Paeniclostridium sordellii uterine infection is dependent on the estrous cycle

Abstract

Human infections caused by the toxin-producing, anaerobic and spore-forming bacterium Paeniclostridium sordellii are associated with a treatment-refractory toxic shock syndrome (TSS). Reproductive-age women are at increased risk for P. sordellii infection (PSI) because this organism can cause intrauterine infection following childbirth, stillbirth, or abortion. PSI-induced TSS in this setting is nearly 100% fatal, and there are no effective treatments. TcsL, or lethal toxin, is the primary virulence factor in PSI and shares 70% sequence identity with Clostridioides difficile toxin B (TcdB). We therefore reasoned that a neutralizing monoclonal antibody (mAB) against TcdB might also provide protection against TcsL and PSI. We characterized two anti-TcdB mABs: PA41, which binds and prevents translocation of the TcdB glucosyltransferase domain into the cell, and CDB1, a biosimilar of bezlotoxumab, which prevents TcdB binding to a cell surface receptor. Both mABs could neutralize the cytotoxic activity of recombinant TcsL on Vero cells. To determine the efficacy of PA41 and CDB1 in vivo, we developed a transcervical inoculation method for modeling uterine PSI in mice. In the process, we discovered that the stage of the mouse reproductive cycle was a key variable in establishing symptoms of disease. By synchronizing the mice in diestrus with progesterone prior to transcervical inoculation with TcsL or vegetative P. sordellii, we observed highly reproducible intoxication and infection dynamics. PA41 showed efficacy in protecting against toxin in our transcervical in vivo model, but CDB1 did not. Furthermore, PA41 could provide protection following P. sordellii bacterial and spore infections, suggesting a path for further optimization and clinical translation in the effort to advance treatment options for PSI infection.

Fig 1. Neutralization of TcsL cytotoxicity by C. difficile monoclonal antibodies, PA41 and CDB1, in vitro.

(A-B) Vero cells were treated with serial dilutions of TcsL alone or in the presence of 100 nM PA41 or CDB1. ATP was measured as a readout of viability and normalized to signal from untreated cells. Dunnett’s test for multiple comparisons was used with statistical significance set at a p value of <0.05, where **** represents p <0.0001. (C-D) Vero cells were treated with a serial dilution of PA41 or CDB1 in the presence of 1 pM TcsL (IC₅₀ = 20 pM and ~41 nM, respectively). The baseline for 1 pM TcsL cytotoxicity is indicated by the dotted line.

Fig 2. PA41 and CDB1 neutralization of rTcsL and P. sordellii vegetative bacteria, in vivo, following intraperitoneal injection.

(A) Mouse survival curve following IP injection of 1ng and 2.5ng TcsL. (B) Mouse survival curve following IP intoxication of 2.5ng TcsL alone or in the presence of PA41 (0.75 mg/kg) or CDB1 (0.75 mg/kg). (C) Mouse survival curve following IP infections of 10⁶, 10⁷, 10⁸ CFUs of vegetative P. sordellii strain ATCC 9714. (D) Mouse survival curve following 7.5 mg/kg PA41 or PA50 (control antibody) administered 18h prior to co-IP instillation of 7.5 mg/kg antibody and 1x10⁷ CFUs of ATCC 9714 vegetative bacteria. Log-rank (Mantel-Cox) multiple comparison test was used with statistical significance set at a p value of <0.05.

Fig 3. Transcervical instillation method of recombinant TcsL or vegetative P. sordellii.

(A) Schematic depicting the murine transcervical instillation method. (B) Transcervical intoxication of 5, 25, 50ng rTcsL and intraperitoneal intoxication of 2.5ng rTcsL. (C) Transcervical infection of 10⁷ and 10⁸ CFUs and intraperitoneal infection of 10⁷ CFUs of vegetative ATCC 9714 P. sordellii.

Fig 4. Development of hormonal transcervical instillation method of recombinant TcsL or vegetative P. sordellii.

