Specific gene targeting in Spiroplasma citri: improved vectors and production of unmarked mutations using site-specific recombination
Sybille Duret,
Aurélie André and
Joël Renaudin
UMR 1090 Génomique Développement et Pouvoir Pathogène, INRA, Université de Bordeaux 2, Centre INRA de Bordeaux, 71 avenue Edouard Bourlaux, BP 81, 33883 Villenave d'Ornon Cedex, France
Correspondence
Joël Renaudin
renaudin{at}bordeaux.inra.fr
 |
ABSTRACT
|
---|
In Spiroplasma citri, where homologous recombination is inefficient, specific gene targeting could only be achieved by using replicative, oriC plasmids. To improve the probability of selecting rare recombination events without fastidious, extensive passaging of the transformants, a new targeting vector was constructed, which was used to inactivate the crr gene encoding the IIA component of the glucose phosphotransferase system (PTS) permease. Selection of recombinants was based on a two-step strategy using two distinct selection markers, one of which could only be expressed once recombination had occurred through one single crossover at the target gene. According to this strategy, spiroplasmal transformants were screened and multiplied in the presence of gentamicin before the crr recombinants were selected for their resistance to tetracycline. In contrast to the wild-type strain GII-3, the crr-disrupted mutant GII3-gt1 used neither glucose nor trehalose, indicating that in S. citri the glucose and trehalose PTS permeases function with a single IIA component. In addition, the feasibility of using the transposon 
TnpR/res recombination system to produce unmarked mutations in S. citri was demonstrated. In an arginine deiminase (arcA-disrupted) mutant, the tetM gene flanked by the res sequences was efficiently excised from the chromosome through expression of the TnpR resolvase from a replicative oriC plasmid. Due to oriC incompatibility, plasmid loss occurred spontaneously when selection pressure was removed. This approach will be helpful for constructing unmarked mutations and generating multiple mutants with the same selection marker in S. citri. It should also be relevant to other species of mollicutes.
Abbreviations: ADI, arginine deiminase; CK, carbamate kinase; CDS, coding sequences; HNPP, 2-hydroxy-3-naphthoic acid-2'-phenylanilide phosphate; OTC, ornithine carbamoyltransferase; PPLO, pleuropneumonia-like organisms; PTS, phosphotransferase system
 |
INTRODUCTION
|
---|
Spiroplasmas and phytoplasmas are plant-pathogenic bacteria belonging to the class Mollicutes, a group of wall-less micro-organisms phylogenetically related to low G+C content, Gram-positive bacteria (Weisburg et al., 1989
). Plant-pathogenic mollicutes are associated with many diseases affecting economically important crops such as ornamentals, vegetables, fruit trees and grapevine (Lee et al., 2000
; Seemüller et al., 2002
). Both phytoplasmas and spiroplasmas are restricted to the phloem sieve tubes of affected plants and are transmitted by phloem sap-feeding, leafhopper or psyllid vectors, in which they also multiply. Spiroplasmas are, in contrast to phytoplasmas, cultivable in vitro. Spiroplasma citri, the causal agent of citrus stubborn disease, has been cultured since 1970 (Saglio et al., 1971
, 1973
). Over the years it has become a model organism for studying the relationships of plant mollicutes with their hosts, the plant and the insect vector (Bové et al., 2003
; Fletcher et al., 1998
). The ongoing determination of spiroplasma genome sequences (http://www.genome.ou.edu/spiro.html) promises to increase our knowledge of the spiroplasmal biology. However, an efficient methodology to generate defined mutations is essential for studying gene function. In S. citri, mutagenesis has been achieved in a random or targeted manner (Duret et al., 1999
; Foissac et al., 1997
; Gaurivaud et al., 2000b
; Jacob et al., 1997
; Renaudin, 2002
), leading to the identification of genes involved in insect transmission and pathogenicity (André et al., 2005
; Boutareaud et al., 2004
; Duret et al., 2003
; Foissac et al., 1997
; Gaurivaud et al., 2000a
). Although double-crossover allelic exchange is easy to perform with many bacteria, it remains very difficult with others and in particular with mollicutes (Dybvig & Volker, 1996
; Razin et al., 1998
; Renaudin, 2002
). A suicide vector approach using recombinant plasmids unable to replicate has been successfully used in Acholeplasma laidlawii (Dybvig & Woodward, 1992
), Mycoplasma genitalium (Dhandayuthapani et al., 1999
) and Mycoplasma gallisepticum (Markham et al., 2003
). In S. citri, where homologous recombination is inefficient, specific gene targeting could only be achieved by using replicative, oriC plasmids as disruption vectors (Duret et al., 1999
; Gaurivaud et al., 2000b
; Renaudin, 2002
). However, plasmid recombination required extensive passaging of the transformants, and in most cases recombination occurred at oriC rather than at the target gene. To enhance the probability of recombination at the target gene, oriC vectors in which the oriC fragment was reduced to the minimal sequences required for replication were constructed (Lartigue et al., 2002
). These plasmids have been successfully used to inactivate genes spi and ptsG, encoding spiralin, the major membrane protein and the IICB component of the glucose phosphotransferase system (PTS) permease, respectively (Lartigue et al., 2002
; Duret et al., 2003
; André et al., 2005
). However, due to low recombination frequencies, a minimal length of homologous sequences was required for selecting recombination events. For example, attempts to inactivate the crr gene (encoding the glucose PTS permease IIA component) by recombination within a 220 bp fragment were unsuccessful (A. André & J. Renaudin, unpublished data).
This report describes a new gene targeting vector allowing selection of rare recombination events. The two-step strategy involves two distinct selection markers. First transformants are screened for their resistance to gentamicin, and then site-specific recombinants are selected for their resistance to tetracycline, which can only be expressed through recombination at the target gene. The usefulness of this strategy was established by the construction of a crr-disrupted mutant, the characterization of which demonstrated that glucose and trehalose PTS permeases of S. citri share a single IIA component. Further functional genetic studies will certainly require complementation and multiple mutation experiments. In S. citri, however, such experiments are limited by the very small number of suitable selectable markers. In fact, only genes aacA-aphD of Tn4001 and tetM of Tn916 conferring resistance to gentamicin and tetracycline are currently used. S. citri is also sensitive to chloramphenicol but oriC plasmids carrying the chloramphenicol acetyltransferase gene of Tn9 transform spiroplasma cells at low frequencies (Renaudin, 2002
). To avoid this limitation, the production of unmarked mutants was investigated. An unmarked, arcA-disrupted mutant was produced by using the transposon 
TnpR/res recombination system (Reed, 1981
). In this system, the resolvase TnpR mediates the efficient resolution of the cointegrate intermediate generated during transposition by binding-specific recombination sites (res) (Grindley, 2002
). In our study, the efficient elimination of the tetM resistance gene flanked with the res sequences was obtained through expression of the resolvase from an oriC plasmid.
 |
METHODS
|
---|
Bacterial strains, transformation and growth conditions.
