Vulvovaginal candidiasis (VVC), predominantly caused by Candida albicans, is a common fungal infection in women of reproductive age, affecting approximately 75% of the otherwise healthy female population at least once during their lifetime (Sobel, 2007; Yano et al., 2019). Fulminant VVC is typically described as itching, burning, redness of the vulva, and vaginal mucosa accompanied by cottage cheese-like vaginal discharge (Sobel, 1992). Despite the high incidence worldwide, current therapies for VVC are largely limited to topical or oral antifungal agents, primarily azoles, most of which are effective for treatment of acute episodes via fungistatic, not fungicidal, mode of action (Sobel, 1992). Incomplete clearance of infection can result in a rapid relapse, causing prolongation of debilitating symptoms and healthcare burden especially in those susceptible to recurrent VVC (RVVC), defined as ≥4 episodes of VVC per year (Sobel, 1997).

Historically, susceptibility to VVC/RVVC was believed to be attributed to systemic and local immunodeficiencies similar to other forms of candidiasis (Fidel, 2002; Yano et al., 2012), in which infections occur due to compromised T-cell immunity (oropharyngeal candidiasis; Pirofski and Casadevall, 2009; Fidel, 2011; Gaffen et al., 2011), immunosuppression from neutropenia (candidemia; Walsh and Gamaletsou, 2013; Mohammadi and Foroughifar, 2016), or direct invasion of the organisms into the bloodstream through abdominal trauma or indwelling catheters (disseminated candidiasis; Vonk et al., 2006; Koh et al., 2008). However, more recent findings from a live challenge study and a series of work using a mouse model revealed that an acute inflammatory response by polymorphonuclear neutrophils (PMNs) occurs during vaginal infection with strong correlation with severity of VVC symptomatology (Fidel et al., 2004; Yano et al., 2010; Peters et al., 2014). Subsequent studies in mice identified paramount mediators of PMN migration and its positive feedback response, namely, IL-1β and S100A8, as hallmarks of symptomatic infection (Yano et al., 2014; Willems et al., 2018). In addition, the inflammatory response is also associated with tissue damage as measured by lactate dehydrogenase (LDH; Yano et al., 2014; Willems et al., 2018). Hence, VVC/RVVC is now considered an immunopathology driven by a strong PMN response to C. albicans (Yano et al., 2018). After nearly three decades of extensive research, however, mechanisms of protection against C. albicans vaginal infection are not well understood.

Despite the robust recruitment and abundance, PMNs have no apparent contribution to fungal clearance in the vaginal environment while fully retaining pro-inflammatory properties (Yano et al., 2010). A recent study using mouse strains exhibiting susceptible vs. resistant conditions of experimental VVC indicated that vaginal heparan sulfate (HS), which is present at high levels under the estrogen-responsive state, serves as a competitive ligand for a PMN cell surface receptor Mac-1, thereby interfering with the binding to C. albicans Pra-1 (Yano et al., 2017). Hence, the PMNs fail to kill C. albicans but retain the activated inflammatory state in the vaginal mucosa. Given the novel mechanism of vaginal PMN dysfunction, termed “Neutrophil Anergy” (Yano et al., 2018), therapeutic approaches should focus on preventing or reducing VVC immunopathology so as to eliminate the symptomatic condition and reduce recurrent episodes.

Leukotrienes (LTs) are a group of eicosanoid inflammatory mediators derived from arachidonic acid via the 5-lipoxygenase (5-LO) pathway (Dixon et al., 1990). The enzyme 5-LO is an essential component of the biosynthesis of all LT products in various leukocytes, including neutrophils, and deletion of 5-LOX results in complete LT deficiency in mice (Reid et al., 1990). In response to inflammatory stimuli, such as infections, arachidonic acid is liberated from the cell membrane and converted via the sequential actions of 5-LO and downstream enzymes into LTB₄ or cysteinyl leukotrienes (cys-LTs) comprising of LTC₄, LTD₄, and LTE₄ (Dixon et al., 1990). LTB₄ is known to be a potent mediator of neutrophil chemotaxis into target tissues, and excess production can lead to neutrophil-mediated pathology as seen in type-2 diabetes and rheumatoid arthritis (Chen et al., 2006; Mathis et al., 2007; Spite et al., 2011). Similarly, cys-LTs are produced by a variety of activated leukocytes, including neutrophils, mast cells, and eosinophils, and are predominant effectors in pathogenesis of asthma (Niimi, 2013). Various LT receptor antagonists (LTRAs) have been developed, of which cys-LTRAs zafirlukast and montelukast are best characterized, currently approved drugs for the prevention and treatment of asthma (Suissa et al., 1997; Reiss et al., 1998). Other LTRA agents include etalocib, a selective LTB₄ receptor antagonist, that has been under investigation (Bhatt et al., 2017).

