Borrelia burgdorferi rel Is Responsible for Generation of Guanosine-3'-Diphosphate-5'-Triphosphate and Growth Control
http://www.100md.com
感染与免疫杂志 2005年第8期
Departments of Microbiology and Immunology Pathology, New York Medical College, Valhalla, New York 10595
ABSTRACT
The global transcriptional regulator (p)ppGpp (guanosine-3'-diphosphate-5'-triphosphate and guanosine-3',5'-bisphosphate, collectively) produced by the relA and spoT genes in Escherichia coli allows bacteria to adapt to different environmental stresses. The genome of Borrelia burgdorferi encodes a single chromosomal rel gene (BB0198) (B. burgdorferi rel [relBbu]) homologous to relA and spoT of E. coli. Its role in (p)ppGpp synthesis, bacterial growth, and modulation of gene expression has not been studied in detail. We constructed a relBbu deletion mutant in an infectious B. burgdorferi 297 strain and isolated an extrachromosomally complemented derivative of this mutant. The mutant did not synthesize relBbu mRNA, RelBbu protein, or (p)ppGpp. This synthesis was restored in the complemented derivative, confirming that relBbu is necessary and sufficient for (p)ppGpp synthesis and degradation in B. burgdorferi. The relBbu mutant grew well during log phase in complete BSK-H but reached lower cell concentrations in the stationary phase than the wild-type parent, suggesting that (p)ppGpp may be an important factor in the ability of B. burgdorferi to adapt to stationary phase. Deletion of relBbu did not eliminate the temperature-elicited OspC shift, nor did it alter bmp gene expression or B. burgdorferi antibiotic susceptibility. Although deletion of relBbu eliminated B. burgdorferi virulence for mice, which was not restored by complementation, we suggest that relBbu-dependent accumulation of (p)ppGpp may be important for in vivo survival of this pathogen.
INTRODUCTION
The stringent response is a regulatory response that allows bacteria to adapt to a lack of nutrients and other environmental stresses (6). It causes accumulation of guanosine-3'-diphosphate-5'-triphosphate (pppGpp) and guanosine-3',5'-bisphosphate (ppGpp), collectively referred to as (p)ppGpp or "magic spots." These nucleotides are synthesized by enzymatic phosphorylation of GDP and GTP to ppGpp and pppGpp, respectively, using ATP as a phosphate donor (6). In Escherichia coli, two different but highly homologous proteins are involved in (p)ppGpp synthesis: RelA, bound to ribosomes and activated by the presence of uncharged tRNA at the ribosomal A site which generally synthesizes (p)ppGpp in response to amino acid limitation (47), and SpoT, a cytosolic (p)ppGpp synthetase (15) which is responsive to changes in the availability of carbon, phosphate, and fatty acids as well as to changes in temperature and osmolarity (6, 34). SpoT is also a (p)ppGpp hydrolase (21, 31). Many gram-positive bacteria have only a single rel ortholog which exhibits both (p)ppGpp synthetase and hydrolase activity (29).
(p)ppGpp acts as a global transcriptional regulator. The general effect of the stringent response is a decrease in rRNA, tRNA, and protein synthesis and a decrease in growth rate that results in bacterial adaptation to an environment scarce in nutrients (6). (p)ppGpp also influences many other bacterial physiological functions including competence (23), morphological and physiological differentiation and production of clavulanic acid and cephamycin C (25, 26), production of actinorhodin and undecylprodigiosin antibiotics (46), thermotolerance (48), adaptation to oxidative stress (30), and sensitivity to antibiotics. As regards the latter, E. coli strains able to synthesize (p)ppGpp in either a RelA- or a SpoT-dependent manner show a greater resistance to antimicrobials than strains that cannot produce (p)ppGpp (17). In E. coli, production of (p)ppGpp is also required for the accumulation of inorganic polyphosphate needed for degradation of proteins during starvation mediated by the Lon protease (28), and lack of inorganic polyphosphate synthesis in enteric bacteria is accompanied by a reduction in virulence (27).
The relA and spoT gene products and (p)ppGpp mediate important aspects of virulence in a number of pathogens (8, 20, 34). For example, Vibrio cholerae relA is involved in the ability of V. cholerae to display pathogenicity in in vitro and in vivo models of infection (20), Mycobacterium tuberculosis rel-mediated adaptation to stationary phase is critical to long-term persistence of M. tuberculosis in mice (8), and mutations in Salmonella enterica serovar Typhimurium relA and spoT result in attenuation in animals (34). Not all rel mutants show decreased virulence. Mutation of Listeria monocytogenes relA, while accompanied by loss of (p)ppGpp synthesis and a decrease in osmotolerance, does not diminish in vivo virulence (32).
Borrelia burgdorferi is the etiologic agent of Lyme disease (44). During its life cycle, this spirochete thrives in both ticks and mammals and is able to adapt to various environmental conditions, including changes in temperature, pH, osmolarity, and nutrient availability (10, 41). Preliminary work by ourselves and others has indicated that the stringent response is likely to be present in B. burgdorferi (3, 4, 7). The B. burgdorferi genome contains a single chromosomal rel gene (BB0198) (B. burgdorferi rel [relBbu]) that shows 35 and 39% amino acid homology, respectively, to E. coli relA and spoT (13). Microarray studies have shown increased expression of relBbu mRNA in B. burgdorferi grown in BSK-H at pH 7.5 and 23°C (unfed tick conditions) compared to B. burgdorferi grown in BSK-H at pH 6.8 and 37°C (fed tick conditions) or in rat peritoneal chambers (37). (p)ppGpp synthesis by B. burgdorferi and the ability of relBbu to complement a relA-spoT E. coli double mutant for growth on minimal medium (3, 4, 7) provide additional support for the existence of a functional stringent response in B. burgdorferi mediated by the product of relBbu. We previously suggested that the stringent response in B. burgdorferi could modulate expression of bmpD, a lipoprotein gene jointly transcribed with the ribosomal protein genes rpsL and rpsG (3, 11). The discovery that B. burgdorferi RpoS (S) is responsible for the expression of OspC, the adhesin crucial for the ability of B. burgdorferi to infect the mammalian host (18, 33) and evidence from other bacterial systems that (p)ppGpp is a positive regulator of rpoS expression (16), strongly suggests that (p)ppGpp might modulate expression of lipoproteins involved in the B. burgdorferi life cycle and virulence (22).
To confirm that relBbu is necessary for (p)ppGpp production in B. burgdorferi and to clarify its role in B. burgdorferi physiology, relBbu was deleted in infectious B. burgdorferi 297. This deletion was complemented extrachromosomally, and the B. burgdorferi derivatives obtained were assayed for relBbu mRNA transcription, RelBbu protein synthesis, and the ability to produce (p)ppGpp. Growth characteristics, OspA and OspC protein levels, bmp mRNA levels, sensitivity to antibiotics, and virulence in mice were also compared between wild-type, mutant, and complemented strains. The relBbu gene was clearly responsible for (p)ppGpp synthesis. Its deletion affected B. burgdorferi growth in vitro and ablated in vivo virulence in mice. In contrast, deletion of relBbu did not alter OspA protein levels, bmp gene expression, or B. burgdorferi antibiotic susceptibility and did not eliminate the temperature-elicited OspC shift, suggesting that the role of (p)ppGpp in B. burgdorferi physiology is different from that in E. coli.
MATERIALS AND METHODS
Bacterial strains and culture conditions. Transformable clone BbAH130 of infectious B. burgdorferi 297 was kindly provided by M. V. Norgard, University of Texas Southwestern Medical Center. B. burgdorferi was maintained at 34°C in BSK-II (38) or BSK-H (Sigma-Aldrich, St. Louis, MO). Both media were supplemented with 6% rabbit serum (Sigma). BSK-II was used for preparing electrocompetent B. burgdorferi cells and for the selection of transformants. BSK-H was used to compare growth and levels of mRNA, protein, and (p)ppGpp in B. burgdorferi and its derivatives. The B. burgdorferi relBbu mutant was grown in the presence of kanamycin, 400 μg/ml (Sigma). Complemented B. burgdorferi was grown in the presence of kanamycin, 400 μg/ml, and streptomycin, 50 μg/ml (Sigma). E. coli DH5 (GIBCO/Life Technologies, Grand Island, NY) grown in the presence of spectinomycin, 100 μg/ml (Sigma), was used for propagation of the pKFSS1 and pKFSS1-relBbu plasmids.
Generation of relBbu deletion. Deletion of B. burgdorferi relBbu was done by homologous recombination with a relBbu inactivation construct (Fig. 1) using PCR-based fusions (43). In brief, nucleotides 195216 to 195852 (primers 1 and 2) (Table 1) and nucleotides 197417 to 198004 (primers 3 and 4) (Table 1) were amplified from B. burgdorferi chromosomal DNA, and the kanamycin resistance gene aph(3')-IIIa from Enterococcus faecalis with its own promoter was amplified from plasmid pBLS500 (primers III and IV) (Table 1). All PCR assays used an initial denaturation at 94°C for 2 min and 32 cycles of 94°C for 15 s, annealing at 56°C for 20 s, and extension at 68°C for 3 min. The three amplicons were fused (43), and the final PCR product containing the relBbu inactivation construction was precipitated with ethanol and resuspended in sterile water at 4 μg/μl. Forty micrograms was used for each transformation (38).
Construction of B. burgdorferi pKFSS1-relBbu. A PCR fragment containing full-size relBbu with 446 bp upstream from its start codon and 109 bp downstream from its stop codon was amplified from B. burgdorferi genomic DNA using forward primer ST1, containing a BamHI restriction site, and reverse primer ST2, containing a PstI restriction site (Table 1). After digestion with BamHI and PstI (New England Biolabs, Inc., Beverly, MA), the resulting fragment was ligated into pKFSS1 (12) digested with the same enzymes. The resulting plasmids were propagated in E. coli DH5 to obtain pKFSS1-relBbu encoding relBbu under its own promoter; 10 μg of this plasmid was electroporated into B. burgdorferi (38). Transformation of B. burgdorferi with pKFSS1 was used as a control.
Confirmation of relBbu inactivation and complementation in B. burgdorferi. DNA from B. burgdorferi transformants was purified (High Pure PCR template preparation kit; Roche Diagnostics Corporation, Indianapolis, IN) and analyzed by PCR (primers R/SpPROT1S and R/SpPROT1EndA) (Table 1 and Fig. 1) to confirm the presence of the relBbu deletion or complementation. The PCR fragment obtained with primers relBbu 6kbL and relBbu 6kbR (Table 1; Fig. 1) was sequenced using primers R/SpPROT1S and R/SpPROT1EndA (Table 1; Fig. 1) to provide further confirmation of the relBbu deletion (Herbert Irving Comprehensive Cancer Center, Columbia University, New York, NY). DNA from complemented B. burgdorferi strains was transformed into E. coli DH5. DNA from E. coli clones selected on spectinomycin plates was subjected to restriction analysis using BamHI and PstI for comparison with the plasmids originally used for electroporation of B. burgdorferi.
Estimation of pKFSS1-relBbu copy number. A competitive PCR assay was used to determine the pKFSS1-relBbu copy number in complemented B. burgdorferi 297. Competitors for flaB and relBbu genes containing internal deletions were constructed as previously described (3, 39). Serial twofold dilutions of competitors were mixed with fixed amounts of total 297 DNA to perform quantitative PCR for flaB and relBbu. Primers 49 and 50 (Table 1) were used for flaB detection; primers ST3 and ST4 (Table 1) were used for relBbu detection. Target DNA and competitor DNA were assumed to be present at equimolar concentrations in those reactions when the competitor and target PCR products were at similar intensities in agarose gels (11).
Determination of B. burgdorferi plasmid content. The plasmid content in B. burgdorferi relBbu mutants and complemented strains was compared with that of the wild-type parental strain using PCR as described previously (24).
