Estrogen Dysregulation, Intraocular Pressure, and Glaucoma Risk

Abstract

Characterized by optic nerve atrophy due to retinal ganglion cell (RGC) death, glaucoma is the leading cause of irreversible blindness worldwide. Of the major risk factors for glaucoma (age, ocular hypertension, and genetics), only elevated intraocular pressure (IOP) is modifiable, which is largely regulated by aqueous humor outflow through the trabecular meshwork. Glucocorticoids such as dexamethasone have long been known to elevate IOP and lead to glaucoma. However, several recent studies have reported that steroid hormone estrogen levels inversely correlate with glaucoma risk, and that variants in estrogen signaling genes have been associated with glaucoma. As a result, estrogen dysregulation may contribute to glaucoma pathogenesis, and estrogen signaling may protect against glaucoma. The mechanism for estrogen-related protection against glaucoma is not completely understood but likely involves both regulation of IOP homeostasis and neuroprotection of RGCs. Based upon its known activities, estrogen signaling may promote IOP homeostasis by affecting extracellular matrix turnover, focal adhesion assembly, actin stress fiber formation, mechanosensation, and nitric oxide production. In addition, estrogen receptors in the RGCs may mediate neuroprotective functions. As a result, the estrogen signaling pathway may offer a therapeutic target for both IOP control and neuroprotection. This review examines the evidence for a relationship between estrogen and IOP and explores the possible mechanisms by which estrogen maintains IOP homeostasis.

1. Introduction

Glaucoma is the leading cause of irreversible blindness in the world, impacting approximately 80 million individuals (Budenz et al., 2013; Doshi et al., 2008; Lee and Mackey, 2022; Paterson and Miller, 1963; Rudnicka et al., 2006; Tham et al.; Ziai et al., 1994). Approximately three-fourths of glaucoma cases are diagnosed as primary open-angle glaucoma (POAG) (Liu and Allingham, 2017). Although POAG is characterized by features shared by all forms of glaucoma, including retinal ganglion cell (RGC) death, optic nerve (ON) degeneration, and progressive visual field loss, it is unique in that it lacks a distinguishable cause (Liu and Allingham, 2017). While the cause is not well-defined, several risk factors for POAG have been identified, including genetic factors, aging, positive family history, African ancestry, and the only modifiable risk factor – elevated intraocular pressure (IOP) (Liu and Allingham, 2017).

IOP is defined as the pressure exerted on the wall of the eye by the aqueous humor (AH), a fluid inside the anterior segment of the eye that supplies nutrients to and removes wastes from the avascular cornea, lens, and trabecular meshwork (Figure 1A) (Liu and Allingham, 2017; Stamer, 2012). IOP is maintained within a narrow range over a lifetime by the balance of AH production and outflow (Liu and Allingham, 2017; Stamer, 2012). AH outflow may occur either through the conventional or unconventional outflow pathway (Stamer, 2012), with approximately 90% in older eyes occurring through the conventional outflow pathway comprised of the trabecular meshwork (TM), Schlemm’s canal (SC), and eventually the episcleral vein (Stamer, 2012).

Figure 1. Aqueous humor production and outflow.

(A) After being produced by the ciliary body, aqueous humor flows between the lens and iris into the anterior segment where it nourishes the lens, iris, and cornea. The fluid then leaves the eye either through the conventional trabecular meshwork pathway or through the unconventional uveoscleral pathway. Figure adopted from (Ito and Walter, 2013) according to the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/3.0/). (B) Aqueous humor continues beyond Schlemm’s canal into collector channels, aqueous veins, and the episcleral vein. Figure adopted from (Lee et al., 2019) according to the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by-nc-nd/4.0/).

