Expression of miR-21 and miR-93 in abattoir-derived bovine ovarian follicles Authors: Merve Yılmaz, Mert Sarılar, Aslıhan Baykal Uğur, Gamze Evkuran Dal Published: 2026 ABSTRACT AbstractBackgroundCystic ovarian follicles (COF) are a major reproductive disorder in cows, leading to impaired fertility and substantial economic losses. The pathogenesis of COF is multifactorial and involves neuroendocrine dysregulation as well as local ovarian dysfunction. Increasing evidence suggests that microRNAs (miRNAs) play critical roles in follicular development and granulosa cell function. In this study investigated the expression levels of miR-21, miR-93, and BMPR2 in cystic ovarian follicles. The study examined healthy (n = 35) and cystic (n = 35) ovarian tissues. The aim of this study was to evaluate the expression levels of miR-21, miR-93, and BMPR2 in bovine cystic ovarian follicles and to investigate their potential association with the cystic ovarian condition.ResultsHistopathological evaluation confirmed characteristic degenerative changes in cystic follicles, including thinning and sparse cellularity of the granulosa layer and atrophic alterations of the theca layer. Quantitative real-time PCR analysis revealed significantly higher expression levels of miR-21 (p = 0.013) and miR-93 (p < 0.001) in cystic ovarian tissues compared with healthy controls, whereas BMPR2 expression did not differ significantly between groups (p = 0.725). Receiver operating characteristic (ROC) curve analysis demonstrated moderate discriminative power for miR-21 (AUC = 0.672) and high discriminative performance for miR-93 (AUC = 0.899). Correlation analysis showed a significant positive association between miR-93 and BMPR2 expression in the cystic group, while no significant correlations were observed in healthy ovaries.ConclusionsThe findings indicate that miR-21 and miR-93 are upregulated in bovine cystic ovarian follicles, with miR-93 showing superior diagnostic potential compared with miR-21. The positive association between miR-93 and BMPR2 expression in cystic ovaries suggests a possible involvement of miR-93 in BMP-related regulatory pathways during cyst formation.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s12917-026-05616-3.Keywords: MicroRNA, Cystic ovarian follicles, Cows, miR-21, miR-93 Introduction Cystic ovarian follicles (COF) develop when one or more follicles fail to ovulate, continue to grow, and maintain steroidogenic activity [1]. These structures are defined as follicle-like structures, at least 2.5 cm in diameter, located in one or both ovaries in the absence of luteal tissue, and surviving for at least ten days [1, 2]. The mechanism of COF formation is multifactorial, and the most widely accepted hypothesis is neuroendocrinological dysfunction of the hypothalamic-pituitary-gonadal axis [3]. The second key mechanism is ovarian dysfunction. Abnormal production of growth factors in granulosa cells, irregular secretion of extracellular matrix proteins (vitronectin, fibronectin, etc.), and changes in Luteinizing Hormone /Estrogen receptor expression inhibit ovulation, thus creating a predisposition to cyst formation [4]. Within the ovarian regulatory system, Oocyte-derived Bone Morphogenetic Protein-15 (BMP15) and Growth Differentiation Factor-9 (GDF-9) which belong to the Transforming Growth Factor beta superfamily, play a critical role in regulating granulosa and cumulus cell functions. These factors exert their effects via the Bone Morphogenetic Protein Receptor Type II in order to initiate intracellular signal transduction [5–9]. In living organisms, the carrier of genetic information is deoxyribonucleic acid (DNA). Genetic information encoded in DNA is transcribed into Ribonucleic Acid (RNA) through transcription, and protein synthesis occurs after this process [10]. RNA molecules that constitute a significant portion of the genome and lack protein-coding capacity are defined as non-coding RNAs [11]. MicroRNAs (miRNAs), a class of non-coding RNAs, are approximately 22 nucleotides in length and are generated through a biogenesis process involving the enzymes Drosha and Dicer [12]. miRNAs have been studied in various mammalian species such as humans, cows, sheep, buffaloes, etc. Current studies indicate