-labeled IS1-IS3 probe. Lanes 1 to 12, DNA restrictedwith HindIII (lanes 1 and 8), HindIII plus HincII (lane 2), HincII (lane 3), HincIIplus Sau3AI (lane 4), HpaI plus Sau3AI (lane 5), HpaI (lane 6), HpaI plusHindIII (lane 7), HindIII plus Sau3AI (lane 9), Sau3AI (lane 10), ScaI (lane 11),and AluI (lane 12). Lane 13, plasmid pMUT restricted with HindIII. (B) Re-striction map of the chromosomal DNA of S. citri mutant G540 at the transposoninsertion site. H, HindIII; Hp, HpaI; N, NsiI; S, Sau3AI; Sc, ScaI. The blackrectangles at the top indicate the positions of the IS1-IS3 probe, and the openrectangle indicates the position of the CJ6-CJ7 probe. Gmr, gentamicin resis-tance gene. Sizes in both panels are indicated in kilobase pairs.

4804 JACOB ET AL. J. BACTERIOL.


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

Sequence analysis of the region around the transposon in-sertion site. To characterize the region around the transposoninsertion site, the restricted genomic DNA of GII-3 was firsthybridized with the CJ6-CJ7 probe, located immediately down-stream of the insertion site of the transposon. The hybridiza-tion patterns presented in Fig. 3 show that the probe hybrid-ized with an EcoRI fragment of approximately 14 kbp (lane 1),a HincII fragment of 5.2 kbp (lane 2), an EcoRV fragment of8.6 kbp (lane 5), a HindIII fragment of 8.1 kbp (lane 7), and anNsiI fragment of 5.8 kbp (lane 9). As expected from the re-striction map in Fig. 2B, the probe also hybridized with anNsiI-HindIII fragment of 2 kbp (Fig. 3, lane 8). This fragment,encompassing the target site of the transposon, was recoveredfrom a genomic DNA library of GII-3, and its nucleotidesequence was determined. Combining the nucleotide sequenceof this fragment with those of the 3.5-kbp HpaI and 2.5-kbpHindIII fragments of G540 (Fig. 2B) led to a sequence of 4242nucleotides around the insertion site of the transposon.

Taking into account the fact that, in spiroplasmas, UGAcodes for tryptophan (7, 36), we found this sequence to containthree ORFs (ORFs 1, 2, and 3), each starting at an ATGinitiation codon located downstream of a possible ribosomebinding site. The nucleotide sequence and the deduced aminoacid sequences of the putative translation products are pre-sented in Fig. 4. The first ORF (ORF 1), starting at nucleotide166 and finishing at nucleotide 2412, encodes a 749-amino-acidpolypeptide which has significant homology with stringent fac-tors involved in the synthesis of the phosphorylated nucleo-tides ppGpp and pppGpp. In particular, it has 37.5% identityand 61.1% similarity with the (p)ppGpp 39-pyrophosphohydro-lase of Streptococcus equisimilis (30), a gram-positive bacte-rium with a low G1C content, and 34.6% identity and 58.5%similarity with the E. coli (p)ppGpp 39-pyrophosphohydrolaseencoded by the spoT gene (39). ORF 3 (nucleotides 3834 to4242) codes for a putative polypeptide which has striking ho-mology (77.2% identity and 84.6% similarity in a 136-amino-acid overlap) with the alanyl-tRNA synthetase of Mycoplasmacapricolum (2). The target sequence, TATTTTTA, was locatedwithin ORF 2. This ORF (nucleotides 2509 to 3735) encodes aputative polypeptide of 409 amino acids for which no signifi-cant homology with known proteins was found. The predictedpolypeptide is essentially hydrophobic and possesses an N-terminal sequence showing common features with signal pep-tides of eubacteria (25). The putative signal peptide of the

