Hippo Signaling in the Ovary: Emerging Roles in Development, Fertility, and Disease

Abstract

Emerging studies indicate that the Hippo pathway, a highly conserved pathway that regulates organ size control, plays an important role in governing ovarian physiology, fertility, and pathology. Specific to the ovary, the spatiotemporal expression of the major components of the Hippo signaling cascade are observed throughout the reproductive lifespan. Observations from multiple species begin to elucidate the functional diversity and molecular mechanisms of Hippo signaling in the ovary in addition to the identification of interactions with other signaling pathways and responses to various external stimuli. Hippo pathway components play important roles in follicle growth and activation, as well as steroidogenesis, by regulating several key biological processes through mechanisms of cell proliferation, migration, differentiation, and cell fate determination. Given the importance of these processes, dysregulation of the Hippo pathway contributes to loss of follicular homeostasis and reproductive disorders such as polycystic ovary syndrome (PCOS), premature ovarian insufficiency, and ovarian cancers. This review highlights what is currently known about the Hippo pathway core components in ovarian physiology, including ovarian development, follicle development, and oocyte maturation, while identifying areas for future research to better understand Hippo signaling as a multifunctional pathway in reproductive health and biology.

Graphical Abstract

Graphical Abstract.

ESSENTIAL POINTS.

• The Hippo signaling pathway is an evolutionarily conserved pathway that regulates organ size via the control of cellular proliferation, differentiation, apoptosis, and cell fate

• Regulation of the Hippo signaling cascade is cell-type and context dependent and is facilitated by various signals including the extracellular matrix, mechanical stress, G protein–coupled receptors, growth factor receptor tyrosine kinases, protein kinases and phosphatases, and cellular metabolism

• The Hippo pathway and its effector YAP1 are critical for fine tuning ovarian physiology including cell fate determination, follicular activation, growth and differentiation, and steroidogenesis

• In vivo mouse models and in vitro culture systems show that dysregulation of the Hippo pathway resulting in YAP1 over activation in granulosa cells contributes to loss of tissue homeostasis and reduced fertility

• YAP1 is implicated in the ovarian PCOS phenotype and in metabolic disorders

• Modulation of upstream Hippo pathway components leading to YAP1 activation leads to follicle activation and may improve assisted reproduction outcomes

• Hyperactivation of YAP1 is implicated in the development and progression of ovarian cancer and resistance to chemotherapeutics

Introduction to Hippo Signaling

The ovary is a dynamic structure, undergoing extensive tissue remodeling with each reproductive cycle. This cyclic remodeling of the ovary includes the processes of folliculogenesis and atresia, ovulation, luteal formation and regression, and the accompanying vasculature and microenvironment changes, all of which are required for successful reproduction (1-3). Paracrine and endocrine-induced alterations in ovarian morphology begin during the development of the gonads and continues as the organ matures. The adult ovary undergoes multiple structural and developmental processes during the reproductive life span of women. Cellular proliferation and differentiation are regulated by a host of paracrine, autocrine, and endocrine factors. Errors in these processes either during fetal life or after menarche can result in ovarian dysgenesis, infertility, and pathology (4).

The follicle is the functional unit of the ovary. The most basic follicular structure is the primordial follicle, consisting of an oocyte surrounded by a few squamous granulosa cells and basal lamina (5). Intraovarian growth factors and secretion of pituitary gonadotropins influence the cyclic growth of ovarian follicles from gonadotropin-independent growth to gonadotropin-dependent growth (6, 7). The transition from primordial follicle to primary follicle occurs with a dramatic increase in cytoplasmic and nuclear volume in the oocyte along with changes in morphology and proliferation of the surrounding granulosa cells (5, 8). As the follicle progresses from a primary to a secondary stage follicle, the granulosa cells surrounding the oocyte continue to proliferate and increase in size (5). Other structural changes begin to occur during the transition to secondary follicle, such as the generation of the theca cell layer (5). The theca cells proliferate from a population of unspecialized mesenchymal cells in the ovarian stroma to form a layer adjacent to the basal lamina (9). Additionally, the follicle begins to vascularize, enabling the follicle to now secrete and respond to steroid hormones (5). In preantral follicles, theca cells respond to luteinizing hormone (LH), while follicle-stimulating hormone (FSH) binds to differentiated granulosa cells, resulting in the production and release of estradiol (10). The granulosa cells of the growing antral follicles compete for FSH, and a dominant follicle emerges, leaving the other follicles to undergo atresia. Estradiol secretion reaches its peak in mature follicles, triggering the LH surge that is required for ovulation (11). Postovulation, the remaining granulosa and theca cells of the ruptured follicle become luteinized and vascularized to form the corpus luteum, contributing to the increase in circulating progesterone and other hormones responsible for the establishment of pregnancy and early embryo development (12, 13).

The regulation of follicle development and subsequent luteal formation via endocrine signaling mechanisms have been well established (14-16). Research on intracellular signaling pathways in the ovary has identified several pathways that regulate follicle cell development, growth, and cell fate such as adenylyl cyclase/cAMP/protein kinase A (PKA) and phospholipase C/calcium/protein kinase C (PKC) (17-19), Notch (20), Hedgehog (21-23), and Wnt (24-26) signaling, members of the TGF-B superfamily (27-30), many receptor tyrosine kinases (RTKs) (31-35), and nuclear receptors (36). A new intracellular signaling pathway called the Hippo pathway (37, 38) has emerged having considerable influence on cellular differentiation and steroidogenesis in the ovary and will be the subject of this review.

The Hippo signaling pathway, also referred to as Salvador/Warts/Hippo pathway was initially discovered as an integral regulator of imaginal disc size in Drosophila melanogaster (39) and later studies implicated the Hippo pathway in Drosophila ovarian development (40-44). The Hippo pathway also interacts with several of the pathways mentioned above to facilitate ovarian development and function as demonstrated by comprehensive studies in the Drosophila ovary (42, 44-49). This evolutionarily conserved pathway regulates organ size (50) via control of cell proliferation (51), apoptosis (52), and stem cell self-renewal (53), with dysfunction of the pathway contributing to a wide range of diseases (54), tissue overgrowth (51), and tumorigenesis (54). The foundation of this pathway consists of a protein kinase cascade (Fig. 1) of several negative regulators, inclusive of a complex formed by mammalian serine/threonine-protein kinase 4 and 3 (commonly called MST1/2) and Salvador family WW domain-containing protein 1 (SAV1). The MST/SAV1 complex phosphorylates and activates large tumor suppressors 1 and 2 (LATS1/2) and its regulatory protein Mps one binder 1 (MOB1). The activated LATS1/2 complex then phosphorylates yes-associated protein 1 (YAP1) and WW domain-containing transcription regulator 1 (WWTR1, also known as TAZ), the main effectors of this signaling pathway (55). Elevated phosphorylation of YAP1 and TAZ promotes interaction with 14-3-3 proteins, thus sequestering them to the cytoplasm or targeting the proteins for ubiquitin-mediated proteolytic degradation (56). Unphosphorylated YAP1 and TAZ translocate and accumulate in the nucleus when Hippo signaling is inactive, binding to TEA domain (TEAD) transcription factors, which promote the expression of genes involved in cell proliferation and survival (57).

Figure 1.

Diagrammatic representation of the Hippo signaling pathway and mechanism of action in mammalian cells. The Hippo signaling pathway has many upstream regulators, from left to right: G protein–coupled receptors (GPCRs), extracellular matrix (ECM) stiffness and mechanical stress, integrins, and growth factor receptor tyrosine kinases (RTK). When core Hippo components MST1/MST2 and LATS1/LATS2 and their respective regulatory proteins SAV1 and MOB1 are phosphorylated, the Hippo pathway is active, suppressing the transcriptional activity of YAP1/TAZ by phosphorylation and sequestering YAP1/TAZ to the cytoplasm where they bind to 14-3-3 proteins or undergo protein degradation. When the Hippo pathway is inactive, dephosphorylated YAP1/TAZ enter the nucleus and bind with TEA domain transcription factors, TEADs. YAP1 and TAZ can also interact with other nuclear transcription factors. Additionally, other signaling pathways (WNT, TGF-β, and Notch) may interact with YAP1 and TAZ in order to regulate target gene expression. Abbreviations: EGF, epidermal growth factor; G, guanine nucleotide binding protein; LATS1/2, large tumor suppressor 1 and 2; MAP4K, mitogen-activated protein kinase kinase kinase kinase; MOB1, Mps one binder 1; MST1/2, mammalian STE20-like; PI3K, phosphoinositide 3-kinase; SAV1, Salvador family WW domain-containing protein 1; STRIPAK, striatin-interacting phosphatase and kinase; TAZ, transcriptional coactivator with PTZ-binding motif also called WW domain-containing transcription regulator 1, WWTR1; TGF, transforming growth factor; YAP1, yes-associated protein 1, ZO-2, zona occludens-2.

Due to lack of direct DNA-binding activity, YAP1 and TAZ bind to a wide range of transcription factors including TEADs, TP73, ERBB4, EGR1, TBX5, SMADs, and RUNXs in order to regulate target gene expression (57). Regulation of the Hippo signaling cascade is cell-type and context dependent and is facilitated by various signals including cell polarity (58), mechanical stress (59), G protein–coupled receptors (60), microRNAs (61), and cellular metabolism (62). Hippo signaling can also be facilitated by contact-dependent or contact-independent mechanisms (58). One of the contact-dependent mechanisms involves sequestration of YAP1 at cellular junctions such as adherens junctions and tight junctions (56, 63, 64). Dependent on the response to cellular stimuli, the role of YAP1 and the respective binding transcription factor may provoke a range of responses, including proliferation or apoptosis, and differentiation or epithelial-mesenchymal transition.

The regulation of the Hippo kinase cassette (MST1/2, SAV, LATS1/2, MOB1) and effector molecules of the Hippo pathway, YAP1 and TAZ, have been reviewed recently (65, 66). The canonical Hippo pathway has a number of inputs that act directly or indirectly to modulate the activity of the Hippo kinase cascade (Fig. 1). The mitogen-activated protein kinase kinase kinase kinase (MAP4K) acts to coordinate MST1/2 and LATS1/2 activation serving to phosphorylate and inhibit YAP1 and TAZ. Conversely, MST1/2 and MAP4K are negatively regulated by a large protein complex called STRIPAK (striatin-interacting phosphatase and kinase) containing the protein phosphatase (PP2A), which dephosphorylates these protein kinases and inhibits their activity. STRIPAK complex interactions with the MST1/2 or MAP4K occur in response to conditions that inhibit the Hippo pathway (67).

Mechanical cues, such as extracellular matrix (ECM) stiffness and shear stress, are also potent regulators of YAP1/TAZ (68, 69). Active YAP1/TAZ enter the nucleus when cells are grown on a stiff ECM or are spread across a large surface. In contrast, YAP1/TAZ are inactive when cells are in a soft ECM or are compressed into a small area. Attachment of cells to the ECM activates Rho-GTPases, limiting LATS-dependent YAP1 phosphorylation. Disruption of F-actin blocks the effect of ECM attachment on YAP1 nuclear localization. ECM stiffness is important for cell proliferation and differentiation, and YAP1/TAZ activity plays a role in these cellular processes.