Most current therapies target AH production by the ciliary body (CB) or AH outflow through the unconventional outflow pathway, leading to only modest results (Stamer, 2012; Tanna and Johnson, 2018). Furthermore, many of these therapies lead to deleterious side effects (Stamer, 2012; Tanna and Johnson, 2018). Therapies targeting the conventional outflow pathway offer better efficacy since it is the presumed site of dysfunction in IOP regulation (Stamer, 2012; Tanna and Johnson, 2018). Intense investigation has been devoted to developing novel therapies targeting the TM, such as netarsudil (Rhopressa^®), a Rho kinase inhibitor that lessens glaucoma-related TM fibrosis (Stamer, 2012; Sturdivant et al., 2016). However, due to common side effects, netarsudil remains a third line medication for glaucoma. Thus, novel therapeutic targets in the TM and other outflow tissues are still required. The current focus has been directed toward identifying genes and pathways complicit in TM dysfunction, IOP dysregulation, and POAG pathogenesis.

Interestingly, dysregulated sex hormone levels, especially decreased estrogen levels, have gained increasing support as a possible glaucoma risk factor. In general, estrogen levels have been inversely related to IOP levels and glaucoma risk. For example, the onset of menopause, a low estrogen state, increases IOP and glaucoma risk (Hulsman et al., 2001; Patel et al., 2018; Qureshi, 1995; Vajaranant and Pasquale, 2012). An overall shorter fertility duration marked by either late menarche or early menopause/ovariectomy also increases glaucoma risk (Harris et al., 2000; Hulsman et al., 2001; Vajaranant et al., 2014). In contrast, the use of postmenopausal hormones (PMH) among menopausal women decreases their IOP and glaucoma risk as compared to those without PMH (Altintas et al., 2004; Lang et al., 2002; Patel et al., 2018; Sator et al., 1998; Sator et al., 1997; Sorrentino et al., 1998; Uncu et al., 2006; Wojtowiec et al., 2016). Furthermore, IOP is lower during hyper-estrogenic phases of the menstrual cycle and pregnancy (Bahadir Kilavuzoglu et al., 2018; Green et al., 1988; Phillips and Gore, 1985; Qureshi, 1995; Qureshi et al., 1996; Weinreb et al., 1988; Yucel et al., 2005; Ziai et al., 1994). Beyond the epidemiological studies, genetic studies identified a significant association of a set of single nucleotide polymorphisms (SNPs) in a panel of 23 estrogen metabolic and signaling pathway genes (i.e., HSD11B1, HSD3B1, CYP1B1, SULT1E1, UGT2B11, SRD5A1, ESR1, AKR1D1, CYP3A4, CYP11B1, HSD17B3, AKR1C4, CYP17A1, ESR2, CYP11A1, CYP19A1, CYP1A1, CYP1A2, HSD17B1, SULT2B1, COMT, ARSD, and STS) with POAG and high-tension glaucoma (Pasquale et al., 2013). In addition, the loss of aromatase, the enzyme responsible for converting androgens to E2, causes elevated IOP and RGC loss in female homozygous mice by 12 weeks of age (Chen et al., 2018). Importantly, estrogen receptors and metabolizing enzymes are known to be expressed in multiple ocular tissues where estrogen signaling has been shown to exert protective effects (Deschênes et al., 2010; Lang et al., 2002; Munaut et al., 2001; Ogueta et al., 1999; Patel et al., 2018; Rocha et al., 2000; Suzuki et al., 2001). For example, estrogen appears to play a neuroprotective role in the retina (Cascio et al., 2015; Douglass et al., 2023; Fotesko et al., 2022; Levi and Brimble, 2004; Nuzzi et al., 2018, 2019). Together these lines of evidence strongly support the potential role of estrogen signaling in IOP regulation and protection against POAG risk. Thus, the estrogen signaling pathway offers a promising therapeutic target for POAG. This review will focus on the impact of the estrogen signaling pathway on IOP regulation only. For the retinal or neuroprotective effects of estrogen signaling, please refer to numerous reviews describing the effects of estrogen on the retina (Cascio et al., 2015; Douglass et al., 2023; Fotesko et al., 2022; Levi and Brimble, 2004; Nuzzi et al., 2018, 2019).