RNA isolation, RT-PCR, and competitive RT-PCR. Total RNA from late-log-phase B. burgdorferi was isolated with TRizol reagent (Invitrogen Life Technology, Carlsbad, CA) according to the manufacturer's recommendations and was treated with RQ1 RNase-free DNase (Promega Corporation, Madison, WI) to eliminate DNA contamination. Reverse transcription (RT)-PCR was performed using the Access RT-PCR system (Promega) according to the manufacturer's recommendations (3, 11). Primers ST3 and ST4 (Table 1) were used for relBbu mRNA detection; primers 49 and 50 (Table 1) were used for detection of the constitutively expressed flaB gene as a control to show the presence of RNA in a sample (11). Primers BB0199RT_R and BB0199RT_F (Table 1) were used for detection of mRNA for the BB0199 gene. Primers and conditions for competitive RT-PCR for bmp genes have been described previously (3, 11).
Detection of B. burgdorferi proteins. Anti-RelBbu antibodies were raised in rabbits immunized with recombinant RelBbu protein (A. V. Bryksin, T. N. Orlova, and F. C. Cabello, unpublished data). Rabbit anti-FlaB antibodies were provided by J. D. Radolf, University of Connecticut, Farmington, CT; mouse monoclonal anti-OspA antibodies (H5332) were provided by A. G. Barbour, University of California, Irvine, CA; mouse monoclonal anti-OspC antibodies were provided by R. D. Gilmore, Centers for Disease Control and Prevention, Fort Collins, CO; and mouse monoclonal anti-RpoS antibodies (6A7-101-H11) were provided by M. V. Norgard, University of Texas Southwestern Medical Center, Dallas, TX. FlaB, OspA, OspC, RelBbu, and RpoS proteins were detected in B. burgdorferi lysates by Western blot analysis and the ECF Western blotting kit (Amersham Biosciences, Piscataway, NJ) and quantified using a Storm 860 PhosphorImager and ImageQuaNT software (Molecular Dynamics, Sunnyvale, CA).
Detection of (p)ppGpp. B. burgdorferi (5 x 106 cells/ml) was incubated in BSK-H at 34°C containing 10 μCi/ml of uniformly labeled [32P]orthophosphate (Amersham) for 2 days. Labeled cells were harvested from 10-ml cultures, and (p)ppGpp was extracted and chromatographed on cellulose polyethyleneimine thin-layer chromatography (TLC) plates (Selecto Scientific, Suwanee, GA.) (3). Plates were air dried, exposed to a phosphor screen (Molecular Dynamics) for 12 to 24 h, and scanned using a Storm 860 PhosphorImager.
B. burgdorferi growth assays. Wild-type, relBbu, and complemented B. burgdorferi strains (104 cells/ml) were grown at 34°C in BSK-H for 30 days. Cell numbers were determined by dark-field microscopy every other day during the first 2 weeks of growth. To measure the effect of cell death in B. burgdorferi stationary-phase cultures, the number of viable organisms present in each culture was determined by limiting dilution at day 30 and compared to the number of cells counted microscopically at day 14. Limiting dilution was performed using serial fourfold dilutions of 30-day-old cultures with fresh BSK-H in 96-well plates (180 μl/well; initial dilution, 1:4; final dilution, 1:16,777,216). The highest dilution yielding positive cultures was determined after two additional weeks of growth at 34°C, and the number of cells/ml in the culture on day 30 was calculated assuming that one cell was sufficient to produce cell growth in a well. The data are presented as the cell survival index (as a percentage) = viable cells/mlDay 30 (by limiting dilution)/Total cells/mlday 14 (by microscopy) x 100 and are presented as means ± standard deviations. Limiting dilution analysis and microscopy were done in duplicate in each of two independent experiments. Results were analyzed statistically using a one-way analysis of variance.
B. burgdorferi temperature shift. B. burgdorferi was grown at 34°C in BSK-H to 107 cells/ml, then inoculated at an initial concentration of 106 cells/ml, and grown at 23°C. After 1 week at 23°C, B. burgdorferi was transferred to fresh BSK-H at 23°C, 34°C, and 37°C at an initial concentration of 103 cells/ml. Cells were collected when their concentration reached 3 x 107 cells/ml for the wild-type parent and 107 cells/ml for relBbu. In other temperature shift experiments, B. burgdorferi was grown at 34°C as described above, then transferred to BSK-H at 37°C, and collected at early stationary phase when cell concentrations reached 108 cells/ml for the wild-type parent, 2 x 107 cells/ml for the relBbu mutant and the mutant complemented with the empty pKFSS1 vector, and 5 x 107 cells/ml for the mutant complemented with pKFSS1-relBbu.
Antibiotic sensitivity. MICs of ampicillin, ciprofloxacin, erythromycin, tetracycline, and vancomycin (Sigma) were determined for wild-type, relBbu, and complemented B. burgdorferi in BSK-H. B. burgdorferi (105 cells/ml) in BSK-H was added to serial threefold dilutions of each antibiotic (initial concentration, 600 μg/ml; final concentration, 0.01 μg/ml) or no antibiotic in 96-well plates. The plates were incubated for 2 weeks at 32°C in a humidified 3% CO2 atmosphere. The MIC was defined as the smallest antibiotic concentration preventing B. burgdorferi growth under these conditions.
Infection of mice with B. burgdorferi. Four groups of 4-week-old C3H/HeN mice (Charles River Laboratory, Wilmington, MA) were infected intradermally with B. burgdorferi 297 wild type (five mice), relBbu mutant (four mice), relBbu mutant transformed with pKFSS1 (four mice), and complemented relBbu mutant (four mice) using 104 cells/mouse. Two weeks after infection, the mice were sacrificed, and ear punches and blood samples (100 μl) were taken and cultured for 2 weeks at 34°C in 1.5 ml of BSK-H containing 50 μg/ml of rifampin (Merrell Dow Pharmaceuticals, Inc., Cincinnati, OH) and 2.5 μg/ml of amphotericin B (Sigma) before being microscopically examined for growth of B. burgdorferi.
RESULTS
Isolation of a relBbu mutant in B. burgdorferi. Construction of the relBbu deletion is shown in Fig. 1. A PCR fragment containing the relBbu flanking regions surrounding a kanamycin resistance gene was electroporated into infectious B. burgdorferi 297. This electroporation yielded 20 kanamycin-resistant clones with a deleted relBbu gene. Of the 2,004 nucleotides of the wild-type relBbu gene, the mutants had 160 nucleotides at the 5' terminus and 280 nucleotides at the 3' terminus of relBbu, with the central 1,564 nucleotides deleted and replaced by the kanamycin resistance gene. The deletion in the relBbu gene was confirmed by PCR analysis with primers R/SpPROT1S and R/SpPROT1EndA (Fig. 1) which showed that relBbu produced a shorter PCR fragment than the fragment generated by the same primers in the wild-type B. burgdorferi (Fig. 2A). Only 2 of the 20 mutants had all the plasmids of parental B. burgdorferi 297. PCR revealed that the parental and these two relBbu B. burgdorferi mutants lacked the lp56, lp38, and cp32-8 plasmids but otherwise contained the other expected B. burgdorferi plasmids (data not shown) (24). Chromosomal DNA fragments in these two relBbu mutants were PCR amplified with primers relBbu6kbL and relBbu6kbR (Fig. 1), and the deletion of relBbu was confirmed by DNA sequencing. Both mutants were fully characterized genetically and functionally. Identical results were obtained with both; only the data generated from one of them is shown below. Deletion of the relBbu gene did not prevent mRNA expression of the downstream BB0199 gene (Fig. 1). Its RNA could be detected by RT-PCR in wild-type and relBbu B. burgdorferi (Fig. 2B), showing the absence of a polar effect of the relBbu mutation.
Complementation of relBbu mutant. To complement the relBbu deletion, the relBbu strain was electroporated with pKFSS1-relBbu carrying the relBbu gene under its own promoter or with the empty pKFSS1 cloning vector as a control. The presence of the wild-type and deleted relBbu alleles was determined by PCR analysis of the complemented B. burgdorferi derivatives using primers R/SpPROT1S and R/SpPROT1EndA (Fig. 1). The mutant complemented with pKFSS1-relBbu produced both mutant- and wild-type-sized amplicons, while the strain transformed with the pKFSS1 cloning vector produced only the mutant-sized amplicon (Fig. 2A). To demonstrate that the plasmids used for complementation were not modified in B. burgdorferi, DNA from complemented B. burgdorferi derivatives was transformed into E. coli. Restriction analysis confirmed that plasmid DNA from spectinomycin-resistant E. coli clones and plasmid DNA originally used for complementation were identical (Fig. 2C).
Detection of relBbu mRNA and RelBbu protein. RT-PCR analysis showed that relBbu mRNA was produced in wild-type B. burgdorferi grown in BSK-H at 34°C, while the relBbu mutant had no relBbu mRNA synthesis (Fig. 3). Transformation of relBbu with the pKFSS1-relBbu plasmid but not with the pKFSS1 cloning vector restored relBbu mRNA synthesis in the mutant (Fig. 3).
Western blot analysis with rabbit anti-rRelBbu protein detected a 78-kDa band corresponding to RelBbu protein in lysates of the wild-type B. burgdorferi grown in BSK-H at 34°C (Fig. 4B). This band was absent in lysates of relBbu (Fig. 4B) and reappeared in lysates of relBbu transformed with pKFSS1-relBbu. Transformation with the pKFSS1 vector alone did not restore RelBbu synthesis in relBbu (Fig. 4B). The intensity of the RelBbu band detected in the complemented derivative with pKFSS1-relBbu was stronger than that detected in wild-type B. burgdorferi (Fig. 4B). Image analysis of six independent Western blots indicated that the intensity of this band was threefold greater in the complemented relBbu mutant than in wild-type B. burgdorferi, suggesting increased synthesis of RelBbu protein in complemented B. burgdorferi (Fig. 4C).
To determine whether the increased intensity of the RelBbu band was merely a consequence of the pKFSS1-relBbu copy number, the numbers of relBbu genes in wild-type and complemented B. burgdorferi were compared by competitive PCR (39). Competitive PCR for flaB was used as an internal control for copy number, since both strains have only a single flaB gene. Wild-type and complemented B. burgdorferi had similar numbers of relBbu copies (Fig. 2D), and since wild-type B. burgdorferi has only one relBbu gene, the copy number of the pKFSS1-relBbu plasmid was also one per B. burgdorferi cell.
(p)ppGpp synthesis. Both ppGpp and pppGpp could be easily detected in thin-layer chromatograms in wild-type B. burgdorferi 297 grown at 34°C in BSK-H. Spots corresponding to ppGpp or pppGpp were not detected in the relBbu mutant (Fig. 5A and B), while (p)ppGpp synthesis was restored in relBbu complemented with pKFSS1-relBbu. There was no ppGpp nor pppGpp in the mutant transformed with the empty pKFSS1 vector (Fig. 5A and B). (p)ppGpp levels, as indicated by the intensity of the spots, appeared to be the same in wild-type and complemented B. burgdorferi despite the different amounts of RelBbu protein in these two strains. These results confirm that relBbu is involved in (p)ppGpp synthesis in B. burgdorferi.
Growth in culture. In a variety of bacteria, including E. coli, (p)ppGpp levels affect cell growth (6, 26, 32). We therefore compared the growth of the wild type, the relBbu mutant, and the complemented mutant at 34°C in BSK-H (Fig. 6). Figure 6A shows that there was little difference in cell numbers between the wild type and the relBbu mutant during the first 8 days of culture. The average doubling time was 5.2 h for wild-type B. burgdorferi 297 and 7.8 h for the relBbu mutant during this time, but this difference was not statistically significant (analysis of variance, P > 0.05). However, at 10 to 14 days of culture, when the cultured cells were approaching stationary phase, cell densities in the wild-type cultures were approximately 10-fold greater than those in relBbu, and this difference was significant (analysis of variance, P < 0.01). Complementation of the relBbu mutant with pKFSS1-relBbu increased stationary-phase cell densities significantly, about fourfold, at 10 to 14 days, but a significant twofold difference in cell densities between complemented and wild-type cell remained (analysis of variance, P < 0.01). There was no difference in cell growth between relBbu and relBbu transformed with pKFSS1 vector alone.