G protein–coupled receptors (GPCRs) also regulate the Hippo pathway. Serum, lysophosphatidic acid (LPA), and sphingosine-1-phosphate (S1P) activate YAP1 and TAZ via their Gα12/13 and Gαq/11 GPCRs that activate Rho GTPase, leading to increases in F-actin. In contrast, Gαs–coupled receptors via activation of protein kinase A (PKA) inhibit Rho GTPase and reduce actin stiffness resulting in repression of YAP1 and TAZ (70). Activated PKA can also directly phosphorylate LATS, leading to YAP1 phosphorylation (71).

The Hippo pathway can be inhibited by growth factors like insulin-like growth factor 1 (IGF1) and the epidermal growth factor (EGF) family of ligands (EGF, AREG) that activate receptor tyrosine kinases leading to the activation of phosphoinositide 3-kinase (PI3K) and AKT signaling. Activation of PI3K dismantles the Hippo kinase complex which inactivates LATS and consequently increases levels of YAP1 (72). Additionally, once activated AKT can directly phosphorylate MST1/2 preventing its activation and allowing YAP1 and TAZ to be active (73).

LATS1/2 kinases phosphorylate YAP1 at 5 serine motifs [Ser61, Ser109, Ser127, Ser164, Ser397 (also annotated as Ser381)] and TAZ at 4 serine motifs (Ser66, Ser89, Ser117, Ser311) (56, 74-76) (Fig. 2). Other protein kinases can also stimulate YAP1 and TAZ phosphorylation either directly or indirectly. Independent of Hippo activation, the energy sensor AMP-activated protein kinase (AMPK) can directly phosphorylate YAP1 at Ser61, Ser94, and Thr119 (77, 78). Phosphorylation at the Ser94 site disrupts the interaction between YAP1 and TEAD, thus limiting YAP1-mediated transcription (77, 78). AMPK can also affect YAP1 activity indirectly by phosphorylation of Angiomotin-like 1 (AMOTL1), which can promote LATS1/2 activity and sequester YAP1 to the cytoplasm (79). On the other hand, LATS1/2 kinase can be activated in AMPK knockout cells in response to energy stress (78), indicating that cellular energy stress can also activate the Hippo pathway via an AMPK-independent mechanism. Many other protein kinases such as PKA, cyclin-dependent kinase 1 (CDK1), Jun N-terminal kinases (JNK), and Src family tyrosine kinases have been shown to phosphorylate YAP1 and TAZ (70, 80), suggesting that YAP1/TAZ can be regulated by mechanisms independent of Hippo pathway kinases.

Figure 2.

Structural motifs and phosphorylation sites in YAP1/TAZ. YAP1 is phosphorylated by LATS1/2 at 5 serine residues (Ser61, Ser109, Ser127, Ser164, and Ser397) and TAZ at 4 serine residues (Ser66, Ser89, Ser117, and Ser311) (green circles). Phosphorylation of YAP1 Ser127 results in the restriction of YAP1 to the cytoplasm due to binding with 14-3-3 and phosphorylation at Ser397 activates a phosphodegron degradation motif leading to reduced levels of YAP1 and TAZ. Likewise, phosphorylation of TAZ Ser89 sequesters TAZ to the cytoplasm via binding with 14-3-3 and Ser 311 activates a phosphodegron degradation motif. Phosphorylation of YAP1 by AMP-activated protein kinase (AMPK) occurs at multiple sites shown by red circles. Phosphorylation at Ser 94 via AMPK disrupts the YAP-TEAD interactions, resulting in reduction of the transcription and YAP1 target genes. Protein kinase A (PKA) activates LATS and results in LATS mediated phosphorylation of YAP1 and TAZ. Phosphorylation of YAP1 at Ser 112 by PKA enables YAP1 to translocate to the nucleus (blue circle). Abbreviations: cc, coiled-coil domain; ww, WW domain.

Investigations of Hippo signaling in the ovary demonstrate that the pathway is an integral regulator of ovarian physiology, fertility, and pathology, although the precise role and concomitant mechanisms remain obscure. Emerging studies have demonstrated a role for Hippo pathway components in follicle activation and growth (81-88), infertility caused by reproductive disorders such as polycystic ovary syndrome (PCOS) (89, 90), premature ovarian insufficiency (82, 91), and ovarian cell neoplasia (92-95). To date, most studies done in the ovary have focused on the role of YAP1. In this review, we summarize the studies that have investigated the Hippo pathway and core components in ovarian physiology across different stages of ovarian follicle development and oocyte maturation with a focus on the mammalian Hippo pathway (Table 1), and we identify gaps in our understanding of this fascinating area of reproductive biology. Although some key molecular mechanisms that regulate Hippo signaling have been identified, many aspects of how these are modulated under physiological or stress conditions remain poorly explored.

Table 1.

Hippo pathway core components

Mammalian gene	Binding domains
Mst1/2	Ste20 Ser/Thr kinase
Sav1	WW
Lats1/2	NDR Ser/Thr kinase
Mob1	Cys2-His2 zinc binding/Mob1/phocein
Yap/Taz (Wwtr1)	WW/PDZ and TEAD

Methods

An extensive examination of the literature was performed to ensure a comprehensive review on Hippo signaling and ovarian function. The search was completed using the PubMed database (https://www.ncbi.nlm.nih.gov/pubmed) through April 2022. The search terms included Hippo pathway/signaling and general keywords (ovary, follicle, reproductive tract, ovarian cancer) to specific keywords (granulosa cell, oocyte, YAP/YAP1, LATS1/2, WWTR1, TAZ, SAV, MOB, polycystic ovarian syndrome, steroidogenesis) based on the authors’ knowledge on the topic. Further, the reference lists of identified articles were manually reviewed.

Hippo Signaling in the Ovary

Cellular Localization of Hippo Pathway Components

The dynamic functions of the Hippo pathway have led to studies investigating the capacity of the Hippo signaling cascade to regulate mammalian germ cell proliferation, follicle development, and luteal formation. Gene expression studies and immunostaining approaches demonstrate that MST1/2, LATS1/2, YAP1, and phospho-YAP1 are present in all follicle stages (primordial, primary, secondary, antral) within the oocyte, granulosa, and theca cells, as well as present in some atretic follicles and the corpus luteum (83, 84, 86, 96-100). Expression of the Hippo component YAP1 is observed in the granulosa and luteal cells of the mammalian ovary, with YAP1 localization to the nucleus of granulosa cells in follicles from the primary stage through the preovulatory stages (85). Postovulation, YAP1 is predominately localized to the cytoplasm in differentiated luteal cells (84, 85, 101).

Microarray analysis identified the expression of all Hippo signaling components in the bovine ovary. No differences were observed between mRNA expression levels of MST1, MST2, LATS2, MOB1A, and YAP1 in granulosa cells and theca cells of large antral follicles or the large and small steroidogenic luteal cells, their respective follicular counterparts (84). However, when cell types were analyzed in a group, MST2 transcripts were 4-fold more abundant than MST1 in the granulosa cells, theca cells, small luteal cells, and large luteal cells (84). Further, LATS1 abundance was ~5-fold greater than LATS2 in the bovine ovary, although LATS1 transcripts were reduced by around 40% in large luteal cells compared with granulosa cells (84). Notably SAV1 mRNA expression in bovine follicles was upregulated during the follicular to luteal transition. Similarly, TAZ mRNA transcripts were increased ~7.5 fold in bovine large luteal cells when compared with bovine granulosa cells, possibly contributing to the increase in luteal cell size as granulosa cells differentiate into luteal cells (84). These findings suggest that MST2 and LATS1 are the predominant upstream Hippo signaling components in the bovine ovary. Whether this relationship holds true for other species remains to be determined.

Reproductive aging in females is characterized by a considerable diminution in the number and quality of follicles and healthy oocytes, in addition to changes in the composition and architecture of the ovarian stroma (102). In mice, whole ovarian expression of MST1 declines with age and YAP1 levels increase before dropping off between 5 months and 16 months of age (96). Interestingly, ovarian expression of LATS2 initially decreases with age in the ovary, but a notable increase in expression occurs after 5 months, coinciding with the time that YAP1 levels decrease and the number of follicles decreases (96). Another study investigating the expression of Hippo pathway components during ovarian aging in mice determined that mRNA and protein expression of MST1 and LATS2, and protein abundance of phospho-YAP1 decreases with physiological ovarian aging (103). In contrast, YAP1 mRNA and protein expression gradually increases before falling after 10 months of age (103). In a model of pathologically induced ovarian aging, investigators observed that YAP1 and MST1 mRNA and protein levels are decreased, although an increase in LATS2 protein and mRNA is observed (103). Future studies are needed to understand the endocrine and paracrine controls that contribute to the dynamic changes in Hippo signaling components during ovarian development and aging. Furthermore, studies are needed to determine how YAP1/TAZ contributes to ovarian gene expression and function over the reproductive lifespan and how the changing ovarian stromal environment (such as increases in fibrotic matrix) contributes to mechanosensing and Hippo activation.

Regulation of Follicular Development

Current understanding of the role of the proximal Hippo pathway component MST1/2 and its adaptor protein SAV1 in the ovary is limited. Most of the experimental work has been focused on SAV1, the core binding partner of MST1 and MST2. In vitro experiments demonstrated that SAV1 alone is unable to phosphorylate the downstream protein kinase LATS1; however, in combination with MST1/2, SAV1 promotes phosphorylation of LATS1 (104). Studies using the avian follicle as a model system revealed that during avian ovarian follicle development, SAV1 mRNA transcript levels are highest in small follicles (< 1 mm) and decline as follicular development progresses to form the large preovulatory follicles (> 9 mm) (104). Similar to mammalian follicle selection, follicle-stimulating hormone receptor (FSHR) transcripts and protein levels are maintained and required within the granulosa cells of ovarian prehierarchical follicles in the hen (105). Overexpression of SAV1, which promotes Hippo signaling in hen granulosa cells, decreased cell proliferation and downregulated mRNA transcripts for FSHR, steroidogenic acute regulatory protein (STAR), and growth differentiation factor 9 (GDF9) (104). Genetic knockdown of SAV1 using small interfering RNA (siRNA)-attenuated Hippo activation, causes an increase in YAP1 activity and results in increased FSHR, STAR, and GDF9 mRNA expression and increased granulosa cell proliferation (104). These observations are of relevance to the mammalian ovary, as FSHR expression is critical for antral follicle development and GDF9 has been demonstrated to promote FSH-induced progesterone production and upregulation of STAR (106), an essential protein required for steroidogenesis. Taken together, these observations indicate that the upstream Hippo component SAV1/MST1/2 appears to play a suppressive role in ovarian granulosa cell proliferation, which may serve to prevent follicle selection during follicle development. It seems likely that SAV1 activation of Hippo signaling limits YAP1 activity, thus, reducing cell proliferation in the ovary. Further studies evaluating the physiological inputs to the MST/SAV complex will provide needed insight into the regulation of Hippo signaling and its control of ovarian function.

The LATS1/2 protein kinases are downstream of MST1/2 and are responsible for phosphorylating the Hippo effectors YAP1 and TAZ. Studies across several mammalian species have provided important clues about the LATS1/2 protein kinases in the ovary. Studies by Pisarska et al (107), identified LATS1 as a forkhead box L2 (FOXL2) interacting protein in the granulosa cells of small/medium follicles in the mouse ovary. This interaction expands the mechanistic role of the Hippo pathway, as FOXL2 is involved in virtually all stages of ovarian development and function (108), including cellular differentiation and estrogen and progesterone synthesis (107, 109). Furthermore, in vitro studies show that LATS1 can directly phosphorylate FOXL2, citing a potential role for LATS1 as a mechanism for controlling the transcriptional repressor activity of FOXL2 during follicle development (107). Although these early studies did not examine YAP1 or TAZ, they highlight the possible involvement of LATS protein kinases in noncanonical Hippo signaling (110).

