Evidence that BosR (BB0647) Is a Positive Autoregulator in Borrelia burgdorferi

Abstract

Borrelia burgdorferi survives in nature through a complex tick-mammalian life cycle. During its transit between ticks and mammalian hosts, B. burgdorferi must dramatically alter its outer surface profile in order to interact with and adapt to these two diverse niches. It has been established that the regulator BosR (BB0647) in B. burgdorferi plays important roles in modulating borrelial host adaptation. However, to date, how bosR expression itself is controlled in B. burgdorferi remains largely unknown. Previously, it has been shown that DNA sequences upstream of BosR harbor multiple sites for the binding of recombinant BosR, suggesting that BosR may influence its own expression in B. burgdorferi. However, direct experimental evidence supporting this putative autoregulation of BosR has been lacking. Here, we investigated the expression of bosR throughout the tick-mammal life cycle of B. burgdorferi via quantitative reverse transcription (RT)-PCR analyses. Our data indicated that bosR is expressed not only during mouse infection, but also during the tick acquisition, intermolt, and transmission phases. Further investigation revealed that bosR expression in B. burgdorferi is influenced by environmental stimuli, such as temperature shift and pH change. By employing luciferase reporter assays, we also identified two promoters potentially driving bosR transcription. Our study offers strong support for the long-postulated function of BosR as an autoregulator in B. burgdorferi.

INTRODUCTION

Lyme disease, caused by the spirochetal bacterium Borrelia burgdorferi, remains the most prevalent vector-borne illness in the United States (1). B. burgdorferi is transmitted to humans through Ixodes ticks (2, 3). In the United States, Ixodes scapularis is the major transmission vector for B. burgdorferi in the northeastern, Mid-Atlantic, and north-central areas, whereas Ixodes pacificus is the carrier along the Pacific Coast. The arthropod tick has three life stages, larva, nymph, and adult, and takes three blood meals during its 2-year life cycle. Usually, newly hatched larvae acquire spirochetes when they first feed on small infected rodents (the acquisition phase). The engorged larvae then molt into nymphs (the intermolt phase). When resting nymphs take a second blood meal, they grow and mature into adults. During the second feeding, B. burgdorferi multiplies within the tick midgut and migrates to the salivary glands, where it is then injected into mammalian dermal tissue to initiate infection in a mammalian host (the transmission phase). Following transmission, B. burgdorferi replicating within the skin can migrate to distant organs, such as the heart, to joints, and to the nervous system. Clinically, bacterial infection and dissemination culminate in a multisystem inflammatory illness with a broad spectrum of clinical manifestations, including carditis, arthritis, and neuroborreliosis (4, 5).

To establish infection and cause disease, B. burgdorferi must maintain its tick-mammalian life cycle of colonization and persistence within ticks and mammalian infection and dissemination. Typically, B. burgdorferi achieves this through altering its outer surface coat in response to specific environmental stimuli (4–8). For example, in order to successfully survive and persist in ticks during the acquisition and/or the larval-nymphal intermolt phase, B. burgdorferi must express on its surface tick-specific determinants, such as outer surface protein A (OspA) (9–11), OspB (12), and BptA (13). In contrast, upon taking a blood meal, B. burgdorferi prepares itself by transitioning to expressing an array of other (host-associated) factors, such as OspC, decorin binding protein B (DbpB), DbpA, BBK32, and BBA64 (14–23); these determinants appear to allow the spirochetes to be tick-mammal transmitted, enter the host dermis, and establish the local infection. Once residing in an animal host, B. burgdorferi expresses another set of new determinants to facilitate tissue colonization and to counter host immunity (4–8).

It has now been established for more than 10 years that the RpoN (σ⁵⁴)-RpoS (σ^S) regulatory pathway in B. burgdorferi plays a central role in modulating key adaptive changes during the transit of the spirochete between its tick vector and its mammalian host(s) (4, 19, 24–28). In general, signals, such as temperature shift (from 23°C to 37°C), pH reduction (from pH 7.6 to pH 6.8), and a marked replicative burst of spirochetes, that B. burgdorferi encounters in nymphs during blood feeding activate the transcription of the σ⁵⁴ (RpoN)-dependent rpoS. Once RpoS is produced, it promotes the expression of mammalian host-associated factors and, in some instances, represses the expression of tick-specific determinants (4, 28). Although the molecular details remain largely unknown, two major factors, including a putative bacterial enhancer binding protein (bEBP), Rrp2 (26, 29–34), and BosR (BB0647), a nonclassical Fur/PerR homologue (35–39), are required for the sensing of environmental signals to trigger the RpoN-RpoS regulatory pathway. The latter is a particularly remarkable feature of the borrelial RpoN-RpoS pathway as, to our knowledge, this is the only example where a secondary (other than the bEBP) accessory molecule (BosR) is required for the activation of σ⁵⁴ in a bacterium.

