INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Lyme disease is a chronic infection caused by certain species of the Borrelia burgdorferi sensu lato complex. In North America, B. burgdorferi is the primary species associated with disease in humans. The ability of the Lyme disease spirochetes to maintain chronic infection indicates that they are capable of immune evasion. To date, two different genes or gene families have been implicated in immune evasion, vls and the ospE gene family (designated family 162 by the Institute for Genomic Research [TIGR]) (20, 23). Recent studies have demonstrated that the ospE gene family undergoes mutation during infection, leading to the generation of OspE variants that are antigenically distinct from the proteins expressed by the preinfection spirochete population (20). The mutations that develop in the ospE genes are of two types, point mutations that alter the amino acid sequence, and recombination events between ospE alleles that generate polymorphic OspE-related proteins. The vlsE gene also undergoes mutation during infection (23). It is thought that vlsE is involved in unidirectional recombination with a series of vls pseudogenes, leading to the modification of the vlsE sequence that is expressed. The resulting variants are thought to encode antigenically distinct proteins.

The process of immune evasion in the Lyme disease spirochetes, as mediated by antigenic variation, differs from the well-described system of the relapsing fever spirochetes (3). During relapsing fever, a single Vmp is produced at high levels and becomes a dominant antigen of the outer membrane. In contrast, it is not yet clear if OspE and VlsE are dominant proteins of the spirochetal cell surface during infection in mammals. Hence, it is premature to conclude that they play a similar dominant role in immune evasion as has been demonstrated for the Vmps. The process of immune evasion during infection with the Lyme disease spirochetes is likely to be multifactorial and may be mediated by several different genes or gene families.

The ospE gene family is one of three gene families whose members are flanked at their 5' end by a highly conserved, promoter-carrying sequence element called the upstream homology box (UHB) element (1, 2, 5, 11, 21). The focus of this study is the ospF gene family (designated family 164 by TIGR), which in B. burgdorferi B31MI contains three members, BBO39, BBR42, and BBM38 (TIGR designations). The members of this family, their general properties, and alternative nomeclatures that have been assigned are listed in Table 1. It should be noted that TIGR has placed a fourth gene in this family, BBS41. However, evolutionary analyses have suggested that this gene is a peripheral member of the ospF family (11), and as a result this gene was not analyzed as part of this report. All of the ospF gene family members are carried by plasmids belonging to the cp32 plasmid family (4, 19).

From the variable sequence and molecular properties of the UHB-flanked genes, we hypothesized that mutational and recombination events occur frequently in these genes (11, 21), resulting in the generation of new UHB-flanked gene variants that encode proteins with altered antigenic characteristics. In an analysis of the ospE gene family, we demonstrated this hypothesis to be correct (20). Since all of the UHB-flanked genes encode potentially surface-exposed lipoproteins, it follows that mutational events in members of the UHB-flanked ospF and 163 gene families could also lead to the development of new antigenic variants that could contribute to immune evasion. In addition, differential expression of UHB-flanked genes could also alter the antigenic characteristics of the Lyme disease spirochetes. While discrepancies exist in the literature regarding the expression patterns of the UHB-flanked genes, it has been clearly demonstrated that the UHB-flanked genes encode immunogenic proteins that are expressed at some point during infection (1, 6, 17, 22).

In this report we focus our analyses on the ospF gene family and have sought to assess the genetic stability of these genes and their expression patterns during infection. The analyses presented here demonstrate that the frequency of mutation in these genes is low during infection in mice. Analyses of the humoral immune response to these proteins revealed that there is a temporal component to this response, suggesting that different members of the family are expressed at different times during infection. These studies shed further light on the hypothesized role of the OspF protein family during infection and specifically in immune evasion.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial isolates, cultivation, and experimental infection of mice. B. burgdorferi B31MI, kindly provided to us by MedImmune Inc. (Gaithersburg, Md.), was used for these analyses because the entire genome sequence has been determined for this isolate (7). This isolate was cultivated in BSK-H complete medium (Sigma) at 33°C. We previously confirmed the infectivity of this clone in C3H-HeJ mice (20). All clonal populations derived from B31MI used in this study were generated as part of an earlier analysis of the genetic stability of the ospE genes (20). To obtain postinfection clonal populations, subsurface plating of cultures obtained from ear punch biopsies was performed as previously described (20). The plates were incubated at 33°C under 3.2% CO₂ in a humidified CO₂ incubator. Approximately 2 to 3 weeks later, individual colonies were cored from the plates using a sterile Pasteur pipette. The colony-containing plugs of agarose were inoculated into 2 ml of complete BSK-H medium and cultivated at 33°C.