2. Glaucoma and Aqueous Humor Outflow

2.1. Trabecular Meshwork

In order to understand how estrogen signaling may affect glaucoma etiology, we must first examine the anatomic structure of the AH outflow pathway, which encapsulates the TM, SC, and the distal vasculature. The TM is traditionally divided into three layers from interior to exterior: uveal, corneoscleral, and juxtacanalicular (JCT) (American Academy of, 2019; Stamer and Clark, 2017; Tamm and Fuchshofer, 2007). The outermost portion of the JCT neighbors the basement membrane of the inner wall of SC (American Academy of, 2019; Stamer and Clark, 2017). This region of the JCT, the inner wall of SC, and the basement membrane between them contribute to a large majority of outflow resistance (American Academy of, 2019; Carreon et al., 2017; Overby et al., 2009; Stamer, 2012; Stamer and Acott, 2012; Stamer and Clark, 2017).

As the first tissue that the AH must pass in the conventional outflow tract, the TM is exposed to constant levels of stress, including metabolic and oxidative stress and mechanical stretch (Acott et al., 2014; Acott et al., 2021; Hirt and Liton, 2017; Stamer and Clark, 2017). The systemic pulse and movements, such as blinking, transiently affect AH outflow and TM stretch (Coleman and Trokel, 1969; Hirt and Liton, 2017; Johnstone, 2004; Stamer and Clark, 2017). Furthermore, as IOP increases, the amount of mechanical stretch experienced by TM cells increases (Acott et al., 2014; Acott et al., 2021; Hirt and Liton, 2017). In normal eyes, TM cells experience 10-20% stretch while glaucomatous TM could experience as high as 50% stretch (Acott et al., 2014; Grierson and Lee, 1975a, b).

Glaucomatous TM is characterized by reduced TM cellularity, increased ECM accumulation, increased contractility, and an overall increased stiffness (Acott et al., 2014; Acott et al., 2021; Lütjen-Drecoll, 2005; Stamer, 2012; Stamer and Clark, 2017; Zhang et al., 2007). Ocular hypertension and glaucomatous changes in the TM can be induced by pro-fibrotic transforming growth factor β2 (TGFβ2) (Fleenor et al., 2006; Gottanka et al., 2004; Stamer and Clark, 2017). In addition to TGFβ2, corticosteroids frequently used to lower systemic or ocular inflammation are known to induce elevated IOP after extended use. The use of corticosteroids like dexamethasone is a well-recognized cause of glaucoma. Steroid-induced glaucoma is recognized as a subtype of secondary open-angle glaucoma, distinct from primary open-angle glaucoma in that its cause is known (Hall, 2000; Jonas et al., 2017; King et al., 2013; Liu and Allingham, 2017; Weinreb et al., 2014). The effects of dexamethasone on TM cells have been found to share similar fibrotic characteristics to that of TGFβ2-treated TM cells (Clark et al., 1995; Stamer and Clark, 2017; Steely et al., 1992). Interestingly, corticosteroids like dexamethasone are a part of the steroid class of chemical compounds that share a common four-ring hydrocarbon structure. These molecules include several different steroid types, including mineralocorticoids, glucocorticoids, and sterol sex hormones, including testosterone, progesterone, and estrogen.

2.2. Schlemm’s Canal

Schlemm’s canal (SC) is a continuous vessel encircling the anterior segment distal to the TM. This vessel has both vascular and lymphatic characteristics (Aspelund et al., 2014; Carreon et al., 2017; Kizhatil et al., 2014; Park et al., 2014; Ramos et al., 2007). The inner wall of SC is comprised of a single layer of endothelial cells adjacent to the JCT (Overby et al., 2009; Swain et al., 2021). The pressure within the anterior segment is greater than the venous pressure inside of SC’s lumen, creating a gradient across the SC inner wall, which promotes the formation of invaginations in the inner wall (Overby et al., 2009; Stamer and Acott, 2012; Stamer et al., 2015; Swain et al., 2021). These invaginations are termed giant vacuoles and are sites of active fluid flow across the inner wall of SC. If giant vacuoles get too large, a transcellular pore or “relief valve” forms (Carreon et al., 2017; Overby et al., 2009; Stamer and Acott, 2012; Stamer et al., 2015; Swain et al., 2021). Importantly, pores in the inner wall have been shown to exist in lower numbers in glaucomatous outflow tissue (Allingham et al., 1992; Carreon et al., 2017; Johnson et al., 2002; Overby et al., 2009; Stamer and Acott, 2012; Stamer et al., 2015; Swain et al., 2021).