Currently, very little is known about the precise physiological functions of BosR. BosR was originally annotated as a homologue of the iron-sensing bacterial global regulator Fur or the peroxide stress response regulator PerR (40–42). However, it has been reported that, unlike other bacteria, B. burgdorferi may not import or utilize iron (43). Therefore, BosR seems unlikely to be involved in the regulation of iron transport, storage, or use in B. burgdorferi. Rather, we and others recently reported that BosR functions as a critical virulence factor required by B. burgdorferi to infect mammals (35, 37, 38), which is linked to BosR's essentiality for activation of the central RpoN-RpoS pathway. More specifically, BosR binds to a DNA element called the BosR box in the rpoS promoter region and then activates the transcription of σ⁵⁴-dependent rpoS (37, 44, 45). In addition, our microarray analyses (38) indicated that BosR may also influence the expression of many other genes (other than the RpoS regulon) potentially involved in borrelial physiology and metabolism, thereby substantiating its function as a potential global regulator in B. burgdorferi. In general, the levels of global regulators in a wide variety of microorganisms are strictly maintained, as aberrant expression of these factors typically interrupts bacterial physiological processes and therefore is often toxic for bacterial growth (46–48). For example, it had been reported that abnormal production of BosR or RpoS in B. burgdorferi results in cell death, quick clearance of infection, and a dramatic reduction in borrelial infection potential (49, 50). Thus, to maintain its enzootic life cycle, B. burgdorferi must tightly regulate BosR and the BosR regulon during its tick-mammal transit.

To date, how bosR expression is controlled in B. burgdorferi remains largely unknown. In particular, very little is known regarding the factors involved in the regulation of bosR expression. Previously, gel shift data indicated that recombinant BosR (rBosR) protein bound to the 5′ DNA sequence upstream of bosR (38, 42), implying that BosR, like its Fur/PerR homologues, may function as an autoregulator. However, to date, experimental evidence has been lacking to support the autoregulation of bosR, probably due to the inability to obtain a bosR deletion mutant of B. burgdorferi. To address these important information gaps, we examined bosR expression using a luciferase reporter assay and a set of strategically constructed bosR mutants.

MATERIALS AND METHODS

Ethics statement.

This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All animal procedures were approved by the Institutional Animal Care and Use Committee at the University of Texas (UT) Southwestern Medical Center.

Bacterial strains and culture conditions.

The bacterial strains used in this study are listed in Table 1. B. burgdorferi strain 297 (51) was used as the wild-type (WT) strain. The isogenic bosR mutant OY08 was constructed in our previous study (37). Barbour-Stoenner-Kelly II (BSK-II) medium or BSK-H medium (Sigma Chemical Co., St. Louis, MO) (52) supplemented with 6% rabbit serum (Pel-Freeze Biologicals, Rogers, AR) was used to cultivate B. burgdorferi. When appropriate, kanamycin (Kan) or streptomycin (Strep) was added to the BSK medium at a final concentration of 160 or 150 μg per ml, respectively. Spirochetes were routinely cultured in BSK medium (pH 7.6) at 37°C in a 5% CO₂ incubator. Bacterial growth was monitored by enumerating spirochetes via dark-field microscopy. To determine the effects of environmental factors on gene expression, B. burgdorferi was cultivated, as previously described, under various conditions, such as at 23°C or in BSK medium at pH 6.8 (53–55). To avoid potential effects on gene expression by in vitro serial passages of B. burgdorferi (56), spirochetes were collected from cultures with less than three serial passages throughout the study. Escherichia coli strain TOP10 (Thermo Fisher Scientific, Grand Island, NY) was used as the cloning host for plasmid construction.

TABLE 1.