ORF 2 product comprises an N-terminal region of 8 aminoacids positively charged by the presence of two lysine residues,a hydrophobic core of 14 amino acids, and a C-terminal regionwith a polar glycine residue and two alanine at positions 21and 23 from the putative cleavage site (Fig. 5). Upstream ofORF 2, the two sequences TATAAT (nucleotides 2480 to2485) and TTGTTT (nucleotides 2457 to 2462) are reminis-cent of the 210 and 235 consensus sequences of eubacterialtranscription promoters functioning with the general sigmafactor, sA in Bacillus subtilis or s70 in E. coli. In addition,downstream of ORF 2, the presence of inverted repeat se-quences (nucleotides 3744 to 3760 and 3764 to 3780) followedby a stretch of uridylic residues (in the mRNA) strongly sug-gests a rho-independent terminator. According to these data,ORF 2 would be transcribed as a monocistronic mRNA ofapproximately 1.3 kb.

Expression of ORF 2 was monitored by Northern blot hy-bridization of total RNAs extracted from GII-3 and G540 withthe CJ6-CJ7 probe. The hybridization patterns showed that inthe case of GII-3, the probe did hybridize with a 1.3-kb mRNAand that no such mRNA could be detected in the motilitymutant G540 (data not shown). These results show that, asexpected, the Tn4001 insertion abolished transcription of ORF2 in G540. Moreover, the finding of a 1.3-kb mRNA in the caseof GII-3 suggests that ORF 2 is not part of an operon butinstead is transcribed as a monocistronic mRNA.

Transformation of G540 with the wild-type ORF 2. To en-sure that disruption of ORF 2 was indeed responsible for themotility defects of G540, we attempted to restore the motilephenotype by transformation of the mutant cells with the wild-type ORF 2. To that purpose, the wild-type ORF 2 was recov-ered from plasmid pCJ3 as a 1.55-kbp Sau3AI fragment (Fig.4, nucleotides 2341 to 3893) and was inserted at the BamHIsite of the oriC plasmid pBOT1. The recombinant plasmid,named pCJ6, was used to transform the motility mutant G540,and the transformants were selected by plating on SP4 agarmedium containing 2 mg of tetracycline per ml. As a control,G540 cells were transformed with the ORF 2-free pBOT1plasmid. Interestingly, G540 cells transformed with plasmidpCJ6 formed dispersed colonies with satellite colonies (Fig.6a), whereas cells transformed with the pBOT1 vector stillformed dense, unsatellited colonies (Fig. 6b). It should benoted that, due to expression of tetracycline resistance, colo-nies formed by G540 cells showed irregular outlines. Similarcolonies have been previously observed when S. citri ASP1 cellswere transformed to tetracycline resistance with plasmidpBOT1 (35). However, treatment with Dienes stain clearlyshowed that only the G540 cells transformed with the pCJ6plasmid carrying ORF 2 formed diffuse colonies (cf. Fig. 6c andd). These observations suggest that the cells transformed withORF 2, like the wild-type spiroplasmas, were capable of trans-lational movement in agar medium. Indeed, when grown inliquid medium, they displayed rotational activity, as seen bydark-field microscopy, while those transformed with the vectoralone did not.

In order to exclude the possibility that the motile spiroplas-mal transformants could be revertants having lost the transpo-son, PCR amplifications were performed to determine whetherthese transformants still carried the transposon within ORF 2and whether they did contain the pCJ6 plasmid. Broth culturesof the wild-type strain GII-3, the motility mutant G540, G540transformed with plasmid pBOT1, and G540 transformed withthe pCJ6 plasmid containing the wild-type ORF 2 were sub-jected to amplification with four different pairs of primers,primer pairs CJ6-CJ7, IS4-CJ8, CJ6-CJ17, and EV4-CJ2 (Fig.7). As primers CJ6 and CJ7 are both located downstream of

FIG. 3. Southern blot hybridization between genomic DNA of S. citri GII-3and the 32P-labeled CJ6-CJ7 probe. The DNA was restricted with EcoRI (lane1), HincII (lane 2), Sau3AI (lane 3), MboI (lane 4), EcoRV (lane 5), HindIII plusEcoRV (lane 6), HindIII (lane 7), HindIII plus NsiI (lane 8), NsiI (lane 9),EcoRV plus NsiI (lane 10), and AluI (lane 11). Sizes are indicated in kilobasepairs.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

