Introduction

Trichomonas vaginalis is a flagellated protozoan that infects the human urogenital tract, leading to trichomoniasis, the most common non-viral sexually transmitted infection worldwide. According to the World Health Organization, approximately 276 million new cases are diagnosed each year1. Infected women often experience symptoms such as vaginal discharge, pain, and inflammation. In contrast, most infected men remain asymptomatic, with only a small percentage developing urethritis or prostatitis1. In addition to its clinical symptoms, infection with T. vaginalis is linked to an increased risk of acquiring human immunodeficiency virus (HIV), persistence of human papillomavirus (HPV), and the development of cervical cancer2. Although trichomoniasis is not directly fatal, its significance in epidemiology highlights the necessity for ongoing research.

The vaginal microbiota forms a key defense against pathogens by maintaining mucosal homeostasis and preventing colonization through mechanisms such as competitive exclusion and immune modulation3. The vaginal community has been classified into five community-state types (CSTs). CST I, II, and V, dominated by Lactobacillus crispatus, L. gasseri, and L. jensenii, are protective, whereas CST III and CST IV, characterized by L. iners and diverse anaerobes, respectively, are linked to dysbiosis and infection4,5. In healthy women, Bifidobacterium breve is also abundant, particularly in CST I and II6,7.

B. breve is a Gram-positive anaerobe commonly found in the gut, vagina, and oral cavity8. As a probiotic commensal bacterium, it promotes epithelial barrier integrity, modulates immune responses, and produces lactic acid, which helps maintain acidic vaginal pH and suppresses pathogen overgrowth9,10,11. Lactic acid not only inhibits bacteria associated with bacterial vaginosis but also regulates epithelial function and reduces susceptibility to infections such as Chlamydia trachomatis. Notably, the abundance of B. breve significantly decreases during T. vaginalis infection12. This reduction suggests a disruption of normally protective microbial members during parasite overgrowth. However, despite its well-established beneficial roles in vaginal homeostasis, the influence of B. breve on T. vaginalis biology and the mechanisms underlying their interaction remain unclear. This knowledge gap highlights the importance of characterizing how B. breve affects parasite physiology and its potential contribution to microbial community shifts during infection.

The interplay between vaginal microbes and T. vaginalis can profoundly affect infection outcomes. Lactobacillus gasseri reduces host cell damage by decreasing the adhesion of T. vaginalis onto host epithelium, and the secreted products exhibit parasiticidal activity against trichomonad infection13,14. Similarly, bacterial extracellular vesicles (EVs), which carry proteins, nucleic acids, lipids, and metabolites15, modulate parasite adhesion and host cell interaction. EVs from commensal bacteria promote host defense16,17, whereas those from BV-associated bacteria, such as G. vaginalis, enhance parasite internalization15. Furthermore, both CST-IV bacteria and T. vaginalis compromise tight junction integrity, increasing paracellular permeability, and disrupting the epithelial barrier18,19,20,21,22,23.

In addition to microbial interactions, lipid metabolism is vital for protozoan survival and virulence. Fatty acids are essential structural components of membranes, energy sources, and signaling molecules24. Fatty acid synthesis (FAS) pathways vary among protozoa, with some species utilizing FAS-I or FAS-II systems25,26,27. Fatty acids also participate in host-pathogen interactions; for instance, Trypanosoma cruzi manipulates host lipid signaling to facilitate invasion28,29. Unlike other protozoa, T. vaginalis lacks a de novo FAS pathway and depends on exogenous fatty acids, linking its pathogenicity to host lipid availability30. Given that commensal bacteria can influence local lipid availability and metabolism, they may indirectly shape T. vaginalis physiology.

In light of the essential roles of vaginal microbial communities and parasite lipid metabolism in shaping T. vaginalis infection dynamics, elucidating how B. breve influences parasite behavior is of particular importance. This study therefore investigates the interactions between B. breve and T. vaginalis, examines the resulting effects on parasite growth and physiology, and assesses the consequences for host–parasite interactions. By characterizing these microbe–parasite dynamics, our work provides new insights into the ecological and metabolic factors that may contribute to parasite overgrowth and may inform the development of microbiome-informed strategies for managing trichomoniasis.

Methods

Parasite and bacterial cultures

T. vaginalis (ATCC 30236) was cultured in YI-S medium30, and serum-free YI-S (horse serum replaced by ddH₂O) was used for experimental treatments. The medium contained a vitamin mixture (#18), which was sterile-filtered prior to use. B. breve (ATCC 15700) was maintained anaerobically in Bifidobacterium broth, streaked to isolate single colonies, and preserved at −80°C in 50% glycerol. These cultures served as the basis for parasite-bacterium interaction studies.