To determine whether cell death was responsible for the lower cell concentrations of the relBbu mutant cultures in the stationary phase, wild-type, mutant, complemented mutant, and vector-transformed cultures were grown for 30 days and the numbers of viable organisms remaining in these cultures at that time were determined by limiting dilution. After a further 2 weeks of growth, the highest dilution that gave positive cultures for each B. burgdorferi strain was used to calculate a survival index which took into account the differences in cell concentrations reached by each strain on day 14 (Fig. 6B). The numbers of surviving cells at 30 days of culture in the wild-type, mutant, and complemented strains were not significantly different (analysis of variance, P = 0.7) and constituted approximately 5 to 9 cells per 10,000 cells observed microscopically at 14 days of culture. These experiments showed that the relBbu mutation did not lead to increased cell death in stationary-phase cultures but rather decreased growth abilities prior to this point during the late exponential growth phase.
Expression of bmp genes. bmpD is transcribed together with the ribosomal protein genes rpsL and rpsG, and its expression might be modulated by the stringent response (3, 11). Measurement of bmp mRNA by competitive RT-PCR showed no differences in transcription of bmpA, bmpB, bmpC, and bmpD genes between the wild type and the relBbu mutant of B. burgdorferi (data not shown), indicating that bmp gene transcription is not modulated by (p)ppGpp.
Effects of temperature shift on OspC production. B. burgdorferi increases OspC levels when B. burgdorferi is transferred from 23°C to 34 to 37°C conditions of growth (42, 45). RpoS has been shown to be responsible both for this increase in OspC levels with a shift in temperature and for activation of transcription from the ospC promoter at elevated temperatures (22). Because (p)ppGpp is a positive regulator of RpoS in E. coli (16), ablation of (p)ppGpp synthesis in the relBbu mutant might alter the expression pattern of OspC. OspC protein was produced at 34°C and 37°C and was essentially absent at 23°C in mid-log-phase wild-type and relBbu cultures grown in vitro in BSK-H (Fig. 7A). However, OspC levels were at least twofold lower in the relBbu mutant than in wild-type B. burgdorferi grown at elevated temperatures (Fig. 7A). To determine whether (p)ppGpp was needed for high-level OspC expression, OspC levels in wild-type, relBbu, and complemented B. burgdorferi were determined in organisms grown at 37°C to early stationary phase, where the difference in growth between the wild type and the relBbu mutant might further increase any possible effect of (p)ppGpp on OspC expression. Stationary-phase OspC levels were higher in wild-type B. burgdorferi grown at 37°C than in the relBbu mutant (Fig. 7B), confirming the results obtained with mid-log-phase cultures, but complementation of the relBbu mutant did not raise OspC levels, suggesting that (p)ppGpp is not involved in regulation of OspC synthesis. RpoS was detected in the immunoblots of the wild-type and relBbu mutant B. burgdorferi growing at elevated temperature but not at 23°C (data not shown), indicating that RpoS expression does not require (p)ppGpp.
Antibiotic susceptibility of wild-type, relBbu mutant, and complemented B. burgdorferi. In E. coli, (p)ppGpp synthesis is required for resistance to several antibiotics (17). We therefore examined whether the absence of (p)ppGpp in the relBbu mutant modified B. burgdorferi antibiotic susceptibility. The MICs of ampicillin, erythromycin, tetracycline, and vancomycin were similar for wild-type B. burgdorferi and the relBbu mutant (Table 2). The MIC of ciprofloxacin was three times higher for the wild-type strain than for the relBbu mutant, but the MIC of ciprofloxacin for the complemented derivative was identical to that of the relBbu mutant. Since complementation of the relBbu mutant did not restore the ciprofloxacin MIC to the wild-type level, we concluded that (p)ppGpp was not involved in the change in the ciprofloxacin susceptibility of the mutant.
Infection of mice with wild-type, relBbu, and complemented B. burgdorferi. C3H/HeN mice were infected intradermally with the B. burgdorferi 297 wild type, relBbu mutant, relBbu mutant transformed with pKFSS1, or relBbu mutant complemented with pKFSS1-relBbu. All the strains used for infection had been shown to carry all plasmids of the parental B. burgdorferi 297 (data not shown). Two weeks after infection, B. burgdorferi could be cultured only from ear punches and blood samples of mice injected with wild-type B. burgdorferi (four of five positive ear punch cultures and five of five positive blood cultures). B. burgdorferi could not be cultured from ear punches and blood of any of the mice receiving relBbu or its derivatives transformed with pKFSS1 or pKFSS1-relBbu.
DISCUSSION
relA and spoT homologues are responsible for synthesis of (p)ppGpp, the alarmone involved in the bacterial stringent response (6). We have shown that relBbu is the only gene involved in generation of (p)ppGpp in B. burgdorferi. Deletion of relBbu did not eliminate mRNA synthesis from the BB0199 downstream gene of unknown function but completely abolished the production of relBbu mRNA, RelBbu protein, and (p)ppGpp in infectious B. burgdorferi 297. Synthesis of these molecules during growth of B. burgdorferi in BSK-H was restored after introduction of an extrachromosomal copy of the wild-type allele of relBbu into the mutant. Our findings are not inconsistent with the possible existence of different relBbu regulators that may affect (p)ppGpp production under certain conditions. The apparently increased amounts of RelBbu protein in the complemented B. burgdorferi strain (Fig. 4) could not be explained by the different strengths of the relBbu gene promoter in the chromosome and the plasmid, since relBbu was cloned into pKFSS1 under its own promoter (located about 300 nucleotides upstream from the relBbu ATG translational start) (4). These increased amounts of RelBbu could also not be a result of multiple copies of the introduced plasmid, since we found only a single copy of the vector per complemented B. burgdorferi genome. The possibility exists that differences in DNA supercoiling for the relBbu gene located on the pKFSS1-relBbu plasmid and on the B. burgdorferi chromosome might ultimately be responsible for the different amounts of RelBbu protein observed (1, 9). The increased amounts of RelBbu did not increase (p)ppGpp levels in complemented B. burgdorferi (Fig. 5), most probably because RelBbu has two activities, (p)ppGpp synthetic and hydrolytic (6), and both increase simultaneously with increased amounts of protein. Alternatively, the increase in the production of RelBbu might not be accompanied by increased levels of (p)ppGpp because the excess protein might be inactively bound to the ribosomes (47).
The relBbu gene and (p)ppGpp are not essential for in vitro growth of B. burgdorferi, since the relBbu mutant was able to multiply in BSK-H (Fig. 6). B. burgdorferi thus differs from Staphylococcus aureus, since rel mutants cannot be isolated in S. aureus because the relSau gene and (p)ppGpp are essential in this species (14). Nevertheless, the relBbu mutant did show an important growth defect because it only achieved 1/10 the cell concentration of wild-type B. burgdorferi in the stationary phase in vitro. It is not clear whether this growth defect resulted from a decrease in growth during the exponential phase or is a result of the inability of this mutant to thrive during the stationary phase, since the mutant does not display increased cell death during the stationary phase (Fig. 6). The growth deficit of B. burgdorferi lacking (p)ppGpp is quite different from that of E. coli lacking (p)ppGpp. In E. coli, experimental evidence indicates that (p)ppGpp can suppress expression of the ftsZ gene, an essential gene for cell division and septation (35), and the level of (p)ppGpp is inversely correlated with the rate of initiation of new rounds of DNA replication (40).
We previously reported that lack of (p)ppGpp synthesis in B. burgdorferi grown in the presence of tick cells was associated with changes in the transcription of bmpD, a gene transcribed with the ribosomal protein genes rpsL and rpsG, compared to that seen in B. burgdorferi grown in BSK-H (3, 11). Unexpectedly, we found no alteration of bmp gene expression in the relBbu mutant compared to wild-type B. burgdorferi, indicating that perhaps the changes seen in the expression of these genes during B. burgdorferi growth in the presence of tick cells were more the result of their interaction with the tick cells than of the disappearance of (p)ppGpp (3). More extensive experiments to examine bmp gene modulation in the relBbu mutant and its complemented derivatives growing in the presence of tick cells will be needed to clarify these differences.
It is well established that (p)ppGpp can regulate gene expression in E. coli through a positive control cascade mediated by the alternative bacterial factor RpoS (6). RpoS is involved in the OspC temperature-mediated shift in B. burgdorferi (22), which has been shown to be important for transmission of B. burgdorferi from ticks to mammals (18, 33). Although OspC expression decreased in B. burgdorferi relBbu, the OspC temperature shift was not eliminated by the lack of (p)ppGpp. This suggests that (p)ppGpp does not modulate RpoS functions in B. burgdorferi. Detection of RpoS in immunoblots in the wild-type 297 and relBbu strains (data not shown) also confirmed that RpoS expression does not require (p)ppGpp. The absence of modulation of RpoS-dependent genes by (p)ppGpp in B. burgdorferi is surprising, since there is a homologue of dksA in the B. burgdorferi genome (BB0168) (13) whose gene product mediates RpoS induction by (p)ppGpp in E. coli (2). An absence of modulation of RpoS levels by (p)ppGpp has also been found in S. enterica serovar Typhimurium (34), where (p)ppGpp can modulate expression of virulence-associated genes by an RpoS-independent pathway (34). A recent report has proposed that there are RpoS-independent virulence regulons in B. burgdorferi (5) and that (p)ppGpp may have a role in their regulation, but this requires further study.
(p)ppGpp mediates resistance of E. coli to various antibiotics by lowering bacterial metabolism and growth rate as well as by RpoS activation (17, 19, 36). Our experiments indicate that the MICs of several classes of antibiotics are similar in B. burgdorferi in the presence and absence of (p)ppGpp (Table 2). The absence of this phenotype could be explained by the fact that, in contrast to what occurs in E. coli, the lack of (p)ppGpp did not change the B. burgdorferi growth rate substantially and did not affect the synthesis of RpoS, two elements involved in (p)ppGpp-mediated antibiotic resistance in E. coli. The decreased MIC of ciprofloxacin observed in the relBbu mutant and its complemented derivative could be the result of a spontaneous mutation. Alternatively, the difference in cell densities between complemented and wild-type B. burgdorferi strains might be responsible for the increased susceptibility to ciprofloxacin in the complemented strain.
Since mutation of rel has been shown to affect virulence in several bacterial pathogens (8, 20, 34), we compared the ability of wild-type, relBbu, and complemented B. burgdorferi to infect mice after needle inoculation. The loss of infectivity for relBbu suggested that relBbu is necessary for B. burgdorferi to produce infection in mice. Since complementation, even in the presence of all parental B. burgdorferi plasmids, did not restore infectivity in mice, there is a possibility of a defect that cannot be determined by the present techniques. This suggests that other complementation approaches (for example, intrachromosomal) might be needed to confirm that (p)ppGpp is involved in B. burgdorferi virulence. The decreased levels of OspC, a major lipoprotein necessary for infection of mammalian hosts (18, 33), might partially explain the loss of infectivity in the relBbu mutant, although complementation restored neither OspC levels nor infectivity. Increased RelBbu levels in complemented B. burgdorferi, even though they did not lead to increased (p)ppGpp amounts, might be responsible for the failure of the complementation to restore infectivity in mice. Another explanation for the failure of complementation to restore the infectiousness of the relBbu strain could be loss of the complementing plasmid harboring the relBbu wild-type gene in mice (12).
In summary, we have shown that relBbu is responsible for (p)ppGpp synthesis and affects the growth and probably the infectivity of B. burgdorferi. Our work also provides evidence that the absence of (p)ppGpp in B. burgdorferi was not accompanied by the same phenotypes as in E. coli. This suggests that the role of this mediator in B. burgdorferi physiology is different from its role in E. coli. Our results indicate that (p)ppGpp does not regulate RpoS-dependent promoters and does not influence susceptibility to antibiotics in B. burgdorferi. Further studies will be needed to ascertain the role of (p)ppGpp in B. burgdorferi and to define the stringent response in this pathogen.
ACKNOWLEDGMENTS
This work was supported by a grant to F.C.C. (R01 AI41-056-7).
We thank M. Cashel, National Institutes of Health, Bethesda, MD, for advice and discussions; M. V. Norgard, University of Texas Southwestern Medical Center, Dallas, TX, for B. burgdorferi 297, clone BbAH130, and anti-RpoS antibodies; D. S. Samuels, University of Montana, Missoula, MT, for pKFSS1; J. D. Radolf, C. H. Eggers, and M. J. Caimano, University of Connecticut, Farmington, CT, for anti-FlaB antibodies and for extensive discussions; A. G. Barbour, University of California, Irvine, CA, for anti-OspA antibodies (H5332); and R. D. Gilmore, CDC, Fort Collins, CO, for anti-OspC antibodies. We also thank Radha Iyer and Darya Terekhova, New York Medical College, Valhalla, NY, for help with analysis of B. burgdorferi plasmid composition.
