Unveiling the impact of Leptospira TolC efflux protein on host tissue adherence, complement evasion, and diagnostic potential

Abstract

The TolC family protein of Leptospira is a type I outer membrane efflux protein. Phylogenetic analysis revealed significant sequence conservation among pathogenic Leptospira species (83%-98% identity) compared with intermediate and saprophytic species. Structural modeling indicated a composition of six β-strands and 10 α-helices arranged in two repeats, resembling bacterial outer membrane efflux proteins. Recombinant TolC (rTolC), expressed in a heterologous host and purified via Ni-NTA chromatography, maintained its secondary structural integrity, as verified by circular dichroism spectroscopy. Polyclonal antibodies against rTolC detected native TolC expression in pathogenic Leptospira but not in nonpathogenic ones. Immunoassays and detergent fractionation assays indicated surface localization of TolC. The rTolC's recognition by sera from leptospirosis-infected hosts across species suggests its utility as a diagnostic marker. Notably, rTolC demonstrated binding affinity for various extracellular matrix components, including collagen and chondroitin sulfate A, as well as plasma proteins such as factor H, C3b, and plasminogen, indicating potential roles in tissue adhesion and immune evasion. Functional assays demonstrated that rTolC-bound FH retained cofactor activity for C3b cleavage, highlighting TolC's role in complement regulation. The rTolC protein inhibited both the alternative and the classical pathway-mediated membrane attack complex (MAC) deposition in vitro. Blocking surface-expressed TolC on leptospires using specific antibodies reduced FH acquisition by Leptospira and increased MAC deposition on the spirochete. These findings indicate that TolC contributes to leptospiral virulence by promoting host tissue colonization and evading the immune response, presenting it as a potential target for diagnostic and therapeutic strategies.

Keywords: C3b; ECM; Leptospira; MAC; OMP; TolC; complement; efflux protein; factor H.

Fig 1

Evolutionary analysis of TolC protein of Leptospira. (A) Phylogenetic analysis of TolC protein in different Leptospira spp. The amino acid sequence of the L. interrogans serovar Copenhageni TolC protein was retrieved from the NCBI protein database, and a total of 26 orthologs of TolC protein were retrieved through NCBI protein BLAST. The obtained sequences were aligned, and the phylogenetic tree was constructed using the MEGA, version 11 program. The tree was generated using the maximum likelihood algorithm and 1,000 bootstrap replications. The pathogenic, intermediate, and saprophytic Leptospira are displayed in red, blue, and green, respectively. The percentage of the tree in which the associated taxa clustered together is shown next to the branches. A bootstrap value greater than 50 at each cluster shows the reliability of the data. The tree was drawn to scale, with branch lengths measured in the number of substitutions per site. The resulting phylograms show the high level of sequence conservation for TolC among pathogenic Leptospira serovars. (B) The modeled structure of the leptospiral TolC subunit. The N-terminal region, consisting of the signal peptide (residues 1–28) and the unstructured region (residues 29–59), is depicted in maroon and blue, respectively. The β-barrel domains (S1–S6) are shown in magenta and the α-helical domains (H1–H10) in green.

Fig 2

Characterization of the TolC protein of Leptospira. (A) The recombinant TolC protein is displayed on a 12% SDS-polyacrylamide gel stained with Coomassie brilliant blue stain. The recombinant protein is purified using Ni-NTA affinity column chromatography under native conditions. (B) Far-ultraviolet circular dichroism (CD) spectra of rTolC. The CD spectra (over a wavelength range of 190–250 nm) are shown as an average of three scans. (C) Titer of the polyclonal antibody raised against rTolC. Pooled sera from mice inoculated with rTolC protein were used to determine the titer of polyclonal antibodies capable of detecting rTolC using enzyme-linked immunosorbent assay (ELISA). Sera obtained before the immunization of the mice (pre-immune sera) was used as a control for the evaluation of antibody titer, and data are presented as means ± SEM from two independent experiments. (D) Immunoblot analysis of native and recombinant TolC antigen with anti-TolC antibody. The mice anti-TolC anti-sera can detect the purified rTolC and the native TolC protein in pathogenic (L. interrogans serovar Copenhageni and serovar Lai) and saprophytic (L. biflexa serovar Patoc) Leptospira serovars. The protein standard molecular size marker (M) is indicated on the left side of the SDS-polyacrylamide gel and immunoblot.