that miRNAs can be detected in blood, milk, and various body fluids and may be associated with various physiological and pathological processes [13–18]. miRNAs have been shown to play crucial roles in follicular development and corpus luteum function within the bovine ovary [19, 20]. miR-21 is one of the miRNAs that plays a critical role in regulating the expression of genes associated with cell proliferation and the suppression of apoptosis in numerous biological processes in cattle, such as folliculogenesis, nuclear function, embryo development, milk production, and immunity [14, 21, 22]. miR-93 has been reported to be associated with pregnancy, milk production, and immune function in cattle [23–25]. Furthermore, miR-21 and miR-93 have been implicated in the pathophysiology of polycystic ovary syndrome (PCOS) in women. This condition shares several features with cystic ovarian follicles in cattle, including anovulation and altered follicular dynamics [15, 26–28]. Given the critical role of BMPR2 in BMP-related signaling and the known regulatory functions of miRNAs in ovarian physiology, a potential interaction between altered miRNA expression and BMPR2-mediated pathways may be relevant to cystic follicle formation. In bovine cumulus cells, specific miRNAs have been shown to regulate BMPR2 expression and to influence cellular processes such as proliferation and apoptosis. Therefore, evaluating miR-21, miR-93, and BMPR2 together may provide further insight into the molecular alterations associated with bovine cystic ovarian follicles [29]. It was hypothesized that the expression levels of miR-21 and miR-93 would be altered in bovine cystic ovarian follicles and that these alterations might be associated with BMPR2 expression. The aim of this study was to evaluate alterations in the expression levels of miR-21, miR-93, and BMPR2 in COFs of cows, with the objective of providing insight into molecular changes associated with follicular cyst formation. Materials and methods Ovarian collection and storage Ovarian samples were collected from a local abattoir. Prior to slaughter, the age and breed of the cows were recorded. To ensure homogeneity, only tissue samples obtained from Holstein cows aged 5–7 years and weighing 500–700 kg were included in the preliminary evaluation. Ovaries containing follicles exceeding 25 mm in diameter were provisionally classified as cystic ovaries based on gross morphological assessment (Fig. 1) [30]. In contrast, healthy control samples were collected from ovaries containing large antral follicles with a diameter of approximately 8–15 mm, consistent with physiologically developing follicles and without any macroscopic signs of cystic or luteal structures. Two separate tissue specimens involving the ovarian cortex and follicular wall were obtained from each healthy and cystic ovary. The first sample was transferred into Eppendorf tubes, rapidly frozen in liquid nitrogen to preserve RNA integrity, and stored at − 80 °C for subsequent molecular analyses. The other sample was preserved in formaldehyde solution and submitted for histopathological examination. To ensure objectivity of the evaluations, the study was conducted using a blinded method. Histopathological evaluations were performed by a single experienced veterinary pathologist who was blinded to the group allocation. An image of an ovary containing a follicle larger than 25 mm Histopathological analysis Tissue samples were fixed in 10% buffered formalin solution for 24 h. Subsequently, sections taken from the prepared paraffin blocks were stained with Hematoxylin & Eosin (H&E). All sections were examined under a light microscope (Nikon Eclipse Ci-L, Japan). H&E staining allowed clear evaluation of follicle wall thickness and luteinization [31]. Total RNA isolation RNA isolation from all ovarian tissue samples was performed according to the Hybrid-R™ RNA Isolation Kit (GeneAll, Seoul, Korea) protocol. To evaluate the quantity and purity level of the isolated RNA, measurements were performed using NanoDrop™ 2000/2000c Spectrophotometers (Thermo Fisher Scientific, USA). Complementary DNA (cDNA) synthesis Following RNA synthesis, cDNA synthesis was performed for BMPR2 using the cDNA Synthesis Kit (ABT, Atlas Biotechnology, Türkiye #dT20) according to the protocol described in Table 1. For miR-21 and miR-93, synthesis was performed using Stem-Loop Primers according to the protocols described in Table 2. Sample mixtures placed in 96-well plates were then processed by Applied Biosystems SimpliAmp Thermal Cycler (Thermo Fisher Scientific, Waltham, USA) under the conditions described in Table 3. Table 1Components and volumes used for cDNA synthesis for BMPR2ComponentVolume5X Reaction Buffer4 µlReverse Transcriptase (200 U/µL)1 µLRNase Inhibitor (10 U/µL)0.5 µLRandom Hexamers (50 µM)1 µL10 mM dNTP Mix1 µLTemplate + RNase Free Water12.5 µL Components and volumes used for cDNA synthesis for BMPR2 Table 2Components and volumes used for cDNA synthesis for miR-21 and miR-93ComponentsVolume5X Reaction Buffer4 µlReverse Transcriptase (200 U/µL)1 µLRNase Inhibitor (10 U/µL)0.5 µLStem-loop Primer (5 µM)2 µL10 mM dNTP Mix1 µLTemplate + RNase Free Water11.5 µL Components and volumes used for cDNA synthesis for miR-21 and miR-93 Table 3RT-PCR reaction required for cDNA synthesisTemperature (oC)Time (minute)Cycle25515015185514∞∞ RT-PCR reaction required for cDNA synthesis Real time PCR Real-Time PCR (RT-PCR) was performed for miR-21, miR-93 and BMPR2 using 2X qPCR SYBR-Green MasterMix (without ROX) (ABT, Atlas Biotechnology, Türkiye # Q03-01-05) (Table 4). Sample mixtures were prepared in a total volume of 20 µL and placed in a 96-well plate using the Applied Biosystems QuantStudio™ 5 RT-PCR System (Thermo Fisher Scientific, Waltham, MA, USA) and the reaction was carried out as shown in Table 5. SYBR-Green was used as the fluorescent dye. For miRNA expression analysis, U6 was used as the endogenous reference for normalization, whereas β-actin (ACTB) served as the housekeeping gene for normalization of BMPR2 mRNA expression. Primer sequences used for miR-21, miR-93, BMPR2, and reference genes are provided in Table 6. Table 4Components and volumes for RT-PCR reactions performedComponentsVolumeA.B.T.™ 2X qPCR SYBR- Green MasterMix (without ROX)10 µLForward Primer (10 µM)1Reverse Primer (10 µM)1specific cDNA TemplateIt is added in a sample volume of 1000 ng/1 µL.RNase-Free Distilled WaterIt is added until the total volume reaches 20 µL Components and volumes for RT-PCR reactions performed Table 5RT-PCR reactionPCR StepTemperature (oC)Time (second)CycleInitial Denaturation95300 s.1Denaturation9515 s.35Anneal5830 s.35Melt Curve60–952–5 s. /steps.- RT-PCR reaction Table 6Primer sequencesPrimer sequencesbta-miR-21(F)TAGCTTATCAGACTGATGTTGACTbta-miR-21(R)It was used from the kit.bta-miR-21(sl)GAAAGAAGGCGAGGAGCAGATCGAGGAAGAAGACGGAAAGAAATGTGCGTCTTCGCCTTCTTTCAGTCAACAbta-miR-93 (F)CAAAGTGCTGTTCGTGCAGGTAbta-miR-93 (R)It was used from the kit.bta-miR-93 (sl)GAAAGAAGGCGAGGAGCAGATCGAGGAAGAAGACGGAAAGAATGTGCGTCTCGCTTCTTTCTACCTGCAU6It was used from the kit.BMPR2 (F)CTTGCGCTTGGCTCATTCCTBMPR2 (R)ATCTCGATGGGAAATTGCAGGTACTB (F)GCAGATGTGGATCAGCAAGCACTB (R)CAGCTAACAGACTGATTGTTGACT Primer sequences Statistical analysis Statistical analyses were performed using Jamovi version 2.3.21. Normality was assessed using the Shapiro-Wilk test to determine whether the fold-change values of miR-21, miR-93, and BMPR2 were normally distributed. Non-parametric analysis was conducted using the Mann-Whitney U test to assess differences between groups. A p-value < 0.05 was considered statistically significant. Results Histopathological analysis Histological evaluation revealed that ovarian tissue in the healthy group showed no pathological alterations, and follicles exhibited typical morphology. In these follicles, the granulosa cell layer was uniform, multilayered and continuous, and the theca layer was clearly delineated from the surrounding stromal tissue (Fig. 2). In follicular cysts, the granulosa cell layer is markedly thinned and characterised by sparse cellularity, while the theca layer exhibits atrophic changes with a stromal-like appearance (Fig. 3). Luteal cysts are characterised by cyst walls predominantly composed of luteal cells and widespread luteinised tissue. Ovaries histologically diagnosed as luteal cysts, as well as those exhibiting other pathological conditions such as oophoritis or tumors, were excluded from the study. Following microscopic examination and definitive