FIG. 4. Nucleotide sequence of the 4,242-bp HpaI-HindIII fragment of S. citri GII-3 and derived amino acid sequences of the putative polypeptides. EcoRV,HindIII, HpaI, NsiI, Sau3AI, and ScaI restriction sites, as well as the starts of ORFs 1, 2, and 3 and the ribosome binding sites (RBS), are indicated. ,235., and,210., 235 and 210 regions of a putative transcription promoter. Primers CJ2, CJ6, CJ7, CJ8, and CJ17 are indicated with arrows. The transposon target siteTATTTTTA is underlined.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

the insertion site of the transposon (Fig. 7B), they allowedamplification of DNA from all of the cultures tested (Fig. 7A,panel a, lanes 2 to 6). Figure 7A, panel b, shows that amplifi-cation with primer pair IS4-CJ8 also gave positive results for allof the G540 cultures tested (lanes 3 to 6) but not for thewild-type strain GII-3 (lane 2). Since primers IS4 and CJ8 arelocated, respectively, within IS256 of Tn4001 and upstream ofORF 2, these results indicated that the G540 cells transformedwith plasmid pCJ6 (Fig. 7A, panel b, lanes 5 and 6) still con-tained the transposon within ORF 2. This was confirmed by thefact that DNA from GII-3 could be amplified with primer pair

CJ6 and CJ17, located, respectively downstream and upstreamof the transposon target site (Fig. 7A, panel c, lane 2), whereasDNAs from G540 and G540 transformed with pBOT1 couldnot be (lanes 3 and 4). The presence of the pCJ6 plasmid inG540 cells transformed with this plasmid was revealed by am-plification with primer pair CJ6-CJ17 (Fig. 7A, panel c, lanes 5and 6), as well as with primer pair EV4-CJ2 (Fig. 7A, panel d,lanes 5 and 6). In the latter case, no amplification could bedetected, either with GII-3 (lane 2), with G540 (lane 3), or withG540 transformed with pBOT1 (lane 4). These data clearlyindicate that, in G540 cells, recovery of the motile phenotypewas due to transformation by the wild-type ORF 2 present inplasmid pCJ6 and not to excision of the transposon.

DISCUSSION

Motility and helical morphology are the most distinctiveproperties of spiroplasmas, especially since these organismspossess no flagella or axial filaments. Spiroplasmas are capableof translational motility in viscous media, and they displaychemotaxis (10, 11). Motility movements of spiroplasmas, ro-

FIG. 5. Amino acid sequence of the N terminus of the ORF 2-encodedpolypeptide.

FIG. 4—Continued.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

tation about the helix axis, and flexing of the cell body weredescribed as early as 1973 (12). However, little is known aboutthe motility mechanism. The molecular components that per-mit spiroplasma cells to maintain a helical shape and to displayrotatory and flexional motility have not been identified. Spiro-plasmas have been shown to possess fibrils, which were thoughtto be involved in the mechanism of spiroplasma motility (45,50). However, even though the fibrillar protein has been char-acterized (44, 45) and its gene has been sequenced (51), thereis presently no experimental evidence to ascertain the role ofthis protein in the motility of spiroplasmas. The eventual un-derstanding of the structural and energy transduction compo-nents of spiroplasmal motility will depend on the availability ofrelevant mutants.