Strains and plasmids used in this study

Strain or plasmid	Description	Source
B. burgdorferi
297	Infectious, low-passage B. burgdorferi strain	51
OY08	297; ΔbosR::Kan	37
OY20	297 transformed with pOY63	53
OY23	297 transformed with pOY69	53
OY39	297 transformed with pOY90	This study
OY44	297 transformed with pOY106	This study
OY68	OY08 transformed with pOY69	This study
OY69	OY08 transformed with pOY106	This study
OY79	297 transformed with pOY89.2	This study
OY81	OY08 transformed with pOY89.2	This study
OY82	OY08 transformed with pOY90	This study
OY318	297 transformed with pOY463	This study
OY319	OY08 transformed with pOY463	This study
E. coli
TOP10	F− mcrA Δ(mrr-hsdRMS-mcrBC) f80lacZΔM15 ΔlacX74 recA1 araD139 Δ(ara-leu)7697 galU galK rpsL (Strr) endA1 nupG	Thermo Fisher Scientific
Plasmids
pGEM-Teasy	TA cloning vector	Promega
pOY16	B. burgdorferi flaB cloned into pGEM-T Easy	27
pPROEX HTb	Cloning vector	Thermo Fisher Scientific
pOY21	bosR (PCR product from primers 26F and 26R) cloned into pPROEX HTb	This study
pOY63	Promoterless lucBb+ from pJD48 cloned into pJD54 at BglII and HindIII sites	53
pOY69	pOY63::dbpBA minimal promoter	53
pOY89.2	pOY63::UPbb0648 (PCR product from primers 91F and 91.2R)	This study
pOY90	pOY63::UPbosR (PCR product from primers 92F and 92R)	This study
pOY106	pOY63::ospC minimal promoter (PCR product from primers 108F and 108R)	This study
pOY463	pOY63::UPbb0648-bb0648-UPbosR (PCR product from primers 91F and 92R)	This study

Generation of luciferase reporter vectors.

All reporter constructs were created based on the shuttle plasmid pOY63 (53), which harbors a B. burgdorferi codon-optimized, promoterless luciferase gene (luc) (57). DNA fragments containing putative bosR promoter regions were amplified from B. burgdorferi strain 297 genomic DNA via high-fidelity PCR using Pfx50 DNA polymerase (Thermo Fisher Scientific). The primers are listed in Table S1 in the supplemental material. After digestion with appropriate restriction enzymes, DNA was cloned into plasmid pOY63 linearized with the same restriction enzymes. These DNA inserts contain the native ribosome-binding site (RBS) and partial coding sequences of bb0648 or bosR. More specifically, the insert 91F/91.2R in pOY89.2 contains 129 bp of 5′ bb0648, whereas the insert 91F/92R or 92F/92R in pOY463 or pOY90, respectively, contains 66 bp of 5′ bosR. Therefore, the resultant reporter constructs are luciferase transcriptional fusion vectors. These constructs were transformed into WT strain 297 or the bosR mutant OY08 to yield merodiploids, and luciferase activity was monitored to assess gene transcription.

Luciferase assay.

Luciferase activity was determined as previously described by using the luciferase assay system (Promega Corp., Madison, WI) (53, 57). Briefly, after spirochetes were harvested by centrifugation at 10,000 × g for 10 min, the pellets were suspended in cell culture lysis buffer containing 25 mM Tris phosphate (pH 7.8), 2 mM dithiothreitol, 2 mM 1,2-diaminocyclohexane-N,N,N′,N′-tetraacetic acid, 10% glycerol, 1% Triton X-100, 1.25 mg/ml lysozyme, and 2.5 mg/ml bovine serum albumin. The lysate was then incubated with the luciferase assay reagent, and the relative luciferase units (RLU) were measured using a Centro LB 960 luminometer (Berthold Technologies, Oak Ridge, TN). The results are presented as RLU per 1 × 10⁶ spirochetes. At least three independent tests were performed, and the results were analyzed using the Student t test.

SDS-PAGE and immunoblot analysis.

