Bacterial strains and growth conditions
Escherichia coli strains BL21 (DE3) Star and TOP10 (Invitrogen, São Paulo, SP, Brazil) were grown at 37°C in Lysogeny Broth (LB) medium, with 100 µg.ml−1 ampicillin when required. M. bovis BCG (Pasteur and recombinant strains) were cultivated at 37°C in Middlebrook 7H9 broth (Difco, BD, São Paulo, SP, Brazil) supplemented with 0.5% glycerol, 0.05% Tween 80 and 10% Oleic acid, Albumin, Dextrose Complex (OADC) or in 7H10 agar (Difco, BD, São Paulo, SP, Brazil) with 10% OADC. The BCG media was supplemented with kanamycin, 25 µg.ml−1, when needed. Low-passage virulent L. interrogans serogroup Icterohaemorrhagiae serovar (sv.) Copenhageni strain (st.) Fiocruz L1–130 (WDCM1012) was cultivated at 30°C in Ellinghausen–McCullough–Johnson–Harris (EMJH) medium supplemented with Leptospira EMJH enrichment (Difco, BD, São Paulo, SP, Brazil).
In silico structure predictions and epitope selection
Three proteins (LIC10896, LIC10964 and LIC12374) from L. interrogans sv. Copenhageni st. Fiocruz L1-130 (Taxid:267671), previously identified as leptospiral TonB-Dependent receptors, were selected for evaluation in this study. Three-dimensional (3D) structures were predicted by threading using the I-TASSER server (Yang and Zhang 2015). Models with the highest C-Score were selected for epitope prediction. The 3D protein structures were visualized using USCF Chimera 1.12 (Pettersen et al. 2004). Sequences for the CD4+ T cell epitopes, with high affinity to 14 different human major histocompatibility complex class II (MHC-II) alleles, were predicted using NetMHCII 2.2(Nielsen and Lund 2009) at the default settings. For linear (continuous) B cell epitope prediction, amino acid sequences were analyzed using BepiPred 1.0 (Larsen et al. 2006). Only T cell epitopes with a strong binding affinity (IC50 < 50 nM) and amino acid residues with a score higher than 0.35 in the B cell epitope analysis were considered. Epitopes were manually mapped onto the predicted 3D models of each βb-OMP in USCF Chimera 1.12. These models were aligned with their best structural template obtained from the Orientations of Proteins in Membranes (OPM) database (Lomize et al. 2012). All surface-exposed fragments containing B and T CD4+ cell epitopes were used for antigen construction. Orthologs for each L. interrogans TBDR were identified by BlastP (NCBI) in 10 pathogenic (P1 clade) Leptospira spp. proteomes (Online Resource 1, Table S1). Protein sequences with >70% identity and 40% query coverage were considered orthologs. Identity between the L. interrogans polypeptides and their orthologs was determined after multiple sequence alignments (MSA) using MUSCLE (EMBL-EBI) (Edgar et al. 2004).
Recombinant protein expression and purification
Selected surface-exposed fragments were assembled in silico using Vector NTI Advance® 11.5 (Invitrogen) without linkers, to construct the fusion proteins rTBDRchi, rLIC10896chi, rLIC10964chi and rLIC12374chi (Online Resource 2). E. coli codon optimized coding sequences (CDS) for each fusion protein were synthesized by Epoch Life Science (Texas, USA) and cloned into E. coli expression vector pAE (Ramos et al. 2004) using BamHI and HindIII restriction sites. Recombinant plasmids were used to transform E. coli BL21 (DE3) C41 strain (Invitrogen, São Paulo, SP, Brazil) by heat-shock. When cultures reached mid-log phase (0.6-0.8 absorbance at OD600), expression was induced by adding 0.5 mM IPTG (isopropyl-β-D-thiogalactopyranoside) and incubated at 37°C with agitation. After 4h, cells were harvested by centrifugation (7,000 × g, 15 min, 4°C). Pellets were washed in phosphate-buffered saline (PBS) and resuspended in lysis buffer (20 mM sodium phosphate, 0.5 M NaCl and 20 mM Imidazole, pH 8.0) before sonication with six 30-s pulses on ice. After lysis, insoluble proteins were harvested by centrifugation (10,000 × g, 1 h, 4˚C), and resuspended in denaturing binding buffer (8M urea, 20 mM sodium phosphate, 0.5 M NaCl, and 20 mM Imidazole, pH 8,0) and incubated under constant agitation for 16 h. After a final centrifugation (10,000 × g, 1 h, 4˚C), recombinant proteins were purified using the AKTA Start automated chromatography system (GE Healthcare). Briefly, the supernatant was applied to a nickel-charged HisTrap FF column (GE Healthcare, São Paulo, SP, Brazil), and after washing with 20 column volumes, recombinant fusion proteins were eluted over a total of 20 ml using a 0-100% gradient of elution buffer (8M urea, 20 mM sodium phosphate, 0.5 M NaCl, and 0.5 M Imidazole, pH 8,0). Purified proteins were then dialyzed against PBS at 4°C for 24 h and stored at -80˚C or 4°C.