histopathological evaluation, ovaries obtained from cows were divided into two groups: healthy ovaries (n = 35) and cystic ovaries (n = 35). Each ovary represented an independent specimen from a different animal. Healthy ovarian tissue. A: A Graafian follicle containing an ovum (red arrow). H&E, Bar = 400 μm. B: Blue arrow represents the granulosa layer, star represents the theca layer. H&E, Bar = 100 μm A Histopathological image of a cystic ovary, follicular cyst. A large cystic cavity and thinned follicular wall are observed. H&E, Bar = 400 µm. B Higher magnification of the follicular cyst. The granulosa layer is thinned markedly (blue arrow), and the theca layer is stained similarly to the surrounding ovarian stroma. H&E, Bar = 100 µm. C Histopathological image of a follicular cyst wall showing thinning of the follicular layers. H&E, Bar = 200 µm. D Higher magnification showing a markedly thinned granulosa layer (red arrow). H&E, Bar = 90 µm Total RNA isolation RNA concentrations and optical density ratios of 260 nm / 280 nm obtained from ovarian tissue samples are presented (Additional files 1). miR-21, miR-93 and BMPR2 expression Relative expression fold changes in healthy and cystic ovarian tissues were calculated using the method (2−ΔΔCt). The average expression fold changes were 1.40 and 2.89 for miR-21, 1.20 and 3.46 for miR-93, and 1.33 and 1.32 for BMPR2 in healthy and cystic ovarian tissues, respectively (Fig. 4). A significant difference was found between healthy and cystic groups for miR-21 (p = 0.013). Similarly, a significant difference was detected between the two groups for miR-93 expression (p < 0.001). No significant difference was determined for BMPR2 (p = 0.725). Relative expression levels of miR-21, miR-93, and BMPR2 in healthy and cystic ovarian follicle groups. Bars represent mean ± SD. Different letters above the bars indicate statistically significant differences between groups for each gene (p < 0.05). No significant difference was observed between healthy and cystic groups for BMPR2 ROC analysis for miR‑21 and miR-93 The ability of miRNAs to discriminate between healthy and cystic ovarian tissues was evaluated using receiver operating characteristic (ROC) curve analysis to further assess their diagnostic accuracy (Fig. 5). miR-21 demonstrated moderate discriminative performance (AUC = 0.672), whereas BMPR2 showed no significant diagnostic value (AUC = 0.525). The highest discriminative performance was observed for miR-93 (AUC = 0.899). The optimal cut-off value was determined using the Youden index, defined as the point at which the sum of sensitivity and specificity was maximised. At a cut-off value of 1.78, miR-93 showed a sensitivity of 80%, a specificity of 82.86%, and a Youden index of 0.629. (Additional Files 2). ROC curve analysis graphs of miR-21, miR-93, and BMPR2 Correlation of BMPR2 with miR‑21 and miR‑93 expression According to the correlation analysis results, no statistically significant relationship was found between miR-21, miR-93, and BMPR2 in the healthy group. However, a significant and positive correlation was determined between miR-93 and BMPR2 in the cystic group (Pearson r = 0.490, p = 0.003; Spearman rho = 0.647, p < 0.001) (Additional Files 3). Discussion A modelling study estimating annual global losses reported an economic loss of approximately $600 million associated with ovarian cysts. This loss was linked to reduced milk yield, impaired fertility performance, and increased culling rates. This condition adversely affects not only the performance of individual cows but also herd-level productivity and the economic sustainability of the farm [32]. In bovine ovaries, the granulosa cell layer is highly variable. It is typically single-layered in primordial and primary follicles, becomes multilayered in secondary follicles (approximately 3–4 layers), and further increases in thickness in antral follicles; however, the number of granulosa cell layers may vary considerably among individual follicles [33, 34]. The progressive loss of granulosa cells, including their complete absence in certain areas, has been described as a characteristic feature of cystic ovaries, and the findings of the present study align with this degenerative pattern during