While the inner wall of the SC is characterized by pore formation, the outer wall contains openings for aqueous humor egress into the collector channels (Carreon et al., 2017; Overby et al., 2009). If pressure changes intraocularly are too high, the SC lumen will narrow and may even collapse (Johnstone and Grant, 1973; Overby et al., 2009). When high pressures cause the space between the inner and outer wall to diminish, the resulting shear stress on SC endothelium leads to nitric oxide (NO) production, which causes relaxation of the TM and improved outflow facility (Aliancy et al., 2017; Ashpole et al., 2014; Dismuke et al., 2014; Dismuke et al., 2008; McDonnell et al., 2018; McDonnell et al., 2020; Reina-Torres et al., 2021; Stamer et al., 2011).

2.3. Distal Vessels

Although the TM and SC have traditionally been the primary focus of studies on outflow dysfunction, there has been a growing interest in the vessels distal to SC in recent years. After AH enters SC, it follows a pressure gradient out of approximately 30 collector channels that exit from the outer wall of SC (Carreon et al., 2017; Overby et al., 2009). From there, the fluid continues to follow the pressure gradient into the deep scleral plexus, intrascleral collector channels, aqueous veins, and finally the episcleral vein (Figure 1B) (Carreon et al., 2017; Overby et al., 2009). As these vessels are lined with vascular endothelial cells, they respond to NO signaling (Reina-Torres et al., 2021). The amount of outflow resistance provided by these vessels and the effects of NO signaling on this resistance have been the subject of intense inquiry.

3. Estrogen Signaling

Estrogen is one of five types of steroid hormones, including glucocorticoids, mineralocorticoids, androgens, and progestogens. Estrogen may be found in four different isoforms (i.e., estradiol, estrone, estriol, and estretrol), with 17β-estradiol (E2) being the primary active form (Fuentes and Silveyra, 2019). These various isoforms are ultimately derived from cholesterol and are produced predominantly by granulosa cells found in ovarian follicles (Fuentes and Silveyra, 2019). However, lower levels of estrogen may be produced by other organs in both sexes, including adipose tissue, brain, breast, bone, adrenal glands, and liver, as well as testes in males (Barakat et al., 2016; Fuentes and Silveyra, 2019; Simpson et al., 2002). Whether produced by the ovaries or other tissues, estrogen synthesis requires the action of aromatase, making this enzyme a prime target for inhibition of estrogen signaling (Fuentes and Silveyra, 2019; Simpson et al., 2002). While aromatase and several other enzymes are responsible for estrogen production, estrogen isoforms also are metabolized by several cytochrome P450 enzymes, including CYP1B1, mutations for which are responsible for some forms of primary congenital glaucoma and juvenile glaucoma (Bejjani et al., 1998; Fuentes and Silveyra, 2019; Libby et al., 2003; Stoilov et al., 1998; Stoilov et al., 1997; Vasiliou and Gonzalez, 2008).