Escherichia coli TG1 (supE hsd
5 thi
(lacproAB) F'[traD36 proAB+ lacIq lacZ
M15] or DH10B [F mcrA
(mrrhsdRMSmcrBC)
80lacZ
M15
lacX74 recA1 araD139
(araleu)7697 galK rpsL(StrR) endA1 nupG] were used as the host strains for cloning experiments and plasmid propagation. E. coli C600 [F e14 (mcrA) supE44 thi-1 thr-1 leuB6 lacY1 tonA21] was used for construction and propagation of plasmids carrying res sequences. E. coli TG1 competent cells were transformed by heat shock at 42 °C, whereas C600 and DH10B were transformed by electroporation. Transformants were selected by plating on LB agar medium containing 50 µg ampicillin or chloramphenicol ml1. S. citri GII-3 wild-type strain was originally isolated from its leafhopper vector Circulifer haematoceps captured in Morocco (Vignault et al., 1980
). Spiroplasmas were grown at 32 °C in SP4 medium from which fresh yeast extract was omitted or in HSI medium in which the horse serum was replaced by 1 % pleuropneumonia-like organism (PPLO) serum fraction (Whitcomb, 1983
). Electrotransformation of S. citri has been described previously by Stamburski et al. (1991)
. Transformants were selected in SP4 medium containing 2 µg tetracycline ml1 or 100 µg gentamicin ml1. Selection of the crr-disrupted mutant, unable to use glucose, was performed in SP4 medium in which glucose had been replaced by fructose (SP4Fru).
Construction of plasmids.
The S. citri oriC plasmids pBOG, pC1/2, pC55 and pSR2 have been described elsewhere (Lartigue et al., 2002
; Renaudin, 2002
). Plasmid pSD1 is identical to pSD2 (Lartigue et al., 2002
), except that the oriC fragment is in the opposite orientation. Plasmid pSD6 was obtained by inserting res sequences of the 
transposon at the BamHI and BglII sites, respectively, upstream and downstream of the tetracycline resistance cassette of pSD1. The res sequences were retrieved from plasmid pCG118 (Malaga et al., 2003
) by digestion with BamHI and BglII. Plasmid pSRCAT was obtained by inserting the cat gene (as the 765 bp fragment resulting from digestion of Cat1/Cat2 PCR product with BamHI and BglII) at the BglII site of pSR2, downstream of the spiralin gene promoter. Plasmid pSD12 was obtained through insertion of the 163 bp oriC fragment of pC1/2 at the BamHI site of pSRCAT. To construct pSD25, the resolvase gene (tnpR) was amplified from pCG123 (Malaga et al., 2003
) with primer pair TNP5/TNP6 and inserted at the BglII site of pSR2. Then, the 923 bp BamHIBglII fragment containing tnpR downstream of the spiralin gene promoter was retrieved from pSD25 and inserted into the BamHI site of pSD12 to yield pSD261/262 (depending on tnpR orientation). To construct pSRG, the aacA-aphD gene encoding the gentamicin and kanamycin resistance determinant of Tn4001 was amplified from pBOG with primers Gmr1 and Gmr2. After restriction with BamHI plus BglII, the 1969 bp DNA fragment was inserted into the BglII site of pSR2 to yield pSRG. The 2303 bp PstI cassette of pSRG containing the aacA-aphD coding sequence immediately downstream of the spiralin gene promoter was then inserted into the PstI site of pC55 to yield pGOT1/2, depending on the insert orientation. Plasmids pGOT1/2, pSD61 and pGOT18 are described in Results.
DNA analyses.
Genomic DNA from spiroplasma cells was isolated with the Wizard genomic DNA purification kit (Promega). Restricted DNA was fractionated by agarose gel electrophoresis, transferred to positively charged nylon membranes by the alkali procedure, and hybridized with appropriate [digoxigenin]dUTP-labelled probes using standard procedures (Sambrook et al., 1989
). Hybridization signals were detected with anti-digoxigeninalkaline phosphatase conjugate and HNPP (2-hydroxy-3-naphthoic acid-2'-phenylanilide phosphate) as the substrate, following the supplier's instructions (Roche Diagnostics). Fluorescent signals were detected using a Fluor-S Multimager phosphoimager (Bio-Rad). The arcA, crr, oriC and res probes were generated by PCR amplification with primer pairs ArcA3/ArcA4, GPA1/GPA3, EV13/EV14 and Res1/Res2, respectively. Primers used in this study are listed in Table 1
.
Carbohydrate and arginine catabolism.
In S. citri, carbohydrate fermentation results in the production of lactic and acetic acids that are released into the medium causing acidification (Miles, 1992
). In this study, sugar fermentation was monitored by following the pH decrease during spiroplasma growth as described previously (André et al., 2005
). S. citri cells grown to the mid-exponential phase in SP4 medium were harvested by centrifugation (15 min, 12 000 g, 18 °C), washed twice and resuspended in HEPES/sorbitol buffer [8 mM HEPES, pH 7·4, 10 % (w/v) sorbitol]. Two millilitres of sugar-free HSI medium supplemented with 0·5 % glucose, fructose, trehalose or sorbitol was inoculated with 5x108 washed cells. Sugar fermentation was followed by pH measurements every 24 h for 7 days. To determine the ability to hydrolyse arginine, spiroplasmas grown in SP4 were harvested as described above except that they were washed three times in HEPES/sorbitol buffer. Ten millilitres of HSI medium supplemented with 0·1 % fructose and/or 1 % arginine was inoculated with 5x108 washed cells. Sugar fermentation and/or arginine hydrolysis were followed by pH measurements for 14 days.
Experimental transmission assay.
