Chlamydiatrachomatis serovar L2 434/Bu (Ct) was kindly provided and characterized by the Unidad de Estudios de Clamidias, UBA, Buenos Aires, Argentina. Serovar D was generously contributed by the Instituto Malbran, Buenos Aires, Argentina. Additionally, a fluorescent Ct L2 strain with green fluorescence, containing the p2TK2-SW2IncDProm-RSGFP-IncDTerm serovar L2 construct (GFP-Ct), was kindly supplied by Dr. Derré (26). Ct bacteria were cultivated within HeLa cells (ABAC, Argentina) using high glucose Dulbecco’s Modified Eagle’s Medium (GIBCO, Thermo Fisher Scientific, USA) supplemented with 10% (v/v) fetal bovine serum (FBS) sourced from Internegocios SA, Argentina. The growth medium was further enriched with 1.55 mg/mL glucose (Biopack, Argentina) and 0.3 mg/mL L-glutamine (ICN Biomedicals Inc, USA), and antibiotics were omitted. Cultures were maintained in a 5% CO2 environment at 37°C. Chlamydial elementary bodies (EBs) were harvested, purified, and quantified following the methodology previously outlined by Del Balzo et al. (27).

The design, cloning, and purification of plasmid construction pVAX1−FPmpD for DNA immunization and the production of recombinant PmpD protein fragment (rFPmpD) for boosters were previously described (25). Cationic liposomes (Lp) were prepared by the ethanol injection method using 4 mM as the final lipid concentration in 7:2:2 mol/mol ratio of DPPC: Chol:SA. The ODN-CpG (CpG) with the sequence: 5’-tcgtcgtttgtcgttttgtcgtt-3’ was obtained by chemical synthesis (Invitrogen, USA), with phosphodiester bonds (17). The gemini lipopeptide AG2-C16 (Gem) was synthesized using 9-fluorenyl-methoxycarbonyl (Fmoc) chemistry in the solid phase, according to Grippo et al., 2019 (22). The purified DNA was resuspended in sterile sodium acetate buffer (0.05 M; pH 5.3) at 2.5 μg/μL. The protein formulations were prepared alternatively as: free rFPmpD, rFPmpD+Lp+CpG, rFPmpD+Lp+Gem, or rFPmpD+Lp, in sodium acetate buffer (0.05M; pH 5.3). Final concentrations were: 7.5 nmol/mL CpG, 400 μM Gem, 0,125 μg/μL (intranasal route) and 0,05 μg/μL (subcutaneous route) rPmpD.

Female BALB/c mice (Icivet Litoral, Argentina) and female C57BL/6 mice (The Jackson Laboratory, USA), aged six to eight weeks, were housed in a controlled environment that met specific pathogen-free (SPF) criteria. The housing conditions included a humidity level of 50-60%, a temperature range of 20-24°C, 60 air exchanges per hour within the cages, and a 12/12 hour light/dark cycle with lights turning on at 8 am. The mice were provided with ad libitum access to drinking water and a standardized mouse diet. All experimental procedures were conducted in accordance with the guidelines outlined in the Guide for the Care and Use of Laboratory Animals (ILAR 2010). Animal protocols were granted approval by both the Institutional Advisory Committee on Research Ethics and Security (FBCB-UNL, Argentina) and the Institutional Committee on Laboratory Animal Care and Use (CICUAL, UNCUYO, Argentina). This study adheres to the standards set by the ARRIVE guidelines (https://arriveguidelines.org).

