Development of a Single-Plasmid-Based Regulatable Gene Expression System for Borrelia burgdorferi

Abstract

We developed a single-plasmid-based regulatable protein expression system for Borrelia burgdorferi. Expression of a target gene is driven by P_ost, a hybrid B. burgdorferi ospA-tetO promoter, from a recombinant B. burgdorferi plasmid constitutively expressing TetR. The system was tested using the green fluorescent protein (GFP) as a reporter. Under noninducing conditions, recombinant B. burgdorferi cells were nonfluorescent, no GFP protein was detected, and residual, small amounts of transcript were detectable only by reverse transcription-PCR but not by Northern blot hybridization. Upon induction with anhydrotetracycline, increasing levels of GFP transcript, protein, and fluorescence were observed. This tight and titratable promoter system will be invaluable for the study of essential borrelial proteins. Since target protein, operator, and repressor are carried by a single plasmid, the system's application is independent of a particular strain background.

Since its first description in 1982 (6), the Lyme disease spirochete Borrelia burgdorferi has become a model organism for its phylum for studies on bacterial virulence mechanisms, membrane structure, physiology, and metabolism. An increasing number of molecular tools have been developed (reviewed in reference 23), including fluorescent fusion proteins for use in both cytoplasmic and extracytoplasmic compartments (10, 28).

On the basis of studies determining a minimal bacterial genome (14, 21), about 400 or more of the 1,813 genes encoded by the B. burgdorferi type strain B31 (11, 12) can be expected to be essential for growth on its rich Barbour-Stoenner-Kelly II (BSK-II) laboratory medium (2, 37). One currently available but indirect approach to test the indispensability of a B. burgdorferi gene is to attempt its disruption by an antibiotic resistance cassette; essential genes can be knocked out only if the cell is supplied with an additional functional copy in trans on a recombinant plasmid (7, 17). A more direct and elegant approach uses a tightly regulatable promoter construct either to deplete the wild-type protein or to selectively overexpress a dominant negative mutant. Environmental gene regulation by, e.g., temperature or pH is counter-indicated due to its pleiotropic effects (9, 32). Escherichia coli lac operon-based expression systems responding to induction by isopropyl β-d-1-thiogalactopyranoside have been adapted for use in B. burgdorferi by either optimizing the codon usage of the lac repressor gene lacI (4) or generating a lac operator-modified B. burgdorferi flaB promoter in a specific, engineered strain background expressing LacI (13).

Tetracycline-regulated expression is similarly well characterized on the molecular level and provides tight control and sensitive induction by compounds that diffuse passively across biological membranes without the requirement of uptake proteins. In addition, tetracycline has been used in eukaryotes as well as in both gram-positive and -negative bacteria (20, 29, 34, 36). A typical regulatory setup consists of a constitutively expressed tet repressor, TetR, binding to the tet operator, tetO, within the tet promoter, tetP. Binding of tetracycline to TetR releases the repressor, leading to expression of the reporter protein from tetP. Two modifications of this system make it particularly versatile: (i) the use of the tetracycline analog anhydrotetracycline (ATc), which has a 30-fold higher affinity for TetR and lower antibiotic activity (15) and thus allows for its use in tetracycline-sensitive bacteria and for transient induction of bacterial genes during in vivo experiments (19, 35); and (ii) the additional experimental flexibility provided by the use of revTetR, a TetR G96E L205S mutant, which binds ATc not as an inducer but as a corepressor (18). Here, we describe the successful adaptation and evaluation of a tet promoter system in B. burgdorferi. It allows for a single-plasmid-based, tight, and titratable expression of a specific target gene.

MATERIALS AND METHODS

Strains and culture conditions.

E. coli strains TOP10 (Invitrogen) or XL-10 Gold (Stratagene) were used for recombinant plasmid construction and maintenance and were grown in selective Luria-Bertani broth or on solid plates (26). B. burgdorferi B31-e2 (1) and B313 (24) are clones of type strain B31 (ATCC 35210). B313 lacks the ospA-harboring 54-kb linear plasmid lp54 (25, 39). B. burgdorferi cells were transformed by electroporation using 10 μg of plasmid DNA following established protocols (27, 33). Transformants were selected in solid BSK-II medium containing 200 μg/ml kanamycin and expanded in selective liquid BSK-II medium at 34°C in a 5% CO₂ atmosphere (2, 37). For induction of expression, anhydrotetracycline hydrochloride ([ATc] IBA GmbH, Göttingen, Germany) was added at final concentrations of 2 ng/ml to 2 μg/ml to liquid B. burgdorferi cultures with a cell density of 1 × 10⁷ cells/ml.

