Emerging Roles for Noncoding RNAs in Female Sex Steroids and Reproductive Disease

Abstract

The “central dogma” of molecular biology, that is, that DNA blueprints encode messenger RNAs which are destined for translation into protein, has been challenged in recent decades. In actuality, a significant portion of the genome encodes transcripts that are transcribed into functional RNA. These noncoding RNAs (ncRNAs), which are not transcribed into protein, play critical roles in a wide variety of biological processes. A growing body of evidence derived from mouse models and human data demonstrates that ncRNAs are dysregulated in various reproductive pathologies, and that their expression is essential for female gametogenesis and fertility. Yet in many instances it is unclear how dysregulation of ncRNA expression leads to a disease process. In this review, we highlight new observations regarding the roles of ncRNAs in the pathogenesis of disordered female steroid hormone production and disease, with an emphasis on long noncoding RNAs (lncRNAs) and microRNAs (miRNAs). We will focus our discussion in the context of three ovarian disorders which are characterized in part by altered steroid hormone biology – diminished ovarian reserve, premature ovarian insufficiency, and polycystic ovary syndrome. We will also discuss the limitations and challenges faced in studying noncoding RNAs and sex steroid hormone production. An enhanced understanding of the role of ncRNAs in sex hormone regulatory networks is essential in order to advance the development of potential diagnostic markers and therapeutic targets for diseases, including those in reproductive health. Our deepened understanding of ncRNAs has the potential to uncover new applications and therapies; however, in many cases, the next steps will involve distinguishing critical ncRNAs from those which are merely changing in response to a particular disease state, or which are altogether unrelated to disease pathophysiology.

Introduction

Sex steroid hormones have broad physiologic roles and are essential for maintenance of reproductive capacity, including proper human development and function, sexual differentiation, secondary sex characteristics, sexual behavior patterns, cardiovascular health, various metabolic processes, and brain function. Sex steroid synthesis is a complex, multistep process that requires coordinated activity of multiple substrates and enzymes in the gonads (ovaries or testes) and other tissues (such as the adrenal glands). Tight spatiotemporal control of steroid production is essential for multiple processes involved in reproduction, including appropriate follicular development, ovulation, and the development of a synchronous endometrial lining for implantation. Many female reproductive disease states - including polycystic ovary syndrome, early pregnancy loss and failed pregnancy, and endometriosis - can be characterized by aberrations in steroid hormone levels.

Over the past decade, a growing body of evidence has shown that noncoding RNAs (ncRNAs) can regulate the physiologic mechanisms of sex steroid biosynthesis and secretion. These noncoding RNAs, which are not transcribed into protein, play critical roles in a wide variety of biological processes. Literature regarding the role of ncRNAs as critical post-transcriptional regulators in many diseases has rapidly expanded. Moreover, an expanding breadth of clinical and translational studies has emerged with respect to ncRNAs which sheds new light on the existing basic research in this field. However, there are many challenges to overcome in order to properly understand the effects of aberrant ncRNA production on female steroid hormone biology and disease. Here, we provide an updated narrative review summarizing the role of differentially expressed ncRNAs in the pathogenesis of disordered female steroid hormone production and highlighting new observations regarding these roles. We will review these ncRNA regulators and frame our discussion in the context of recent discoveries linking ncRNAs to the pathogenesis of female reproductive disorders which are characterized in part by altered steroid hormone biology, including diminished ovarian reserve, premature ovarian insufficiency, and polycystic ovary syndrome. We will highlight miRNAs and lncRNAs as the most well-studied ncRNAs in the area of sex hormone regulation and production (Figure 1). We will also discuss the limitations and challenges faced in studying noncoding RNAs and sex steroid hormone production. An enhanced understanding of the role of ncRNAs in sex hormone regulation networks is essential in order to advance development of potential diagnostic markers and therapeutic targets for diseases, including those in reproductive health.