In our studies, we have used Tn4001 mutagenesis to producemotility mutants of S. citri GII-3. One mutant, G540, wasselected on the basis of its inability to move through the agarmatrix. When plated on solid medium, G540 formed smooth,unsatellited colonies resembling those formed by S. citri ASP1,a nonhelical and nonmotile natural variant of S. citri (46).However, unlike ASP1 cells, G540 cells had retained theirhelical morphology. In liquid medium, the growth rate of G540was similar to that of the wild-type strain GII-3. However,G540 cells displayed flexing of the cell body at a low frequencyand were devoid of rotational movement about the helix axis.Interestingly, G540 has remained phenotypically stablethroughout multiple passages in broth medium. This observa-tion correlates well with the finding that the transposon wasstably maintained in the chromosome for at least 100 genera-tions, even in the absence of gentamicin selection pressure(data not shown). The stability of Tn4001 insertions in S. citriGII-3 grown in SP4 medium has been previously reported fortwo of the three spiroplasmal transformants tested (18). In thenonphytopathogenic mutant GMT553 also, the transposon wasfound to be stably maintained when the mutant was propa-gated in vitro (21). However, when GMT553 was transmittedby leafhoppers to periwinkle plants and allowed to multiply inthe plants, revertants having lost the transposon could be de-tected in the plants within 4 to 5 weeks after insect transmis-sion (19). These data suggest that stability of the Tn4001 in-sertions in the spiroplasmal mutants could depend not only on

the insertion site of the transposon but also on the selectionpressure imposed on the mutant by the environmental condi-tions. Accordingly, the stable maintenance of Tn4001 in mu-tant G540, i.e., absence of reversion, might indicate that inliquid medium, motile spiroplasmas have no advantages overnonmotile cells. However, the possibility that, in S. citri,Tn4001 excises quite rarely and thus that the lack of findingrevertants of G540 might be unrelated to selection cannot beruled out.

In the case of G540, the finding of a motility-deficient spi-roplasma that has retained its helical morphology indicatesthat helical morphology and motility are genetically indepen-dent. Similar conclusions have been reached previously for thenitrous acid-generated mot1 mutant of S. melliferum (8).

Southern blot hybridizations of genomic DNA with Tn4001-specific probes showed that a single copy of the transposon waspresent in the chromosome of G540. In order to identify themutated DNA sequence that confers the nonmotile phenotypeto mutant G540, we have determined the nucleotide sequenceof chromosomal DNA flanking the site of transposon insertion.The target site (TATTTTTA) was identified as directly re-peated sequences bordering the transposon. In the previouslydescribed S. citri mutants GMT470 and GMT553 (19), thetarget sites were also found to be 8 nucleotides long. However,in agreement with the low insertion specificity of Tn4001, thesetarget sites have different sequences (21, 31). Similar data havebeen recently reported for Tn4001 insertional mutants of M.pneumoniae (27).

In the motility mutant G540, we found the target site to belocated within an ORF (ORF 2) encoding a polypeptide of 409

FIG. 6. Colonies of S. citri grown in 1% agar SP4 medium before (a and b)and after (c and d) staining. Staining was performed by overlaying the plates with5 ml of a 3% Dienes stain solution. After a 5- to 10-min staining period, theplates were extensively washed with water before being photographed. (a and c)Motility mutant G540; (b and c) G540 transformed with plasmid pCJ6. In panela, the arrows indicate satellite colonies. Magnification, 320.

FIG. 7. (A) Ethidium bromide-stained agarose gel of DNAs amplified withprimer pairs CJ6-CJ7 (a), IS4-CJ8 (b), CJ6-CJ17 (c), and EV4-CJ2 (d). Lanes: 1,no template DNA; 2, DNA of S. citri GII-3; 3, DNA of motility mutant G540; 4,DNA of G540 transformed with plasmid pBOT1; 5 and 6, DNAs from twoindividual clones of G540 transformed with plasmid pCJ6. Sizes are indicated inbase pairs. (B) Locations of primers CJ2, CJ6, CJ7, CJ8, CJ17, IS4, and EV4 inthe chromosomal DNAs of GII-3 and G540 and in plasmids pBOT1 and pCJ6.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

