Regulation of Expression of the Borrelia burgdorferi β₃-Chain Integrin Ligand, P66, in Ticks and in Culture

Abstract

Borrelia burgdorferi is maintained in an infection cycle between mammalian and arthropod hosts. Appropriate gene expression by B. burgdorferi at different stages of this cycle is probably essential for transmission and establishment of infection. The B. burgdorferi β₃ integrin ligand P66 is expressed by the bacteria in mammals, laboratory culture, and engorged but not unfed ticks. No in vitro culture conditions in which P66 expression reflected that in the unfed tick were found, suggesting that there are aspects of B. burgdorferi-tick interaction that remain unexplored.

An important recent development in Lyme disease research has been the demonstration that, like other pathogenic bacteria, Borrelia burgdorferi alters the expression of a number of genes in response to environmental cues (3, 19-21, 26, 27). This is of great interest because the Lyme disease spirochetes must be able to survive in mammalian hosts and in arthropod vectors, which represent very different environments. B. burgdorferi is also remarkable for its ability to cause persistent disseminated infection in immunocompetent mammals (24) and therefore may alter gene expression as a means to evade clearance by the immune system. Even within a particular arthropod or mammalian host, the spirochete encounters different tissues, which vary in chemical content and complexity. Each of these environments may influence the profile of bacterial gene expression; appropriate gene expression is likely to be required for survival within the different environments encountered by B. burgdorferi and for transmission between ticks and mammals.

B. burgdorferi is acquired by a larval tick by feeding on an infected mammal. The tick then molts to the nymphal stage and takes in a second blood meal. Between feedings, B. burgdorferi resides in the tick midgut, in which the pH is slightly alkaline and approximately that of human blood (pH ∼7.4) (26). The tick is subject to greater variations in ambient temperature when it is unattached than while it is feeding, but the temperature range encountered by the unattached tick is generally below that of the mammalian body. While attached and feeding, the tick is warmed to the surface temperature of the mammal, the pH of the midgut falls to slightly below neutral (pH, ∼6.8) (26), and the spirochetes migrate from the midgut to the salivary glands for transmission. The saliva of the tick is quite alkaline (pH ∼9.7) (26).

One protein that is differently expressed in ticks relative to mammals is the lipoprotein OspA, which is expressed at high levels by B. burgdorferi in the midguts of unfed ticks and under standard laboratory culture conditions but is either not expressed or expressed by a minority of the bacteria in the midguts of feeding ticks or entering a mammal (13, 21, 22). This finding is consistent with the proposed role of OspA in the attachment of B. burgdorferi to tick midgut cells (14). In contrast to that of OspA, expression of a second lipoprotein, OspC, is increased by B. burgdorferi in the midgut and salivary glands of a feeding tick as the bacteria are being transmitted to the mammal (9, 13, 20, 21).

P66, also known as OMS66, is a candidate B. burgdorferi ligand for mammalian β₃-chain integrins (5). Previous work had demonstrated that P66 is expressed in mammalian hosts, as a large percentage of Lyme disease patient sera recognize the protein (7, 12). No previous studies have evaluated P66 expression in ticks, which is of interest for several reasons. First, P66 has been identified as a porin, suggesting an important role in nutrient acquisition (23). If P66 is involved in nutrient assimilation, expression of the protein in the tick midgut may provide clues to the metabolic activity of the B. burgdorferi cells residing therein. Second, the potential role of P66 as an adhesin in the tick midgut has never been explored, but integrins are likely to be present in ticks and expression of P66 would be required for any Borrelia-integrin interaction to occur in the tick vector. Finally, it is possible that P66 has a functional role in, and is expressed in, only the mammalian environment.

Analysis of P66 expression by B. burgdorferi in the tick midgut.