Co-culture assay and B. breve pre-treated T. vaginalis

Co-culture experiments were conducted to investigate the direct interactions between T. vaginalis and B. breve. Mid-log phase bacteria and parasites were adjusted to 108 CFU and 107 cells, respectively. The ratio used in the co-culture assay followed that of the previous study12. Both organisms were combined in serum-free YI-S and incubated anaerobically for four hours. Cell viability and bacterial counts were monitored at sequential time points to assess the mutual growth dynamics. Each experiment was performed with three biological replicates. After the same treatment, T. vaginalis cells (1.5 × 105 cells) were subsequently used for co-culture with HeLa cells (2.5 × 10⁶ cells) (Figure 3).

Tri-culture assay (pre-treatment): prior colonization by B. breve

HeLa cells were purchased from ATCC, The culture medium is Dulbecco’s Modified Eagle Medium (DMEM; Gibco, Thermo Fisher Scientific, Cat# 12100046) with 10 % fetal bovine serum and 1 % antibiotic (GeneDireX, Cat# CC502-0100) were used as a host model to evaluate the effects of the microbes on epithelial cells. HeLa cells (2.5 × 10⁶ cells/mL) were seeded in 24-well plates and incubated for 24 h before treatment. For the pretreated B. breve experiments, cells were pretreated with 5 × 10⁷ CFU B. breve (live or heat-killed) followed by the addition of 2.5 × 10⁶ T. vaginalis cells in a total volume of 1 mL serum-free DMEM. To establish comparative conditions, T. vaginalis or B. breve was cultured alone under the same conditions. Untreated HeLa cells cultured in serum-free DMEM served as the negative control. The tri-cultures were incubated for 6 or 12 hours to compare host responses under different conditions. Each experiment was performed with three biological replicates.

Tri-culture assay (co-treatment): simultaneous competition

For the simultaneous competition experiments, HeLa cells were co-treated with 5 × 10⁷ CFU B. breve (live or heat-killed) and 2.5 × 10⁶ T. vaginalis cells in a total volume of 1 mL serum-free DMEM. To establish comparative conditions, T. vaginalis or B. breve was cultured alone under the same conditions. Untreated HeLa cells cultured in serum-free DMEM served as the negative control. The tri-cultures were incubated for 6 or 12 hours to compare host responses under different conditions. Each experiment was performed with three biological replicates.

Microscopy and staining

Confocal microscopy was used to assess morphological alterations and microbial interactions. After co-culture with B. breve, T. vaginalis samples and T. vaginalis which cultured in serum-free YI-S served as the negative control were stained with 4’,6-diamidino-2-phenylindole (DAPI) and visualized using confocal microscopy with a 100× oil immersion and a 405 nm laser. Choose an image size of 1,024 × 1,024 pixels and a scanning speed of 8.0 μs/ pixel to obtain high-quality images.

HeLa cells were treated according to the previously described procedure, followed by Giemsa staining for tri-culture assays to evaluate structural changes. The supernatants were collected for downstream cytokine analysis. HeLa cells were washed three times with phosphate-buffered saline (PBS) and fixed with a fixation solution for 30 minutes at room temperature, followed by overnight incubation with Giemsa solution. Images were quantitatively analyzed using CellProfiler, which measured the host cell area while excluding parasite interference. This analysis provided insights into the effects of microbes on host cell morphology.

RNA sequencing and analysis

T. vaginalis co-cultured with B. breve for 4 hours was collected with pre-warmed PBS washing steps, and T. vaginalis alone (negative control) was transferred into centrifuge tubes and centrifuged at a low speed (2,500 rpm for 5 minutes) to minimize bacterial cells co-pelleting. The supernatant was removed, and TRIzol reagent was added and mixed thoroughly. Total RNA was purified (Zymo Research, #R2052), and poly(A)-tailed mRNA was enriched using the Illumina Poly(A) Selection Kit (#20777928). The purified mRNA was fragmented and converted into double-stranded cDNA using the Illumina cDNA Synthesis Module (#20761354), followed by library construction (end repair, A-tailing, adaptor ligation, and PCR enrichment) with the Illumina RNA Prep, Ligation kit (#20778789). The library was constructed using Illumina Transcriptomic profiling of T. vaginalis was performed to identify the changes in gene expression induced by the experimental treatments. Total RNA was sequenced on the Illumina NovaSeq platform. Each experiment was performed with three biological replicates. Processed reads were mapped to the T. vaginalis G3 reference genome (TrichDB-65), and differential gene expression was analyzed using the DESeq2 package. Enrichment analyses, including ORA, GSEA, and GSVA, were performed using R to determine the affected biological pathways. Pathway maps were generated using KEGG (Kyoto Encyclopedia of Genes and Genomes)31.