amino acids. The predicted polypeptide is mainly hydrophobicand possesses an N-terminal sequence characteristic of eubac-terial signal peptides (25). These data suggest that the trans-lation product of ORF 2, which is disrupted by the presence ofthe transposon in the motility mutant G540, might be an exo-protein having a role in protein secretion. However, searchingfor possible homologies of this polypeptide with previouslyreported proteins gave negative results. In particular, no ho-mology was found with any of the proteins of the mollicutes M.genitalium and M. pneumoniae, whose complete genome se-quences have been determined (20, 24). These results are inagreement with the fact that mycoplasmas and spiroplasmasdisplay different types of motility (26), gliding for mycoplasmasand rotation and undulation for spiroplasmas. They suggest theORF 2-encoded protein to be specific to spiroplasmas. Se-quencing of the region around ORF 2 led to identification oftwo additional ORFs, ORF 1 and ORF 3. ORF 1 is upstreamof ORF 2 and encodes a polypeptide which has significanthomology with stringent factors encoded by the relA and spoTgenes. In the stringent response, these proteins are involved inthe metabolism of the phosphorylated nucleotides ppGpp andpppGpp (5). ORF 3 is downstream of ORF 2 and is undoubt-edly the gene encoding alanyl-tRNA synthetase.

To confirm that the nonmotile phenotype of G540 was dueto the insertion of the transposon into ORF 2, we tried toachieve complementation of the mutant with a wild-type DNAfragment containing ORF 2. The 1.55-kbp Sau3AI fragmentwas chosen according to Northern blot hybridization datashowing that ORF 2 was transcribed as a monocistronicmRNA of 1.3 kb. In previous studies, we have reported theusefulness of the oriC plasmid pBOT1 for cloning and expres-sion of foreign genes in S. citri (14, 34, 35). Therefore, we haveused this plasmid as the vector to introduce the wild-type ORF2 into the motility mutant G540. Expression of the wild-typeORF 2 in the G540 transformants was revealed by the occur-rence of fuzzy, satellited colonies, suggesting that the motilephenotype was restored in these transformants. Indeed, thespiroplasma cells grown from these colonies in liquid mediumwere found to display rotation about the helix axis. In addition,we were able to show that these cells, in contrast to the un-transformed G540, did carry the 1.3-kb ORF 2 mRNA. Thepresence of a single copy of Tn4001 in the G540 chromosomeand the functional complementation of this mutant with thewild-type ORF 2 indicates that disruption of ORF 2 is respon-sible for the motility defect of G540. Hence, for the first time,a gene involved in the motility of S. citri has been identified.We propose to name this gene corresponding to ORF 2 scm1,for S. citri motility gene 1. Sequences homologous to the scm1gene have been detected in all S. citri strains tested, includingthe nonhelical, nonmotile strain ASP1 (data not shown). How-ever, transformation of S. citri ASP1 with plasmid pCJ6 failedto restore cell motility, indicating that in this natural variant,the nonmotile phenotype is not due to a mutation in the scm1gene.

ACKNOWLEDGMENTS

This investigation was supported by the Ministere de l’Enseigne-ment Superieur et de la Recherche (ACC-SV6 no. 9506077) and by theConseil Regional d’Aquitaine (CCRDT no. 950307005).

We are grateful to X. Foissac for providing the plasmid pMUT.

REFERENCES

1. Altshul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990.Basic local aligment search tool. J. Mol. Biol. 215:403–410.

2. Bork, P., C. Ouzounis, G. Casari, R. Schneider, C. Sander, M. Dolan, W.Gilbert, and P. M. Gillevet. 1995. Exploring the Mycoplasma capricolumgenome: a minimal cell reveals its physiology. Mol. Microbiol. 16:955–967.

3. Bove, J. M. 1993. Molecular features of mollicutes. Clin. Infect. Dis.17(Suppl 1):S10–S31.

4. Byrne, M. E., D. A. Rouch, and R. A. Skurray. 1989. Nucleotide sequenceanalysis of IS256 from the Staphylococcus aureus gentamycin-tobramycin-kanamycin-resistance transposon Tn4001. Gene 81:361–367.

