Spiralins of Spiroplasma citri and Spiroplasma melliferum: amino acid sequences and putative organization in the cell membrane

Spiralin is the major membrane protein of the helical mollicute Spiroplasma citri. A similar protein occurs in the membrane of Spiroplasma melliferum, an organism related to S. citri. The gene encoding spiralin has been sequenced. A restriction fragment of the spiralin gene has been used as a probe to detect the gene encoding S. melliferum spiralin. A 4.6-kilobase-pair ClaI DNA fragment from S. melliferum strongly hybridized with the probe. This fragment was inserted in pBR322 and cloned in Escherichia coli. It was further subcloned in the replicative forms of M13mp18 and M13mp19, and its nucleotide sequence was determined (GenBank accession number M33991). An open reading frame showing 88.6% base sequence homology with the S. citri spiralin gene could be identified and was assumed to be the gene encoding S. melliferum spiralin. The deduced amino acid sequence of the protein had 75% homology with the spiralin sequence. In particular, the two proteins possess a stretch of 20 amino acids which can form an alpha-helix, in which all polar amino acids occupy approximately one-third of the axial projection down the helix. On the basis of these data and published data, we propose a topological model for the structural organization of the spiralin in the cell membrane of spiroplasmas.     

Spiroplasma citri virus SpV1-derived cloning vector: deletion formation by illegitimate and homologous recombination in a spiroplasmal host strain which probably lacks a functional recA gene.

We have previously described the use of the replicative form (RF) of Spiroplasma citri virus SpV1 as a vector for expressing an epitope of the P1 adhesin protein from Mycoplasma pneumoniae in S. citri (A. Marais, J. M. Bové, S.F. Dallo, J. B. Baseman, and J. Renaudin, J. Bacteriol. 175:2783-2787, 1993). We have now studied the structural instability of the recombinant RF leading to loss of the DNA insert. Analyses of viral clones with deletions have shown that both illegitimate and homologous recombination were involved in deletion formation. For one such clone, deletion has occurred via a double crossing-over exchange between the circular free viral RF and SpV1 viral sequences present in the S. citri host chromosome. The homologous recombination process usually requires the RecA protein. However, characterization of the recA gene of the S. citri R8A2 host strain revealed that over two-thirds of the open reading frame of the recA gene was deleted from the C-terminal part, indicating that this particular strain is probably RecA deficient.     

Integrative and free Spiroplasma citri oriC plasmids: expression of the Spiroplasma phoeniceum spiralin in Spiroplasma citri.

The replication region (oriC) of the Spiroplasma citri chromosome has been recently sequenced, and a 2-kbp DNA fragment was characterized as an autonomously replicating sequence (F. Ye, J. Renaudin, J. M. Bové, and F. Laigret, Curr. Microbiol. 29:23-29, 1994). In the present studies, we have combined this DNA fragment, containing the dnaA gene and the flanking dnaA boxes, with a ColE1-derived Escherichia coli replicon and the Tet M determinant, which confers resistance to tetracycline. The recombinant plasmid, named pBOT1, was introduced into S. citri cells, in which it replicated. Plasmid pBOT1 was shuttled from E. coli to S. citri and back to E. coli. In S. citri, replication of pBOT1 did not require the presence of a functional dnaA gene on the plasmid. However, the dnaA box region downstream of the dnaA gene was essential. Upon passaging of the S. citri transformants, the plasmid integrated into the spiroplasmal host chromosome by recombination at the replication origin. The integration process led to duplication of the oriC sequences. In contrast to the integrative pBOT1, plasmid pOT1, which does not contain the E. coli replicon, was stably maintained as a free extrachromosomal element. Plasmid pOT1 was used as a vector to introduce into S. citri the G fragment of the cytadhesin P1 gene of Mycoplasma pneumoniae and the spiralin gene of Spiroplasma phoeniceum. The recombinant plasmids, pOTPG with the G fragment and pOTPS with the spiralin gene, were stably maintained in spiroplasmal transformants. Expression of the heterologous S. phoeniceum spiralin in S. citri was demonstrated by Western immunoblotting.     