Metabolomics analysis

Metabolomic analyses were conducted to characterize the lipid alterations in parasite-bacterium interactions. After co-culture, the culture medium was transferred to centrifuge tubes and centrifuged at 2,500 rpm for 5 min. The resulting pellets were resuspended in 1 mL of 80% methanol and stored at −80 °C until analysis. For lipid extraction, 300 μL of the methanolic cell suspension was mixed with 30 μL of methanol, followed by the addition of 900 μL of methyl tert-butyl ether (MTBE) containing an internal standard (Avanti Polar Lipids, Alabaster, AL, USA). The mixture was vortexed for over 1 min and incubated at room temperature for 60 min. Subsequently, 156 μL of water was added, and the sample was centrifuged at 12,000 ×g for 30 min at 4 °C. The upper organic phase was collected and evaporated to dryness under a gentle nitrogen stream. After KOH-mediated lipid hydrolysis, the dried residue was reconstituted in 200 μL of isopropanol/acetonitrile/water (2:1:1, v/v/v) and centrifuged again at 12,000 ×g for 30 min at 4 °C. The resulting supernatant was used for fatty acid analysis by ultra-high-performance liquid chromatography (UHPLC) coupled with a Xevo TQ-S triple quadrupole mass spectrometer (Waters Corp., Milford, MA, USA). Protein-normalized data were used to quantify the fatty acid profiles, providing insights into the lipid metabolic reprogramming associated with microbial interactions. Chromatographic separation was performed on a Waters ACQUITY CORTECS C18 column (2.1 mm × 150 mm, 1.6 μm) maintained at 60 °C. The mobile phases consisted of solvent A (acetonitrile/water, 20:80, v/v, with 0.01% formic acid) and solvent B (isopropanol/acetonitrile, 50:50, v/v, with 0.01% formic acid). The flow rate was 0.50 mL/min. The elution gradient started at 50% B, increased linearly to 70% B over 4 min, then to 98% B at 6.1 min, and was maintained for 1.3 min before returning to initial conditions and equilibrating for over 2 min. Mass spectrometric detection was performed in negative electrospray ionization (ESI) mode with a capillary voltage of 2000 V and a sampling cone voltage of 45 V. The desolvation gas flow was 1000 L/h, cone gas flow 150 L/h, desolvation temperature 500 °C, and source temperature 150 °C. Quality control (QC) samples and pooled QC samples (prepared by mixing aliquots of all samples) were analyzed throughout the run, with one QC injection after every 10 samples. Protein concentrations in the analyzed samples were determined using the Bradford assay and used to normalize the lipid extraction data. Each experiment was performed with four biological replicates.

Image analysis

Quantitative analysis of HeLa cell morphology was performed using the CellProfiler software. Images from Giemsa-stained cells were segmented using adaptive thresholding to accurately measure the cell area, while non-cellular elements and parasites were excluded. This approach allows for an objective assessment of host cell structural changes.

Cytokine measurement

The concentrations of IL-6 and IL-8 in culture supernatants were quantified using sandwich ELISA using an IL-6 ELISA assay (#430501; BioLegend, San Diego, CA, USA) and IL-8 ELISA assay (#431501; BioLegend, San Diego, CA, USA)as described previously32,33. Tri-culture supernatants were collected and centrifuged at 1,500 rpm to remove dead cells and debris. This assay detects cytokines through specific antibody binding and enzymatic color development, with the sample concentrations calculated from standard curves. Read the absorbance at 450 nm and 570 nm. The absorbance at 570 nm can be subtracted from that at 450 nm to correct for optical imperfections or background. These measurements provide a sensitive evaluation of inflammatory responses under various experimental conditions. Final cytokine levels are displayed as fold change of the control in the figures. Each experiment was performed with three biological replicates.

Statistical analysis

The Student’s t-test was used to quantify the experimental results using GraphPad Prism 10. Asterisks represent the significance of each assay as determined by the P-value (* p < 0.05; ** p < 0.01; *** p < 0.001).