REFERENCES
1. Alverson, J., S. F. Bundle, C. D. Sohaskey, M. C. Lybecker, and D. S. Samuels. 2003. Transcriptional regulation of the ospAB and ospC promoters from Borrelia burgdorferi. Mol. Microbiol. 48:1665-1677.
2. Brown, L., D. Gentry, T. Elliott, and M. Cashel. 2002. DksA affects ppGpp induction of RpoS at a translational level. J. Bacteriol. 184:4455-4465.
3. Bugrysheva, J., E. Y. Dobrikova, H. P. Godfrey, M. L. Sartakova, and F. C. Cabello. 2002. Modulation of Borrelia burgdorferi stringent response and gene expression during extracellular growth with tick cells. Infect. Immun. 70:3061-3067.
4. Bugrysheva, J., E. Y. Dobrikova, M. L. Sartakova, M. J. Caimano, T. J. Daniels, J. D. Radolf, H. P. Godfrey, and F. C. Cabello. 2003. Characterization of the stringent response and relBbu expression in Borrelia burgdorferi. J. Bacteriol. 185:957-965.
5. Caimano, M. J., C. H. Eggers, K. R. Hazlett, and J. D. Radolf. 2004. RpoS is not central to the general stress response in Borrelia burgdorferi but does control expression of one or more essential virulence determinants. Infect. Immun. 72:6433-6445.
6. Cashel, M., D. R. Gentry, V. J. Hernandez, and D. Vinella. 1996. The stringent response, p. 1458-1496. In F. C. Neidhardt, R. I. Curtiss, J. L. Ingraham, E. C. C. Lin, K. B. Low, B. Magasanik, W. S. Reznikoff, M. Riley, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella: cellular and molecular biology, 2nd ed. ASM Press, Washington, D.C.
7. Concepcion, M. B., and D. R. Nelson. 2003. Expression of spoT in Borrelia burgdorferi during serum starvation. J. Bacteriol. 185:444-452.
8. Dahl, J. L., C. N. Kraus, H. I. Boshoff, B. Doan, K. Foley, D. Avarbock, G. Kaplan, V. Mizrahi, H. Rubin, and C. E. Barry III. 2003. The role of RelMtb-mediated adaptation to stationary phase in long-term persistence of Mycobacterium tuberculosis in mice. Proc. Natl. Acad. Sci. USA 100:10026-10031.
9. Deng, S., R. A. Stein, and N. P. Higgins. 2004. Transcription-induced barriers to supercoil diffusion in the Salmonella typhimurium chromosome. Proc. Natl. Acad. Sci. USA 101:3398-3403.
10. De Silva, A. M., and E. Fikrig. 1997. Arthropod- and host-specific gene expression by Borrelia burgdorferi. J. Clin. Investig. 99:377-379.
11. Dobrikova, E. Y., J. Bugrysheva, and F. C. Cabello. 2001. Two independent transcriptional units control the complex and simultaneous expression of the bmp paralogous chromosomal gene family in Borrelia burgdorferi. Mol. Microbiol. 39:370-378.
12. Frank, K. L., S. F. Bundle, M. E. Kresge, C. H. Eggers, and D. S. Samuels. 2003. aadA confers streptomycin resistance in Borrelia burgdorferi. J. Bacteriol. 185:6723-6727.
13. Fraser, C. M., S. Casjens, W. M. Huang, G. G. Sutton, R. Clayton, R. Lathigra, O. White, K. A. Ketchum, R. Dodson, E. K. Hickey, M. Gwinn, B. Dougherty, J. F. Tomb, R. D. Fleischmann, D. Richardson, J. Peterson, A. R. Kerlavage, J. Quackenbush, S. Salzberg, M. Hanson, R. van Vugt, N. Palmer, M. D. Adams, J. Gocayne, and J. C. Venter. 1997. Genomic sequence of a Lyme disease spirochaete, Borrelia burgdorferi. Nature 390:580-586.
14. Gentry, D., T. Li, M. Rosenberg, and D. McDevitt. 2000. The rel gene is essential for in vitro growth of Staphylococcus aureus. J. Bacteriol. 182:4995-4997.
15. Gentry, D. R., and M. Cashel. 1995. Cellular localization of the Escherichia coli SpoT protein. J. Bacteriol. 177:3890-3893.
16. Gentry, D. R., V. J. Hernandez, L. H. Nguyen, D. B. Jensen, and M. Cashel. 1993. Synthesis of the stationary-phase sigma factor s is positively regulated by ppGpp. J. Bacteriol. 175:7982-7989.
17. Greenway, D. L., and R. R. England. 1999. The intrinsic resistance of Escherichia coli to various antimicrobial agents requires ppGpp and s. Lett. Appl. Microbiol. 29:323-326.
18. Grimm, D., K. Tilly, R. Byram, P. E. Stewart, J. G. Krum, D. M. Bueschel, T. G. Schwan, P. F. Policastro, A. F. Elias, and P. A. Rosa. 2004. Outer-surface protein C of the Lyme disease spirochete: a protein induced in ticks for infection of mammals. Proc. Natl. Acad. Sci. USA 101:3142-3147.
19. Gustafson, C. E., S. Kaul, and E. E. Ishiguro. 1993. Identification of the Escherichia coli lytB gene, which is involved in penicillin tolerance and control of the stringent response. J. Bacteriol. 175:1203-1205.
20. Haralalka, S., S. Nandi, and R. K. Bhadra. 2003. Mutation in the relA gene of Vibrio cholerae affects in vitro and in vivo expression of virulence factors. J. Bacteriol. 185:4672-4682.
21. Hernandez, V. J., and H. Bremer. 1991. Escherichia coli ppGpp synthetase II activity requires spoT. J. Biol. Chem. 266:5991-5999.
22. Hübner, A., X. Yang, D. M. Nolen, T. G. Popova, F. C. Cabello, and M. V. Norgard. 2001. Expression of Borrelia burgdorferi OspC and DbpA is controlled by a RpoN-RpoS regulatory pathway. Proc. Natl. Acad. Sci. USA 98:12724-12729.
23. Inaoka, T., and K. Ochi. 2002. RelA protein is involved in induction of genetic competence in certain Bacillus subtilis strains by moderating the level of intracellular GTP. J. Bacteriol. 184:3923-3930.
24. Iyer, R., O. Kalu, J. Purser, S. Norris, B. Stevenson, and I. Schwartz. 2003. Linear and circular plasmid content in Borrelia burgdorferi clinical isolates. Infect. Immun. 71:3699-3706.
25. Jin, W., H. K. Kim, J. Y. Kim, S. G. Kang, S. H. Lee, and K. J. Lee. 2004. Cephamycin C production is regulated by relA and rsh genes in Streptomyces clavuligerus ATCC27064. J. Biotechnol. 114:81-87.
26. Jin, W., Y. G. Ryu, S. G. Kang, S. K. Kim, N. Saito, K. Ochi, S. H. Lee, and K. J. Lee. 2004. Two relA/spoT homologous genes are involved in the morphological and physiological differentiation of Streptomyces clavuligerus. Microbiology 150:1485-1493.
27. Kim, K. S., N. N. Rao, C. D. Fraley, and A. Kornberg. 2002. Inorganic polyphosphate is essential for long-term survival and virulence factors in Shigella and Salmonella spp. Proc. Natl. Acad. Sci. USA 99:7675-7680.
28. Kuroda, A., K. Nomura, R. Ohtomo, J. Kato, T. Ikeda, N. Takiguchi, H. Ohtake, and A. Kornberg. 2001. Role of inorganic polyphosphate in promoting ribosomal protein degradation by the Lon protease in E. coli. Science 293:705-708.
29. Mittenhuber, G. 2001. Comparative genomics and evolution of genes encoding bacterial (p)ppGpp synthetases/hydrolases (the Rel, RelA and SpoT proteins). J. Mol. Microbiol. Biotechnol. 3:585-603.
30. Mostertz, J., C. Scharf, M. Hecker, and G. Homuth. 2004. Transcriptome and proteome analysis of Bacillus subtilis gene expression in response to superoxide and peroxide stress. Microbiology 150:497-512.
31. Murray, K. D., and H. Bremer. 1996. Control of spoT-dependent ppGpp synthesis and degradation in Escherichia coli. J. Mol. Biol. 259:41-57.
32. Okada, Y., S. Makino, T. Tobe, N. Okada, and S. Yamazaki. 2002. Cloning of rel from Listeria monocytogenes as an osmotolerance involvement gene. Appl. Environ. Microbiol. 68:1541-1547.
33. Pal, U., X. Yang, M. Chen, L. K. Bockenstedt, J. F. Anderson, R. A. Flavell, M. V. Norgard, and E. Fikrig. 2004. OspC facilitates Borrelia burgdorferi invasion of Ixodes scapularis salivary glands. J. Clin. Investig. 113:220-230.
34. Pizarro-Cerda, J., and K. Tedin. 2004. The bacterial signal molecule, ppGpp, regulates Salmonella virulence gene expression. Mol. Microbiol. 52:1827-1844.
35. Powell, B. S., and D. L. Court. 1998. Control of ftsZ expression, cell division, and glutamine metabolism in Luria-Bertani medium by the alarmone ppGpp in Escherichia coli. J. Bacteriol. 180:1053-1062.
36. Rand, J. D., S. G. Danby, D. L. Greenway, and R. R. England. 2002. Increased expression of the multidrug efflux genes acrAB occurs during slow growth of Escherichia coli. FEMS Microbiol. Lett. 207:91-95.
37. Revel, A. T., A. M. Talaat, and M. V. Norgard. 2002. DNA microarray analysis of differential gene expression in Borrelial burgdorferi, the Lyme disease spirochete. Proc. Natl. Acad. Sci. USA 99:1562-1567.
38. Samuels, D. S. 1995. Electrotransformation of the spirochete Borrelia burgdorferi. Methods Mol. Biol. 47:253-259.
39. Sartakova, M., E. Dobrikova, and F. C. Cabello. 2000. Development of an extrachromosomal cloning vector system for use in Borrelia burgdorferi. Proc. Natl. Acad. Sci. USA 97:4850-4855.
40. Schreiber, G., E. Z. Ron, and G. Glaser. 1995. ppGpp-mediated regulation of DNA replication and cell division in Escherichia coli. Curr. Microbiol. 30:27-32.
41. Schwan, T. G., and J. Piesman. 2002. Vector interactions and molecular adaptations of Lyme disease and relapsing fever spirochetes associated with transmission by ticks. Emerg. Infect. Dis. 8:115-121.
42. Schwan, T. G., J. Piesman, W. T. Golde, M. C. Dolan, and P. A. Rosa. 1995. Induction of an outer surface protein on Borrelia burgdorferi during tick feeding. Proc. Natl. Acad. Sci. USA 92:2909-2913.
43. Shevchuk, N. A., A. V. Bryksin, Y. A. Nusinovich, F. C. Cabello, M. Sutherland, and S. Ladisch. 2004. Construction of long DNA molecules using long PCR-based fusion of several fragments simultaneously. Nucleic Acids Res. 32:e19.
44. Steere, A. C., J. Coburn, and L. Glickstein. 2004. The emergence of Lyme disease. J. Clin. Investig. 113:1093-1101.
45. Stevenson, B., T. G. Schwan, and P. A. Rosa. 1995. Temperature-related differential expression of antigens in the Lyme disease spirochete, Borrelia burgdorferi. Infect. Immun. 63:4535-4539.
46. Sun, J., A. Hesketh, and M. Bibb. 2001. Functional analysis of relA and rshA, two relA/spoT homologues of Streptomyces coelicolor A3(2). J. Bacteriol. 183:3488-3498.
47. Wendrich, T. M., G. Blaha, D. N. Wilson, M. A. Marahiel, and K. H. Nierhaus. 2002. Dissection of the mechanism for the stringent factor RelA. Mol. Cell 10:779-788.