Recombinant plasmids.

All plasmids generated and used in this study are derivatives of pBSV2 (33), which replicates autonomously in both E. coli and B. burgdorferi and confers resistance to kanamycin. Plasmids used as target DNA for sequence overlap extension PCR (SOE-PCR) (16) were the following: (i) pASK-IBA14 (IBA GmBH, Göttingen, Germany) with the E. coli tetPO tetracycline promoter and operator region, as well as the tetR repressor gene; (ii) pBSVΦ(ospAp-gfp) (10) with the outer surface protein A (ospA) promoter P_ospA driving green fluorescent protein (cycle 3 [c3] GFP) expression; and (iii) pBSV2 harboring the flaB promoter P_flaB. SOE-PCR was carried out using either Platinum Taq or Platinum Pfx polymerase (Invitrogen) using the primers listed in Table 1. Site-directed mutagenesis was carried out using a QuikChange kit (Stratagene).

TABLE 1.

Oligonucleotides used in this study

Name	Target	Sequence (5′ to 3′)a
KpntetPO-fwd	5′ end of tetPO	GGTACCATCGAATGGCCAGATGATTAATTC
tetPO_gfp-rev	3′ end of tetPO	CAGAATTGCCCTTTCATTTTTTGCCCTCGTTATC
tetPO_gfp-fwd	5′ end of gfp	GATAACGAGGGCAAAAAATGAAAGGGCAATTCTG
gfp_flaBP-rev	3′ end of gfp	CACAAGAGGCGACAGACATCATTATTTGTAGAGCTCATC
gfp_flaBP-fwd	5′ end of flaB promoter	GAGCTCTACAAATAATGATGTCTGTCGCCTCTTGTG
flaBP_tetR-rev	3′ end of flaB promoter	CTTTTATCTAAACGAGACATCATATGTCATTCCTCCATG
flaBP_tetR-fwd	5′ end of tetR	CATGGAGGAATGACATATGATGTCTCGTTTAGATAAAAG
KpntetR-rev	3′ end of tetR	GGTACCTCATTAAGACCCACTTTCACATTTAAG
Sacgfp-rev	3′ end of gfp	TTGTAGAGCTCATCCATGCCATGTG
SacPospA-fwd	5′ end of ospA promoter	GAATTCGAGCTCAAGTCCCAAAACTGGGAC
ospAtetO-rev	3′ end of ospA promoter	TATATTCTCCTTTTTCTCTATCACTGATAGGGACAAGTATAATTATATTATAAGATTAAC
ospAtetO-fwd	5′ end of gfp	ATAATATAATTATACTTGTCCCTATCAGTGATAGAGAAAAAGGAGAATATATTATGAAAG
gfpRTPCR-fwd	5′ internal of gfp	TGGCCAACACTTGTCACTACTTTC
gfpRTPCR-rev	3′ internal of gfp	AGCTCATCCATGCCATGTGTAATC
pBRori-fwd	pBR origin of replication	GCGTAATCTGCTGCTTGC AAAC
pBRori-rev	pBR origin of replication	AAATCGACGCTCAAGTCAGAGG

^aRestriction sites are italicized.

pCRW50 carries fusions of tetPO to the c3 GFP gene and P_flaB to tetR. tetPO was amplified from pASK-IBA14 using primer pair KpntetPO-fwd and tetPO_gfp-rev. The c3 GFP gene was amplified from pBSVphi(ospAp-gfp) using the primer pair tetPO_gfp-fwd and gfp_flaBP-rev. The overlapping tetPO and c3 GFP gene amplicons were fused using the flanking primer pair KpntetPO-fwd and gfp_flaBP-rev. P_flaB was amplified from pBSV2.vsp1 (38) using the primer pair gfp_flaBP-fwd and flaBP_tetR-rev. tetR was amplified from pASK-IBA14 using the primer pair flaBP_tetR-fwd and KpntetR-rev. P_flaB and tetR were fused together by SOE-PCR using the flanking primers gfp_flaBP-fwd and KpntetR-rev. In a final SOE-PCR, the tetPO-c3 GFP gene-P_flaB-tetR cassette was fused by using the primer pair KpntetPO-fwd and KpntetR-rev and ligated into pCR2.1-TOPO (Invitrogen). The cassette was then excised with KpnI and ligated into pBSV2.