Figure 1. Associations between ncRNAs and steroid hormone production.

miRNAs and ncRNAs are grouped by studies demonstrating associations with estradiol and progesterone production and GC proliferation and apoptosis. NcRNAs which promoted estradiol or progesterone production are shown in green; those with suppressive effects are shown in red. For proliferation/apoptosis, ncRNAs with pro-proliferative effects are shown in green; those with apoptotic effects are shown in red. NcRNAs with multiple studies/overlapping effects are bolded.

Overview of noncoding RNA classes

The “central dogma” of molecular biology, that is, that DNA blueprints encode messenger RNAs which are destined for translation into protein, has been challenged in recent decades. In actuality, a significant portion of the genome encodes transcripts that are transcribed into functional RNA. Research over the last two decades has revealed new classes of ncRNAs, commonly grouped by length as small ncRNA (sncRNA, <200 nucleotides) and long ncRNA (lncRNA, >200 nucleotides)¹. Small noncoding RNAs are ncRNAs less than 200 nucleotides, including miRNA, small nucleolar RNA (snoRNA), small nuclear RNA (snRNA), small interfering RNA (siRNA), transfer RNA (tRNA) and piwi-interacting RNA (piRNA) ². Perhaps the most well-characterized small RNAs are miRNAs, 21–22 nucleotide ncRNAs that suppress protein expression either by silencing mRNA translation or leading to target mRNA degradation. miRNAs recognize their target mRNA 3’UTR sites by their first 8 residues on the 5’end (the “seed sequence”) and form Watson-Crick base pairing³. The class of small noncoding RNAs known as endogenous siRNAs also function to alter mRNA translation via cleavage of target mRNAs. Like miRNAs, siRNAs recognize their target mRNA 3’UTR sites; however, unlike miRNAs, they require full complementarity in order to suppress translation of the target transcript.

The function of lncRNAs, which are more than 200 nucleotides in length, is diverse. LncRNAs can be further subdivided into long intron ncRNA, long intergenic ncRNA (lincRNA), circular RNAs (circRNAs)⁴, and natural antisense transcripts^5–8. Most lncRNAs are located in the nucleus, where they can act as molecular scaffolds to assist in alternative splicing or modification of chromatin structure^9–12. However, emerging evidence suggests that certain lncRNAs function in the cytoplasm in activities such as regulating translation, promoting or inhibiting messenger RNA (mRNA) degradation, and acting as miRNA sponges^13,14. LncRNAs can interact with DNA, RNA, and proteins, and act in a variety of functions, including as molecular scaffolds¹⁵, guides to a specific target locus¹⁶, decoys or sponges¹⁷, and enhancers of transcriptional activity¹⁸.

piRNAs, a class of small RNAs found almost exclusively in germ cells^19,20, form RNA-protein complexes by binding to a specific class of Ago proteins known as Piwi proteins; these piRNA-protein complexes are involved in epigenetic and post-transcriptional gene silencing of transposable elements in germ cells (particularly spermatogenic cells). piRNAs are synthesized from long, single-stranded RNA precursor sequences, are larger than miRNAs (26–31 nt), and their processing does not require Dicer; in many respects their mechanism of biogenesis still remains unclear. While two studies have reported differential expression of piwi-interacting RNAs in GCs and oocytes from women with DOR^21,22, female mouse Piwi mutants do not display defective oocytes, in contrast to Piwi protein mutant male mice, which exhibit altered spermatogenesis and depletion of spermatogonia²³. Thus, while essential for male fertility, piRNAs do not appear to be essential for female gametogenesis and fertility.

Noncoding RNAs in ovarian reproductive disease states

Premature ovarian insufficiency

In women with premature ovarian insufficiency (POI), the typically gradual process of follicle loss is catastrophically accelerated, causing loss of fertility and early menopause. The great majority of these cases are idiopathic²⁴, although ovarian failure can be accelerated by exposure to gonadotoxic agents such as chemotherapy^25–28. The resultant early menopause and infertility is devastating for women. Primary ovarian insufficiency is characterized by the development of abnormal menstruation, increase in serum gonadotropins and decrease in serum estradiol (E₂) due to depletion or dysfunction of ovarian follicles in women less than 40 years old.

