Host-specific functional compartmentalization within the oligopeptide transporter during the Borrelia burgdorferi enzootic cycle

Abstract

Borrelia burgdorferi must acquire all of its amino acids (AAs) from its arthropod vector and vertebrate host. Previously, we determined that peptide uptake via the oligopeptide (Opp) ABC transporter is essential for spirochete viability in vitro and during infection. Our prior study also suggested that B. burgdorferi employs temporal regulation in concert with structural variation of oligopeptide-binding proteins (OppAs) to meet its AA requirements in each biological niche. Herein, we evaluated the contributions to the B. burgdorferi enzootic cycle of three of the spirochete's five OppAs (OppA1, OppA2, and OppA5). An oppA1 transposon (tn) mutant lysed in the hyperosmolar environment of the feeding tick, suggesting that OppA1 imports amino acids required for osmoprotection. The oppA2tn mutant displayed a profound defect in hematogenous dissemination in mice, yet persisted within skin while inducing only a minimal antibody response. These results, along with slightly decreased growth of the oppA2tn mutant within DMCs, suggest that OppA2 serves a minor nutritive role, while its dissemination defect points to an as yet uncharacterized signaling function. Previously, we identified a role for OppA5 in spirochete persistence within the mammalian host. We now show that the oppA5tn mutant displayed no defect during the tick phase of the cycle and could be tick-transmitted to naïve mice. Instead of working in tandem, however, OppA2 and OppA5 appear to function in a hierarchical manner; the ability of OppA5 to promote persistence relies upon the ability of OppA2 to facilitate dissemination. Structural homology models demonstrated variations within the binding pockets of OppA1, 2, and 5 indicative of different peptide repertoires. Rather than being redundant, B. burgdorferi's multiplicity of Opp binding proteins enables host-specific functional compartmentalization during the spirochete lifecycle.

Fig 1. In vitro characterization of oppA1 and oppA2 mutants and complements.

(A) Schematics of the wt, oppA1tn and oppA2tn mutants and the corresponding complements in cp26. The transposon (tn) insertion sites (bp of coding sequence) are indicated. aacC and aadA confer gentamycin and streptomycin resistance, respectively. qRT-PCR primers specific to the tn insertion site are noted with small black arrows. (B-C) Transcript copy numbers of oppA1-3 (mean ± SEM, normalized to flaB) in (B) wt, oppA1tn, and oppA1c and (C) wt, oppA2tn, and oppA2c determined from triplicate biological replicates and quadruplicate technical replicates. Statistical analyses were performed using unpaired Student’s t test.

Fig 2. oppA1 is essential in ticks.

Spirochete burdens as assessed by (A) colony counts (mean ± SEM) and (B) qPCR in midguts of larvae naturally fed on mice infected with wt, oppA1tn, and oppA1c. Each data point represents a separate pool of larvae. (C) Detection of spirochetes by immunofluorescence analysis of midguts from replete larvae fed on mice infected with wt, oppA1tn, or oppA1c (400x total magnification). Spirochete burdens (mean ± SEM) assessed by (D) colony counts and (E) qPCR in midguts of larvae immersed in wt, oppA1tn, or oppA1c cultures and subsequently fed on naïve C3H/HeJ. Each data point represents a separate pool of larvae. Statistical analysis of tick studies was determined by unpaired Student’s t test.

Fig 3. oppA2 is essential for dissemination within the mammal.

(A-C) Maps of skin sample sites and culture results for mice dorsally-infected with 1 x 10⁴ wt (A), oppA2tn (B), or oppA2c (C) (n = 5). Inoculation sites are designated with a white asterisk. Dotted circles denote capsule placement for larval feeding. (D) Immunoblot analysis of infected mouse sera against B. burgdorferi whole cell lysates.

Fig 4. oppA2tn survives comparably to wt during ex vivo blood culture.

(A) Darkfield microscopy (DF) and epifluorescence microscopy of Hoechst- (HS—Viable) and propidium iodide-stained (PI–Dead) wt, oppA2tn, and oppA2c spirochetes after 24 hrs ofincubation in whole mouse blood, 400x magnification. (B) Viable spirochetes recovered from blood culture assessed by semi-solid plating at timepoints 0, 24, and 48 hrs.

Fig 5. oppA2 is dispensable during larval acquisition.

(A) Two preliminary experiments measuring spirochete burdens (mean ±SEM) assessed by colony counts in midguts from larvae naturally fed on mice infected with wt, oppA2tn, or oppA2c. (B) Colony counts (mean ±SEM) from midguts of larvae immersion fed with wt, oppA2tn, or oppA2c cultures and subsequently fed on naïve C3H/HeJ. (C) Colony counts (mean ±SEM) from midguts of larvae naturally fed on mice infected with wt-dorsally, oppA2tn-dorsally, oppA2tn-ventrally, or oppA2c-dorsally. (D) Maps of skin sample sites and culture results and immunoblot analysis using B. burgdorferi whole cell lysates of sera from mice ventrally-infected with 1 x 10⁴ oppA2tn (n = 5). Inoculation sites are designated with a white asterisk. Dotted circle denotes capsule placement for larval feeding.

Fig 6. oppA2 is dispensible for survival in the tick midgut and transmission.

(A-B) Colony counts (mean ±SEM) from midguts of (A) flat and (B) fed nymphs infected with wt or oppA2tn. (C) Colony counts (mean ±SEM) from midguts of fed nymphs infected with wt, oppA2tn, or oppA2c. Each data point represents a separate pool of ticks. Statistical analysis of tick studies was determined by unpaired Student’s t test. (D-F) Maps of skin sample sites and culture results for mice dorsally-infected via nymph feeding with wt (A), oppA2tn (B), or oppA2c (C) (n = 5). Dotted circles denote capsule placement for nymphal feeding.

Fig 7. oppA5 is dispensable in ticks.

(A) Colony counts (mean ±SEM) of midguts from larvae immersion fed with wt and oppA5tn, (B) subsequent molted flat nymphs, and (C) fed nymphs. Each data point represents a separate pool of ticks. Statistical analysis was determined by unpaired Student’s t test.

Fig 8. The binding pockets of OppA1-5 demonstrate variations in electrostatic distribution.

(A) Ribbon model of OppA4 crystal structure (PDB: 4GL8) [26] showing domains I (purple–hinge region), II (yellow), and III (green) and the Ala₄ liganded peptide (black). (B)Electrostatic distributions unliganded (open) OppA1-5, modeled against unliganded E. coli OppA (PDB: 3TCH) [24]. Electrostatic models have been rotated 45° on the z-axis to show the permease-binding regions (region framing the binding site as clearly seen on the positively charged surface of OppA3) or the binding pockets (clearly distinguished as the positively charged pocket along the center of OppA1). Figure was adapted from Groshong et al. [14].