https://orcid.org/0000-0002-9454-9234
Figures
Abstract
Trichomonas vaginalis is the causative agent of the common sexually transmitted disease, trichomoniasis, which affects more than a hundred million people worldwide. Metronidazole and tinidazole, agents belonging to the 5-nitroheterocyclic class of antimicrobials, are most often used to treat infection, but increased resistance has been reported and adverse effects of these drugs can be significant. Consequently, an urgent need exists for the development of novel drug entities against trichomoniasis. Critical for antimicrobial drug development is the demonstration of in vivo efficacy. Murine models of vaginal T. vaginalis infection are unreliable for unknown reasons. Meanwhile, murine infections with the related bovine pathogen, Tritrichomonas foetus, tend to be more robust, although susceptibility to different antimicrobials might differ from T. vaginalis. Here, we explored the utility of T. foetus infection as a surrogate model for drug development against T. vaginalis. Four different T. foetus strains caused robust vaginal infection in young mice, while none of four diverse T. vaginalis strains did. Comparison of drug susceptibility profiles revealed that T. foetus and T. vaginalis were similarly susceptible to a range of 5-nitroheterocyclic and gold(I) compounds. By comparison, proteasome inhibitors were 10- to 15-fold less active against T. foetus than T. vaginalis, although one of the proteasome inhibitors, bortezomib, had low micromolar activity or better against multiple strains of both trichomonads. Different strains of T. foetus were used to demonstrate the utility of the murine vaginal infection models for in vivo efficacy testing, including for bortezomib and a gold(I) compound. The differences in susceptibility to proteasome inhibitors may be partially explained by differences in the proteasome subunit sequences between the two trichomonads, although the functional relevance of the proteasome was similar in both organisms. These findings indicate that T. foetus can serve as a reliable surrogate model for T. vaginalis in vitro and in murine infections in vivo, but caution must be exercised for specific drug classes with targets, such as the proteasome, that may display genetic divergence between the trichomonads.
Citation: Nieskens NM, Miyamoto Y, Hurysz BM, O’Donoghue AJ, Eckmann L (2024) Vaginal Tritrichomonas foetus infection in mice as an in vivo model for drug development against Trichomonas vaginalis. PLoS ONE 19(10): e0308672. https://doi.org/10.1371/journal.pone.0308672
Editor: Nneoma Confidence JeanStephanie Anyanwu, New York Medical College, UNITED STATES OF AMERICA
Received: February 1, 2024; Accepted: July 29, 2024; Published: October 1, 2024
Copyright: © 2024 Nieskens et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: The studies were supported by National Institutes of Health (https://www.nih.gov/) grants AI158612 (LE, AO), DK120515 (LE) and T32 GM007752 (BMH). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Trichomoniasis, caused by infection with the protozoan parasite Trichomonas vaginalis, is the most common, non-viral sexually transmitted disease in the world with a global incidence of ~156 million, and a global prevalence of 5.4% [1]. The parasite can infect different genitourinary tissues, including the urethra, vagina, and cervix of women, and the prostate and urethra of men [2]. Infected women can experience a broad range of symptoms, including abnormal vaginal discharge, odor, and itching, vulvar or vaginal erythema, and colpitis macularis (“strawberry cervix”) [3]. Many women remain infected for extended periods [3]. Infection in men can lead to mild to moderate urethral discharge and urethral inflammation, but symptoms tend to be less severe than in women [1, 4]. The global prevalence of T. vaginalis infection is almost 10 times lower in men compared to women [1], which may reflect a lack of testing due to milder infection symptoms or could be related to an abbreviated course of infection. Nonetheless, lack of symptoms does not protect an individual from the complications associated with infection. For example, T. vaginalis infection can increase susceptibility to human immunodeficiency virus (HIV) infection by 50% [5]. Furthermore, a positive correlation between HIV infection and trichomoniasis symptom severity was found in men [6]. In women, infection has also been linked to an increased risk of cervical cancer, low infant birth weight, preterm birth, and stillbirths [7, 8]. In men, a decrease in sperm viability and motility has been observed with the risk to decrease fertility [9]. To avoid potential complications, it is important to treat infection upon diagnosis, particularly since infection can last for months to years [10].
Metronidazole and tinidazole are two antimicrobials currently FDA-approved for the treatment of trichomoniasis [11]. They belong to the class of 5-nitroheterocyclic drugs. Metronidazole, first developed in 1959, is the primary drug for trichomoniasis treatment worldwide [12]. Tinidazole, a structural derivative of metronidazole, was FDA-approved for trichomoniasis treatment in 2004 [11]. Both are prodrugs that must be reduced to a free radical inside microbial cells before they can covalently bind to multiple protein and nucleic acid targets. Adduct formation leads to inhibition of DNA synthesis and many vital protein functions, eventually killing the organism. Both drugs primarily target anaerobic microorganisms, including T. vaginalis, as these organisms possess enzymatic pathways with low enough redox potential to donate electrons to the drug molecules [12]. For metronidazole, cure rates range from 86–97% [13–15]. Despite its general effectiveness, an estimated 2–10% of all clinical trichomoniasis cases are caused by strains that display resistance to metronidazole and other 5-nitroheterocyclic drugs [16–18]. Beyond loss of efficacy due to resistance, metronidazole can have significant adverse effects that compromise its use and effectiveness. Thus, a large percentage of women treated with the drug have reported nausea, vomiting, unpleasant taste, diarrhea, headache, vaginal irritation, and yeast infections [19, 20]. Currently no other drug classes are FDA-approved to treat trichomoniasis. With the threat of growing resistance, and significant adverse effects, development of alternative antimicrobials in other drug classes remains urgent for treatment of trichomoniasis.
Library screens of approved FDA drugs and drug candidates have been done to identify new activities against T. vaginalis but have not resulted in promising candidates [21]. De novo drug design against unexploited targets is currently the most realistic strategy for development of new antitrichomonal agents. One such strategy is exemplified by proteasome inhibitors [22]. The proteasome is an enzymatic complex present in all eukaryotic and some prokaryotic cells, whose function is to degrade and recycle proteins that are misfolded or not needed [23]. The enzyme complex consists of a barrel-like structure, consisting of two pairs of rings with seven α-subunits and seven β-subunits. Peptide substrates are fed through the center of the barrel where they are cleaved into smaller fragments. Three of the 14 proteasome subunits are catalytically active: β1 (caspase-like), β2 (trypsin-like), and β5 (chymotrypsin-like). Each cleaves the substrate at its preferred site via hydrolysis [24]. Proteasome inhibitors were first developed as anti-inflammatory agents and as novel agents for cancer treatment [25]. For example, the proteasome inhibitor bortezomib, was FDA-approved for the treatment of multiple myeloma and mantle cell lymphoma. More recently, proteasome inhibitors have also been found to be effective against Plasmodium falciparum, the protozoan responsible for malaria [26]. Importantly, several new proteasome inhibitors were shown to be highly selective for the parasite over human cells [27]. T. vaginalis also possess a proteasome that can be targeted for the development of new agents against the parasite, since several proteasome inhibitors were active against the parasite in vitro and have selectivity over human cells [22].