Animals were randomly assigned to one of these groups: FPmpD, FPmpD-Lp-CpG or FPmpD-Lip-Gem (BALB/c and C57BL/6; n=6 and n=5 per group, respectively). A DNA prime-protein boost strategy was applied, with a total of three doses, three weeks apart. Mice from the three groups received an ear-pinnae intradermal first dose of 50 μg FPmpD-pVAX1. Two protein boosts were administered by the intranasal (2.5 μg) and subcutaneous (5 μg) routes, simultaneously. For the intranasal immunization free rFPmpD was administered to the FPmpD group, and rFPmpD+Lp was used for the groups FPmpD-Lp-CpG and FPmpD-Lp-Gem. For the subcutaneous injections, free rFPmpD, rFPmpD+Lp+CpG or rFPmpD+Lp+Gem were administered. The adjuvant control groups (BALB/c and C57BL/6; n=5 and n = 3, respectively) received empty plasmid (pVAX1) in the first dose followed by two doses of the adjuvants, without protein (Lp-Gem or Lp-CpG). Two additional control groups (BALB/c and C57BL/6; n = 5 and n = 4, respectively) were inoculated with pVAX1 in the first dose, and acetate buffer in the second and third doses (Control). Blood was collected from the submandibular vein seven days (d) before immunization and ten days after each inoculation and sera were obtained and stored at − 20°C until use. Vaginal fluid (VF) was collected by vaginal washing using PBS ten days after the last dose, with the addition of protease inhibitors cocktail Set I (Merck, Germany). The collected samples were stored at − 80°C until use. Upon sacrifice, which occurred 30 d post last dose, mice were subjected to cardiac puncture for blood collection, under anesthesia. Anesthesia was induced using 100 mg/kg Ketamine (Hollyday-Scott SA, Argentina) and 10 mg/kg Xylazine (König SA, Argentina). The sera were stored at -20°C until use. Timelines that represent experimental protocols and the experimental groups with the vaccinal formulations administered are displayed in Figure 1.

Specific antibodies against rFPmpD was measured by an indirect enzyme-linked immunosorbent assay (ELISA) both in sera and VF samples. In brief, 96-well polystyrene plates (Microlon™ 600, Greiner Bio-One, Austria) were coated overnight at 4°C with rFPmpD (1 μg/100μL) diluted in carbonate buffer. Following this, non-specific binding sites were blocked using PBS supplemented with 5% skimmed milk. Sera were diluted 1:100, and VF was diluted 1:3 in PBS-1%-skimmed milk. The plates were incubated for 1 hour at 37°C with the respective peroxidase-conjugated anti-mouse antibodies: anti-mouse IgG (Jackson, USA), anti-mouse IgG1 (Santa Cruz Biotechnology, Inc., USA), anti- mouse IgG2a (Abcam Inc., USA), anti-mouse IgG2c (Abcam Inc., USA), or anti-mouse IgA (Sigma-Aldrich, USA), as appropriate. Plates were washed 3 times after each step using PBS-0.05% Tween 20. Finally, the plates were incubated with the substrate (hydrogen peroxide) and the chromogen tetramethylbenzidine (Thermo Scientific, USA). The reaction was stopped by adding 0.5 M H₂SO₄. The Optical Density (OD) was read at 450 nm using a microplate reader (Mulstiskan EX, Labsystems, USA). The obtained results are presented in terms of OD. The antibody levels in serum samples collected 30 d post-last inoculation from FPmpD-vaccinated (FPmpD, FPmpD-Lp-Gem and FPmpD-Lp-CpG), and non-vaccinated mice (Control, Lp-CpG, Lp-Gem) were quantified by titration. The cut-off value was determined by calculating the mean OD of the control group and adding three times the standard deviation.

Female BALB/c mice from all experimental groups were mated with untreated male mice aged 14–16 weeks with proven fertility. Each cage contained three female mice paired with a single male mouse. Mating took place after a ten-day period following the last administered dose. The presence of vaginal plugs or spermatozoa in the vaginal secretions was monitored daily, and the day of detection marked the beginning of pregnancy (designated as day 0). The female mice were then separated and kept until day 18 of pregnancy, when they were sacrificed. Uteri and ovaries were collected, and various reproductive parameters were recorded, including the counts of corpora lutea, implantations, resorptions, as well as both live and dead fetuses. Subsequently, the following parameters were calculated: Fertility potential (efficiency of implantation): Calculated as (number of implantation sites/number of corpora lutea) × 100. Rate of preimplantation loss: Calculated as [(number of corpora lutea - number of implantations)/number of corpora lutea] × 100. Rate of postimplantation loss: Calculated as [(number of implantations - number of live fetuses)/number of implantations] × 100, following the methodology described by Perobelli et al. (28).