Production of anti-rTBDRchi hyperimmune sera
Four-week-old female Wistar rats (Rattus norvegicus) were injected intraperitoneally with 50 µg rTBDRchi emulsified in complete Freund adjuvant for the first dose and incomplete Freund adjuvant for the remaining two doses. Two weeks after the second boost, animals were euthanized by exsanguination, and hyperimmune sera were obtained by centrifugation (3,500 × g, 15 min, 4°C) and stored at -20°C.
Construction of recombinant M. bovis BCG and expression analysis
Using pAE/rTBDRchi as template, the rTBDRchi CDS was amplified by PCR using primers described in Online Resource 1 (Table S2). The PCR product was digested using XbaI and PstI restriction enzymes (New England BioLabs) and cloned into the BCG expression vector pUP500/PpAN (Oliveira et al. 2019b), previously digested with Spel e PstI (New England BioLabs), generating pUP500/PpAN:rTBDRchi plasmid. Recombinant pUP500/PpAN:rTBDRchi was transformed into M. bovis BCG Pasteur cells by electroporation, as previously described (Parish and Stoker 1995). Cells were plated onto 7H10 agar (Difco, BD, São Paulo, SP, Brazil) containing 25 µg.ml−1 of kanamycin and incubated at 37°C for 21 days. Resistant colonies were inoculated in selective Middlebrook 7H9 broth (Difco) and cultivated at 37ºC to evaluate recombinant expression by immunoblot. Briefly, whole-cell lysates were prepared from rBCG:TBDRchi grown to mid-logarithmic phase at 37°C (OD600 = 0.8), when cells were collected by centrifugation (4,000 × g, 15 min) and resuspended in Tris-HCl (pH 8), adjusting the cell density to 107 CFU.ml−1. rBCG:TBDRchi whole-cell lysates were separated in 12% polyacrylamide gels and transferred to nitrocellulose membranes (Hybond ECL, GE Healthcare, Illinois, United States). Membranes were incubated overnight with rat anti-rTBDRchi sera at 1:100 dilution, followed by anti-rat HRP-conjugated secondary antibodies (Sigma Aldrich, Missouri, United States) diluted 1:5,000. Immunoblots were developed using Pierce ECL Western Blotting Substrate (Thermo-Fisher, Illinois, USA).
Vaccination and challenge of hamsters
rBCG:TBDRchi and BCG Pasteur (negative control) strains were cultivated in vitro and collected by centrifugation (4,000 × g, 15 min). Cells were resuspended in PBS, adjusting concentration to 107 cells/ml (considering 1 OD600 = 1 x 108 cells/ml), and immediately used. Male Syrian hamsters (Mesocricetus auratus), 4–6 weeks old, were subcutaneously inoculated with 106 cells of BCG Pasteur (n=13 per experiment) or recombinant BCG:TBDRchi (n=10 per experiment). An additional control group (n=4 per experiment) was immunized (intramuscular) with a leptospiral whole-cell bacterin (108 heat-inactivated leptospires in 100 µl PBS). Two doses of each formulation were administered 21 days apart. Thirty days after the second boost, hamsters were infected intraperitoneally with 103 leptospires (L. interrogans sv. Copenhageni st. Fiocruz L1-130), equivalent to five times the average 50% endpoint dose (ED50), as previously determined (Oliveira et al. 2019a). Hamsters were monitored daily for clinical signs of acute leptospirosis for 30 days after challenge. Criteria for humane euthanasia (CO2 narcosis) was established as previously described (Coutinho et al. 2011), and includes: >10% weight loss, prostration, bristling, apathy, and lack of appetite. Blood samples were collected one day before the first immunization (day zero) and before challenge (day 51). A total of two independent experiments were performed. All animal procedures were conducted according to the rules and regulations of the Animal Experimentation Ethics Committee at UFPel.
