Introduction

Recurrent pregnancy loss, clinically termed as recurrent spontaneous abortion, poses significant emotional suffering for individuals enduring the condition, as it is considered one of the most complex pregnancy complications, affecting nearly 1–3% of couples trying to conceive [1]. Regardless of the differences in diagnostic criteria, RPL is most commonly described as two or more consecutive pregnancy losses prior to the 20 weeks of gestation. However, this is still a debatable definition, especially with respect to the criteria such as minimum threshold for the number of pregnancy losses, consecutiveness of losses and the cutoff for the gestational age [2]. To resolve this contention, WHO has recommended a guideline for defining the miscarriage as the loss of a pregnancy prior to the twenty second week bearing a fetal weight of less than five hundred grams [3].

The etiology of RPL is complex and multifactorial involving genetic abnormalities, anatomical malformation, endocrine dysfunction, placental pathology, infection, substance abuse (tobacco and alcohol), environmental factors, psychological stress and immunological disorders [4]. Among these, the genetic factors play significant role in RPL pathogenesis with numerous SNPs particularly microRNAs (miRNAs), being studied intensively in recent studies [5,6]. MiRNAs are small non-coding RNAs that function at the post-transcriptional level by binding to complementary sequences in the 3′-untranslated region of specific messenger RNAs, leading to mRNA degradation or inhibition of translation [7–9]. The microRNAs play significant roles in numerous pathological and physiological processes including cell division, differentiation, apoptosis, immune response and tissue remodeling. In addition to their broad influence, changes in the miRNA expression have been implicated in pregnancy-related complications such as preeclampsia, implantation failure and RPL [10–13].

According to many studies, approximately 30% of human genome is estimated to be regulated by miRNAs, underscoring their broad influence. Many miRNAs are highly expressed in reproductive tissues, suggesting their critical roles in maintaining successful pregnancies [5,14]. Beyond their role in reproduction, miRNA polymorphisms have emerged as important genetic modifiers in a wide spectrum of human diseases. In cancers, specific miRNA variants can alter the regulation of tumor suppressors and oncogenes, thereby affecting cell proliferation, apoptosis, and metastasis [15]. Similarly, in cardiovascular disease, miRNA polymorphisms have been linked to aberrant lipid metabolism, endothelial dysfunction, and vascular remodeling, while in diabetes, such variants influence insulin signaling pathways and β-cell survival, collectively highlighting their pleiotropic contribution to complex disorders including RPL [16,17].

Moreover, miRNA polymorphisms have also been associated with autoimmune diseases and reproductive pathologies [18]. Previous studies have associated SNPs in miRNAs like miR-149-rs2292832, miR-499-rs3746444, miR-10a-rs3809783, miR-323b-rs56103835, miR-125a-rs41275794/rs12976445 and miR-146a-rs2910164 with increased risk of RPL in different populations [19,20]. For instance, a SNP in pri-miR-10a has been shown to alter its maturation and potentially influence downstream gene regulation, contributing to RPL susceptibility in a Han-Chinese population [19]. Likewise, genetic variants in miR-146a and miR-196a2 have been linked to altered immune responses and inflammatory pathways in Iranian women with idiopathic RPL [20]. Notably, miR-146a has been involved in regulating the Fas gene, a critical mediator of oocyte apoptosis, suggesting a possible mechanistic link to RPL [21,22]. Based on these functional roles and previous genetic associations, this study investigated the association among six miRNA polymorphisms (miR-10-A > T, miR-125-G > A, miR-146a-C > G miR-149-T > G, miR-323b-C > T and miR-499-A > G) and idiopathic RPL in a cohort of Saudi women.

Materials and methods

Subjects selection

This study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of King Saud University (22/0629/IRB). Written informed consent was obtained from all participants, including both patients and healthy controls, who voluntarily agreed to donate peripheral blood samples for genetic and biochemical analyses. Recruitment was carried out at the outpatient department of King Khaled University Hospital.