The expression of P66 in Ixodes scapularis ticks, which transmit B. burgdorferi between mammalian hosts, was evaluated at key points in the tick life cycle using indirect fluorescent antibody staining (Fig. 1 and 2). To ensure that P66 was detectable by this protocol, expression was first evaluated with in vitro-cultivated spirochetes. Strain B31 (2) (uncloned) cultured in BSK-H complete medium (Sigma Chemical, St. Louis, Mo.) was smeared onto glass slides. The smears were allowed to dry at room temperature and fixed with gentle flame heat and then with acetone for 20 min. Spirochetes were stained first with a rabbit antiserum directed against the integrin-binding domain of B. burgdorferi strain N40 P66 (5) (1:50 dilution) and then with goat anti-rabbit immunoglobulin G (IgG) rhodamine isothiocyanate (1:25 dilution; Kirkegaard & Perry Laboratories, Inc., Gaithersburg, Md.). The slides were then stained with mouse monoclonal antibody H9724 (hybridoma supernatant, neat) directed against flagellin (1) followed by goat anti-mouse IgG fluorescein isothiocyanate (1:50 dilution; Kirkegaard & Perry). The smears were immersed in each antibody and incubated at 37°C for 30 to 45 min, washed in phosphate-buffered salin (PBS), and dried before the next antibody was applied. After being stained, the smears were covered with glycerol in PBS and a glass coverslip. All samples were examined with a Nikon Eclipse 800 epifluorescence microscope at a ×400 magnification with fluorescein or rhodamine emission filters to detect spirochetes and determine the presence of P66. All B. burgdorferi strain B31 cells examined expressed both P66 and flagellin. Similar results have been obtained with strain N40 (clone D10/E9 [4, 6]) cultivated in both BSK and MKP media (not shown). MKP medium is slightly less rich than BSK, but B. burgdorferi organisms grown in MKP containing human serum bind to integrin α_IIbβ₃ more efficiently than bacteria grown in BSK (4); consequently, MKP is the standard medium used for studies of integrin binding.

FIG. 1.

Expression of P66 in the midguts of Ixodes scapularis ticks. Midguts from ticks at various stages of the feeding cycle were fixed and stained with anti-P66 and antiflagellin antibodies. (A) Midgut of an unfed tick showing spirochetes stained with antiflagellin; (B) same field as in panel A, stained with anti-P66; (C) midgut of a fed nymphal tick showing a cluster of spirochetes stained with anti-P66; (D) single spirochete in a fed nymph stained with anti-P66. Scale bars = 30 μm.

FIG. 2.

Quantification of spirochetes expressing P66 at different stages of the tick feeding cycle. Midguts from ticks at various stages of the feeding cycle were fixed and stained with anti-P66 and antiflagellin antibodies. All spirochetes appeared to stain with the antiflagellin antibody; the percentage of the number of tick midguts examined at each time point that also stained with the anti-P66 antibody is shown as an average and standard deviation. The numbers of individual tick midguts examined at each time are indicated in parentheses.

The expression of P66 in tick midguts was then examined. Ixodes scapularis ticks were from uninfected and infected (with B. burgdorferi strain B31) colonies maintained at the Rocky Mountain Laboratories, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Hamilton, Montana. Ticks were fed on white mice, Mus musculus, in a room with 12 h of light and 12 h of dark. Ticks not feeding were maintained at 98% relative humidity and 21 to 22°C with 14 h of light and 10 h of dark. We examined spirochetes in ticks that had been infected as larvae while feeding on an infected mouse, which in turn had been infected by tick bite, as previously described (15). Engorged larvae were examined shortly after completing their blood meal (day 0) and 7 and 16 days later. Unfed and freshly engorged nymphs were also examined. The midgut was removed from each tick and placed into a 10-μl drop of PBS on a glass microscope slide. The organ was teased apart and smeared with microforceps while being viewed with a dissecting microscope. The smears were fixed, stained, and viewed as described above. A total of 3,798 spirochetes were examined in 41 ticks.

Virtually all (99.34% ± 1.95%) of the B. burgdorferi strain B31 cells in the midguts of larval ticks immediately after drop off expressed P66 (Fig. 1 and 2), demonstrating that the protein is expressed as the bacteria are acquired from the mammal by the tick in the first blood meal. P66 expression in the midguts of larval ticks was also examined at 7 and 16 days after drop off, as the ticks digested the blood meal and progressed toward the molt to the nymphal stage. At 7 days, 91.18% ± 7.18% of the B. burgdorferi cells detectable with the antiflagellin antibody also expressed P66, while at 16 days, only 20.9% ± 11.25% of the B. burgdorferi cells expressed P66 (Fig. 2). In the midguts of unfed nymphal ticks, no P66 was detected (Fig. 1 and 2). As the nymphal ticks fed on an uninfected mouse, P66 expression was sharply up-regulated, with 76.1% ± 18.56% of the B. burgdorferi cells in the midgut expressing the protein at drop off (Fig. 2). Using different techniques, two other groups also recently reported that P66 is expressed by B. burgdorferi in feeding nymphal ticks (11, 18), but unlike with this study, neither group reported quantification of P66 expression by B. burgdorferi in ticks at multiple points in the life cycle.

