Tritrichomonas foetus trophozoites were maintained in trypticase-yeast extract-maltose (TYM) medium supplemented with adult bovine serum (10%) and 100X penicillin-streptomycin (1%) (Martin et al., 2021). Cultures were maintained at 35°C and regularly subcultured to maintain cell concentrations of approximately 4x10⁵ trophozoites/mL (Schaut et al., 2017). Cultures were maintained in sterile 15 mL centrifuge tubes filled completely with media and capped tightly to create an anaerobic environment. All experiments utilized IA-1 (field) strain parasite (Bader et al., 2016). Studies were conducted under all applicable local laws and approved by the Iowa State University Biosafety and IACUC committees under protocols 18–217 and 21-264.

Cross bred Holstein-Angus heifer calves (70 kg, 6–8 weeks of age) were purchased from commercial dairy operations in Iowa, USA. No twins or twin sibling calves were used in the study. Four independent experiments (A – D) were conducted, each at a different point in time. In experiments A and B, calves were group housed indoors in a controlled ambient environment. In experiments C and D, calves were group house in an outdoor environment. Experiment A included 3 heifer calves, experiments B and C included 6 heifer calves, and experiment D included 12 heifer calves. Experiment A utilized a minimal number of animals to form a basis for power analyses to determine the number of animals needed in subsequent studies. In subsequent experiments the proportion of animals infected and parasite counts were used to determine the number of animals required.

An overview of the procedural timeline for each experiment is shown in Figure 1 . Using a 15-gauge feeding needle attached to a syringe, heifers were infected with trophozoites (experiment A = 1x10⁸, B = 4.3x10⁷, C = 9.4x10⁸, D = 2x10⁸) in a total volume of 1.5–2 mL TYM delivered directly into the vaginal vault. Variation in the infectious dose was due to availability of parasite cultures at the time of each experiment. In experiment B calves with consecutive negative cultures were reinfected with 3.75x10⁷ (calf B.D) or 3.1x10⁸ (calves B.B and B.D) trophozoites. The rationale for reinfection was that it would not be worthwhile to sacrifice a negative animal for necropsy; such animals would not allow us to determine which parts of the reproductive tract were colonized.

To assess infection status, calves were sampled by swabbing the vaginal vault with a TYM-soaked cotton tip applicator (15cm) inserted and retracted 4 times while simultaneously rotating the applicator. The cotton tip applicator was placed into a microcentrifuge tube containing 1.5 mL TYM culture medium supplemented with Modified Preston Campylobacter selective supplement (Oxoid LTD, Hampshire, England).

Reproductive tract swabs were placed in TYM culture media, maintained at 35°C and observed microscopically every 24 hrs until motile T. foetus trophozoites were observed. Cultures were considered negative if no live organisms were observed within 7 days. Live organisms were identified morphologically. Figure 2 provides an example of a typical trophozoite as observed in this study. Viable organisms demonstrated forward motion. Positive cultures were verified as needed by PCR described below.

Aliquots of all samples from experiment A vaginal swabs were submitted to the Iowa State University Veterinary Diagnostic Laboratory for PCR testing. In subsequent studies (B and D), culture positive samples were subjected to confirmatory PCR. DNA was extracted from 500 µL aliquots of culture media inoculated with sample swab using the QIAmp DNA Mini kit (Qiagen), in accordance with manufacturer instruction for cells in suspension. Conventional PCR using published primer pair TFR3/TFR4 was performed ( Table 1 ) (Felleisen et al., 1998). Negative culture samples were used as a negative control in these reactions. PCR was carried out in a 20 μL volume with 2 μl DNA template, 0.5 µM forward and reverse primers, and 1x Platinum™ SuperFi™ II PCR Mastermix (Invitrogen). PCR conditions consisted of initial denaturation at 98°C for 30s, followed by 30 cycles consisting of 10s denaturation (98°C), 10s annealing (60°C), 1 min elongation (72°C); final elongation was carried out for 5 min (72°C). Agarose gel electrophoresis was performed to confirm amplification of bands of the expected size (~350 base pairs). Culture material from experiment C was utilized for transcriptomic analysis used in a separate study; PCR was not performed on samples from this experiment.