The study cohort comprised 50 women diagnosed with recurrent pregnancy loss (RPL), defined according to the American Society for Reproductive Medicine (ASRM) criteria as two or more consecutive clinical pregnancy losses before 20 weeks of gestation, confirmed by ultrasound or histopathological examination [23]. Inclusion criteria were women aged 20–45 years with a history of ≥2 idiopathic consecutive miscarriages. Exclusion criteria included uterine malformations (septate uterus, fibroids), parental chromosomal abnormalities, endocrine disorders (uncontrolled diabetes, thyroid dysfunction, hyperprolactinemia), autoimmune diseases (antiphospholipid syndrome, systemic lupus erythematosus), coagulation disorders, TORCH infections, exposure to teratogenic drugs, smoking or alcohol abuse.

The control group consisted of 50 age-matched women with at least one natural live birth, no history of miscarriage, regular menstrual cycles, and absence of infertility. The same exclusion criteria applied to the RPL group were also applied to the control group to eliminate potential confounding factors.

The anonymized demographic and clinical data of healthy controls and RPL patients are provided as Supporting Information (S1 File and S2 File). The demographic, clinical, and haematological characteristics of the study population are summarized in Table 1. No significant differences were observed between cases and controls with respect to age and body mass index (BMI), whereas the number of pregnancies differed significantly between the groups (p < 0.001).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Demographic, clinical and hematological characteristics of population study.

https://doi.org/10.1371/journal.pone.0336432.t001

DNA sequencing and PCR amplification

PCR (polymerase chain reaction) was used to magnify the target regions containing single nucleotide polymorphisms (SNPs) in particular miRNAs. DNA samples were amplified using conventional polymerase chain reaction (PCR) using optimized thermal cycling parameters. Primer sequences and thermocycling conditions for each SNPs are given in Tables 2 and 3 respectively. A thermal cycler was used to perform PCR amplifications under controlled conditions. PCR amplicons were separated by agarose gel electrophoresis on 1.5% stained with SYBER Safe DNA Gel Stain and then visualized under UV transilluminator. Purified PCR amplicons were sequenced in the forward direction using Sanger sequencing on an ABI 3730xl DNA analyzer. The sequencing results were further visualized and analyzed using SnapGene software (GSL Biotech) to confirm the presence of complete nucleotide sequences and alignments. Representative sequencing chromatograms for SNP genotyping are provided in S1 Fig.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Primer sequences used for amplification of selected miRNA polymorphisms.

https://doi.org/10.1371/journal.pone.0336432.t002

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. PCR thermocycling conditions used for amplification of miRNA polymorphisms.

https://doi.org/10.1371/journal.pone.0336432.t003

Results

The genotype distribution of six mi RNA SNPs (miR-10 A > T, miR-125 G > A, miR-146a C > G, miR-149 T > C, miR-323b C > T, and miR-499 A > G) were examined in both RPL and control groups are displayed in Table 4. Representative sequencing chromatograms confirming SNP genotypes are shown in S1 Fig. Anonymized individual-level genotype and clinical data for all study participants are provided in the Supporting Information (S1 File and S2 File). Genotypes within the control groups were constant with Hardy-Weinberg Equilibrium (HWE) for miR-125 G > A, miR146a C > G, miR-323b C > T polymorphisms. and miR-499 A > G. However, miR-146a C > G showed a deviation from HWE in the control group.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. The genotype distribution of miR10 A > T, miR-125 G > A, miR125 T > C miR-146aC > G, miR-149T > C, miR-323b C > T, and miR-499 A > G Polymorphisms in the case and control groups.

https://doi.org/10.1371/journal.pone.0336432.t004

Discussion

MicroRNAs (miRNAs) are crucial post-transcriptional regulatory elements that play vital roles in complex biological processes, including centromere function, cell proliferation, stem cell differentiation and division, fat and cholesterol metabolism and most importantly female reproduction. They play most essential role in regulating oogenesis and spermatogenesis, as well as their precise control of embryo implantation and maternal-fetal communication [24–26]. Several studies have demonstrated the role of miRNAs in the female reproductive system across different mammalian species such as bovine, that have shown different expressions of miRNAs between small and large follicles or between healthy and atretic follicles. These miRNAs are involved in follicular cell proliferation, steroidogenesis, luteinization and oocyte maturation. Moreover, during the third day of the estrous cycle, miRNAs were also exhibited different expressions involved in the Wnt signaling, transforming growth factor (TGF-beta) signaling, axon guidance and apoptosis. By the day seven, the differentially expressed miRNAs shifted to the metabolism of vitamins, cofactors, lipids, lipoproteins amino acids such as cysteine and methionine [27,28].