Regulation of P66 expression by temperature and pH.

Environmental conditions that might replicate what occurs in unfed versus feeding ticks and therefore might regulate P66 expression by B. burgdorferi were investigated with in vitro culture. All B. burgdorferi sensu lato strains tested produce the protein under standard in vitro culture conditions (pH 7.5, 34°C) (Fig. 3), but the expression of this protein had not previously been examined systematically under alternative culture conditions. Both pH and temperature affect the expression of other B. burgdorferi proteins, so the effects of each of these variables were determined for P66 expression in B. burgdorferi strains N40 and B31. Cultures of each strain at a density of approximately 10⁸ spirochetes/ml were diluted 1:100 into fresh BSK medium at the standard pH (7.5) or at a pH adjusted to 6.5 or 8.5. Duplicate cultures were then placed at either 34°C or ambient temperature (∼24°C) and grown to approximately 10⁸ bacteria/ml. The cultures at pH 8.5 were considerably less healthy (as judged by bacterial motility and morphology) than those at pH 6.5 or 7.5 and were therefore collected at a density of approximately 2.5 × 10⁷ bacteria/ml. The cells were harvested by centrifugation and analyzed for expression of P66 and flagellin by immunoblot analysis and reverse transcriptase (RT)-PCR. For the immunoblots, B. burgdorferi cells were washed once with PBS supplemented with 0.2% (wt/vol) bovine serum albumin (BSA) and then twice with PBS alone. The pellet was resuspended in PBS. The total protein content of approximately 2.5 × 10⁷ bacteria was fractionated by gel electrophoresis in the presence of sodium dodecyl sulfate and β-mercaptoethanol (10). For each sample set, one gel was stained with Coomassie brilliant blue and a duplicate gel was transferred to a polyvinylidene difluoride membrane and probed with rabbit anti-P66 antiserum (5) (1:20,000) and with antiflagellin antibody H9724 hybridoma supernatant (1) (1:20,000). In some experiments, the membranes were also probed with a rabbit antiserum directed against OspC (21) (dilution, 1:10,000). The blots were then probed with anti-rabbit IgG and anti-mouse IgG antibodies conjugated to alkaline phosphatase (dilution, 1:10,000; Promega, Madison, Wis.). Reactive bands were revealed with a mixture of 5-bromo-4-chloro-3-indolylphosphate (50 μg/ml) plus nitroblue tetrazolium (100 μg/ml; Sigma).

FIG. 3.

P66 is expressed by diverse Lyme spirochetes in in vitro culture. Representatives of B. burgdorferi sensu stricto (N40, HB19, and G39/40), Borellia garinii (PBr and VS102), and Borellia afzelii (VS461) were grown to late exponential phase in MKP medium supplemented with human serum. The total protein contents of approximately 2.5 × 10⁷ to 5 × 10⁷ bacteria were separated on 12.5% polyacrylamide gels under denaturing conditions. One gel was stained with Coomassie blue; the second was transferred to a polyvinylidene difluoride membrane and probed with anti-P66 and antiflagellin (anti-fla) antibodies. The positions of molecular mass markers (lane M) in kilodaltons are shown on the left.

Levels of P66 protein were not grossly affected by either temperature or pH or by any combination of these two factors tested (Fig. 4). Slight differences (2- to 3-fold) were revealed when 10-fold-fewer spirochetes were analyzed by immunoblotting (data not shown). For both B. burgdorferi strains, OspC was increased at 34 rather than 24°C, consistent with results reported previously (20, 26).

FIG. 4.

Effects of growth temperature and pH on P66 expression in B. burgdorferi strains N40 and B31. Immunoblots were processed as described for Fig. 3. RNA samples were treated with (+) or without (−) RT in the presence of random hexamer primers, and the resulting cDNAs were amplified by PCR with gene-specific primers. The PCR products were separated by electrophoresis through 10% polyacrylamide gels in Tris-borate-EDTA buffer. Lanes M, DNA size markers; lanes NTC, no template control for the PCR; lanes +control, total DNA from the appropriate B. burgdorferi strain as the template for the PCR. Quantitative RT-PCR analyses of the same samples are shown in Table 1. anti-fla, antiflagellin antibody. Growth temperatures are in degrees Celsius.