Critical for antimicrobial drug development is the demonstration of in vivo efficacy. Ideally, animal models of infection can be used that employ the very same pathogens that infect humans. In the case of trichomoniasis, several reports have shown that mice can be infected vaginally with T. vaginalis [28–30], but infections tend to be tenuous and unreliable for poorly understood reasons. As an alternative, mouse infection models with surrogate trichomonad pathogens can be used if it can be shown that those infections are robust, and the pathogens respond to antimicrobial candidates in similar ways to the human target pathogen. For example, vaginal infection of mice with the bovine pathogen, Tritrichomonas foetus, have been shown to cause sustained infection [28]. T. foetus is a protozoan parasite that causes tritrichomoniasis in cattle, a similar disease to human trichomoniasis. The parasite is closely related to T. vaginalis, being in the same phylum [31]. Despite its promise, the broader utility of this model for drug development against T. vaginalis remains to be established. Therefore, the aims of this study were to establish infectious characteristics in mice, similarities in in vitro drug responses, structural target similarities of a set of T. foetus strains in relation to T. vaginalis and establish the overall utility and limits of the murine T. foetus model as a basis for current and future drug development against T. vaginalis.
Materials and methods
Trichomonad strains
T. foetus strains 003 (ATCC 30003), 166 (ATCC 30166), 232 (ATCC 30232), 924 (ATCC 30924), and T. vaginalis strain G3 (ATCC PRA-98) were obtained from the American Type Culture Collection (Manassas, VA). T. vaginalis strains R88, MSA1172, S520 and S1489 were kindly provided by Dr. William Secor (Centers for Disease Control and Prevention, GA). T. foetus strain D1 and T. vaginalis strain F1623 were described in prior studies [32, 33]. Trichomonads were grown in trypticase-yeast extract-maltose (TYM) medium, supplemented with 10% horse serum (Omega), 0.74% ammonium iron (II) sulfate hexahydrate (Sigma-Aldrich) and 1% penicillin-streptomycin (Sigma-Aldrich), and adjusted to pH 6.2. Cultures were grown axenically at 37°C under anaerobic conditions.
Antimicrobial compounds
The following 5-nitro heterocyclic were purchased commercially (molecular masses in parentheses): Metronidazole (171 Da; Sigma-Aldrich), ronidazole (200 Da; APExBIO), tinidazole (247 Da; TCI America), nitazoxanide (307 Da; AdipoGen), and nithiamide (187 Da; MedChemExpress). The following gold(I) compounds (CPD), whose naming follows our prior usage [34], were also purchased: Auranofin (679 Da; Enzo Life Sciences), (Tri-n-ethylphosphine)gold(I) chloride (351 Da, CPD4; Sigma-Aldrich), (Tri-n-methylphosphine)gold(I) chloride (309 Da, CPD10; Sigma-Aldrich) and myochrysine (390 Da; Sigma-Aldrich). The following gold(I) compounds were synthesized and characterized as described previously [34]: (Tri-n-propylphosphine)gold(I) chloride (357 Da, CPD11), (Triisopropylphosphine)gold(I) chloride (357 Da, CPD12), Bis(triethylphosphine)gold(I) chloride (433 Da, CPD14), and Bis(tri-n-propylphosphine)gold (I) chloride (517 Da, CPD15). The proteasome inhibitors bortezomib (384 Da), ixazomib (361 Da) and carfilzomib (720 Da) were obtained from MedChemExpress. The benzimidazole compounds lansoprazole (369 Da), omeprazole (345 Da), albendazole (281 Da) and fenbendazole (299 Da) were also obtained from MedChemExpress. Pantoprazole (383 Da) was purchased from APExBIO and disulfiram (297 Da) from TCI America.
Mice
BALB/cJ mice (obtained from The Jackson Laboratory) were bred in-house. Female mice were used within one week of weaning at 3–6 weeks of age. Groups of mice were housed in single plastic cages with free access to feed and water.
The study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the University of California San Diego Institutional Animal Care and Use Committee (Protocol Number: S00205). Mice were handled with care, only by research staff fully trained in animal handling, at all times to minimize distress. Mouse health was monitored daily, as well as each time a sample was taken. Vaginal trichomonad infections in mice cause no pain, distress, or suffering, obviating any need for humane endpoints. Experimental endpoints were based solely on the infectious dynamics relative to the study objectives of assessing drug effects on infectious load. All animals were euthanized on the last day of sample collection by controlled CO2 inhalation, followed by cervical dislocation.
Vaginal infections and treatment
Infections were performed with different strains of either T. vaginalis or T. foetus as indicated under the specific experiments in the figures. Trichomonads were cultured overnight, centrifuged, and resuspended at a density of 2 x 108 trophozoites/mL. For all in vitro and in vivo experiments, numbers of viable trophozoites were determined under a phase-contrast microscope. Viability was >95% as assessed by observation of the characteristic parasite motility. Five μL of the parasite suspension (containing 106 trophozoites) were slowly administered intravaginally to mice with a 20 μL micropipette. Mice were held briefly upside-down after agent administration to minimize leakage.
In select experiments (as indicated in the figure legends), mice were pretreated one day before infection with one of following pretreatment regimens: a) 5 μL intravaginal administration of 10 mg/mL ammonium iron (III) citrate (Sigma-Aldrich); b) 100 μL intraperitoneal (i.p.) injection of 10 mg/mL ammonium iron (III) citrate; c) 100 μL i.p. injection of 4 mg/mL 17β-estradiol (Sigma-Aldrich); or d) 100 μL i.p. injection of 4 mg/mL 17β-estradiol (Sigma-Aldrich) and 100 μL i.p. injection of 1 mg/ml dexamethasone (Sigma-Aldrich). The ammonium iron (III) citrate solution for intravaginal administration was made in PBS containing 1% hypromellose, while all formulation for i.p. injections were prepared in PBS. Control mice received no pretreatment.
For drug efficacy tests in vivo, suspensions of bortezomib or gold (I) CPD4 were prepared at 10 mg/mL (= 1% w/v) in 0.2% hypromellose in PBS. Hypromellose solution without drugs was used as a control. Five μL of the drug suspensions (containing 0.05 mg of drug) were administered intravaginally as outlined above for the parasite inoculations. A total of five doses were administered: One dose on day 1 after infection, two doses each on days 2 and 3 after infection with 6–8 h between treatments on those days.
Infectious load was determined 3–4 days after infection for all experiments except for a time course study with T. foetus strain 232, where mice were sampled on days 1, 3, 6, and 9. For assessment of infection, mice were lavaged intravaginally by slowly dispensing 30 μL of TYM medium into the vagina, taking it up again, and repeating the process two more times. The washes were diluted to 120 μL in TYM medium, and 10 μL of the diluted washes were placed into a hemocytometer for counting of motile trophozoites under a phase-contrast microscope. Infectious load is expressed as total number of trophozoites per mouse. For washes without countable trophozoites, the remaining volume was cultured in 2 mL of TYM medium containing 100 units/ml penicillin and 100 μg/ml streptomycin for five days at 37°C, and regularly monitored for cell growth.