Although most studies evaluating miRNA profiles in POI have been performed in murine models, small-scale human studies have examined miRNA expression within small cohorts of patients with POI and in limited ethnic groups. Most available studies on the association between ncRNAs and POI have focused on miRNA expression as it relates to aberrant ovarian follicle growth, which promotes sex steroid hormone dysregulation. Several miRNAs have been shown to be dysregulated in POI, although mechanistic data is limited. In a study of differentially expressed miRNAs in 140 women of Han Chinese ancestry compared to 140 controls, 51 miRNAs were found to be differentially expressed (22 upregulated and 29 down regulated)²⁹. This finding is of particular interest because miR-22–3p downregulates FSH secretion and promote GC apoptosis, and also suppresses estrogen receptor 1 (ESR1) and phosphatases and tensin homolog (PTEN), which are considered potential POI candidate genes^29,30. Another study evaluating miRNA expression in women with POI demonstrated increased miR-146a in the plasma and GC of POI patients relative to controls. miR-146a targets IRAK1 and TRAF6 through the caspase cascade pathway and promotes GC apoptosis ³¹. Studies have also explored the relationship between miRNA polymorphisms and the incidence of POI. One study showed that the gene-gene interaction of three specific miRNA polymorphisms expressed in ovarian tissue, miR-27a A> G, miR-423 C> A and miR-608 G> C, is associated with an increased risk of POI ³². Another study identified a specific polymorphism in miR-449b (rs10061133 A> G), a miRNA tissue-specific to ovaries and testes, as associated with POI³³. Target genes of miR-449b include CDK4,6,15, HDAC1, E2F1 and TP53, which are involved in cycle regulation and apoptosis. Both miR-449b and another miRNA, miR-320, specifically inhibit E2F1, a transcription factor which regulates GC steroidogenesis³⁴.

Studies on the mechanistic regulation of sex steroid hormone synthesis by miRNAs in POI are far more limited. The enzymes involved in sex steroid synthesis are members of the cytochrome P450 (CYP) family; Nakajima et al summarized the regulation of cytochrome P450s and nuclear receptors by miRNAs and their association with a variety of diseases including cancer, diabetes, and cardiovascular diseases and for brevity these will not be re-reviewed here³⁵. The rate-limiting step for production of sex steroid hormones, including estradiol and progesterone, is the conversion of cholesterol to pregnenolone (a steroid precursor) through a side chain cleavage enzyme. This step of steroidogenesis is controlled by steroidogenic acute regulatory protein (StAR). During the early phase of the menstrual cycle, theca cell expression of StAR allows for follicular production of androgens (a substrate for follicular estrogen production). At the time of luteinization, granulosa cell (GC) StAR expression undergoes a dramatic upregulation, promoting conversion to a progesterone-dominant microenvironment, which allows endometrial decidualization and support of a growing pregnancy. Activation of the cAMP signaling pathway and other transcription driven processes are dominant mechanisms for rapidly increasing StAR transcripts and promoting steroid production. However, noncoding RNA-mediated mechanisms have been increasingly explored as regulators of StAR and POI. The forkhead box L2 gene is mutated in patients with blepharophimosis-ptosis-epicanthus inversus syndrome (BPES), a condition which results in POI ³⁶. Foxl2 expression in the ovary regulates StAR as well as several other steroidogenic genes (including Cyp19a1, Cyp11a1, and Cyp11a1), and is essential for granulosa cell function and steroid hormone biosynthesis (reviewed in ^37,38). The miRNA miR-133b binds the 3’ untranslated region of Foxl2 mRNA; miR-133b overexpression downregulates Foxl2 expression in human and mouse granulosa cells, inhibiting Foxl2-mediated transcriptional repression of StAR and CYP19A1 and promoting estradiol production³⁹. In human granulosa lutein cells, miR-96 knockout inhibits progesterone production via the transcription factor FOXO1, another forkhead box protein⁴⁰. Hu et al showed that miR-132, which is expressed in ovarian granulosa and adrenal cells, attenuates progesterone production through dual mechanisms: by decreasing StAR protein levels and promoting expression of 3β-HSD and 20α-HSD, leading to inactivation of progesterone via conversion to the biologically inactive 20α-OHP⁴¹.