Involvement of a Spiroplasma citri plasmid in the erythromycin-resistance transfer

An erythromycin-resistant strain (M4 Er-1) was selected from Spiroplasma citri M4+. The transfer by transformation of the erythromycin-resistance character to the erythromycin-sensitive S. citri strain R8A2+ was studied. Transfer became effective and reproducible when cells were treated with alkali cations plus polyethylene glycol. Comparison of the efficiency of transformation of the erythromycin-sensitive strain S. citri R8A2+ by total and extrachromosomal DNA purified from the erythromycin-resistant strain M4 Er-1 showed that the plasmid pM42 was able to transfer the erythromycin-resistance. pM42 was mapped with restriction endonucleases and found to be related to the pMH1 plasmid previously isolated from S. citri MH. Hybridization analysis of DNA from sensitive and resistant strains has shown that a sequence from pM42, analogous to a sequence from pMH1, was integrated at a specific locus in the chromosome of the erythromycin-resistant cells, i.e., of the transformed R8A2 cells and of the spontaneous mutant M4 Er-1 strain.     

Antigenic relatedness between the spiralins of Spiroplasma citri and Spiroplasma melliferum.

Four spiralins were compared by rocket immunoelectrophoresis, quantitative immunoblotting techniques, and the spiroplasma deformation test with the use of antispiralin (polyclonal) monospecific antibodies. This investigation revealed that the spiralins of Spiroplasma citri and S. melliferum are antigenically related and that probably no more than two epitopes simultaneously saturable with antibodies are shared by the two proteins. One at least of these epitopes is accessible to antibodies on the spiroplasma cell surface.     

Short, interspersed, and repetitive DNA sequences in Spiroplasma species

Small fragments of DNA from an 8-kbp plasmid, pRA1, from a plant pathogenic strain of Spiroplasma citri were shown previously to be present in the chromosomal DNA of at least two species of Spiroplasma. We describe here the shot-gun cloning of chromosomal DNA from S. citri Maroc and the identification of two distinct sequences exhibiting homology to pRA1. Further subcloning experiments provided specific molecular probes for the identification of these two sequences in chromosomal DNA from three distinct plant pathogenic species of Spiroplasma. The results of Southern blot hybridization indicated that each of the pRA1-associated sequences is present as multiple copies in short, dispersed, and repetitive sequences in the chromosomes of these three strains. None of the sequences was detectable in chromosomal DNA from an additional nine Spiroplasma strains examined.     

Molecular detection of Spiroplasma apis and Spiroplasma melliferum in bees

Spiroplasma apis and Spiroplasma melliferum are known as honey bee pathogens and are detected by unspecific methodologies like culturing or dark field microscopy. We developed a multiplex PCR being able to differentiate between both species and detect the genus Spiroplasma. This PCR can directly be used on culture samples or on DNA extracted bees. By PCR on cultured samples we were able to identify S. apis in Bombus pratorum and S. melliferum in Bombus pascuorum.     

Expression in Spiroplasma citri of an epitope carried on the G fragment of the cytadhesin P1 gene from Mycoplasma pneumoniae.

We have previously described the use of the replicative form (RF) of Spiroplasma citri virus SpV1 as a vector for cloning and expressing foreign genes in S. citri, an organism which reads UGA as a tryptophan codon (C. Stamburski, J. Renaudin, and J.M. Bové, J. Bacteriol. 173:2225-2230, 1991). We now report cloning and expression in S. citri of the G fragment of cytadhesin P1 gene from Mycoplasma pneumoniae. The G fragment was inserted in the SpV1 RF downstream of a synthetic ribosome binding site and introduced into S. citri by electroporation. Northern (RNA) blot analyses showed that in S. citri, the G fragment was transcribed from an SpV1 RF promoter as a 1.2-kb mRNA. The translation product was detected by Western blotting (immunoblotting) with a rabbit antiserum raised against total proteins from M. pneumoniae (strain FH) and was proved to be P1 specific by using monoclonal antibodies specific for the G region of the P1 protein. The apparent molecular mass of the polypeptide (24.5 kDa) indicates that in S. citri, the G fragment was fully translated in spite of the seven UGA codons present in the reading frame.     