In vitro drug testing
Drug stocks were prepared as 10 mM solutions in DMSO. A 75 μM working stock was prepared in PBS and 1:3 serial dilutions were made in TYM medium in 96-well plates (post-dilution volume 40 μL/well). Parasite cultures were grown overnight in a 15 mL tube, centrifuged and resuspended to a density of 7 x 105–2 x 106 trophozoites/mL. Ten μL of this suspension were added into each well (final culture volume 50 μL/well), resulting in a final number of 7,000–20,000 trophozoites per well and final drug concentrations from 20 μM to 1 nM. Plates were incubated at 37°C for 24 hours under anaerobic conditions (AnaeroPack; Mitsubishi Gas Chemical). Viable cell numbers were determined with a luciferase-based ATP assay (Promega) [35]. Luminescence was measured with the SpectraMax M2e plate reader, using the SoftMax Pro program (version 7.0.3).
To assay the minimum lethal concentration (MLC), drugs were added to 96-well plates, and serially 2-fold diluted in TYM medium. Parasites were added to each well at a density of 7 x 105/mL (T. foetus D1) or 106 trophozoites/mL (T. vaginalis R88), resulting in a final number of 28,000 or 40,000 trophozoites per well, respectively. Plates were incubated at 37°C for 24 hours under anaerobic conditions. Cultures (200 μL) were then transferred to medium-filled 15 mL tubes (representing a 1:75 dilution of the overnight culture to ensure adequate dilution of the drugs in those cultures) and kept at 37°C for a maximum of five days for observation of any cell growth indicative of survivors of the initial drug exposure.
Proteasome analysis
For sequence analysis of proteasome subunits, the genome and proteome sequences of the following parasite strains were used: T. foetus CC09-1 (GenBank AIF54491–504), T. vaginalis G3 (BioProject PRJNA938429, PRJNA16084) and P. falciparum NF54 (BioProject PRJNA422809), along with that of humans (BioProject PRJNA807723) and mice (BioProject PRJNA169). A phylogenetic tree was generated using MEGA software (version 11.0.13). Support values were calculated by bootstrapping with 100 replications. Protein sequences were aligned with the MUSCLE algorithm and an unrooted tree was constructed using the Le Gascuel model, with gamma distribution and invariant sites. Gaps were partially deleted, so that sites with missing data were only removed when necessary. Proteasome subunit designation followed the findings in [36].
For immunoblotting, T. foetus D1 and T. vaginalis F1623 cultures were incubated at 37°C for 1 hour with 0.3 μM or 3 μM of bortezomib, or with DMSO as a control. Lysates were made with radioimmunoprecipitation (RIPA) buffer (Cell Signaling Technology), supplemented with Halt protease inhibitor cocktail (Thermo Scientific). Proteins were separated on a 4–20% SDS-PAGE gel (Bio-Rad), transferred onto a PVDF membrane (Millipore) and blocked with 5% nonfat dry milk. The membrane was incubated with a primary rabbit anti-ubiquitin polyclonal antibody (Proteintech, Cat No. 10201-2-AP) at a 1:10,000 dilution overnight and a secondary horseradish peroxidase (HRP)-linked anti-rabbit IgG antibody (Cell Signaling Technology, Cat No. 7074S) at a 1:20,000 dilution for 30 minutes. Proteins were detected with a chemiluminescent HRP substrate (Millipore).
For labeling of catalytically active proteasome subunits with an activity-based probe, total cell lysates of T. foetus D1 or T. vaginalis F1623 were incubated at 37°C for 1 hour with 0.1–10 μM of bortezomib, or with DMSO as a control. Afterwards, 2 μM of the fluorescent proteasome probe, MV151 (R&D Systems, Cat No. I-190) [37] was added and lysates were further incubated at 37°C for 2 hours. Proteins were separated on a 4–20% SDS-PAGE gel, and bands were visualized by fluorescent imaging with excitation at 470 nm and emission at 530 nm.
Statistics
All statistical analyses were performed with GraphPad Prism (version 10.0.0). For compound activity screening, luminescence values were normalized against drug-free controls and plotted against the log drug concentration in each well to create a concentration-response curve for each drug. The half-maximal inhibitory concentration (IC50), or concentration of drug needed to inhibit trophozoite growth and viability by 50%, and negative log10 of the IC50 (pIC50) were calculated. The assay sensitivity for pIC50 was set at 4.7, equivalent to the highest drug concentration tested in the assays. Each experiment was independently repeated three times per compound and per trichomonad strain.
For in vitro data comparisons, parametric tests (ANOVA, t-tests) were used. For comparison of multiple groups, a one-way ANOVA was performed, and if a significant difference was found, Tukey’s multiple comparisons test was applied to specific pairs. For in vivo data comparisons, a non-parametric Mann-Whitney test was used.
Results
Murine vaginal infection model with diverse strains of T. foetus and T. vaginalis
Previous studies had suggested that murine vaginal infections with T. vaginalis are characterized by low infectivity and/or short-lived infections, even with various manipulations of the host designed to increase susceptibility [28]. It is possible that marginal infectivity is related to the particular strains of T. vaginalis used for the infections, so we tested a range of four diverse strains of T. vaginalis for their ability to cause vaginal infection in young female mice. Three to four days after intravaginal inoculation, vaginal washes showed no detectable trophozoite counts for any of the tested strains (Fig 1A). Cultures of the washes for up to 5 days were uniformly negative, underlining that none of the T. vaginalis strains led to even low-level infection (Fig 1A). Furthermore, systemic (i.p.) or topical (vaginal) administration of iron, which improves T. vaginalis growth in culture [38], or systemic administration of estrogen alone or estrogen plus dexamethasone, which have been suggested to improve susceptibility of mice to T. vaginalis infection [28, 39], did not promote infectivity (Fig 1B). These experiments suggest that T. vaginalis could not infect mice under any of our current conditions, regardless of particular strains or conditioning regimens.
Young (3–4 weeks) female BALB/cJ mice were intravaginally inoculated with the indicated strains of T. vaginalis (A, B) or T. foetus (C, D). In the experiments in panel B, mice were pretreated by vaginal instillation (topical) of iron or intraperitoneal (i.p.) injection of iron, estrogen, or estrogen and dexamethasone (dex) before infection, at dosages detailed in the Methods. Vaginal lavages were taken on day 3 or 4 of infection (A-C) or on the indicated days (D), and motile trophozoites were counted in a hemocytometer. Data points represent counts from individual mice, dashed lines indicate the detection threshold of the assays, and red diamonds show geometric means. For select samples without countable trophozoites, lavages were cultured in trichomonad growth medium with antibiotics (blue symbols). Positive cultures are indicated by solid blue dots, while negative cultures (no growth) are depicted by open blue dots.
As an alternative to T. vaginalis, we explored the related trichomonad, T. foetus, which was originally derived from cows and has been used for mouse infections [22, 28, 40]. Four different T. foetus strains were obtained and tested for their vaginal infectivity in young female mice. All four strains were able to cause infection without any pre-treatment with infectivity close to 100% (i.e., the percentage of mice with trophozoites detectable by direct counting or culture over total number of inoculated mice) (Fig 1C). To examine the time course of infection, we selected the T. foetus strain 232, which had the highest infectivity and/or highest trophozoite loads, and sampled mice every 2–3 days. All mice remained infected throughout the 9 days of the experiment, indicating a lack of spontaneous clearance during that period (Fig 1D). These findings demonstrate that T. foetus constitutes a more reliable murine model of vaginal trichomonad infection than T. vaginalis.