Intriguingly, an open-label pilot study of zafirlukast indicated that regimens with 20mg twice daily for 24weeks provided symptom relief in women with RVVC (White et al., 2004). Considering the extensive efforts required for development of novel therapeutics, drug repurposing has tremendous advantages in comparison with traditional approaches to drug discovery. Therefore, the purpose of this study was to evaluate the effects of LTRA agents (etalocib, zafirlukast, and montelukast) and LT deficiency (5-LOX knockout mice) on the vaginal immunopathogenic response in a mouse model of VVC.

Given the wide use of antifungals that often fail to provide complete clearance and long-term relief for VVC, there remains a need for more effective antifungals or immunotherapies that can alleviate the symptoms of vaginitis. In cases of severe RVVC, constant need for maintenance antifungal therapy leads to negative quality-of-life and significant healthcare burden for affected women. Here, our aim was to determine whether repurposing of two established cys-LT receptor antagonists, montelukast and zafirlukast, as well as a second-generation selective LTB₄ receptor antagonist, etalocib, could offer a novel immunotherapeutic strategy for long-term symptom relief. Results showed that none of the drugs were able to ameliorate the parameters of VVC immunopathology to any measurable level. Recognizing the dysfunction in PMNs against Candida in the vaginal environment (Yano et al., 2017), a reduction in PMNs to any extent by the drugs would not have been expected to modulate vaginal fungal burden, but we did expect a reduction to be reflected in the inflammatory parameters (IL-1β, S100A8, LDH). However, not only were the drugs ineffective, the lack of effects on the immunopathology was further supported by similar immunopathologic parameters in inoculated 5-LOX^−/− mice compared to wild-type mice.

Together, these findings were rather surprising given the previous open-label pilot study demonstrating clinical efficacy in maintaining a symptom-free period following a 24-week treatment with zafirlukast and for up to 1year following the final dose (White et al., 2004). In fact, this animal study was borne out of the pilot study so as to inform the design for a larger clinical trial. While recognizing the limited sample size of the pilot study, the discrepancy to the animal study could be the robustness of the mouse model including the requirement for exogenous 17β-estradiol. There is evidence that sex hormones influence 5-LO production and subsequent biosynthesis of LTs, though the effect is predominantly derived from the disparate testosterone levels (Pergola et al., 2008). Importantly, neither 17β-estradiol nor progesterone levels had appreciable effects on LT expression in neutrophils from human blood in response to a variety of stimuli for 5-LO induction (Pergola et al., 2008). Furthermore, a study conducted in mouse zymosan-induced peritonitis revealed similar levels of LTB₄ and LTC₄ in peritoneal exudates of females and orchidectomized males compared to sham males that exhibited significantly lower peritoneal LT levels (Rossi et al., 2014). Hence, androgens appear more downregulatory for the 5-LO activity. Taken together, although the animal model is robust, it is unlikely that the estrogen treatment adversely affected the action of the LTRA drugs or LT biosynthesis.

Taking into account a strong PMN response as the chief mediator of VVC immunopathology, etalocib presented a strong potential owing to its selective action on LTB₄-mediated neutrophil responses (Ford-Hutchinson et al., 1980; Jackson et al., 1999; Silbaugh et al., 2000; Chen et al., 2006). To our disappointment, etalocib failed to dampen the PMN response during infection. Due to its poor solubility, however, our experimental concentrations were limited to a fairly low range (≤1mg/kg). It is possible that etalocib could exert an anti-inflammatory effect on the vaginal mucosa at higher concentrations, possibly by use of a salt formulation with increased solubility (not commercially available; Kuwabara et al., 2000; Silbaugh et al., 2000). Interestingly, although a study conducted in patients with psoriasis indicated that etalocib was safe and well-tolerated (Pelt et al., 1998), the development of the drug for treatment of inflammatory indications was discontinued. A more recent study concluded that etalocib was not effective in preventing recurrent psoriasis (Mommers et al., 2000).