Microinjection of S. citri into the leafhopper (C. haematoceps) vector and transmission to the periwinkle (Catharanthus roseus) host plant have been described previously (Foissac et al., 1996
, 1997
). Briefly, the insects were injected with 105106 spiroplasma cells and then caged on young periwinkle plants (with 12 insects per plant and 510 plants per spiroplasma strain) for a 2 week transmission period. Plant symptoms were recorded for 8 weeks after transmission. Culture of S. citri from plants and insects were as described previously (Foissac et al., 1996
, 1997
; Duret et al., 2003
). Titres of S. citri in the insects were determined at the end of the transmission period. In the plants, titres were determined 4 weeks after the transmission period.
 |
RESULTS
|
---|
Construction of the disruption vector pGOT18 and isolation of the crr-disrupted mutant
The pGOT1 vector was designed in order to facilitate the selection of recombinants in specific gene disruption experiments. This plasmid was constructed by inserting the 2303 bp PstI cassette of pSRG, containing the gentamicin resistance gene aacA-aphD downstream of the spiralin gene promoter, into the S. citri oriC plasmid pC55. The resulting plasmid pGOT1 comprises two distinct selection markers, one of which can only be expressed once recombination has occurred at the target gene (Fig. 1
). While the gentamicin resistance gene is constitutively transcribed from the spiralin gene promoter, the promoterless tetM gene cannot be transcribed from the free plasmid. The possible read-through transcription of tetM from the lacZ promoter is further prevented by two copies of the transcription terminator of the fibril protein gene. The crr gene encodes the IIA component of the glucose PTS permease, the organization of which has been described previously (André et al., 2003
). For inactivation, the internal crr fragment was obtained by PCR amplification of S. citri genomic DNA with primer pair GPA7/GPA8. After restriction with BamHI plus BglII, the 220 bp fragment was inserted into the BamHI site of pGOT1 to yield the disruption plasmid pGOT18 (Fig. 1
). S. citri GII-3 was electroporated with 2 µg pGOT18 and the transformation mixture was plated onto SP4Fru plates containing either 100 µg gentamicin ml1 or 2 µg tetracycline ml1. Gentamicin-resistant colonies were obtained at a frequency of approximately 106 transformants µg1 c.f.u.1. In contrast, no (less than 109) tetracycline-resistant colonies were obtained, indicating that the tetM gene was not or was very poorly expressed from the free pGOT18. Gentamicin-resistant colonies were grown in liquid medium and then plated onto tetracycline agar medium. Tetracycline-resistant colonies were obtained at a frequency of 12x108. Six tetracycline-resistant transformants were grown in the presence of the antibiotic for three passages and their DNAs were analysed by Southern blot hybridization with a crr probe (Fig. 2
). In clones 2, 5, 6 and 7 (lanes 36) the probe hybridized with two HindIII fragments of 2·3 and 3 kbp corresponding to the wild-type crr and the free pGOT18 fragment, respectively. In contrast, these fragments were not detected in clones 8 and 9 (lanes 7 and 8), suggesting that pGOT18 had integrated into the crr region of the chromosome as early as the third passage. In these clones, the sizes (1·6 and 3·7 kbp) of the hybridizing fragments were consistent with plasmid recombination at crr, as illustrated in Fig. 1
. The crr-disrupted mutant was triply cloned and named GII3-gt1.

View larger version (18K):
[in this window]
[in a new window]
|
Fig. 1. Partial restriction maps of plasmids pGOT1 and pGOT18, and schematic representation of pGOT18 integration by recombination at the chromosomal crr gene. O, S. citri oriC; p tetM, tetM gene of Tn916 without promoter; crr, truncated crr; T, transcription terminator of the S. citri fibril protein gene. B, BamHI; E, EcoRI; V, EcoRV; H, HindIII; Hp, HpaI; N, NsiI; P, PstI. Block arrows indicate direction of transcription.
|
|

View larger version (49K):
[in this window]
[in a new window]
|
Fig. 2. Southern blot hybridization between restricted DNA of S. citri and the crr probe. Purified pGOT18 (lane 1), and genomic DNAs from S. citri GII-3 (lane 2) and six transformants (lanes 38) were restricted by HindIII. The crr probe consisted of a DNA fragment amplified with primer pair GPA1/GPA3. Sizes are indicated in kbp.
|
|
Sugar catabolism in GII3-gt1
Sugar fermentation in S. citri results in lactic acid production, and therefore can be monitored by measuring the pH decrease of the culture medium during spiroplasma growth. We have compared the spiroplasma growth of the crr-disrupted mutant GII3-gt1 to that of the wild-type GII-3 in HSI medium supplemented with sorbitol, fructose, glucose or trehalose (Fig. 3a, b, c and d
, respectively). As expected, the wild-type strain GII-3 was found to acidify the medium supplemented with fructose, glucose and trehalose but not sorbitol (used as the control). In the case of mutant GII3-gt1, pH decrease was observed in the presence of fructose but not in the presence of glucose or trehalose. In the presence of fructose the pH curves of GII-3 and GII3-gt1 were indistinguishable. These results clearly indicated that GII3-gt1 was unable to use glucose and trehalose as well, indicating that the crr-encoded, glucose-PTS permease IIAGlc polypeptide is required for glucose and trehalose import.
Insect transmission and pathogenicity of GII3-gt1
To determine whether the inability to import glucose and trehalose affected spiroplasmal pathogenicity, the S. citri mutant GII3-gt1 was experimentally transmitted to periwinkle plants via injection into the leafhopper vector C. haematoceps. Insects were microinjected with cultures of S. citri GII-3 (used as the control) or GII3-gt1, and caged on young periwinkle plants for 2 weeks. After the transmission period, the spiroplasma titres in the insects were determined, and plant symptoms were recorded for 6 weeks (Table 2
). The results showed that, similarly to the wild-type strain GII-3, the S. citri mutant GII3-gt1 multiplied in the injected leafhoppers, and was transmitted to periwinkle plants, in which it multiplied and induced severe symptoms. Characterization of spiroplasmas isolated from these symptomatic plants proved that symptom expression was due to multiplication of GII3-gt1, and not to the presence of contaminants or spiroplasmas having reverted to the wild-type phenotype. However, while all five plants (of five) were infected in the case of S. citri GII-3, only four plants of ten developed symptoms in the case of mutant GII3-gt1. Furthermore, symptoms appeared with a delay (12 weeks) suggesting that, in these plants, a smaller amount of spiroplasmas was injected by the insects. Nevertheless, the low transmission efficiency of the mutant (as compared with the wild-type) was not due to the failure to multiply in the leafhopper. In spite of its inability to use trehalose, which is the major sugar in the insect haemolymph (Wyatt, 1967
), GII3-gt1 was found to multiply at approximately the same rate as the wild-type strain, reaching a plateau within 4 days after injection (Fig. 4
). After 2 weeks (i.e. at the end of the transmission period), the GII3-gt1 titre in the insects (6·3x105 c.f.u. per insect) was slightly lower than that (1·2x106 c.f.u. per insect) of GII-3, but higher than the minimal spiroplasma titre (104105 c.f.u. per insect) required for efficient transmission. In the light of previous studies (André et al., 2005
; Boutareaud et al., 2004
, Duret et al., 2003
), the delay in symptom production and the fact that only some of the plants were infected strongly suggested that GII3-gt1 was affected in its transmission rather than in its ability to multiply in the insect haemolymph. These data indicate that, in the insect, the ability of the spiroplasma to multiply does not simply rely on its ability to import glucose or trehalose. However, the lower transmission efficiency of the GII3-gt1 mutant suggests that glucose and/or trehalose might play a role in the insect transmission process. In this respect, it has been hypothesized that the failure to use glucose might hinder spiroplasmal multiplication in the salivary gland cells, resulting in decreased transmission efficiency (André et al., 2005
).