In experiment A, DNA was also extracted from fresh vaginal, cervical, and uterine tissue samples collected at necropsy using the Qiagen DNeasy Blood and Tissue kit (Valencia, CA) according to manufacturer’s instructions for tissue samples. An approximately 700-bp fragment of T. foetus cysteine protease 2 (TfCP2) was amplified using the published primer pair S0295/S0296 ( Table 1 ) (Šlapeta et al., 2012). PCR was carried out in a 25 μL volume with 2 μl DNA template, 1x PCR buffer (GoTaq Flexi buffer, Promega, Madison WI), 3 mM MgCl2, 0.2 mM dNTPs, 0.5 μM of each primer, and 2 U of Taq polymerase (GoTaq Flexi, Promega, Madison WI). PCR conditions were as follows: initial denaturation at 95°C for 5 min followed by 35 cycles of 15s denaturation (95°C), 15s annealing (55°C), and 30s elongation (72°C), with a final elongation step of 5 min at 72°C. Agarose gel electrophoresis was performed to visualize amplicons.

Samples of vaginal vault, cervix, and uterus were collected from calves at necropsy. Samples were fixed in 10% neutral buffered formalin (NBF), embedded in paraffin blocks, and processed through routine histologic methods (Chelladurai et al., 2021). Samples were H&E stained with Signature Series™ hematoxylin (Fisher Scientific) for five min and Richard-Allan Scientific™ eosin (Fisher Scientific) for one minute. Slides from group D were evaluated by an ACVP boarded pathologist (TAH). Group D was selected for histopathologic review because these animals were subject to fewer sampling instances, reducing the possibility of iatrogenic inflammation. Gram and Grocott’s methenamine silver (GMS) staining were performed on all tissues to rule out bacterial and fungal causes of inflammation.

Immunohistochemistry was performed to investigate the presence of galectin-1 on vaginal, cervical, and uterine tissue samples. Bovine trachea and lung, known to express galectin-1, were used as positive controls (Kaltner et al., 2002). For all IHC procedures, five-micrometer sections on positively charged slides were utilized. Antigen retrieval was performed using Decloaker (Biocare Medical, DV2004) in a decloaking chamber. Tissues were blocked with Peroxidase Blocking Reagent (Biocare Medical, IPB5000G20) for 5 min and Background Punisher (Biocare Medical, IP974G20) for 5 min. Galectin-1 Recombinant Monoclonal antibody, JM13-37 (Invitrogen, catalog number MA5-32779) was applied at a dilution of 1:400 in DaVinci Green (Biocare Medical, PD900H) with a 2-hour incubation. Rabbit-on-Canine HRP Polymer (Biocare Medical, RC542L) was used as a secondary with a 30-minute incubation. Signal was visualized using DAB Chromagen Kit (Biocare Medical, IPK5010) incubated on tissues for 5 min. Tissues were counterstained with a 1:10 dilution of CAT Hematoxylin (Biocare Medical, CATHE-M) for 5 min, dehydrated in a series of graded ethanol, and coverslipped.

Heifer calves remained infected for as long as 6 weeks (42 days), following a single inoculation ( Figure 3 ). Since calves were sacrificed at this point, the total possible duration of infection in juvenile heifers is unknown. Results from the present series of experiments show that at least 60% of prepubescent heifer calves are culture positive at 4 weeks post inoculation, when experiments were carried out through this timepoint. Additionally, in the longer-term, single inoculation experiment (C), 5/6 calves were still culture positive at the 6-weeks post infection. Because samples were incubated for up to 7 days to determine infection status, enrichment time varied among samples, therefore, comparing the proportion of culture positive animals is more meaningful than the number of trophozoites recovered ( Figure 3 ). The duration of experiments in this study varied. The rationale was that after we confirmed we could infect heifers with the parasite, we wished to verify specific parts of the reproductive tract that were colonized. As the duration of the infection became apparent, subsequent experiments were allowed to proceed for a longer timeframe. The rationale for not observing the time needed to clear the infection as is seen in naturally infected animals is that the location of parasites in the reproductive tract could not be determined if the host eliminated the infection.

Vaginal, uterine, and cervical tissue samples collected from the calves in Experiment A were PCR positive for T. foetus using TfCP2 primers ( Table 2 ). In experiment C, cultures collected from the lumen of the uterus of 3/6 calves at necropsy demonstrated live organisms, confirming that viable organisms are colonizing beyond the site of inoculation (vaginal vault), traveling through the cervix, and colonizing the body of the uterus. Uterine cultures were not collected in experiments B and D.