For RT-PCR analysis of gene expression, the bacterial cells were washed once in PBS and then resuspended in PBS and extracted with the Trizol reagent (Gibco, Gaithersburg, Md.) by using 1.0 ml of Trizol per 10⁸ bacteria. The RNA fraction was isolated according to the manufacturer's instructions. Residual DNA in the RNA preparations was digested with RNase-free DNase (QIAGEN, Valencia, Calif.), and the RNA was repurified by use of an RNeasy kit (QIAGEN). Yields of RNA were estimated by measuring the absorbance at 260 nm. First-strand synthesis was performed with Superscript II RT (Gibco) and random hexamer primers (Tufts University Protein and Nucleic Acid Core Facility). Each cDNA sample was then amplified by PCR with primers that target p66 (5′-ATTCTCTTGATTTCTGTTGCCCGTG and 5′-CACTAAAAGCGGAAGGCAAAAAAGGCGA) and flaA (5′-AAAGTCACACAGTTCAAAAGAGC and 5′-GATTCTTCAGGTTTTTCACTCTC). Products were analyzed by electrophoresis through 10% acrylamide gels in Tris-borate-EDTA buffer. For quantitative RT-PCR, the cDNA samples were amplified by use of an ABI Prism 7700 sequence detection system and the SYBR Green PCR master mix of the manufacturer (PE Biosystems/ABI, Foster City, Calif.) with the same primers. All quantitative PCRs included a standard curve of known amounts of genomic DNA isolated from the appropriate B. burgdorferi strain. By this method, approximately 1.5 genome equivalents of DNA were easily detected by both primer sets. Although this approach does not allow the precise quantification of the number of RNA molecules for either template in a given sample, it does allow the assessment of how the relative levels of the two transcripts change with varying environmental conditions.

As was true for the immunoblot analyses of P66 and flagellin expression, RT-PCR analyses of p66 and flaA transcript levels performed under nonquantitative conditions revealed no gross differences between any of these in vitro culture conditions (Fig. 4). When quantitative RT-PCR was employed to analyze the same samples, however, several differences were revealed (Table 1). In both B31 and N40 cells, increasing pH was associated with decreasing p66/flaA transcript ratios. Interestingly, the two B. burgdorferi strains showed opposite responses to growth at different temperatures. In strain N40, the p66/flaA ratios at all pH values were higher when the bacteria were grown at 34 than at 24°C, while the opposite was observed for strain B31. One possible explanation for this result is that there are strain variations in the responses of B. burgdorferi to environmental signals in vitro. The fact that strains N40 and B31 also differ in in vitro levels of expression of the fibronectin-binding protein BBK32 (17, 25) supports this hypothesis. The p66/flaA ratios for strain N40 grown in MKP medium were also higher at 34 than at 24°C (data not shown). In MKP medium at 34°C, however, the p66/flaA ratio for strain N40 increased with increasing pH. The relative levels of expression of P66 and p66 (protein and mRNA, respectively) at the two temperatures were consistent with previous results that demonstrated that strain N40 binding to integrin α_IIbβ₃ is more robust when the bacteria are grown at 34°C than at ambient temperature (4). Although p66 expression at the transcriptional level is affected by pH and temperature, these changes are not sufficient to cause gross changes at the protein level. Under the in vitro culture conditions said to mimic those of unfed ticks (pH 7.5 and ambient temperature [19, 26]), strain B31 expressed higher levels of p66 than under the conditions thought to mimic those of mammals. The conditions thought to mimic those in the unfed tick clearly do so only to a limited extent, in that P66 expression was not abolished in either of the B. burgdorferi strains that we examined.

TABLE 1.