pCRW53 carries a tetO-modified P_ospA (P_ost) driving expression of the c3 GFP gene (Fig. 1). The P_ost promoter fragment was amplified from pBSVΦ(ospAp-gfp) with the primer pair SacPospA-fwd and ospAtetO-rev. The c3 GFP gene was amplified from pBSVΦ(ospAp-gfp) using the primer pair ospAtetO-fwd and Sacgfp-rev. The P_ost-c3 GFP gene fusion was obtained by SOE-PCR with flanking primers SacPospA-fwd and Sacgfp-rev. The resulting amplicon was digested with SacI and ligated with an SacI-cut pCRW50 plasmid backbone fragment.

FIG. 1.

Plasmid map of pCRW53. The plasmid is based on a pBSV2 plasmid backbone, replicates autonomously in E. coli (pBR origin of replication) and B. burgdorferi (paralogous gene families [PF] 57, 50, and 49 flanked by inverted repeats [IR]), and confers resistance to kanamycin (P_flgB-Kan^r cassette) (33). The hybrid P_ost promoter consists of the B. burgdorferi ospA promoter P_ospA and the 19-bp tet operator (tetO; gray box) inserted downstream of its +1 transcriptional start site, replacing 19 nucleotides of the 5′ untranslated ospA mRNA shown above tetO. +1, −10, and −35 sequences of P_ospA are indicated in bold. The formyl-methionyl (fMet) ospA start codon is indicated in bold italics. The tetracycline repressor gene tetR is driven by the B. burgdorferi P_flaB promoter.

Epifluorescence microscopy and flow cytometry.

Cultured B. burgdorferi cells were harvested, washed once, and resuspended in phosphate-buffered saline containing 5 mM MgCl₂. For an initial qualitative analysis, cells were observed under epifluorescence using a Nikon Eclipse E600 microscope fitted with a V-2A and fluorescein isothiocyanate-HYQ filter block and a Retiga EXi camera (QImaging). For a quantitative analysis, 200 μl of cells at approximately 5 × 10⁷ cells/ml were subjected to flow cytometry at 100 lb/in² using a 100-μm nozzle at the slowest flow rate on a BD LSRII instrument equipped with FACSDiva, version 4.4, software (BD Biosciences). Gating was determined by plotting log forward scatter versus log side scatter, using running buffer alone to determine the forward scatter threshold. Nonrecombinant B. burgdorferi cells were used to determine background fluorescence. About 100,000 events were counted for each sample. The FlowJo program suite, version 7.2.2 (Treestar), was used for data analysis.

Protein gel electrophoresis and immunoblot analysis.

Proteins were separated by sodium dodecyl sulfate-12% polyacrylamide electrophoresis and visualized by Coomassie blue staining. For immunoblot analysis, proteins were electrophoretically transferred to nitrocellulose membranes (Immobilon-NC; Millipore) using a Transblot-SD Semi-Dry Transfer Cell (Bio-Rad) as described previously (39). Membranes were blocked and incubated with antibodies in 5% nonfat dry milk, 20 mM Tris-500 mM NaCl, and 0.05% Tween 20 as described previously (38). Antibodies used were anti-TetR rabbit polyclonal antiserum (1:500; Abcam), anti-GFP rabbit polyclonal rabbit antiserum (1:3,000; Invitrogen), or monoclonal antibody against B. burgdorferi FlaB (1:10; H9724) (3). Secondary antibodies were alkaline phosphatase-conjugated goat anti-mouse heavy and light chain immunoglobulin G or mouse anti-rabbit immunoglobulin G (heavy chain) (Sigma). The alkaline phosphatase substrate for chemiluminescent detection was CDP-Star (GE Healthcare). A Fujifilm LAS-4000 Luminescent Image Analyzer was used for data acquisition and analysis.

RNA isolation and Northern blot analysis.