Genetic manipulations of LATS protein kinases reveal that precise regulation of the Hippo pathway is required to maintain cellular identity. Studies in the female mouse show that global loss of the Lats1 allele does not impact the number of germ cells in mutant ovaries vs wild-type mouse ovaries at birth, although greater germ cell apoptosis was observed as the mouse ages, resulting in premature loss of follicles (111). In this same study, Lats1^−/− newborn ovaries that were cultured in vitro exhibited abnormal follicle development, with a reduction in primordial and activated follicles and development of ovarian cysts (111). Similarly, the Lats1 null adult ovary displayed a reduction in the number of follicles, although they exhibit less granulosa cell apoptosis (112). Moreover, the adult ovaries lack corpora lutea, indicating a disruption in ovulation. The phenotype extends beyond the ovary with adult Lats1^−/− female mice lacking mammary glands and developing ovarian stromal tumors (112). Overexpression of LATS2 in avian granulosa cells repressed granulosa cell proliferation and downregulated mRNA transcripts for FSHR, STAR, cytochrome P450 family 11 subfamilies A member 1 (CYP11A1), estrogen receptor isoform 1 (ESR1/ERα) and isoform 2 (ESR2/ERβ) (99). Inversely, genetic knockdown of LATS2 via siRNA resulted in enhanced granulosa cell proliferation and increased abundance of transcripts for STAR, CYP11A1, ESR1, and ESR2 (99). Interestingly, the overexpression of LATS2 increased transcripts of MST1, MST2, TEAD1, and TEAD3 while downregulating transcripts of YAP1 and SAV1. Conversely, the knockdown of LATS2 in avian granulosa cells decreased MST1, MST2, TEAD1, and TEAD3 mRNA transcript abundance while increasing YAP1 and SAV1 mRNA transcripts (99). Further, the loss of YAP1 in avian granulosa cells overexpressing LATS2 nullified the inhibitory effects of LATS2 on granulosa cell proliferation and steroidogenesis related gene expression, suggestive of a YAP1-LATS2 feedback loop (99). Other investigators have reported that LATS2 and YAP1 form a negative feedback loop to control YAP1 activity and prevent human ovarian surface epithelial cells from malignant transformation (113).

To examine the role of LATS1/2 more specifically in the ovary, investigators have developed mouse models with targeted deletion of both Lats1 and Lats2. Targeted deletion of both Lats1 and Lats2 caused a loss of granulosa cell identity and function, as well as a change in granulosa cell morphology (114). Using a mouse model of granulosa cell-specific deletion of Lats1 and Lats2 (Cyp19a1-Cre; Lats1^fl/fl; Lats2^fl/fl) investigators observed premature ovarian failure, demonstrated by an enlarged ovary (~10-fold compared with wild-type control), sterile phenotype, and multi-lineage transdifferentiation of granulosa cells into seminiferous tubule-like structures and bone, suggestive of epithelial-to-mesenchymal transition (114). In vitro experiments with granulosa cells isolated from Lats1^fl/fl; Lats2^fl/fl mice and infected with Ad-Cre to induce recombination resulted in significant loss of Lats1/2 transcripts and subsequent reductions in LATS1 protein and YAP1 phosphorylation. As expected, this resulted in an increase in total YAP1 protein expression in granulosa cells. The elevation of YAP1 resulted in decreased expression of mRNA transcripts for Wnt4, Fshr, Lhcgr, and Cyp19a1, genes previously identified as necessary for granulosa cell function and differentiation. The LATS1/2 deficient granulosa cells also underwent similar multi-lineage transdifferentiation as observed in vivo (114).

Granulosa cell-specific deletion of Lats1/2 under control of the anti-Müllerian hormone receptor type 2 (Amhr2) promoter (Lats1^fl/fl; Lats2^fl/fl; Amhr2-Cre) resulted in a reduction of follicle numbers in adult female mice, although corpora lutea were observed, indicative of functional ovulation (115). Further, histological evaluation revealed altered granulosa cell morphology and differentiation reminiscent of epithelial-to-mesenchymal transition (115). Interestingly, the ovarian surface epithelium contained cells that were suggestive of a neoplasm, but immunohistochemical analysis utilizing ovarian cancer markers did not indicate the presence of ovarian tumor cells, nor did the cells increase in number or become invasive as the animals aged (115). Collectively, these findings point to a prominent role for the Hippo pathway LATS protein kinases in maintaining ovarian homeostasis during ovarian follicle development. The differences observed between the 2 phenotypes could be accredited to the timing and activation of the promotor gene in vivo. The Amhr2-Cre would be active much earlier in ovarian development (embryonic day 12.5) than the Cyp19a1-Cre, which is active in granulosa cells of large preantral and antral follicles and luteal cells. Additionally, because LATS protein kinases may also target transcriptional regulators other than YAP1 and TAZ, additional approaches are needed to determine if the observed ovarian phenotypes were due solely to Hippo signaling or represent a combination of events in response to canonical and noncanonical Hippo signaling.

Most studies have focused on exploring the role of YAP1, the downstream effector of the Hippo signaling pathway. Active Hippo signaling results in phosphorylation of YAP1 and TAZ, resulting in sequestration of the transcriptional co-regulators in the cytoplasm. Studies in the KGN human granulosa tumor cell line revealed that forced expression of wild-type YAP1 or constitutively active YAP1^S127A (this mutation prevents phosphorylation and maintains YAP1 activity) stimulates cell proliferation (85, 92). Similarly, ectopic expression of YAP1 drives cell cycle progression and proliferation of primary human granulosa cells and human granulosa cell lines in 3D spheroid culture systems (101). In contrast, genetic knockdown of YAP1 or pharmacological inhibition of YAP1/TEAD interactions in primary human granulosa cell cultures and KGN cells resulted in the suppression of cell proliferation and increased cell apoptosis (101). More evidence that suppression of the Hippo pathway is essential for granulosa cell proliferation was provided by studies employing siRNA targeting YAP1 and/or TAZ in primary cultures of bovine granulosa cells. Knockdown of either YAP1 or TAZ attenuated cell proliferation (84). Another study utilizing bovine granulosa cells determined that upon stimulation with bone morphogenic protein 2 (BMP2), YAP1 was dephosphorylated and translocated to the nucleus, resulting in increased mRNA abundance of the downstream YAP1/TEAD target and proliferation-related gene connective tissue growth factor (CTGF; also known as cellular communication network factor 2, CCN2) (116). Furthermore, disruption of the YAP1/TEAD complex with verteporfin treatment disrupted bovine granulosa cell proliferation (84, 116).

Genetic studies in mice also provide evidence for an important role for YAP1 in ovarian follicular development. Granulosa cell-specific deletion of Yap1 in mice (Foxl2-Cre crossbred with Yap1^fl/fl) disrupted ovarian follicle development, resulting in reduced ovarian size with increased follicular atresia, and decreased litter size and offspring number (85). However, this group observed no impact on follicle development in conditional Yap1 knockout driven by the Cyp19a1-Cre promoter (85). A reason for the differences observed in these models could be explained by the very early depletion of Yap1 when follicles are forming and developing in the Foxl2-Cre mice compared to the Cyp19a1-Cre mice in which Yap1 is deleted in granulosa cells only after the formation of preantral follicles. The lack of a phenotype in the Cyp19a1-Cre mice could potentially be attributed to either a compensatory role for Taz, in which there would be little to no impact of the loss of Yap1 on follicle development; or changes in the expression of other transcriptional regulators during follicular development (36).

Given the Hippo pathway’s role in controlling tissue size and development, it is not a surprise to find that disruption of the Hippo pathway has been implicated in follicular activation, particularly in nonphysiological conditions. In mouse ovaries cultured in vitro, lentiviral knockdown of Yap1 suppressed follicle growth, resulting in more primordial follicles and fewer primary follicles, while overexpression of Yap1 results in follicle activation (83). Moreover, pharmacological inhibition of Yap1 via verteporfin disrupted follicle formation in the adult mouse ovary in vivo and the immature mouse ovary in vitro (85). Similarly, in vitro stimulation of the PI3K pathway, and subsequent treatment with a chemotherapeutic drug in cultured ovaries resulted in an increase of nonphosphorylated (active) YAP1 and follicle activation (87). On the contrary, the lentiviral overexpression of Yap1 in a chemotherapeutic-induced infertile mouse model increased the thickness of the ovarian surface epithelium, increased recruitment of follicles, and resulted in an increase in pup birth rate relative to respective controls. In this same study, the knockdown of Yap1 in a wild-type model resulted in no changes in the ovarian surface epithelium but resulted in decreased numbers of primordial follicles and lower birthrates relative to controls (117). These reports only begin to reveal the roles of each Hippo pathway component and how they are regulated during early follicle development and differentiation (Table 2).

Table 2.

Knockout/overexpression studies of Hippo pathway core components and ovarian phenotypes

Study (first author, year)	Model system	Gene targeted	Overexpression or knockout	Findings
Fu et al., 2014	KGN cell line	YAP1	OE	Increase in granulosa cell proliferation and migration
			KO	Inhibition of cell proliferation; changes in cell morphology
Ye et al., 2017	Cyclophosphamide (CPA)-induced infertile mouse model	Yap1	OE	Increased follicle number, litter size/birth rate, and ovarian surface epithelium thickness
			KO	Decreased follicle number and pup birthrate
He et al., 2019	Primary human ovarian surface epithelium cells	YAP1	OE	Increase in cell proliferation; eventual cell senescence
			KO	No changes in cell proliferation
		TAZ	KO	No changes in cell proliferation
		YAP1/TAZ	KO	Decrease in cell number
		LATS2	KO	Increase in cell proliferation; loss of cell senescence
Lv et al., 2019	Foxl2-cre; Yap1fl/fl mouse model	Yap1	KO	Decrease in ovarian follicle number/formation; decrease in ovarian size; increased number of atretic follicles; decreased litter size
	Mouse model—in vitro ovary culture	Yap1	Inhibition via VP	Disruption of follicle formation
	Mouse model—in vivo	Yap1	Inhibition via VP	Reduced corpora lutea; increased granulosa cell apoptosis
Plewes et al., 2019	Primary bovine granulosa cells	YAP1	KO	Reduced granulosa cell proliferation
		TAZ	KO	Decreased granulosa cell proliferation
Hu et al., 2019	Mouse model—in vitro ovary culture	Yap1	OE	Decrease in primordial follicle number; increase in primary follicle number
			KO	Increase in primordial follicle number; decrease in primary follicle number
Hu et al., 2019	Tri-ortho-cresyl phosphate (TOCP) exposed mouse model	Yap1	OE	Increase in follicle numbers; decreased atretic follicles; increased fertility
Lv et al., 2020	Primary human granulosa cells from IVF patients	YAP1	OE	Increased granulosa cell proliferation and growth; loss of granulosa cell identity
			KO	Decreased granulosa cell proliferation and growth; increased cell apoptosis
	KGN cell line	YAP1	OE	Promotion of cell cycle progression; decrease in apoptotic cells; loss of granulosa cell identity
			KO	Reduced proliferation; increased apoptosis
	HGrC1 cell line	YAP1	OE	Increased granulosa cell proliferation and growth; decreased apoptosis; loss of granulosa cell identity
			KO	Suppressed granulosa cell proliferation
St John et al., 1999	Lats1 -/- mouse model	Lats1	KO	Reduced follicle number; disruption in ovulation
Sun et al., 2015	Lats1 -/- mouse model	Lats1	KO	Abnormal follicle development; development of ovarian cysts; increased germ cell apoptosis
Tsoi et al., 2019	Lats1 fl/fl ; Lats2fl/fl; Cyp19a1-cre mouse model	Lats1/2	KO	Enlargement of the ovary; infertility; loss of granulosa cell identity
	Lats1 fl/fl ; Lats2fl/fl; Cyp19a1-cre mouse isolated granulosa cells	Lats1/2	KO	Loss of granulosa cell identity and function
St Jean et al., 2019	Lats1 fl/fl ; Lats2fl/fl; Amhr2-cre mouse model	Lats1/2	KO	Reduced follicle number; alterations in granulosa cell morphology and identity; thickening of the ovarian surface epithelium
Sun et al., 2021	Primary hen granulosa cells	LATS2	OE	Decrease in granulosa cell proliferation
			KO	Increase in granulosa cell proliferation
Lyu et al., 2016	Primary hen granulosa cells	SAV1	OE	Decrease in granulosa cell proliferation
			KO	Increase in granulosa cell proliferation

Abbreviations: KO, knockout; OE, overexpression; VP, verteporfin.