Spirochetes were collected via centrifugation and washed thrice in PBS. The pellets were suspended in an appropriate volume of SDS sample buffer and boiled for 5 min. The samples were then loaded onto a 12.5% acrylamide gel. The resolved proteins were either stained with Coomassie brilliant blue R-250 or electrotransferred onto a nitrocellulose membrane for immunoblot analysis as previously described (37, 58). BosR, RpoS, and OspC were detected using anti-BosR polyclonal antibody (38), anti-RpoS monoclonal antibody 6A7-101 (26), or anti-OspC monoclonal antibody 1B2-105A (59), respectively. The immunoblots were developed colorimetrically using 4-chloro-1-naphthol as the substrate or by chemiluminescence using the ECL Plus Western blotting detection system (Amersham Biosciences). Images were documented using a Fujifilm LAS-3000 Imager (Fujifilm), and semiquantitative analyses were performed using MultiGauge V3.0 software (Fujifilm).

RNA isolation and quantitative reverse transcription (qRT)-PCR analysis.

Total RNA was isolated from spirochetes using TRIzol (Thermo Fisher Scientific) according to the manufacturer's protocol. After purification, the RNA was treated with RNase-free DNase I (GenHunter Corporation, Nashville, TN) to remove genomic DNA, followed by a second round of purification using an RNeasy minikit (Qiagen, Valencia, CA). cDNA was generated using the SuperScript III first-strand synthesis System (Thermo Fisher Scientific) according to the instructions. Quantitative PCR (qPCR) was performed using Platinum SYBR green qPCR SuperMix-UDG (Thermo Fisher Scientific). The relative quantification (ΔΔC_T) method was used to calculate gene expression changes between B. burgdorferi strains (26). All data were normalized against the B. burgdorferi flaB gene or the aadA gene (encoding aminoglycoside-3″-adenylyltransferase) carried in the cloning vectors as the endogenous control.

Analyses of bosR expression in infected ticks and mice via qRT-PCR.

Preparation of B. burgdorferi-infected ticks and mice, sample collection, RNA extraction, cDNA synthesis, and qRT-PCR analyses were performed as previously described (27). Briefly, groups of 3 to 5 C3H/HeN mice were infected with B. burgdorferi via intradermal needle injection at 10⁵ spirochetes per mouse. After 7, 14, 21, 28, or 50 days of infection, mice were sacrificed, and 30- to 50-mg tissue samples were collected. To prepare B. burgdorferi-infected ticks, larvae were fed on infected mice (∼100 larvae per mouse), and ∼50 fed ticks were collected when feeding was completed (i.e., fed larvae). The other ∼50 fed larvae were stored in an incubator for 3 weeks, and ∼25 ticks were collected as fed intermolt larvae. The remaining fed larval ticks were allowed to molt to nymphs. The newly molted nymphs were then allowed to feed on naive mice (∼25 ticks per mouse; tick transmission phase). The nymphs were collected at 24, 48, or 72 h postinfestation and stored in liquid nitrogen until they were processed for RNA extraction. As a control, flat larvae were also collected for RNA extraction and subsequent gene expression analysis.

Primer Express software (Thermo Fisher Scientific) was used to design specific primers (see Table S1 in the supplemental material) for B. burgdorferi genes. Primers were then validated via an absolute quantification test using B. burgdorferi genomic DNA. Standard curves created for all the primers had a slope of −3.3 ± 0.3 (data not shown). For quantification of amplicons, bosR was first amplified by PCR using primers 26F and 26R. Following purification, bosR was cloned into pPROEX HTb (Thermo Fisher Scientific) at the BamHI and EcoRI sites, resulting in pOY21. In addition, the B. burgdorferi flaB gene was amplified by PCR using primers 14F and 14R and cloned into the pGEM-T Easy vector (Promega, Madison, WI), resulting in pOY16 (27). Recombinant plasmid DNA was then purified and serially 10-fold diluted to generate a standard curve. Copies of bosR transcripts were calculated using the Absolute Quantification Analysis program (Applied Biosystems) and normalized against copies of flaB (27). Note that for the data shown in Fig. 1A, it was not possible to perform multiple trials for statistical analyses due to limited mRNA starting material.

FIG 1.

Expression of bosR in infected ticks (A) and mice (B). Ticks, i.e., flat (uninfected) larvae, fed larvae, intermolt larvae, and fed nymphs during the transmission phase at 24, 48, and 72 h postfeeding, were collected. Tissue samples were collected from infected mice on different days (day 7, 14, 21, 28, or 50) postinfection. RNA was isolated from pooled ticks and animal tissues, and qRT-PCR was employed to analyze gene expression. The values represent the average copy number normalized per 100 copies of B. burgdorferi flaB transcripts. B, bladder; H, heart; J, joint; S, skin.

