Development of Genetic System to Inactivate a Borrelia turicatae Surface Protein Selectively Produced within the Salivary Glands of the Arthropod Vector

Background Borrelia turicatae, an agent of tick-borne relapsing fever, is an example of a pathogen that can adapt to disparate conditions found when colonizing the mammalian host and arthropod vector. However, little is known about the genetic factors necessary during the tick-mammalian infectious cycle, therefore we developed a genetic system to transform this species of spirochete. We also identified a plasmid gene that was up-regulated in vitro when B. turicatae was grown in conditions mimicking the tick environment. This 40 kilodalton protein was predicted to be surface localized and designated the Borrelia repeat protein A (brpA) due to the redundancy of the amino acid motif Gln-Gly-Asn-Val-Glu. Methodology/Principal Findings Quantitative reverse-transcriptase polymerase chain reaction using RNA from B. turicatae infected ticks and mice indicated differential regulation of brpA during the tick-mammalian infectious cycle. The surface localization was determined, and production of the protein within the salivary glands of the tick was demonstrated. We then applied a novel genetic system for B. turicatae to inactivate brpA and examined the role of the gene product for vector colonization and the ability to establish murine infection. Conclusions/Significance These results demonstrate the complexity of protein production in a population of spirochetes within the tick. Additionally, the development of a genetic system is important for future studies to evaluate the requirement of specific B. turicatae genes for vector colonization and transmission.


Introduction
Tick-borne pathogens must efficiently adapt to the arthropod vector and mammalian host to ensure survival of the microorganism. Species of relapsing fever spirochetes are such pathogens that colonize and are transmitted by argasid ticks. Borrelia turicatae, an agent of tick-borne relapsing fever distributed throughout the southern United States and Latin America, is maintained in enzootic cycles by Ornithodoros turicata [1]. After an infectious bloodmeal the spirochetes initially colonize the midgut, and given the short duration of the bloodmeal (5-60 min) [2,3], salivary gland colonization is necessary for subsequent transmission. Furthermore, with efficient transovarial and transstadial transmission, the vector itself is considered the reservoir for the spirochetes [4].
To understand B. turicatae adaptation within the tick, identifying and characterizing the genes that are up-regulated during vector colonization is important. A microarray analysis identified a gene, which we designated the Borrelia repeat protein A (brpA), to be up-regulated when spirochetes were grown at 22uC, a temperature similar to the tick (unpublished results). Originally reported as BTA018 [5], BrpA was predicted to contain a signal peptide, suggesting the protein's surface localization. Given the significance of microbial surface proteins in pathogenesis, the BrpA was further investigated. The surface localization and production of the protein was determined, and a system to inactivate the brpA was developed to evaluate the necessity of the gene during the tick-mammalian infectious cycle. These results demonstrate the up-regulation of a B. turicatae gene within the vector, differential protein production during midgut and salivary gland colonization, and the ability to inactivate B. turicatae genes by allelic exchange.

Ethical statement
All work was performed in adherence to the United States Public Health Service Policy on Humane Care and Use of Laboratory Animals and the Guide for the Care and Use of Laboratory Animals. Murine studies were conducted in accordance with the Mississippi State University Institutional Animal Care and Use Committee, protocol # 11-091. Animal husbandry was provided in adherence to the United States Public Health Service Policy on Humane Care and Use of Laboratory Animals and the Guide for the Care and Use of Laboratory Animals. Rabbit serum was produced by Cocalico Biologicals, Inc and all animal protocols were approved by the Animal Care and Use Committee (animal assurance number A369-01).
The entire brpA locus in seven B. turicatae isolates was amplified using DNA sequences from B. turicatae 91E135 to design the primers brpA F flank and brpA R flank, which flank the gene (Table 1), and the internal sequencing primers brpA seq F1 and brpA seq R1 (Table 1). PCR amplification was performed with a DNA Engine Tetrad (Bio-Rad, Hercules, CA, USA) as previously described [9] using the GoTaq Flexi DNA Polymerase kit (Promega, Madison, WI, USA). Amplicons were visualized in an agarose gel containing GelRed (Phenix Research Products, Candler, NC, USA), and processed with the Qiagen PCR Purification kit (Qiagen Inc, Valencia, CA, USA). Sequencing reactions for brpA, were performed as previously described [6] and deposited to GenBank under sequence accession numbers KC859623-KC859629.
In silico translation of brpA and amino acid alignments were performed using the MacVector 6.0 software package (Oxford Molecular, Oxford, UK). The algorithms used to predict the presence of a signal peptide, surface localization, and conserved motifs were the Basic Local Alignment Search Tool (BLAST) from NCBI, Motif Scan [10] on the MyHits web server, LipoP 1.0 [11], and ScanProsite [12]. The Vector NTI suite was used for amino acid alignments of BrpA [13].