5. Cashel, M., D. R. Gentry, V. J. Hernandez, and D. Vinella. 1996. Thestringent response, p. 1458–1496. In F. C. Neidhart, R. Curtiss III, J. L.Ingraham, E. C. C. Lin, K. B. Low, B. Magasanik, W. S. Reznikoff, M. Riley,M. Schaechter, and M. E. Umbarger (ed.), Escherichia coli and Salmonella:cellular and molecular biology, 2nd ed. ASM Press, Washington, D.C.

6. Chirgwin, J. M., A. E. Przybyla, R. J. MacDonald, and W. J. Rutter. 1979.Isolation of biologically active ribonucleic acid from sources enriched inribonuclease. Biochemistry 18:5294–5299.

7. Citti, C., L. Marechal-Drouard, C. Saillard, J. H. Weil, and J. M. Bove. 1992.Spiroplasma citri UGG and UGA tryptophan codons: sequence of the twotryptophanyl-tRNAs and organization of the corresponding genes. J. Bacte-riol. 174:6471–6478.

8. Cohen, A. J., D. L. Williamson, and P. R. Brink. 1989. A motility mutant ofSpiroplasma melliferum induced with nitrous acid. Curr. Microbiol. 18:219–222.

9. Cole, R. M., J. G. Tully, T. G. Popkin, and J. M. Bove. 1973. Morphology,ultrastructure and bacteriophage infection of the helical mycoplasma-likeorganism (Spiroplasma citri gen. nov., sp. nov.) cultured from stubborn dis-ease of citrus. J. Bacteriol. 115:367–386.

10. Daniels, M. J., and J. M. Longland. 1984. Chemotactic behavior of spiro-plasmas. Curr. Microbiol. 10:191–194.

11. Daniels, M. J., J. M. Longland, and J. Gilbart. 1980. Aspects of motility andchemotaxis in spiroplasmas. J. Gen. Microbiol. 118:429–436.

12. Davis, R. E., and J. F. Worley. 1973. Spiroplasma: motile, helical microor-ganism associated with corn stunt disease. Phytopathology 63:403–408.

13. Devereux, J., P. Haeberli, and O. Smithies. 1984. A comprehensive set ofsequence analysis programs for the VAX. Nucleic Acids Res. 12:387–395.

14. Duret, S., J. M. Bove, and J. Renaudin. 1996. Spiroplasma citri oriC plasmids.Expression of heterologous spiralins in Spiroplasma citri. IOM Lett. 4:351.

15. Dybvig, K., and L. L. Voelker. 1996. Molecular biology of mycoplasmas.Annu. Rev. Microbiol. 50:25–57.

16. Feinberg, A. P., and B. Vogelstein. 1983. A technique for radiolabelling DNArestriction endonuclease fragments to high specific activity. Anal. Biochem.132:6–13.

17. Feinberg, A. P., and B. Vogelstein. 1984. A technique for radiolabelling DNArestriction endonuclease fragments to high specific activity. Addendum.Anal. Biochem. 137:266–267.

18. Foissac, X., C. Saillard, and J. M. Bove. 1997. Random insertion of Tn4001in the genome of Spiroplasma citri strain GII3. Plasmid 37:80–86.

19. Foissac, X., C. Saillard, J. L. Danet, P. Gaurivaud, C. Pare, F. Laigret, andJ. M. Bove. 1997. Mutagenesis by insertion of transposon Tn4001 into thegenome of Spiroplasma citri: characterization of mutants affected in plantpathogenicity and transmission to the plant by the leafhopper vector Circu-lifer haematoceps. Mol. Plant-Microbe Interact. 10:454–461.

20. Fraser, C. M., J. D. Gocayne, O. White, M. D. Adams, R. A. Clayton, R. D.Fleishmana, C. J. Bult, A. R. Kerlavage, G. Sutton, J. M. Kelley, J. L.Fritchman, J. F. Weidman, K. U. Small, M. Sandusky, J. Fuhrmann, D.Nguyen, T. R. Utterback, D. M. Sandek, C. A. Phillips, J. M. Merrick, J. F.Tomb, B. A. Dougherty, K. F. Bott, P. C. Hu, T. S. Lucier, S. N. Peterson,H. O. Smith, C. A. Hutchison III, and J. C. Veuter. 1995. The minimal genecomplement of Mycoplasma genitalium. Science 270:397–403.

