Functions of RhoA in the female reproductive system

The oviduct (in rodents)/fallopian tube (in humans) is the site of fertilization and early embryo development and transport. Our unpublished immunohistochemistry data show RhoA expression on day 3.5 post-coitum (D3.5; embryo implantation is initiated at ~D4.0 in mice) in the mouse oviductal epithelium, and our unpublished microarray data revealed that RhoA is among the most abundant genes in the D3.5 mouse oviduct. The Human Protein Atlas designates RHOA RNA expression to ciliated and secretory cells, endothelia and lymphatic endothelia, smooth muscle and fibroblasts, as well as T cells, plasma cells, granulocytes, and macrophages within human fallopian tubes [ 76 ] . However, the functions of RhoA in the oviduct/fallopian tube have not been systematically investigated. In a study, Ras-related nuclear protein (RAN), which promotes membrane targeting and stabilization of RhoA, was detected in small extracellular vesicles derived from human fallopian tube epithelial cells [ 83 ] . This observation may implicate RhoA signaling as a mediator of cell–cell communication in the oviduct/fallopian tube, such as during fertilization, early embryo development, and embryo transport. It has been demonstrated that CDC42, one of the three most-studied Rho GTPases, is critical for embryo transport in the mouse oviduct [ 84 ] . The potential in vivo functions of RhoA in the oviduct/fallopian tube require further investigation, and we are currently investigating the in vivo function of RhoA in the oviduct during early pregnancy using a RhoA f/f Pgr Cre/+ mouse model [ 22 ] .