It is clear that activation of the Hippo pathway results in the cessation of granulosa cell growth, while inhibition of the Hippo pathway results in the activation of YAP1 and proliferation of granulosa cells with subsequent follicle growth. Further, aberrant expression of YAP1 can contribute to lineage reprogramming in ovarian cells, potentially disrupting follicular development and overall ovarian homeostasis. Taken together, the evidence suggests that the Hippo pathway in ovarian somatic cells plays a prominent role in follicle development and function.

Hippo Components in Ovarian Germ Cells and Oocytes

Successful reproduction depends on proper oocyte development and maturation. Definitive temporal gene and protein expression and their respective cellular localization in germ cells is required for oocyte development. Single cell transcriptome analysis of oocytes collected from human in vitro fertilization (IVF) patients has identified YAP1 in mature oocytes (118). Immunohistochemical evaluation of the upstream Hippo signaling component MST1 in human ovarian cortical strips demonstrated a dynamic shift in MST1 expression in the oocyte during follicular development, with localization occurring in the oocyte cytoplasm in primordial and primary follicles and a gradual translocation to the oocyte nucleus in secondary and antral follicles (119-121). In mice, localization of YAP1 protein and mRNA varies within oocytes based on developmental stage. YAP1 is evenly distributed in the cytoplasm of murine oocytes at the germinal vesicle stage, with expression shifting to the nucleus after fertilization (121). Similar observations demonstrate that YAP1 expression is restricted to the cytoplasm in murine embryonic germ cells and throughout oocyte development, with phosphorylation at Ser112 regulated by protein kinase A (PKA) (119). Upon dephosphorylation of YAP1 at Ser112, YAP1 translocates to the nucleus of the growing oocyte, although it is unable to accumulate, suggesting that YAP1 is not intrinsically involved in murine oogenesis (119). Further support for the thesis that Yap1 is nonessential for oogenesis in mice is derived from experiments in which Yap1 was selectively deleted in oocytes by crossing Yap1-floxed mice with transgenic Gdf9-Cre mice (121). Oocyte-specific deletion of Yap1 had no impact on folliculogenesis, oocyte maturation, spindle organization, or fertilization (121). Importantly, upon fertilization, oocyte Yap1 is vital for embryonic development (122). Interestingly, expression of the coactivator Taz is minimal in the mouse oocyte (121). It appears therefore, that Hippo signaling in ovarian somatic cells rather than germ cells directs follicular growth and oocyte maturation.

Studies using murine cumulus oocyte complexes (COCs) reveal that the oocyte and other factors that control COC expansion have an important impact on the Hippo pathway (123). Removal of oocytes from COCs of immature mice resulted in increases in Sav1, Mob1b, and Lats1/2 transcripts in cumulus cells and the subsequent re-introduction of oocytes to the cumulus cells restored expression to the levels observed in intact COCs (123). To determine whether growth factors secreted from the oocyte, like GDF9 and BMP15, contributed to the regulation of Hippo signaling within the cumulus cell, COCs were treated with SB431542 to inhibit oocyte-initiated GDF9/BMP15-mediated SMAD2/3 signaling. Similar to the removal of oocytes, treatment with the inhibitor increased Sav1 and Lats2 mRNA, although no comparable increases in Lats1 or Mob1b were observed (123). Treatment of COCs with verteporfin to disrupt the YAP1/TEAD interaction hindered oocyte stimulation of granulosa cell proliferation in vitro and increased expression of transcripts that support progesterone secretion in the absence of ovulatory signaling (123). Together, this suggests that inhibition of YAP1/TEAD association can lead to premature differentiation of cumulus cells (123). These studies provide valuable clues about the role of the oocyte in controlling Hippo signaling and the role of Hippo signaling in the process of cellular differentiation.

In vitro stimulation of murine COCs with EGF which mimics the post-LH surge ovulatory increase of EGF-like growth factors, increased the abundance of many Hippo-related transcripts (Mob1b, Mst1/2, Lats1/2, and Taz), but not Yap1. In this study, levels of proteins for these components as determined by Western blot analysis displayed only modest changes or were greatly reduced in the intact COCs, especially LATS1 and YAP1 (123). In contrast, in vitro treatment of bovine granulosa cells with the YAP1/TEAD inhibitor verteporfin inhibited EGF stimulation of ERK1/2 and AKT phosphorylation (100), consistent with the findings in human granulosa cells, where YAP1 interacts with the EGFR singling pathway to promote cell proliferation (85). In vivo modeling of ovulation in the mouse with human chorionic gonadotropin treatment revealed increases in phosphorylation of YAP1 and TAZ after 24 hours and a significant reduction in total YAP1, but not TAZ protein (123). Understanding the role of the Hippo pathway during the ovulatory transition and response to gonadotropin control is pivotal for the development of new therapeutic approaches for improving fertility or contraception.

It is well accepted that human females, as well as females from other mammalian species, are born with a finite number of oocytes. However, a theory posits that mammalian ovarian stem cells exist as a potential source of germ cell renewal (124-127). Given the role of the Hippo pathway in stem cell renewal, some studies suggest an association between Hippo core components and ovarian stem cells (103, 117). Co-expression of the Hippo components Lats2, Mst1, and Yap1 is observed in vitro in isolated murine ovarian stem cells expressing the germ cell marker mouse vasa homologue (Mvh; also known as DEAD box protein 4, DDX4). The overexpression of Yap1 in these cells resulted in an increased abundance of specific markers for ovarian stem cells, Mvh and Oct4 (also known as POU Class 5 Homeobox1, POU5F1), suggesting that Yap1 promotes the differentiation of ovarian stem cells (117). Conversely, the knockdown of Yap1 via shRNA resulted in reduced abundance of Mvh and Oct4 and a decrease of multiplication efficiency in cells depleted of Yap1 (117), indicating that Yap1 expression is positively correlated with the proliferation and differentiation of murine ovarian stem cells.

Impacts of Hippo Signaling on Endocrine Function and Cellular Metabolism

Impacts on Endocrine Function

Hormone responsive estradiol biosynthesis is a hallmark of granulosa cells contributing to the timing and control of the reproductive cycle. In vitro studies utilizing siRNA-mediated knockdown of YAP1 in bovine granulosa cells revealed an 80% reduction in FSH-induced estradiol production relative to granulosa cells treated with nontargeting siRNA controls (84). Suppression of YAP1 activity via pharmacological inhibition or gene silencing also suppresses estradiol and progesterone production in KGN cells and primary human granulosa cells (85, 92). Paradoxically, other investigations using the human KGN cell line and primary human granulosa cells reported that overactivation of YAP1 also repressed production of estradiol and progesterone and eliminated accumulation of lipid droplets, presumptively as a result of the dedifferentiation of gonadotropin-responsive granulosa cells (85, 101). These divergent findings suggest that physiologic levels of YAP1 are required to balance normal proliferative and steroidogenic activities of granulosa cells. Forced overexpression of YAP1 results in cellular reprogramming and loss of granulosa identity. These observations are consistent with findings in murine granulosa cells showing that targeted deletion of Lats1/2 resulted in dedifferentiation or transdifferentiation of granulosa cells (114, 115). There is a great opportunity for future studies to identify the nuclear binding partners of YAP1 and TAZ and to determine how Yap1/Taz control transcriptional reprogramming during phases of follicular development and differentiation.

Studies performed on murine granulosa cells showed that treatment with testosterone and estradiol enhanced the expression and activity of Yap1 (97). As demonstrated by studies employing a luciferase reporter reflecting YAP1 activity, testosterone induces translocation of Yap1 from the cytoplasm to the nucleus within 30 minutes of treatment before being redistributed to the cytoplasm after 24 hours. Testosterone’s ability to stimulate transcription of the Yap1-activity reporter was prevented in cells depleted of YAP1 by treatment with siRNA targeting Yap1. This response was accompanied by a reduction in testosterone-stimulated proliferation of granulosa cells (97).

Studies utilizing genetic mouse models have also demonstrated that alteration in Hippo signaling adversely impacts gonadotropin levels and ovarian hormones. Global deletion of Lats1 in female mice results in decreased serum levels of LH and prolactin, although FSH levels remain unaffected (112). A recent study utilizing a mouse model of gonadotrope-specific deletion of Yap1 and Taz (GnRH receptor-IRES-Cre) revealed that Yap1 and Taz can regulate LH release, as demonstrated by increased levels of circulating LH after loss of Yap1 and Taz function (128). These mice also exhibited a hyperfertility phenotype characterized by higher ovulation rates and larger litter sizes (128). Further, in cultured pituitary cells from the Yap1 and Taz knockout increased both basal and GnRH-induced LH secretion, while inhibition of Yap1 via verteporfin also increased LH secretion in a similar manner in LβT2 gonadotrope-like cells (128).

Other studies show that ovarian specific deletion of Lats1 and Lats2 (Cyp19a1-Cre) results in increased ovarian weight, loss of granulosa cell identity, decreased progesterone levels (~6 fold lower than respective controls), and increased levels of FSH and LH, indicative of potential ovarian failure (114). Ovarian targeted Yap1 shRNA in mice decreased serum estradiol and FSH levels while lentiviral overexpression of Yap1 generates the opposite effect, increasing estradiol and FSH serum levels (117). Together, these findings implicate Hippo signaling and YAP1 in moderating the endocrine function of the ovary. The mechanism(s) behind how the Hippo pathway regulates steroidogenesis are subject to future investigation.