cystic follicle development [35, 36]. It has been reported that cystic ovarian disease is also associated with molecular changes occurring at the local ovarian level; in particular, disruptions in receptor-mediated signaling pathways and the life and renewal processes of follicular cells are effective in this process [37]. In addition to these morphological alterations, studies in cattle have shown that cystic follicles are associated with changes in intraovarian regulatory factors. Altered expression of TGFB isoforms, particularly increased TGFB1 and TGFB2 levels in cystic follicles compared with tertiary and atretic follicles, has been reported. Given that TGFB isoforms regulate steroidogenesis, granulosa cell proliferation, and apoptosis, these alterations may contribute to granulosa cell degeneration, follicular persistence, and the pathogenesis of bovine cystic ovarian disease [38]. Several studies have shown that arrest in follicular development, altered apoptosis dynamics in granulosa cells, and disruptions in steroidogenesis are closely associated with changes in miRNA expression profiles [26–28]. Although reproductive physiology differs between species, the common mechanism underlying the development of PCOS in women and follicular cysts in cows is thought to be associated with alterations in the regulatory roles of miRNAs. Studies in both cows and women have shown that miR-21 is an anti-apoptotic miRNA acting on the ovaries [39–41]. Apoptosis of granulosa cells is considered a key determinant in the initiation of follicular atresia. Suppression of apoptosis is thought to delay physiological follicular regression, thereby contributing to the formation of cyst-like structures [42]. In line with this concept, altered insulin-related signaling has also been reported in bovine COF. Gareis et al. showed reduced p-mTOR and p-ERK1/2 expression in cystic follicles, suggesting impairment of pathways involved in granulosa cell proliferation, survival, differentiation, and steroidogenic activity. These findings further support the view that follicular persistence in COF is associated with dysregulation of local ovarian signaling mechanisms [43]. In cows, miR-21 is considered a protective miRNA in ovarian follicles, contributing to follicular cell survival and participating in ovulation-associated regulatory pathways. It has been reported that miR-21 expression increases in granulosa cells following the LH surge, and this increase contributes to the inhibition of apoptosis in granulosa cells in association with the suppression of caspase-3 activation [19, 44]. While another study demonstrated that miR-21 supresses autophagy in bovine granulosa cells, apoptosis was not directly assessed. It was suggested that the autophagic effect occurs through inhibition of the PI3K/AKT signaling pathway by targeting VEGFA by miR-21 [45]. In this study, the upregulation of miR-21 in cystic follicles is considered to be associated with granulosa cell dysfunction and impaired follicular maturation. Consistent with the current literature, it suggests that miR-21 plays a role in cyst formation and that this increased expression may contribute to persistent follicle development by suppressing follicular atresia. Although large-scale RNA-seq analyses of granulosa cells in follicular cysts in cows have reported the differentiation of numerous miRNAs, miR-21 and miR-93 were not included in these differential expression lists [30]. Specifically, there is currently no study in the literature that directly examines the relationship between miR-93 and the pathogenesis of cystic ovaries in cows. It has been reported that miR-93 is significantly increased in granulosa cells of PCOS patients, and this increase stimulates the apoptosis process in granulosa cells, leading to cellular dysfunction [46]. Another study reported that miR-93 was significantly up-regulated in rats with PCOS compared to the control group [47]. In the present study, the significant increase in miR-93 in the cystic ovarian group suggests potential molecular similarities between the cystic ovarian follicles observed in cows and PCOS in women. This finding indicates the presence of BMP pathway activity in cystic cases. Studies in cattle have shown that the BMP signaling