The RhoA is a small GTPase that was first discovered in the mollusk Aplysia and then in humans in 1985 by Madaule and Axel [ 1 ] . RhoA functions by cycling between an active GTP-bound state and an inactive GDP-bound state, changing its conformational state and relative activity. The GTPase activity of RhoA is modulated by GTPase-activating proteins (GAPs) and guanine nucleotide exchange factors (GEFs), with further regulatory effects by guanine nucleotide-dissociation inhibitors (GDIs) [ 2 ] . GAPs inactivate RhoA by increasing the rate of GTP hydrolysis, returning RhoA to its inactive form, whereas GEFs provide a counter-mechanism by expediting the release of GDP, thereby allowing new GTP to bind RhoA. Specific GAPs and GEFs can be turned on or off by various molecular signals, allowing for finely tuned control of RhoA activity. RhoA activity is associated with its cellular localization. For example, p115RhoGEF binds to active G protein-coupled receptors (GPCRs) in the plasma membrane and activates RhoA upon ligand binding [ 3 ] . Common GPCR ligands for RhoA activation include thrombin, lysophosphatidic acid (LPA), sphingosine-1-phosphate (S1P), and thromboxane A2 [ 4 , 5 ] . When bound to GDP, RhoA is sequestered in the cytoplasm by RhoGDIs in its latent state. RhoA is fully activated upon translocation to the membrane and dissociation from the GDI. Dissociation from GDI may be facilitated by a GDI displacement factor (GDF), which binds to RhoGDI to facilitate RhoA release and activation [ 6 ] , or by phosphorylation events initiated by kinases, such as protein kinase C (PKC) or p21-activated kinase (PAK) [ 7 ] . After activation, RhoA is either located in the cytosol or associated with the plasma membrane via its prenyl group [ 3 ] . Although GTP-bound RhoA is assumed to be active, an inactive conformation of GTP-bound RhoA has been reported [ 8 ] . In addition, GTP-bound RhoA may be localized in the cytosol or at the membrane; however, membrane-localized GTP-bound RhoA is required for the activation of RhoA downstream signaling [ 9 ] (Fig. 1 ). Further, RhoA can localize within the nucleus for actin polymerization and DNA damage response, with the involvement of various GEFs and GAPs present within the nucleus [ 10 ] . Intracellular RhoA signaling. RhoA may be activated via GPCR-induced G protein activation of RhoGEF, which exchanges GDP for GTP to activate RhoA. Active RhoA may interact with a variety of downstream effectors, including ROCK1/2. RhoA is inactivated by RhoGAP, which accelerates the hydrolysis of GTP to GDP. Inactive RhoA is sequestered by RhoGDI, which forms a complex with GDP-RhoA. Phosphorylation events initiated by kinases such as PKC or PAK phosphorylate sites on RhoGDI or actions of RhoGDF initiate dissociation and free GDP-RhoA to continue the cycle. GPCR: G protein-coupled receptor; PAK: p12-activated kinase; PKC: protein kinase C; RhoA: Ras homolog gene family member A; RhoGAP: Rho GTPase activating protein; RhoGDI: Rho guanine nucleotide-dissociation inhibitors; RhoGEF: Rho guanine nucleotide exchange factors; ROCK1/2: Rho-associated coiled-coil kinases 1/2. The figure was generated using BioRender. RhoA activity is also regulated at transcriptional and post-translational levels. The transcription factors Myc, hypoxia inducible factor 1 (HIF-1), signal transducer and activator of transcription 6 (STAT6), and nuclear factor-kappa B (NF-κB) are involved in RhoA transcription in various physiological contexts [ 3 ] . Myc may also be involved in a negative feedback mechanism to post-transcriptionally block RhoA/ROCK signaling, which is involved in cytoskeletal function [ 11 ] . RhoA may be modified via ubiquitylation of lysine; phosphorylation of tyrosine, threonine, or serine; nitration of tyrosine; and AMPylation of tyrosine at conserved sites. Nitration of tyrosine Y34 can reduce GDP affinity, whereas AMPylation of Y34 can block interactions with downstream effector proteins, thus serving as an additional control mechanism over RhoA activation/inactivation [ 12 ] . Serine/threonine phosphorylation at S/T26 also inhibits RhoA, whereas lysine ubiquitylation is thought to increase protein stability. The tyrosine residues Y34 and Y64 are phosphorylated under normal conditions. Active RhoA can activate its downstream proteins Rho-associated coiled-coil kinases 1 and 2 (ROCK1/2, or ROK, Rho-kinase, and p160ROCK, colloquially referred to as ROCK(s)), p140mDia (mDia), and other targets [ 13 ] . ROCK proteins regulate actin organization, apoptosis, tissue and organ development, cell proliferation, and cytokinesis [ 14 ] . mDia inhibition is relieved by a conformational change upon RhoA binding, enabling mDia to play important roles in actin polymerization and cytokinesis [ 13 ] . ROCK proteins are ubiquitously expressed, with high ROCK-1 expression reported in all tissues except the brain and muscle, whereas ROCK-2 expression is higher in the brain, muscle, heart, lungs, and placenta [ 15 ] . RhoA signaling has several functions, including the regulation of cell proliferation, survival, migration, adhesion, cytokinesis, stress fiber formation, apical constriction, stem cell lineage fate, vesicle trafficking, and gene expression [ 4 , 16 – 21 ] . Despite its ubiquitous expression in eukaryotic cells, RhoA has essential tissue- and