48. Yang, X., and E. E. Ishiguro. 2003. Temperature-sensitive growth and decreased thermotolerance associated with relA mutations in Escherichia coli. J. Bacteriol. 185:5765-5771.(Julia V. Bugrysheva, Anto)
ABSTRACT
The global transcriptional regulator (p)ppGpp (guanosine-3'-diphosphate-5'-triphosphate and guanosine-3',5'-bisphosphate, collectively) produced by the relA and spoT genes in Escherichia coli allows bacteria to adapt to different environmental stresses. The genome of Borrelia burgdorferi encodes a single chromosomal rel gene (BB0198) (B. burgdorferi rel [relBbu]) homologous to relA and spoT of E. coli. Its role in (p)ppGpp synthesis, bacterial growth, and modulation of gene expression has not been studied in detail. We constructed a relBbu deletion mutant in an infectious B. burgdorferi 297 strain and isolated an extrachromosomally complemented derivative of this mutant. The mutant did not synthesize relBbu mRNA, RelBbu protein, or (p)ppGpp. This synthesis was restored in the complemented derivative, confirming that relBbu is necessary and sufficient for (p)ppGpp synthesis and degradation in B. burgdorferi. The relBbu mutant grew well during log phase in complete BSK-H but reached lower cell concentrations in the stationary phase than the wild-type parent, suggesting that (p)ppGpp may be an important factor in the ability of B. burgdorferi to adapt to stationary phase. Deletion of relBbu did not eliminate the temperature-elicited OspC shift, nor did it alter bmp gene expression or B. burgdorferi antibiotic susceptibility. Although deletion of relBbu eliminated B. burgdorferi virulence for mice, which was not restored by complementation, we suggest that relBbu-dependent accumulation of (p)ppGpp may be important for in vivo survival of this pathogen.
INTRODUCTION
The stringent response is a regulatory response that allows bacteria to adapt to a lack of nutrients and other environmental stresses (6). It causes accumulation of guanosine-3'-diphosphate-5'-triphosphate (pppGpp) and guanosine-3',5'-bisphosphate (ppGpp), collectively referred to as (p)ppGpp or "magic spots." These nucleotides are synthesized by enzymatic phosphorylation of GDP and GTP to ppGpp and pppGpp, respectively, using ATP as a phosphate donor (6). In Escherichia coli, two different but highly homologous proteins are involved in (p)ppGpp synthesis: RelA, bound to ribosomes and activated by the presence of uncharged tRNA at the ribosomal A site which generally synthesizes (p)ppGpp in response to amino acid limitation (47), and SpoT, a cytosolic (p)ppGpp synthetase (15) which is responsive to changes in the availability of carbon, phosphate, and fatty acids as well as to changes in temperature and osmolarity (6, 34). SpoT is also a (p)ppGpp hydrolase (21, 31). Many gram-positive bacteria have only a single rel ortholog which exhibits both (p)ppGpp synthetase and hydrolase activity (29).
(p)ppGpp acts as a global transcriptional regulator. The general effect of the stringent response is a decrease in rRNA, tRNA, and protein synthesis and a decrease in growth rate that results in bacterial adaptation to an environment scarce in nutrients (6). (p)ppGpp also influences many other bacterial physiological functions including competence (23), morphological and physiological differentiation and production of clavulanic acid and cephamycin C (25, 26), production of actinorhodin and undecylprodigiosin antibiotics (46), thermotolerance (48), adaptation to oxidative stress (30), and sensitivity to antibiotics. As regards the latter, E. coli strains able to synthesize (p)ppGpp in either a RelA- or a SpoT-dependent manner show a greater resistance to antimicrobials than strains that cannot produce (p)ppGpp (17). In E. coli, production of (p)ppGpp is also required for the accumulation of inorganic polyphosphate needed for degradation of proteins during starvation mediated by the Lon protease (28), and lack of inorganic polyphosphate synthesis in enteric bacteria is accompanied by a reduction in virulence (27).
The relA and spoT gene products and (p)ppGpp mediate important aspects of virulence in a number of pathogens (8, 20, 34). For example, Vibrio cholerae relA is involved in the ability of V. cholerae to display pathogenicity in in vitro and in vivo models of infection (20), Mycobacterium tuberculosis rel-mediated adaptation to stationary phase is critical to long-term persistence of M. tuberculosis in mice (8), and mutations in Salmonella enterica serovar Typhimurium relA and spoT result in attenuation in animals (34). Not all rel mutants show decreased virulence. Mutation of Listeria monocytogenes relA, while accompanied by loss of (p)ppGpp synthesis and a decrease in osmotolerance, does not diminish in vivo virulence (32).
Borrelia burgdorferi is the etiologic agent of Lyme disease (44). During its life cycle, this spirochete thrives in both ticks and mammals and is able to adapt to various environmental conditions, including changes in temperature, pH, osmolarity, and nutrient availability (10, 41). Preliminary work by ourselves and others has indicated that the stringent response is likely to be present in B. burgdorferi (3, 4, 7). The B. burgdorferi genome contains a single chromosomal rel gene (BB0198) (B. burgdorferi rel [relBbu]) that shows 35 and 39% amino acid homology, respectively, to E. coli relA and spoT (13). Microarray studies have shown increased expression of relBbu mRNA in B. burgdorferi grown in BSK-H at pH 7.5 and 23°C (unfed tick conditions) compared to B. burgdorferi grown in BSK-H at pH 6.8 and 37°C (fed tick conditions) or in rat peritoneal chambers (37). (p)ppGpp synthesis by B. burgdorferi and the ability of relBbu to complement a relA-spoT E. coli double mutant for growth on minimal medium (3, 4, 7) provide additional support for the existence of a functional stringent response in B. burgdorferi mediated by the product of relBbu. We previously suggested that the stringent response in B. burgdorferi could modulate expression of bmpD, a lipoprotein gene jointly transcribed with the ribosomal protein genes rpsL and rpsG (3, 11). The discovery that B. burgdorferi RpoS (S) is responsible for the expression of OspC, the adhesin crucial for the ability of B. burgdorferi to infect the mammalian host (18, 33) and evidence from other bacterial systems that (p)ppGpp is a positive regulator of rpoS expression (16), strongly suggests that (p)ppGpp might modulate expression of lipoproteins involved in the B. burgdorferi life cycle and virulence (22).
To confirm that relBbu is necessary for (p)ppGpp production in B. burgdorferi and to clarify its role in B. burgdorferi physiology, relBbu was deleted in infectious B. burgdorferi 297. This deletion was complemented extrachromosomally, and the B. burgdorferi derivatives obtained were assayed for relBbu mRNA transcription, RelBbu protein synthesis, and the ability to produce (p)ppGpp. Growth characteristics, OspA and OspC protein levels, bmp mRNA levels, sensitivity to antibiotics, and virulence in mice were also compared between wild-type, mutant, and complemented strains. The relBbu gene was clearly responsible for (p)ppGpp synthesis. Its deletion affected B. burgdorferi growth in vitro and ablated in vivo virulence in mice. In contrast, deletion of relBbu did not alter OspA protein levels, bmp gene expression, or B. burgdorferi antibiotic susceptibility and did not eliminate the temperature-elicited OspC shift, suggesting that the role of (p)ppGpp in B. burgdorferi physiology is different from that in E. coli.
MATERIALS AND METHODS
Bacterial strains and culture conditions. Transformable clone BbAH130 of infectious B. burgdorferi 297 was kindly provided by M. V. Norgard, University of Texas Southwestern Medical Center. B. burgdorferi was maintained at 34°C in BSK-II (38) or BSK-H (Sigma-Aldrich, St. Louis, MO). Both media were supplemented with 6% rabbit serum (Sigma). BSK-II was used for preparing electrocompetent B. burgdorferi cells and for the selection of transformants. BSK-H was used to compare growth and levels of mRNA, protein, and (p)ppGpp in B. burgdorferi and its derivatives. The B. burgdorferi relBbu mutant was grown in the presence of kanamycin, 400 μg/ml (Sigma). Complemented B. burgdorferi was grown in the presence of kanamycin, 400 μg/ml, and streptomycin, 50 μg/ml (Sigma). E. coli DH5 (GIBCO/Life Technologies, Grand Island, NY) grown in the presence of spectinomycin, 100 μg/ml (Sigma), was used for propagation of the pKFSS1 and pKFSS1-relBbu plasmids.
Generation of relBbu deletion. Deletion of B. burgdorferi relBbu was done by homologous recombination with a relBbu inactivation construct (Fig. 1) using PCR-based fusions (43). In brief, nucleotides 195216 to 195852 (primers 1 and 2) (Table 1) and nucleotides 197417 to 198004 (primers 3 and 4) (Table 1) were amplified from B. burgdorferi chromosomal DNA, and the kanamycin resistance gene aph(3')-IIIa from Enterococcus faecalis with its own promoter was amplified from plasmid pBLS500 (primers III and IV) (Table 1). All PCR assays used an initial denaturation at 94°C for 2 min and 32 cycles of 94°C for 15 s, annealing at 56°C for 20 s, and extension at 68°C for 3 min. The three amplicons were fused (43), and the final PCR product containing the relBbu inactivation construction was precipitated with ethanol and resuspended in sterile water at 4 μg/μl. Forty micrograms was used for each transformation (38).
Construction of B. burgdorferi pKFSS1-relBbu. A PCR fragment containing full-size relBbu with 446 bp upstream from its start codon and 109 bp downstream from its stop codon was amplified from B. burgdorferi genomic DNA using forward primer ST1, containing a BamHI restriction site, and reverse primer ST2, containing a PstI restriction site (Table 1). After digestion with BamHI and PstI (New England Biolabs, Inc., Beverly, MA), the resulting fragment was ligated into pKFSS1 (12) digested with the same enzymes. The resulting plasmids were propagated in E. coli DH5 to obtain pKFSS1-relBbu encoding relBbu under its own promoter; 10 μg of this plasmid was electroporated into B. burgdorferi (38). Transformation of B. burgdorferi with pKFSS1 was used as a control.
Confirmation of relBbu inactivation and complementation in B. burgdorferi. DNA from B. burgdorferi transformants was purified (High Pure PCR template preparation kit; Roche Diagnostics Corporation, Indianapolis, IN) and analyzed by PCR (primers R/SpPROT1S and R/SpPROT1EndA) (Table 1 and Fig. 1) to confirm the presence of the relBbu deletion or complementation. The PCR fragment obtained with primers relBbu 6kbL and relBbu 6kbR (Table 1; Fig. 1) was sequenced using primers R/SpPROT1S and R/SpPROT1EndA (Table 1; Fig. 1) to provide further confirmation of the relBbu deletion (Herbert Irving Comprehensive Cancer Center, Columbia University, New York, NY). DNA from complemented B. burgdorferi strains was transformed into E. coli DH5. DNA from E. coli clones selected on spectinomycin plates was subjected to restriction analysis using BamHI and PstI for comparison with the plasmids originally used for electroporation of B. burgdorferi.
Estimation of pKFSS1-relBbu copy number. A competitive PCR assay was used to determine the pKFSS1-relBbu copy number in complemented B. burgdorferi 297. Competitors for flaB and relBbu genes containing internal deletions were constructed as previously described (3, 39). Serial twofold dilutions of competitors were mixed with fixed amounts of total 297 DNA to perform quantitative PCR for flaB and relBbu. Primers 49 and 50 (Table 1) were used for flaB detection; primers ST3 and ST4 (Table 1) were used for relBbu detection. Target DNA and competitor DNA were assumed to be present at equimolar concentrations in those reactions when the competitor and target PCR products were at similar intensities in agarose gels (11).
Determination of B. burgdorferi plasmid content. The plasmid content in B. burgdorferi relBbu mutants and complemented strains was compared with that of the wild-type parental strain using PCR as described previously (24).
RNA isolation, RT-PCR, and competitive RT-PCR. Total RNA from late-log-phase B. burgdorferi was isolated with TRizol reagent (Invitrogen Life Technology, Carlsbad, CA) according to the manufacturer's recommendations and was treated with RQ1 RNase-free DNase (Promega Corporation, Madison, WI) to eliminate DNA contamination. Reverse transcription (RT)-PCR was performed using the Access RT-PCR system (Promega) according to the manufacturer's recommendations (3, 11). Primers ST3 and ST4 (Table 1) were used for relBbu mRNA detection; primers 49 and 50 (Table 1) were used for detection of the constitutively expressed flaB gene as a control to show the presence of RNA in a sample (11). Primers BB0199RT_R and BB0199RT_F (Table 1) were used for detection of mRNA for the BB0199 gene. Primers and conditions for competitive RT-PCR for bmp genes have been described previously (3, 11).