The intravaginal infection mouse model was performed according to Lujan et al. (6). On day 16 after the last dose, C57BL/6 mice of all the experimental groups were intravaginally challenged using 1.5 × 10⁵ GFP-Ct serovar L2 EBs (6). Ct serovar L2 was employed as described in Shaw et al. (29, 30). To synchronize the estrous cycle of the C57BL/6, a single subcutaneous injection of 2.5 mg of medroxyprogesterone acetate (Holliday-Scott SA, Argentina) were administered seven d prior to the GFP-Ct infection. Cervicovaginal swabs were collected at seven- and 14 d post-challenge from each experimental group using disposable microapplicators (Multi-Brush, Denbur Inc, USA). These swabs were then cultured on a monolayer of HeLa cells. The quantification of infectious bacterial progeny present in the cervicovaginal secretions was performed by assessing Inclusion Forming Units (IFU) using confocal microscopy and flow cytometry, according to Vromman et al. (7). The reduction in IFU upon vaccination was considered protection. Protection was examined in swab samples on day seven based on previous data (25, 31, 32).

On the 14th day after infection (14 d p.i.), the mice were euthanized, and their uterine horns were dissected for both morphological and histological examination. The length of the uterine horns was quantified. Subsequently, one uterine horn from each mouse was fixed in a solution of 4% paraformaldehyde and then subjected to embedding in paraffin using standard protocols. Thin tissue slices, 5 μm in thickness, were prepared and stained using hematoxylin and eosin. The stained slices were observed under a Leica DMi8 microscope, which was equipped with an MShot color camera, facilitating detailed histological analysis. The resulting images were further processed using M-Shot analysis system software v 1.1.6 (Micro-Shot, China).

The uterine sections (5 μm thick) were immunostained for the detection of macrophages and the expression of TNFα. At 4°C, the samples were overnight incubated with primary antibodies: anti-CD68, 1:200 (Invitrogen, USA); anti-TNFα, 1:50 (Invitrogen, USA). Then, the sections were incubated 1 h at room temperature with biotinylated anti-mouse (Vectastain Elite ABC universal kit, Vector Laboratories, USA) and 1:5.000 biotinylated anti-rat (Vector Laboratories, USA), respectively. Biotinylated antibodies were detected with Vectastain Elite ABC universal Kit, peroxidase (Vector Laboratories, USA), and finally, the reactions were developed using an ImmPACT DAB Peroxidase Substrate (Vector Laboratories, USA).

Total RNA was isolated from a single uterine horn using TRIzol® Reagent (Invitrogen, USA). The extracted RNA was quantified by measuring OD at 260 nm employing a NanoDrop Lite spectrophotometer (Thermo Scientific, USA). Subsequently, one microgram of total RNA from each sample was retrotranscribed to cDNA using MMLV reverse transcriptase (Promega, USA). Quantitative Real-Time PCR was employed to determine the relative expression levels of TNFα, IL-1, IL-6, IFNγ, IL-10 mRNA. Specific primers were used for each target gene (as outlined in Table 1). Each qPCR was performed using HOT FIREPol® EvaGreen® qPCR Mix Plus (Solis Biodyne, Estonia) and a primer concentration of 0.5 μM for each primer. PCR amplifications were performed on a Corbett Rotor Gene 6000 Real-Time Thermocycler (Corbett Research Pty Ltd Sydney, Australia). The amplification protocol initiated with a denaturation step (12 min at 95°C), followed by 40 cycles of denaturation (20 sec at 95°C), annealing (30 sec at the temperature specified for each primer pair in Table 1), and extension (20 sec at 72°C). To ensure specificity, a melting curve analysis was routinely conducted for each reaction. The relative gene expression was calculated using the 2-DDCT method p and GAPDH as a housekeeping gene for normalization of the relative mRNA levels.