The Hippo and ERK signaling pathways may intersect in the control of ovarian function. The RAS-ERK1/2 signaling cascade is activated in the ovary upon the LH surge prior to ovulation (35). In various cell types, mitogen-activated protein kinases (MAPKs) may work in parallel with MST1/2 in the regulation of LATS1/2 (129). Using a superovulation model, in vivo human chorionic gonadotropin (hCG) injection into prepubertal mice increased the abundance of phosphorylated MST1/2 and phosphorylated YAP1 in granulosa cells of preovulatory follicles and in luteal cells after ovulation, thus decreasing nuclear YAP1. It is likely that ERK signaling mediated the response because in vivo deletion of Erk1/2 in granulosa cells (Erk1^-/-; Erk2^fl/fl; Cyp19a1-Cre) preserved YAP1 protein expression after hCG injection (97). Similar results were observed in vitro after forskolin/phorbol 12-myristate 13-acetate (PMA) treatment (mimicking the LH surge) and use of the ERK1/2 inhibitor U0126 (97). Treatment of murine granulosa cells with forskolin and PMA resulted in translocation of YAP1 from the nucleus to cytoplasm and the upregulation of genes associated with ovulation. Overexpression of Yap1 hindered their expression while the subsequent knockdown of Yap1 induced upregulation of the LH target genes (97). In a similar fashion, blocking YAP1 activity by direct injection of the YAP1/TEAD inhibitor verteporfin into the bovine follicle in vivo, obstructed ovulation in a dose-dependent manner (100). In bovine granulosa cells treated with verteporfin in vitro, downregulation of YAP1 target genes CTGF and CYR61 was observed, along with decreases in EGFR mRNA abundance (100). These important observations suggest that Hippo signaling plays a role in maintaining the state of cellular differentiation that allows for proper timing and execution of the ovulatory response to gonadotropins (Table 3). The ways in which these pathways interact to control the ovulatory process in large animal models, nonhuman primates, and humans await discovery.

Table 3.

Knockout/overexpression studies of Hippo pathway core components and impacts on endocrine function

Study (first author, year)	Model system	Gene targeted	Overexpression/ knockout	Hormone measured	Findings
Fu et al., 2014	KGN cell line	YAP1	KO	E2 (conditioned media)	Reduces FSH-induced E2 production
Ye et al., 2017	Lentiviral OE infertile mouse model	Yap1	OE	E2, FSH (serum)	Increased FSH; no changes in E2
	Lentiviral KO wild-type mouse model	Yap1	KO	E2, FSH (serum)	Decline in E2, FSH
Lv et al., 2019	KGN cell line 3D culture system	YAP1	OE	E2, P4 (conditioned media)	Reduced E2, P4
Plewes et al., 2019	Primary bovine granulosa cells	YAP1	KO	E2 (conditioned media)	Reduces FSH-induced E2 production
Lv et al., 2020	KGN cell line	YAP1	OE	E2, P4 (conditioned media)	Reduces FSH/FSK-induced E2, P4 production
	Human granulosa cells from IVF patients	YAP1	OE	E2, P4 (conditioned media)	Reduces FSH/FSK-induced E2, P4 production
Lalonde-Larue et al., 2022	Yap1 fl/fl ; Taz fl/fl ; Gnrhr GRIC mouse model	Yap1/Taz	KO	FSH, LH (serum)	Increases in LH; no changes in FSH
St. John et al., 1999	Lats1 -/- mouse model	Lats1	KO	LH, PRL, FSH (serum)	Deficiency in LH, PRL; no changes in FSH
Tsoi et al., 2019	Lats1 fl/fl ; Lats2fl/fl; Cyp19a1-Cre mouse model	Lats1/2	KO	E2, P4, LH, FSH (serum)	Decrease in P4; Increases in LH, FSH; E2 undetected (below assay threshold)

Abbreviations: E2, estradiol; FSH, follicle-stimulating hormone; FSK, forskolin; KO, knockout; LH, luteinizing hormone; OE, overexpression; P4, progesterone; PRL, prolactin.

Impacts on Cellular Metabolism

Similar to events that activate Hippo signaling, environmental stimuli (e.g., energy stress, nutrients, oxygen) and upstream protein kinases (growth factor tyrosine kinases, Ca²⁺/calmodulin protein kinases, PKA) regulate cellular metabolism, which leads to a complex perspective on their regulation of YAP1 activity (130, 131). One such protein kinase is 5′-AMP-activated protein kinase (AMPK). AMPK activates catabolic processes such as autophagy, glucose transport, glycolysis, and glycogen synthesis or gluconeogenesis and is expressed in oocytes, ovarian follicle cells, and luteal cells (132-137). AMPK is activated in response to energy stress, i.e., reduced levels of ATP (elevated AMP/ATP ratio) (138). AMPK activation can also be triggered by upstream kinases, such as liver kinase B1 (LKB1) or in response to calcium flux via calcium/calmodulin dependent kinase kinase 2 (CAMKK2) (138, 139). Importantly, the activation of AMPK can be offset by gonadotropin activation of PKA signaling, which limits AMPK activity (133, 134). Studies suggest that AMPK serves to limit YAP1 activity. AMPK can directly phosphorylate YAP1, disrupting the interaction between YAP1 and the transcription factor TEAD, thus limiting YAP1-mediated transcription (77, 78). AMPK can also promote LATS1/2 activity and phosphorylation of YAP1, which reduces transcription activity by sequestering YAP1 in the cytoplasm (74). On the other hand, LATS1/2 kinase can be activated in response to energy stress in cells deficient in AMPK cells (78), indicating that cellular energy stress can also activate the Hippo pathway via an AMPK-independent mechanism.

Activation of AMPK inhibits gonadotropin-stimulated progesterone synthesis and induces autophagy in ovarian granulosa and luteal cells (132-134, 137, 140). Studies performed in the mouse ovary found that energy stress via glucose depletion increased AMPK activation and phosphorylation of MST1/2 (140). However, it did not affect phosphorylation of LATS1 and YAP1 (Ser127 or Ser94) when compared to ovaries treated with high levels of glucose (141). In this study, treatment with 5-aminoimidazole-4-carboxamide-1-β-D-ribofuranoside (AICAR), a known activator of AMPK, also elevated phosphorylation of LATS1, but paradoxically decreased phosphorylation of YAP1 at the site required for inactivation (Ser127) (141). Treatment with AICAR can also decrease progesterone production in granulosa and luteal cells (132-134, 137, 140). Thus, it is possible that dysregulation of AMPK activity can affect posttranslational modifications of YAP1 and ultimately change ovarian steroidogenesis. As FSH is a main driver of follicle development, Puri et al suggested that observed effects of FSH on YAP1 can be associated with cell differentiation rather than cell growth (142). A caveat with the studies mentioned above is that direct AMPK-YAP1 interactions were not established.

Another protein kinase responsible for regulating metabolic events in ovarian cells is mechanistic target of rapamycin (mTOR), which in contrast to AMPK, promotes anabolic processes, including ribosome biogenesis and protein, nucleotide, fatty acid, and lipid synthesis. In general, the activation of mTOR is dependent on nutrients and growth factors via (PI3K)/AKT and/or mitogen-activated protein kinase (MAPK)1/3 signaling pathways (143, 144). As described earlier for the Hippo signaling pathway, the mTOR signaling pathway is also important for follicular activation, growth and differentiation, and ovulation (145). The roles of Hippo and mTOR in organ size regulation are well established by their respective functions in controlling cell number and cell size (146). Several reports indicate that components of the Hippo pathway play an important role in the regulation of mTOR activity. For example, deletion of the upstream regulator of neurofibromin 2 (NF2), which promotes Hippo signaling, results in the activation of mTORC1 and stimulates growth of various cell types (147-149). Additionally, Yap1 regulates IGF1 (150) and the EGF ligand AREG, which activate mTOR and contribute to granulosa cell proliferation (85). Similar to depletion of Nf2, loss of Mst1 or Mst2 leads to mTORC1 activation (151). More recent studies indicate that mTORC2 can directly phosphorylate YAP1 at serine 436, positively regulating YAP1 activity, YAP1 dependent gene expression, cell growth, and motility in glioblastoma cells (149). Other lines of evidence suggest a reciprocal regulation of mTOR by YAP1 signaling. Nutrients such as amino acids (e.g., glutamine, arginine, leucine) are sufficient to activate mTOR (152). Experiments in systems other than the ovary indicate that activated, nuclear-localized YAP1 promotes the transcription of key metabolic enzymes. For instance, YAP1 stimulates the expression of glutamine synthetase, catalyzing the transformation of glutamate into glutamine, which then activates mTOR (153, 154). YAP1 can also upregulate expression of amino acid transporters that enhance mTOR activity (155). Thus, ample evidence suggests that both mTOR and Hippo signaling regulate ovarian growth, but the exact mechanisms operating during phases of follicular development remain to be solved experimentally.

The gonadotropins FSH and LH activate mTOR via the cAMP/PKA signaling pathway in ovarian granulosa and luteal cells, respectively (156, 157). Inhibition of mTOR activity impairs follicular function (158), but not luteal function or cellular viability (132), suggesting that mTOR activation is required for follicular maturation. Activation of AMPK, which activates Hippo, also inhibits mTOR and impairs the ability of FSH and LH to stimulate granulosa and luteal steroidogenesis (156, 159). How mTOR and AMPK pathways converge to regulate Hippo signaling in ovarian cells is complex and has not been carefully examined. Both FSH and LH can inhibit AMPK activity in their respective target cells (133, 134, 160) suggesting that the levels of YAP1 and TAZ phosphorylation likely fine tune the overall growth- and differentiation-promoting actions of gonadotropins. Studies are needed to fully characterize the YAP1 and TAZ regulated transcriptomes in response to hormones and under conditions of metabolic stress to identify genes and pathways critical for altering ovarian function and cellular fate.

Dysregulated energy metabolism represents a major contributing cause of female infertility (161). Lipids are an important source of mitochondrial energy production and cholesterol is a key precursor for steroid synthesis. Cellular lipid homeostasis is therefore essential for female fertility and regulated by a complex interplay between de novo synthesis, export, and storage (162). Multiple studies have demonstrated that excess cholesterol and fatty acids can negatively affect development competence of oocytes (163-173). It has been recently demonstrated that multiple members of the Hippo family influence lipid metabolism (174). Sterol regulatory-element binding proteins (SREBPs), key transcription factors that regulate the expression of genes involved in cholesterol synthesis, were found to be regulated by the Hippo components MST1 and LATS2 (175, 176). Overexpression of MST1, which would activate LATS2, led to the downregulation of SREBPs (176). Inhibition of Hippo signaling by in vitro knockdown of LATS2 caused activation of SREBP and cholesterol accumulation. Mice harboring a liver specific knockout of LATS2 displayed constitutive activation of SREBP, leading to aberrant lipid accumulation even when maintained under normal diet (175). Furthermore, the Hippo effector molecule, YAP1 has been shown to directly interact with SREB-1c and SREBP2 on the promoter’s fatty acid synthase (FASN) and 30-hydroxylmethyl glutaryl coenzyme A reductase (HMGCR) leading to stimulation of cholesterol synthesis (177). During the ovulatory process, LH stimulates expression of SREBP, resulting in de novo cholesterol synthesis (178). Lipid metabolism may also influence Hippo signaling by influencing the functional activity of YAP1 and TAZ. One such pathway is the mevalonate synthesis pathway, which converts acetyl-CoA into lipid precursors for generation of cholesterol and steroid hormones (179). Inhibition of the mevalonate pathway by statins, which inhibit HMGCR, leads to the downregulation of YAP1 activity, which was bought about by obstructing the activity of Rho GTPase through its effect on the F-actin cytoskeleton (180). In other studies, increased cholesterol levels in pathological conditions were shown to promote TAZ activity through a calcium-mediated signaling pathway (181). Generally, the YAP1/TAZ elevations in lipid synthesis were associated with increased cellular proliferation. Because cellular lipid homeostasis is essential for female fertility, an understanding of how Hippo signaling controls ovarian fatty acid and cholesterol metabolism during the ovarian reproductive cycle is critical to better understand follicular development and oocyte quality as well as the underpinnings of timely cholesterol production for cellular growth, differentiation, and steroidogenesis.