Other studies have utilized chemotherapy-induced gonadotoxicity as a mechanism to study a potential causal and/or protective role of miRNAs in POI and dysregulated sex steroid hormone production. In one study utilizing ovarian tissue from a rat POI model induced by 4-vinylcyclohexene diepoxide (a chemotherapeutic agent), a total of 83 differentially expressed miRNAs were identified (63 up-regulated and 20 down-regulated) after gonadotoxin exposure and induction of POI. Of the miRNAs up-regulated in POI tissue, miR-27b and miR-190 were found to suppress steroidogenesis, and miR-151 and miR-672 found to promote GC apoptosis. Two of the down-regulated miRNAs, miR-29a and miR-144, were found to target phospholipase A2 group 4a (Pla2g4a), the enzyme producing arachidonic acid (the cyclooxygenase-2 substrate)⁴². A study using a cyclophosphamide-induced POI rat model found that overexpression of miR-21 in MSCs resulted in increased ovarian weight and follicle count, increased E₂ level and decreased FSH level⁴³, and decreased apoptosis of chemotherapy-damaged GC via targeting of PTEN and programmed cell death protein 4 (PDCD4). In another study, miR-10a derived from amniotic fluid stem cells had an anti-apoptotic effect on chemotherapy damaged GCs⁴⁴, while in work using a cisplatin-induced POI mouse model, miR-125a-5p overexpression resulted in lower progesterone and estradiol levels via targeting of the STAT3 pathway. These results suggest that miR-125a-5p may have a role in the pathogenesis of aberrant steroidogenesis involved in POI⁴⁵. Lastly, an induced POI mouse model utilizing tripterygium glycosides (TGs) demonstrated that miR-15a promotes apoptosis of chemotherapy-damaged GC via inhibiting the Hippo-YAP / TAZ pathway, leading to POI⁴⁶. However, while these studies suggest a potential protective function of miRNA in POI, and while RNA-based therapeutics are already in use in other fields^47–50, to date no studies exist evaluating the effect of miRNA-based therapies in the clinical setting of POI.

Diminished ovarian reserve and infertility

Infertility and pregnancy loss are traumatic for women and families^51,52. Frequently, infertility and pregnancy loss can be directly attributed to ovarian aging. With age, a drastic decline in the quantity of follicles and oocytes (the “ovarian reserve”) occurs⁵³, leading to infertility. Additionally, women are waiting longer to conceive, compounding these problems⁵⁴. While not directly tied to infertility, several miRNAs have been shown to repress estradiol synthesis, a marker of granulosa cell function. Overexpression of 172 synthetic miRNA precursors by transient transfection in primary human granulosa cells identified 51 miRNAs that suppressed E₂ release, including miR-15a, miR-24, let-7d, let-7 g, miR-125a, miR-125b, miR-98, and miR-29a; none stimulated E₂ release and the mechanism for these effects was not studied⁵⁵. In another study, miR-1275 repressed estradiol synthesis and resulted in GC apoptosis by targeting and inhibiting liver receptor homolog (LRH) −1, a CYP19A1 transcription factor⁵⁶. Other miRNAs shown to indirectly regulate CYP19A1, including miR-320, miR-764–3p, and miR-383, directly target E2F1 and steroidogenic factor (SF)-1, decreasing E₂ production by downstream suppression^34,57.