Southern blot analysis
Total genomic DNA was separated by reverse-field agarose gel electrophoresis and transferred to a MagnaGraph Nylon Transfer Membrane (Osmonics Inc., Minnetonka, MN, USA) as previously described [14,15]. brpA seq F2 and brpA seq R2 (Table 1) were used to generate the probe for Southern blotting because the primers are within a conserved region of the gene. Hybridization probes were produced with the PCR DIG probe synthesis kit following the manufacturer's instructions (Roche Applied Science, Indianapolis, IN, USA). Probe amplification, hybridization, and development of the Southern blots were performed as previously described [6,15].

O. turicata colony
The ticks used in these studies originated from Kansas and were previously reared at the Rocky Mountain Laboratories for many years. We have fed adult O. turicata on naïve mice, performed immunofluorescent assays, and quantitative PCR to confirm that the ticks are uninfected. The ticks were housed in 15 or 50 ml ventilated tubes at 27uC and 85% relative humidity (RH) using a saturated solution of KCL [16].
An infected cohort of ticks was obtained by feeding second nymphal stage O. turicata on Swiss Webster mice that were needle inoculated intraperitoneally with 1610 5 wild type or mutant spirochetes. The following day spirochetes were detected in the blood and the ticks were allowed to engorge. Ticks were stored separately at 27uC and 85% RH.
RNA isolation and analysis of B. turicatae in infected mouse blood, ticks, and spirochetes in vitro cultivated at 22uC and 35uC RNA was extracted from three cohorts of three to five infected O. turicata ticks. The ticks were placed into 1.5 ml RNAse free centrifuge tubes (Qiagen), flash frozen using liquid nitrogen, and triturated into a powder using a pestle. Ground ticks were suspended in 100 ml of RNAlater RNA Stabilizing Reagent (Qiagen), centrifuged through the QIAshredder (Qiagen) column, and RNA was isolated using the RNeasy Mini kit (Qiagen) following the manufacturer's instructions including the on-column DNA digestion.
To obtain B. turicatae RNA from spirochetes grown in vivo, cohorts of five infected ticks were fed on Swiss Webster mice. Blood was obtained from the animals by tail nick, and spirochetes were visualized by dark field microscopy within four to five days after tick bite. Mice were sedated using 25 mg/ml Ketamine and 7.6 mg/ml Rompun at a dosage of 0.1 ml per 25 gm of body weight, and exsanguinated by intra-cardiac puncture. Whole blood was centrifuged at 5,0006 g for 10 min at 35uC and the serum containing the spirochetes was removed and placed into a clean RNAse-free 1.5 ml centrifuge tube. B. turicatae was pelleted by centrifugation at 15,0006 g for 20 min, supernatant was removed, and the spirochete-enriched pellet was suspended in 100 ml of RNAlater RNA Stabilizing Reagent (Qiagen), flash frozen using liquid nitrogen, and stored at 280uC until RNA was extracted. Spirochetes were centrifuged over a QIAshredder column (Qiagen) and RNA was isolated as described above.
B. turicatae RNA was isolated from low passage cultures (passage five) grown at 22uC and 35uC in mBSK medium. Spirochetes