Tissues (vaginal vault, cervix, and uterus) from all calves in experiment D were evaluated for the following criteria: epithelial degeneration, epithelial proliferation, intraluminal exudate, inflammation severity, inflammation distribution, hemorrhage, and edema. A total histologic score (THS) was determined based on the sum of individual criteria scores for each tissue ( Table 3 ). Representative lesions are shown in Figure 4 . Moderate histopathologic changes were observed in the vaginal sections of 5/6 (83.3%) of calves. These changes were characterized by scant to small amounts of fibrinosuppurative intraluminal exudate, moderate epithelial attenuation to focal ulceration, lymphoplasmacytic to suppurative vaginitis, and moderate scattered submucosal lymphoplasmacytic nodules and perivascular cuffs. The remaining calf (D.B) had similar but less severe histopathologic lesions in the vagina. Mild histopathologic changes were present in the cervical sections of 4/6 (66.7%) calves. These changes were characterized by mild epithelial attenuation, mild lymphoplasmacytic cervicitis, and scattered lymphoplasmacytic nodules. The remaining calves had similar but more moderate inflammatory lesions. Mild histopathologic changes were present in the uterine sections of 4/6 (66.7%) calves. These changes were characterized by mild epithelial attenuation, mild superficial endometritis, and increased lymphoplasmacytic to lymphohistiocytic endometrial nodules. Of the remaining calves, one had moderate changes, and one had severe changes in the uterus that contained similar inflammatory infiltrates. All sections were negative for GMS staining, indicating the absence of fungal organisms. Two calves (D.A and D.F) had rare gram-positive cocci in luminal mucus of the uterus (D.A) and cervix (D.F). The presence and character of this bacteria was consistent with normal flora of the mucus layer.

Immunohistochemistry staining confirmed the presence of galectin-1, a putative host cell receptor for trichomonads throughout the epithelium and submucosa of the vagina, cervix, and uterus of prepubescent heifer calves ( Figure 5 ). This is the first report of galectin-1 expression in these tissues in prepubescent heifers.

Following infection with trophozoites directly in the vaginal vault, a majority of calves were consistently culture positive four weeks post-inoculation. Importantly, viable organisms were also present in cultures from the uterine lumen which indicated successful growth and survival. Interestingly, our results show that even in a prepubescent animal, trophozoites are able to penetrate through the cervix, whereas sperm cells are unable to do so when a cow is in a period of anestrus (Nabors and Linford, 2015). Significantly, the susceptibility of heifers in our study demonstrated that hormonally driven changes in the female reproductive tract are not a prerequisite for infection. Experimental infection studies in mature heifers and cows typically involve hormone synchronization to ensure animals are in the same stage of the estrous cycle (Stockdale et al., 2007; Fuchs et al., 2017; Ortega-Mora et al., 2022). The use of prepubescent animals removes hormonal influence and the risk of failure of synchronization attempts as experimental variables. Taken together, the series of experiments presented herein demonstrate the ability of prepubescent heifer calves to maintain T. foetus infections for at least 6 weeks without the need for exogenous hormones or immunosuppressive treatments. Our study reveals the exciting potential of a natural host calf model to study the biology of the parasite.

Calves were not culture positive at all time points during the study. While in experimental context, parasite culture has the benefit of confirming the presence of viable organisms, one drawback to the testing method is reduced sensitivity compared to PCR. Infection rates in this study are consistent with those reported in post-pubertal heifer infection challenge studies. Anderson et al. reported that of 6 heifers challenged with T. foetus, 4 were culture positive at weekly intervals, with 2 failing to produce a positive culture result (BonDurant et al., 1993; Anderson et al., 1996). Additionally, previous studies have also noted that challenged heifers may go between culture negative and positive status from week to week, meaning that a single negative culture may not reflect the true infection status (BonDurant et al., 1993). This is broadly accepted and the reasoning for the standard requirement of three, consecutive weekly negative culture samples prior to permitting the movement of bulls across state lines where T. foetus movement regulations are in place (Schroeder et al., 2023).

This is the first evaluation of microscopic lesions caused by T. foetus in prepubescent heifers. Histopathologic evaluation revealed lesions and cellular infiltrates consistent with findings in previous studies in sexually mature cattle, including the presence of lymphoid nodules and lymphoplasmacytic infiltrates (Parsonson et al., 1976; Anderson et al., 1996). The most recent description of histopathologic changes associated with T. foetus infections in cattle was observed nearly 30 years ago (Anderson et al., 1996). Gram and GMS staining confirmed that inflammation was not the result of fungal or bacterial pathogens and therefore supported the hypothesis that the observed lesions and inflammation were due to T. foetus. Observing similar microscopic lesions in calves and sexually mature cattle further supports the utility of our infection model.