Studies in humans and mice have also revealed low expression levels of the let-7b gene within granulosa cells of mature oocytes compared to the immature oocytes, suggesting the potential role in oocyte maturation [29,30]. Let-7b was first discovered in Caenorhabditis elegans and is highly conserved in humans and consists of 12 members of miRNA involved in development, stem cell biology, aging and metabolism [31]. miR-146a is one of the important regulatory molecules that play a role in controlling the expression of many genes particularly FOXL2, CCND2 and FAS [32]. FOXl2 (Forkhead box L2) belongs to FOX gene family, and highly conserved expression in adult human ovaries and play an effective role in ovarian and cellular differentiation [33,34]. FOXL2 also shows its highest levels of expression in the endometrium during the proliferative phase [35]. Its inactivation in murine models leads to infertility due to myometrial disorganization and vascular defects [36]. Similarly, CCND2 its predominantly expressed in granulosa cells and is essential for their proliferation during ovarian folliculogenesis [37]. It encodes cyclin D2 that facilitates in the transition from the G1 to the S phase during the cell cycle and its deficiency stops the cell cycle [20]. Moreover, miR-146a negatively regulates FAS expression by binding to its 3′-UTR which reduces apoptotic signaling [38]. FAS is reduced in women who are conflicting from recurrent pregnancy loss and unexplained infertility which results in the downregulation of apoptotic signaling in the epithelial cells during embryo implantation [39]. miR-146a itself expression is controlled by NF-κB signaling pathway and its upregulation enhances the survival and viability of mesenchymal stem cell (MSC) [40,41]. Suzuki et al. showed that treatment of MSCs with diazoxide (DZ) increased miR-146a expression and promoted cell survival via NF-κB pathway. Inhibition of NF-κB pathway or miR-146a knockdown mitigated this effect, indicating the critical function of miR-146a in cell survival [42]. Functionally, SNP rs2910164 within miR-146a may alter the stem-loop structure of its precursor, thereby reducing processing efficiency and mature miRNA abundance, which could dysregulate FOXL2, CCND2, and FAS expression, affecting folliculogenesis and endometrial receptivity [43].

miR149a is considered as a pro-apoptotic miRNA by inhibiting the expression of serine/threonine protein kinase AKT1 and transcription factor E2F1, both regulate cell proliferation and survival [44]. AKT1 is a multi-functional protein and involved in numerous cellular processes like growth and metabolism, while E2F1 regulates the G1/S phase transition and apoptosis [45]. SNP rs2292832 within miR-149 has been shown to influence Drosha and Dicer processing and strand selection, potentially altering its ability to downregulate AKT1 and E2F1. This dysregulation may disrupt the balance between granulosa cell proliferation and apoptosis, a key factor in successful embryo implantation [46]. miR-499 directly targets SOX6 gene which enhances the activity of terminal binding protein 2 (CtBP2), a repressor of transcription of fibroblast growth factor-3 (FGF-30). FGF-30 is essential for stimulating cell proliferation and differentiation, especially during the early stages of embryonic tissue development [47]. Similarly, miR-125 plays a critical role in organogenesis and adult tissues development, and associated with the TGF-β signaling which is vital for reproductive function and immune modulation. miR-125 also regulates angiogenesis in the endometrium and immune cell differentiation (T helper cells-Th17, T cells-Treg), and has been linked to abnormal pregnancy outcomes (preeclampsia when overexpressed) [48,49].

