The hybrid histidine kinase Hk1 is part of a two-component system that is essential for survival of Borrelia burgdorferi in feeding Ixodes scapularis ticks

Abstract

Two-component systems (TCS) are principal mechanisms by which bacteria adapt to their surroundings. Borrelia burgdorferi encodes only two TCS. One is comprised of a histidine kinase, Hk2, and the response regulator Rrp2. While the contribution of Hk2 remains unclear, Rrp2 is part of a regulatory pathway involving the spirochete's alternate sigma factors, RpoN and RpoS. Genes within the Rrp2/RpoN/RpoS regulon function to promote tick transmission and early infection. The other TCS consists of a hybrid histidine kinase, Hk1, and the response regulator Rrp1. Hk1 is composed of two periplasmic sensor domains (D1 and D2), followed by conserved cytoplasmic histidine kinase core, REC, and Hpt domains. In addition to its REC domain, Rrp1 contains a GGDEF motif characteristic of diguanylate cyclases. To investigate the role of Hk1 during the enzootic cycle, we inactivated this gene in two virulent backgrounds. Extensive characterization of the resulting mutants revealed a dramatic phenotype whereby Hk1-deficient spirochetes are virulent in mice and able to migrate out of the bite site during feeding but are killed within the midgut following acquisition. We hypothesize that the phosphorelay between Hk1 and Rrp1 is initiated by the binding of feeding-specific ligand(s) to Hk1 sensor domain D1 and/or D2. Once activated, Rrp1 directs the synthesis of cyclic dimeric GMP (c-di-GMP), which, in turn, modulates the expression and/or activity of gene products required for survival within feeding ticks. In contrast to the Rrp2/RpoN/RpoS pathway, which is active only within feeding nymphs, the Hk1/Rrp1 TCS is essential for survival during both larval and nymphal blood meals.

Fig. 1.

Expression profiling of hk1 and rrp1. Values represent the average flaB-normalized transcript copy number ± standard error of the mean for each gene. Values for hk1 were significantly different (P < 0.05) for the following comparisons: flat nymph versus fed larvae and fed nymph, fed larvae versus fed nymph and DMC, and fed nymph versus DMC. Values for rrp1 were significantly different (P < 0.05) for the following comparisons: fed larvae versus flat nymph, fed nymph and DMC, and DMC versus flat nymph and fed larvae. hk1 was expressed at significantly (P < 0.05) higher levels than rrp1 in the same sample under all four conditions examined. The sequences of the forward and reverse primers used to detect hk1 and rrp1 are provided in Table 2, and their locations are shown in Fig. 2A (hk1-F and hk1-R are designated by arrows 3 and 4, respectively; rrpl-F and rrpl-R are designated by arrows 5 and 6, respectively).

Fig. 2.

(A) Strategy for inactivation of hk1. The hk1 coding sequence plus upstream and downstream flanking regions was amplified from strain 297 using primers ups.hk1-5′ and dwns.hk1-3′ (1 and 2). The hk1 coding sequence was disrupted by insertion of a P_flgB-aadA antibiotic resistance cassette into an HpaI restriction site present within the endogenous hk1 gene, yielding pMC1389. Only the relevant portion of pMC1389 is shown. Insertion of the hk1 KO allele was confirmed in strain 297 and B31 mutant isolates using primers hk1-KO-5′ and hk1-KO-3′ (7 and 8) (see Fig. S2 in the supplemental material). Primer sequences are provided in Table 2. (B) Hk1-deficient spirochetes continue to express Rrp1. Semiquantitative RT-PCR was performed on RNAs isolated from a wild-type (WT) (297 and B31) and hk1 KO strains (Bb508, Bb807, and Bb1197) using primers specific for rrp1 (5 and 6) and flaB. RT indicates the absence (−) or presence (+) of reverse transcriptase in the reaction mixture. Purified genomic DNA (gDNA) was used as positive controls. (C) Detection of Rrp1 by immunoblotting. Whole-cell lysates of CE162 (WT 297), Bb508 (hk1-KO), and a previously characterized B31 5A13 Δrrp1 isolate (58) were immunoblotted with rat polyclonal antisera directed against Rrp1 (58) with FlaB used as a loading control. α, anti.

Fig. 3.