While diminished ovarian reserve affects 10% of women seeking fertility treatment⁵⁸, methods of determining ovarian reserve all have their drawbacks, and all available tests are considered diagnostic rather than prognostic^59–61. Thus, miRNAs have also been explored as mechanistic and prognostic indicators of diminished ovarian reserve. Differential expression profiles of miRNAs have been linked with dysregulated steroidogenesis and diminished ovarian reserve (DOR) in plasma and serum^29,62, whole ovaries⁶³, follicular fluid⁶⁴, granulosa cells^62,65,66, and oocytes⁶⁷ (Figure 1). One study of women grouped by different levels of ovarian reserve function showed twenty conserved and three new miRNAs upregulated in GC from follicular fluid in the DOR group. Subsequent testing showed that one of the conserved miRNAs in the DOR group, miR-23a, targets sirtuin (SIRT)1 and promotes apoptosis in GCs by inhibiting the ERK1/2 signaling pathway⁶⁶. Decreased expression of miR-106a, which promotes apoptosis signal regulating kinase (ASK)1^62,, was found in serum and GC of women with DOR. In another study examining differential expression of miRNAs in GC in young women with DOR, a group of differentially expressed miRNAs was associated with regulation of cellular growth and proliferation primarily via the MAPK, Wnt and TGF-beta pathways⁶⁵. Importantly, while differences in miRNA expression correlate with clinical outcomes in these patient populations, it is not yet clear whether aberrant miRNA expression causes or is a result of physiological differences within the follicular environment. Detailed mechanistic data regarding effects on sex steroid production is not available for most miRNAs associated with DOR, although most miRNAs identified were associated with GC cell proliferation and survival which would be expected to impact GC steroidogenesis. miR-100–5p and miR-16–5p also target IGF1R (insulin-like growth factor 1) which has numerous actions in the ovary including enhancement of progesterone biosynthesis (reviewed in ^68–70).

Little data exists on the role of siRNAs in female steroidogenesis directly; however, studies using conditional knockouts of ncRNA processing pathways have attempted to delineate the importance of siRNAs in ovarian folliculogenesis. Dicer is an RNase III enzyme that processes double-stranded RNA precursors into miRNAs and siRNAs. Multiple RNA interference pathways utilize Dicer. Selective knockout of Dicer in fetal oocytes during prophase I results in disruption of folliculogenesis, POI and infertility, indicating that ncRNAs are essential for normal folliculogenesis and may play a key role in the occurrence of POI⁷¹. However, mouse oocytes express both miRNAs and siRNAs, whose synthesis is also Dicer-dependent. In mice deficient in DGCR8, an RNA-binding protein required solely for miRNA biogenesis, siRNA production remains intact, and folliculogenesis is not disrupted, oocyte development progresses normally ⁷². This finding has been interpreted to reflect that it may be siRNAs, rather than miRNAs, that are ultimately essential for normal follicular development.

As expected, lncRNAs are also much less well studied with respect to steroid hormone production and fertility. The lncRNA H19 belongs to a highly conserved imprinted gene cluster, containing genes essential for growth and development. We have shown that H19, via the miRNA intermediary, let-7, represents a novel mechanism by which StAR can be regulated at the post-transcriptional level. Loss of H19 in vitro in mouse granulosa cells and in vivo in an H19KO mouse model results in disrupted StAR and progesterone production and subfertility^73,74. H19 acts as a molecular “sponge” for let-7, binding and modulating its availability¹⁷; let-7 functions as a negative regulator of target genes ¹⁷. We showed that StAR is a novel let-7 target gene⁷³, revealing a novel mechanism by which H19 can support the rate-limiting step in steroidogenesis via let-7. We also observed that loss of H19 also results in decreased ovarian expression of anti-Müllerian hormone (AMH), which regulates the size of the follicular pool by inhibiting the initial recruitment of primordial follicles into the growing pool and modulating the sensitivity of growing follicles to follicle stimulating hormone numbers. The AMH mRNA contains putative binding sites for let-7⁷⁵, and let-7 transfection leads to decreased AMH expression in GCs⁷⁵, suggesting a potential role for H19 and let-7 in the regulation of AMH and ovarian reserve⁷⁵. Previous studies have demonstrated that let-7 expression is increased in GCs from women with diminished ovarian reserve⁶² and in plasma from women with POI ²⁹. Steroid receptor RNA activator (SRA), another lncRNA, enhances transcription of steroid receptors including the progesterone receptor (reviewed in ⁷⁶) and increases expression of Cyp11a1 and progesterone production in mouse granulosa cells in vitro⁷⁷. Loss of Neat1, a lncRNA which localizes to nuclear bodies and is expressed in corpus luteum, results in infertilty with normal ovulation⁷⁸, a phenotype which is rescued by administration of progesterone⁷⁸, suggesting a role for Neat1 in luteal function and progesterone production.