E2 may bind to three different estrogen receptors, estrogen receptor 1 or α (ESR1 or ERα), estrogen receptor 2 or β (ESR2 or ERβ), and G protein-coupled estrogen receptor 1 (GPER1 or GPR30), each of which mediates estrogen signaling through separate mechanisms (Figure 2) (Fuentes and Silveyra, 2019). ERα and ERβ primarily reside in the cytoplasm while GPR30 is associated with the plasma membrane (Björnström and Sjöberg, 2005; Fuentes and Silveyra, 2019). Dimers of ERα and/or ERβ act predominantly through a canonical genomic pathway whereby they translocate from the cytoplasm to the nucleus after binding E2 (Björnström and Sjöberg, 2005; Fuentes and Silveyra, 2019; Klinge, 2001; Lara-Castillo, 2021; Marino et al., 2006; O’Malley, 2005). Once in the nucleus, activated ERα and ERβ dimers bind to enhancer regions called estrogen response elements (EREs) to initiate transcription of downstream target genes (Björnström and Sjöberg, 2005; Fuentes and Silveyra, 2019; Lara-Castillo, 2021; Nilsson et al., 2001). Alternatively, dimers of ERα and/or ERβ may act through an indirect genomic pathway by recruiting transcription factors to the promoters of target genes (Aranda and Pascual, 2001; Björnström and Sjöberg, 2005; Fuentes and Silveyra, 2019; Göttlicher et al., 1998; Lara-Castillo, 2021). In this case, the estrogen receptors do not directly interact with genomic DNA sequences (Aranda and Pascual, 2001; Björnström and Sjöberg, 2005; Fuentes and Silveyra, 2019; Göttlicher et al., 1998; Lara-Castillo, 2021).

Figure 2. Estrogen signaling pathways.

Estrogen signaling may occur through genomic or non-genomic pathways resulting from the interaction of estradiol with one of the three estrogen receptors (i.e., ERα/ESR1, ERβ/ESR2, or GPR30/GPER1). These estrogen receptors may impact transcription by (1) direct interactions with estrogen response elements (EREs), (2) transcription factor (TF) recruitment, or (3) phosphorylation cascades. Figure adapted from (Lara-Castillo, 2021) according to the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/).

Besides direct or indirect genomic signaling, estrogen signaling may occur in a non-genomic manner. In this pathway, GPR30 located on the cellular membrane, or to a lesser extent membrane-associated ERα and ERβ, binds with E2 and activates G protein complex-induced kinase cascades, including the cAMP/PKA, PI3K/Akt, PLC/PKC, and Ras/Raf/MAPK pathways (Björnström and Sjöberg, 2005; Fuentes and Silveyra, 2019; Lara-Castillo, 2021). In turn, these pathways may cause a variety of cellular effects, including changes in transcription (Fuentes and Silveyra, 2019). The genomic and non-genomic pathways may overlap and engage in crosstalk (Björnström and Sjöberg, 2005; Fuentes and Silveyra, 2019). Furthermore, signaling through the different receptors may result in opposing effects (Prossnitz and Barton, 2014).

Further complicating estrogen signaling, ESR1 and ESR2 genes may produce shortened isoforms of these estrogen receptors (Fuentes and Silveyra, 2019). These truncated receptors lack the normal transcriptional activity of full-length receptors, have an inhibitory effect on ERα, and/or are able to induce alternate signaling events (Fuentes and Silveyra, 2019). In addition to the complicated effects of alternate isoforms, full-length estrogen receptors are known to bind ligands other than E2, including endogenous (estrone, estriol, and estretrol) and non-endogenous ligands (metalloestrogens, phytoestrogens, xenoestrogens, and selective estrogen receptor modulators (SERMs)) (Fuentes and Silveyra, 2019). These ligands may agonize or antagonize estrogen signaling (Fuentes and Silveyra, 2019). In particular, SERMs, such as tamoxifen, may act as either an agonist or an antagonist in a tissue-specific manner and their direction of effect may also depend upon which receptor they bind (Fuentes and Silveyra, 2019; Shang and Brown, 2002; Sharma and Prossnitz, 2017). Several selective agonists and antagonists have been developed for each of the three estrogen receptors (Table 1). Selective ERα agonists include PPT, although PPT has also been shown to agonize GPR30 to a lesser degree (Petrie et al., 2013; Prossnitz and Barton, 2014; Prossnitz and Hathaway, 2015; Stauffer et al., 2000). Selective ERβ agonists include DPN (Meyers et al., 2001; Prossnitz and Barton, 2014; Prossnitz and Hathaway, 2015). G1, GPER-L1, and GPER-L2 serve as selective GPR30 agonists, while G15, G36, CIMBA, PBX1, PBX2, C4PY serve as selective GPER1 antagonists, with G15 causing slight agonism of ERα and ERβ (Bologa et al., 2006; DeLeon et al., 2020; Dennis et al., 2009; Dennis et al., 2011; Lappano et al., 2015; Maggiolini et al., 2015; Prossnitz and Barton, 2014; Prossnitz and Hathaway, 2015; Rouhimoghadam et al., 2020; Sharma and Prossnitz, 2017; Singh et al., 2022).