Spiralin polymorphism in strains of Spiroplasma citri is not due to differences in posttranslational palmitoylation.

Spiralin is defined as the major membrane protein of the helical mollicute Spiroplasma citri. According to the S. citri strain used, spiralin shows polymorphism in its electrophoretic mobility. The spiralin gene sequences of eight S. citri strains were determined by direct sequencing of the PCR-amplified genes. All spiralins were found to be 241 amino acids long, except for the spiralin of strain Palmyre, which is 242 amino acids long. The molecular masses calculated from these sequences did not explain the differences observed in the electrophoretic mobilities. In all of the spiralins examined, the first 24 N-terminal amino acids were conserved, including a cysteine at position 24, and had the features of typical signal peptides of procaryotic lipoproteins. When S. citri strains were grown in the presence of [3H]palmitic acid, at least 10 proteins, including spiralin, became labeled. In the presence of globomycin, a lipoprotein signal peptidase inhibitor in eubacteria, apparently unprocessed spiralin could be detected. Formic acid hydrolysis of the [3H]palmitic acid-labeled spiralins of four representative S. citri strains yielded two peptide fragments for each spiralin, as expected from the gene sequence. On fragment was [3H]palmitic acid labeled, and it had almost the same electrophoretic mobility irrespective of the spiralins used. Samples of the unlabeled peptide fragments from the four representative strains had slightly different electrophoretic mobilities (delta Da approximately equal to 800 Da); however, these were much smaller than those of the whole spiralins before formic acid hydrolysis (delta Da approximately equal to 8,000 Da). These results suggest that spiralin polymorphism in S. citri is not due to differences in posttranslational modification by palmitic acid and is certainly a structural property of the whole protein or could result from an unidentified posttranslational modification of spiralin.     

Resistance of Spiroplasma citri Lines to the Virus SVTS2 Is Associated with Integration of Viral DNA Sequences into Host Chromosomal and Extrachromosomal DNA

Spiroplasmavirus SVTS2, isolated from Spiroplasma melliferum TS2, produces plaques when inoculated onto lawns of Spiroplasma citri M200H, a derivative of the type strain Maroc R8A2. S. citri strains MR2 and MR3, originally selected as colonies growing within plaques on a lawn of M200H inoculated with SVTS2, were resistant to SVTS2. Genomic DNA fingerprints and electrophoretic protein profiles of M200H, MR2, and MR3 were similar, but three proteins present in M200H were missing or significantly reduced in both resistant lines. None of these three polypeptides reacted with antiserum against S. citri membrane proteins, indicating that they probably are not surface-located virus receptors. Electroporation with SVTS2 DNA produced 1.5 x 10(sup5) transfectants per (mu)g of DNA in M200H but none in MR2 or MR3, suggesting that resistance may result from inhibition of viral replication. The digestion patterns of the extrachromosomal double-stranded (ds) DNA of these lines were similar. Three TaqI fragments of MR2 extrachromosomal DNA that were not present in M200H extrachromosomal DNA hybridized strongly to an SVTS2 probe, and two of these fragments plus an additional one hybridized with the MR3 extrachromosomal DNA, indicating that a fragment of SVTS2 DNA was present in the extrachromosomal ds DNA of MR2 and MR3 but not of M200H. When the restricted genomes of all three lines were probed with SVTS2 DNA, strong hybridization to two EcoRI fragments of chromosomal MR2 and MR3 DNA but not M200H DNA indicated that SVTS2 DNA had integrated into the genomes of MR2 and MR3 but not of M200H. When MR3 extrachromosomal ds DNA containing a 2.1-kb SVTS2 DNA fragment was transfected into M200H, the transformed spiroplasmas were resistant to SVTS2. These results suggest that SVTS2 DNA fragments, possibly integrated into the chromosomal or extrachromosomal DNA of a previously susceptible spiroplasma, may function as viral incompatibility elements, providing resistance to superinfection by SVTS2.     