Results

B. breve enhanced the growth of T. vaginalis

B. breve is considered the second most abundant beneficial commensal species in the human vaginal microbiota, and a marked reduction in its abundance has been reported in STD patients co-infected with T. vaginalis6,12. However, its role during T. vaginalis infection remains underexplored. To investigate potential interactions between the two organisms, T. vaginalis and B. breve were co-cultured at a 1:10 ratio under anaerobic conditions for 4 h, as previously described12,34. Viable T. vaginalis counts and B. breve colony-forming units (CFUs) were assessed at 0, 1, 2, and 4 hours.

Parasite viability did not differ significantly from the control during the first 2 hours (Fig. 1A). By 4 hours, T. vaginalis showed a significant increase in viable counts compared with the untreated control (Fig. 1B). In contrast, B. breve CFUs declined significantly by the 2-hour time point and continued to decrease thereafter (Fig. 1B). These observations indicated that B. breve enhances the short-term growth of T. vaginalis, while its own viability was adversely affected during co-culture. Based on the enhanced protozoan viability observed at 4 hours, this condition was selected for subsequent experiments.

Fig. 1
Fig. 1The alternative text for this image may have been generated using AI.
Full size image

Growth of T. vaginalis and reduction in B. breve colony-forming units (CFU) during co-culture. T. vaginalis and B. breve were co-cultured in serum-free YI-S medium at an initial ratio of 1:10. Monocultures of each organism served as controls. The X-axis indicates co-culture duration, and the Y-axis shows the relative counts or CFU of T. vaginalis (A) and B. breve (B) compared with their respective controls. Statistical significance: *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Direct interaction between T. vaginalis and B. breve during co-culture

To visualize the interaction between T. vaginalis and B. breve, confocal microscopy was performed following a 4-hour co-culture, the time point at which a significant increase in trichomonad viability was observed. T. vaginalis cells were stained with DAPI, fixed, and mounted for imaging. Representative confocal images are presented in Fig. 2A–F, with continuous video recordings available in Movie S1. In the co-culture group, B. breve cells were found closely surrounding the surface of T. vaginalis (Fig. 2D–F), suggesting a direct physical association between the two organisms under the in vitro culture conditions examined.

Fig. 2
Fig. 2The alternative text for this image may have been generated using AI.
Full size image

Localization of B. breve on the surface of T. vaginalis during co-culture. T. vaginalis and B. breve were co-cultured for 4 hours and stained with DAPI for confocal microscopy. Panels (A–C): T. vaginalis monoculture. Panels (D–F): Co-culture of T. vaginalis and B. breve.

B. breve pre-colonization does not confer cytoprotection against T. vaginalis infection

Previous studies have suggested that vaginal commensal bacteria may exert cytoprotective effects or contribute to maintaining microbiota stability during T. vaginalis infection14,35. To clarify the role of B. breve in this context, we first evaluated whether pre-exposure of T. vaginalis to B. breve modulates the inflammatory response in HeLa cells. HeLa cells were incubated for 24 h with T. vaginalis that had been pretreated with or without B. breve for 4 hours, and IL-6 secretion was quantified. As shown in Fig. 3A, T. vaginalis infection increased in IL-6 secretion approximately two-fold relative to uninfected controls. Notably, IL-6 levels were further elevated in cells infected with B. breve-pretreated T. vaginalis, suggesting that B. breve did not attenuate the inflammatory response and may even enhance it.

Fig. 3
Fig. 3The alternative text for this image may have been generated using AI.
Full size image

B. breve does not exhibit cytoprotective activity during T. vaginalis infection. (A) The secretions of IL-6 from HeLa cells infected with T. vaginalis and B. breve-treated T. vaginalis were determined after 24 hours of incubation. The treated cells were washed, incubated, stained with Giemsa, and monitored under a microscope: (B) upper left, HeLa cells only; upper right HeLa cells treated with B. breve for 3 hours and cultured in serum- and antibiotic-free DMEM for 12 hours; lower left, HeLa cells were co-cultured with T. vaginalis in serum- and antibiotic-free DMEM for 12 hours; lower right, HeLa cells were pretreated with B. breve in serum- and antibiotic-free DMEM for 3 hours, followed by infection with T. vaginalis under the same conditions for 12 hours. (C) The quantified chart of (B). (D) The same settings as B were applied for heat-killed B. breve (HK-Bb) instead of viable B. breve. (E) The quantified chart of (D). Statistical significance: *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Next, we investigated whether B. breve exerts a cytoprotective effect when pre-colonizing host epithelial cells. HeLa cells were pretreated with viable bacteria prior to T. vaginalis infection, with heat-killed B. breve used as an inactivated control. No significant difference in cell confluence (Fig. 3B–E) was observed between infected and uninfected cells pretreated with viable B. breve. Similarly, projected cell area measurements indicated that pretreatment with heat-killed B. breve did not alter the cell morphology induced by T. vaginalis (Fig. 3B and D).