Author Summary
Relapsing fever spirochetes are a global yet neglected pathogen causing recurrent febrile episodes, nausea, vomiting, and pregnancy complications including miscarriage. Most species of tick-borne relapsing fever spirochetes are maintained in enzootic cycles, and given an approximately 20 year life span, the arthropod vector for Borrelia turicatae represents a reservoir for the pathogens. While B. turicatae has adapted mechanisms to efficiently colonize and survive within the vector, the genes necessary during the tick-mammalian infectious cycle are unknown. We have identified a gene that was designated the Borrelia repeat protein A (brpA). brpA was up-regulated in a portion of the spirochetes colonizing Ornithodoros turicata, the vector for B. turicatae. Developing a system to delete the gene in B. turicatae enabled the evaluation of the necessity of brpA. With the genetic system established for B. turicatae, a better understanding of the genetic constituents required during the tick-mammalian infectious cycle may be obtained. grown in 50 ml cultures were centrifuged at 8,0006 g for 15 min, the pellets suspended in 100 ml of RNAprotect (Qiagen), flash frozen using liquid nitrogen, and stored at 280uC. Spirochetes were processed and RNA was isolated as described above.
RNA integrity was determined using the Agilent RNA 6000 Pico kit and the Agilent 2100 Bioanalyzer (Agilent Technologies, Waldbroon, Germany) following the manufacturer's instructions. RNA integrity values of 8.0 or higher were used for quantitative reverse transcriptase-PCR analysis (qRT-PCR).

qRT-PCR
Duplex qRT-PCR was performed to determine the upregulation of brpA in spirochetes grown at 22uC and within the tick compared to spirochetes grown at 35uC and in murine blood. Primer and probe sets (Table 1) for brpA (brpA F, brpA R, and brpA probe) and B. turicatae flagellin (flaB F, flaB R, and flaB probe) were used at a final concentration of 5 mM (probe) and 10 mM (primer) (TIB MOLBIL, LLC, Adelphia, NJ, USA). The reporter fluorophores for the flaB and brpA probes were YAK and 6-FAM, respectively, while the quencher was BBQ (Table 1). To determine similar genomic equivalents of flaB and brpA, duplex assays were performed with a DNA template (10 -0.1 ng) isolated from in vitro grown spirochetes using the primer and probe sets stated above.
qRT-PCR assays were performed using the AgPath-ID TM One-Step RT-PCR reagents (Life Technologies) and the StepOne Plus ABI 96-well (Life Technologies) following the manufacturer's instructions. Assays were performed in triplicate using 10, 1, and 0.1 ng of RNA. To determine if DNA contaminated the RNA samples, qRT-PCR assays were also performed without the RT. flaB was used as the normalization gene [17], and differential expression was determined using the 2 2DDCT calculation.
Expressing B. turicatae brpA and flaB as recombinant fusion proteins in Escherichia coli Forward (brpA F exp) and reverse (brpA R exp) primers (Table 1) were used to amplify brpA using genomic DNA from B. turicatae 91E135. PCR amplification, cloning into the pET102/Directional TOPO vector, DNA sequencing, expression, and recombinant protein purification were performed as previously described [9]. brpA was expressed as a thioredoxin, 66-His-labeled fusion protein.
flaB was expressed as a hemagglutinin fusion protein using a modified pEXP1-DEST expression vector (Invitrogen). Briefly, a Cm R -ccdB-attR2 fragment was amplified by PCR from pEXP1-DEST with Accuprime pfx DNA polymerase (Invitrogen) using primers pEXP1-HA-F and pEXP1-HA-R (Integrated DNA Technologies, Coralville, IA, USA), which incorporated a Cterminal sequence encoding the hemagglutinin (HA) epitope. The Cm R -ccbB-attR2-HA PCR product was cloned using the In-Fusion kit (BD Clontech, Mountain View, CA, USA) into EcoRI and HindIII-digested pEXP1-DEST to create pEXP1-HA-DEST. The cloning process resulted in the deletion of the HindIII restriction site in pEXP1-DEST and generation of a new HindIII restriction site 59 of the HA-tag sequence. flaB was amplified by PCR from B. turicatae genomic DNA with Accuprime pfx DNA Bolded nucleotides indicate the restriction enzyme site. doi:10.1371/journal.pntd.0002514.t001 polymerase using primers flaB F exp and flaB R exp ( Table 1). The amplicon was cloned using the In-Fusion kit into BsrGI and HindIII-digested pEXP1-HA-DEST to create pEXP1-HA::flaB.
Producing antisera to recombinant BrpA (rBrpA) and recombinant FlaB (rFlaB) Rabbit anti-rBrpA or chicken anti-rFlaB immune serum was produced by Cocalico Biologicals, INC. Preimmunization serum samples were collected, and two rabbits and two chickens were immunized twice intraperitoneally with 50 mg of rBrpA and rFlaB, respectively, using complete Freund's adjuvant. Three subsequent immunizations were performed at two week intervals using incomplete Freund's adjuvant. Serum samples were collected and tested for specificity to rBrpA and rFlaB by immunoblotting.