Analysis of the leptospiral burden in hamster kidneys post-challenge
At the end of the experiment, kidneys were collected aseptically in order to evaluate the leptospiral burden post-challenge. Kidney samples were macerated in 10 ml EMJH medium supplemented with 10% Leptospira enrichment supplement. After one hour at 30°C, 500 µl of the macerate was inoculated into 10 ml of fresh EMJH and incubated at 30ºC for up to 8 weeks. Cultures were monitored weekly by dark-field microscopy. DNA extraction from kidney samples, followed by qPCR was performed as previously described (Dorneles et al. 2020). Briefly, DNA was extracted from approximately 40 mg of kidney tissue using SV Genomic DNA Purification Kit (Promega, Brazil) and quantified using Qubit 2.0 fluorometer (Thermo Fisher). Quantitative real-time PCR (qPCR) detection of leptospiral DNA was performed in a LightCycler 96 system (Roche, Basel, Switzerland), using 200 ng of total DNA, 0.4 µM of each primer specific to lipL32 (Online Resource 1, Table S2) and SYBR Green PCR Master Mix (Applied Biosystems, São Paulo, SP, Brazil). Reactions were performed in triplicate as previously described (Dorneles et al. 2020). A standard curve was generated from 106 to 10 copies of lipL32 and samples were considered negative when their qPCR quantitation cycle (Cq) were higher than that obtained for the lowest concentration of 10 copies. Absolute quantification analysis using LightCycler 96 software (Roche) was performed to determine the number of leptospire genome equivalents per reaction, and thereafter converted to copies per µg of total DNA.
Evaluation of the humoral immune response
Humoral immune response elicited after vaccination was evaluated by indirect ELISA (Enzyme-Linked Immunosorbent Assay), as previously described (da Cunha et al. 2019). First, 96 well microtiter plates (Polysorp, Nunc, São Paulo, Brazil) were coated with 1 µg of purified rTBDRchi per well at 4ºC overnight, in 0.1 M carbonate-bicarbonate buffer, pH 9.6. Plates were washed three times with PBS-0.05% Tween 20 (PBST) and then blocked with 200 µl of 5% non-fat dry milk in PBS-T, for 1 h at 37ºC. Hamsters’ sera were diluted 1:50 in PBST and evaluated in triplicate. After incubation for 1h at 37ºC, plates were washed and HRP-conjugated anti-Syrian hamster IgG (Sigma Aldrich, Missouri, United States) was added at 1:5,000 dilution for 1 h at 37°C. After a final wash, reactions were developed using o-phenylenediamine dihydrochloride (Sigma-Aldrich, Missouri, United States) and hydrogen peroxide, and stopped after 15 min with 4N H2SO4. Absorbance was read at 492 nm wavelength in a microplate reader (Mindray MR-96A, São Paulo, Brazil).
Immunoblots were performed as previously described, with small changes (Groshong et al. 2021). Briefly, samples containing 1 µg of each recombinant protein (rTBDRchi, rLIC10896, rLIC10964, and rLIC12374) or L. interrogans st. Fiocruz L1-130 whole-cell lysate (108/lane), were analyzed by SDS-PAGE and transferred to nitrocellulose membranes (GE Healthcare, Illinois, United States) at 20 V for 25 min using a semi-dry transfer (Bio-Rad). Membranes were blocked for 1 h at room temperature with 5% non-fat dried milk diluted in PBS-T, and probed overnight at 4°C with a pool of hamster sera obtained after immunization, at a 1:200 dilution in blocking buffer. After washing 5 times with PBS-T, membranes were incubated for 1 h at RT with an HRP-conjugated goat anti-hamster IgG antibody (Sigma Aldrich, Missouri, United States) diluted in blocking buffer (1:6,000). Following six washes with PBS-T, immunoblots were developed using the Pierce ECL Western Blotting Substrate (Thermo-Fisher, Illinois, USA) detection reagent, and results were visualized using C-Digit Blot Scanner (LI-COR, Nebraska, USA).
Statistical analysis
Significant protection against lethal leptospirosis was evaluated using Fisher’s exact test (two-tailed). Comparison of humoral immune response elicited by different experimental groups was performed using analysis of variance (one way-ANOVA with Tukey multiple comparison). For all analyses, P-values < 0.05 were considered significant. Graphical representations and statistical analysis were performed using GraphPad Prism 8.