View this table:
[in this window]
[in a new window]
|
Table 2. Transmission of S. citri GII-3 (wild-type) and GII3-gt1 (crr-disrupted mutant) to periwinkle (Catharanthus roseus) plants via injection into the leafhopper vector (Circulifer haematoceps)
|
|
Gene organization of the S. citri arginine deiminase (ADI) operon
With the aim to create an insertional mutant unable to hydrolyse arginine, we first characterized the gene cluster encoding the ADI pathway of S. citri. Briefly, from the data of the ongoing S. citri genome sequencing project, sequences encoding putative polypeptides with high similarities to enzymes of the arginine catabolism pathway were identified. The 5·2 kbp region of the chromosome (GenBank accession no. DQ004462) was found to contain four coding sequences (CDS), which we named arcA, arcB, arcC and arcD, as they encode proteins homologous to ADI, ornithine carbamoyltransferase (OTC), carbamate kinase (CK) and arginineornithine antiporter (ArcD), respectively (Fig. 5
). The best BLAST hits were with ADI from Clostridium perfringens (50 % identity, 70 % similarity), OTC from Mycoplasma capricolum (63 % identity, 77 % similarity), CK from Mycoplasma mycoides (59 % identity, 73 % similarity) and ArcD from Mesoplasma florum (40 % identity, 61 % similarity). All four CDS start with an ATG initiation codon preceded by a typical ribosome-binding sequence and end with a TAA (arcA, arcC, arcD) or TAG (arcB) stop codon. The occurrence of very short (less than 50 nt) intergenic regions, as well as the absence of transcription terminator-like structures in between the arc genes, strongly suggests that they are transcribed as a single transcription unit. An imperfect inverted repeat located immediately downstream of arcC might represent the transcription terminator. The gene order, arcABDC, of the S. citri ADI operon is similar to those of Bacillus licheniformis (Maghnouj et al., 1998
) and C. perfringens (Ohtani et al., 1997
), except that no gene homologous to arcR (encoding the arginine regulator) could be identified, either in the vicinity of the ADI operon or somewhere else in the chromosome. The presence of direct and inverted repeat sequences in the intergenic region upstream of arcA suggests that this operon might be regulated. However, no palindromic sequences resembling arcR binding sites were identified.

View larger version (18K):
[in this window]
[in a new window]
|
Fig. 5. Partial restriction map and gene organization of the S. citri ADI operon. Comparison with the ADI gene clusters of Clostridium perfringens, Pseudomonas aeruginosa and Mycoplasma penetrans. Arrowheads indicate direct and inverted repeat sequences. E, EcoRI; Hp, HpaI; N, NsiI; T, transcription terminator structure.
|
|
Construction of the disruption vector pSD61 and isolation of the arcA-disrupted mutant
Plasmid pSD61 was constructed from pSD6, in which the tetM gene was flanked by the res sequences. The arcA internal fragment was obtained by PCR amplification of S. citri genomic DNA with primer pair ArcA1/ArcA2. After restriction by BamHI and BglII, the 380 bp fragment was inserted into the BamHI site of pSD6 to yield pSD61 (Fig. 6
), which was used to transform S. citri GII-3. Tetracycline-resistant colonies were obtained at a frequency of 5x106 transformants µg1 c.f.u.1. Ten transformants were propagated in liquid medium containing tetracycline for 10 passages. To check for plasmid recombination at arcA, genomic DNAs from five transformants were analysed by Southern blot hybridization with the arcA probe. The results indicated that one transformant still carried pSD61 as free extrachromosomal DNA, whereas the other four carried pSD61 sequences integrated into the chromosome. Plasmid recombination was found to occur either at the spiralin gene promoter (one transformant) or at oriC (two transformants). The hybridization pattern of the fifth transformant indicated a mixed population of cells, some in which plasmid recombination had occurred at oriC and others in which recombination had occurred at arcA (data not shown). Indeed, subcloning this transformant yielded two distinct profiles, one of which was consistent with plasmid recombination at arcA (Fig. 6
) and the other with recombination at oriC. The relevant clones were named GII3-arg1 and GII3-ori1, respectively. As shown in Fig. 7
, the wild-type arcA fragments (3·5 kbp NsiI, 2·5 kbp HpaI, 6·1 kbp EcoRI and 6·55 kbp HindIII), which were detected in S. citri GII-3 (lanes 1, 7, 10 and 14) and GII3-ori1 (lanes 3, 9, 12 and 16), were not detected in GII3-arg1 (lanes 2, 8, 11 and 15), indicating that, in this mutant, recombination had occurred within the arcA region. In mutant GII3-arg1, the sizes of the restriction fragments hybridizing with the arcA probe, such as the 9 kbp HpaI and the two EcoRI fragments of 1·6 and 11 kbp (lanes 8 and 11), were consistent with integration of pSD61 into the chromosome by one crossover recombination at arcA, as illustrated in Fig. 6
.

View larger version (24K):
[in this window]
[in a new window]
|
Fig. 6. Schematic representation of pSD61 recombination at the arcA gene. Partial restriction maps and gene organization of the arcA regions of S. citri mutants GII3-arg1 and GII3-arg2. Positions of primers ArcA3, Tet7 and Rev (universal reverse) are indicated by open arrows. B, BamHI; E, EcoRI; H, HindIII; Hp, HpaI; N, NsiI; P, PstI.
|
|

View larger version (91K):
[in this window]
[in a new window]
|
Fig. 7. Southern blot hybridization between restricted DNAs from S. citri GII-3 (wild-type), GII3-arg1 and GII3-ori1, and the arcA probe. Purified pSD61 (lanes 13 and 17) and genomic DNAs from GII-3 (lanes 1, 4, 7, 10 and 14), GII3-arg1 (2, 5, 8, 11 and 15) and GII3-ori1 (lanes 3, 6, 9, 12 and 16) were restricted by NsiI (lanes 13), EcoRV (lanes 46), HpaI (lanes 79), EcoRI (lanes 1013) and HindIII (lanes 1417). Sizes of DNA fragments are indicated in kbp.