PCR analysis of ospF gene family members in clonal populations of B. burgdorferi B31MI. To PCR amplify the ospF gene family members of B. burgdorferi B31MI and its clonal derivative, isolated genomic DNA (~50 ng) was used as the template with the primers listed in Table 2. DNA was isolated as previously described (10). In some cases template DNA was obtained by collecting well-isolated B. burgdorferi colonies (derived from spirochete cultures recovered from ear punch biopsies), transferring them into 2 ml of complete BSK-H medium, and cultivating them to mid-log phase. Then 100-µl aliquots were removed, and cells were pelleted, washed with phosphate-buffered saline (PBS), and resuspended in 100 µl of H₂O. The cell suspension was boiled for 10 min and centrifuged to pellet debris, and 1 µl of the supernatant was used as the template in PCR. PCR was performed with Taq polymerase (Promega) for 30 cycles in an MJ Research PTC100 thermal cycler with a hot bonnet. Reaction volumes were 30 µl, and final primer set concentrations were 1 pmol of primer pair per µl. Cycling conditions were as follows: 1 cycle of 5 min at 94°C, followed by 30 cycles of 1 min at 94°C, 1 min at 50°C and 1.5 min at 72°C. The resulting amplicons were analyzed by agarose gel electrophoresis.

Rapid screening for ospF mutations. ospF-related genes were amplified from B. burgdorferi B31MI and postinfection clonal populations using a variety of primer sets as described above. As a control for these analyses, we also amplified and performed single-nucleotide polymorphism (SNP) analyses on the vlsE gene, which is known to undergo mutation during infection in mice. The purified ospF and vlsE amplicons obtained from these clonal populations then served as the template for SNP analyses as previously described (20). It is important to note that the vlsE and ospF SNP analyses were performed using the same set of clones. The SNP approach is essentially a limited sequencing approach in that only one of the four dideoxynucleotide incorporation reactions is performed. Comparison of the resulting ladders on a 6% polyacrylamide-8 M urea sequencing gel provides a rapid means for screening for mutations. To perform the SNP analyses, the Excel-Sequencing kit (Epicentre Technologies) and 5'-end, ³²P-labeled primers were used.

Cloning and sequence analysis of PCR amplicons. To determine the complete sequence of representative ospF gene family members from different postinfection clonal populations, the amplicons were TA cloned into the pGEM-T-Easy vector as described by the manufacturer (Promega). To identify Escherichia coli clones harboring ospF-carrying recombinant (r-) plasmids, the cells were plated onto Luria-Bertani (LB) plates (amplicillin, 50 µg ml¹), and individual colonies were picked with sterile toothpicks and resuspended in 100 µl of distilled H₂O. The resuspended cells were boiled for 10 min, and 1 µl of the cell lysate was used as the template in PCR with gene-specific PCR primer sets. R-plasmids carrying the appropriate inserts were used as the template for sequence analysis. Sequencing was accomplished using end-labeled primers and the Excel-Sequencing kit as described by the manufacturer (Epicentre Technologies). Sequencing reactions were run on 6% polyacrylamide-8 M urea gels, and autoradiography was performed. The determined sequences were translated using the Translate program, and both the nucleotide and amino acid sequences were aligned using the Pileup program and manually adjusted. These programs are contained within the Wisconsin-GCG sequence analysis package.