Detection of B. burgdorferi proteins. Anti-RelBbu antibodies were raised in rabbits immunized with recombinant RelBbu protein (A. V. Bryksin, T. N. Orlova, and F. C. Cabello, unpublished data). Rabbit anti-FlaB antibodies were provided by J. D. Radolf, University of Connecticut, Farmington, CT; mouse monoclonal anti-OspA antibodies (H5332) were provided by A. G. Barbour, University of California, Irvine, CA; mouse monoclonal anti-OspC antibodies were provided by R. D. Gilmore, Centers for Disease Control and Prevention, Fort Collins, CO; and mouse monoclonal anti-RpoS antibodies (6A7-101-H11) were provided by M. V. Norgard, University of Texas Southwestern Medical Center, Dallas, TX. FlaB, OspA, OspC, RelBbu, and RpoS proteins were detected in B. burgdorferi lysates by Western blot analysis and the ECF Western blotting kit (Amersham Biosciences, Piscataway, NJ) and quantified using a Storm 860 PhosphorImager and ImageQuaNT software (Molecular Dynamics, Sunnyvale, CA).
Detection of (p)ppGpp. B. burgdorferi (5 x 106 cells/ml) was incubated in BSK-H at 34°C containing 10 μCi/ml of uniformly labeled [32P]orthophosphate (Amersham) for 2 days. Labeled cells were harvested from 10-ml cultures, and (p)ppGpp was extracted and chromatographed on cellulose polyethyleneimine thin-layer chromatography (TLC) plates (Selecto Scientific, Suwanee, GA.) (3). Plates were air dried, exposed to a phosphor screen (Molecular Dynamics) for 12 to 24 h, and scanned using a Storm 860 PhosphorImager.
B. burgdorferi growth assays. Wild-type, relBbu, and complemented B. burgdorferi strains (104 cells/ml) were grown at 34°C in BSK-H for 30 days. Cell numbers were determined by dark-field microscopy every other day during the first 2 weeks of growth. To measure the effect of cell death in B. burgdorferi stationary-phase cultures, the number of viable organisms present in each culture was determined by limiting dilution at day 30 and compared to the number of cells counted microscopically at day 14. Limiting dilution was performed using serial fourfold dilutions of 30-day-old cultures with fresh BSK-H in 96-well plates (180 μl/well; initial dilution, 1:4; final dilution, 1:16,777,216). The highest dilution yielding positive cultures was determined after two additional weeks of growth at 34°C, and the number of cells/ml in the culture on day 30 was calculated assuming that one cell was sufficient to produce cell growth in a well. The data are presented as the cell survival index (as a percentage) = viable cells/mlDay 30 (by limiting dilution)/Total cells/mlday 14 (by microscopy) x 100 and are presented as means ± standard deviations. Limiting dilution analysis and microscopy were done in duplicate in each of two independent experiments. Results were analyzed statistically using a one-way analysis of variance.
B. burgdorferi temperature shift. B. burgdorferi was grown at 34°C in BSK-H to 107 cells/ml, then inoculated at an initial concentration of 106 cells/ml, and grown at 23°C. After 1 week at 23°C, B. burgdorferi was transferred to fresh BSK-H at 23°C, 34°C, and 37°C at an initial concentration of 103 cells/ml. Cells were collected when their concentration reached 3 x 107 cells/ml for the wild-type parent and 107 cells/ml for relBbu. In other temperature shift experiments, B. burgdorferi was grown at 34°C as described above, then transferred to BSK-H at 37°C, and collected at early stationary phase when cell concentrations reached 108 cells/ml for the wild-type parent, 2 x 107 cells/ml for the relBbu mutant and the mutant complemented with the empty pKFSS1 vector, and 5 x 107 cells/ml for the mutant complemented with pKFSS1-relBbu.
Antibiotic sensitivity. MICs of ampicillin, ciprofloxacin, erythromycin, tetracycline, and vancomycin (Sigma) were determined for wild-type, relBbu, and complemented B. burgdorferi in BSK-H. B. burgdorferi (105 cells/ml) in BSK-H was added to serial threefold dilutions of each antibiotic (initial concentration, 600 μg/ml; final concentration, 0.01 μg/ml) or no antibiotic in 96-well plates. The plates were incubated for 2 weeks at 32°C in a humidified 3% CO2 atmosphere. The MIC was defined as the smallest antibiotic concentration preventing B. burgdorferi growth under these conditions.
Infection of mice with B. burgdorferi. Four groups of 4-week-old C3H/HeN mice (Charles River Laboratory, Wilmington, MA) were infected intradermally with B. burgdorferi 297 wild type (five mice), relBbu mutant (four mice), relBbu mutant transformed with pKFSS1 (four mice), and complemented relBbu mutant (four mice) using 104 cells/mouse. Two weeks after infection, the mice were sacrificed, and ear punches and blood samples (100 μl) were taken and cultured for 2 weeks at 34°C in 1.5 ml of BSK-H containing 50 μg/ml of rifampin (Merrell Dow Pharmaceuticals, Inc., Cincinnati, OH) and 2.5 μg/ml of amphotericin B (Sigma) before being microscopically examined for growth of B. burgdorferi.
RESULTS
Isolation of a relBbu mutant in B. burgdorferi. Construction of the relBbu deletion is shown in Fig. 1. A PCR fragment containing the relBbu flanking regions surrounding a kanamycin resistance gene was electroporated into infectious B. burgdorferi 297. This electroporation yielded 20 kanamycin-resistant clones with a deleted relBbu gene. Of the 2,004 nucleotides of the wild-type relBbu gene, the mutants had 160 nucleotides at the 5' terminus and 280 nucleotides at the 3' terminus of relBbu, with the central 1,564 nucleotides deleted and replaced by the kanamycin resistance gene. The deletion in the relBbu gene was confirmed by PCR analysis with primers R/SpPROT1S and R/SpPROT1EndA (Fig. 1) which showed that relBbu produced a shorter PCR fragment than the fragment generated by the same primers in the wild-type B. burgdorferi (Fig. 2A). Only 2 of the 20 mutants had all the plasmids of parental B. burgdorferi 297. PCR revealed that the parental and these two relBbu B. burgdorferi mutants lacked the lp56, lp38, and cp32-8 plasmids but otherwise contained the other expected B. burgdorferi plasmids (data not shown) (24). Chromosomal DNA fragments in these two relBbu mutants were PCR amplified with primers relBbu6kbL and relBbu6kbR (Fig. 1), and the deletion of relBbu was confirmed by DNA sequencing. Both mutants were fully characterized genetically and functionally. Identical results were obtained with both; only the data generated from one of them is shown below. Deletion of the relBbu gene did not prevent mRNA expression of the downstream BB0199 gene (Fig. 1). Its RNA could be detected by RT-PCR in wild-type and relBbu B. burgdorferi (Fig. 2B), showing the absence of a polar effect of the relBbu mutation.
Complementation of relBbu mutant. To complement the relBbu deletion, the relBbu strain was electroporated with pKFSS1-relBbu carrying the relBbu gene under its own promoter or with the empty pKFSS1 cloning vector as a control. The presence of the wild-type and deleted relBbu alleles was determined by PCR analysis of the complemented B. burgdorferi derivatives using primers R/SpPROT1S and R/SpPROT1EndA (Fig. 1). The mutant complemented with pKFSS1-relBbu produced both mutant- and wild-type-sized amplicons, while the strain transformed with the pKFSS1 cloning vector produced only the mutant-sized amplicon (Fig. 2A). To demonstrate that the plasmids used for complementation were not modified in B. burgdorferi, DNA from complemented B. burgdorferi derivatives was transformed into E. coli. Restriction analysis confirmed that plasmid DNA from spectinomycin-resistant E. coli clones and plasmid DNA originally used for complementation were identical (Fig. 2C).
Detection of relBbu mRNA and RelBbu protein. RT-PCR analysis showed that relBbu mRNA was produced in wild-type B. burgdorferi grown in BSK-H at 34°C, while the relBbu mutant had no relBbu mRNA synthesis (Fig. 3). Transformation of relBbu with the pKFSS1-relBbu plasmid but not with the pKFSS1 cloning vector restored relBbu mRNA synthesis in the mutant (Fig. 3).
Western blot analysis with rabbit anti-rRelBbu protein detected a 78-kDa band corresponding to RelBbu protein in lysates of the wild-type B. burgdorferi grown in BSK-H at 34°C (Fig. 4B). This band was absent in lysates of relBbu (Fig. 4B) and reappeared in lysates of relBbu transformed with pKFSS1-relBbu. Transformation with the pKFSS1 vector alone did not restore RelBbu synthesis in relBbu (Fig. 4B). The intensity of the RelBbu band detected in the complemented derivative with pKFSS1-relBbu was stronger than that detected in wild-type B. burgdorferi (Fig. 4B). Image analysis of six independent Western blots indicated that the intensity of this band was threefold greater in the complemented relBbu mutant than in wild-type B. burgdorferi, suggesting increased synthesis of RelBbu protein in complemented B. burgdorferi (Fig. 4C).
To determine whether the increased intensity of the RelBbu band was merely a consequence of the pKFSS1-relBbu copy number, the numbers of relBbu genes in wild-type and complemented B. burgdorferi were compared by competitive PCR (39). Competitive PCR for flaB was used as an internal control for copy number, since both strains have only a single flaB gene. Wild-type and complemented B. burgdorferi had similar numbers of relBbu copies (Fig. 2D), and since wild-type B. burgdorferi has only one relBbu gene, the copy number of the pKFSS1-relBbu plasmid was also one per B. burgdorferi cell.
(p)ppGpp synthesis. Both ppGpp and pppGpp could be easily detected in thin-layer chromatograms in wild-type B. burgdorferi 297 grown at 34°C in BSK-H. Spots corresponding to ppGpp or pppGpp were not detected in the relBbu mutant (Fig. 5A and B), while (p)ppGpp synthesis was restored in relBbu complemented with pKFSS1-relBbu. There was no ppGpp nor pppGpp in the mutant transformed with the empty pKFSS1 vector (Fig. 5A and B). (p)ppGpp levels, as indicated by the intensity of the spots, appeared to be the same in wild-type and complemented B. burgdorferi despite the different amounts of RelBbu protein in these two strains. These results confirm that relBbu is involved in (p)ppGpp synthesis in B. burgdorferi.
Growth in culture. In a variety of bacteria, including E. coli, (p)ppGpp levels affect cell growth (6, 26, 32). We therefore compared the growth of the wild type, the relBbu mutant, and the complemented mutant at 34°C in BSK-H (Fig. 6). Figure 6A shows that there was little difference in cell numbers between the wild type and the relBbu mutant during the first 8 days of culture. The average doubling time was 5.2 h for wild-type B. burgdorferi 297 and 7.8 h for the relBbu mutant during this time, but this difference was not statistically significant (analysis of variance, P > 0.05). However, at 10 to 14 days of culture, when the cultured cells were approaching stationary phase, cell densities in the wild-type cultures were approximately 10-fold greater than those in relBbu, and this difference was significant (analysis of variance, P < 0.01). Complementation of the relBbu mutant with pKFSS1-relBbu increased stationary-phase cell densities significantly, about fourfold, at 10 to 14 days, but a significant twofold difference in cell densities between complemented and wild-type cell remained (analysis of variance, P < 0.01). There was no difference in cell growth between relBbu and relBbu transformed with pKFSS1 vector alone.
To determine whether cell death was responsible for the lower cell concentrations of the relBbu mutant cultures in the stationary phase, wild-type, mutant, complemented mutant, and vector-transformed cultures were grown for 30 days and the numbers of viable organisms remaining in these cultures at that time were determined by limiting dilution. After a further 2 weeks of growth, the highest dilution that gave positive cultures for each B. burgdorferi strain was used to calculate a survival index which took into account the differences in cell concentrations reached by each strain on day 14 (Fig. 6B). The numbers of surviving cells at 30 days of culture in the wild-type, mutant, and complemented strains were not significantly different (analysis of variance, P = 0.7) and constituted approximately 5 to 9 cells per 10,000 cells observed microscopically at 14 days of culture. These experiments showed that the relBbu mutation did not lead to increased cell death in stationary-phase cultures but rather decreased growth abilities prior to this point during the late exponential growth phase.