21. Gaurivaud, P., and F. Laigret. Unpublished data.22. Hanahan, D. 1983. Studies on transformation of Escherichia coli with plas-

mids. J. Mol. Biol. 166:557–580.23. Hedreyda, C. T., and D. C. Krause. 1995. Identification of a possible cytad-

herence regulatory locus in Mycoplasma pneumoniae. Infect. Immun. 63:3479–3483.

24. Himmelreich, R., H. Hilbert, H. Plagens, E. Pirki, B. Li, and R. Hermann.1996. Complete sequence analysis of the genome of the bacterium Myco-plasma pneumoniae. Nucleic Acids Res. 24:4420–4449.

25. Izard, J. W., and D. A. Kendall. 1994. Signal peptides: exquisitely designedtransport promoters. Mol. Microbiol. 13:765–773.

26. Kirchhoff, H. 1992. Motility, p. 289–306. In J. Maniloff, R. N. McElhaney,L. R. Finch, and J. B. Baseman (ed.), Mycoplasmas: molecular biology andpathogenesis. American Society for Microbiology, Washington, D.C.

27. Krause, D. C., T. Proft, C. T. Hedreyda, H. Hilbert, H. Plagens, and R.Hermann. 1997. Transposon mutagenesis reinforces the correlation betweenMycoplasma pneumoniae cytoskeletal protein HMW2 and cytadherence. J.Bacteriol. 179:2668–2677.

28. Marais, A., J. M. Bove, S. F. Dallo, J. B. Baseman, and J. Renaudin. 1993.Expression in Spiroplasma citri of an epitope carried on the G fragment ofcytadhesin P1 gene from Mycoplasma pneumoniae. J. Bacteriol. 175:2783–2787.

29. Marais, A., J. M. Bove, and J. Renaudin. 1996. Spiroplasma citri virus SpV1-derived cloning vector: deletion formation by illegitimate and homologousrecombination in a spiroplasmal host strain which probably lacks a functionalrecA gene. J. Bacteriol. 178:862–870.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

30. Mechold, U., K. Steiner, S. Vettermann, and H. Malke. 1993. Genetic orga-nization of the streptokinase region of the Streptococcus equisimilis H46Achromosome. Mol. Gen. Genet. 241:129–140.

31. Pare, C., and C. Saillard. Unpublished data.32. Reddy, S. K., W. G. Rasmussen, and J. B. Baseman. 1996. Isolation and

characterization of transposon Tn4001-generated, cytadherence-deficienttransformants of Mycoplasma pneumoniae and Mycoplasma genitalium.FEMS Immunol. Med. Microbiol. 15:199–211.

33. Renaudin, J., and J. M. Bove. 1994. SpV1 and SpV4, spiroplasma viruseswith circular, single-stranded DNA genomes, and their contribution to themolecular biology of spiroplasmas. Adv. Virus Res. 44:429–463.

34. Renaudin, J., and J. M. Bove. 1995. Plasmid and viral vectors for genecloning and expression in Spiroplasma citri, p. 167–178. In S. Razin and J. G.Tully (ed.), Molecular and diagnostic procedures in mycoplasmology. Aca-demic Press, San Diego, Calif.

35. Renaudin, J., A. Marais, E. Verdin, S. Duret, X. Foissac, F. Laigret, andJ. M. Bove. 1995. Integrative and free Spiroplasma citri oriC plasmids: ex-pression of the Spiroplasma phoeniceum spiralin in Spiroplasma citri. J. Bac-teriol. 177:2800–2877.

36. Renaudin, J., M. C. Pascarel, C. Saillard, C. Chevalier, and J. M. Bove. 1986.Chez les spiroplasmes le codon UGA n’est pas non-sens et semble coderpour le tryptophane. C. R. Acad. Sci. Ser. III 303:539–540.

37. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: alaboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold SpringHarbor, N.Y.

38. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequencing withchain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463–5467.

39. Sarubbi, E., K. E. Rudd, H. Xiao, K. Ikehara, M. Kalman, and M. Cashel.1989. Characterization of the spoT gene of Escherichia coli. J. Biol. Chem.264:15074–15082.

40. Simoneau, P., and J. Labarere. 1990. Construction of chimeric antibioticresistance determinants and their use in the development of cloning vectorsfor spiroplasmas, p. 66–74. In G. Stanek, G. H. Cassell, J. G. Tully, and J. G.Whitcomb (ed.), Recent advances in mycoplasmology. Fischer Verlag, Stutt-gart, Germany.

41. Staden, R., and A. D. McLachlan. 1982. Codon preference and its use in

identifying protein coding regions in long DNA sequences. Nucleic AcidsRes. 10:141–156.

42. Stamburski, C., J. Renaudin, and J. M. Bove. 1990. Characterization of apromoter and a transcription terminator of Spiroplasma melliferum virusSpV4. J. Bacteriol. 172:5586–5592.

43. Stamburski, C., J. Renaudin, and J. M. Bove. 1991. First step toward avirus-derived vector for gene cloning and expression in spiroplasmas, organ-isms which read UGA as a tryptophan codon: synthesis of chloramphenicolacetyltransferase in Spiroplasma citri. J. Bacteriol. 173:2225–2230.

44. Townsend, R., and D. B. Archer. 1983. A fibril protein antigen specific toSpiroplasma. J. Gen. Microbiol. 129:199–206.

45. Townsend, R., D. B. Archer, and K. A. Plaskitt. 1980. Purification andcharacterization of spiroplasma fibrils. J. Bacteriol. 142:694–700.

46. Townsend, R., P. G. Markham, K. A. Plaskitt, and M. J. Daniels. 1977.Isolation and characterization of a non-helical strain of Spiroplasma citri.J. Gen. Microbiol. 100:15–21.

47. Vignault, J. C., J. M. Bove, C. Saillard, R. Vogel, A. Faro, L. Venegas, W.Stemmer, S. Aoki, R. E. McCoy, A. S. Al-Beldawi, M. Larue, O. Tuzcu, M.Ozsan, A. Nhami, M. Abassi, J. Bonfils, G. Moutous, A. Fos, F. Poutiers, andG. Viennot-Bourgin. 1980. Mise en culture de spiroplasmes a partir demateriel vegetal et d’insectes provenant de pays circum mediterraneens et duProche Orient. C. R. Acad. Sci. Ser. III 290:775–780.

48. Weisburg, W. G., J. G. Tully, D. L. Rose, J. P. Petzel, H. Oyaizu, D. Yang, L.Mandelco, J. Sechrest, T. G. Lawrence, J. Van Etten, J. Maniloff, and C. R.Woese. 1989. A phylogenetic analysis of the mycoplasmas: basis for theirclassification. J. Bacteriol. 171:6455–6467.

49. Whitcomb, R. F. 1983. Culture media for spiroplasmas. Methods Mycoplas-mol. 1:147–159.

50. Williamson, D. L., P. R. Brink, and G. W. Zieve. 1984. Spiroplasma fibrils.Isr. J. Med. Sci. 20:829–835.

51. Williamson, D. L., J. Renaudin, and J. M. Bove. 1991. Nucleotide sequenceof the Spiroplasma citri fibril protein gene. J. Bacteriol. 173:4353–4362.

52. Ye, F., J. Renaudin, J. M. Bove, and F. Laigret. 1994. Cloning and sequenc-ing of the replication origin (oriC) of the Spiroplasma citri chromosome andconstruction of autonomously replicating artificial plasmids. Curr. Microbiol.29:23–29.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

isolation, characterization, and complementation of a motility … · isolation, characterization,...

Documents