To further assess inflammatory responses, IL-6 and IL-8 levels were measured at 6 and 12 h post-infection. As expected, IL-6 levels did not differ significantly between infected and uninfected cells, regardless of B. breve pretreatment (Fig. 4). In contrast, IL-8 levels were transiently elevated at the 6-hour time point in HeLa cells infected with T. vaginalis pretreated following B. breve pretreatment, suggesting a short-term stimulatory effect. In addition, IL-6 levels increased in the heat-killed B. breve group (Supplementary Fig. S1), likely reflecting immune activation by bacterial structural components.

Fig. 4
Fig. 4The alternative text for this image may have been generated using AI.
Full size image

Pre-colonization with B. breve does not alter IL-6 or IL-8 secretion in HeLa cells treated with T. vaginalis. HeLa cells were pretreated with viable B. breve followed by T. vaginalis infection. The IL-6 (A and B) and IL-8 (C and D) concentrations were measured by ELISA at 6 and 12 hours, respectively. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001; ns, not significant.

In summary, these findings suggest that B. breve does not confer cytoprotective or anti-inflammatory benefits during T. vaginalis infections. Instead, early exposure to B. breve may transiently enhance pro-inflammatory cytokine responses in host epithelial cells.

Simultaneous exposure to B. breve and T. vaginalis does not alter HeLa cellular responses.

To determine whether B. breve competes with T. vaginalis for attachment to epithelial cells and thereby affects parasite cytoadherence and the subsequent epithelial damage, HeLa cells were co-treated with both organisms simultaneously and monitored for 6 and 12 hours. Quantification of projected cell area showed no significant difference between the co-treatment group and the T. vaginalis-only group, indicating that B. breve did not alter the extent of T. vaginalis-induced cytopathic effects (Fig. 5A–B). Comparable results were observed in cells co-treated with heat-killed B. breve (Fig. 5C–D), further supporting the lack of competitive interference with T. vaginalis-mediated epithelial damage.

Fig. 5
Fig. 5The alternative text for this image may have been generated using AI.
Full size image

Unchanged cell morphology in HeLa cells during simultaneous treatment of B. breve and T. vaginalis. The treated cells were washed, incubated, stained with Giemsa, and monitored under a microscope: (A) upper left, HeLa cells only; upper right, HeLa cells treated with B. breve for 12 hours; lower left, HeLa cells were co-cultured with T. vaginalis for 12 hours; lower right, HeLa was co-treated with B. breve and T. vaginalis for 12 hours. (B) The quantified chart of A. (C) The same settings as A were applied for heat-killed B. breve (HK-Bb) instead of viable B. breve. (D) The quantified chart of C. *p < 0.05; **p < 0.01.

To evaluate potential cytoprotective effects, IL-6 and IL-8 levels were measured in the co-treated groups. At 6 hours, cells co-treated with viable B. breve and T. vaginalis exhibited higher levels of both cytokines than cells infected with T. vaginalis alone (Fig. 6). A similar increase was observed in the heat-killed B. breve co-treatment group (Supplementary Fig. S2), suggesting that B. breve—whether viable or inactivated—may transiently provoke early inflammatory responses during co-exposure.

Fig. 6
Fig. 6The alternative text for this image may have been generated using AI.
Full size image

Simultaneous treatment of B. breve and T. vaginalis altered IL-6 or IL-8 secretion transiently in HeLa cells. HeLa cells were co-infected with viable B. breve and T. vaginalis for tri-culture. The IL-6 (A and B) and IL-8 (C and D) concentrations were measured by ELISA at 6 and 12 hours, respectively. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001; ns, not significant.

Taken together, these findings indicate that the presence of B. breve does not mitigate epithelial disruption caused by T. vaginalis. Instead, simultaneous exposure appeared to enhance early pro-inflammatory cytokine production, highlighting the potential inflammation-promoting role of B. breve during the initial phase of T. vaginalis colonization.