Differential growth characteristics of T. foetus and T. vaginalis
To explore potential reasons for the differential murine vaginal infectivity of T. vaginalis and T. foetus, we examined two parasite characteristics that may plausibly be involved: maximal growth rate and tolerance to different pH environments. The T. vaginalis strains had significantly slower growth rates in culture with longer doubling times compared to the T. foetus strains despite optimal growth conditions and media (Fig 2A). Furthermore, T. vaginalis growth was inhibited by even modestly alkaline pH in vitro, while T. foetus was able to grow mostly unhindered under these conditions (Fig 2B). These data indicate that the intrinsic properties of the two trichomonad species exhibit significant differences in regard to growth and pH tolerance.
Trophozoites of the indicated strains were cultured in vitro in trypticase-yeast extract-maltose (TYM) medium. (A) Cell numbers were counted before and after a 24-hour time incubation period and doubling times were calculated. Data are mean ± SE; *p<0.05 (unpaired t-test for all T. vaginalis strains vs. all T. foetus strains). (B) Media were adjusted to the indicated pH and trophozoites were grown in each medium for 24 hours and counted. Data are mean ± SE.
To test whether the differential pH sensitivity of the trichomonad strains might be responsible for their differential infectivity in vivo, we measured the vaginal pH in 3–6 week old mice before and after treatment with estrogen alone or with estrogen and dexamethasone [28]. Mean vaginal pH was 6.21 ± 0.04 in the untreated group (mean ± SE, n = 7 mice), 6.49 ± 0.09 for the estrogen-treated group (n = 7), and 6.71 ± 0.05 for the mice treated with estrogen and dexamethasone (n = 9). A Kruskal-Wallis and Dunn’s multiple comparisons test showed significance between the estrogen and dexamethasone versus the no treatment group. However, these data show that the vaginal pH was below 7.0 in all groups, thereby excluding differential pH tolerance as a possible explanation for the differential murine vaginal infectivity of T. vaginalis vs T. foetus.
Activity profiles of different antimicrobial drug classes against diverse T. foetus and T. vaginalis strains
To compare susceptibility to antimicrobial drugs between the two trichomonad species, we screened different strains against four major drug classes that are clinically used in trichomoniasis or have been shown to have significant in vitro activity against T. vaginalis: 5-nitroheterocyclic drugs [41], gold(I) compounds [34], proteasome inhibitors [22], and benzimidazoles [21]. The most consistently active compounds were 5-nitroheterocyclic compounds and gold(I) compounds (S1 Table). By comparison, the tested proteasome inhibitors had excellent activity against the T. vaginalis strains but were significantly less potent against the T. foetus strains (S1 Table). The benzimidazoles and another compound with reported in vitro activity against T. vaginalis, disulfiram [21], had only minimal or no activity against either T. foetus or T. vaginalis (S1 Table).
Comparison of drug susceptibilities of T. foetus and T. vaginalis
Comparison of specific drugs within the three major drug classes with the greatest potencies revealed that all tested drugs were less active in T. foetus, although differences were observed between the drug classes (Fig 3). 5-Nitroheterocyclic drugs and gold(I) drugs displayed, on average, a ~3-fold lower activity (mean ΔpIC50 = 0.47–0.48) in the T. foetus strains compared to the T. vaginalis strains, but this difference did not reach statistical significance (Fig 3). By comparison, the tested proteasome inhibitors were ~14-fold less active (mean ΔpIC50 = 1.15) against T. foetus relative to T. vaginalis, which was statistically significant (Fig 3).
Activity (pIC50) of the indicated compounds from each of the indicated three drug classes against multiple strains of T. foetus and T. vaginalis was determined by 24 h growth and survival assay (the detailed data are provided in S1 Table). Each point represents the mean pIC50 of four (T. foetus) or three (T. vaginalis) different strains for the drugs listed below each graph. ΔpIC50 was calculated for each drug class by subtracting the mean T. foetus pIC50s from the mean T. vaginalis pIC50s. Red lines represent geometric means. Significance was calculated with Tukey’s multiple comparisons test (*p <0.01).
Pairwise comparisons of strains were conducted to discover whether any individual T. foetus strain had similar drug susceptibility profiles to any of the T. vaginalis strains. Strong positive correlations were seen among the different T. foetus strains for all three major drug classes (Fig 4A, Table 1). In contrast, only modest average correlations were observed between strains of T. foetus vs. T. vaginalis (Table 1), which particularly affected the proteasome inhibitors and gold(I) compounds (Fig 4B). This might be partially explained by the only modest average correlations of drug susceptibilities among the different T. vaginalis strains (Table 1). In particular, the T. vaginalis strain S1489 displayed a very different drug susceptibility profile than the other T. vaginalis strains (Fig 4C, Table 1), thereby markedly skewing the correlation data for both T. vaginalis and T. foetus (Table 1).
The graphs show correlations of the pIC50 values for individual drugs between the indicated pairs of trichomonad strains (the detailed data on drug activities used for these analyses are provided in S1 Table). Each point represents one drug and is color-coded by drug class. Plots are shown with a linear regression line (black) and line of identity (dotted red).
Together, these data indicate that T. foetus is generally less susceptible than T. vaginalis to all tested drugs, but this difference is particularly prominent and significant for the class of proteasome inhibitors.
MLCs of selected drugs against T. foetus and T. vaginalis
To further characterize antimicrobial drug activity in the two trichomonad species, we determined MLCs by culturing cells for 24 hours in the presence of titrated drug concentrations, diluting the cultures 75-fold (to minimize the impact of residual drugs), and testing for survivors by outgrowth over the ensuing 5 days. The MLCs for metronidazole as a representative 5-nitroheterocyclic compound, and auranofin as a representative gold(I) compound, were very similar for T. foetus or T. vaginalis, reflecting the trends of the respective IC50 findings (Table 2). MLCs were generally 3- to 10-fold higher than IC50, consistent with the notion that complete parasite killing requires higher drug concentrations than growth inhibition by 50%. In contrast, MLC values could not be determined for the two tested proteasome inhibitors, even when tests were done with >100-fold higher drug concentrations than IC50 (Table 2). Even higher drug concentrations could not be tested in the MLC assays due to toxicity of the solvent DMSO. These results confirm that the antimicrobial activities of 5-nitroheterocyclic and gold(I) compounds are not significantly different between the two trichomonad species. The data also show that the tested proteasome inhibitors are unlikely to be trichomonacidal for either T. foetus or T. vaginalis.