cell type–specific functions, such as protection from heart injuries, regulation of lens epithelium morphology and the hematolymphoid system, diverse functions in different parts of the nervous system [ 17 ] , progesterone (P4) steroidogenesis in the corpus luteum (CL) [ 22 ] , and regulation of immune responses [ 23 , 24 ] , that have been revealed in conditional RhoA knockout mouse models. RhoA signaling can be studied using various methods, both in vitro and in vivo . Classical approaches before the development of live fluorescent probes measured active RhoA activity using an active RhoA-specific antibody, analyzed through pulldown assays and visualized through localized fluorescence [ 25 ] . Modern techniques have since been developed, allowing for live spatiotemporal reporting of active RhoA. The first technique uses a binding domain specific to RhoA-GTP fused to a fluorescent probe with the ability to relocalize to the native location of active Rho GTPase [ 26 ] . Although groundbreaking, this technique has the following disadvantages: competition with endogenous effectors and no preferential binding between RhoA, RhoB, and RhoC. A second method has been developed that uses fluorescence resonance energy transfer (FRET), which relies on exogenous RhoA fused to a RhoA-binding domain that may accomplish FRET upon activation, allowing localized reporting of activated exogenous RhoA [ 27 ] . However, this method has the disadvantage of only measuring exogenous RhoA, which may localize and migrate differently when overexpressed compared with endogenous RhoA. Both approaches are robust for in vitro labeling of GTP-bound RhoA. Notably, the approach of using FRET-competent RhoA can also be modified for in vivo use by creating a RhoA-FRET mouse model that utilizes a lox-stop-lox site coupled with Cre recombinase for conditional expression in specific tissues [ 28 ] . Other in vitro strategies include mutant overexpression of RhoA or inactivation of RhoA, RhoB, and RhoC via Clostridium botulinum C3 toxin; however, these approaches may result in the loss of some of the overarching signaling pathways and motifs owing to the use of immortalized cell lines and general in vitro conditions, which may alter the biochemical environment [ 17 ] . The cyclic nature of GTPases makes it challenging to study the real-time functions of active RhoA, especially in vivo . The Cre-loxP system can be used to conditionally delete RhoA from specific cells or tissues. Elucidating the upstream and downstream signaling regulators and effectors of RhoA is more difficult as there are a variety of GEFs, GAPs, and GDIs that are specific to cell types and have varied cascades to impart the effects of RhoA [ 3 , 4 ] . An informed approach is essential when knocking out one of these proteins to study its function. For example, knockout of the GEF Ect2 in mice was informed by its interphase nuclear localization and release during cytokinesis, and Ect2 knockout caused impaired cell proliferation and migration [ 29 ] . This finding is consistent with the known functions of RhoA in cytokinesis. However, some effectors may also display phenotypes atypical to established RhoA functions, such as the RhoGEF rgnef , the knockout of which in MEF cells blocks focal adhesion formation [ 30 ] , whereas RhoA knockout MEF cells showed no alterations in focal adhesions [ 31 ] . ROCK kinases also present an opportunity to evaluate RhoA signaling, and deletion of either of these kinases or their own downstream targets may be a promising approach to elucidate the latent functions of RhoA. For example, deletion of ROCK-1 or ROCK-2, or mutation of their downstream target, MAP3K1, results in an eyes-open-at-birth phenotype in mice, which is atypical for murine development. Similar results were observed with RhoGEF deletion; however, RhoA deletion did not result in alterations in eyelid development [ 32 , 33 ] . The existing literature illustrates the possibility of multiple pathways that act on RhoA and its regulators and effectors, necessitating an investigation of the GTPase itself and an independent investigation of upstream and downstream targets. The mammalian FRS generally consists of two ovaries, two fallopian tubes/oviducts, a uterine body composed of one central compartment or two uterine horns, a placenta/placentas that transiently develops during pregnancy, a cervix, a vagina, and mammary glands. The functions of the FRS are under the central control of the hypothalamic–pituitary–ovarian (HPO) axis and are regulated by local factors [ 34 – 38 ] . RhoA is expressed in cells of the FRS and RhoA signaling has been noted in FRS tissues/cells [ 39 – 48 ] . These findings have led to a greater understanding of the in vivo functions of RhoA. This is the first review of the function of RhoA in the FRS. It is organized as follows: ovary, oviduct, uterus/corpus uterus, cervix, placenta, vagina, and mammary gland. Both physiological functions and pathological implications are discussed (Fig. 2 ). Summary of reported functions of RhoA in the female reproductive system. CL: corpus luteum; KO: knockout; P4: progesterone; PCOS: polycystic ovarian syndrome; RhoA: Ras homolog gene family member A; StAR: steroidogenic acute regulatory protein; TEB: terminal end buds. The figure was generated using BioRender.