Fig 3

Cellular localization of the TolC protein in Leptospira. (A) Immunogold electron microscopy of Leptospira. L. interrogans were treated with antisera (1:100) specific to TolC protein, ErpY-like protein (positive control), rLipL31 (negative control), or preimmune mice serum (negative control). Specific binding of the antibodies was probed with the gold-conjugated anti-mouse or anti-rabbit secondary polyclonal IgG (1:30). The arrows indicate the specific binding of IgG−colloidal-gold conjugates to the surface-exposed protein of spirochete. (B) Immunofluorescence assay to detect TolC protein. Leptospires were treated with various antisera (1:100) generated against TolC, ErpY-like protein (positive control), LipL31 (negative control), and preimmune mice serum (negative control). Fluorescein isothiocyanate (FITC)-conjugated secondary antibodies (1:100) were used to detect the surface-bound antibodies (top panels). Leptospiral cells were identified by staining the nucleic acid with propidium iodide (PI) (middle panels). The co-localization of the proteins and cells is shown by the merged images (bottom panels). (C) Phase partitioning of whole-cell protein of L. interrogans serovar Copenhageni using Triton X-114 and its immunoblot with the anti-TolC antibody. Live leptospires were subjected to 2% Triton X-114 to partition proteins into protoplasmic cylinder (PC), detergent (Det), and aqueous-phase (Aq) fractions. Phase-partitioned proteins were resolved on 12% SDS-polyacrylamide gel, electroblotted, on nitrocellulose membrane, and probed with anti-TolC (top), anti-ErpY-like (middle), or anti-LipL31 (bottom) antibodies. Whole-cell lysate of L. interrogans serovar Copenhageni (W) was used as a marker. The protein standard molecular size marker (M) is indicated on the left side of each blot. (D) Protease accessibility assay using proteinase K (PK). Live leptospire suspensions were incubated PK at the indicated time points. The suspensions were harvested, washed, resuspended in PBS, and coated on a microtiter plate. The presence of TolC, ErpY-like, and LipL31 on the bacterial surface was detected using specific antibodies. The bars represent mean ± SEM from three independent experiments performed with two technical replicates (**P < 0.01).

Fig 4

The rTolC protein is recognized by leptospirosis-positive sera from various hosts. The recombinant proteins rTolC and rLoa22 (positive control) were coated onto microtiter plate wells, and their reactivity with human, bovine, and canine sera testing MAT-positive or -negative for leptospirosis was measured by ELISA. The mean absorbance values obtained from each serum sample are indicated. Solid black horizontal lines represent the mean absorbance for each group. The dashed black horizontal lines denote the assay’s cutoff value [(mean absorbance of leptospirosis MAT-negative samples) + (2 × standard deviations)]. (A) ELISA to detect rTolC antigen (400 ng/well) using MAT-positive (n = 38) and -negative (n = 14) serum samples of humans (1:100) for leptospirosis. (B) Detection of rLoa22 (400 ng) antigen using MAT-positive and -negative human serum samples. (C) Immunoblot of antigens (rTolC and rLoa22) using the pooled human MAT-positive (left) and MAT-negative serum (Right). (D) Detection of rTolC using bovine serum samples testing MAT-positive (n = 40) or -negative (n = 15). (E) ELISA to detect rLoa22 antigen using bovine serum samples. (F) Immunoblot using pooled bovine serum testing MAT-positive or -negative for leptospirosis. (G) ELISA is used to detect rTolC antigen using a canine serum to test MAT-positive (n = 15) or -negative (n = 10) samples. (H) ELISA to detect the rLoa22 antigen using canine serum samples. (I) Immunoblot using pooled canine serum testing MAT-positive or -negative. The protein standard molecular size marker (M) is indicated on the left side of each blot.

Fig 5

Recombinant TolC protein binds to the host extracellular matrix and plasma components. The wells of a microtiter plate were coated with 1 µg of ECM components (laminin, fibronectin, collagen, chondroitin sulfate A, hyaluronic acid, heparan sulfate, and elastin) or plasma components (thrombin, fibrinogen, plasminogen, factor H, C3b, C4BP, and factor I) and binding of rTolC (0.1–6 µM) to the immobilized host components was analyzed through ELISA using anti-TolC antibodies. The plasma fetuin (highly glycosylated protein) and BSA (nonglycosylated protein) were negative controls for ligands. The rLoa22, known to interact moderately with host ECM components, was included in the assay as a negative-control protein. (A) Interaction between rTolC protein (1 µg) and host ECM and plasma components. The bar represents mean ± SEM from three independent experiments performed in triplicate. For statistical analyses, the binding of rTolC protein with host ligands was compared to that with fetuin or BSA by the two-tailed t-test (***P < 0.001). (B) Dose-dependent binding of the rTolC protein to host components. (C) The saturation data for the interaction between rTolC and plasminogen was fitted using the Hill equation, yielding a dissociation constant (K_D) of 336 ± 26 nM. (D) The saturation data for the interaction between rTolC and FH was fitted using a Hill curve, with the K_D determined to be 391 ± 45 nM. (E) The saturation data of the rTolC-C3b interaction was fitted using a Hill curve, yielding a K_D of 570 ± 59 nM.