In vivo efficacy testing of drugs in murine T. foetus infection models
To determine the utility of the new T. foetus vaginal infection models for drug efficacy testing, and to evaluate whether selected drugs are efficacious in vivo, we tested two representative compounds with excellent in vitro activity, the gold(I) compound, CPD4, and the proteasome inhibitor, bortezomib, using two different T. foetus strains. Mice infected with T. foetus 232 were topically treated five times over three days with CPD4 and the infectious load was determined on day 4 (Fig 5A). None of the CPD4-treated mice had countable trophozoites, while ~70% of control mice were infected with an average of ~103 trophozoites per mouse (Fig 5B). Similarly, mice infected with the T. foetus strain D1 and treated with bortezomib showed no countable trophozoites, while all of the control mice had detectable infection (Fig 5C). These results demonstrate that the murine vaginal infection models are not only effective with different T. foetus strains, but are suitable for efficacy testing of new drug candidates against trichomonads. The results also reinforce the fact that CPD4 and bortezomib can clear T. foetus infection in an in vivo model.
(A) Timeline of infection and treatment. Mice were treated five times over the course of 2.5 days, starting pno day after infection. (B/C) Mice were infected with the indicated T. foetus strains and treated intravaginally with 0.05 mg/dose (5 μL of a 10 mg/mL suspension) of either the gold(I) compound CPD4 (B) or bortezomib (C), or were left untreated as controls. After 4 days, trophozoite numbers were determined. Each point represents an individual mouse. Dashed black lines represent the detection threshold. Red lines represent geometric means. Significance was calculated by Mann-Whitney test (*p <0.01).
Characterization of T. foetus 20S proteasome as a drug target
Bortezomib showed good in vitro and in vivo activity against T. foetus, yet it and the other proteasome inhibitors were markedly less potent against the different T. foetus strains than T. vaginalis (S1 Table). This discrepancy raised the question whether the T. foetus proteasome is substantially different from that of T. vaginalis. Analysis of the T. foetus genome identified all 14 proteasome subunits, seven α subunits (α1–7) and seven β subunits (β1–7), as expected for all eukaryotes (Fig 6). Sequence comparison of the subunits showed good homology (50–87% amino acid identity) with T. vaginalis, but much less homology (31–57%) with the human and murine counterparts, or even the protozoan parasite P. falciparum (Fig 6). By comparison, the human and murine proteasome subunits are more closely related to each other than the T. foetus subunits are to T. vaginalis. These findings indicate that structural differences exist between the T. foetus and T. vaginalis proteasome subunits, potentially offering an explanation for the differential activity of the tested proteasome inhibitors against the two trichomonad species.
Phylogenetic analysis of 20S proteasome α- and β-subunits from T. foetus (TF, bolded), T. vaginalis (TV), Plasmodium falciparum (PF), Mus musculus (MM), and Homo sapiens (HS). Branches are labeled with UniProt accession numbers when available, or otherwise GenBank accession numbers. Subunit groups are highlighted with the same color. Bootstrap values were calculated with 100 replications. Nodes supported with values ≥95% are marked with color-coded circles. Distance scale is shown on the bottom left.
To confirm that the T. foetus proteasome is functionally similar to T. vaginalis, we analyzed the accumulation of ubiquitin-labeled proteins after proteasome inhibition by western blotting. For both trichomonads, bortezomib caused concentration-dependent increases in the abundance of ubiquitin-labeled proteins (Fig 7A). Furthermore, we used an activity-based probe, MV151 [37] for labeling of the catalytic proteasome subunits of the parasites. For T. foetus, labeling of β1 and β5 was strongly inhibited by bortezomib, while β2 labeling was only minimally affected at the highest concentrations (Fig 7B). By comparison, for T. vaginalis, inhibition was most prominent for the β1 subunit followed by β5 and, to a lesser degree, β2 (Fig 7B). Together, these findings show that bortezomib inhibits the proteasome in both T. foetus and T. vaginalis, most likely by targeting the same catalytic subunits.
(A) The depicted trichomonad strains were incubated for 1 hour with the indicated final concentrations of bortezomib or left without inhibitor (0 μM). Total cell lysates were prepared and analyzed by western blotting with an anti-ubiquitin antibody (left panels). As expected, multiple proteins of different sizes are labeled with ubiquitin, leading to the apparent “smear” on the blots. Total protein was visualized by Coomassie Blue staining (right panels). (B) Total cell lysates of T. foetus and T. vaginalis were incubated for 1 hour with the indicated concentrations of bortezomib and subsequently labeled for 2 hour with the fluorescent activity-based proteasome probe, MV151. Lysates were fractionated by SDS-PAGE and labeled proteins were visualized by fluorescent imaging. The band labeled with an asterisk corresponds to a non-specific target of MV151 that could not be competed for with any available proteasome inhibitor and is thus unlikely to be a proteasome subunit. Raw images of the blots and gels are shown in S1 Raw images.
Discussion
Development of new antimicrobial drugs depends heavily on the use of suitable animal models for in vivo efficacy testing [40]. Robust infections with the relevant pathogens for a sufficiently long duration are key features of a good model. For trichomoniasis, infection of amenable animals, particularly rodents such as mice, with the causative pathogen, T. vaginalis, would be desirable, but we found here that murine vaginal infections with any of four T. vaginalis strains did not lead to detectable infection. In contrast, different strains of another trichomonad, T. foetus, were found to reliably infect mice for at least one week. These findings indicate that vaginal T. foetus infections provide a robust murine model suitable for in vivo drug testing.
The striking difference in infectivity between T. vaginalis and T. foetus is unlikely to be related to particular strains of the respective pathogen, since the findings were consistent for four diverse strains of each species. General characteristics of the two trichomonad species are probably more important. For example, we found a difference in doubling time and pH tolerance between the two trichomonad species. Faster growth rates of T. foetus might promote infectivity by counteracting inevitable vaginal losses, while the slower growth of T. vaginalis may not be sufficient to make up for such losses. Consistent with this notion, highly infectious organisms tend to have faster doubling times due to optimization of codon usage bias and increased copies of rRNA and tRNA [42]. Another potential explanation for differential infectivity could be differences in pH tolerance. The murine vagina, unlike the normally acidic environment of the human vagina, has a pH close to neutral or even slightly alkaline [43]. The ability of T. foetus to grow under alkaline pH conditions may promote vaginal infection, while in vitro growth of T. vaginalis was strongly inhibited by alkaline pH and may thus also be compromised in vivo in mice, although our measurements that the murine vaginal pH ranged from 6.5–7 would not suggest that differential pH tolerance can explain differential murine vaginal infectivity of T. foetus vs T. vaginalis.
Other possible explanations are conceivable for differential infectivity. For example, the estrus cycle of mice is characterized by periodic influx of leukocytes [44]. T. foetus may be more resistant to killing by polymorphonuclear leukocytes compared to T. vaginalis [28]. While the presence of vaginal leukocytes has no apparent impact on infectivity for T. foetus, it has been shown to have a negative impact on T. vaginalis infection [28]. Furthermore, the vaginal microbiota may be differentially affected by trichomonad infection and may impact the interactions of different trichomonads with the host. For example, T. vaginalis infection reduces vaginal epithelia-associated Lactobacillus species in humans [45]. In cats, T. foetus infection significantly impacts the abundance of a variety of different phyla present in the fecal microbiota [46]. Perhaps the extent and manner of microbial changes and their impact on host susceptibility to infection in mice may be more conducive to vaginal T. foetus infection.