Probable insensitivity of mollicutes to rifampin and characterization of spiroplasmal DNA-dependent RNA polymerase.

The effect of rifampin on five mollicutes (Spiroplasma citri, Spiroplasma melliferum, Spiroplasma apis, Acholeplasma laidlawii, and Mycoplasma mycoides) was compared with that on Escherichia coli. We found that, in contrast to wild-type E. coli, mollicutes were insensitive to rifampin. DNA-dependent RNA polymerases from S. melliferum and S. apis were purified to the stage where the enzymes were dependent on the addition of exogenous templates for activity. The enzymes were then tested for their sensitivity to rifampin. Spiroplasmal enzymes were at least 1,000 times less sensitive to rifampin than the corresponding E. coli enzyme. This result provides a molecular basis for the resistance of mollicutes to rifampin. The RNA polymerase of S. melliferum was further purified and its subunit composition was investigated. The RNA polymerase has one small and two large subunits. The structure of S. melliferum RNA polymerase therefore resembles that of the eubacterial enzymes in spite of its insensitivity to rifampin.     

Lethal and mutation frequency responses of Spiroplasma citri cells to UV irradiation

The effect of UV irradiation on viability and mutant colony frequency in the Mollicute Spiroplasma citri was investigated at 3 phases of growth. The first UV-induced mutants obtained in Mollicutes were selected: xylitol-resistant (XylR) and arsenic acid-resistant mutants (ArsR). Lethal and mutation frequency responses of S. citri cells increase with the age of the cell cultures. In all UV-irradiated populations, light exposure slightly increases the number of survivors and decreases the induced mutation frequency; liquid holding conditions increase the number of both survivors and mutant colonies. This suggests that, in UV-irradiated S. citri cells maintained under liquid holding conditions, there is no dark reactivation but induction of an error-prone repair system of the SOS type. In S. citri, the error-free light and dark repair systems are inefficient. Results allow the development of a method to select UV-induced mutations usable as markers in genetic studies of Spiroplasma cells.     

Gene disruption through homologous recombination in Spiroplasma citri: an scm1-disrupted motility mutant is pathogenic

To determine whether homologous recombination could be used to inactivate selected genes in Spiroplasma citri, plasmid constructs were designed to disrupt the motility gene scm1. An internal scm1 gene fragment was inserted into plasmid pKT1, which replicates in Escherichia coli but not in S. citri, and into the S. citri oriC plasmid pBOT1, which replicates in spiroplasma cells as well as in E. coli. Electrotransformation of S. citri with the nonreplicative, recombinant plasmid pKTM1 yielded no transformants. In contrast, spiroplasmal transformants were obtained with the replicative, pBOT1-derived plasmid pCJ32. During passaging of the transformants, the plasmid was found to integrate into the chromosome by homologous recombination either at the oriC region or at the scm1 gene. In the latter case, plasmid integration by a single crossover between the scm1 gene fragment carried by the plasmid and the full-length scm1 gene carried by the chromosome led to a nonmotile phenotype. Transmission of the scm1-disrupted mutant to periwinkle (Catharanthus roseus) plants through injection into the leafhopper vector (Circulifer haematoceps) showed that the motility mutant multiplied in the insects and was efficiently transmitted to plants, in which it induced symptoms similarly to the wild-type S. citri strain. These results suggest that the spiroplasmal motility may not be essential for pathogenicity and that, more broadly, the S. citri oriC plasmids can be considered promising tools for specific gene disruption by promoting homologous recombination in S. citri, a mollicute which probably lacks a functional RecA protein.     

A physical and genetic map of the Spiroplasma citri genome.