Immunoblot procedures and LIC and expression of OspF paralogs: analysis of humoral immune response to OspF in experimentally infected mice. The humoral immune response to OspF variants was assessed by immunoblotting. The test antigens for these analyses were generated by PCR amplifying ospF gene family members from B. burgdorferi B31MI-derived clones with primers possessing tail sequences that complement the single-stranded overhangs of the ligase-independent cloning (LIC) pT7Blue-2 LIC vector (Novagen). After PCR was performed, the single-stranded overhangs on the amplicon were generated by treatment with T4 DNA polymerase in the presence of dATP (other deoxynucleoside triphosphates are omitted) as described by the manufacturer (Novagen). To summarize, the 3'-to-5' exonuclease activity of the T4 DNA polymerase will digest the amplicon until it reaches the first A residue under the reaction conditions described above. The first A residue in the primers above corresponds to the first base of the start codon and the last base of the stop codon. When these bases are encountered by the DNA polymerase, the 5'-3' polymerase activity counteracts its exonuclease activity, resulting in an amplicon with single-stranded overhangs that complement the vector. The treated amplicon and pT7Blue2-LIC vector were then annealed and transformed into E. coli NovaBlue Singles competent cells (Novagen), using a standard transformation protocol, where covalent bond formation occurs and the plasmid is propagated. To identify colonies harboring the correct r-plasmid, colonies were selected, placed in 100 µl of H₂O (after generating a master plate with all analyzed colonies), and boiled for 10 min, and 1 µl was used as the template in PCR with the appropriate primer set. Selected recombinants carrying the appropriate plasmid were inoculated into 3 ml of LB (ampicillin, 100 µg ml¹) and grown to an optical density at 600 nm of 0.6, and then protein expression was induced with 0.4 mm IPTG (isopropylthiogalactoside) for 3 h. The expressed protein represents a fusion protein with a 62-amino-acid S-Tag fused to its N terminus. The induced cells were pelleted and cell lysates were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) on a 15% gel.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

PCR analyses of ospF gene family members in postinfection clonal populations. Several studies have demonstrated that borrelial plasmids can be lost during cultivation (14, 16) or during infection in mice (12). Knowledge of the cp32 plasmid composition is essential in order to accurately interpret the data obtained in the course of this study. To determine if the plasmids that carry the ospF gene family members were maintained by the clonal populations analyzed here after 12 weeks of infection in mice, PCR analyses were performed using plasmid- or gene-specific primers (a comprehensive experimental flow chart is shown in Fig. 1). Of relevance to this report are plasmids cp32-4, cp32-6, and cp32-7, which carry the ospF gene family members BBR42, BBM38, and BBO39, respectively (7). All 30 postinfection clones yielded BBO39 and BBR42 amplicons, indicating that they carry cp32-4 and cp32-7 (Fig. 2). All but one of the clones was found to carry cp32-6. Clone 9 was found to have lost cp32-6, as evidenced by the absence of amplification of BBM38. The absence of cp32-6 from this clone was confirmed through hybridization analysis in a separate study (12). Analysis of the size of the amplicons obtained for each gene from each clonal population revealed all to be of the predicted size. This observation suggests that large-scale rearrangements or mutational events did not occur within these genes during cultivation. However, since small-scale rearrangements, insertions, and/or deletions would not be detected by the PCR approach applied above, SNP analyses were performed as described below.

View larger version (28K): [in this window] [in a new window]   FIG. 1.   Schematic of the experimental approach used to assess the genetic stability of the ospF gene family and humoral immune response to OspF-related proteins during infection.

View larger version (52K): [in this window] [in a new window]   FIG. 2.   Analysis of the genetic stability of ospF gene family members in representative postinfection clonal populations of B. burgdorferi B31MI. PCR and SNP analyses were performed. (A) PCR analyses of BBR42, BBM38, and BBO39. Cell lysates of each B. burgdorferi B31MI postinfection clone and the preinfection parental clone (pc) were used as the template in PCR with gene-specific primer sets as described in the text. The amplicons were fractionated in 1% GTG-agarose gels and visualized by staining with ethidium bromide. Size standards are indicated on the left, and the gene targeted for amplification is shown on the right. For the SNP analyses, PCR amplicons for each gene were purified and served as the SNP template as described in the text. A representative segment of the SNP analyses of BBO39, conducted to determine if point mutations developed during infection, is presented in panel B. In addition, a segment of an SNP analysis of the vlsE gene, which is known to undergo mutation during infection, is shown. In both panels, the number designation assigned to each clone analyzed is indicated across the top of the figure.