SDS-PAGE and immunoblotting
All SDS-PAGE and immunoblot assays were performed using Any kD Mini-PROTEAN TGX precast gels (BioRad, Hercules, CA). Proteins were transferred at 100 volts for 1 hr onto polyvinylidene fluoride (PVDF) membranes with the Mini Trans-Blot Cell (BioRad) and antibody binding was detected with the ECL Western blotting reagent (VWR, Atlanta, GA, USA).
To determine the production of rBrpA, uninduced and induced E. coli BL21 Star (DE3) cells (Life Technologies) were electrophoresed and transferred to PVDF membranes following the manufacturer's instructions. PVDF membranes were probed with anti-poly-histidine peroxidase conjugate (Sigma-Aldrich, St. Louis, MO, USA) at a dilution of 1:4,000. The immunogenicity rBrpA was evaluated using serum samples at a 1:200 dilution from mice infected by tick bite.
One microgram of rBrpA and protein lysates from wild type and DbrpA spirochetes grown at 22uC and 35uC were prepared for immunoblotting as previously described [18], electrophoresed, and transferred to PVDF membranes. Immunoblots were probed with rabbit and chicken serum samples generated against rBrpA and rFlaB, respectively, at a 1:200 dilution. To determine the antigenicity of BrpA, immunoblots were also probed with serum from mice infected by tick bite. HRP-rec-protein G (Life Technologies) was used as a secondary molecule when immunoblots were probed with rabbit serum, while goat anti-chicken IgY-HRP (Life Technologies) was used to detect FlaB binding.