Table 1.

Steroid receptors and their agonists and antagonists.

Receptor	Specific Agonists	Specific Antagonists
Estrogen Receptor 1 (Erα; ESR1)	PPT (4,4’,4”-(4-Propyl-[1H]-pyrazole-1,3,5-triyl)trisphenol)	MPP (Methyl-piperidino-pyrazole)
Estrogen Receptor 2 (ERβ; ESR2)	DPN (Diarylpropionitrile)	PHTPP (4-[2-phenyl-5,7-bis(trifluoromethyl)pyrazolo[1,5-a]-pyrimidin-3-yl]phenol)
G Protein-Coupled Estrogen Receptor 1 (GPR30; GPER1)	G1 ((±)-1-[(3aR*,4S*,9bS*)-4-(6-Bromo-1,3-benzodioxol-5-yl)-3a,4,5, 9b-tetrahydro-3H-cyclopenta[c]quinolin-8-yl]-ethanone) GPER-L1 (7-({[2-(diethylamino)ethyl] amino}methyl)-5imino-1,3,6-triphenyl-5,6-dihydropyrimido[4,5-d]pyrimidine-2,4(1H,3H)-dithione) GPER-L2 (1-[bis(phenylthio)methyl] imidazolidine-2thione)	G15 ((3aS*,4R*,9bR*)-4-(6-Bromo-1,3-benzodioxol-5-yl)-3a,4,5,9b-3H-cyclopenta[c]quinoline) G36 ((±)-(3aR*,4S*,9bS*)-4-(6-Bromo-1,3-benzodioxol-5-yl)-3a,4,5,9b-tetrahydro-8-(1-methylethyl)-3H-cyclopenta[c]quinoline) CIMBA (2-cyclohexyl-4-isopropyl-N-(4-methoxybenzyl) aniline) PBX1 (pyrrolobenzoxazinone 1) PBX2 (pyrrolobenzoxazinone 2) C4PY (Meso-octamethylcalix-[4]-pyrrole)

Estrogen signaling is involved in a diverse array of physiological processes with its role most well-known in female sexual development and reproduction (Fuentes and Silveyra, 2019). Estrogen signaling is also involved in the brain, bone, and cardiovascular system, with menopause-associated estrogen loss contributing to neurodegenerative disorders, osteoporosis, and cardiovascular disease (Davis and Baber, 2022; Fuentes and Silveyra, 2019; Lobo and Gompel, 2022; Maioli et al., 2021; Yang et al., 2022). On a molecular level, estrogen signaling is involved in a wide variety of cellular events, including nitric oxide (NO) production (Kim et al., 2008; Kim et al., 2014; Kypreos et al., 2014; Mantione, 2008; Ndzie Noah et al., 2021; Teoh et al., 2020), ECM turnover (Lephart and Naftolin, 2021; Piperigkou and Karamanos, 2020; Schuler and Murdoch, 2021), modulation of the actin cytoskeleton (Babayan and Kramár, 2013; Giretti and Simoncini, 2008; Kramár et al., 2013; Sanchez et al., 2012; Sanchez and Simoncini, 2010), mechanotransduction (Imai et al., 2008; Wang et al., 2023), epithelial-to-mesenchymal transition (Di Zazzo et al., 2019; Gelissen and Huang, 2022; Jeon et al., 2016; Piperigkou and Karamanos, 2020), and mitochondrial metabolism (Guajardo-Correa et al., 2022; Pellegrino et al., 2022; Vasconsuelo et al., 2013). More recently, as discussed earlier in this review, low levels of estrogen signaling have been correlated with elevated IOP and higher glaucoma risk.