Therefore, although there is growing evidence for the mechanism by which ncRNAs may be related to the pathogenesis of DOR and POI, several issues remain to be resolved. First, many clinical studies rely on small sample sizes; larger groups would allow for more robust conclusions. Second, it is unclear whether dysfunction of a single miRNA can cause DOR or POI. Studies utilizing targeted ncRNA knockdown in vitro and in vivo will be helpful in delineating the specific role of loss of function of a single ncRNA in steroidogenesis and follicular growth. Third, the regulatory and functional networks of most miRNAs are still unknown, and software-specific prediction targets for specific miRNAs need to be further validated to help understand the genetic basis of DOR and POI and develop new treatment methods for these conditions.

Polycystic ovary syndrome

Polycystic ovary syndrome (PCOS) is a disorder characterized by dysregulated steroid hormone production. PCOS affects up to 20% of reproductive aged women and is associated with a number of serious metabolic comorbidities including metabolic syndrome, hypertension, dyslipidemia, and diabetes, as well as reproductive consequences including infertility, pregnancy loss, and endometrial cancer. The high prevalence rate of PCOS makes it the most common cause of anovulatory infertility in premenopausal women. Women with PCOS frequently demonstrate elevated levels of virtually all androgens measured⁷⁹ and up to 80% of women with PCOS exhibit clinical manifestations of hyperandrogenism⁸⁰. High androgen levels are strongly associated with many of the morbidities associated with PCOS including follicular arrest (leading to anovulation), central obesity, and development of the metabolic syndrome ⁸⁰. Women with PCOS also exhibit decreased progesterone production, even after treatment with ovulation inducing agents such as metformin⁸¹.

The presence of hyperandrogenemia is strongly associated with many PCOS-related comorbidities including anovulation as well as the development of metabolic syndrome⁸⁰. Studies of miRNA-mediated regulation of androgens have primarily focused on Leydig cells of the testes, which are the primary source of testosterone in males. miR-150 directly inhibits StAR at the post-transcriptional level in Leydig cells⁸². One study showed five miRNAs (miR-29a, miR-29c, miR-142–3p, miR-451 and miR-335) are involved in the regulation of androgens in immature Leydig cells via regulation of basic fibroblast growth factor (bFGF), suggesting that changes in mRNA mediated by bFGF can inhibit LH-stimulated testosterone biosynthesis ⁸³. Other studies have reported that Scavenger receptor class B type 1 (SR-B1), a high-density lipoprotein (HDL) receptor, is essential for the selective uptake of cholesterol esters in steroidogenic cells and is regulated by pre-miRNA-125a and pre-miRNA-455 in Leydig cells. Expression levels of SR-B1 and cholesterol ester uptake was reduced after transfection of pre-miRNA-125a and pre-miRNA-455; suggesting that these miRNAs also have roles in the inhibition of steroidogenesis^84,85. In women, miR-23a expression in serum of women with PCOS is inversely related to serum total testosterone levels, with each doubling of miR-23a expression decreasing the probability of PCOS by 0.01-fold⁸⁶.