Reproductive Toxicity/Endocrine Disruptors

Endocrine-disrupting chemicals are exogenous substances and/or mixtures from natural or synthetic sources that disrupt the normal actions of endogenous hormones. Endocrine-disrupting chemicals can affect fertility via the alteration of ovarian steroidogenesis (182), mimicry of receptor signaling (183, 184), and impaired follicle and oocyte growth (185, 186). Zearalenone is an estrogenic mycotoxin present in contaminated grain products, including those used in animal feeds and breakfast cereals (187). The estrogenic effects of zearalenone on the female reproductive system are responsible for delayed pregnancy and fetal development (188, 189), disrupted germ cell meiotic progression and follicle assembly (190-193), and alterations in the ovarian proteome (194). Studies have shown that low doses of zearalenone can exert estrogen-like effects and carcinogenic properties, which can stimulate the proliferation of cells, while high doses can result in oxidative stress, DNA damage, and apoptosis (195).

In cultured mouse granulosa cells, exposure to zearalenone increased mRNA and protein expression of the Hippo pathway components Yap1, Taz, and Tead and resulted in an abnormal population of large, flattened granulosa cells (196). Bioinformatics analysis of RNA sequencing studies following similar treatment protocols revealed that zearalenone upregulated genes in granulosa cells associated with pathways linked to uncontrolled cell growth. Oral gavage of young female mice with zearalenone for 3 weeks also increased the expression of Yap1, Taz, Tead, and Smad3 in granulosa cells and disrupted ovarian follicle development (196). The impacts of chronic exposure to zearalenone on the ovary are unclear, although alterations in the cell cycle and Hippo signaling may potentially contribute to uncontrolled granulosa/ovarian cell growth.

Perfluoroalkyl and polyfluoroalkyl substances (PFAS) are widespread environmental contaminants found in several industrial and consumer products. These chemicals have been detected in nearly every person sampled in the US National Biomonitoring Program (197). Perfluorooctanoic acid (PFOA) is one of the primary PFAS compounds and is extensively manufactured. Observational studies in humans have shown PFOA to delay the age of menarche, disrupt menstrual cyclicity, and initiate early menopause via premature ovarian insufficiency (198-200). Similar studies in animals have revealed advanced puberty onset, inhibition of sex steroid hormones, and alterations in the composition of the follicular pool (201, 202). A recent study utilizing the immortalized human granulosa cell line HGrC1, a noncancerous human granulosa cell line that mimics granulosa cell behavior in earlier stages of development, revealed increased levels of YAP1 protein and its downstream target CTGF after PFOA exposure (203). The observations of increased YAP1 and CTGF protein expression was paired with increases in cell proliferation and migration (203). Upon inhibition of YAP1 with verteporfin, HGrC1 cell proliferation was repressed, further supporting a role for the Hippo pathway in PFOA-induced cell proliferation.

Another potential source of endocrine-disrupting chemicals includes exposure to plasticizers, chemical additives used in the manufacturing of products widely used in industry, agriculture, medical devices, and in food and beverage containers (204). Tri-ortho-cresyl phosphate (TOCP), a plasticizer commonly used in industry as a plastic softener and flame retardant, causes dose-dependent effects on murine ovarian follicle development, including decreased ovarian weight and follicle numbers relative to ovaries from mice that were exposed to vehicle alone (205). In mice exposed to TOCP, ovarian protein levels of the Hippo pathway components MST1 and LATS2 were increased and YAP1 expression was decreased (205), consistent with Hippo pathway activation. Posttranslational modifications of the Hippo pathway were observed in response to exposure to higher doses of TOCP. The phospho-MST1/MST1 ratio was decreased while in contrast the phospho-YAP1/YAP1 ratio was increased, suggestive of reduced survival of follicles due to activation of the Hippo pathway and subsequent loss of active YAP1-mediated growth-associated gene expression (205). Interestingly, ovarian overexpression of Yap1 in mice treated with TOCP provided a protective effect, evidenced by decreased numbers of atretic follicles and increased primordial, preantral, and antral follicle numbers, as well as increased fertility as measured by generation of live offspring (205). Chemical insults trigger stress responses and activation of Hippo signaling is essential for preserving tissue homeostasis and cell survival upon a chemical insult (Table 4). Taken together, these data further implicate the Hippo pathway as a key mediator of ovarian homeostasis in response to environmental toxicants. Much remains to be discovered about how YAP1 and TAZ interact with ovarian transcription factors to fine tune cell differentiation and cell fate after chemical exposure.

Table 4.

Chemical exposure impacts on Hippo pathway core components

Study (first author, year)	Model system	Chemical exposure	Findings
Zhang et al., 2018	Primary mouse granulosa cells	Zearalenone (30μM)	Altered granulosa cell morphology; increased transcripts of Yap1, Taz, Tead3; increased protein expression YAP1, TAZ
	Mouse model—in vivo	Zearalenone (40-100 μg/kg)	Increased transcripts of Yap1, Taz; decrease in follicle number
Hu et al., 2019	Mouse model—in vivo	Tri-ortho-cresyl phosphate (TOCP)	Increased protein expression MST1, LATS2, decrease YAP1 protein; decline in follicle number; increase in atretic follicles; protective effect of YAP1 OE
Clark et al., 2022	HGrC1 cell line	Perfluorooctanoic acid (PFOA)	Increased cell proliferation and migration; elevated YAP1 and CTGF protein expression

Abbreviations: CTGF, connective tissue growth factor; LATS2, large tumor suppressor 2; MST1, mammalian serine/threonine-protein kinase 4; OE, overexpression; TAZ, WW domain-containing transcription regulator 1 (also known as WWTR1); TEAD, TEA domain transcription factor; YAP1, yes-associated protein 1.

Hippo Signaling in Ovarian Disorders

Polycystic Ovary Syndrome

Polycystic ovary syndrome (PCOS) is a complex endocrine disorder affecting 5% to 15% of women of reproductive age (206). Clinically, women with PCOS have at least 2 of the 3 following characteristics: clinical and/or biochemical hyperandrogenism, ovulatory dysfunction, and polycystic ovarian morphology via ultrasound. PCOS is frequently associated with abdominal adiposity, insulin resistance, obesity, metabolic disorders, and cardiovascular risk factors. The etiology of this syndrome remains largely unknown, but mounting evidence suggests that PCOS might be a complex multigenic disorder with strong epigenetic and environmental influences. PCOS is associated with anovulatory or oligo-ovulatory infertility and increased risk of obstetric complications such as miscarriage, gestational diabetes, and preeclampsia (207); therefore, it is especially important to identify the molecular background of this disorder.

Several independent genomic studies have identified several loci significantly associated with PCOS (90, 208-210). A genome-wide association study (GWAS) identified 3 variants of YAP1 (rs11225138, rs11225161, rs11225166) that correlate with enhanced susceptibility for PCOS (90). Interestingly, genotype-phenotype correlation analysis showed that carriers of the rs11225161 and/or rs11225166 allele have impaired glucose tolerance, suggesting that women carrying these specific alleles of YAP1 may also have an increased risk of developing a metabolic disorder (90). Among the highly expressed genes in the ovaries of women with PCOS was the Hippo pathway component YAP1 (90, 210). Studies show that the YAP1 promoter is hypomethylated in PCOS patients compared to non-PCOS controls, which could contribute to the increased level of YAP1 mRNA and protein observed in PCOS patients (211). As described in previous sections of this review, YAP1 supports appropriate development and growth of ovarian follicles, while dysregulation in the Hippo pathway can lead to elevated expression of YAP1 and result in enlargement of the ovaries (114) and ovarian cysts (111), which are characteristics of PCOS (212).

Hormonal abnormalities in women with PCOS are associated with a higher concentrations of testosterone in blood serum, suggesting that an elevated androgenic environment contributes to the disorder (213). Hyperinsulinemia, a characteristic feature for women with PCOS, contributes to hyperandrogenism by stimulatory effects on androgen production by theca cells (214). In primary human granulosa cells, in vitro treatment with testosterone demonstrated a concentration-dependent inverse correlation with methylation status of the YAP1 promoter, with increases in testosterone concentration resulting in decreases in YAP1 promoter methylation (211). However, no changes in methylation of the YAP1 promoter were observed after treatment with either FSH or LH (211). The suggestion that ovarian androgens can elevate YAP1 expression in human granulosa cells is supported by studies in murine granulosa cells demonstrating that testosterone can enhance the expression and activity of YAP1 as discussed in a previous section of this review (97). Considering the hormonal imbalance in women with PCOS, it is likely that elevated concentrations of testosterone contribute to the elevated expression and activity of YAP1 in human ovarian cells, creating a favorable environment for PCOS development (Table 5). The elevated content of YAP1, observed in the ovaries of women with PCOS, is available for further posttranslational modifications that regulate its activity and consequently the expression of downstream genes.

Table 5.

Hippo pathway and PCOS

Study (first author, year)	Model system	Method	Findings
Li et al, 2012	Human GWAS study	SNP genotyping	YAP1 variants rs11225138, rs11225161, rs11225166 associated with PCOS susceptibility
Jiang et al., 2017	Primary human granulosa cells from IVF patients	Bisulfite sequencing PCR	Hypomethylation of YAP1 promoter; increase in YAP1 mRNA and protein in women with PCOS; elevated levels of testosterone decreased YAP1 promoter methylation; no changes in YAP1 promoter methylation with FSH, LH

Abbreviations: FSH, follicle-stimulating hormone; GWAS, genome-wide association study; IVF, in vitro fertilization; LH, luteinizing hormone; PCOS, polycystic ovary syndrome; PCR, polymerase chain reaction; SNP, single nucleotide polymorphism; YAP1, yes-associated protein 1.

As discussed in earlier sections of this review, the activity of YAP1 can be modulated by numerous upstream kinases, including AMPK via phosphorylation of YAP1 on Ser94, which disrupts the interaction of YAP1 with the transcription factor TEAD, thus repressing YAP1 target genes and restricting cell growth (77). Metformin, a commonly used therapeutic to treat women with PCOS with insulin resistance (215), is also an AMPK activator and has been shown to target YAP1 (216-218). Previous studies demonstrate that AMPK activation inhibits steroidogenesis in ovarian granulosa cells primarily by reducing the ability of gonadotropins to stimulate the transcription of genes required for steroidogenesis (140, 219). Because metformin activates AMPK in ovarian granulosa cells and impairs progesterone production (140), it is tempting to speculate that metformin via AMPK inhibits YAP1 interactions with transcriptional regulators in granulosa cells and directly or indirectly disrupts the steroidogenic machinery. Further, because the ovaries of PCOS patients appear fibrotic, it seems likely that the lower activity of AMPK observed in ovaries of women with PCOS (220, 221) could contribute to an interaction between YAP1 and TEAD and the transcription of genes encoding factors related to connective tissue growth and cellular communication network (CCN) growth factors. More studies should be undertaken to clarify the interactions between Hippo signaling and other signaling pathways to better understand their role in the etiology of PCOS.