Transcriptomics analysis reveals molecular regulations in T. vaginalis during B. breve treatment

Given the observed increase in viable T. vaginalis cells after 4 h of co-culture with B. breve, RNA sequencing was performed to investigate the transcriptional alterations associated with this interaction. A total of 30,648 genes were detected in the control T. vaginalis group and 30,748 genes in the B. breve–treated group (Table 1and Supplementary Table 1). Principal component analysis (PCA) showed clear separation between the two groups, demonstrating high reproducibility and confirming the suitability of the dataset for downstream analyses (Fig. 7A). Differential gene expression analysis identified numerous genes with significant log2 fold changes, which are presented in the heatmap in Fig. 7B.

Table 1 Statistics of transcriptomics analysis.
Full size table
Fig. 7
Fig. 7The alternative text for this image may have been generated using AI.
Full size image

PCA plot and heatmap of the comparative transcriptomics analysis. (A) Principal Component Analysis (PCA) was performed following Differential Expression Gene (DEG) analysis. Red dots represent the control group (triplicate), and blue dots represent the treatment group (co-culture). (B) The annotation bar at the top of the heatmap indicates experimental conditions. Control samples are shown in purple and treatment samples are shown in green. The upper (1) of the heatmap shows upregulated genes (red) in the control group compared to the control group, whereas the lower (2) shows downregulated genes (purple) in the co-culture group. Adjusted p-value (FDR) < 0.05.

To functionally interpret the transcriptomic data, we employed the ProFun platform36, which integrates gene set enrichment analysis (GSEA) and Gene Ontology (GO) analyses. Upregulated genes were enriched in pathways related to DNA replication and chromatin organization, including nucleosome assembly34,37,38, protein–DNA complex formation39,40, and chromatin subunit organization41,42 (Fig. 8A). These results are consistent with our earlier observation that T. vaginalis proliferation increased following B. breve exposure (Fig. 1). In addition, GO network analysis revealed significant enrichment of pathways involved in ATP synthesis–coupled electron transport (Fig. 8B), suggesting enhanced energy production.

Fig. 8
Fig. 8The alternative text for this image may have been generated using AI.
Full size image

Gene Ontology (GO), Gene Set Enrichment Analysis (GSEA), and Gene Set Variation Analysis (GSVA) of significantly regulated genes in B. breve co-cultured with T. vaginalis. (A) This GO plot showed the upregulated genes of T. vaginalis in the co-culture group. The circle size represented the gene number of the specific terms, whereas the color represented the statistical significance. (B) This network highlighted significant enrichment of gene sets related to the indicated biological processes. The circle size represented the gene number of the specific terms, whereas the color represented the statistical significance. (C) The upregulated and downregulated genes involved in biological pathways according to the Kyoto Encyclopedia of Genes and Genomes (KEGG) were shown, with different colors of bars representing the statistical significance (q-value). (D) GSVA represented pathways significantly upregulated in the co-culture group, which were blue, whereas downregulated pathways were green. Gray bars represented no significant differences compared to the control.

Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis further demonstrated that upregulated genes were associated with energy metabolism and putative fatty acid metabolic pathways (Fig. 8C), implicating metabolic reprogramming in parasites. Supporting this interpretation, gene set variation analysis (GSVA) showed increased activity in fatty acid biosynthesis and propionate metabolism pathways (Fig. 8D). Propionate, a short-chain fatty acid (SCFA), is commonly produced by the vaginal microbiota and contributes to pH regulation and immune modulation within the vaginal environment43,44,45.

Collectively, these transcriptomic and functional analyses indicated that co-culture with B. breve enhanced T. vaginalis growth by upregulating genes involved in DNA replication, energy production, and fatty acid metabolism. In particular, the activation of SCFA-related pathways, such as propionate metabolism, suggests that B. breve may influence T. vaginalis physiology via microbiota-derived metabolites.

Metabolomics analysis reinforced the impacts of B. breve on fatty acid metabolism in T. vaginalis

As noted above, transcriptomic analysis revealed significant upregulation of genes involved in fatty acid metabolism in T. vaginalis after co-culture with B. breve, whereas key genes associated with primary energy-generating pathways—including glycolysis and proteolysis/ amino acid catabolism—were not markedly altered (Supplementary Fig. S3). To determine whether these transcriptional changes were reflected at the metabolic level, we performed metabolomic profiling of T. vaginalis.