The presence of galectin-1 throughout the reproductive tract of prepubescent is a novel finding. This lectin is known to be involved in adhesion of the closely related human pathogen, Trichomonas vaginalis, to host cells via parasite lipophosphoglycan (LPG) (Okumura et al., 2008; Ryan et al., 2011). It is known that lipophosphoglycan moieties present on Tritrichomonas foetus are involved in the adhesion process, but the specific host cell receptors remain elusive (Singh et al., 1999). The present study demonstrated galectin-1 expression in the reproductive tract of 6-week-old heifer calves but did not determine if it was essential for parasite adhesion. A recent study also confirmed that galectin-1 was expressed in the penis and prepuce of prepubescent bull calves (Martin et al., 2025). Further investigation is needed to determine if galectin-1 is essential for adhesion T. foetus infection in cattle.

There was a degree of biological variation among our 4 experiments. For example, the proportion of culture positive calves 4 weeks post-inoculation in experiment D was lower than experiments B and C, with 60% (experiment D) and 83% (experiments B and C) culture positive calves at this timepoint. It is important to note that six out of twelve calves in experiment D experienced clinical signs unrelated to T. foetus and consistent with Mycoplasma infection (fever, nasal discharge, head tilt). These animals were treated with antibiotics (tulathromycin) and NSAIDS (meloxicam) and their symptoms subsequently resolved. Three of the six calves treated for Mycoplasma were T. foetus culture negative 4 weeks post-inoculation, accounting for 75% of the culture negative calves in experiment D. Five out of six calves without Mycoplasma were culture positive at 4-weeks post inoculation, consistent with the results of experimental groups B and C. We suspect that pyrexia, and not antibiotic treatment, led to reduced infection rates/durations in this experimental group since there is no evidence T. foetus is susceptible to tulathromycin and T. foetus has poor antimicrobial susceptibility (Sui et al., 2024). Prophylactic antibiotic treatment for pre-existing bacterial infections, prior to parasite inoculation, may be indicated to ensure optimal infection with T. foetus in future studies.

The primary constraints to interpretation of the present report are variations in methods across and within experimental groups. Many of the variations were due to unplanned events and are inherently related to the logistics of conducting studies in cattle in a research environment. Such studies are subject to sudden changes as well as long term planning efforts due to extended gestation periods, fluctuations in cash markets, availability of facilities, and unexpected concomitant disease. These factors tend to be more difficult to control in cattle compared to the level of control that is possible for environmental control in rodent studies. Herein, we have summarized a collection of experiments carried out across a period of several years. While some variables vary slightly across individual experiments, we aimed to present all experiments together in a single cohesive report rather than deliver our results in a patchwork of publications which could be more difficult to interpret.

Experiment A served as the pilot study for this project and was planned to be short-term in order to determine the feasibility of pursuing future experiments. Experiment B aimed to determine the ability of the calf model to maintain longer term infections to demonstrate the potential of the model for use in treatment and vaccine studies. In experiment B the decision to reinfect animals following multiple negative culture results was made in an effort to increase the number of animals that were culture positive at necropsy. Consistent with natural infection, culture and PCR results can lead to intermittent positives, and this was observed in some studies following a single inoculation ( Supplementary Figure S1 ). While we were aware of this phenomenon, we elected to re-inoculate selected calves in experiment B. Our intention was to maximize the number of tissues from culture positive animals which could be utilized for downstream applications such as histology. Although this was not ideal, attempting to ensure all calves were positive at necropsy was necessary to decrease the overall number of animals used in our studies. Experiments C and D were more consistent in approach with regularly spaced sampling time points. The number of animals in experiment D (n=12) necessitated two separate days for necropsy and sample processing. Ideally, the endpoint of the experiment would have been the exact same for all animals.

It should be noted that we did not use the exact same number of trophozoites across all four experiments. The underlying rationale was that we used a vast excess of trophozoites and we used the maximum number of log-phase parasites available to us at the time cohorts of calves became available. When growing trophozoites we seeded culture medium with consistent numbers of trophozoites from the same passage of liquid nitrogen seed stocks, variation in cell revival and replication led to differences in the number of organisms available on planned inoculation days. Due to facility availability and scheduling, we elected to move forward with inoculation rather than attempt to grow additional trophozoites. We did, however, ensure that parasites were in log-phase of growth with vigorous motility prior to infection. The minimum number of T. foetus trophozoites under natural conditions is not known. However, insights regarding infectious dose could be further resolved using our model since it would be unlikely to determine the total number of organisms inhabiting the prepuce of a bull under natural conditions. In our case, for ethical reasons, we elected not to sacrifice additional animals for this purpose alone.