Primary Ovarian Insufficiency

Ovarian Fragmentation

The Hippo signaling pathway is responsive to external stimuli, including a variety of ligands interacting with specific receptors and alterations in the physical environment and cell shape (222). In vitro activation therapy is an alternative for women who have previously experienced a poor response to gonadotrophin-mediated follicle stimulation. In this procedure, cortical strips from the ovary are surgically removed, fragmented, and transplanted back into the patient, or cultured in the presence of protein kinase B (Akt) stimulators prior to implantation (82, 223, 224). In this regard, fragmentation of the ovary has been shown to promote actin polymerization, resulting in disruption of Hippo signaling, reduced phosphorylation of YAP, and elevation of YAP1 abundance in the nucleus (82, 87, 98, 120, 225). The disruption of Hippo signaling leads to the transcription of YAP1-responsive genes including the expression of cellular communication network growth factors (CCN1, 2, 3, 5, and 6) and members of the inhibitor of apoptosis gene family (BIRC1 and 7) in follicular somatic cells, thus, promoting follicular growth (82, 225, 226). Upon fragmentation of the ovary, YAP1 expression was primarily located in the nucleus of the oocyte and granulosa cells in the cultured human ovary wedge sections, while YAP1 expression was observed in both the cytoplasm and nucleus of granulosa cells and oocytes in the intact cultured ovary controls (87). Further studies showed that treatment of murine ovaries or human granulosa cells with drugs that stimulate actin polymerization such as jasplakinolide and sphingosine-1-phosphate resulted in increased nuclear YAP1 localization and expression of the downstream growth factor Ccn2 and activation of follicle growth (225, 227). Likewise, Akt stimulation after ovarian fragmentation promoted greater secondary and antral follicle growth (82). In contrast, the Akt inhibitor MK2206 suppressed follicle growth, which was partially rescued upon overexpression of Yap1 (83). Similarly, in an ovarian cortex model that was microinjected with rhabdomyosarcoma cells, inhibition of YAP1/TEAD interaction via verteporfin and verteporfin plus a chemotherapeutic was able to effectively clear the tumor cells without any impacts on ovarian tissue or follicular integrity as measured by metabolic activity and Neutral red uptake (228). While this study was purely proof-of-principle, it would be interesting to see if these follicles could be activated and their subsequent viability after treatment with verteporfin and chemotherapy drugs. Together, procedures of mechanical stimulation and/or pharmaceutical administration in order to disrupt the Hippo pathway or YAP1/TEAD activity in order to control follicle growth have potential for treating infertility.

In ovarian samples derived from transgender men or oncology patients, in vitro ovarian fragmentation also led to activation of primordial follicles through inhibition of the Hippo pathway (120). The fragmentation and culture of human ovarian tissue reduced protein levels of phosphorylated YAP1 in the granulosa cells of primordial follicles and increased levels of total YAP1 upon follicle growth and activation (120). In parallel with the increase in nuclear YAP1 observed with follicle activation, the downstream YAP1 target CTGF was upregulated (120), lending further support for the role of the Hippo pathway in follicle activation. Interestingly, a notable difference in YAP1 was observed in the samples from oncology patients vs the transgender men, in that nuclear YAP1 expression in granulosa cells from the oncology patients was minimal, suggesting this could be possible due to the etiology of the oncological disease or hormone therapy prior to surgery (86, 120). These data reflect what has been reported in murine studies in which testosterone induces the translocation of YAP1 from the cytoplasm to the nucleus of granulosa cells (97), suggesting conserved regulators and mechanisms between species.

In contrast to the studies described above, another study reported that fragmentation of the human ovarian cortex did not induce changes in actin polymerization, as revealed by failure to observe changes in the ratios of globular and filamentous actin after tissue fragmentation (229). Further, levels of YAP1 and phosphorylated YAP1 proteins were unaltered, as were levels of transcripts for the YAP1 downstream targets CCN2, CCN3, and CCN5 (229). As a result, no apparent changes were observed in the number of growing follicles in fragmented cortex vs intact controls, though the overall number of total follicles was lower in the fragmented pieces relative to intact cortex (229). Further study is needed to clarify the molecular and physiological mechanisms of Hippo regulation and its role in follicle activation after fragmentation in human ovaries. Although some key molecular mechanisms that regulate Hippo signaling have been identified, many aspects of how these are modulated under physiological or stress conditions remain poorly explored.

Ovarian Cell Malignant Neoplasia

Ovarian cancer is the most lethal gynecological malignancy and the fifth leading cause of cancer death in women in the United States. American Cancer Society estimated that more than 21 500 women would be diagnosed with ovarian cancer, and around 14 000 women would die of this disease in 2020 (230). Globally, it is estimated that ovarian cancer claims the lives of more than 184 000 women annually (231). Ovarian cancer is sometimes referred to as a “silent killer” because most patients with ovarian cancer lack specific symptoms and are diagnosed in the advanced stages of the disease. Despite advances in surgical technique, postoperative care, and chemotherapy, ovarian cancer’s overall survival rate has only improved modestly in the past decades. This could be attributed to the fact that the risk factors and the etiology of ovarian cancer are not fully understood.

Ovarian cancer is not one disease. Ovarian cancers represent a spectrum of distinct types of cancers that may originate from organs (or tissues) other than ovaries. For example, accumulating evidence showed that high-grade serous ovarian cancer (HGSOC), the most common and lethal subtype of ovarian cancer, may originate from fallopian tube epithelium (232-234). In the ovary, ovarian cancer can arise from cells such as ovarian surface epithelial cells, granulosa cells, germ cells, theca cells, and specialized gonadal stromal cells. Accumulating evidence has now shown that the Hippo pathway plays a vital role in the development of ovarian cancer from ovarian cells.

Numerous studies demonstrate that YAP1 is an ovarian cancer oncogene (Table 6). YAP1 protein is localized in both the cytoplasm and nucleus in 98% of ovarian serous cystadenocarcinoma patients, while in normal ovarian tissues, only 25% of the samples display nuclear YAP1, and 50% of the samples display only cytoplasmic YAP1 expression (235). Consistently, a high level of nuclear YAP1 is associated with poor patient survival rate and is an independent prognostic marker for ovarian cancer outcome (93, 236-238). The involvement of the disrupted Hippo pathway in the development of ovarian cancer is further supported by multidimensional cancer genetic/genomic studies performed by the Cancer Genome Atlas Research Network (TCGA) (239). Data extracted from TCGA datasets (239) demonstrate that genes encoding the upstream tumor suppressors are frequently deleted, mutated, or downregulated, while genes encoding the downstream oncogenic proteins are frequently amplified or upregulated in the Hippo signaling cascade (95). TCGA data also clearly suggest high nuclear YAP1 is significantly related to a poor prognosis of ovarian serous cystadenocarcinoma (239). In addition to its potential role in ovarian cancer development, hyperactivation of YAP1 increases the resistance of ovarian cancer cells to drug-induced cell apoptosis, suggesting that the Hippo/YAP1 pathway may also be involved in ovarian cancer chemoresistance (238, 240, 241).

Table 6.

Knockout/overexpression studies of Hippo pathway core components and ovarian cancer phenotypes

Study (first author, year)	Model system	Gene targeted	Overexpression/knockout	Findings
St John et al., 1999	Lats1 -/- mouse model	LATS1	KO	Development of ovarian stromal tumors
Xia et al., 2014	A2780 cell line	YAP1	OE	Increased cancer cell proliferation and migration; enhanced anchorage-independent growth; increased chemoresistance
			KO	Decreased cancer cell proliferation and migration; increased chemosensitivity
	OVCAR8 cell line	YAP1	OE	Increased cancer cell proliferation and migration; enhanced anchorage-independent growth; chemoresistance
			KO	Decreased cancer cell proliferation and migration; increased chemosensitivity
He et al., 2015	HOSE-T80 cell line	YAP1	OE	Increased cancer cell proliferation; malignant transformation; enhanced anchorage-independent growth; induced tumorigenesis in vivo
			KO	Suppressed cancer cell proliferation
	TOV21G cell line	YAP1	OE	Increased cancer cell proliferation; malignant transformation; enhanced anchorage-independent growth; induced tumorigenesis in vivo
			KO	Suppressed cancer cell proliferation
Hua et al., 2016	FTSEC cell lines	YAP1	OE	Increased cancer cell proliferation; malignant transformation; enhanced anchorage-independent growth; induced tumorigenesis in vivo
			KO	Suppressed cancer cell proliferation
Lv et al., 2020	HGrC1 cell line	YAP1	OE	Induced tumorigenesis in vivo
	HGL5 cell line	YAP1	OE	Induced tumorigenesis in vivo

Abbreviations: KO, knockout; LATS1, large tumor suppressor 1; OE, overexpression; YAP1, yes-associated protein 1.

Ovarian surface epithelial cells (OSE) or ovarian cortical inclusion cyst epithelial cells have long been considered as the cell of origin of epithelial ovarian cancer, including HGSOC (232). However, recent studies favor the concept that HGSOC may originate from the fallopian tube epithelial cells (FTE) (232, 242-246). Accumulating evidence suggests that OSE cells or epithelial cells from the inclusion cysts may be the cell of origin of the low-grade serous carcinoma (247). Interestingly, there is also evidence supporting the concept that low-grade serous carcinoma could progress to HGSOC (248-252). By introducing the same genetic abnormalities into both FTE and OSE cells, Zhang and colleagues (253) indicated that HGSOC might originate from either cell type. Therefore, the role of ovarian surface epithelial cells in the development of HGSOC cannot be excluded. Upon hyperactivation of YAP1 in immortalized FTE and OSE, the malignant transformation of both types of cells and the development of high-grade ovarian cancer was observed (93, 95). Consistent with these observations, early studies demonstrated that overexpression of YAP1 and phosphorylation-defective YAP1-5SA in immortalized OSE cells resulted in increased cell proliferation, resistance to cisplatin-induced apoptosis, faster cell migration, and anchorage-independent growth, whereas YAP1 knockdown resulted in increased sensitivity to cisplatin-induced apoptosis (237). In ovarian cancer cell lines, ectopic expression of hyperactivated YAP1 promotes cell proliferation and migration, while knockdown of YAP1 suppresses cancer cell growth in vivo and in vitro, suggesting that the Hippo/YAP1 signaling pathway also plays a role in ovarian cancer progression (93, 95).

The uncontrolled growth of granulosa cells also results in ovarian cancer. Ovarian cancers developed from ovarian granulosa cells are called granulosa cell tumors (GCT) and account for about 70% of malignant sex-cord stromal tumors (254). The mechanisms underlying GCT development, recurrence, and metastasis are poorly understood. Previous reports have found that YAP1 is highly expressed in human GCT tissues and overexpression of YAP1 protein significantly stimulates the proliferation and migration of the KGN GCT cell line, demonstrating that the Hippo pathway may be involved in the progression of GCT (92). Interestingly, whole transcriptome sequencing of human samples has demonstrated that a somatic mutation in FOXL2 (C402G; Cys134TrP) is present in almost all morphologically identified adult-type GCTs (255), suggesting that mutant FOXL2 is a potential driver in the pathogenesis of adult-type GCTs. Whether the Hippo pathway interacts with mutated FoxL2 to drive the development of GCT has not been studied.