Cryosectioning and immunofluorescent laser scanning confocal microscopy (IF-LSCM)
IF-LSCM was performed on B. turicatae grown at 22uC, in infected murine blood, and ticks. 1610 8 in vitro grown spirochetes were centrifuged at 10,0006 g for 5 min, the pellet washed with 16PBS MgCl 2 , and suspended in 500 ml of 16PBS MgCl 2 . Thick smears were produced on microscope slides using 10-50 ml suspension of spirochetes, air dried, and fixed in methanol for  Thin smears were produced from infected blood obtained from mice five days after cohorts of B. turicatae-infected ticks had fed on the animals and spirochetes were visible. Infected O. turicata were cryopreserved at 260uC and 20 mm longitudinal sections were cut using the Lecia CM1850 cryostat (Lecia Microsystems Inc, Buffalo Grove, IL, USA). Thin smears and tick sections were fixed in Figure 2. Temperature mediated production and surface localization of BrpA. A rabbit pre-immunization serum sample was used as a negative control (A), while hyperimmune serum generated against rBrpA was used to determine differential protein production at 22uC and 35uC (B). Immunoblots were also probed with chicken serum generated against B. turicatae rFlaB (B lower panel). Surface localization assays were performed by immunoblotting using the rabbit serum generated against rBrpA (C upper panel) and chicken serum generated against rFlaB (C lower panel). Molecular mass is depicted in kilodaltons on the left of each blot. The asterisk and arrowhead indicate the expected size of BrpA and FlaB, respectively. doi:10.1371/journal.pntd.0002514.g002 Proteinase K treatment of B. turicatae B. turicatae was grown at 22uC to 1610 8 spirochetes per ml and treated with proteinase K as previously described [19]. B. turicatae was harvested after 30, 60, 90, and 120 min and a final concentration of 5 mg/ml Phenylmethylsulfonyl fluoride in isopropanol was added to the spirochetes to inhibit proteinase K. Samples were electrophoresed, transferred to PVDF membranes, and membranes were probed with rabbit anti-rBrpA or chicken anti-rFlaB serum as described above. Assays were performed three times.

Developing a construct for gentamycin resistance
The nucleotide sequence for the Borrelia hermsii flagellar basal body rod protein (flgB) promoter [8] was used to identify homologous sequence in B. turicatae. The B. turicatae flgB promoter was amplified with primers flgB promoter F and flgB promoter R, AgeI and NdeI sites at the 59 and 39 end of the amplicon, respectively. The B. turicatae flgB promoter was ligated to the 59 end of the gentamycin acetyl transferase gene (NCBI accession #: U22104) after digesting the pBhSV-2::flgB gent [8] with AgeI and NdeI, replacing the B. hermsii flgB promoter with the B. turicatae promoter, forming pBhSV-2::Bt P flgB -gent.

Constructing the suicide vector
Primers brpA F (2) 1000 bp and brpA R (+) 1000 bp (Table 1) and the Expand high-fidelity polymerase (Roche, Indianapolis, IN) were used to amplify brpA including 1000 base pairs up-and downstream of the gene. The amplicon was cloned into the pCR-XL-TOPO plasmid (Life Technologies) and transformed into Top10 E. coli (Life Technologies). NheI and AvrII restriction sites were added onto the 59 end of primers brpA F del and brpA R del (Table 1), respectively, and brpA was removed from the pCR-XL-TOPO by PCR amplification. P flgB-gent was amplified from pBhSV-2::Bt P flgB -gent using flgB promoter F and gent R SpeI primers (Table 1), adding AvrII and SpeI restriction sites, and cloned into the pCR-XL-TOPO containing brpA flanking DNA, forming the deletion construct.

Transformation of B. turicatae
B. turicatae 91E135, passed seven times in mBSK medium at 35uC, was transferred into 500 ml of fresh mBSK and competent cells were made as previously described for B. hermsii [8]. Spirochetes were electroporated in 0.2 cm cuvettes (BioRad) and transferred into 5 ml of mBSK medium for 24 hr at 35uC. The 5 ml cultures were placed into 40 ml of mBSK containing a final concentration of 40 mg/ml of gentamycin. Within 5 days live spirochetes were visualized in the 45 ml cultures and 1 ml was transferred into 4 ml of fresh mBSK medium containing gentamycin. Once the bacteria attained a density of 1610 8 spirochetes/ml, the transformants were cloned in 96-well flat-bottom plates by limiting dilution from 1610 4 -1610 23 spirochetes/ml, and clonality was determined by Poisson statistics. PCR was performed to confirm the deletion of brpA, while immunoblotting was performed on protein lysates from DbrpA mutants grown at 22uC to confirm the protein was no longer produced. For PCR analysis, genomic DNA was extracted with the Wizard Genomic DNA Purification kit (Promega, Madison, WI, USA) and PCR was performed as described above. For SDS-PAGE and immunoblotting, protein lysates were prepared as previously described [18], and assays were performed as described above.