S.V., Y.L., and X.Y. conceived the idea; S.V. initiated the drafting of the manuscript as her senior thesis; J.K.S. wrote the manuscript with assistance from all coauthors; J.K.S. drew the figures; all authors contributed to the construction and revision of the manuscript; and X.Y. supervised the entire process.

This is the first comprehensive review of studies of RhoA in the FRS. It provides an emerging picture of the spatiotemporal functions of RhoA in both physiological and pathological conditions. Significant knowledge gaps still exist regarding the cell type–specific and stage-specific functions of RhoA in the FRS, especially in vivo . Cell type–specific RhoA upstream and downstream signaling pathways associated with each RhoA function remain to be elucidated. Spatial transcriptomic technology enables the spatiotemporal transcriptome analysis of genes. However, this technology has limitations in studying RhoA, as RhoA activity is mainly regulated at the post-transcriptional level. The Cre-loxP system is a powerful tool for studying the loss of cell type–specific gene functions. However, the use of this technology for studying cell type–specific functions of molecules in RhoA signaling pathways in FRS has been limited owing to the ubiquitous expression of many molecules in RhoA signaling pathways, including RhoA, as well as by the lack of suitable cell-type–specific Cre mice. For example, during our attempts to generate a uterine epithelial–specific RhoA conditional knockout mouse model using two commonly used Cre mice, Wnt7A Cre mice and Ltf Cre mice, the former yielded RhoA f/f Wnt7A Cre mice with hydrocephalus and postnatal death [ 164 ] , while the latter was genetically impossible owing to the localization of both RhoA and Ltf on chromosome 9. As the FRS is under hormonal control, the pathways through which RhoA signaling mutually interacts with hormonal signaling is an important area that requires extensive investigation. RhoA has been implicated in many pathological conditions in the FRS; therefore, targeting RhoA signaling to develop therapeutic strategies needs to be explored.