Other studies had some success with vaginal infection of mice with T. vaginalis, although rates of infection were generally low and variable [28–30]. Beyond specific mechanistic explanations for this variable infectivity, it is conceivable that differential infectivity may be related to the host ranges of the two trichomonads. T. vaginalis is strictly a human parasite [47], while T. foetus can infect cows, cats, and pigs [48]. The apparently narrow host range of T. vaginalis may indicate a requirement for particular host features that are not present in other species, while the broader host range of T. foetus may indicate that infection depends less on particular host characteristics. It is also important to note that none of the prior work, contrary to our present study, had assessed infection by direct trichomonad counts but relied on cultures of vaginal washes. It is possible that cell numbers were too low for counting but sufficient for obtaining positive cultures, which are likely to be one or two logs more sensitive than counts. It is also possible that different conditioning or pretreatment protocols, such as multiple subcutaneous estradiol injections, or co-infection with lactobacilli could have sensitized mice to vaginal T. vaginalis infection beyond what we employed in the current work [28–30]. Finally, although we tested four different stains, differences in T. vaginalis strains could still matter for murine infectivity, so it remains possible that a heretofore untested strain can, in fact, infect mice robustly.
For a microorganism to be a useful surrogate model for drug development against a related but different infectious agent, both must have similar responses to the drugs of interest. We observed such congruency in the drug susceptibility of T. foetus and T. vaginalis to 5-nitroheterocyclic drugs and gold(I) compounds. Both drug classes are thought to act on a broad range of targets [49, 50]. In the case of 5-nitroheterocyclic drugs, the prodrug forms are first reductively activated, then form covalent adducts on multiple target molecules, including different proteins as well as nucleic acids [51]. The molecular targets responsible for cytocidal activity of 5-nitroheterocyclic drugs have not been defined, but it is likely that inactivation of multiple targets has synergistic effects that irreversibly damage the target cells [49]. The mechanism of action of gold(I) compounds is not well understood but also involve adduct formation and likely inactivation of multiple targets [50]. The multiplicity of cellular targets for both drug classes can provide an explanation for their similar activities against T. foetus and T. vaginalis, since even differences in the sequence or structure of any particular target are unlikely to interfere with the overall effectiveness of these drugs. It must be noted, however, that while not statistically significant, T. foetus had generally lower pIC50s for each compound of the two drug classes in the in vitro screening assays, which was also observed in prior studies with fewer compounds [22, 40]. Another explanation for differential drug susceptibility could be differential membrane permeability between the two parasites [52], which may allow for increased passive diffusion of small molecules, including potentially the tested antimicrobial drugs whose molecular masses are <800 Da, and thus greater susceptibility of T. vaginalis.
Notably, we found little or no activity differences in the MLC assays, perhaps suggesting that the IC50 growth assays and MLC survival assays test slightly different drug properties or are differentially impacted by parasite characteristics not directly related to the drug action. For example, the IC50 assays could be affected by the faster growth rate of T. foetus over T. vaginalis. Drugs such as 5-nitroheterocyclic compounds that form covalent adducts may be consumed during the culture period, so a faster growing organism may deplete drugs more rapidly, thereby requiring apparently higher IC50 values to achieve 50% maximal growth inhibition. By comparison, MLC assays measure the ability of a drug to kill all target organisms, which typically requires higher drug levels, so the impact of drug consumption would be less apparent. Nonetheless, both IC50 and MLC provide valuable information for comparative assessment of the drug impact of T. foetus and T. vaginalis, and clearly show that the drug susceptibility of T. foetus closely resembles T. vaginalis in regard to 5-nitroheterocyclic and gold(I) compounds. Consequently, our demonstration of in vivo efficacy for one of the gold(I) compounds is promising for its ultimate in vivo efficacy against T. vaginalis.
Contrary to the 5-nitroheterocyclic and gold(I) compounds, we observed marked differences in the activity of proteasome inhibitors against T. foetus and T. vaginalis. Thus, T. foetus had significantly lower pIC50s for all tested drugs in this class compared to T. vaginalis. The target of these compounds is well defined and involves one or several of the three catalytic subunits of the proteasome [53, 54]. Differences in the sequence or fine structure of the relevant proteasome subunits between the two trichomonad species, which clearly exist based on our phylogenetic analysis, can probably impact the binding of specific inhibitors. In this context, it is important to note that the tested inhibitors were originally developed against the human proteasome [55] but not specifically selected for activity against either T. vaginalis or T. foetus. Screening of large proteasome inhibitor libraries may well identify compounds that are equally active against both trichomonads. It is therefore possible, that inhibitors could be found that are more active against T. foetus than T. vaginalis and even human cells, which could represent a promising starting point for new veterinary drugs against bovine tritrichomoniasis [56]. Nevertheless, based on the phylogenetic and functional analyses, we were able to conclude that no fundamental differences exist in the 20S proteasomes of T. foetus and T. vaginalis. We could also demonstrate that one proteasome inhibitor, bortezomib, has low micromolar or better activity against both T. foetus and T. vaginalis and can eliminate T. foetus infection in vivo, underlining that drug candidates with adequate activity against T. foetus can be active in vivo even if their relative potency in T. foetus does not closely mimic the findings in T. vaginalis.
While this study focused on female mice, it is also important to acknowledge that males are carriers of T. vaginalis and at times may suffer symptoms. Male animal models of T. vaginalis infection via urethral injection and anesthesia have been established in mice [47] and rats [57]. In future studies, it would be of interest to evaluate drug efficacy in these models for potential treatment of male trichomoniasis.
In summary, the T. foetus murine infection models described here have excellent utility as surrogate in vivo models for drug development campaigns for new drugs against T. vaginalis, particularly if multiple molecular drug targets are predicted. Caution should be exercised, however, for compounds with singular targets if those targets differ significantly between the two trichomonad species. In such situations, it would be advisable to conduct careful in vitro compound testing beforehand to confirm that drug activity is not significantly different between the two organisms or at last remains sufficiently high in T. foetus that in vivo efficacy testing is not likely to yield false negative results.
Supporting information
S1 Table. Activity of antimicrobial compounds against diverse T. foetus and T. vaginalis strains.
T. foetus and T. vaginalis cultures were incubated for 24 hours with a range of drug concentrations, ATP content was assayed as a measure of cell growth and viability, and pIC50 values were calculated from the resulting concentration-response curves. Data are shown as mean ± SE of three or more independent experiments. IC50 values were derived from the mean pIC50. The highest tested drug concentration was 20 μM, so lack of growth inhibition at that concentration is listed as IC50 >20 μM.
https://doi.org/10.1371/journal.pone.0308672.s001
(PDF)
Acknowledgments
BMH was supported in part by the UCSD Graduate Training Program in Cellular and Molecular Pharmacology.