A total of 38 fatty acids containing 12 or more carbon atoms were identified and categorized into saturated fatty acids (SFAs) and unsaturated fatty acids (UFAs). After normalization to protein content, 25 metabolites were classified as UFAs and 13 as SFAs. As shown in Fig. 9, levels of several fatty acids were significantly reduced in the B. breve–treated group relative to the control, including 8 UFAs and 7 SFAs. This reduction suggests increased utilization of both SFA and UFA substrates, likely reflecting enhanced fatty acid metabolic activity.

Fig. 9
Fig. 9The alternative text for this image may have been generated using AI.
Full size image

The overall decrease in fatty acid concentrations in T. vaginalis co-cultured with B. breve was revealed by metabolomic analysis. Fatty acid compositions of monocultured (control, green) and co-cultured T. vaginalis with B. breve (treatment, red) for 4 hours were determined by metabolomics analysis. Four independent samples in each group were analyzed. Orange color blocks represent genes with log2 fold change > 0, while blue color blocks represent those with log₂ fold change < 0. *p < 0.05; **p < 0.01.

These findings indicate that T. vaginalis undergoes metabolic adaptation in response to B. breve, increasing its reliance on fatty acid metabolism—possibly to support the energetic and biosynthetic demands associated with accelerated DNA replication and proliferation. This metabolic shift aligns with the observed increase in T. vaginalis cell numbers following co-culture.

Discussion

Probiotics are commonly used to promote vaginal health46; however, their interactions with protozoan pathogens, such as T. vaginalis, are still poorly understood. In this study, we investigated the role of B. breve in T. vaginalis infection, particularly its effect on host epithelial responses. HeLa cells were used in this research to explore the functional associations between B. breve and L. gasseri, and they also served as a commonly used epithelial model for studying T. vaginalis–host interactions11,47,48,49. Consistent with its known immunomodulatory properties50,51, B. breve modulates the production of IL-6 and IL-8 in HeLa cells. However, when HeLa cells were pretreated with B. breve before T. vaginalis infection, cytokine production remained elevated rather than attenuated. Similarly, B. breve did not prevent epithelial damage or reduce T. vaginalis adherence, regardless of whether it was administered before or simultaneously with the parasites. These findings suggest that B. breve does not exert cytoprotective or anti-inflammatory effects. While B. breve alone elicited cytokine release, the biological relevance of this effect during T. vaginalis infections remains unclear. Future studies using immune cells or cytokine-conditioned media could clarify whether B. breve-induced signals alter immune cell behavior or influence protozoan pathogenesis.

Confocal imaging revealed a close physical association between B. breve and T. vaginalis, raising the possibility of phagocytosis. T. vaginalis is known to internalize yeast via actin-dependent mechanisms27, and our transcriptomic data showed upregulation of phagocytosis-related genes (Supplementary Fig. S4). Similar mechanisms have been described for bacterial uptake in Entamoeba histolytica, and the enrichment of genes annotated to ER-mediated phagocytosis suggests a potential contribution of intracellular membranes to the engulfment process52. Although no direct evidence of bacterial internalization was obtained, short-term co-culture with live-cell imaging may clarify whether B. breve is internalized or remains surface-bound.

Although B. breve is generally regarded as beneficial, its effects are context-dependent. Previous studies have focused on its intestinal role, where it functions synergistically with other probiotics to suppress pathogenic bacteria and promote microbial balance53,54,55,56. In contrast, its role within the vaginal niche remains unclear. In our short-interval co-culture model, B. breve did not protect epithelial cells from T. vaginalis-induced cytopathic effects, and its abundance decreased during the co-culture period. These findings differ from the behavior of protective species like L. gasseri, which inhibits T. vaginalis adhesion and preserves epithelial integrity13,14,57. Thus, although B. breve may support the growth of other beneficial species, its direct impact on T. vaginalis infection appears limited and, in certain contexts, may even be detrimental.

The pro-inflammatory cytokines IL-6 and IL-8 act as a ‘double-edged sword’ during T. vaginalis infection. While their transient elevation is vital for neutrophil recruitment and parasite clearance58,59, probiotics like Bifidobacterium can also induce basal secretion to prime mucosal immunity and enhance barrier function60. In this context, the differential IL-6 responses observed between Figures 3 and 4 likely reflect differences in experimental context and timing, as sustained exposure to T. vaginalis following bacterial pre-incubation (24 h) may amplify epithelial inflammatory signaling, whereas short-term pre-treatment of HeLa cells with B. breve alone (6–12 h) is insufficient to elicit a comparable response. Conversely, T. vaginalis exploits hyperinflammation for iron acquisition and tissue invasion60, thereby increasing susceptibility to secondary infections61. Thus, the ability of beneficial microbes to modulate these cytokine spikes is likely essential for vaginal homeostasis62. Nevertheless, further functional studies are required to determine whether such cytokine elevations ultimately favor host defense or serve as a parasitic survival strategy.