SNP pattern analysis of ospF gene family members in clonal populations after infection in mice. SNP analyses offer a rapid means for assessing whether mutational events have occurred in a gene of interest. By this approach, one simply compares sequencing ladders generated using one of the four nucleotides for a series of templates. In this case, the BB039, BBM38, and BBR42 ddATP ladders, obtained from the analysis of amplicons derived from a series of 30 postinfection clonal populations, were compared. This approach allows one to determine if changes have developed among clonal populations in ospF gene family members during infection in mice. The clones selected for analysis represent a subset of those that we characterized previously with regard to the genetic stability of the ospE genes during infection (20). These clones were found to either undergo recombination events or develop mutations in ospE during a 3-month infection time frame in C3H-HeJ mice. As an additional positive control for our ability to detect mutations by this approach, we also conducted SNP analyses of the vlsE gene, which is known to experience mutation during infection in C3H-HeJ mice (23). In contrast to the vlsE and ospE genes, mutations were not detected in the BBR42, BBM38, and BBO39 genes in these 30 different clonal populations (Fig. 2). From these analyses, it can be concluded that in B. burgdorferi B31MI, members of the ospF gene family are genetically stable during infection.

Analysis of humoral immune response to OspF protein family members. To allow an assessment of the specificity and development of the humoral immune response to OspF protein family members, the BBO39, BBM38, and BBR42 genes were PCR amplified using primers generated for LIC cloning and expressed in E. coli. To verify that the cloned genes were expressed in E. coli, immunoblot analyses were performed using an S-protein directed against the S-Tag segment of the r-fusion proteins (Fig. 3). R-proteins of the appropriate size were detected for each of the constructs, confirming expression. The relatively equivalent amounts of protein detected in the immunoblot analyses indicate that expression levels were comparable for each. This observation is of importance for evaluating the data presented below.

View larger version (55K): [in this window] [in a new window]   FIG. 3.   Immunoblot analysis of r-OspF protein family members expressed in E. coli. Each member of the ospF gene family was cloned and expressed as an S-tag fusion protein using ligase-independent cloning methods as described in the text. Proteins from E. coli cultures that were induced to express the r-proteins with IPTG were fractionated by SDS-PAGE and transferred to a PVDF membrane by electroblotting. The membrane on the left was screened with anti-S-Tag protein HRP conjugate, while the membrane on the right was screened with a polyclonal anti-OspF antiserum (generated with a gene of B. burgdorferi N40 origin). Immunoblot methods are described in the text. Molecular size standards are indicated on the left.

Analysis of temporal pattern of humoral immune response to OspF protein family members during infection of C3H-HeJ mice with B. burgdorferi B31MI. To determine if there is a temporal aspect to the humoral immune response to OspF-related proteins, serum samples were collected at 2-week intervals up to 12 weeks from mice that had been infected by needle inoculation with B. burgdorferi B31MI pc. Lysates of E. coli cells that had been induced with IPTG to express each r-protein were used as the test antigen in immunoblot analyses. The serum samples were first preabsorbed with E. coli DH5 cells. Immunoblot analyses using sera from two mice revealed a weak IgG response to BBO39 by week 4 of infection (data not shown) and a strong response by week 6 (Fig. 4). BBR42 was found to be only weakly immunoreactive, while BBM38 was not reactive with these infection-derived sera even after 12 weeks of infection. The absence of an early response to BBR42 and BBM38 in both mice suggests that, in contrast to BBO39, these proteins are not expressed during early infection. An alternative interpretation is that BBR42 and BBM38 are expressed but that these proteins are only weakly immunogenic. However, earlier studies have demonstrated that the OspE-, and OspF-related proteins are immunogenic in mice (1, 6, 9, 13, 17, 22).