References
- 1. Sinka K. The Global Burden of STIs. Clin Dermatol. 2023 Dec 23. pmid:38142791
- 2. Soper D. Trichomoniasis: under control or undercontrolled? Am J Obstet Gynecol. 2004 Jan 1;190(1):281–90. pmid:14749674
- 3. Wølner-Hanssen P, Krieger JN, Stevens CE, Kiviat NB, Koutsky L, Critchlow C, et al. Clinical Manifestations of Vaginal Trichomoniasis. J Am Med Assoc. 1989 Jan 27;261(4):571–6. pmid:2783346
- 4. Krieger JN. Trichomoniasis in men: old issues and new data. Sex Transm Dis. 1995;22(2):83–96. pmid:7624817
- 5. Mirmonsef P, Krass L, Landay A, Spear GT. The Role of Bacterial Vaginosis and Trichomonas in HIV Transmission Across The Female Genital Tract. Curr HIV Res. 2012 Apr;10(3):202–10.
- 6. Price MA, Miller WC, Kaydos-Daniels C, Hoffman IF, Chilongozi D, Martinson FE, et al. Trichomoniasis in Men and HIV Infection: Data from 2 Outpatient Clinics at Lilongwe Central Hospital, Malawi. J Infect Dis. 2004 Oct 15;190(8):1448–55. pmid:15378437
- 7. Zhang ZF, Graham S, Yu SZ, Marshall J, Zielezny M, Chen YX, et al. Trichomonas vaginalis and cervical cancer: A prospective study in China. Ann Epidemiol. 1995 Jul 1;5(4):325–32.
- 8. Cotch MF, Pastorek JGI, Nugent RP, Hillier SL, Gibbs RS, Martin DH, et al. Trichomonas vaginalis Associated With Low Birth Weight and Preterm Delivery. Sex Transm Dis. 1997 Jul;24(6):353.
- 9. Gopalkrishnan K, Hinduja IN, Kumar TC. Semen characteristics of asymptomatic males affected by Trichomonas vaginalis. J In Vitro Fert Embryo Transf. 1990 Jun;7(3):165–7.
- 10. Kissinger P. Epidemiology and Treatment of Trichomoniasis. Curr Infect Dis Rep. 2015 Apr 30;17(6):31. pmid:25925796
- 11. Fung HB, Doan TL. Tinidazole: a nitroimidazole antiprotozoal agent. Clin Ther. 2005 Dec;27(12):1859–84. pmid:16507373
- 12. Löfmark S, Edlund C, Nord CE. Metronidazole is still the drug of choice for treatment of anaerobic infections. Clin Infect Dis. 2010 Jan 1;50 Suppl 1:S16–23. pmid:20067388
- 13. Lossick JG. Single-dose metronidazole treatment for vaginal trichomoniasis. Obstet Gynecol. 1980 Oct;56(4):508–10. pmid:7422197
- 14. Hager WD, Brown ST, Kraus SJ, Kleris GS, Perkins GJ, Henderson M. Metronidazole for Vaginal Trichomoniasis: Seven-Day vs Single-Dose Regimens. J Am Med Assoc. 1980 Sep 12;244(11):1219–20.
- 15. Aubert JM, Sesta HJ. Treatment of vaginal trichomoniasis. Single, 2-gram dose of metronidazole as compared with a seven-day course. J Reprod Med. 1982 Dec;27(12):743–5. pmid:7161754
- 16. Kirkcaldy RD, Augostini P, Asbel LE, Bernstein KT, Kerani RP, Mettenbrink CJ, et al. Trichomonas vaginalis Antimicrobial Drug Resistance in 6 US Cities, STD Surveillance Network, 2009–2010. Emerg Infect Dis. 2012 Jun;18(6):939–943. pmid:22608054
- 17. Schmid G, Narcisi E, Mosure D, Secor WE, Higgins J, Moreno H. Prevalence of Metronidazole-Resistant Trichomonas vaginalis in a Gynecology Clinic. Obstet Gynecol Surv. 2001 Nov;56(11):693.
- 18. Schwebke JR, Barrientes FJ. Prevalence of Trichomonas vaginalis Isolates with Resistance to Metronidazole and Tinidazole. Antimicrob Agents Chemother. 2006 Dec;50(12):4209–10.
- 19. Ayinde O, Ross JD. The frequency and duration of side-effects associated with the use of oral metronidazole; A prospective study of VITA trial participants. Int J STD AIDS. 2023 Oct 1;34(12):897–902. pmid:37395087
- 20. Kissinger P, Muzny CA, Mena LA, Lillis RA, Schwebke JR, Beauchamps L, et al. Single-dose versus 7-day-dose metronidazole for the treatment of trichomoniasis in women: an open-label, randomised controlled trial. Lancet Infect Dis. 2018 Nov;18(11):1251–9. pmid:30297322
- 21. Goodhew EB, Secor WE. Drug library screening against metronidazole-sensitive and metronidazole-resistant Trichomonas vaginalis isolates. Sex Transm Infect. 2013 Sep 1;89(6):479–84. pmid:23794105
- 22. O’Donoghue AJ, Bibo-Verdugo B, Miyamoto Y, Wang SC, Yang JZ, Zuill DE, et al. 20S Proteasome as a Drug Target in Trichomonas vaginalis. Antimicrob Agents Chemother. 2019 Oct 22;63(11): pmid:31451503
- 23. Coux O, Tanaka K, Goldberg AL. Structure and Functions of the 20s and 26s Proteasomes. Annu Rev Biochem. 1996;65(1):801–47. pmid:8811196
- 24. Kisselev AF, Akopian TN, Castillo V, Goldberg AL. Proteasome Active Sites Allosterically Regulate Each Other, Suggesting a Cyclical Bite-Chew Mechanism for Protein Breakdown. Mol Cell. 1999 Sep 1;4(3):395–402. pmid:10518220
- 25. Raedler L.Velcade (Bortezomib) Receives 2 New FDA Indications: For Retreatment of Patients with Multiple Myeloma and for First-Line Treatment of Patients with Mantle-Cell Lymphoma. Am Health Drug Benefits. 2015 Mar;8(Spec Feature):135–40. pmid:26629279
- 26. Reynolds JM, El Bissati K, Brandenburg J, Günzl A, Mamoun CB. Antimalarial activity of the anticancer and proteasome inhibitor bortezomib and its analog ZL3B. BMC Clin Pharmacol. 2007 Oct 23;7:13. pmid:17956613
- 27. Tschan S, Brouwer AJ, Werkhoven PR, Jonker AM, Wagner L, Knittel S, et al. Broad-Spectrum Antimalarial Activity of Peptido Sulfonyl Fluorides, a New Class of Proteasome Inhibitors. Antimicrob Agents Chemother. 2013 Aug;57(8):3576–84. pmid:23689711
- 28. Cobo ER, Eckmann L, Corbeil LB. Murine Models of Vaginal Trichomonad Infections. Am J Trop Med Hyg. 2011 Oct 1;85(4):667–73. pmid:21976570
- 29. McGrory T, Garber GE. Mouse intravaginal infection with Trichomonas vaginalis and role of Lactobacillus acidophilus in sustaining infection. Infect Immun. 1992 Jun;60(6):2375–9.
- 30. Abraham MC, Desjardins M, Filion LG, Garber GE. Inducible immunity to Trichomonas vaginalis in a mouse model of vaginal infection. Infect Immun. 1996 Sep;64(9):3571–5.