During short-term co-incubation, T. vaginalis remained the dominant organism in the culture. By 24 hours, however, the interaction dynamics had shifted markedly, with B. breve becoming the predominant species and exhibiting substantial overgrowth—even in serum-free medium. This reversal indicates that prolonged co-culture conditions favor bacterial expansion and fundamentally alter the microbe–parasite balance, thereby limiting the interpretability of extended in vitro assays (data not shown). Consequently, analyses focused on the early time points provide a more accurate representation of the initial bacterium–protozoan interaction.

Transcriptomic analysis revealed significant upregulation of genes associated with DNA replication and fatty acid metabolism, whereas pathways related to glycolysis and amino acid catabolism were not markedly altered (Supplementary Fig. S3). These transcriptional findings are consistent with the observed increase in T. vaginalis cell numbers following B. breve exposure. GO and KEGG analyses further highlighted enrichment in pathways associated with ATP synthesis, nucleosome assembly, and fatty acid utilization. In addition, GSVA and GSEA indicated activation of SCFA–related pathways, particularly propionate metabolism, which is known to influence microbial growth as well as host immune regulation43,44,45.

To prevent contamination of the T. vaginalis transcriptome by B. breve RNA, poly(A)-tailed mRNA was selectively captured during library preparation. Additionally, a low-speed centrifugation step was performed before harvesting the trichomonad cells to minimize the co-pelleting of bacterial cells. Consistent with these precautionary measures, the mapping rate across all transcriptomic samples remained stable (Table 1), supporting the integrity and purity of our T. vaginalis transcriptome dataset.

Transcriptomic analysis revealed enrichment of pathways associated with SCFA metabolism. However, SCFAs cannot be reliably quantified using conventional LC–MS without chemical derivatization due to their low molecular weights and poor ionization efficiency49. Therefore, to evaluate whether these transcriptional shifts corresponded to measurable metabolic changes, we performed LC-MS–based profiling of long-chain fatty acids (≥C12), which constitute the readily detectable portion of the parasite lipid pool. In total, 38 fatty acids were identified, including 25 unsaturated and 13 saturated species. Among these, 15 metabolites (8 UFAs and 7 SFAs) were significantly decreased following B. breve co-culture, suggesting enhanced fatty acid consumption by the parasite. These findings support the interpretation that T. vaginalis shifts its metabolism toward the utilization of exogenous fatty acids, potentially conserving intracellular energy for replication and contributing to increased growth. Future work will explore whether specific metabolites, such as propionate, or B. breve-conditioned media, can induce phenotypic changes in T. vaginalis. This approach aims to distinguish between contact-dependent mechanisms and effects mediated by B. breve secreted factors.

Notably, glycerophospholipid biosynthesis was downregulated in co-cultured T. vaginalis. Glycerophospholipids are central to membrane integrity, protein function, and cellular signaling63,64,65, and alterations in lipid composition have been shown to influence protozoan virulence, as reported in Leishmania and Plasmodium, where lipid alterations modulate immune evasion and motility66,67,68. We therefore hypothesize that, in the presence of B. breve, T. vaginalis may reduce its own lipid synthesis to conserve biosynthetic energy and instead rely on available fatty acids to support membrane biogenesis and proliferation.

Conclusions

Our study demonstrates that B. breve interacts directly with T. vaginalis, promoting its growth and triggering transcriptional and metabolic reprogramming, particularly in fatty acid metabolism and DNA replication. Despite its known probiotic functions, B. breve did not confer cytoprotective or anti-inflammatory benefits to host cells challenged with T. vaginalis. Instead, our findings suggest that B. breve may help sustain microbial balance rather than acting as a direct antagonist against protozoan pathogens in the vagina. These results highlight the complexity of host–microbe–protozoan interactions and underscore the need to further explore the context-specific roles of vaginal probiotic in infection-related ecological dynamics (Fig. 10).

Fig. 10
Fig. 10The alternative text for this image may have been generated using AI.
Full size image

The proposed role of B. breve during T. vaginalis infection. During the co-culture period, the phenotype of T. vaginalis was altered through direct contact with B. breve, while the CFU of B. breve was reduced. On the other hand, B. breve might not exert a cytoprotective role but rather help maintain microbiota balance.