Total RNA was isolated from 35-ml cultures using an RNeasy Mini Kit (Qiagen). Prior to isolation, the RNAs were fixed using an RNA Protect Kit (Qiagen). RNA concentrations were measured using a NanoDrop 1000 spectrophotometer (Thermo Scientific). Total RNA (1.0 μg) was fractionated in a 1.2% formaldehyde-agarose gel and transferred to an Immobilon-NY+ membrane (Millipore) by upward capillary transfer (26). RNA ladders (0.5 to 10 kb or 0.24 to 9.5 kb; Invitrogen) served as size standards. A DNA probe was generated by PCR using the primer pair ospAtetO-fwd and Sacgfp-rev (Table 1) and pBSVΦ(ospAp-gfp) as a template. Probe labeling and Northern blot hybridizations were performed using the Gene Images AlkPhos Direct Labeling and Detection System with CDP-Star (GE Healthcare) according to the manufacturer's instructions. A Fujifilm LAS-4000 Luminescent Image Analyzer was used for data acquisition and analysis.

RT-PCR.

Isolated total RNA was treated with DNase I (Invitrogen). Reverse transcriptase PCR (RT-PCR) reaction mixtures consisted of 0.25 μg of total DNA-free RNA and primers gfpRTPCR-fwd and gfpRTPCR-rev (Table 1) (predicted amplicon of 539 bp) with the GeneAmp EZ rTth RNA PCR kit (Applied Biosystems), according to the manufacturer's instructions. Reaction mixture incubation conditions were 30 min at 55°C for reverse transcription, followed by 2 min at 94°C, 40 cycles of 1 min at 94°C and 1 min at 55°C, and 7 min at 60°C. To check for DNA contamination, Mn(OAc)₂ was omitted for the control (without the RT step). To conserve rTth polymerase, alternative control reactions were carried out with Taq polymerase (New England Biolabs) using pBRori-fwd and pBRori-rev (Table 1) (predicted amplicon of 533 bp), two primers specific for an untranscribed region of pCRW53. Reaction mixture incubation conditions were identical to those described above. A DNA ladder (1 Kb Plus DNA ladder; Invitrogen) served as a size marker.

RESULTS AND DISCUSSION

Given the potential versatility of a tetPO- and TetR-based expression system, we decided to adapt it for use in B. burgdorferi. First, we assessed whether B. burgdorferi was resistant to the ATc concentrations typically used for induction in a preliminary experiment. B. burgdorferi cells did not show any growth defect in the presence of the standard ATc concentration for induction (0.2 μg/ml) or at ATc concentrations that were 100-fold higher. Next, we tested whether the tet promoter is active and regulatable by ATc in B. burgdorferi. E. coli and B. burgdorferi transformants carrying pCRW50 (tetPO-gfp-P_flaB-tetR) were cultured in the presence or absence of ATc. Transformed E. coli cells were green fluorescent under both conditions, suggesting that the tet promoter was leaky due to insufficient levels of TetR. Transformed B. burgdorferi cells were nonfluorescent when grown in BSK-II medium without ATc. In the presence of 0.2 μg/ml ATc, a very low level of green fluorescence was observed by epifluorescence microscopy using a Nikon V-2A filter (data not shown). The level of fluorescence remained low in the presence of 2 and 20 μg/ml ATc. This indicated that the tet promoter was active at a very low level in B. burgdorferi and could be repressed by TetR. The low activity of nonborrelial promoters had been a major obstacle in the initial development of antibiotic markers (5, 31) and was therefore expected.

We hence set out to modify a strong B. burgdorferi promoter that drives expression of the outer surface protein OspA (P_ospA). The −35/−10 spacer region in P_ospA contains 16 nucleotides (30). It was therefore not entirely surprising that insertion of the 19-bp tetO sequence between the promoter's −10 and −35 sequence disrupted activity (data not shown). Replacing tetO in the construct with a truncated 16-bp sequence still containing the palindromic sequence interacting with TetR (22) led to a leaky phenotype (data not shown). This suggested that the truncated tetO sequence did not bind TetR efficiently under noninducing conditions. Thus, we repositioned the operator in subsequent constructs.

In pCRW53, we replaced 19 residues downstream of the P_ospA +1 transcriptional start site with tetO (Fig. 1), resulting in the hybrid P_ost promoter driving the expression of gfp. B. burgdorferi cells carrying pCRW53 derived from both strain B31 clones B31-e2 and B313 were brightly green fluorescent in the presence of ATc yet nonfluorescent in the absence of the inducer. This suggested that the hybrid P_ost promoter, in combination with a constitutively expressed TetR repressor, provided a tight and tetracycline-inducible B. burgdorferi expression system. We therefore decided to evaluate its induction properties in more detail.