Housing for the calves in this study varied among the four experiments. Experimental groups A and B were housed indoors while groups C and D were house outdoors. Housing decisions were based in part on facility availability as well as a preference for outdoor housing after pilot studies demonstrated the tractability of this model under more controlled conditions. Tritrichomonas foetus is not known to survive outside the reproductive tract, so environmental contamination is highly unlikely to have influenced our results. Likewise, the cervix of prepubertual heifers is closed, providing a relatively sterile uterine environment. Histological examination of tissues did not reveal additional microbial infection that could have been introduced from the environment. Studies aiming to study the impact of environmental conditions, co-infections, and the vaginal microbiome could be made possible using our model.

In addition to the benefits of this model for advancing the study of T. foetus in the bovine host, this model may also be beneficial in the study of Trichomonas vaginalis, the cause of human trichomoniasis. Trichomoniasis is the most common non-viral sexually-transmitted infection globally (Van Der Pol, 2007). Currently, human trichomoniasis research relies on a T. foetus infected mouse model (Nieskens et al., 2024). The calf infection model would allow for inferences to be made in a natural host, rather than a rodent model. Natural host models can provide a more accurate representation of the disease and more physiologically relevant information than rodent models. Cattle are known to be good models for human reproductive system research including research regarding placental insufficiency and ovarian function (Pramod et al., 2025). In female T. vaginalis infections, inflammation of the reproductive tract is characterized by the presence of lymphocytes and macrophages, similar to the findings in this study and other T. foetus infection studies (Edwards et al., 2016). Histologically, the human and bovine uterus share the presence of simple columnar epithelium with ciliated and secretory cells dispersed throughout the epithelium (Gálfiová et al., 2022; Senn et al., 2024). The dominance of Lactobacilli spp. in the female reproductive tract is unique to humans and would present one disparate factor in the bovine model (Miller et al., 2016; Gupta et al., 2024). Comparative pathology studies are needed to evaluate the utility of a T. foetus calf model as a surrogate for human trichomoniasis and confirm advantages over the currently used mouse infection model.

Our studies also highlighted the need to optimize a method for quantitating severity of T. foetus infection. While we made efforts to collect samples in a similar manner, parasite culture in TYM led to trophozoite counts that varied significantly among positive animals (e.g., 1 vs 1,000,000 trophozoites/mL). Based on the methods employed in this study, it is unclear if the observed level of variation reflects biological differences among infected calves or is simply related to the limits of our sampling methods. Swab depth, duration, and number of rotations could all impact the quality of the sample and should be evaluated. Clinically, status of infection is simply characterized as positive or negative, and the importance of trophozoite numbers on disease severity or progression is not established. However, quantification of infection is important for assessing success of parasitological studies or interventions such as treatment and vaccination. Possible methods to achieve this could include cycle threshold (CT) values for qPCR, lesion scoring, or quantification of T. foetus antigen via ELISA. While some of these methods have been used to detect parasites from naturally infected animals, their use in quantification of disease is not well-established. Future studies in our model will aim to investigate infectious dose as well as quantification of parasite burden with lesion severity.

Our findings will also enable us to address new questions regarding characterization and optimization of a heifer calf model for bovine trichomonosis. Future studies should aim to evaluate the duration of infection, as the results of these studies did not extend beyond 6 weeks. With 83% of calves remaining culture positive at 6 weeks, it is unclear if heifer calves can maintain infections far beyond this timepoint. While the paradigm is that cows are only infected transiently, there are some reports of persistently infected cows (BonDurant, 2005). The present study utilized a single, high-dose inoculum. The optimal inoculation regimen and dose should be explored to aid in the optimization of this model. Our experiments relied on a single field strain (IA-1), of T. foetus. Ideally, infection challenges with other parasite strains should be performed and whereby infection dynamics of various strains could be compared. Lastly, these experiments utilized cross-bred dairy calves. It is possible that differences in breed susceptibility exist, and thus experimental infection of additional cattle breeds could be explored to aid in model optimization. While some optimization is needed, the success of our experiments underscores a new ability to answer research questions that were previously difficult to obtain.

In this study, we tested the hypothesis that heifer calves are a suitable natural host model for bovine trichomonosis and challenged the dogma that estrus and ovulation are required for infection of the bovine female reproductive tract with T.foetus. The results of our series of experiments supported our hypothesis and demonstrate significant promise of the heifer calf model. The ability of calves to reliably maintain infections for at least 6 weeks with viable organisms colonizing and replicating in the reproductive tract is a significant finding. Our natural host model fills a significant knowledge gap and allows for further understanding of the parasite-host relationship, as well as providing a straightforward method to assess novel treatments, vaccines, and diagnostic platforms.