The results are presented as the mean ± standard error of the mean (SEM). The normality of the distribution was evaluated using the Kolmogorov-Smirnov test. Statistical differences among groups were assessed through appropriate methods, including two-tailed t-test or one-way and two-way ANOVA or Kruskal-Wallis, with multiple comparisons, using GraphPad Prism 8.0.1 software (GraphPad, USA). Statistical significance was determined with a threshold of p values less than 0.05.

The systemic humoral immune response was analyzed by measuring the anti-PmpDIgG, IgG1, and IgG2a/c serum levels before immunization (pre-immune), and ten days after each dose in BALB/c (Figures 2A, C) and C57BL/6 (Figures 2B, D) mice as indicated in Methods (Figure 1). The prime-boost regimen elicited demonstrable increments of anti-PmpD antibodies with the successive doses that became significant compared to the control after the second dose for IgG and IgG1 in BALB/c mice vaccinated with FPmpD (p<0.05), for IgG in the three groups of vaccinated C57BL/6 mice, and for IgG2c in the FPmpD-Lp-Gem and FPmpD-Lp-CpG groups of C57BL/6 mice vaccinated with adjuvants. After the third dose, all the vaccine formulations induced a significant further increase in the antibody levels, except for IgG1 in the FPmpD-Lp-Gem group of C57BL/6 mice (Figure 2). Specific anti-PmpD IgG, IgG1, or IgG2a/c antibodies were absent in the pre-immune sera of the experimental groups, as well as in the groups that received adjuvants without protein (Lp-Gem and Lp-CpG) during the entire trial period (Table 2A). The titers of IgG and the subclasses (IgG1 and IgG2a/c) anti-PmpD antibodies ten days after the last dose in the groups that received free rFPmpD or rFPmpD with adjuvants were between 10⁴ and 10⁵ dils^-1 in animals of both strains (Figures 2C, D). FPmpD-Lp-CpG elicited higher IgG2 titers than FPmpD-Lp-Gem in both strains. However, none of the adjuvant groups showed a significant difference compared to the FPmpD group. We then proceeded to examine the efficacy of the FPmpD formulations in enhancing humoral immunity within the mucosal environment of the genital tract. To assess this, we quantified anti-PmpD IgA and IgG antibodies in vaginal wash samples collected ten days following the last immunization. We observed significantly higher levels of IgG in vaginal washes from mice vaccinated with free FPmpD (FPmpD; p<0.05) and with FPmpD-Lp-CpG (p<0.01) in BALB/c (Figure 3A) and C57BL/6 (Figure 3C) mice, compared to control ones. While the FPmpD formulations, both with and without adjuvants, did induce specific anti-PmpD IgA antibodies in the mucosal genital tract of certain BALB/c mice, the mean IgA levels observed did not exhibit significant differences when compared to the non-vaccinated control mice (as shown in Figures 3B, D). Pre-immune vaginal washes and the Lp-Gem and Lp-CpG groups displayed no presence of specific IgG or IgA throughout the duration of the study (as outlined in Table 2B). These results collectively indicate that FPmpD elicits a robust systemic and mucosal IgG humoral immune response, while the inclusion of any of the tested adjuvants does not seem to enhance this response.

We conducted an assessment of the potential impact of vaccination on mice fertility considering that an intensified immune response in the mucosal region might negatively influence the functioning of reproductive organs (33). The fertility potential, evaluated by implantation efficiency, remained unaltered in vaccinated mice when compared to the non-vaccinated control group (Kruskal-Wallis test; p > 0.05; Table 3). Furthermore, we computed the pre-implantation loss rate by counting the number of corpora lutea and implantations, and the post-implantation loss rate by considering the number of implantations and live fetuses. Our analysis revealed no notable distinctions in either the pre-implantation or post-implantation loss rates between the vaccinated and non-vaccinated mice. This suggests that the FPmpD prime-boost strategy did not hinder fertility (Kruskal-Wallis test; p > 0.05; Table 3). Moreover, there were no significant differences in the fertility parameters between any of the adjuvant control groups (Lp-Gem, Lp-CpG) and the non-vaccinated mice (Kruskal-Wallis test; p>0.05). In summary, the scheme of immunization with the vaccine formulations assayed in this study does not alter fertility.