Statistical analysis.

All data from qRT-PCR and the luciferase assays were expressed as means and standard deviations (SD) and analyzed using an unpaired Student t test, in which statistical significance was determined to be at a P value of <0.05.

RESULTS AND DISCUSSION

Expression of bosR throughout the tick-mammal life cycle of B. burgdorferi.

Despite the critical roles of BosR in B. burgdorferi virulence regulation, how bosR is expressed in infected animals and ticks has remained unexplored. To address this question, we examined the expression of bosR throughout the tick-mammal infectious cycle of B. burgdorferi. To this end, ticks during the acquisition (i.e., fed larvae), larva-nymph intermolt (i.e., intermolt larvae), and transmission (i.e., 24-, 48-, or 72-hour-postfeeding nymph) phases were collected. As previously described (27), total RNA was extracted, and gene copy numbers in these samples were measured by qRT-PCR via absolute quantification. As shown in Fig. 1A, bosR was expressed in ticks during the acquisition, intermolt, and transmission phases, whereas bosR expression was undetectable in uninfected flat larvae (as a negative control). More specifically, approximately 4.4 or 10.4 copies of bosR transcripts per 100 flaB transcripts were detected in fed larvae or intermolt larvae, respectively (Fig. 1A). In ticks during the transmission phase, there were increasing levels of bosR transcription during blood feeding. Specifically, in nymphs after 24, 48, or 72 hours of feeding on mice, approximately 3.8, 4.2, or 13.4 copies of bosR transcripts per 100 flaB transcripts, respectively, were detected (Fig. 1A).

bosR transcription was also assessed in tissues isolated from infected mice. To this end, mouse tissues (skin, heart, joint, and bladder) were collected at various times (7, 14, 21, 28, or 50 days) postinfection. As shown in Fig. 1B, bosR was highly expressed in infected mice throughout the mouse infection phase. Specifically, in skin and bladder samples, bosR expression was detected relatively early (7, 14, and 21 days postinfection) but was not detected during later phases of infection (28 or 50 days postinfection) (Fig. 1B). In contrast, low levels of bosR expression were observed in heart tissues during early infection (7, 14, 21, and 28 days postinfection). However, during later phases of infection (50 days postinfection), bosR was highly expressed by B. burgdorferi in the heart (Fig. 1B). These data not only further substantiate the important functions of BosR during mammalian infection by B. burgdorferi, but also imply a critical role(s) of BosR, and perhaps other BosR-dependent genes, in tissue colonization and/or tissue tropism at different stages of infection.

We previously found that rpoS expression is triggered only in nymphs during borrelial transmission and remains active in B. burgdorferi-infected animals (27). Our data from the current study show that bosR is also highly expressed under these conditions, providing further evidence for the dependence of rpoS expression on BosR. We also observed high levels of bosR expression in ticks during the acquisition and intermolt phases. However, rpoS expression is undetectable in these ticks, potentially resulting from repression by BadR. Previously, we (58) and others (60) reported that BadR negatively influences rpoS expression by binding to the rpoS promoter. Furthermore, our previous data indicated that badR is also highly expressed in ticks during the acquisition and intermolt phases (58). Therefore, high levels of BadR may occupy the rpoS promoter region, which could inhibit the recruitment of BosR, σ⁵⁴ (RpoN), and Rrp2 to the rpoS promoter. As a result, rpoS expression likely is repressed. Alternatively, inefficient rpoS expression in these ticks may result from other, unknown mechanisms; for example, there may be low levels of phosphorylated Rrp2, inasmuch as bEBPs typically are phosphorylated before they are capable of activating bacterial σ⁵⁴ (61, 62). Regardless, our combined results indicate that varying levels of bosR are expressed in infected ticks and mice during the entire life cycle of B. burgdorferi. These data constitute biologically relevant information regarding the role of BosR in B. burgdorferi's parasitic strategy.

Expression of bosR is influenced by various environmental factors.