View larger version (42K): [in this window] [in a new window]   FIG. 4.   Analysis of temporal pattern and specificity of the humoral response to OspF protein family members during murine infection. Cell lysates of E. coli that were induced to express the r-proteins (as labeled in the figure) with IPTG were fractionated by SDS-PAGE and immunoblotted. Identical membranes were screened with infection-derived sera from mice infected with either B. burgdorferi B31MI pc (A) or B. burgdorferi B31MI c53 (B). In addition, as shown in panel B, the infection sera were also preabsorbed with a lysate of E. coli that had been induced to express r-BBO39. m1 and m2 indicate the specific mouse from which the serum was collected. The time after initial infection at which the sera were collected is indicated. The asterisk indicates that infection serum pc-m2 was actually collected at week 10 (not week 12).

Analysis of specificity of antibody response to BBM38 and BBR42. The experiments above demonstrated that in some mice, an antibody response to BBM38 and BBR42 does not develop until late stages of infection. It is possible that the immunoreactivity of BBM38 and BBR42 with the late stage infection sera could be due to an expansion of the antibody response to BBO39, resulting in cross-immunoreactivity of anti-BBO39 antibodies with these proteins. To test for this possibility, immunoblots of the r-proteins were screened with the infection sera from mice infected with B31MI c53 that had been preabsorbed with lysates of E. coli that had been induced to express BBO39 (Fig. 4). Preabsorption eliminated immunoreactivity with rBBO39 but had no effect on the immunoreactivity of BBR42 and BBM38. These analyses demonstrate that the antibodies that develop late in infection to BBO39 and BBR42 are in fact specific for these proteins and are not cross-reactive antibodies. This observation provides further evidence for the late-stage-specific expression of BBM38 and BBR42 during infection.

Analysis of antibody response to OspF protein family members in mice infected with B. burgdorferi N40 and 297. To determine if other isolates express antigenically related proteins and exhibit a similar temporal pattern of expression, mice were infected with clones of B. burgdorferi N40 and 297 (N40c1 and 297c1, respectively). It is important to note that the genome sequence for these isolates has not been determined, and as a result the composition of the ospF gene family in these isolates has not been fully defined. This is an unavoidable caveat in analyses that deal with isolates other than B31MI. The first step in these analyses was to determine if these isolates carry genes related to BBO39, BBR42, and BBM38. Towards this goal, PCR analyses were performed. PCR amplification was observed with both N40 and 297 using the BBO39-(strong amplification) and BBR42 (moderate amplification)-specific primer sets (data not shown). For BBM38, strong amplification was observed with isolate 297, while no product was obtained from isolate N40. Hence, it appears either that N40 lacks BBM38 or that its sequence is divergent enough that amplification will not occur with the primer set used. In any event, these analyses indicate that the clones of isolates 297 and N40 used here carry sequences related to at least some of these genes.

View larger version (29K): [in this window] [in a new window]   FIG. 5.   Analysis of murine humoral immune response to OspF protein family members in mice infected with heterologous strains. Immunoblots of the r-proteins were generated as described for Fig. 4. The immunoblots were screened with sera collected at either 0, 8, or 12 weeks after inoculation. The mice were infected for 12 weeks with either B. burgdorferi N40 c1 and B. burgdorferi 297 c1.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Evolutionary analyses have demonstrated that B. burgdorferi B31MI carries three well-defined UHB-flanked gene families, ospE, ospF, and family 163 (2, 11). The focus of this report is the ospF gene family, which in B. burgdorferi B31MI comprises three members: BBR42, BBO39, and BBM38 (7). Comparative analyses of ospF-related genes from different isolates have demonstrated that these genes are highly variable, suggesting recent modification by mutational and recombination events (11, 18, 21). Based on this observation and our earlier work that demonstrated that the UHB-flanked ospE gene family undergoes mutation and recombination during infection (20), we hypothesized that similar molecular events occur in the ospF gene family during infection. One possible consequence of these processes could be the generation of antigenic variants that contribute to immune evasion. To test this hypothesis, we analyzed the genetic stability and humoral immune response to individual members of the OspF protein family using the murine model for Lyme disease. Analyses of ospF genetic stability focused on a series of B. burgdorferi clones recovered from C3H-HeJ mice that had been infected for 12 weeks with B. burgdorferi B31MI (20). The clonal populations analyzed in this report were the same populations previously found to undergo mutation in their ospE genes. In this report, we also demonstrate that mutations develop within the vlsE gene of these clones during infection. However, in contrast to ospE and vlsE, SNP analyses of BBO39, BBR42, and BBM38 revealed that the frequency of mutation in these genes during infection is low. As an additional means for testing for the development of mutations, complete sequence analysis of several of the ospF gene family member PCR amplicons was performed. As with the SNP analyses, mutations were not detected. In an earlier analysis, we demonstrated that BBS41 (a peripheral member of the ospF family) is also stable during infection (20). The absence of mutation in the ospF gene family members and the peripherally related BBS41 gene, in a background that is known to undergo mutation during infection, indicates that ospF gene family members are stable during infection in mice.