|
|
Isolation of the unmarked ADI mutant GII3-arg2
In order to produce an unmarked mutant, the ADI mutant GII3-arg1 was transformed with pSD262 carrying the resolvase gene tnpR under the control of the spiralin gene promoter. Following expression of tnpR, the resolvase was expected to excise the DNA fragment in between the res sequences, i.e. the tetM gene, leading to a loss of tetracycline resistance. When S. citri GII3-arg1 was electrotransformed with pSD262, chloramphenicol-resistant colonies were obtained at a frequency of 25x108 transformants µg1 c.f.u.1. Twelve transformants (112) were grown in the presence of chloramphenicol for two passages, and screened for the presence of cat and tetM genes by PCR amplification with primer pairs Cat1/Cat2, ArcA3/Tet7 and ArcA3/Rev (Fig. 8a, b and c
, respectively). As expected, all 12 transformants but one (which probably represents a spontaneous chloramphenicol-resistant mutant, in lane 6) yielded positive amplification with primer pair Cat1/Cat2, indicating that they all contained pSD262 (Fig. 8a
, lanes 35 and 714). Interestingly, while most of the tranformants (9 of 12) yielded positive amplification with primer pair ArcA3/Tet7 (Fig. 8b
, lanes 4 and 714), two of them seemed not to contain the tetM gene, as indicated by the absence of PCR product (Fig. 8b
, lanes 3 and 5). In addition, detection of the 1 kbp ArcA3/Rev PCR product suggested that in all 11 transformants a number of cells had lost the tetM gene (Fig. 8c
, lanes 35 and 714). To further confirm excision of the tetM gene, genomic DNAs were restricted by HindIII and hybridized with the res probe. In mutant GII3-arg1, the two res copies were detected as two HindIII hybridizing fragments of 2 and 2·3 kbp (Fig. 9
, lane 2), whereas in pSD262 transformants 1 and 3, the probe hybridized with one single DNA fragment of 1·9 kbp (Fig. 9
, lanes 3 and 5), a size that was in perfect agreement with excision of the tetM gene through recombination between the two res sequences (see Fig. 6
). In all other transformants (Fig. 9
, lanes 4 and 714) detection of the three fragments of 1·9, 2 and 2·3 kbp suggested the presence of a mixed population of cells, some of which still carried tetM and others not. Indeed, plating these transformants onto SP4 and tetracycline SP4 plates revealed that 2070 % of spiroplasmal cells had lost the tetM gene, as indicated by their sensitivity to the antibiotic. Interestingly enough, due to incompatibility of oriC plasmids, pSD262 carrying tnpR was spontaneously lost during propagation of the transformants in the absence of chloramphenicol. After only four passages, no plasmid could be detected by PCR with primer pair Cat1/Cat2 (data not shown).

View larger version (37K):
[in this window]
[in a new window]
|
Fig. 8. PCR amplification of genomic DNAs from S. citri GII-3, GII3-arg1 and pSD262 transformants with primer pairs Cat1/Cat2 (a), ArcA3/Tet7 (b) and ArcA3/Rev (pUC/M13 reverse) (c). Lane 1, control without DNA; lane 2, DNA from GII3-arg1; lanes 314, DNAs from pSD262 transformants 112.
|
|

View larger version (31K):
[in this window]
[in a new window]
|
Fig. 9. Southern blot hybridization between genomic DNAs from S. citri GII-3, GII3-arg1 and pSD262 transformants and the res probe. Lane 1, GII-3; lane 2, GII3-arg1; lanes 314, pSD262 transformants 112.
|
|
Arginine catabolism in S. citri GII-3 and GII3-arg1
To determine the ability of the arcA-disrupted mutant to metabolize arginine, S. citri GII-3 (wild-type) and GII3-arg1 (mutant) were grown in HSI medium supplemented with various sugars and/or arginine, and the pH curves were compared (Fig. 10
). Carbohydrate fermentation in S. citri essentially results in lactic acid production and, consequently, leads to a pH decrease in the culture medium. When S. citri GII-3 (wild-type) and GII3-arg1 (arcA-disrupted mutant) were grown in the presence of fructose, the pH progressively decreased from 7·6 to 5·5 within 3 days. In contrast, no such pH decrease was observed in the presence of sorbitol, which is not metabolized by S. citri. Interestingly, when S. citri GII-3 was grown in the presence of fructose and arginine, the pH first dropped from 7·5 to 5·75 as a result of fructose fermentation. Then, due to the arginine catabolism and subsequent ammoniac release, the pH progressively increased to reach 7·8 after 14 days (Fig. 10
). In the case of the arcA mutant GII3-arg1, no pH increase was observed, indicating that this mutant was unable to metabolize arginine. It has been shown previously that complete utilization of arginine by S. citri only occurred when an alternative energy source, glucose or fructose, was present to encourage growth (Townsend, 1976
). Accordingly, a minor pH increase was observed when S. citri GII-3 was grown in the absence of fructose. As expected, no such pH increase was observed with mutant GII3-arg1 (Fig. 10
). The unmarked mutant GII3-arg2 displayed an identical phenotype.
Insect transmission and pathogenicity of GII3-arg1
Insect transmission and pathogenicity of the arcA-disrupted mutant was determined through experimental transmission to periwinkle plants as described above for mutant GII3-gt1. Four weeks after transmission, all 10 plants displayed symptoms undistinguishable from those produced by the wild-type strain GII-3. These results indicate that the GII3-arg1 mutant was efficiently transmitted to plants by the leafhopper vector. Determination of spiroplasma titres revealed that GII3-arg1 multiplied in the insects and in the plants similarly to S. citri GII-3 (data not shown). These data indicate that arginine degradation through the ADI pathway is not critical for the spiroplasma to complete its life cycle in the vector insect and in the host plant.
 |
DISCUSSION
|
---|
In the present study, disruption of crr was achieved with a pGOT1-based disruption vector. This plasmid vector carries two selection markers, one of which (the gentamicin resistance gene Gmr) is constitutively expressed, whereas the other (tetM), devoid of promoter, can only be expressed when recombination has occurred at the target gene. In this case, transcription of tetM is driven by the target gene promoter. Due to low recombination frequency, direct selection of recombinants by plating the transformed cells on tetracycline medium was unsuccessful in agreement with the fact that, in the presence of tetracycline, pGOT1 behaves as a suicide plasmid. Selection of recombinants required two steps. First, transformants carrying the disruption plasmid were selected for their resistance to gentamicin, and following propagation, spiroplasma cells in which recombination had occurred at the target gene were selected by plating onto tetracycline plates. The advantage of such a two-step selection procedure lies in the almost unlimited number of cells carrying the disruption vector, increasing the probability for selection of rare recombination events. Unexpectedly, some tetracycline-resistant transformants still carried free plasmid, indicating that in spite of the absence of promoter the tetM gene was somehow transcribed. These data suggest that transcription of tetM could proceed from a DNA sequence carried on the crr DNA fragment. Considering the high A+T content (70 mol%) of the crr DNA fragment, the occurrence of sequences resembling the 35 and 10 regions of eubacterial promoters recognized by the RNA polymerase cannot be excluded.