Expression of p66 and flaA as quantified by real-time RT-PCR

pH	p66/flaA ratioa
	N40		B31
	24°C	34°C	24°C	34°C
6.5	9.46	41.05	21.26	7.41
7.5	7.53	38.26	8.97	2.09
8.5	4.17	12.04	1.76	1.11

^aB. burgdorferi strains N40 (clone D10E9) and B31 (uncloned) were grown in BSK complete medium (adjusted to the indicated pH values) to a density of approximately 0.5 × 10⁸ to 1 × 10⁸/ml. RNA extraction, cDNA synthesis, and quantitative PCR were performed as described in the text. Known amounts of genomic DNA from both strains were amplified in parallel. ABI sequence detector software generated critical threshold (C_t) values for each sample, and the average of results for three replicates was calculated. Based on the standard curve (C_t versus the number of B. burgdorferi genomes per reaction) generated for each primer set for each set of reactions, the cDNA content of each unknown sample was generated based on its average C_t value. The flaA signal was normalized to 1 for calculation of p66/flaA ratios.

Regulation of P66 expression by growth phase and nutrient limitation.

Bacterial gene expression can also be influenced by medium constituents, culture density, and time in post-exponential phase. B. burgdorferi strain N40 inoculated into MKP medium adjusted to pHs 6.5, 7.5, and 8.0 was harvested for analysis of P66 expression at several time points during incubation at 34 and 24°C, with a particular focus on the transition from late exponential phase to stationary phase, a time of important changes in gene expression in other bacteria. This transition to stationary phase is likely to reflect the status of B. burgdorferi in the midgut of a fed larval tick as it progresses to the molt, particularly in days 12 to 20 postrepletion (16). In contrast to what was observed with BSK medium, cultures in MKP medium were much less healthy (as assessed by bacterial motility and morphology) at pH 6.5 than at the two higher values. The reasons for this difference have not been identified. Through the course of this experiment, the levels of P66 protein (by immunoblotting) did not grossly vary in comparison to those of either flagellin (Fig. 5, showing pH 7.5 samples) or total protein (data not shown). Similarly, conventional RT-PCR analyses of these samples did not reveal any obvious differences in p66/flaA transcript levels (data not shown). Quantitative RT-PCR analysis, however, did reveal a decrease in the p66/flaA transcript levels with increasing time in culture, i.e., during postexponential growth (Table 2). The major conclusion from these results is that, although p66 mRNA levels do decline, neither the p66/flaA ratio nor P66 protein levels drop to undetectable levels as the bacteria enter stationary phase.

FIG. 5.

P66 expression in late-exponential to stationary-phase growth. B. burgdorferi strain N40 was grown at 34°C in MKP medium to the densities indicated and then harvested for protein and RNA analyses. Protein gel electrophoresis and immunoblotting were performed as described for Fig. 3. Quantitative RT-PCR data from the same samples are shown in Table 2. anti-fla, antiflagellin antibody.

TABLE 2.

p66/flaA expression with increasing time in in vitro culture^a

No. of days postinoculation	No. of bacteria per ml (107)	p66/flaA ratio
1	4.48	7.65
2	6.2	3.39
3	8.3	1.77
4	9.48	0.83
6	9.43	0.45
8	6.68	0.35

^aB. burgdorferi strain N40 (clone D10E9) was inoculated into MKP medium and incubated at 34°C for the number of days indicated. The bacteria were enumerated by dark-field microscopy. After reaching saturation, this strain clumps, decreasing the apparent culture density. p66 and flaA transcript levels were quantified by real-time RT-PCR as described for Table 1.

The expression of p66 at the transcriptional level is affected by temperature, pH, and time in culture, but none of these in vitro culture variations were found to affect P66 protein levels as dramatically as was observed in the tick. The potential roles of mammal-derived medium components, which are also present in the blood meal, were therefore tested. B. burgdorferi strain N40 was grown to late exponential phase (∼7.5 × 10⁷ cells/ml) in MKP medium and then collected by centrifugation and resuspended in MKP or the same base medium devoid of the serum, gelatin, and BSA that are normally constituents of the medium. Although the lifespan of B. burgdorferi is limited under these conditions, viable bacteria were harvested at 24, 48, and 72 h after transfer into this mammalian-component-free medium, and P66 expression was analyzed. Again, no significant differences were discernable at the protein level (Fig. 6). When the samples were analyzed by quantitative RT-PCR (Table 3), the p66/flaA ratio was considerably lower in bacteria grown in mammalian-component-free MKP than in standard MKP medium at 34°C at the 24-h time point, but the difference vanished at later time points and the cultures at 24°C showed little difference. In general, the ratio decreased with increasing time, which, if it were possible to continue the experiment with living bacteria, might begin to reflect what occurs as a larval tick digests its meal and progresses to the nymphal stage.