Tick transmission studies and quantitative PCR (qPCR)
Cohorts of second stage O. turicata nymphs were infected with wild type or mutant B. turicatae by feeding the ticks on spirochetemic mice that were previously inoculated with 1610 5 bacteria. For subsequent transmission studies, cohorts of five ticks infected with wild type spirochetes or DbrpA mutants were allowed to feed on groups of five mice sedated with 25 mg/ml Ketamine and 7.6 gm/ml Rompun at a dosage of 0.1 ml per 25 gm of body weight. To evaluate transstadial transmission, the studies were repeated with the same cohort of ticks after they molted. In both studies, 2.5 ml of blood was collected daily by tail nick for 12 consecutive days, placed into 47.5 ml of Sidestep Lysis and Stabilization Buffer (Agilent Technologies), and stored at 280uC. qPCR on infected blood was performed as previously described [20]. Primer and probe sets (TIB MOLBIOL, LLC) were designed for B. turicatae flaB ( Table 1). All samples were assayed in triplicate, and a standard curve was developed using in vitro grown spirochetes at concentrations of 1610 2 to 1610 8 bacteria/ml. qPCR assays were performed with the 96-well ABI Step-One Plus Instrument (Applied Biosystems) in 20 ml reactions containing 26 Brilliant qPCR Master Mix (Agilent Technologies). Standard curves were constructed by plotting CT-values and spirochete concentrations, and the equation for the best-fit line was calculated using Microsoft Excel. R 2 values for the best-fit line were $0.98.

Results
Plasmid mapping and amino acid sequence analysis of brpA from B. turicatae isolates brpA was first identified by a microarray analysis to be upregulated in spirochetes grown at 22uC, a growth temperature similar to the tick environment (data not shown). Searching the B. turicatae chromosome (GenBank accession number CP000048) and partially assembled plasmid sequences indicated that brpA was plasmid-encoded [5], and Southern blotting further mapped brpA to the large linear plasmid (data not shown).
Amplifying the brpA locus from the seven isolates of B. turicatae produced a product of the expected size, and the sequenced amplicons confirmed the presence of the gene in all isolates. The amino acid sequence alignment indicated high sequence identity at the amino and carboxy terminus of the protein (Figure 1). Interestingly, the alignments also revealed a redundancy in the amino acid sequence Gln-Gly-Asn-Val-Glu, with TCB-1 displaying the greatest number of repeats (Figure 1).

In silico analysis of BrpA
Sequence analysis of the brpA locus indicated the gene was not associated with an operon and was predicted to be 40 kilodaltons (data not shown). BLAST analysis of BrpA identified that the NxE motif within Gln-Gly-Asn-Val-Glu was similar to putative calcium binding proteins and the Plasmodium myosin-like proteins. The motif also shared similarity to the Anaplasma marginale Major surface protein 1 (Msp 1) [21]. BrpA was predicted to contain a signal peptide and lipoprotein motif, suggesting the protein was surface localized. brpA is up-regulated at 22uC and within infected ticks Prior to determining differential brpA expression in vitro and in vivo, a series of assays evaluated the efficacy of flaB as the normalizing gene. flaB has been used for expression normalization in B. hermsii [17], and we confirmed similar transcript levels of B. turicatae flaB using RNA from spirochetes grown at 22uC and 35uC ( Figure S1A). Performing the assays without the RT enzyme validated that the RNA samples were not contaminated with DNA (data not shown). Consequently, we assumed that flaB was equally expressed by B. turicatae during tick and mammalian infection. Moreover, using B. turicatae genomic DNA in the assays confirmed comparable C T values for flaB and brpA, indicating similar genomic equivalents for these genes ( Figure S1B).
qRT-PCR using RNA isolated from B. turicatae grown in vitro at 22uC and 35uC, in infected murine blood and infected ticks indicated that brpA was up-regulated at the lower temperature and in ticks ( Table 2). Performing the assays without the RT enzyme demonstrated that the RNA preparations were free of DNA contamination (data not shown).
Immunogenicity, production, and surface localization of BrpA during B. turicatae cultivation at 22uC Immunoblot analysis using serum samples from mice infected by tick bite indicated that rBrpA was not immunogenic (data not shown). Moreover, immunoblotting using serum samples from rabbits immunized with rBrpA demonstrated up-regulation of the native protein when spirochetes were grown at 22uC when compared to 35uC (Figure 2 A and B). Reactivity against FlaB with the samples indicated that similar amounts of spirochete lysates were electrophoresed (Figure 2B insert). The susceptibility of BrpA to digestion with proteinase K suggested that the protein was localized on the outer surface of the spirochetes (Figure 2 C). Detecting the periplasmic protein, FlaB, verified that the outer membrane of B. turicatae remained intact during the proteinase K treatments (Figure 2 C lower insert).
Performing IF-LSCM using rabbit and chicken serum generated against rBrpA and rFlaB, respectively, demonstrated that BrpA was present in a portion (24.6%) of the spirochete population during cultivation at 22uC (Figure 3 A-C). Rabbit and chicken pre-immunization serum samples were not reactive to BrpA and FlaB, respectively ( Figure 3 D and E). Collectively, these results further suggested that BrpA is a tick-associated surface protein.