pathway, particularly BMP2, BMP4, and BMP6, is altered during folliculogenesis and the early stages of follicular persistence, suggesting that these changes may contribute to the pathogenesis of cystic ovarian disease [48]. A study in bovine cumulus cells stimulated with BMP15/GDF9 demonstrated miRNA-mediated regulation of BMPR2, confirming that miR-93 suppresses BMPR2 expression as shown by both luciferase assay and RT-qPCR. This finding indicates that BMPR2 is a miRNA-regulated target in the bovine ovary [9]. In the present study, BMPR2 expression was found to increased in parallel with elevated miR-93 levels in the cystic group. ROC-based studies in women with PCOS have demonstrated the diagnostic value of specific miRNAs [15, 27, 49]. Similarly, the findings of the present study suggest that miR-93 has greater diagnostic potential than miR-21 in bovine cystic ovarian follicles. Overall, these findings indicate that miR-93 may represent a relevant molecular marker in bovine cystic ovarian pathology. The higher diagnostic potential of miR-93 compared with miR-21, together with its significant positive association with BMPR2 expression, suggests that miR-93 may be closely related to the molecular alterations observed in cystic ovarian follicles. However, the present findings do not support immediate field-level diagnostic use. If validated in clinically accessible samples, such as blood or follicular fluid, miR-93 may contribute to the molecular characterisation and classification of cystic ovarian conditions in cattle in future studies. Nevertheless, this study has several limitations. The regulatory effects of miR-21 and miR-93 on BMPR2 were assessed solely at the expression level, and the lack of functional validation (e.g., luciferase reporter assay) precludes confirmation of a direct target relationship. Furthermore, the lack of in vivo experimentation, the use of abattoir-derived samples, and the absence of serum or follicular fluid hormone measurements limit a more comprehensive interpretation of the functional status of the follicles. Conclusion In conclusion, this study potentially identifies miR-93 as a candidate molecular marker in bovine cystic ovarian follicles. Its higher diagnostic performance compared with miR-21 and its positive association with BMPR2 expression suggest that it may be involved in the molecular alterations associated with cystic follicle formation. If validated in clinically accessible samples, such as blood or follicular fluid, miR-93 may eventually contribute to the molecular characterization, classification, or future identification of cystic ovarian conditions in cattle. To our knowledge, this is the first study to evaluate miR-93 in bovine cystic ovarian follicles, thereby providing a basis for future functional and validation studies. Supplementary Information Supplementary Material 1. Supplementary Material 2. Supplementary Material 3. Acknowledgements This study is derived from the doctoral thesis of Merve YILMAZ entitled “Examination of the Effects of miR-21 and miR-93 on Ovarian Cyst Formation in Cows” which was completed and submitted in January 2026. The authors would like to thank Kıvılcım Sönmez for her contributions to the histopathological analyses, Bülent Ekiz for his support in the statistical analyses, and Atlas Biotechnology Laboratory and Kübra Öztürk for their assistance with the molecular analyses and Veterinarian Ebru ERDAL for her valuable assistance during the abattoir sampling procedures. Authors’ contributions MY: Writing – review & editing, Writing – original draft, Data curation, Conceptualization. GED, ABU, MS: Methodology, Writing GED: Supervision. Funding This research did not receive any private grants from any funding organization. Data availability The authors declare that all data supporting the findings of this study are included within the article. Declarations Ethics approval and consent to participate The Animal Experiments Local Ethics Committee of İstanbul University-Cerrahpaşa confirmed that ethical approval was not required for this study (Approval No: 1050387 2024/42). Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The authors declare that all data supporting the findings of this study are included within the article.