FIG. 6.

Effects of mammal-derived medium components on P66 expression. B. burgdorferi strain N40 was inoculated from a culture at approximately 7.5 × 10⁷ bacteria per ml into MKP medium with human serum, gelatin, and BSA (+) or into the same medium base without these mammal-derived components (−). After 24, 48, or 72 h, the bacteria were prepared for immunoblot and RNA (Table 3) analyses. anti-fla, antiflagellin antibodies.

TABLE 3.

p66 and flaA expression in medium devoid of serum, gelatin, and BSA^a

Growth temp (°C)	No. of days postinoculation	p66/flaA ratio
		M+	M−
34	1	119	11
	2	6	5
	3	0.43	0.71
24	1	1	0.56
	2	0.59	0.63
	3	0.53	0.77

^aB. burgdorferi strain N40 was grown to a density of approximately 7.5 × 10⁷ bacteria/ml in standard MKP medium containing serum, gelatin, and BSA; pelleted by centrifugation; and then resuspended either in the same medium (M+) or in MKP devoid of mammalian-component derived serum, gelatin, and BSA (M−). After incubation for the times and at the temperatures indicated, the bacteria were harvested, RNA was isolated, and gene expression was analyzed by quantitative RT-PCR.

Conclusions.

P66, a B. burgdorferi ligand for β₃-chain integrins (5), is expressed in the mammalian host, as evidenced by the immune responses mounted against this protein by Lyme disease patients (7, 12). In addition, antibodies directed against native P66 are protective against B. burgdorferi infection (8). We have now demonstrated that P66 is expressed as B. burgdorferi is acquired from a mammalian host by the vector tick and that expression wanes to reach undetectable levels in the spirochetes in unfed nymphal ticks. The protein is again expressed by a majority of the spirochetes after the nymphal tick has taken a blood meal, suggesting that P66 expression is associated with the mammalian environment. This alteration of expression is not solely attributable to temperature, pH, or growth phase in in vitro culture. The pattern of P66 expression is also clearly different from the expression of other B. burgdorferi proteins that have been thoroughly analyzed to date (13, 18, 20, 21).

Parameters that do influence p66 expression in culture include temperature, pH, and growth phase, but none of these factors affect P66 protein levels as dramatically as factors seen in the tick vector of B. burgdorferi. In the single genome-scale microarray analysis of B. burgdorferi gene expression that has been published (19), the level of p66 (BB0603) expression was found to be approximately 3.5-fold lower in bacteria grown under culture conditions that mimic those of the fed tick (37°C, pH 6.8) than in bacteria grown in dialysis chambers implanted in rats, a condition that is thought to approximate the conditions encountered by the bacteria during mammalian infection. Levels of p66 expression were not found to be significantly different in bacteria grown under conditions that mimic those of the unfed tick (23°C, pH 7.5) and under the so-called fed-tick conditions. With strain B31 we obtained similar results, with little difference in p66 mRNA levels being shown between bacteria grown at pH 6.5 and 34°C and bacteria grown at pH 7.5 and 24°C. In contrast, with strain N40 the difference was more dramatic between these two conditions, as the bacteria grown at pH 6.5 and 34°C had a >5-fold-higher p66/fla ratio than bacteria grown at pH 7.5 and 24°C. Under no condition, however, did the bacteria show the phenotype of B. burgdorferi in the unfed tick midgut. Our results demonstrate the limitations of in vitro culture conditions, even those that replicate the temperatures and pHs of the tick in different circumstances, which are substitutes for assessing B. burgdorferi gene expression in the arthropod vector at different points in its life cycle.

P66 expression may be inactivated in the unfed tick by a repressor that is expressed in the unique environment of the unfed tick midgut or may be activated in response to mammalian factors that are taken in during the blood meal and are present in the mammalian host (and in laboratory culture). The tick midgut may also present an environment in which P66, but not flagellin or OspA, is either degraded or not translated. Our in vitro culture data, together with the data obtained from vector ticks, suggest that there are interesting facets to the regulation of P66 expression in ticks versus in mammals and in ticks versus in in vitro culture that have never been explored. In addition, while certain in vitro culture conditions do affect the expression of some B. burgdorferi genes, they do not completely reflect the differences between ticks and mammals. A number of as-yet-unidentified factors will likely be found to play important roles in the regulation of expression of B. burgdorferi genes.