(A) Timeline of hormonal synchronization of murine estrous cycles following subcutaneous administration of medroxyprogesterone acetate (Day -5) to induce diestrus or beta-estradiol (Day -2) to induce estrus. Transcervical Instillation of rTcsL or P. sordellii was performed on Day 0 and animals were weighed/monitored for six to eight days. (B) Survival curve of mice subcutaneously injected with medroxyprogesterone acetate or beta-estradiol followed by transcervical intoxication with 50ng TcsL. (C) Survival curve of mice subcutaneously injected with medroxyprogesterone acetate followed by transcervical intoxication with 5, 10, 20, 50, 500ng TcsL. (D) Survival curve of mice subcutaneously injected with medroxyprogesterone acetate or beta-estradiol followed by transcervical infection with 10⁷ CFUs vegetative P. sordellii ATCC9714. (E) Survival curve of mice subcutaneously injected with medroxyprogesterone acetate followed by transcervical infection with 10², 10⁴, 10⁵,10⁶, 10⁷ CFUs vegetative P. sordellii ATCC9714. Log-rank (Mantel-Cox) multiple comparison test was used with statistical significance set at a p value of <0.05.

Fig 5. Neutralization studies of monoclonal antibodies, PA41 and CDB1, following transcervical intoxication of rTcsL.

(A) Timeline of neutralization studies: subcutaneous administration of medroxyprogesterone acetate on Day -5 to induce diestrus. Intraperitoneal injections of 7.5 mg/kg PA41, 7.5 mg/kg CDB1, or PBS on Days -5, -3, -1. Transcervical intoxication of rTcsL on Day 0. Animals weighed/monitored for seven days. (B) Survival curve of animals in diestrus treated with PA41 or CDB1 on days -5, -3, -1, and transcervically intoxicated with 10ng TcsL on Day 0. (C) Survival curve of animals in diestrus treated with PA41, on days -5, -3, -1, and transcervically intoxicated with 50ng TcsL on Day 0. Log-rank (Mantel-Cox) multiple comparison test was used with statistical significance set at a p value of <0.05. (D) Histological scoring of edema, acute inflammation, and epithelial injury of uterine tissues at time of death or end of study from mice transcervically instilled with 50ng TcsL in the presence or absence of PA41 and 10ng TcsL in the presence of absence of CDB1. Mann-Whitney test was used with statistical significance set at a p value of <0.05. (E) White blood cells count following Complete Blood Counts. Neutrophil (NE), lymphocyte (LY) and hematocrit (HCT) blood cell percentages of (F) TcsL/PA41 and (G) TcsL/CDB1. Kruskal-Wallis multiple comparison test was used with statistical significance set at a p value of <0.05.

Fig 6. Prophylactic and therapeutic administration of PA41 following transcervical P. sordellii vegetative bacterial or spore infection, respectively.

(A) Timeline showing subcutaneous administration of medroxyprogesterone acetate on Day -5 to induce diestrus, intraperitoneal injection of 7.5 mg/kg PA41 or PBS on Day -1, and transcervical infection of 10⁶ vegetative P. sordellii bacteria on Day 0. Animals were weighed/monitored for four days. (B) Survival curve of diestrus animals prophylactically treated with PA41 or PBS and TC inoculated with 2x10⁵ CFUs vegetative bacteria. (C) Survival curve following transcervical inoculation of 10⁵ and 10⁶ spores in mice in diestrus. (D) Timeline showing subcutaneous administration of medroxyprogesterone acetate on Day -5 to induce diestrus, transcervical infection of 10⁶ spores on Day 0 and intraperitoneal injection of 7.5 mg/kg of PA41 or PBS one-, three-, and five-days post-infection. Animals were weighed/monitored for 7–10 days. (E) Survival curve of diestrus animals treated with PA41 or PBS following TC inoculation of 10⁶ spores. Gehan-Breslow-Wilcoxon test was used with statistical significance set at a p value of <0.05. Gehan-Breslow-Wilcoxon p = 0.0424, Log-rank p = 0.0624.