Prior to this report, little was known about the patterns of expression of the OspF protein family during infection. As a means of assessing the expression patterns and to characterize the specificity of the humoral response, each of the ospF family members from B. burgdorferi B31MI was cloned and expressed in E. coli for use in immunoblot analyses. After confirming that the constructs were expressed at similar levels in E. coli, the r-proteins were screened with a polyclonal anti-OspF antiserum generated using r-protein of B. burgdorferi N40 origin (9). This antiserum recognized r-forms of BBO39 and BBR42 but not BBM38, indicating that BBM38 is antigenically distinct from other OspF family member proteins.

To assess the expression patterns and the specificity of the immune response to the OspF proteins during infection, immunoblots of these proteins were screened with infection sera from several mice that were collected over the course of infection. Sera collected after 6 weeks of infection from mice infected with B. burgdorferi B31MI pc were found to possess antibodies to BBO39 but not to BBR42 or BBM38. This observation demonstrates that anti-BBO39 antibodies are not immunologically cross-reactive with BBM38 or BBR42. A significant response to BBM38 and BBR42 did not develop in any of the B31MI pc-infected mice analyzed, even after 12 weeks of infection. A similar trend was also observed in the mice infected with B. burgdorferi B31MI c53. While a strong IgG response did develop by week 12 in c53-infected mice, it is clear from both experiments that a significant IgG response to BBR42 and BBM38 develops considerably later than to BBO39. As all OspF-related proteins that have been analyzed to date have been demonstrated to be immunogenic in mice (17), it is possible that the delayed response to BBR42 and BBM38 results from the temporal expression of these proteins during infection. One possible explanation for the detection of antibodies to BBR42 and BBM38 in the B31MI c53-infected mice was that there is some degree of cross-reactivity of the anti-BBO39 antibodies with these proteins. However, preabsorption of these infection sera with lysates of E. coli that had been induced to express BBO39 confirmed that the IgG response to these proteins was specific and not due to cross-reactivity.

To determine if infection with other isolates results in the late-stage production of antibodies that recognize BBO39, BBR42, or BBM38, mice were infected with the B. burgdorferi isolates N40 and 297. During the early stages of infection, these mice produced antibodies that recognize r-BBO39 but not BBR42 or BBM38. The absence of antibodies to BBR42 and BBM38 in sera collected during early infection suggests either that these proteins are not expressed or that 297 and N40 encode variants of these proteins that are antigenically distinct from that carried by B31MI. Immunoblot analyses of late-stage infection sera revealed that mice infected with N40 but not 297 develop anti-BBR42 antibodies. As in the mice infected with B31MI clones pc and c53, the antibody response to these proteins suggests temporal expression of BBR42. Antibodies to BBM38 were not detected in any of the serum samples from the N40- or 297-infected mice. As described above, we were able to amplify BBO39 and BBR42 from both 297 and N40 but could only amplify BBM38 from isolate 297. The absence of antibodies to BBM38 and BBR42 in the 297 infection sera may indicate that these proteins are not expressed by this isolate during infection or that they are expressed at a time point later than 12 weeks. Since BBM38 could not be amplified from isolate N40, no conclusion can be reached at this time about the lack of anti-BBM38 antibodies in the N40 infection-derived sera.