The mammary glands are vital postpartum structures that depend heavily on developmental regulation during both puberty and pregnancy for successful function. Branching morphogenesis is initiated during puberty by growth hormones and estrogen, among other factors, and the actions of progesterone and prolactin during pregnancy initiate further development of mammary glands [ 143 ] . In vitro murine studies have shown that mammary branching morphogenesis is partly dependent on ROCK-mediated RhoA signaling, with the inhibition of RhoA/ROCK signaling resulting in a disorganized, hyperbranched phenotype and incomplete myoepithelial cell coverage [ 144 , 145 ] . P190A-RhoGAP and P190B-RhoGAP are negative regulators of RhoA. Terminal end buds (TEB) are vital structures that form at the end of a developing mammary duct and drive ductal development. Overexpression or deletion of these isoforms in mice disrupts TEB architecture and proliferation and alters mammary gland differentiation and E2 and P4 receptor expression [ 146 – 148 ] . Knockout of the RhoGEF Net1, an activator of RhoA, has also been shown to result in impaired murine mammary gland development, with reduced phosphorylation of myosin light chain and myosin light chain phosphatase, along with decreased ERα expression [ 149 ] . The Human Protein Atlas indicates that RhoA mRNA and protein expression in human mammary glands is primarily localized to adipocytes with modest glandular expression [ 76 ] . These studies demonstrate that the regulation of RhoA, such as the coordination of other structural developments, is essential for mammary gland development. Species-specific regulation of RhoA activity and functions in the mammary gland requires further investigation. Mammary glands provide nutrition for newborn infants. In cultured murine HC11 mammary epithelial cells, RhoA is proposed as a parathyroid hormone–related protein (PTHrP) serotonylation target involved in newborn mineralization to release calcium during lactation [ 150 ] . High activation levels of the RhoA pathway also impair prolactin signaling in first-/second-passage primary murine mammary epithelial cells [ 151 ] . The proposed anti-lactogenic pathways of RhoA include RhoA/ROCK kinase inhibition of insulin signaling in primary cultured mid-pregnant murine mammary epithelial cells [ 152 ] and Rho downstream effector protein kinase N1 (PKN1) signaling impairing tight junction sealing via glucocorticoid blockage, as well as limiting milk secretion after parturition in a mouse model with constitutive activation of PKN1 [ 153 ] . The overexpression and uncontrolled behavior of RhoA in vitro may provide insights into a vital therapeutic pathway to ensure successful nutrient delivery to infants and establish and maintain sufficient lactogenesis. Breast cancer is one of the most common types of cancer among women, and metastasis to other organs and tissue systems represents one of the most prominent causes of cancer-related mortality. Breast cancer subtypes can be primarily categorized into four subvariants: luminal A-type, luminal B-type, HER-2 overexpressing, and basal-like, with categorization based on varying expression patterns of ERα, PR, and HER-2 [ 154 ] . Both upregulation and downregulation of RhoA appear to have implications in breast cancer. RhoA (along with Rac1 and Cdc42) expression was considerably higher in malignant human breast tissues than in normal breast tissues from the same patients; however, the mRNA and protein levels of RhoA appeared to be uncorrelated, and delayed RhoA degradation in malignant breast tissues could be the reason [ 155 ] . A similar phenomenon of increased RhoA expression in malignancy was observed in 4T1 (derived from spontaneous murine mammary tumors) and MCF10 (immortalized human mammary epithelium) cell lines. In these cell lines, miR-182 expression was positively correlated with metastasis. Missing in metastasis (MIM) is a target of miR-182 and a suppressor of RhoA. The inhibition of MIM by miR-182 leads to increased actin stress fiber formation, a hallmark of RhoA activity, and an increased metastatic phenotype [ 156 ] . Confirmatory examination with the RhoA inhibitor C3 exoenzyme showed an effect similar to that of phenotypic MIM expression, blocking cell invasion and cytoskeletal remodeling [ 156 ] . Other microRNAs, such as miR-146a [ 157 ] and miR-155 [ 158 ] , have also been reported to regulate RhoA expression/activity in breast cancer cell lines. In addition, the overexpression of RhoGDI, the protein responsible for the sequestration of inactive RhoA, results in a reversible increase in breast cancer resistance to etoposide and doxorubicin [ 158 ] . Indeed, RhoA activation promotes drug resistance in breast cancer cells [ 159 ] , and RhoA/ROCK signaling mediates the effects of doxorubicin on cell migration and invasion via increased myosin light chain 2 phosphorylation in MCF7 and BT-474 cells [ 160 ] . RhoA has also been shown to play a role in the cellular functions of disheveled 2 (Dvl2) and disheveled-associated activator of morphogenesis 1 (Daam1), which are downstream molecules of Wnt5a signaling [ 161 ] , and diaphanous-related formin-3 (DIAPH3), a cytoskeletal regulator and RhoA inhibitor, with decreased expression in luminal A-type, luminal B-type, HER-2 + , and triple-negative (no ERα, PR, or HER-2) clinical human breast cancer tissues [ 162 ] . Although activation of RhoA signaling is positively correlated with breast cancer formation and progression in most cases, RhoA knockdown in a syngeneic triple-negative breast cancer murine model had no impact on tumor formation or proliferation; however, this knockdown significantly increased the rate of invasion into lymph nodes and metastasis to the lungs [ 163 ] . The results of this study contrast with those observed in primary human tissues and both human and murine cell lines and may represent either an alternative pathway of RhoA activity in breast cancer or a species-specific difference revealed in vivo . These reports reveal the complexity of developing therapeutic strategies for treating not only breast cancers but all types of cancers in different patients, because of individual variations, such as variations in oncogenic origin, genetics, and environment. Nevertheless, RhoA signaling has cellular effects on cancer cells, and targeting RhoA signaling (inhibiting or activating) via microRNAs or chemotherapy remains a potential strategy for cancer therapy.

This study was supported by the following grants: NIH R03HD097384, R03HD100652, and R01HD114750 to X.Y.

All authors declare no conflicts of interest.

The authors thank the Office of the Vice President for Research, Interdisciplinary Toxicology Program, Department of Physiology and Pharmacology at the University of Georgia and the National Institutes of Health (NIH) for their financial support.

Not applicable.