In order to assess the protective efficacy of the FPmpD vaccine formulations against the dissemination of Ct, intravaginal challenge was administered to mice using 1.5x10⁵ inclusion forming units (IFU) of GFP-Ct. Cervicovaginal swabs were collected seven and 14 d.p.i, and the number of IFU was quantified. The reduction in IFU upon vaccination was indicative of protection. The samples obtained from the non-vaccinated mice that were challenged with Ct (Control+I) displayed a substantial presence of numerous and large chlamydial inclusions following cultivation on HeLa cells (Figure 4A, left panel). In contrast, some samples derived from the vaccinated mice generated some inclusions of smaller size, while others barely displayed inclusions when placed on HeLa cells (Figure 4A). There was a remarkablereduction in infectivity of the vaginal washes (expressed as Log₁₀ of IFU, mean ± SEM) of the FPmpD+I group (55912 ± 32898) compared to the non-vaccinated mice (Control+I) (169760 ± 27603) (67% reduction) (Figure 4C). In contrast, the vaccinated groups with adjuvants only displayed slightly lower bacterial burden (FPmpD-Lp-Gem: 150384 ± 21673 and FPmpD-Lp-CpG:106417 ± 43352) than in the non-vaccinated group (Control+I) (11% and 37% reduction, respectively). On day 14 p.i., the IFUs shedding was undetectable even in the non-vaccinated infected mice, suggesting a natural resolution of the infection (data not shown). Therefore, among the three vaccine formulations evaluated, the FPmpD vaccine without adjuvants showed the best performance in increasing chlamydial clearance and reducing the bacterial shedding, thus achieving better control of the Ct infection.

One goal of vaccination is to elicit a durable immune response against the antigen through the induction of memory cells. In this context, we examined the capability of FPmpD-based vaccine formulations to sustain the humoral response induced after the Ct infection. In vaccinated mice, although a minor decline in antibody levels was evident between ten days and 30 d post 3rd dose, specific IgG were detectable and remained at significantly high levels in the FPmpD-Lp-CpG group (Figure 5A). The three formulations induced sustained levels of IgG1, and IgG2a that continued significantly increased in the groups vaccinated with FPmpD and with FPmpD-Lp-CpG (Figure 5A). The group FPmpD-Lp-Gem exhibited only significantly high levels of IgG1 at 30 d post 3rd dose compared to control, non-vaccinated uninfected mice (Figure 5A). Subsequently, we assessed the production of anti-PmpD antibodies in the group of animals that underwent the challenge. Among FPmpD-vaccinated and infected mice (FPmpD+I; FPmpD-Lp-Gem+I; FPmpD-Lp-CpG+I), the concentrations of IgG, IgG1, and IgG2c antibodies 14 days subsequent to intravaginal Ct infection displayed similarity to those measured ten days after the third dose administration (Figure 5B). Conversely, the initial infection alone did not result in notable levels of anti-PmpD antibodies in the non-vaccinated infected group (Control+I), which exhibited similar antibody levels to unvaccinated uninfected mice. Consequently, vaccination with or without adjuvants induced a rapid and potent humoral immune response subsequent to the Ct challenge, with high levels of IgG2c likely suggesting stimulation of Th1 cells. Notably, this response was substantially more elevated than the immune response initiated by the primary infection alone (Figure 5B). These results show that the FPmpD formulations assayed in this study are able to elicit specific antibodies, lasting at least 30 d after the last dose. Moreover, no differences were observed among formulations, suggesting that the prime-boost strategy with free rFPmpD does not need adjuvants or immunostimulants to induce a strong humoral immune response.