The apparent temporal expression of BBR42 and BBM38 during infection is consistent with and may explain some earlier data reported by Nguyen et al (13). It was demonstrated that antibodies to an OspF-related protein of N40 could be detected in only 14% of Lyme disease patients with "early Lyme disease." In contrast, an antibody response to OspF was detected in 58% of the patients with late Lyme disease. It was also demonstrated that antibodies to B. burgdorferi N40 OspF did not develop in mice infected with N40 until 90 days into the infection. The time frame of development of the humoral immune response to OspF-related proteins in this earlier report is consistent with that reported here for BBR42 and BBM38. At the time that Nguyen et al. conducted their analyses, a completed genome sequence for B. burgdorferi was not available, and it had not yet been demonstrated that OspF belongs to a paralogous protein family. As a result, it was not possible for those authors to specifically determine which ospF allele(s) was or was not being expressed. By exploiting the genome sequence, we have been able to demonstrate which specific members of the OspF family are expressed at which relative stages of infection.

In an earlier analysis, Stevenson et al. reported that an IgG response to the OspF family members BBO39 and BBM38 could be detected in a mouse infected with B. burgdorferi for 4 weeks (17). Consistent with this, we also detected an IgG response to BBO39 during early infection. However, in contrast, we did not detect an early IgG response to BBM38. The observed immunoreactivity of r-BBM38 (referred to as ErpK in the aforementioned study) noted by Stevenson et al. was relatively weak, and the size of the immunoreactive protein was not consistent with that predicted for BBM38. A second important point is that upon analysis of serum samples from human Lyme disease patients, Stevenson et al. reported that only 2 of the 10 patient sera analyzed had anti-BBM38 antibodies. This observation may be inconsistent with the authors' conclusion that all OspE- and OspF-related proteins are expressed early during infection. It is possible that the absence of a detectable response to BBM38 in these human serum samples could reflect the time point at which the sera were collected. The data presented here would suggest that if the sera were collected during early infection, an IgG response to BBM38 might not be observed due to the expression of this gene during later stages of infection. Unfortunately, the time point during infection at which the human sera were collected was not known to the authors, and hence it is not possible to further assess these earlier data. The basis for the discrepancies regarding the expression patterns of the OspF proteins reported here and by Nguyen et al. (13) with that reported by Stevenson and colleagues is unclear. Note that BBR42 expression was not specifically analyzed by Stevenson et al., and hence we cannot compare and contrast data regarding the expression of this specific protein.

In an attempt to identify the potential molecular basis for the differential expression of BBO39, BBR42, and BBM38, the UHB elements for these genes, which harbor the promoter elements, were aligned and compared (Fig. 6). A 192-nucleotide stretch located upstream from the translational start codon of BBM38 and BBR42 exhibits 93.7 and 80% identity, respectively, with that of BBO39. In addition, although the precise locations of the 35 and 10 elements have not been experimentally defined, a putative consensus 10 is evident in all. A 30-nucleotide stretch immediately upstream of the 10 exhibits only minor sequence variation. Nonetheless, it is possible that these minor sequence changes could influence the transcription of these genes. It is also possible that all of the OspF genes, including BBM38 and BBR42, are transcribed during early infection but that protein production is posttranscriptionally regulated. Recent findings by Hefty et al. suggest that some UHB-flanked genes of B. burgdorferi 297 are transcribed in the mammal but that protein production is posttranscriptionally repressed (8). The precise mechanism(s) by which temporal transcription or protein production occurs will require further experimentation and analysis. In closing, the data presented here further illustrate the dynamics of protein expression in Lyme disease spirochetes. The study of the temporal component of protein expression will provide significant insight into the nature of the host-pathogen relationship over the course of disease.

View larger version (21K): [in this window] [in a new window]   FIG. 6.   Alignment of UHBs of B. burgdorferi B31MI ospF gene family members. The sequences were obtained from the genome sequence of B. burgdorferi B31MI and aligned, and the alignment was manually refined. The putative translational start codons and 10 and 35 elements are indicated. Note that two translational start codons are indicated for BBM38. The more upstream ATG listed by TIGR as the start codon is likely to be incorrect, as there is no ribosome-binding site (RBS) upstream from this codon. However, a second ATG resides 12 bp downstream, and this start is preceded by a consensus ribosome-binding site. Note that BBR42 is the downstream gene in an operon with BBF40. BBR42 itself is not 5' flanked by a UHB element; hence, the UHB element flanking BBR40 is shown.