Production of BrpA by spirochetes in infected blood and ticks
Cryopreserving, sectioning (Figure 4 A), and performing IF-LSCM on infected O. turicata indicated increased BrpA production by B. turicatae in the tick salivary glands compared to the midgut. Probing tick sections with anti-rFlaB serum followed by the  Counting spirochetes in three thin smears of infected murine blood, the midgut, and salivary glands from three ticks indicated that 29.3% of the spirochetes within the salivary glands produced BrpA compared to only 0.67% of the spirochetes in the midgut, and 10.0% in infected blood.

Inactivation of B. turicatae brpA
Given the differential expression of B. turicatae brpA during the tick-mammalian infectious cycle, a genetic system was developed to assess the necessity of this gene for tick colonization and establishing murine infection. brpA deletion constructs were developed using the Topo-XL vector ( Figure 5 A-E). Sequencing the deletion vector confirmed that the 1,000 base pairs up-and down-stream of P flg -gent were in frame and identical to the genomic DNA sequence (data not shown), and brpA was inactivated by allelic exchange (Figure 6 A-D). Transforming B. turicatae demonstrated the ability to genetically manipulate low passage spirochetes, and two clones were produced by limiting dilution, DbrpA #1 and DbrpA #2. PCR analysis using primers localized within the gentamicin acetyl transferase gene and the 1,000 base pair region downstream of the gene produced a 1,300 base pair amplicon, which indicated the insertion of P flgB -gent ( Figure 6 A and B). Moreover, a primer set within brpA demonstrated the displacement of the gene in the mutants (Figure 6 C), while amplification with the flaB primers indicated similar amounts of DNA were used for PCR (Figure 6 D). Immunoblot analysis of mutant spirochetes confirmed that the protein was absent from the bacteria grown at 22uC compared to wild type B. turicatae grown at that temperature ( Figure 7A and B).