A physical and genetic map of the Spiroplasma citri genome has been constructed using several restriction enzymes and pulsed field gel electrophoresis. A number of genes were subsequently localized on the map by the use of appropriate probes. The genome size of the spiroplasma estimated from restriction fragments is close to 1780 kbp, the largest of all Mollicutes studied so far. It contains multisite insertions of Spiroplasma virus 1 (SpV1) sequences. The physical and genetic map of the S. citri genome shares several features with that of other Mollicutes, especially those in the Mycoplasma mycoides cluster. This supports the finding that S. citri and these Mycoplasma spp. are phylogenetically related.     

Isolation, characterization, and complementation of a motility mutant of Spiroplasma citri.

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

Lipid composition and lipid metabolism of Spiroplasma citri.

In a horse serum-based medium containing a full complement of fatty acids, cells of Spiroplasma citri were seen to preferentially incorporate palmitic acid. In the same medium, which had a steryl ester-to-sterol ratio of 3.64, a steryl ester-to-sterol ratio of 0.23 was seen in the cells, cholesterol being preferentially incorporated over cholesteryl ester. Like most other mycoplasmas, S. citri was shown to be unable to synthesize fatty acids or esterify cholesterol. The neutral lipids of S. citri grown in a medium containing horse serum consisted of free cholesterol, cholesteryl ester, free fatty acids, triglycerides and diglycerides. All polar lipids were phospholipids, with no glycolipids detected. These phospholipids, which are characteristic of many mycoplasmas, are phosphatidyl glycerol, diphosphatidyl glycerol, and their lyso derivatives. Sphingomyelin was also incorporated when cells were grown on horse serum. A sterol requirement for the growth of S. citri was confirmed using a serum-free medium supplemented with bovine serum albumin, palmitic acid, and various concentrations of sterols dissolved in Tween 80. The addition of palmitic acid stimulated growth but was not essential for growth. S citri was shown to grow best on cholesterol and beta-sitosterol and was able to grow on stigmasterol and ergosterol to a lesser degree. No growth was obtained using mevalonate, deoxycholate, or taurodeoxycholate as an alternative to sterol. S. citri was also able to grow when palmitic acid was replaced with oleic acid, linoleic acid, or linolenic acid. Alterations in the lipid composition of the growth medium and hence in the lipid composition of S. citri induced changes in the characteristic helical morphology of the cells, concurrent with loss of cell viability. Culture, age, and pH were also factors in determining cell morphology and viability.     

Comparison of the membrane composition of Spiroplasma citri and the corn stunt Spiroplasma.

Components of membranes isolated from Spiroplasma citri and corn stunt spiroplasma grown at 28 degrees C were analyzed. On a protein basis, lipid phosphorus was lower and cholesterol was higher in S. citri. Only minor differences between the two species were found in fatty acid composition, reduced nicotinamide adenine dinucleotide diaphorase, and adenosine triphosphatase.     

Sugar import and phytopathogenicity of Spiroplasma citri: glucose and fructose play distinct roles

We have shown previously that the glucose PTS (phosphotransferase system) permease enzyme II of Spiroplasma citri is split into two distinct polypeptides, which are encoded by two separate genes, crr and ptsG. A S. citri mutant was obtained by disruption of ptsG through homologous recombination and was proved unable to import glucose. The ptsG mutant (GII3-glc1) was transmitted to periwinkle (Catharanthus roseus) plants through injection to the leaf-hopper vector. In contrast to the previously characterized fructose operon mutant GMT 553, which was found virtually nonpathogenic, the ptsG mutant GII3-glc1 induced severe symptoms similar to those induced by the wild-type strain GII-3. These results, indicating that fructose and glucose utilization were not equally involved in pathogenicity, were consistent with biochemical data showing that, in the presence of both sugars, S. citri used fructose preferentially. Proton nuclear magnetic resonance analyses of carbohydrates in plant extracts revealed the accumulation of soluble sugars, particularly glucose, in plants infected by S. citri GII-3 or GII3-glc1 but not in those infected by GMT 553. From these data, a hypothetical model was proposed to establish the relationship between fructose utilization by the spiroplasmas present in the phloem sieve tubes and glucose accumulation in the leaves of S. citri infected plants.