miR-10 family plays a pivotal role in repressing granulosa cell (GC) proliferation and induced apoptosis by targeting Brain-derived neurotrophic factor (BDNF), a key regulator of early follicular development and ovulation. Moreover, the miR-10 family and the TGF-β pathway form a negative feedback regulatory mechanism in granulosa cells [50]. Essential ovarian hormones and growth factors such as FSH, FGF9 and TGF-β pathway ligands (TGFβ1, Activin A, BMP4 and BMP15) inhibits miR-10a expression in GCs. Conversely, the miR-10 family suppresses multiple key genes within the TGF-β signaling pathway [51,52]. Furthermore, miR-323b negatively regulates paired-box 8 (Pax8), a transcription factor essential for embryo development, central nervous system, angiogenesis, immune regulation, and tumor metastasis as well as for the regulation of cell proliferation and differentiation [53,54]. However, patients with RPL show lower Pax8 expression and higher miR-323 levels [55]. These functional associations indicate that SNPs in miRNA genes may contribute to RPL pathogenesis by altering miRNA maturation, stability, or target binding, ultimately affecting key reproductive processes such as folliculogenesis, implantation, and immune modulation.

In our study, we investigated the association between certain SNPs in miRNA genes and prevalence of RPL in Saudi women, focusing on following variants miR-10-rs3809783, pri-miR-125a-rs41275794 and rs12976445, miR-146a-rs2910164, miR-149-rs2292832, miR-323b-rs56103835, miR-499A-rs3746444. Our results showed a significant positive relationship (p < 0.05) between susceptibility to RPL and two SNPs: miR-149T > C (rs2292832) and miR-146aC > G (rs2910164). For miR-146aC > G, our findings contrast with a meta-analysis of three studies which reported no statistically significant association with RPL [20,53,54]. However, we observed significantly higher genotype frequencies of this polymorphism in RPL cases compared to healthy controls (p < 0.05). Regarding miR-149T > C, our results align with a study conducted by reporting a significant association with RPL risk [56]. In contrast, they conflict with other reports that found no such association. For pri-miR-125a-rs12976445, miR-323b-rs56103835 and miR-499A-rs3746444, differential prevalence was observed between RPL cases and controls, but these differences were not statistically significant (p > 0.05). Previous studies and meta-analysis have reported that these SNPs may contribute to the risk of RPL [51,57,58], but our results did not support statistically significant association in Saudi population.

Possible reasons for observed deviation from Hardy–Weinberg equilibrium (HWE) in certain loci include small sample size, potential population stratification within studied cohort or technical errors in genotyping. However, all genotypes were rechecked to minimize chances of such errors. Additionally, unmeasured environmental exposures (smoking status, dietary habits, body mass index), lifestyle factors and other genetic variants may have acted as residual confounders, potentially influencing the observed associations.

This study has several important limitations. First, the small sample size may have reduced statistical power, particularly for detecting SNPs with small to modest associations. Second, deviation from Hardy–Weinberg equilibrium in some SNPs could reflect sampling bias or stratification. Third, the restriction to a Saudi population enhances population homogeneity but limits the generalizability of the findings to other ethnic groups. Fourth, the cross-sectional case–control design precludes causal inference. Finally, the absence of functional assays limits our ability to directly determine whether the identified SNPs alter miRNA maturation, stability, or target regulation. Future research should address these limitations by incorporating larger, multi-ethnic cohorts, collecting detailed environmental and lifestyle data, performing in vitro functional studies to elucidate the mechanistic role of these genetic variants.

Future studies should focus on multicenter cohorts with larger sample sizes to validate our findings across diverse populations. Moreover, functional studies are required to investigate the biological role of the identified miRNA SNPs in regulating gene expression and reproductive pathways. Integration of genomic, transcriptomic, and epigenetic approaches may further enhance our understanding of the molecular basis of recurrent pregnancy loss.

In conclusion, our study suggests that the miR-146aC > G and miR-149 T > C polymorphisms are significantly associated with increased risk of RPL in the Saudi population. These findings highlight the potential of miRNA related polymorphisms as biomarkers for reproductive outcomes and contribute to the increasing amount of data on genetic factors underlying RPL.

Supporting information