We have shown previously that in S. citri, the glucose PTS permease enzyme II was split into two distinct polypeptides IIAGlc and IICBGlc encoded by two separate genes crr and ptsG, and that the trehalose PTS permease did not possess its own IIA component (André et al., 2003
). By using a yeast two-hybrid system, we also showed that the IIAGlc domain bound not only the IIBGlc but also the IIBTre domain, suggesting that glucose and trehalose permeases shared a single IIA domain (André et al., 2003
). The finding that the crr-disrupted mutant GII3-gt1 used neither glucose nor trehalose definitely demonstrates that, in S. citri, glucose and trehalose PTS permeases function with a single IIA domain. In spite of its inability to import these two sugars, the crr-disrupted mutant GII3-gt1 multiplied in the leafhopper vector and was transmitted to periwinkle plants, in which it induced symptoms. In insects, trehalose is the main sugar in the haemolymph but glucose and fructose also are present (Florkin & Jeuniaux, 1974
). Therefore, it is likely that multiplication of GII3-gt1 in the leafhopper vector mainly relies on the use of fructose. However, fructose import is not an absolute requirement for S. citri multiplication in the insect, as the fructose operon mutant GMT553 multiplies to high titre in the leafhopper (Gaurivaud et al., 2000a
). These data attest the capability of S. citri to adapt to carbohydrate changes in its environment. We have shown previously that insect transmission of the ptsG-disrupted mutant GII3-glc1, which is unable to import glucose, was less efficient than that of the wild-type strain GII-3 (André et al., 2005
). As expected, transmission of GII3-gt1 also was found to be poorly efficient. In vitro, S. citri metabolizes glucose, fructose and trehalose equally well (Chang et al., 1994
). However, when both fructose and glucose are present, S. citri uses fructose preferentially (André et al., 2005
). As a result, while the S. citri mutant GMT553, which is unable to use fructose, is non-pathogenic to plants (Foissac et al., 1997
; Gaurivaud et al., 2000a
), GII3-glc1 induces symptoms identical to those produced by the wild-type strain, indicating that glucose import is not essential for pathogenicity (André et al., 2005
). Similarly, the crr-disrupted mutant GII3-gt1 proved to be highly pathogenic to periwinkle plants, showing that trehalose import, like glucose import, is not an absolute requirement for pathogenicity.
Arginine metabolism leading to the synthesis of ATP through the ADI pathway is considered to be the primary energy conserving route in non-glycolytic mollicutes (Pollack et al., 1997
). However, carbohydrates and arginine can be metabolized concomitantly by glycolytic mollicutes, including the plant pathogen S. citri (Townsend, 1976
; Igwegbe & Thomas, 1978
; Stevens et al., 1984
; Pollack et al., 1997
). The ADI pathway comprises three reactions catalysed by ADI, OTC and CK, and converts arginine to ornithine, ammonia and CO2, with concomitant generation of ATP. In mollicutes, the ADI gene clusters display very diverse gene organizations depending on the mollicute species. A rapid survey using the Molligen software (Barré et al., 2004
) revealed that M. penetrans lacks arcD with the organization arcABC, M. mycoides lacks arcA, with arcBD and arcC being located on two distinct transcription units, and M. gallisepticum has two separated copies of arcA and no arcB, arcC and arcD. In S. citri, we found the relevant genes arcA, arcB and arcC to be clustered in a single operon, together with a fourth gene (arcD) encoding an arginineornithine antiporter, and located in between arcB and arcC. As reported previously (Townsend, 1976
), we found that complete utilization of arginine only occurred when an alternative energy source in the form of glucose or fructose was present to encourage growth. Inactivation of the arcA gene through homologous recombination completely abolished the use of arginine. However, the arcA-disrupted mutant GII3-arg1 was found to multiply in vitro, in its leafhopper vector, and in its host plant to approximately the same rate as the wild-type strain GII-3. These results confirm that, in S. citri, the ADI pathway is not the major energy-generating system and might not be essential for the spiroplasma to complete its life cycle. In most bacteria, energy depletion is an essential signal for inducing the ADI pathway (Cunin et al., 1986
). In S. citri also, the ADI pathway is inducible (Igwegbe & Thomas, 1978
) and therefore might play a role in nutrient stress response. From the arcA-disrupted mutant GII3-arg1, we have produced an unmarked arcA mutant, free of the tetM selection marker, by using the TnpR/res recombination system of the E. coli 
transposon. The site-specific recombinase TnpR has been used as a reporter of gene expression in Vibrio cholerae (Camilli et al., 1994
) and for producing unmarked mutations in mycobacteria (Bardarov et al., 2002
; Malaga et al., 2003
). In our experiments, expression of tnpR driven by the spiralin gene promoter resulted in efficient excision of the tetM marker flanked with the res sequences. Indeed, tetM excision was detected in all pSD262 transformants after only two successive propagations. Moreover, plating onto tetracycline agar plates revealed that one of these transformants yielded no (less than 104) tetracycline-resistant colonies, indicating that tetM had been excised in a large majority of cells. Production of unmarked mutations usually requires the use of counterselectable markers (Reyrat et al., 1998
) for curing the plasmid from which the resolvase is expressed (Malaga et al., 2003
). However, in mycobacteria, plasmid loss could also be achieved using a delivery system made of a pair of replicating plasmids, which are incompatible (Pashley et al., 2003
). In our study, we have used the incompatibility of S. citri oriC plasmids as the selection pressure for plasmid loss. During propagation of transformants in the absence of chloramphenicol, the spontaneous loss of the oriC plasmid carrying the resolvase gene avoided the use of a counterselectable marker. Our results demonstrate that the TnpR resolvase was functional in spiroplasmas and catalyses site-specific recombination between two res sequences in direct orientation on the spiroplasmal chromosome.