Montelukast and zafirlukast are currently approved medications indicated for the prevention and treatment of chronic bronchial asthma (Choi and Azmat, 2021) and have been investigated for other inflammatory conditions such as allergic rhinitis, inflammatory bowel disease, chronic obstructive pulmonary disease (COPD), and urticaria (hives; Mahgoub et al., 2003; Riccioni et al., 2007; Khan and Lynch, 2012). Cys-LT-mediated pathology is commonly manifested as increased vascular permeability and mucus production, mainly by activation of mast cells, eosinophils, and basophils that are not part of VVC immunopathology (Peters-Golden and Henderson, 2007; Theron et al., 2014). However, the transient reduction in PMN migration was encouraging and prompted us to further verify the effect of the drugs in vaginal inflammatory markers (IL-1β, S100A8) and tissue damage (via LDH). Unfortunately, these other parameters could not support any measurable reduction in PMNs indicative of any efficacy. Given no firm evidence of montelukast and zafirlukast efficacy in vivo despite employing optimal or the highest concentrations possible, together with the possibility of not achieving an effective dose of etalocib, results from the drug studies remained somewhat insufficient to reach an absolute conclusion on potential for clinical use. Since 5-LO is essential for the biosynthesis of all LT derivatives, 5-LOX^−/− mice are resistant to LT-associated inflammatory conditions, such as PMA-induced skin edema and immune complex peritonitis (i.v. ovalbumin followed by i.p. anti-ovalbumin IgG), in which 5-LOX^−/− mice exhibited significantly reduced PMN infiltrates compared to wild-type mice (Chen et al., 1994). Hence, the 5-LOX^−/− mice represented a strong parallel design to aid the conclusions of the drug studies. Results revealed that 5-LOX^−/− and wild-type mice exhibited similar outcomes in all immunopathologic parameters of vaginitis. In addition, similar results were obtained regardless of the C. albicans strains (a laboratory strain vs. a vaginal isolate) at varying inocula, confirming a lack of variables potentially introduced by fungal factors. Moreover, the 5-LOX^−/− mice in the C57BL/6 background had a normal PMN response comparable to our standard design using C3H mice. Thus, we showed similar results of LT deficiency/antagonism that were reproducible in two different strains of mice with distinct MHC haplotypes. While this indeed represents strong evidence overall, the literature suggests regulation of pro-inflammatory cytokines by 5-LO in response to pathogens, as well as interaction of S100A8/A9 with arachidonic acid and its metabolites. A transient reduction seen in the levels of IL-1β and S100A8 could be due to deletion of 5-LOX that may have had an indirect impact on protein expression or secretion. Support for this includes (i) reduced systemic cytokine production including IL-1β in 5-LOX^−/− mice in a cecal ligation and puncture (CLP)-induced peritonitis model (Monteiro et al., 2014), (ii) S100A8/A9 sequestration by arachidonic acid that possibly becomes accumulated due to a lack of active 5-LO (Kannan, 2003), or (iii) downregulation of S100A8 expression in 5-LOX^−/− mice in the absence of LTA₄ which was shown to scavenge S100A8/A9 proteins for protection from hydrolytic degradation (Rector and Murphy, 2009). While these alternative explanations are plausible in our model, all data realistically point to LTs being dispensable to VVC immunopathogenesis. Thus, agents that target LTs are not likely viable options as therapeutics to reduce symptomatic VVC.

Despite the contradictory results between the mouse model and open-label pilot study (White et al., 2004), there still may be some motivation for a phase III clinical trial with zafirlukast in cases of VVC or RVVC. Since vaginal PMN infiltrates were not evaluated in the pilot study, as in most clinical settings, the magnitude of vaginitis symptoms can become highly subjective or variable between women. Hence, a trial that includes multi-parameter examination of VVC/RVVC symptomatology via vaginal PMN infiltrates and inflammatory mediators could serve as the most reliable indicator of the immunopathology and confirmatory correlates to general signs and symptoms. Additionally, although we feel that RVVC is an exacerbated immunopathogenesis under similar mechanisms to VVC, results from our model that only evaluated acute VVC should not be presumed to apply to RVVC without further experimentation.

In summary, we have demonstrated that pharmacological inhibition or genetic deletion of LT production had no influence on the outcomes of fungal loads and all inflammatory parameters of vaginitis in the mouse model. These data therefore suggest that LTs are dispensable to VVC immunopathology. Based on the detectable and increased presence of vaginal LTB₄ during acute VVC shown here, as well as during experimental vaginal Trichomonas vaginalis infection (Eida et al., 2015), it will be interesting to determine the mechanisms of LT production and receptor expression, both locally and systemically, during vaginal infection. Unfortunately, the assessment of LTs (locally or systemically) in the presence of the drugs is not informative since the drugs act as receptor antagonists and not synthesis inhibitors. Irrespective though it will be interesting to identify what pathways are involved in PMN migration under targeted antagonism/inhibition. This together with current strategies to overcome/rescue the PMN dysfunction (neutrophil anergy) in order to achieve adequate vaginal antifungal activity will be critical to reducing the immunopathology and provide relief to the large number of women affected by VVC/RVVC.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal study was reviewed and approved by Institutional Animal Care and Use Committee (IACUC) of the LSU Health-New Orleans.

PF conceived and designed the experiments. JY performed the experiments. FW provided mouse breeders. JY and PF analyzed the data and wrote the manuscript. DW, AS, and FW edited the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported by the LSUHSC Foundation. The Foundation had no role in study design, data collection, and interpretation or the decision to submit the work for publication.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.