Varying levels of bosR expression in infected ticks and mice suggested that gene expression is influenced by tick or host environmental factors that B. burgdorferi encounters during its transit between ticks and mammals. To date, in vitro culture conditions have been widely used to partially mimic the tick or mammal environment (4, 28, 55, 63–65). For example, regarding the midgut of unfed nymphs, the lower cell density conditions can be reproduced by in vitro culture at a lower temperature (23°C) and/or at early stages of borrelial growth. In contrast, cultures at elevated temperature (37°C), reduced pH (pH 6.8), and late-log or stationary growth phase ostensibly partially mimic the environment observed in the midgut of nymphs taking a blood meal. Previous studies indicated that the RpoN-RpoS regulatory pathway is switched on in nymphs during blood feeding, as well as in B. burgdorferi grown at 37°C (reaching higher cell density or the stationary phase). Moreover, the repression of the RpoN-RpoS pathway in unfed nymphs has also been reproduced in B. burgdorferi cultivated at 23°C in BSK medium at pH 7.6. Along these lines, we also examined the effects of these environmental factors on bosR expression. To this end, B. burgdorferi was cultivated under different conditions, such as 37°C versus 23°C or pH 7.6 versus pH 6.8 (Fig. 2). Analyses of whole-cell lysates by SDS-PAGE and immunoblotting indicated that, relative to gene expression in cultures at 23°C (lane 1), BosR production was highly induced under conditions of elevated temperature (37°C) (Fig. 2A and B, lanes 2). Compared with gene expression in spirochetes cultured at pH 7.6 (lane 2), BosR synthesis was also greatly increased at reduced pH (pH 6.8) (Fig. 2A and B, lanes 3). Changes in gene expression were also quantitated via qRT-PCR analyses. As shown in Fig. 2C and D, transcription of bosR was found to be highly induced at 37°C (versus 23°C) or pH 6.8 (versus pH 7.6). In addition, in cultures at 37°C or pH 6.8, RpoS and RpoS-dependent OspC were also highly induced (Fig. 2), corroborating the activation of rpoS transcription by BosR. These data are in good agreement with the above-mentioned in vivo data showing that bosR expression (Fig. 1A) and the RpoN-RpoS pathway (27) are induced in nymphs during feeding.

FIG 2.

Expression of bosR is influenced by various environmental factors. B. burgdorferi was cultivated in BSK-II medium. The culture was collected when growth reached the late log phase. Gene expression was analyzed by SDS-PAGE (A), immunoblotting (B), and qRT-PCR (C and D). (A and B) Spirochetes were cultivated in BSK medium at pH 7.6 and 23°C (lane 1), pH 7.6 and 37°C (lane 2), or pH 6.8 and 37°C (lane 3); ∼4 × 10⁷ spirochetes were loaded onto each lane of an SDS-12.5% polyacrylamide gel. The arrow in panel A indicates the position of OspC. In panel B, specific antibodies used in the immunoblots are indicated. (C and D) In qRT-PCR analyses, data were collected from at least three independent tests and normalized using the B. burgdorferi flaB gene as an internal control. The ratios of gene expression between spirochetes cultured at 37°C and 23°C (C) or at pH 6.8 and pH 7.6 (D) are presented as mean measurements and SD. The asterisks indicate statistical significance using the Student t test (P < 0.05).

Analyses of the promoter elements driving bosR transcription.

In a previous study (38), we reported that bosR is within an operon with bb0648 and bb0646. To investigate further the regulation of bosR expression in B. burgdorferi, we analyzed the 5′ putative regulatory region upstream of bb0648 using BPROM (SoftBerry), a bacterial promoter prediction program. A typical bacterial σ⁷⁰ promoter (P1) (Fig. 3A) harboring canonical −10/−35 elements was identified; this promoter may drive the transcription of the bb0648-bosR-bb0646 operon. In addition, the 5′ region upstream of bosR (UPbosR) has been predicted to be a tandem intergenic spacer (IGS) harboring putative cis-regulatory sequences (66). In particular, one putative promoter was also predicted in UPbosR in a previous study (42). In our study, we also identified the putative σ⁷⁰ promoter (i.e., P2) harboring typical −35 and −10 elements (Fig. 3B) when analyzing the UPbosR region using the BPROM program. However, it has remained an open question whether the P2 promoter is functional (to drive bosR transcription) in B. burgdorferi.