In summary, we have developed new genetic tools for reverse genetic studies in S. citri. For specific gene targeting through homologous recombination, the use of pGOT1-based disruption vectors was shown to improve the probability to select rare recombination events through a two-step procedure. As an example, we produced the crr-disrupted mutant GII3-gt1, the characterization of which proved the glucose and trehalose PTS permeases to share a unique IIA component. Also, we have shown that the transposon 
site-specific recombination system functions in S. citri. An unmarked ADI mutant was produced through excision of the tetM gene, allowing the successive use of the same antibiotic to produce multiple mutations. These new tools, like the formerly described oriC plasmids (Renaudin, 2002
; Renaudin & Lartigue, 2005
), should be applicable to a wide variety of mollicute species.
 |
ACKNOWLEDGEMENTS
|
---|
This work was funded by INRA and Région Aquitaine (Grant B05763). Support for A. A. was provided by the Ministère de l'Enseignement Supérieur et de la Recherche. We thank our colleagues J. L. Danet for injecting spiroplasma cultures into the insects and P. Bonnet for growing plants and insects. We are grateful to Dr C. Guilhot for providing the E. coli strain C600 as well as plasmids pCG118 and pCG123.
 |
REFERENCES
|
---|
André, A., Maccheroni, W., Doignon, F., Garnier, M. & Renaudin, J. (2003). Glucose and trehalose PTS permeases of Spiroplasma citri probably share a single IIA domain, enabling the spiroplasma to adapt quickly to carbohydrate changes in its environment. Microbiology 149, 26872696.[CrossRef][Medline]
André, A., Maucourt, M., Moing, A., Rolin, D. & Renaudin, J. (2005). Sugar import and phytopathogenicity of Spiroplasma citri: glucose and fructose play distinct roles. Mol Plant Microbe Interact 18, 3342.[Medline]
Bardarov, S., Bardarov, S., Jr, Pavelka, M. S., Jr, Sambandamurthy, V., Larsen, M., Tufariello, J., Chan, J., Hatfull, G. & Jacobs, W. R., Jr (2002). Specialized transduction: an efficient method for generating marked and unmarked targeted gene disruptions in Mycobacterium tuberculosis, M. bovis BCG and M. smegmatis. Microbiology 148, 30073017.[Medline]
Barré, A., de Daruvar, A. & Blanchard, A. (2004). MolliGen, a database dedicated to the comparative genomics of mollicutes. Nucleic Acids Res 32, D307D310.[Abstract/Free Full Text]
Boutareaud, A., Danet, J. L., Garnier, M. & Saillard, C. (2004). Disruption of a gene predicted to encode a solute binding protein of an ABC transporter reduces transmission of Spiroplasma citri by the leafhopper Circulifer haematoceps. Appl Environ Microbiol 70, 39603967.[Abstract/Free Full Text]
Bové, J. M., Renaudin, J., Saillard, C., Foissac, X. & Garnier, M. (2003). Spiroplasma citri, a plant pathogenic mollicute: relationships with its two hosts, the plant and the leafhopper vector. Annu Rev Phytopathol 41, 483500.[CrossRef][Medline]
Camilli, A., Beattie, D. T. & Mekalanos, J. J. (1994). Use of genetic recombination as a reporter of gene expression. Proc Natl Acad Sci U S A 91, 26342638.[Abstract/Free Full Text]
Chang, C. J., Renaudin, J. & Bové, J. M. (1994). Nutritional requirements of Spiroplasma citri. IOM Lett 3, 520.
Cunin, R., Glansdorff, N., Piérard, A. & Stalon, V. (1986). Biosynthesis and metabolism of arginine in bacteria. Microbiol Rev 50, 314352.[Medline]
Dhandayuthapani, S., Rasmussen, W. G. & Baseman, J. B. (1999). Disruption of gene mg218 of Mycoplasma genitalium through homologous recombination leads to an adherence-deficient phenotype. Proc Natl Acad Sci U S A 96, 52275232.[Abstract/Free Full Text]
Duret, S., Danet, J. L., Garnier, M. & Renaudin, J. (1999). Gene disruption through homologous recombination in Spiroplasma citri: an scm1-disrupted motility mutant is pathogenic. J Bacteriol 181, 74497456.[Abstract/Free Full Text]
Duret, S., Berho, N., Danet, J. L., Garnier, M. & Renaudin, J. (2003). Spiralin is not essential for helicity, motility, or pathogenicity but is required for efficient transmission of Spiroplasma citri by its leafhopper vector Circulifer haematoceps. Appl Environ Microbiol 69, 62256234.[Abstract/Free Full Text]
Dybvig, K. & Volker, L. L. (1996). Molecular biology of mycoplasmas. Annu Rev Microbiol 50, 2557.[CrossRef][Medline]
Dybvig, K. & Woodward, A. (1992). Construction of recA mutants of Acholeplasma laidlawii by insertional inactivation with a homologous DNA fragment. Plasmid 28, 262266.[CrossRef][Medline]
Fletcher, J., Wayadande, A., Melcher, U. & Ye, F. (1998). The phytopathogenic mollicute-insect vector interface: a closer look. Phytopathology 88, 13511358.
Florkin, M. & Jeuniaux, C. (1974). Haemolymph: composition. In the Physiology of Insecta, vol. V, pp. 255307. Edited by M. Rockstein. New York: Academic Press.
Foissac, X., Danet, J. L., Saillard, C., Whitcomb, R. F. & Bové, J. M. (1996). Experimental infection of plants by spiroplasmas. In Molecular and Diagnostic Procedures in Mycoplasmology, vol. 2, pp. 385389. Edited by S. Razin & J. G. Tully. New York: Academic Press.
Foissac, X., Saillard, C., Danet, J. L., Gaurivaud, P., Paré, C., Laigret, F. & Bové, J. M. (1997). Mutagenesis by insertion of transposon Tn4001 into the genome of Spiroplasma citri: characterization of mutants affected in plant pathogenicity and transmission to the plant by the leafhopper vector Circulifer haematoceps. Mol Plant Microbe Interact 10, 454461.
Gaurivaud, P., Danet, J. L., Laigret, F., Garnier, M. & Bové, J. M. (2000a). Fructose utilization and phytopathogenicity of Spiroplasma citri. Mol Plant Microbe Interact 13, 11451155.[Medline]
Gaurivaud, P., Laigret, F., Verdin, E., Garnier, M. & Bové, J. M. (2000b). Fructose operon mutants of Spiroplasma citri. Microbiology 146, 22292236.[Medline]
Grindley, N. D. F. (2002). The movement of Tn3-like elements: transposition and cointegrate resolution. In Mobile DNA II, pp. 272302. Edited by N. L. Craig, R. Craigie, M. Gellert & A. M. Lambowitz. Washington, DC: American Society for Microbiology.