DbrpA mutants are transmissible by O. turicata ticks
Cohorts of uninfected O. turicata were fed to repletion on mice infected by needle inoculated with wild type or mutant spirochetes, and salivary gland colonization was confirmed on a portion of ticks by IF-LSCM (data not shown). Ticks infected with wild type or mutant spirochetes were fed on naïve mice and transmission was detected by microscopy and qPCR in two animals within three days ( Figure 8). All animals were infected by the fourth day after tick bite, followed by a day of quiescence, and spirochetes repopulated the blood on the eight day. Both wild type and DbrpA mutant spirochetes attained similar densities within the blood. One animal infected with wild type spirochetes was euthanized before the completion of the experiment, for causes unrelated to this study. Similar results were obtained when transmission experiments were repeated with cohorts of ticks that had molted, which indicated that the mutant spirochetes were maintained transstadially (data not shown). Overall, these results demonstrated that brpA was dispensable for tick colonization and establishing infection in the murine model we used. Given the absence of a phenotype in the mutant, complementation was not performed. Figure 8. Spirochete densities within murine blood after feeding ticks infected with DbrpA mutants or wild type spirochetes. The white (mutant infected mice) and grey (wild type infected mice) shaded box and whisker plots represent the lower and upper quartiles, median, and lower and upper extremes of spirochete densities on a given day. The number below each plot and the zero values above the x-axis represents the number of animals infected with mutant or wild type spirochetes on the given day. doi:10.1371/journal.pntd.0002514.g008

Discussion
The genomes of Borrelia spp. display differential gene expression throughout the tick-mammalian infectious cycle [22][23][24][25][26][27][28]. However, the genes up-regulated by B. turicatae during vector colonization are largely unknown. We report the identification of a plasmid localized gene, brpA, which was up-regulated during spirochete propagation at 22uC and within O. turicata salivary glands. In the B. hermsii-Ornithodoros hermsi model of relapsing fever borreliosis, differential production of the variable tick protein (Vtp) and the arthropod-associated lipoprotein (Alp) was demonstrated during vector colonization [22,28]. While nearly all B. hermsii produced Vtp and Alp, BrpA was produced by a subset of spirochetes within the salivary glands of O. turicata, further demonstrating the complexity of spirochete gene expression within the vector.
With the surface proteome of spirochetes implicated in host adaptation [29][30][31][32][33], determining the requirement of identified surface proteins during the tick-mammalian infectious cycle is important toward deciphering pathogenic mechanisms. Expanding on work accomplished in Borrelia burgdorferi, Battisti and colleagues first demonstrated the feasibility of site-specific mutagenesis by electro-transformation in B. hermsii [8,34]. Interestingly, in contrast to B. burgdorferi, B. hermsii and B. turicatae retain their plasmids and infectivity after continuous passage in vitro [15,35,36]. The genetic stability during cultivation aided us to develop a genetic system to inactivate and evaluate the necessity of the gene product of brpA within the tick and mammal.
Deleting brpA demonstrated that the mutant spirochetes infected mice by needle inoculation, colonized O. turicata, and reached sufficient densities within the salivary glands to ensure subsequent transmission. Our results also indicated that brpA was not required for transstadial transmission. With a 30-100% rate of vertical transmission of B. turicatae [37], future studies will assess BrpA production within infected eggs and transovarial transmission of mutant spirochetes.
Our findings question the necessity of B. turicatae to tightly regulate and retain brpA. The gene was dispensable for vector colonization and transmission to mice, and the protein was only detected within a subpopulation of spirochetes. These findings suggest that B. turicatae may possess redundant mechanisms, and BrpA may have been replaced with a functionally similar protein.
Alternatively, the ecology and maintenance of B. turicatae in nature is complex, and a given population of spirochetes within the vector may be preadapted to infect a specific vertebrate host. For example, B. turicatae isolates have been acquired from infected domestic dogs in Texas and Florida [6,38], implicating wild canids to the susceptibility and possible role in the maintenance of the spirochetes. Also, O. turicata have been obtained from caves in Texas and gopher tortoise burrows in Florida [39][40][41], suggesting the role of bats and reptiles in the enzootic cycle, respectively. With the immunological and physiological differences between vertebrates, specific phenotypes within a population of spirochetes may have a selective advantage in a given host. The animal model we used to study the role of BrpA in B. turicatae was the Swiss Webster mouse, yet brpA may be necessary during the bacteria's enzootic cycle, and the experimental murine model may be limiting. As additional animal models are established and with the developed genetic system for B. turicatae, a better understanding of the genes required for spirochete maintenance will hopefully be achieved.