FIG 3.

Prediction of putative promoters P1 and P2 for bosR expression. (A) Sequences of the bb0649-bb0648 intergenic region. The arrows indicate the start codons (ATG) of bb0648 and bb0649 (on the complementary strand). The proposed −10 and −35 elements of the P1 promoter are underlined. (B) Architecture of the 5′ regulatory region upstream of bosR. The stop codon (TAA) of bb0648 is marked by an asterisk. The arrow indicates the start codon (ATG) of bosR. The proposed −10 and −35 elements of the P2 promoter are underlined. Putative BosR binding sites are boxed. (C) Diagram of luciferase reporter constructs. The lines represent DNA fragments cloned into the promoterless luciferase reporter vector pOY63; approximate start and end sites of the DNA relative to their genomic locations are shown. The gray ellipses represent the putative promoters P1 and P2 in the sequences upstream of bb0648 and bosR, respectively.

Recently, a luciferase reporter system was developed and employed to investigate gene expression in B. burgdorferi (57). Compared with other, traditional methods for measuring gene expression, the quantitative luciferase assay is more sensitive and convenient and is relatively simple. To study the regulation of bosR expression in B. burgdorferi, we first amplified a DNA fragment containing the putative promoter P1, bb0648, the 5′ region upstream of bosR (UPbosR), the predicted RBS for the translation initiation of BosR, and the first 66-bp sequence of the bosR open reading frame (ORF) (from the ATG start codon) (Fig. 3C). This fragment was then cloned into the promoterless luc reporter vector pOY63. In the resultant shuttle vector, pOY463 (Fig. 3C), the fragment P1-bb0648-bosR was fused in frame with the promoterless luc gene, allowing the transcription levels of bosR to be monitored directly by measuring luciferase activities. This construct was transformed into the low-passage, virulent WT strain 297. As shown in Fig. 4, luciferase activity was detected in the strain harboring pOY463, but not in the strain harboring the cloning vector pOY63. These data suggest that pOY463 harbors a functional promoter (i.e., P1) and that this construct can be used to investigate further the trans-acting factors involved in the regulation of bosR expression in B. burgdorferi.

FIG 4.

Luciferase expression in B. burgdorferi transformed with luciferase reporter constructs. B. burgdorferi was grown in BSK-II medium at 37°C (pH 7.6), and spirochetes were harvested at late-log phase. Luciferase activity is indicated. Bb, B. burgdorferi. The results of three independent experiments are presented as mean values and SD. The asterisks indicate statistical significance using Student's t test (P < 0.05 compared with luciferase activity in B. burgdorferi harboring the promoterless vector pOY63). As shown in Fig. 3C, pOY463 contains P1-bb0648-UPbosR-bosR-luc, pOY90 contains UPbosR-bosR-luc, and pOY89.2 contains P1-bb0648-luc; pOY63 is the promoterless luc vector.

To determine whether the P2 promoter is functional in B. burgdorferi, we amplified from B. burgdorferi a DNA fragment containing UPbosR and the first 66-bp sequence of the bosR ORF (Fig. 3C). When this fragment was cloned into pOY63, we obtained the luciferase reporter construct pOY90. We next introduced pOY90 into the WT strain 297 and determined that luciferase was expressed in the resultant strain (Fig. 4). These data suggest that the promoter in UPbosR likely is capable of driving bosR transcription, at least under the conditions tested.

We also amplified another DNA fragment starting from 5′ of bb0649 to 5′ of bb0648 (harboring P1 and 129 bp of bb0648) by PCR. The DNA was then cloned into pOY63, resulting in another shuttle vector, pOY89.2 (Fig. 3C). In pOY89.2, the DNA fragment P1-bb0648 was fused in frame with the promoterless luc, allowing the expression level of bb0648 to be monitored by luciferase assays. When pOY89.2 was introduced into WT strain 297, luciferase activity was detected (Fig. 4), albeit at a much lower level than from pOY463, suggesting that bb0648 sequences or sequences in UPbosR contain other elements (such as P2) influencing the efficiency of bosR transcription or translation.

BosR encodes a positive autoregulator.