Expression of bmp genes. bmpD is transcribed together with the ribosomal protein genes rpsL and rpsG, and its expression might be modulated by the stringent response (3, 11). Measurement of bmp mRNA by competitive RT-PCR showed no differences in transcription of bmpA, bmpB, bmpC, and bmpD genes between the wild type and the relBbu mutant of B. burgdorferi (data not shown), indicating that bmp gene transcription is not modulated by (p)ppGpp.
Effects of temperature shift on OspC production. B. burgdorferi increases OspC levels when B. burgdorferi is transferred from 23°C to 34 to 37°C conditions of growth (42, 45). RpoS has been shown to be responsible both for this increase in OspC levels with a shift in temperature and for activation of transcription from the ospC promoter at elevated temperatures (22). Because (p)ppGpp is a positive regulator of RpoS in E. coli (16), ablation of (p)ppGpp synthesis in the relBbu mutant might alter the expression pattern of OspC. OspC protein was produced at 34°C and 37°C and was essentially absent at 23°C in mid-log-phase wild-type and relBbu cultures grown in vitro in BSK-H (Fig. 7A). However, OspC levels were at least twofold lower in the relBbu mutant than in wild-type B. burgdorferi grown at elevated temperatures (Fig. 7A). To determine whether (p)ppGpp was needed for high-level OspC expression, OspC levels in wild-type, relBbu, and complemented B. burgdorferi were determined in organisms grown at 37°C to early stationary phase, where the difference in growth between the wild type and the relBbu mutant might further increase any possible effect of (p)ppGpp on OspC expression. Stationary-phase OspC levels were higher in wild-type B. burgdorferi grown at 37°C than in the relBbu mutant (Fig. 7B), confirming the results obtained with mid-log-phase cultures, but complementation of the relBbu mutant did not raise OspC levels, suggesting that (p)ppGpp is not involved in regulation of OspC synthesis. RpoS was detected in the immunoblots of the wild-type and relBbu mutant B. burgdorferi growing at elevated temperature but not at 23°C (data not shown), indicating that RpoS expression does not require (p)ppGpp.
Antibiotic susceptibility of wild-type, relBbu mutant, and complemented B. burgdorferi. In E. coli, (p)ppGpp synthesis is required for resistance to several antibiotics (17). We therefore examined whether the absence of (p)ppGpp in the relBbu mutant modified B. burgdorferi antibiotic susceptibility. The MICs of ampicillin, erythromycin, tetracycline, and vancomycin were similar for wild-type B. burgdorferi and the relBbu mutant (Table 2). The MIC of ciprofloxacin was three times higher for the wild-type strain than for the relBbu mutant, but the MIC of ciprofloxacin for the complemented derivative was identical to that of the relBbu mutant. Since complementation of the relBbu mutant did not restore the ciprofloxacin MIC to the wild-type level, we concluded that (p)ppGpp was not involved in the change in the ciprofloxacin susceptibility of the mutant.
Infection of mice with wild-type, relBbu, and complemented B. burgdorferi. C3H/HeN mice were infected intradermally with the B. burgdorferi 297 wild type, relBbu mutant, relBbu mutant transformed with pKFSS1, or relBbu mutant complemented with pKFSS1-relBbu. All the strains used for infection had been shown to carry all plasmids of the parental B. burgdorferi 297 (data not shown). Two weeks after infection, B. burgdorferi could be cultured only from ear punches and blood samples of mice injected with wild-type B. burgdorferi (four of five positive ear punch cultures and five of five positive blood cultures). B. burgdorferi could not be cultured from ear punches and blood of any of the mice receiving relBbu or its derivatives transformed with pKFSS1 or pKFSS1-relBbu.
DISCUSSION
relA and spoT homologues are responsible for synthesis of (p)ppGpp, the alarmone involved in the bacterial stringent response (6). We have shown that relBbu is the only gene involved in generation of (p)ppGpp in B. burgdorferi. Deletion of relBbu did not eliminate mRNA synthesis from the BB0199 downstream gene of unknown function but completely abolished the production of relBbu mRNA, RelBbu protein, and (p)ppGpp in infectious B. burgdorferi 297. Synthesis of these molecules during growth of B. burgdorferi in BSK-H was restored after introduction of an extrachromosomal copy of the wild-type allele of relBbu into the mutant. Our findings are not inconsistent with the possible existence of different relBbu regulators that may affect (p)ppGpp production under certain conditions. The apparently increased amounts of RelBbu protein in the complemented B. burgdorferi strain (Fig. 4) could not be explained by the different strengths of the relBbu gene promoter in the chromosome and the plasmid, since relBbu was cloned into pKFSS1 under its own promoter (located about 300 nucleotides upstream from the relBbu ATG translational start) (4). These increased amounts of RelBbu could also not be a result of multiple copies of the introduced plasmid, since we found only a single copy of the vector per complemented B. burgdorferi genome. The possibility exists that differences in DNA supercoiling for the relBbu gene located on the pKFSS1-relBbu plasmid and on the B. burgdorferi chromosome might ultimately be responsible for the different amounts of RelBbu protein observed (1, 9). The increased amounts of RelBbu did not increase (p)ppGpp levels in complemented B. burgdorferi (Fig. 5), most probably because RelBbu has two activities, (p)ppGpp synthetic and hydrolytic (6), and both increase simultaneously with increased amounts of protein. Alternatively, the increase in the production of RelBbu might not be accompanied by increased levels of (p)ppGpp because the excess protein might be inactively bound to the ribosomes (47).
The relBbu gene and (p)ppGpp are not essential for in vitro growth of B. burgdorferi, since the relBbu mutant was able to multiply in BSK-H (Fig. 6). B. burgdorferi thus differs from Staphylococcus aureus, since rel mutants cannot be isolated in S. aureus because the relSau gene and (p)ppGpp are essential in this species (14). Nevertheless, the relBbu mutant did show an important growth defect because it only achieved 1/10 the cell concentration of wild-type B. burgdorferi in the stationary phase in vitro. It is not clear whether this growth defect resulted from a decrease in growth during the exponential phase or is a result of the inability of this mutant to thrive during the stationary phase, since the mutant does not display increased cell death during the stationary phase (Fig. 6). The growth deficit of B. burgdorferi lacking (p)ppGpp is quite different from that of E. coli lacking (p)ppGpp. In E. coli, experimental evidence indicates that (p)ppGpp can suppress expression of the ftsZ gene, an essential gene for cell division and septation (35), and the level of (p)ppGpp is inversely correlated with the rate of initiation of new rounds of DNA replication (40).
We previously reported that lack of (p)ppGpp synthesis in B. burgdorferi grown in the presence of tick cells was associated with changes in the transcription of bmpD, a gene transcribed with the ribosomal protein genes rpsL and rpsG, compared to that seen in B. burgdorferi grown in BSK-H (3, 11). Unexpectedly, we found no alteration of bmp gene expression in the relBbu mutant compared to wild-type B. burgdorferi, indicating that perhaps the changes seen in the expression of these genes during B. burgdorferi growth in the presence of tick cells were more the result of their interaction with the tick cells than of the disappearance of (p)ppGpp (3). More extensive experiments to examine bmp gene modulation in the relBbu mutant and its complemented derivatives growing in the presence of tick cells will be needed to clarify these differences.
It is well established that (p)ppGpp can regulate gene expression in E. coli through a positive control cascade mediated by the alternative bacterial factor RpoS (6). RpoS is involved in the OspC temperature-mediated shift in B. burgdorferi (22), which has been shown to be important for transmission of B. burgdorferi from ticks to mammals (18, 33). Although OspC expression decreased in B. burgdorferi relBbu, the OspC temperature shift was not eliminated by the lack of (p)ppGpp. This suggests that (p)ppGpp does not modulate RpoS functions in B. burgdorferi. Detection of RpoS in immunoblots in the wild-type 297 and relBbu strains (data not shown) also confirmed that RpoS expression does not require (p)ppGpp. The absence of modulation of RpoS-dependent genes by (p)ppGpp in B. burgdorferi is surprising, since there is a homologue of dksA in the B. burgdorferi genome (BB0168) (13) whose gene product mediates RpoS induction by (p)ppGpp in E. coli (2). An absence of modulation of RpoS levels by (p)ppGpp has also been found in S. enterica serovar Typhimurium (34), where (p)ppGpp can modulate expression of virulence-associated genes by an RpoS-independent pathway (34). A recent report has proposed that there are RpoS-independent virulence regulons in B. burgdorferi (5) and that (p)ppGpp may have a role in their regulation, but this requires further study.
(p)ppGpp mediates resistance of E. coli to various antibiotics by lowering bacterial metabolism and growth rate as well as by RpoS activation (17, 19, 36). Our experiments indicate that the MICs of several classes of antibiotics are similar in B. burgdorferi in the presence and absence of (p)ppGpp (Table 2). The absence of this phenotype could be explained by the fact that, in contrast to what occurs in E. coli, the lack of (p)ppGpp did not change the B. burgdorferi growth rate substantially and did not affect the synthesis of RpoS, two elements involved in (p)ppGpp-mediated antibiotic resistance in E. coli. The decreased MIC of ciprofloxacin observed in the relBbu mutant and its complemented derivative could be the result of a spontaneous mutation. Alternatively, the difference in cell densities between complemented and wild-type B. burgdorferi strains might be responsible for the increased susceptibility to ciprofloxacin in the complemented strain.
Since mutation of rel has been shown to affect virulence in several bacterial pathogens (8, 20, 34), we compared the ability of wild-type, relBbu, and complemented B. burgdorferi to infect mice after needle inoculation. The loss of infectivity for relBbu suggested that relBbu is necessary for B. burgdorferi to produce infection in mice. Since complementation, even in the presence of all parental B. burgdorferi plasmids, did not restore infectivity in mice, there is a possibility of a defect that cannot be determined by the present techniques. This suggests that other complementation approaches (for example, intrachromosomal) might be needed to confirm that (p)ppGpp is involved in B. burgdorferi virulence. The decreased levels of OspC, a major lipoprotein necessary for infection of mammalian hosts (18, 33), might partially explain the loss of infectivity in the relBbu mutant, although complementation restored neither OspC levels nor infectivity. Increased RelBbu levels in complemented B. burgdorferi, even though they did not lead to increased (p)ppGpp amounts, might be responsible for the failure of the complementation to restore infectivity in mice. Another explanation for the failure of complementation to restore the infectiousness of the relBbu strain could be loss of the complementing plasmid harboring the relBbu wild-type gene in mice (12).
In summary, we have shown that relBbu is responsible for (p)ppGpp synthesis and affects the growth and probably the infectivity of B. burgdorferi. Our work also provides evidence that the absence of (p)ppGpp in B. burgdorferi was not accompanied by the same phenotypes as in E. coli. This suggests that the role of this mediator in B. burgdorferi physiology is different from its role in E. coli. Our results indicate that (p)ppGpp does not regulate RpoS-dependent promoters and does not influence susceptibility to antibiotics in B. burgdorferi. Further studies will be needed to ascertain the role of (p)ppGpp in B. burgdorferi and to define the stringent response in this pathogen.
ACKNOWLEDGMENTS
This work was supported by a grant to F.C.C. (R01 AI41-056-7).
We thank M. Cashel, National Institutes of Health, Bethesda, MD, for advice and discussions; M. V. Norgard, University of Texas Southwestern Medical Center, Dallas, TX, for B. burgdorferi 297, clone BbAH130, and anti-RpoS antibodies; D. S. Samuels, University of Montana, Missoula, MT, for pKFSS1; J. D. Radolf, C. H. Eggers, and M. J. Caimano, University of Connecticut, Farmington, CT, for anti-FlaB antibodies and for extensive discussions; A. G. Barbour, University of California, Irvine, CA, for anti-OspA antibodies (H5332); and R. D. Gilmore, CDC, Fort Collins, CO, for anti-OspC antibodies. We also thank Radha Iyer and Darya Terekhova, New York Medical College, Valhalla, NY, for help with analysis of B. burgdorferi plasmid composition.