Northern blotting and RT-PCR were used to assess regulation on the transcriptional level. For Northern blot analysis (Fig. 2A and B), total RNA was probed with a PCR-generated probe specific for the gfp transcript. A significant signal above background was detected in the RNA from cells grown in the presence of 75 ng/ml ATc. Three major bands were detected by the probe, approximately 0.8 kb, 1.2 kb, and 1.8 kb in size. The 0.8-kb band corresponds to the predicted size of the gfp transcript. The detected 1.8-kb band most likely represents a cotranscript of gfp and tetR (1.0 kb in size including the upstream P_flaB sequence) since both operons face the same direction and are not separated by a transcription terminator sequence. The origin of the intermediate 1.2-kb band is currently unclear. Interestingly, the gfp transcript levels upon P_ost induction remained significantly lower than those detected in RNA isolated from cells harboring pBSVΦ(ospAp-gfp) (10), where gfp is driven by the unmodified P_ospA.

FIG. 2.

Transcriptional analysis of P_ost. (A and B) Northern blot analysis of B31-e2 (A) and B313 (B) transformants. One microgram of total RNA was separated in a formaldehyde agarose gel and visualized with ethidium bromide to ensure equal loading (top panel). The blotted RNA was probed with a PCR-generated gfp fragment (bottom panel). Arrowheads indicate the 0.8-kb and 1.8-kb gfp-tetR (*) transcripts. In panel B, a shorter time exposure of the two rightmost RNA lanes is shown to the right to better visualize the P_ospA transcripts. bg, background B. burgdorferi B31-e2 or B313 strains; P_ospA, background strains carrying pBSVΦ(ospAp-gfp); P_ost, background strains carrying pCRW53. Numbers below the bracket indicate the ATc concentrations in μg/ml. m, size marker in kb (0.5 to 10 kb in panel A or 0.24 to 9.5 kb in panel B; RNA ladders, Invitrogen). (C and D) RT-PCR analysis of B31-e2 (C) or B313 (D) transformants. A total of 0.25 μg of DNase-treated total RNA was used as a template for gfp-specific oligonucleotide primers. To exclude potential residual DNA in the sample, control reactions either omitting the RT reaction step (−RT control) or using pCRW53 DNA-specific primers (PCR control) were run in parallel. pCRW53, positive PCR control using pCRW53 plasmid as a template; m, marker in bp (1 kb Plus DNA ladder; Invitrogen). Other sample labeling is identical to that in panels A and B.

Total DNase-treated RNA was used to assay transcription levels by RT-PCR (Fig. 2C and D). Reactions omitting the RT reaction step or using DNA-specific primers did not yield amplicons, demonstrating that the RNA preparation was free of detectable amounts of DNA. As expected from the Northern blot analysis, gfp transcripts were detected in cells grown at ATc concentrations above 50 ng/ml. Most importantly, only residual, low levels of transcript were detected under noninducing conditions for both strain backgrounds. Together, this indicated that the P_ost promoter is also regulated at the transcriptional level and silenced by TetR under noninducing conditions.

Western blot and flow cytometry analyses were performed to measure protein expression upon induction. In Western blots with whole-cell lysates and specific antibodies, no GFP was detectable in cells harboring pCRW53 grown at ATc concentrations of 10 ng/ml and below (Fig. 3A). Weak GFP bands were observed in lysates from cells grown at 20 ng/ml ATc, with the band intensities increasing significantly toward higher ATc concentrations. In line with the above Northern blot data, the TetR signal seemed to increase in intensity as well. Interestingly, the above described higher levels of gfp transcript in cells harboring pBSVΦ(ospAp-gfp) (10) (Fig. 2A and B) translated into only marginally higher protein levels (Fig. 3A). In a Western blot analysis of whole-cell lysates obtained over a 72-h time course, GFP became detectable within 8 h of induction and reached an apparent plateau at around 40 h postinduction with 200 ng/ml ATc (Fig. 3B). Similarly, green fluorescence was detected by epifluorescence microscopy in pCRW53-harboring B31-e2 cells at 8 h postinduction (data not shown).

FIG. 3.