Igwegbe, E. C. & Thomas, C. (1978). Occurrence of enzymes of arginine dihydrolase pathway in Spiroplasma citri. J Gen Appl Microbiol 24, 261269.
Jacob, C., Nouzières, F., Duret, S., Bové, J. M. & Renaudin, J. (1997). Isolation, characterization, and complementation of a motility mutant of Spiroplasma citri. J Bacteriol 179, 48024810.[Abstract/Free Full Text]
Lartigue, C., Duret, S., Garnier, M. & Renaudin, J. (2002). New plasmid vectors for specific gene targeting in Spiroplasma citri. Plasmid 48, 149159.[CrossRef][Medline]
Lee, I.-M., Davis, R. E. & Gundersen-Rindal, D. E. (2000). Phytoplasma: phytopathogenic mollicutes. Annu Rev Microbiol 54, 221255.[CrossRef][Medline]
Maghnouj, A., de Sousa Cabral, T. F., Stalon, V. & Vander Wauven, C. (1998). The arcABDC gene cluster, encoding the arginine deiminase pathway of Bacillus licheniformis and its activation by the arginine repressor argR. J Bacteriol 180, 64686475.[Abstract/Free Full Text]
Malaga, W., Perez, E. & Guilhot, C. (2003). Production of unmarked mutations in mycobacteria using site-specific recombination. FEMS Microbiol Lett 219, 261268.[CrossRef][Medline]
Markham, P. F., Kanci, A., Czifra, G., Sundquist, B., Hains, P. & Browning, G. F. (2003). Homologue of macrophage-activating lipoprotein in Mycoplasma gallisepticum is not essential for growth and pathogenicity in tracheal organ cultures. J Bacteriol 185, 25382547.[Abstract/Free Full Text]
Miles, R. J. (1992). Catabolism in mollicutes. J Gen Microbiol 138, 17731783.[Medline]
Ohtani, K., Bando, M., Swe, T., Banu, S., Oe, M., Hayashi, H. & Shimizu, T. (1997). Collagenase gene (colA) is located in the 3'-flanking region of the perfringolysin O (pfoA) locus in Clostridium perfringens. FEMS Microbiol Lett 146, 155159.[CrossRef][Medline]
Pashley, C. A., Parish, T., McAdam, R. A., Duncan, K. & Stoker, N. G. (2003). Gene replacement in mycobacteria by using incompatible plasmids. Appl Environ Microbiol 69, 517523.[Abstract/Free Full Text]
Pollack, J. D., Williams, M. V. & McElhaney, R. N. (1997). The comparative metabolism of the mollicutes (mycoplasmas): the utility for taxonomic classification and the relationship of putative gene annotation and phylogeny to enzymatic function in the smallest free-living cells. Crit Rev Microbiol 23, 269354.[Medline]
Razin, S., Yogev, D. & Naot, Y. (1998). Molecular biology and pathogenicity of mycoplasmas. Microbiol Mol Biol Rev 62, 10941156.[Abstract/Free Full Text]
Reed, R. R. (1981). Transposon-mediated site-specific recombination: a defined in vitro system. Cell 25, 713719.[CrossRef][Medline]
Renaudin, J. (2002). Extrachromosomal elements and gene transfer. In Molecular Biology and Pathogenicity of Mycoplasmas, pp. 347370. Edited by S. Razin & R. Herrmann. New York: Kluwer Academic/Plenum Publishers.
Renaudin, J. & Lartigue, C. (2005). OriC plasmids as gene vectors for mollicutes. In Mycoplasmas: Pathogenesis, Molecular Biology, and Emerging Strategies for Control, pp. 330. Edited by A. Blanchard & G. Browning. Norwich, UK: Horizon Scientific Press.
Reyrat, J. M., Pelicic, V., Gicquel, B. & Rappuoli, R. (1998). Counter selectable markers: untapped tools for bacterial genetics and pathogenesis. Infect Immun 66, 40114017.[Free Full Text]
Saglio, P., Laflèche, D., Bonissol, C. & Bové, J. M. (1971). Culture in vitro des mycoplasmes associés au stubborn des agrumes et leur observation au microscope électronique. C R Acad Sci 272, 13871390.
Saglio, P., Lhospital, M., Laflèche, D., Dupont, G., Bové, J. M., Tully, J. G. & Freundt, E. A. (1973). Spiroplasma citri gen. and sp. nov.: a mycoplasma-like organism associated with "stubborn" disease of citrus. Int J Syst Bacteriol 23, 191204.
Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
Seemüller, E., Garnier, M. & Schneider, B. (2002). Mycoplasmas of plants and insects. In Molecular Biology and Pathogenicity of Mycoplasmas, pp. 91115. Edited by S. Razin & R. Herrmann. New York: Kluwer Academic/Plenum Publishers.
Stamburski, C., Renaudin, J. & Bové, J. M. (1991). First step toward a virus-derived vector for gene cloning and expression in spiroplasmas, organisms which read UGA as a tryptophan codon: synthesis of chloramphenicol acetyltransferase in Spiroplasma citri. J Bacteriol 173, 22252230.[Medline]
Stevens, C., Cody, R. M., Gudauskas, R. T. & Patterson, A. (1984). Arginine aminopeptidase activity of phytopathogenic spiroplasmas. Isr J Med Sci 20, 10221024.[Medline]
Townsend, R. (1976). Arginine metabolism by Spiroplasma citri. J Gen Microbiol 94, 417420.[Medline]
Vignault, J. C., Bové, J. M., Saillard, C. & 17 other authors (1980). Mise en culture de spiroplasmes à partir de matériel végétal et d'insectes provenant de pays circum méditerranéens et du Proche Orient. C R Acad Sci III 290, 775780.
Weisburg, W. G., Tully, J. G., Rose, D. L. & 9 other authors (1989). A phylogenetic analysis of the mycoplasmas: basis for their classification. J Bacteriol 171, 64556467.[Medline]
Whitcomb, R. F. (1983). Culture media for spiroplasmas. Methods Mycoplasmol 1, 147159.
Wyatt, G. R. (1967). The biochemistry of sugars and polysaccharides in insects. In Advances in Insect Physiology, vol. IV, pp. 287360. Edited by J. W. L. Beament, J. E. Treherne & V. B. Wigglesworth. New York: Academic Press.
Received 15 April 2005;
revised 11 May 2005;
accepted 12 May 2005.
Copyright © 2005 Society for General Microbiology.