- 31. Iriarte LS, Martinez CI, de Miguel N, Coceres VM. Tritrichomonas foetus Cell Division Involves DNA Endoreplication and Multiple Fissions. Microbiol Spectr. 11(2):e03251–22.
- 32. Skirrow SZ, BonDurant RH. Induced Tritrichomonas foetus infection in beef heifers. J Am Vet Med Assoc. 1990 Mar 15;196(6):885–9.
- 33. Brown DM, Upcroft JA, Dodd HN, Chen N, Upcroft P. Alternative 2-keto acid oxidoreductase activities in Trichomonas vaginalis. Mol Biochem Parasitol. 1999 Jan 25;98(2):203–14.
- 34. Miyamoto Y, Aggarwal S, Celaje JJA, Ihara S, Ang J, Eremin DB, et al. Gold(I) phosphine derivatives with improved selectivity as topically active drug leads to overcome 5-nitroheterocyclic drug resistance in Trichomonas vaginalis. J Med Chem. 2021 May 27;64(10):6608–20. pmid:33974434
- 35. Sule P, Wadhawan T, Carr NJ, Horne SM, Wolfe AJ, Prüβ BM. A combination of assays reveals biomass differences in biofilms formed by Escherichia coli mutants. Lett Appl Microbiol. 2009 Sep;49(3):299–304.
- 36. Gomes AV. Genetics of Proteasome Diseases. Scientifica (Cairo). 2013;2013:637629. pmid:24490108
- 37. Verdoes M, Florea BI, Menendez-Benito V, Maynard CJ, Witte MD, van der Linden WA, et al. A Fluorescent Broad-Spectrum Proteasome Inhibitor for Labeling Proteasomes In Vitro and In Vivo. Chem Biol. 2006 Nov 1;13(11):1217–26. pmid:17114003
- 38. Lehker MW, Alderete JF. Iron regulates growth of Trichomonas vaginalis and the expression of immunogenic trichomonad proteins. Mol Microbiol. 1992;6(1):123–32.
- 39. Meysick KC, Garber GE. Interactions between Trichomonas vaginalis and vaginal flora in a mouse model. J Parasitol. 1992 Feb;78(1):157–60.
- 40. Natto MJ, Hulpia F, Kalkman ER, Baillie S, Alhejeli A, Miyamoto Y, et al. Deazapurine nucleoside analogues for the treatment of Trichomonas vaginalis. ACS Infect Dis. 2021 Jun 11;7(6):1752–64. pmid:33974405
- 41. Nanda N, Michel RG, Kurdgelashvili G, Wendel KA. Trichomoniasis and its treatment. Expert Rev Anti Infect Ther. 2006 Feb 1;4(1):125–35. pmid:16441214
- 42. Vieira-Silva S, Rocha EPC. The Systemic Imprint of Growth and Its Uses in Ecological (Meta)Genomics. PLoS Genet. 2010 Jan 15;6(1):e1000808. pmid:20090831
- 43. Ganesan M, Kadalmani B. Phase Dependent Discrepancy in Murine Vaginal Micro-Enviroment: A Correlative Analysis of pH, Glycogen and Serum Estrogen Upon Exposure to Lapatinib Ditosylate. Int J Pharm Pharm Sci. 2016 Mar 15;8:404–7
- 44. McLean AC, Valenzuela N, Fai S, Bennett SAL. Performing Vaginal Lavage, Crystal Violet Staining, and Vaginal Cytological Evaluation for Mouse Estrous Cycle Staging Identification. J Vis Exp. 2012 Sep 15;(67):4389. pmid:23007862
- 45. Fichorova RN, Buck OR, Yamamoto HS, Fashemi T, Dawood HY, Fashemi B, et al. The Villain Team-Up or how Trichomonas vaginalis and bacterial vaginosis alter innate immunity in concert. Sex Transm Infect. 2013 Sep;89(6):460–6.
- 46. Bierlein M, Hedgespeth BA, Azcarate-Peril MA, Stauffer SH, Gookin JL. Dysbiosis of fecal microbiota in cats with naturally occurring and experimentally induced Tritrichomonas foetus infection. PLoS One. 2021 Feb 19;16(2):e0246957. pmid:33606740
- 47. Molgora BM, Mukherjee SK, Baumel-Alterzon S, Santiago FM, Muratore KA, Jr AES, et al. Trichomonas vaginalis adherence phenotypes and extracellular vesicles impact parasite survival in a novel in vivo model of pathogenesis. PLoS Negl Trop Dis. 2023 Oct 23;17(10):e0011693.
- 48. Dąbrowska J, Keller I, Karamon J, Kochanowski M, Gottstein B, Cencek T, et al. Whole genome sequencing of a feline strain of Tritrichomonas foetus reveals massive genetic differences to bovine and porcine isolates. Int J Parasitol Parasites Wildl. 2020 Mar 1;50(3):227–33.
- 49. Lamp KC, Freeman CD, Klutman NE, Lacy MK. Pharmacokinetics and Pharmacodynamics of the Nitroimidazole Antimicrobials. Clin Pharmacokinet. 1999 May 1;36(5):353–73. pmid:10384859
- 50. Saei AA, Gullberg H, Sabatier P, Beusch CM, Johansson K, Lundgren B, et al. Comprehensive chemical proteomics for target deconvolution of the redox active drug auranofin. Redox Biol. 2020 Mar 3;32:101491. pmid:32199331
- 51. Schmid A Schmid H. Pharmaco-Toxicological Mode of Action of Antimicrobial 5-Nitroimidazole Derivatives. J Vet Med A Physiol Pathol Clin Med. 1999;46(9):517–22. pmid:10605360
- 52. Lloyd D, Harris JC, Biagini GA, Hughes MR, Maroulis S, Bernard C, et al. The plasma membrane of microaerophilic protists: oxidative and nitrosative stress. Access Microbiol. 2004;150(5):1183–90. pmid:15133079
- 53. Khalesi N, Korani S, Korani M, Johnston TP, Sahebkar A. Bortezomib: a proteasome inhibitor for the treatment of autoimmune diseases. Inflammopharmacology. 2021 Oct;29(5):1291–306. pmid:34424482
- 54. Kisselev AF. Site-Specific Proteasome Inhibitors. Biomolecules. 2021 Dec 31;12(1):54. pmid:35053202
- 55. Dimopoulos MA, Richardson P, Lonial S. Treatment Options for Patients With Heavily Pretreated Relapsed and Refractory Multiple Myeloma. Clin Lymphoma Myeloma Leuk. 2022 Jul 1;22(7):460–73. pmid:35148975
- 56. Michi AN, Favetto PH, Kastelic J, Cobo ER. A review of sexually transmitted bovine trichomoniasis and campylobacteriosis affecting cattle reproductive health. Theriogenology. 2016 Mar 15;85(5):781–91. pmid:26679515
- 57. Jang KS, Han IH, Lee SJ, Yoo J, Kim YS, Sim S, et al. Experimental rat prostatitis caused by Trichomonas vaginalis infection. Prostate. 2019;79(4):379–89.