Western blot analysis of P_ost-driven GFP expression. B. burgdorferi whole-cell lysates were separated by sodium dodecyl sulfate-polyacrylamide electrophoresis. Samples were normalized for the constitutively expressed flagellar protein FlaB based on densitometry of a Coomassie blue-stained gel. (A) An equally loaded gel was then used for Western blot analysis using antibodies against FlaB, TetR, and GFP. bg, background B. burgdorferi B31-e2 (upper three panels) or B313 (lower three panels) strain; P_ospA, background strains carrying pBSVΦ(ospAp-gfp); P_ost, background strains carrying pCRW53. Numbers below the bracket indicate the ATc concentrations in μg/ml. (B) Time course of GFP expression by B313 carrying pCRW53. Harvested culture volumes were adjusted to obtain equal cell numbers prior to further densitometry-based loading adjustment as described above. GFP was detected by Western immunoblotting. Numbers indicate hours postinduction with 0.2 μg/ml ATc.

Three independent flow cytometry experiments reproducibly showed an analogous ATc-dependent titration of green fluorescence. Results from a representative experiment are shown in Fig. 4A, and a cumulative titration response curve is shown in Fig. 4B. The fluorescence of uninduced cells was indistinguishable from the background fluorescence of nonrecombinant B. burgdorferi cells (Fig. 4A and B). As in the Northern and Western blot analysis, a major response of P_ost was detected between 50 and 75 ng/ml ATc. The cell populations' mean fluorescence intensity (MFI) continued to increase toward higher ATc-concentrations, plateauing toward 2 μg/ml ATc at average MFIs of 8,150 and 2,170 arbitrary units (AU) in B31-e2 and B313 cells, respectively. B31-e2 and B313 cells harboring pBSVΦ(ospAp-gfp) showed equivalent average MFIs of 13,630 and 13,290 AU, respectively. Together, these data indicate that P_ost in both genetic contexts is close to maximally induced at ATc concentrations of 2 μg/ml but remains attenuated at least 1.5-fold compared to its parent promoter. This attenuation may be in part explained by the following: (i) the tetO-mediated modification of the P_ospA sequence at the transcriptional start site, potentially affecting mRNA transcription and stability as well as transcription factor and ribosome binding; and (ii) the above described concurrent increase in the tetR transcript (Fig. 2), which leads to a slightly higher level of repressor in the induced cells (Fig. 3). The observed attenuation of P_ost in B313 remains to be explained.

FIG. 4.

Flow cytometry analysis of P_ost-driven GFP expression. (A) Flow cytometry diagram of a representative experiment. The percentage of total flow cytometry events for each B. burgdorferi cell population is plotted against the fluorescence intensity (FI) in AU. Curves are labeled according to the legend on the right. bg, background B. burgdorferi strain B31-e2; P_ospA, background strain carrying pBSVΦ(ospAp-gfp); P_ost, background strain carrying pCRW53. Numbers inside the bracket indicate the ATc concentrations in μg/ml. (B) The MFI of P_ost-driven GFP expression was determined for each cell population. P_ost-driven MFIs were expressed as a percentage of the P_ospA-driven MFI in each particular background, and the arithmetic means from three separate experiments for each background were plotted against the ATc concentration. Error bars indicate standard deviations. The scale of the left half of the graph is extended to better illustrate MFI changes at low ATc concentrations.

A previously described hybrid tetracycline-responsive Borrelia promoter, which placed the tet operator upstream of the −10 and −35 sequences of the B. burgdorferi bmpA promoter (8), was shown to be inducible but active even in the absence of inducer. In contrast, the hybrid P_ost promoter developed here represents a significantly tighter and titratable expression system that will be an invaluable genetic tool in generating conditional Borrelia mutants. It will facilitate the depletion of suspected essential proteins or the overexpression of dominant negative mutants. In combination with the flacp promoter developed by Samuels and colleagues (13), P_ost will allow for the independent, simultaneous regulation of two proteins, operons, or regulons. Since all the required elements are encoded on a single recombinant plasmid, the “posting” of target genes should be easily adaptable to a variety of genetic Borrelia backgrounds.

Future development of this promoter system will focus on its broader applicability in different experimental settings by addressing its current limitations. P_ost attenuation due to cotranscription of tetR on pCRW53 may be largely blocked by insertion of a transcription terminator upstream of the tetR operon, permitting higher expression levels of target genes. Since applications of the current hybrid ospA-tetO promoter will likely be limited to environments where wild-type OspA is expressed, i.e., in culture and in the tick host environment, constitutive borrelial promoters will be altered to expand the system's usefulness. Also, native chromosomal and plasmid promoters will be modified in situ to render them responsive to tetracycline repressor molecules expressed from a recombinant plasmid.