REFERENCES
1. Alverson, J., S. F. Bundle, C. D. Sohaskey, M. C. Lybecker, and D. S. Samuels. 2003. Transcriptional regulation of the ospAB and ospC promoters from Borrelia burgdorferi. Mol. Microbiol. 48:1665-1677.
2. Brown, L., D. Gentry, T. Elliott, and M. Cashel. 2002. DksA affects ppGpp induction of RpoS at a translational level. J. Bacteriol. 184:4455-4465.
3. Bugrysheva, J., E. Y. Dobrikova, H. P. Godfrey, M. L. Sartakova, and F. C. Cabello. 2002. Modulation of Borrelia burgdorferi stringent response and gene expression during extracellular growth with tick cells. Infect. Immun. 70:3061-3067.
4. Bugrysheva, J., E. Y. Dobrikova, M. L. Sartakova, M. J. Caimano, T. J. Daniels, J. D. Radolf, H. P. Godfrey, and F. C. Cabello. 2003. Characterization of the stringent response and relBbu expression in Borrelia burgdorferi. J. Bacteriol. 185:957-965.
5. Caimano, M. J., C. H. Eggers, K. R. Hazlett, and J. D. Radolf. 2004. RpoS is not central to the general stress response in Borrelia burgdorferi but does control expression of one or more essential virulence determinants. Infect. Immun. 72:6433-6445.
6. Cashel, M., D. R. Gentry, V. J. Hernandez, and D. Vinella. 1996. The stringent response, p. 1458-1496. In F. C. Neidhardt, R. I. Curtiss, J. L. Ingraham, E. C. C. Lin, K. B. Low, B. Magasanik, W. S. Reznikoff, M. Riley, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella: cellular and molecular biology, 2nd ed. ASM Press, Washington, D.C.
7. Concepcion, M. B., and D. R. Nelson. 2003. Expression of spoT in Borrelia burgdorferi during serum starvation. J. Bacteriol. 185:444-452.
8. Dahl, J. L., C. N. Kraus, H. I. Boshoff, B. Doan, K. Foley, D. Avarbock, G. Kaplan, V. Mizrahi, H. Rubin, and C. E. Barry III. 2003. The role of RelMtb-mediated adaptation to stationary phase in long-term persistence of Mycobacterium tuberculosis in mice. Proc. Natl. Acad. Sci. USA 100:10026-10031.
9. Deng, S., R. A. Stein, and N. P. Higgins. 2004. Transcription-induced barriers to supercoil diffusion in the Salmonella typhimurium chromosome. Proc. Natl. Acad. Sci. USA 101:3398-3403.
10. De Silva, A. M., and E. Fikrig. 1997. Arthropod- and host-specific gene expression by Borrelia burgdorferi. J. Clin. Investig. 99:377-379.
11. Dobrikova, E. Y., J. Bugrysheva, and F. C. Cabello. 2001. Two independent transcriptional units control the complex and simultaneous expression of the bmp paralogous chromosomal gene family in Borrelia burgdorferi. Mol. Microbiol. 39:370-378.
12. Frank, K. L., S. F. Bundle, M. E. Kresge, C. H. Eggers, and D. S. Samuels. 2003. aadA confers streptomycin resistance in Borrelia burgdorferi. J. Bacteriol. 185:6723-6727.
13. Fraser, C. M., S. Casjens, W. M. Huang, G. G. Sutton, R. Clayton, R. Lathigra, O. White, K. A. Ketchum, R. Dodson, E. K. Hickey, M. Gwinn, B. Dougherty, J. F. Tomb, R. D. Fleischmann, D. Richardson, J. Peterson, A. R. Kerlavage, J. Quackenbush, S. Salzberg, M. Hanson, R. van Vugt, N. Palmer, M. D. Adams, J. Gocayne, and J. C. Venter. 1997. Genomic sequence of a Lyme disease spirochaete, Borrelia burgdorferi. Nature 390:580-586.
14. Gentry, D., T. Li, M. Rosenberg, and D. McDevitt. 2000. The rel gene is essential for in vitro growth of Staphylococcus aureus. J. Bacteriol. 182:4995-4997.
15. Gentry, D. R., and M. Cashel. 1995. Cellular localization of the Escherichia coli SpoT protein. J. Bacteriol. 177:3890-3893.
16. Gentry, D. R., V. J. Hernandez, L. H. Nguyen, D. B. Jensen, and M. Cashel. 1993. Synthesis of the stationary-phase sigma factor s is positively regulated by ppGpp. J. Bacteriol. 175:7982-7989.
17. Greenway, D. L., and R. R. England. 1999. The intrinsic resistance of Escherichia coli to various antimicrobial agents requires ppGpp and s. Lett. Appl. Microbiol. 29:323-326.
18. Grimm, D., K. Tilly, R. Byram, P. E. Stewart, J. G. Krum, D. M. Bueschel, T. G. Schwan, P. F. Policastro, A. F. Elias, and P. A. Rosa. 2004. Outer-surface protein C of the Lyme disease spirochete: a protein induced in ticks for infection of mammals. Proc. Natl. Acad. Sci. USA 101:3142-3147.
19. Gustafson, C. E., S. Kaul, and E. E. Ishiguro. 1993. Identification of the Escherichia coli lytB gene, which is involved in penicillin tolerance and control of the stringent response. J. Bacteriol. 175:1203-1205.
20. Haralalka, S., S. Nandi, and R. K. Bhadra. 2003. Mutation in the relA gene of Vibrio cholerae affects in vitro and in vivo expression of virulence factors. J. Bacteriol. 185:4672-4682.
21. Hernandez, V. J., and H. Bremer. 1991. Escherichia coli ppGpp synthetase II activity requires spoT. J. Biol. Chem. 266:5991-5999.
22. Hübner, A., X. Yang, D. M. Nolen, T. G. Popova, F. C. Cabello, and M. V. Norgard. 2001. Expression of Borrelia burgdorferi OspC and DbpA is controlled by a RpoN-RpoS regulatory pathway. Proc. Natl. Acad. Sci. USA 98:12724-12729.
23. Inaoka, T., and K. Ochi. 2002. RelA protein is involved in induction of genetic competence in certain Bacillus subtilis strains by moderating the level of intracellular GTP. J. Bacteriol. 184:3923-3930.
24. Iyer, R., O. Kalu, J. Purser, S. Norris, B. Stevenson, and I. Schwartz. 2003. Linear and circular plasmid content in Borrelia burgdorferi clinical isolates. Infect. Immun. 71:3699-3706.
25. Jin, W., H. K. Kim, J. Y. Kim, S. G. Kang, S. H. Lee, and K. J. Lee. 2004. Cephamycin C production is regulated by relA and rsh genes in Streptomyces clavuligerus ATCC27064. J. Biotechnol. 114:81-87.
26. Jin, W., Y. G. Ryu, S. G. Kang, S. K. Kim, N. Saito, K. Ochi, S. H. Lee, and K. J. Lee. 2004. Two relA/spoT homologous genes are involved in the morphological and physiological differentiation of Streptomyces clavuligerus. Microbiology 150:1485-1493.
27. Kim, K. S., N. N. Rao, C. D. Fraley, and A. Kornberg. 2002. Inorganic polyphosphate is essential for long-term survival and virulence factors in Shigella and Salmonella spp. Proc. Natl. Acad. Sci. USA 99:7675-7680.
28. Kuroda, A., K. Nomura, R. Ohtomo, J. Kato, T. Ikeda, N. Takiguchi, H. Ohtake, and A. Kornberg. 2001. Role of inorganic polyphosphate in promoting ribosomal protein degradation by the Lon protease in E. coli. Science 293:705-708.
29. Mittenhuber, G. 2001. Comparative genomics and evolution of genes encoding bacterial (p)ppGpp synthetases/hydrolases (the Rel, RelA and SpoT proteins). J. Mol. Microbiol. Biotechnol. 3:585-603.
30. Mostertz, J., C. Scharf, M. Hecker, and G. Homuth. 2004. Transcriptome and proteome analysis of Bacillus subtilis gene expression in response to superoxide and peroxide stress. Microbiology 150:497-512.
31. Murray, K. D., and H. Bremer. 1996. Control of spoT-dependent ppGpp synthesis and degradation in Escherichia coli. J. Mol. Biol. 259:41-57.
32. Okada, Y., S. Makino, T. Tobe, N. Okada, and S. Yamazaki. 2002. Cloning of rel from Listeria monocytogenes as an osmotolerance involvement gene. Appl. Environ. Microbiol. 68:1541-1547.
33. Pal, U., X. Yang, M. Chen, L. K. Bockenstedt, J. F. Anderson, R. A. Flavell, M. V. Norgard, and E. Fikrig. 2004. OspC facilitates Borrelia burgdorferi invasion of Ixodes scapularis salivary glands. J. Clin. Investig. 113:220-230.
34. Pizarro-Cerda, J., and K. Tedin. 2004. The bacterial signal molecule, ppGpp, regulates Salmonella virulence gene expression. Mol. Microbiol. 52:1827-1844.
35. Powell, B. S., and D. L. Court. 1998. Control of ftsZ expression, cell division, and glutamine metabolism in Luria-Bertani medium by the alarmone ppGpp in Escherichia coli. J. Bacteriol. 180:1053-1062.
36. Rand, J. D., S. G. Danby, D. L. Greenway, and R. R. England. 2002. Increased expression of the multidrug efflux genes acrAB occurs during slow growth of Escherichia coli. FEMS Microbiol. Lett. 207:91-95.
37. Revel, A. T., A. M. Talaat, and M. V. Norgard. 2002. DNA microarray analysis of differential gene expression in Borrelial burgdorferi, the Lyme disease spirochete. Proc. Natl. Acad. Sci. USA 99:1562-1567.
38. Samuels, D. S. 1995. Electrotransformation of the spirochete Borrelia burgdorferi. Methods Mol. Biol. 47:253-259.
39. Sartakova, M., E. Dobrikova, and F. C. Cabello. 2000. Development of an extrachromosomal cloning vector system for use in Borrelia burgdorferi. Proc. Natl. Acad. Sci. USA 97:4850-4855.
40. Schreiber, G., E. Z. Ron, and G. Glaser. 1995. ppGpp-mediated regulation of DNA replication and cell division in Escherichia coli. Curr. Microbiol. 30:27-32.
41. Schwan, T. G., and J. Piesman. 2002. Vector interactions and molecular adaptations of Lyme disease and relapsing fever spirochetes associated with transmission by ticks. Emerg. Infect. Dis. 8:115-121.
42. Schwan, T. G., J. Piesman, W. T. Golde, M. C. Dolan, and P. A. Rosa. 1995. Induction of an outer surface protein on Borrelia burgdorferi during tick feeding. Proc. Natl. Acad. Sci. USA 92:2909-2913.
43. Shevchuk, N. A., A. V. Bryksin, Y. A. Nusinovich, F. C. Cabello, M. Sutherland, and S. Ladisch. 2004. Construction of long DNA molecules using long PCR-based fusion of several fragments simultaneously. Nucleic Acids Res. 32:e19.
44. Steere, A. C., J. Coburn, and L. Glickstein. 2004. The emergence of Lyme disease. J. Clin. Investig. 113:1093-1101.
45. Stevenson, B., T. G. Schwan, and P. A. Rosa. 1995. Temperature-related differential expression of antigens in the Lyme disease spirochete, Borrelia burgdorferi. Infect. Immun. 63:4535-4539.
46. Sun, J., A. Hesketh, and M. Bibb. 2001. Functional analysis of relA and rshA, two relA/spoT homologues of Streptomyces coelicolor A3(2). J. Bacteriol. 183:3488-3498.
47. Wendrich, T. M., G. Blaha, D. N. Wilson, M. A. Marahiel, and K. H. Nierhaus. 2002. Dissection of the mechanism for the stringent factor RelA. Mol. Cell 10:779-788.
48. Yang, X., and E. E. Ishiguro. 2003. Temperature-sensitive growth and decreased thermotolerance associated with relA mutations in Escherichia coli. J. Bacteriol. 185:5765-5771.(Julia V. Bugrysheva, Anto)