Considering the importance of inducing a balanced local response against infection, we studied the impact of chlamydial infection on the genital tract of vaccinated mice. As described in Materials and Methods, the genital tract of each mouse was dissected 14 d following the Ct challenge, enabling the evaluation of morphological and histological aspects. Representative images of uteri from the experimental groups are presented in Figure 6A. A recognized morphological alteration stemming from Ct infection is the reduction in the length of the uterine horns (34). In line with this, the infection with Ct led to a substantial shortening of uterine horns in the non-vaccinated mice (Control+I) in contrast to non-vaccinated and uninfected mice (Control) or the vaccinated infected FPmpD+I and FPmpD-Lp-Gem+I mice (Figure 6B). As expected, we detected the reduction in the length of uterine horns of Lp-Gem+I in comparison toFPmpD-Lp-Gem+I mice, attributing the protective effect to the recombinant antigen FPmpD. Surprisingly, we observed a shortening of uterine horns of mice from both the FPmpD-Lp-CpG+I and Lp-CpG groups, suggesting that the CpG-containing adjuvant induced by itself a shortening of uterine horns (Figure 6B). As it is shown in the representative images, Ct infection produces a series of changes in the tissue architecture (Figure 6C). The uteri of Control+I, Lp-Gem+I, Lp-CpG+I, FPmpD-Lp-Gem+I and FPmpD-Lp-CpG+I mice exhibited edema and tissue structure loss, while vaccination without adjuvants (FPmpD+I) mitigated the tissue damage caused by the Ct infection. Then, our results suggest that the formulations containing adjuvants (FPmpD-Lp-Gem+I, FPmpD-Lp-CpG+I) were not able to prevent the morphological and histopathological changes in the genital tract linked to Ct infection whereas the FPmpD-based vaccine without adjuvants efficiently reduced the chlamydial infection development in the uterine tissue and congruently decreased Ct-induced damage of the genital tract.

Ct infection is known to produce inflammatory damage to the genital tract (33). Therefore, we evaluated the post-challenge inflammatory response. Given their prominent role as immune response modulators, relative expression of TNFα, IL-1β, IL-6, INF-γ, and IL-10 were evaluated in the uterine tissue 14 d post-Ct challenge. Figure 7A shows the TNFα mRNA relative expression (RE) in mice vaccinated with the adjuvanted formulations (FPmpD-Lp-Gem+I RE:6.99 and FPmpD-Lp-CpG+I RE: 7.11), displaying significantly higher values than in mice vaccinated with free FPmpD (FPmpD+I RE: 0.60) (**p<0.01) and unvaccinated infected mice (Control+I RE: 1.23) (*p<0.05). Regarding the other cytokines evaluated, a decrease in RE was observed in mice vaccinated with free FPmpD (FPmpD+I) compared to the other groups (FPmpD-Lp-Gem+I, FPmpD-Lp-CpG+I, Lp-Gem and Lp-CpG), mainly with Control+ I, although not statistically significant. These results suggest that the free FPmpD-vaccinated group (FPmpD+I) would beneficially regulate the levels of cytokines 14 d post-Ct-infection. Macrophages are major inducers of the proinflammatory cytokine TNFα which, in turn, has been proposed to underpin the immunopathological response triggered by Chlamydia. Therefore, immunohistochemical staining for TNFα and macrophages (CD68 marker) in horn sections from mice on day 14 post-Ct infection was evaluated. We detected increased TNFα protein expression mainly in the groups exposed to the adjuvants (FPmpD-Lp-Gem+I and FPmpD-Lp-CpG+I, Lp-Gem, and Lp-CpG) in both vaccinated and unvaccinated mice (Figure 7B). It is noteworthy that the main localization of TNFα protein is observed in the epithelial cells of the single-layer epithelium and the glands. We detected higher levels of macrophages infiltrating the endometrial stroma by CD68 staining in mice immunized with the adjuvanted formulations (FPmpD-Lp-Gem+I and FPmpD-Lp-CpG+I) compared to the FPmpD-vaccinated mice (FPmpD+I) (Figure 7B). Noticeably, FPmpD+I group presents a minor proportion of CD68 labeling compared to the rest of the experimental groups (Figure 7C). Mice exposed to adjuvants alone (Lp-Gem+I and Lp-CpG+I) displayed greater number of CD68 labeled cells than mice challenged with Ct (Control+I) (Figure 7C). These results are in agreement with the levels of TNFα mRNA (Figure 7A), allowing us to infer a relationship between a macrophage/TNFα response and the histopathological damage previously observed in the same experimental groups (Figure 6).