Unexpectedly, recent studies demonstrated that hyperactivation of YAP1 in well-differentiated or luteinized granulosa cells could induce dedifferentiation and reprogramming of the cells, leading to the development of high-grade ovarian cancer with serous features (101). The hyperactivation of YAP1 stimulated expression of TERT and MYC, genes associated with stemness, suggesting that YAP1 dysregulation may result in the reprogramming of granulosa cells (101). This idea was supported by gene set enrichment analysis showing downregulation of genes associated with granulosa cell differentiation and epithelial-mesenchymal transition and upregulation of genes associated with cell stemness and pluripotency (101). Since granulosa cells are of mesenchymal lineage, these cells may be a candidate cell of origin of the mesenchymal subtype of HGSOC. More studies are required to confirm this speculation.

The Hippo pathway may also play a role in the rare cancers of the ovary. Malignant transformation of germ cells leads to the development of germ cell tumors. Although YAP1 is reported to be highly expressed in the germ cells (121), the role of YAP1 in the development of germ cell tumors has not been studied. Due to the nature of the germ cell tumor and a critical role of YAP1 in the maintenance of cell pluripotency, YAP1 may play a critical role in the germ cell development (256-260). The development of ovarian stromal cell tumors in the LATS1 knockout mice demonstrated that the Hippo pathway’s disruption might also contribute to the development of ovarian stromal tumors (112).

Mutation of the YAP1 oncogene is rare in ovarian cancer, suggesting that additional oncogenic perturbations may account for YAP1 activation and subsequent tumor progression. The exact mechanisms by which the Hippo pathway regulates ovarian cancer development is largely unknown. In primary human ovarian surface epithelial (hOSE) cells, the hyperactivation of YAP1 promotes cell proliferation until cellular senescence occurs after ~7 passages (113). Interestingly, siRNA depletion of either YAP1 or TAZ did not stop the proliferation of cultured hOSE cells, though double knockdown of YAP1 and TAZ effectively reduced cell proliferation (113). The suggested mechanism behind this observation is that sustained YAP1 activity resulted in a compensatory elevation in LATS2 expression, thus creating a negative feedback loop preventing the potential for malignant transformation (113). Hyperactivated YAP1 forms feedforward loops with receptors and signaling pathways in the epidermal growth factor family (ERBB) and fibroblast growth factor receptor family (FGFR) to drive the development of ovarian cancer cells from OSE and fallopian tube secretory epithelial cell (FTSEC) cell lines (93, 95, 261).

The GPCR ligand lysophosphatidic acid, mediated by RhoA-ROCK signaling, suppresses the Hippo pathway to drive the proliferation and migration of immortalized ovarian surface epithelial cells (261). Interestingly, protein kinase C iota (PRKCI) amplification, an early event in ovarian cancer development, has been shown to promote immune suppression by driving nuclear YAP1 accumulation, leading to the expression of proinflammatory cytokines, including tumor necrosis factor (TNF) (262). A more recent study demonstrates that arrestin Beta 1 (ARRB1) regulates YAP1 activity by interacting with Trio Rho guanine nucleotide exchange factor (TRIO), a member of the RhoGEF family that induces nuclear YAP1 accumulation and aberrant expression of TEAD genes in HGSOC. Interestingly, ARRB1 also acts as a nuclear anchor for mutated TP53 and YAP1, leading to the recruitment of diverse transcription factors, thus expanding the repertoire of ARRB1 induced downstream genes (263). Several studies also indicated that downregulation of myosin phosphatase targeting protein 1 (MYPT1) and overexpression of miR-30b or miR-149-5p suppressed the Hippo pathway to induce chemotherapeutic resistance (236, 241), possibly via regulating pluripotency genes influencing ovarian cancer stem cells (238, 240). Taken together, these studies demonstrate the dynamic ability of the Hippo pathway to control tissue homeostasis.

Concluding Remarks and Looking Forward

The Hippo pathway and downstream transcriptional effectors YAP1 and TAZ have emerged as an important signal transduction pathway in ovarian development and function, regulating several events such as cell fate determination, germ and follicle cell proliferation/migration, follicle activation/development, and apoptosis. The observations from the multi-species studies discussed herein have begun to elucidate the functional diversity and molecular mechanisms of Hippo signaling in the ovary, as well as identifying interactions with other signaling pathways and responses to various external stimuli, translating, and generating cellular responses that maintain ovarian tissue growth and homeostasis. It seems likely that transcriptional regulators activated by ovarian paracrine and endocrine factors interact with Yap1 and other Hippo pathway members to control follicle development, ovulation, and differentiation. Employment of approaches combining proteomics and transcriptomics will assist in identifying Hippo pathway components and their targets in specific ovarian compartments under physiological conditions. While several of the studies discussed in this review define distinct roles for Lats1, Lats2, and Yap1 in ovarian physiology, limited studies are available on our understanding of Mob1, Sav1, Mst1, Mst2, and Taz, and their impacts on pathway dynamics and regulation in the ovary remain to be elucidated. Further research is imperative in distinguishing the direct or indirect mechanisms of Hippo regulation in ovarian cells and how these mechanisms are directed during the various stages of follicle development. Moreover, disruption and/or dysregulation of the Hippo pathway has been implicated in the development of ovarian disorders, ovarian cancers, and impaired fertility, thus, providing another avenue for research into ovarian Hippo signaling. These investigations could possibly serve to identify cell-based molecular targets for prevention, diagnosis, and/or treatment of ovarian disorders. Further, understanding the role of Hippo signaling components during follicular maturation is necessary for improving applications in human reproductive technologies.

Despite the rapidly accumulating reports on the Hippo pathway’s contributions to ovarian function, some key questions remain unanswered, and along with new information, new questions are emerging. Listed here are some key questions for further study of this emerging signaling pathway.

• (1) Where are Hippo pathway components localized in ovaries during the reproductive cycle? MST1 and MST2 are the core kinases of the mammalian Hippo tumor suppressor pathway; however, their localization and role during various stages of follicle development, ovulation, and luteal formation and regression are unknown. This may be key to understanding how external forces, hormones, and paracrine factors regulate upstream Hippo pathway components. Studies show that phosphorylated YAP1/TAZ are enriched in the cytoplasm and dephosphorylated YAP1/TAZ are present in the nucleus of oocytes and granulosa and luteal cells under various conditions. However, it remains to be determined where YAP1 and TAZ phosphorylation and dephosphorylation occur. A related question is how do non-LATS1/2 mediated YAP1 and TAZ phosphorylation sites (e.g., via PKA and AMPK) impact ovarian gene transcription?

• (2) Does Hippo signaling regulate follicular and luteal size sensing? What Hippo signaling factors promote or limit follicular growth? Understanding these signals may improve in vivo and in vitro approaches to improve oocyte maturation and increase reproductive success. A related question is how are LATS1/2 regulated by actin remodeling and/or cellular stiffness? Ovarian stiffness and fibrosis contribute to reduced follicular activity, especially in the aging ovary, and approaches to alleviate fibrosis may improve fertility.

• (3) How does the ovary integrate the many signaling mechanisms that impact Hippo signaling? In cancer, multiple studies have shown that there is an interactive regulation between Hippo signaling and metabolic signaling. Does the Hippo pathway act as a metabolic sensor in the ovary? Are endocrine or metabolic disturbances the cause or effect of altered ovarian morphology and infertility associated with PCOS?

• (4) An equally intriguing question is how do gonadotropins fine tune Hippo signaling in vivo to achieve an exquisite balance of cellular proliferation and differentiation? Studies are needed to identify the cell-type specific responses occurring in follicular granulosa and theca cells, luteal steroidogenic and endothelial cells, as well as the ovarian stroma.

• (5) How do YAP1 and TAZ direct ovarian gene expression? As a transcriptional co-regulator, YAP1 interacts with TEADs to regulate a transcription of genes that impact proliferation. However, evidence shows that YAP1 interacts with other transcription factors to provide a unique transcriptional landscape. Studies using state-of-the-art approaches are needed not only to identify YAP1/TAZ targets in ovarian cells, but also to determine how YAP1/TAZ intersect with other key transcription factors that are activated during distinct stages of the ovarian cycle. Equally important, what are the unique contributions of TAZ to ovarian function, as distinct from YAP1? Understanding how YAP1 and TAZ direct ovarian gene expression can provide new opportunities for strategies to improve fertility or contraception.

• (6) How does Hippo signaling become deregulated in cancer? It is apparent that elevations of YAP1 contribute to disordered granulosa cellular identity and that that YAP activation is capable of cellular transformation leading to ovarian cancers. Although mutations in Hippo pathway genes are rare, the cell of origin for the various ovarian cancers may contribute other cancer-driving pathways that regulate the Hippo pathway.

Glossary

Abbreviations

AMPK

adenosine monophosphate–activated protein kinase

CCN1/2/3/5

cellular communication network factor 1/2/3 or 5

COC

cumulus oocyte complex

CTGF

connective tissue growth factor (also known as CCN2)

ECM

extracellular matrix

EGF

epidermal growth factor

FOXL2

forkhead box L2

FSH

follicle-stimulating hormone

FSHR

follicle-stimulating hormone receptor

FTE

fallopian tube epithelial cells

GCT

granulosa cell tumor

GDF9

growth differentiation factor 9

GPCR

G protein–coupled receptor

HGSOC

high-grade serous ovarian cancer

LATS1/2

large tumor suppressors 1 and 2

LH

luteinizing hormone

MAP4K

mitogen-activated protein kinase kinase kinase kinase

MOB1

Mps one binder 1

MST1/2

mammalian serine/threonine-protein kinase 4 and 3

mTOR

mechanistic target of rapamycin

OSE

ovarian surface epithelial cells

PCOS

polycystic ovary syndrome

PI3K

phosphoinositide 3-kinase

PFAS

per/polyfluoroalkyl substances

PFOA

perfluorooctanoic acid

PKA

protein kinase A

SAV1

Salvador family WW domain-containing protein 1

siRNA

small interfering RNA

SREBP

sterol regulatory-element binding protein

STAR

steroidogenic acute regulatory protein

TAZ

WW domain-containing transcription regulator 1 (also known as WWTR1)

TEAD

TEA domain transcription factor

TGF-B

transforming growth factor beta

TOCP

tri-ortho-cresyl phosphate

YAP1

yes-associated protein 1

Financial Support

This work was supported by NIH F32HD106722 (K.L.C.), Olson Center for Women’s Health (J.W.G.), VA IK2BX004911 (M.R.P.), NIH R01CA197976 (C.W./J.S.D.), NIH R01CA201500 (C.W./J.S.D.), NIH P01AG029531 (J.S.D.), NIH R01HD092263 (J.S.D.), VA I01BX004272 (J.S.D.), and the Olson Center for Women’s Health (J.S.D.). J.S.D. is the recipient of a VA Senior Research Career Scientist Award (IK6BX005797).

Author Contributions

K.L.C, J.W.G, and J.S.D contributed to conception, study design, literature analysis, manuscript drafting, and revision; E.P., M.R.P., and G.H. contributed to literature analysis, manuscript drafting, and revision; C.W. contributed to manuscript drafting and editing.

Disclosures

All authors confirm that there is no conflict of interest.