Hk1 is not required for mammalian host adaptation. Whole-cell lysates from CE162 (WT 297) and Bb508 (Hk1 mutant) were separated by SDS-PAGE, stained with silver, and immunoblotted using antisera directed against DbpA (33), Lp6.6/BBA62 (42), and OspE (1) with FlaB used as a loading control. MWM, molecular weight marker in thousands; α, anti.

Fig. 4.

Spirochetes lacking Hk1 are infectious in mice by syringe inoculation. Tissues were collected from infected mice (5 mice per group) at 4 weeks postinoculation with 10⁴ spirochetes of CE162 (WT 297), Bb508, Bb807, B31 5A4 NP1 (WT B31), or Bb1197. Spirochete genome copies, here and elsewhere, were determined using TaqMan assays for spirochetal flaB and murine nidogen (nido). The mean nido-normalized flaB value for all mice within a group is indicated by a horizontal line. The asterisk indicates a statistical difference (P < 0.05) between CE162 and Bb807 in ear tissue.

Fig. 5.

hk1 mutants are killed within larvae fed to repletion on syringe-inoculated mice. Data represent spirochete burdens within larvae fed to repletion on C3H/HeJ mice 2 to 3 weeks following syringe inoculation with wild-type parents (CE162 and B31 5A4 NP1), hk1 mutants (Bb508, Bb807, and Bb1197), or complemented hk1 mutants (Bb1363 and Bb1367). (A) Spirochete genome copies within pools of 15 larvae collected from individual mice (3 mice per isolate) were determined using a TaqMan assay for flaB. Bars represent the mean ± standard error of the mean for each isolate. The normalized flaB values for larvae fed on mice infected with hk1 mutants (hk1 KO strains) was significantly lower (*, P = 0.0021; #, P < 0.0001) than that for larvae fed on mice infected with the corresponding parent or complements (compl). (B) Representative micrographs of larvae fed to repletion on syringe-inoculated mice. Pools of 15 larvae from each mouse (3 mice per group) were assessed by IFA using FITC-conjugated anti-Borrelia antibody.

Fig. 6.

hk1 mutants are killed within the midguts of larvae infected by immersion. Larvae infected by immersion with wild-type 297 and B31 (CE162 and 5A4 NP1), hk1-KO mutants (Bb508, Bb807 and Bb1197), or complement (Bb1363) were fed to repletion on naïve mice. (A) Spirochete genome copies within triplicate pools of 15 fed larvae. Bars represent the mean flaB values per larva ± standard error of the mean for each isolate. Asterisks indicate significantly (P < 0.0001) lower values for hk1 mutants than for the corresponding parent. (B) Representative immunofluorescence micrographs of fed larvae infected by immersion. Pools of 15 fed larvae infected with each strain were assessed by IFA using FITC-conjugated anti-Borrelia antibody.

Fig. 7.

Hk1 is required for survival within fed but not flat nymphal midguts. Composite confocal image showing the distribution of spirochetes within nymphs infected by rectal microinjection with CE162 (WT 297), Bb807 (hk1-KO mutant), Bb1197 (hk1-KO mutant), 5A4 NP1 (WT B31), or complement (Bb1363). Spirochetes were detected within midguts carefully dissected from unfed and fed nymphs, forcibly removed at 36 h postattachment, using FITC-conjugated anti-Borrelia antibody. The tick midgut was counterstained with propidium iodide (red). Scale bar, 20 μm.

Fig. 8.

Working model for Hk1/Rrp1 and Rrp2/RpoN/RpoS TCS during the enzootic cycle. At the onset of feeding, Hk1 senses unique host- and/or tick-derived molecules generated within the feeding site. These feeding-specific molecules must be small enough to rapidly traverse the spirochete's outer membrane and engage Hk1's D1 and D2 periplasmic sensor domains. During acquisition, spirochetes first encounter these molecules as they migrate into the feeding site, while spirochetes within flat nymphs (or larvae infected by immersion) would encounter these ligands solely within the midgut (3). The Hk1/Rrp1-directed synthesis of c-di-GMP initiates an adaptive response that enables spirochetes either to evade killing by noxious substances within midgut epithelium as feeding progresses (21, 37, 50, 73, 74) or to adjust metabolically to growth within the arthropod vector. In contrast to the Rrp2/RpoN/RpoS pathway, which is active (ON) only within feeding nymphs, the Hk1/Rrp1 TCS is essential for survival during both the larval and nymphal blood meals.