Previously, it was reported that the 5′ DNA region upstream of bosR (UPbosR) contains multiple sites for the binding of BosR (37, 42), implying that bosR expression is regulated by BosR itself. However, to date, direct experimental evidence to support this hypothesis has been lacking. To this end, pOY463 (containing P1-bb0648-UPbosR-bosR-luc) was introduced into the bosR mutant OY08. Relative to luciferase activity in the WT strain 297, luciferase expression in the bosR mutant was significantly reduced (Fig. 5A). These data were further corroborated by qRT-PCR analyses (Fig. 5B) (showing a significant decrease in luc transcripts in the bosR mutant). We also introduced pOY90 (containing UPbosR-bosR-luc) or pOY89.2 (containing P1-bb0648-luc) into the bosR mutant OY08. It is noteworthy that rBosR binds to multiple sites (in UPbosR) within the DNA fragments cloned into pOY463 (containing P1-bb0648-UPbosR-bosR-luc) and pOY90 (containing UPbosR-bosR-luc) (37, 42), whereas rBosR does not bind to the fragment cloned into pOY89.2 (containing P1-bb0648-luc) (data not shown). Relative to luciferase levels in the WT strain 297 (with pOY90 containing UPbosR-bosR-luc), significantly lower levels of luciferase were detected in the bosR mutant OY08 (with pOY90) (Fig. 5C). However, pOY89.2 (containing P1-bb0648-luc) expressed comparable levels of luciferase in the WT strain and in the bosR mutant (Fig. 5D). These results suggest that BosR positively influences the expression of bosR in B. burgdorferi.

FIG 5.

BosR positively influences expression of bosR in B. burgdorferi. Luciferase reporter constructs were introduced into the WT strain or the bosR mutant. Spirochetes were grown in BSK-II medium at 37°C (pH 7.6) and harvested at late-log phase. Gene expression was analyzed by luciferase assays (A, C, D, and E) or qRT-PCR (B). (A, C, D, and E) Luciferase expression of the WT or the bosR mutant harboring one reporter construct was determined. Luciferase activity is indicated. Bb, B. burgdorferi. (B) Ratio of luc transcription between the bosR mutant and the WT strain (harboring pOY463). The data were normalized using the aadA gene as an internal control. The results of three tests are shown as means and standard errors of the mean (SEM). The asterisks indicate statistical significance using Student's t test (P < 0.05). As shown in Fig. 3C, pOY463 contains P1-bb0648-UPbosR-bosR-luc, pOY90 contains UPbosR-bosR-luc, and pOY89.2 contains P1-bb0648-luc.

Previous studies have demonstrated that the expression of ospC and dbpBA is facilitated by BosR (via RpoS) (35–37, 39, 58). As controls, we created a luc reporter construct by fusing the minimal promoter driving the transcription of ospC (minPospC) (67) with the promoterless luc in pOY63. In addition, we created another luciferase reporter construct containing the minimal promoter for the dbpBA operon (minPdbpBA) (53). These constructs were then introduced into the WT strain 297 or the bosR mutant OY08. As expected, high expression of luciferase from minPospC or minPdbpBA was detected in the WT strain 297 (Fig. 5E). However, luciferase activity in the bosR mutant was markedly reduced.

Implications and conclusions.

This is the first study, to our knowledge, that investigates the expression of bosR during the natural tick-mammal life cycle of B. burgdorferi. Our data indicate that bosR is expressed throughout all the stages of tick and animal infection, providing further validation of the important role of BosR in B. burgdorferi infectivity and pathogenesis. Our study also strongly suggests that BosR functions as an autoregulator in B. burgdorferi and offers the first direct experimental evidence for a longstanding question in the field of B. burgdorferi gene regulation. The positive regulation by BosR ostensibly is dependent on the multiple BosR binding sites harbored in the DNA sequences upstream of bosR, because pOY463 and pOY90 (containing DNA that binds rBosR), but not pOY89.2 (with DNA that does not bind to rBosR), show significantly higher luciferase activities in the WT strain than in the bosR mutant. Given that BosR binding sites overlap the putative −35 and −10 promoter elements (Fig. 3B), the contribution of each individual site (and potential key nucleotides) to gene expression cannot be examined via conventional methods (such as promoter deletion analysis). Future experiments are thus warranted to develop a new system(s) to investigate the roles of these DNA elements in protein recruitment, gene expression during infection, and disease pathogenesis.

Funding Statement

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.