Fig 6

TolC protein of Leptospira acquires C3b and the catalytically active FH directly from NHS. (A) ELISA results demonstrate that the rTolC protein directly captures FH from NHS. Microtiter plates were coated with rTolC protein, and increasing concentrations of NHS (0.25%–20%) were added. Bound FH was detected using an anti-FH antibody. The rErpY-like protein served as the positive control, whereas the immobilized rTolC, neutralized with anti-TolC antibodies, was used as the negative control. (B). ELISA depicts that rTolC protein sequesters C3b directly from NHS. (C) Far-western blot shows rTolC acquires FH and C3b directly from NHS. The recombinant proteins (rTolC, rLoa22, and rErpY-like protein) were resolved on SDS-polyacrylamide gel and electroblotted onto the nitrocellulose membrane. The membrane was overlaid with 10% NHS, and anti-FH or anti-C3b antibodies were utilized to detect the interaction between rTolC and FH (top middle) or C3b (bottom middle). To demonstrate the equal loading of the recombinant proteins, Ponceau-stained blots are presented (left). The rErpY-like protein (right) was employed as a positive control for binding with FH. (D) ELISA results demonstrate that blocking the TolC protein on the leptospiral surface with anti-TolC antibodies reduces FH acquisition by leptospires. In this assay, leptospires, either treated or untreated with anti-TolC antibodies, anti-ErpY-like antibodies (positive control), or preimmune mouse serum (negative control), were incubated with diluted NHS (20% in PBS) and then immobilized on microtiter plate wells. The bound FH was subsequently detected using anti-FH antibodies. The bars represent mean ± SEM from three independent experiments performed with two technical replicates (*P < 0.05, ***P < 0.001). (E) Immunoblot showing FH bound to the immobilized rTolC retains its cofactor activity to cleave C3b. A microtiter plate was coated with either rTolC, rErpY-like protein (positive control), or rLoa22 (negative control) and then overlaid with FH. Unbound components were washed away. The mixtures of C3b + FI were added to the wells initially overlaid with FH. After 2 h of incubation at 37°C, the supernatant obtained was resolved on 12% SDS-polyacrylamide gel, transferred to the NC membrane, and immunoblotted with anti-C3b antibodies. The reactions omitting FH served as negative controls. The reactions (FH + FI + C3 b) performed in bovine serum albumin (BSA) blocked wells without prior coating with rTolC protein served as positive controls. The binding of catalytically active FH to rTolC protein was indirectly measured by immunoblotting with anti-C3b antibody. (F) Immunoblot showing immobilized rTolC protein blocked with anti-TolC antibodies cannot bind to FH. The rTolC was coated in microtiter plates, blocked with anti-TolC like antibodies, and overlaid with FH. C3b + FI were added to the FH overlaid wells. After 2 h of incubation at 37°C, the supernatants were subjected to immunoblotting. The C3b cleavage fragments (indicated by red and yellow triangles) were detected by immunoblotting with anti-C3b antibody. The protein standard molecular size marker (M) is indicated on the left side of each blot.

Fig 7

Immunoassay shows that the TolC protein disrupts the host complement system. Diluted NHS (20% in GVBMgEGTA, for AP, or 0.5% in GVB++, for CP) were preincubated with varying concentrations (0.5–2 µM) rTolC protein, rErpY-like protein (positive control), rClpP1 (negative control), or BSA (negative control). Thereafter, mixtures were added to plates coated with zymosan A (for AP), aggregated human IgG (for CP), or BSA (control). The MAC deposition (C5b9 complex) was measured (absorbance at 450 nm) using rabbit anti-C5b-9 antibodies. The absorbance values were normalized with the value obtained for BSA-coated wells. Heat-inactivated NHS (ΔNHS) was also included as a negative source of complement (control). For all assays, MAC deposition percentage was calculated considering the absorbance values of reactions with diluted NHS (in zymosan A or aggregated IgG-coated wells) as 100%. (A) Reduction in MAC deposition during AP activation in the presence of rTolC protein. The MAC deposition levels are represented in percentages. (B) rTolC protein supplementation did not result in any significant reduction (%) in MAC deposition during CP activation. The absorbance values obtained from the reactions where NHS was supplemented with the proteins (rTolC, rErpY-like, rClpP1, or BSA) were compared with those of unspiked diluted NHS (in zymosan A or aggregated IgG-coated wells). (C) Immunoassay showing C3b deposition during AP activation by rTolC protein. The activation of the AP or CP was indirectly measured by detecting the C3b deposition in the well surface using rabbit anti-C3b antibodies as described in the “Material and methods” section. (D) Immunoassay showing activation of CP by rTolC. The rErpY-like protein (positive control for AP and negative control for CP) and BSA (negative control) were included as controls. (E) Pathogenic L. interrogans blocked with anti-TolC show increased MAC deposition through the alternative pathway. The deposition of the MAC on the Leptospira surface was serologically detected using an antihuman C5b-9 antibody as described in the “Material and methods” section. The absorbance values obtained for leptospires preincubated with anti-TolC were compared with those of cells preincubated anti-ErpY-like (positive control), preimmune mouse serum (negative control), and untreated cells (negative control). The mean of absorbance values ± SEM obtained from three independent experiments performed in duplicate or triplicates are presented (*P < 0.05, ***P < 0.001).