The bovine uterus, once considered sterile, is now recognized as a dynamic microbial ecosystem that undergoes substantial changes during the postpartum period and plays a critical role in reproductive health. This study investigated the composition and temporal dynamics of the uterine bacterial and fungal microbiota in postpartum Angus cows and showed that dietary grape pomace bolus supplementation was associated with temporal shifts in microbial community composition, including changes in selected bacterial taxa and increased fungal community stability. A total of 19 cows were allocated to control (n = 10) or treatment groups (n = 9), with treated animals receiving slow-release grape pomace boluses administered every three weeks from 21 days prepartum until 60–70 days postpartum. Uterine lavage samples were collected during the first (days 4–12) and ninth (days 63–70) postpartum weeks and analyzed using amplicon-based sequencing targeting the bacterial 16S rRNA (V3–V4) region and the fungal ITS2 region. The uterine microbiota exhibited pronounced temporal changes, with Firmicutes predominating in the early postpartum period and an increased abundance of potentially pathogenic Fusobacteria and Bacteroidetes observed by week nine postpartum, particularly in control cows. Grape pomace supplementation was associated with a more favorable bacterial profile, including a reduced relative abundance of key uterine pathogens such as Fusobacterium necrophorum and Trueperella pyogenes. The uterine mycobiome was dominated by Ascomycota across all groups; however, supplemented cows showed greater fungal community stability and a lower prevalence of opportunistic genera, including Candida spp. In conclusion, dietary grape pomace supplementation may beneficially modulate both bacterial and fungal uterine microbiota during the postpartum period, suggesting its potential as a natural strategy to support uterine microbial homeostasis and reproductive health in beef cows.

The equine reproductive microbiota has emerged as an important factor influencing reproductive health and fertility in both mares and stallions. Traditionally, the equine uterus was considered sterile, and microbial presence was interpreted primarily in the context of infection. However, sequencing-based studies have demonstrated that the reproductive tract, including the uterus and semen, contains detectable microbial communities or microbial DNA signatures, challenging this traditional paradigm. In mares, the vaginal microbiota is consistently dominated by Firmicutes, Bacteroidetes, Proteobacteria, and Actinobacteria, whereas the uterine environment represents a low-biomass niche in which interpretation is complicated by contamination risk and the inability of sequencing-based methods to distinguish viable from nonviable microorganisms. Culture-based studies consistently identify opportunistic pathogens such as Streptococcus equi subsp. zooepidemicus, Escherichia coli, and Pseudomonas aeruginosa in association with endometritis and persistent breeding-induced endometritis, while sequencing-based studies suggest broader community-level dysbiosis rather than simple pathogen presence. In stallions, semen is not sterile and commonly contains taxa such as Porphyromonas, Corynebacterium, Peptoniphilus, and other opportunistic bacteria that may influence sperm quality and microbial transmission to mares during breeding. However, most reported associations remain correlative, and direct longitudinal evidence for persistent stallion-to-mare microbial transmission is limited. This review synthesizes current evidence on microbial composition, hormonal influences, dysbiosis, and reproductive implications of the equine reproductive microbiota, integrating culture-based and sequencing-based findings while emphasizing methodological limitations associated with low-biomass samples. Improved understanding of these microbial ecosystems may support more evidence-based reproductive diagnostics and microbiome-informed fertility management in horses.

Fungal DNA in the equine uterus remains poorly characterized, particularly in mares without overt clinical uterine disease. This exploratory pooled pilot study used ITS2 amplicon sequencing to describe fungal DNA profiles in pooled uterine samples and to perform descriptive comparisons by subsequent pregnancy outcome and sampling method. Conventional microbiological culture findings were summarized as clinical context. samples were collected during naturally occurring estrus and, in mares that were subsequently mated, before the first mating of the studied estrous cycle. Low-volume uterine lavage (LVL) samples from 12 naturally mated Trakehner mares were pooled by outcome into a pregnant-LVL pool (1FR; n = 5) and a non-pregnant-LVL pool (2NFR; n = 7). The paired sampling-method comparison comprised a paired LVL pool (3LVL) and a paired SWAB pool (4SWAB), each constructed from samples collected from the same five mares. After bioinformatic processing, 82,862–130,870 reads and 27–111 final unique amplicon sequence variants were retained per pooled library. ITS2-target quantitative PCR detected 1,107–1,728 target copies/μL of pooled DNA extract. Neither the extraction blank nor the library-preparation blank crossed the qPCR detection threshold within 40 cycles; however, both yielded sequencing reads, and sequence-resolved control analysis could not exclude low-level laboratory or sequencing-related contamination. A targeted sensitivity analysis excluding the sequence feature shared by the library-preparation blank and two biological libraries did not materially alter the descriptive taxonomic composition, diversity, or ordination patterns. Ascomycota predominated across the pooled libraries (50.7–72.5%), followed by Basidiomycota (8.1–24.1%). The pregnant-LVL library had the lowest pool-level Shannon diversity value and occupied an ordination position that differed from the other libraries primarily along the first principal coordinate. The paired LVL and swab libraries also differed descriptively in taxonomic composition and ordination position. No fungal growth was reported in the available routine culture reports. These observations demonstrate the presence of diverse fungal DNA signals but do not establish viable uterine colonization, fungal infection, an association with subsequent pregnancy outcome, or equivalence between sampling methods. Confirmation requires sequencing of individual mares, comprehensive field and laboratory controls, standardized mycological culture, and integration with cytological, histopathological, and reproductive data.

Complex interactions among structural, endocrine, and molecular factors determine mare reproductive potential. This narrative review summarizes current research data on clinical, hormonal, and molecular–biochemical biomarkers associated with mare fertility and evaluates their potential use in reproductive assessment. Clinical assessment remains the foundation of reproductive evaluation and includes age, reproductive history, ultrasonography, cytology, bacteriology, and endometrial biopsy. Hormonal biomarkers such as anti-Müllerian hormone, estradiol, progesterone, and gonadotropins provide additional information on ovarian reserve, follicular activity, ovulation, and luteal competence. Increasing attention has also focused on molecular and biochemical biomarkers related to inflammation, oxidative stress, extracellular matrix remodeling, and embryonic–maternal communication. Despite growing interest in advanced biomarker approaches, many molecular indicators are not yet sufficiently validated for routine clinical use in mares. Small sample sizes, methodological heterogeneity, and a lack of standardized reference ranges often limit available studies. Therefore, interpretation of reproductive biomarkers should remain cautious and clinically contextualized. Available data suggest that integrating structural, endocrine, and molecular information may improve reproductive evaluation and enable more individualized management strategies in mares. However, prospective studies evaluating the predictive value and clinical utility of integrated biomarker models remain limited.