Like women with DOR and POI, women with PCOS exhibit aberrant miRNA expression in serum, follicular fluid, granulosa cells and cumulus cells. One study evaluating circulating miRNA expression in patients with PCOS compared to healthy male and female controls found five differentially expressed miRNA in PCOS patients. An additional 15 other circulating miRNAs found in PCOS patients had similar expression patterns only in male controls and 13 others had similar expression patterns only in female controls, suggesting not only aberrant miRNA expression in PCOS but also potential sexual dimorphism of miRNA expression. The transcription levels of several miRNAs were associated with hirsutism scores, obesity, and metabolic dysfunction indicators⁸⁷, indicating a potential effect of metabolic parameters on obesity and miRNA expression. Women with PCOS also have aberrant follicular fluid levels of multiple miRNAs, including increased expression of miR-9, −18b, −32, −34c, and −135a levels⁸⁸ and decreased miR-145 expression in women with PCOS⁸⁹.

Aberrant GC expression of miRNAs involved in granulosa cell proliferation has also been observed. In granulosa cells from ovarian cortex excised from women with PCOS, expression of miR-93, which attenuates cyclin-dependent kinase inhibitor (CDKN)1A and promotes ovarian GC proliferation⁹⁰, is higher than in controls. miR-145, which regulates cellular metabolism, survival and apoptosis via inhibition of the MAPK/ERK signaling pathway, is decreased in granulosa cells isolated from follicular fluid from women with PCOS; overexpression of miR-145 in culture inhibits cell proliferation and promotes apoptosis of GC. Thus, it has been speculated that this low expression of miR-145 may promote the abnormal proliferation of GCs observed in PCOS women, potentially contributing to the inappropriately high number of simultaneously growing follicles observed in women with PCOS⁸⁹. miRNA expression is also dysregulated in cumulus cells (CCs) of women with PCOS and has been implicated in dysregulated sex steroid hormone production. miR-509–3p, which promotes estradiol secretion by targeting the MAP3K8 pathway, is upregulated in CC of PCOS patients⁹¹. miR-320a expression is reduced in CC of patients with PCOS; miR-320a has been linked to decreased estradiol production through the RUNX2 / CYP11A1 (CYP19A1) cascade⁹².

Little data exists regarding expression patterns of lncRNAs and their relationship with PCOS and aberrant steroidogenesis. In the reproductive tract, circulating H19 is higher in women with PCOS compared to controls⁹³. A single study evaluating lncRNA microarray expression profiles of PCOS granulosa cells as compared to healthy controls have revealed differential transcript expression, with 862 lncRNA transcripts and 998 mRNA transcripts differentially expressed in PCOS GC. The differentially expressed lncRNA-LincRNA human antigen complex group 26 (HCG26) was upregulated in granulosa cells from women with PCOS; inhibition of HCG26 in cultured luteinized granulosa cells were found to have increased expression of CYP19A1 and estradiol production⁹⁴.

Summary and Future Directions

Studies demonstrating dysregulation of ncRNAs in reproductive health-related diseases such as PCOS, POI, and DOR using various human biological fluids including serum, plasma, and follicular fluid have demonstrated that ncRNAs may represent promising biomarkers for diagnosis of reproductive disease. However, the identification of upstream and other regulatory factors of miRNAs, their target genes, and their role in related signaling pathways is essential in order to further clarify the mechanistic role of specific ncRNAs in disordered steroid hormone production and female reproductive disease. In future studies, it will be critical to distinguish between ncRNAs which modulate biological pathways regulating steroid hormone pathophysiology in vivo and those which merely function in vivo, as well as to distinguish those ncRNAs whose expression is changing in response to disease rather than as a driver of disease. Resolution of these questions has the potential to pave the way for new therapeutic strategies affecting key genes in specific regulatory networks involving ncRNAs.