In order for estrogen signaling to have an effect on IOP regulation, various isoforms of estrogen and its receptor(s) (i.e., ERα, ERβ, and GPR30) need to be present in the outflow pathway. The presence of E2 has been identified in aqueous and vitreous humor (Iqbal et al., 1997; Rock et al., 1993; Youngblood et al., 2021; Zhang et al., 2003). The other estrogen isoforms, estrone and 17α-estradiol, have also been measured in these tissues with estrone being present in the retina as well (Iqbal et al., 1997; Stárka et al., 1976). Also, estrogen receptors and steroid metabolizing enzymes are expressed throughout the eye, including the retina, ciliary body, and cornea (Agapova et al., 2006; Colitz et al., 2015; Deschênes et al., 2010; Gelinas and Callard, 1993; Hayashi et al., 2007; Jiang et al., 2019; Kobayashi et al., 2004; Kobayashi et al., 1998; Kumar et al., 2005; Lang et al., 2002; Li et al., 2021; Miyamoto et al., 1999; Munaut et al., 2001; Ofri et al., 1999; Ogueta et al., 1999; Patel et al., 2018; Qureshi et al., 1997; Rocha et al., 2000; Southren et al., 1976; Suzuki et al., 2001; Wickham et al., 2000). Our recent studies confirmed the expression of ERα in human outflow tissue; immunohistochemistry indicated that the receptor was expressed in both the TM and SC (Figure 3) (Youngblood et al., 2021).

Figure 3. Expression of ERα in the TM/SC region of human outflow tissue.

ERα protein was expressed in the trabecular meshwork (TM)/Schlemm’s canal (SC) region of human outflow tissue (n=4, 24-60 years of age) taken from both male (A-C) and female (D-F) post-mortem donors. Representative images of H&E (A, D), negative control (B, E), and positive staining (C, F). ERα, estrogen receptor α; TM, trabecular meshwork; SC, Schlemm’s canal. Scale bar: 50 μm. Figure adopted from (Youngblood et al., 2021) according to the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/).

4. Estrogen and IOP

While IOP and family history/genetics remain the primary risk factors for POAG, dysregulated sex hormone levels, especially decreased estrogen levels, have been gaining increasing support as a possible glaucoma risk factor. There are several lines of support from clinical, epidemiological, and genetic studies as well as a limited number of in vivo and in vitro studies (Table 2). While some studies focus on the neuroprotective effects of estrogen and its active form estradiol (E2) in the retina, this review will focus on studies examining the role of estrogen in IOP regulation.

Table 2.

Several lines of evidence support a role for estrogen signaling in IOP regulation and modulation of glaucoma risk.

Finding	Literature
Males have a higher glaucoma risk than females until females enter menopause.	(Bankes et al., 1968; Budenz et al., 2013; Doshi et al., 2008; Hollows and Graham, 1966; Lee and Mackey, 2022; Paterson and Miller, 1963; Rudnicka et al., 2006; Tham et al., 2014)
Menopause, especially early menopause, increases IOP and glaucoma risk.	(Bankes et al., 1968; Hulsman et al., 2001; Panchami et al., 2013; Patel et al., 2018; Qureshi, 1995; Vajaranant et al., 2014)
Short fertility duration increases glaucoma risk.	(Dewundara et al., 2016; Harris et al., 2000; Hulsman et al., 2001; Lee and Mackey, 2022; Vajaranant et al., 2014)
Hormone therapy decreases IOP and glaucoma risk.	(Al-Lozi et al., 2023; Altintas et al., 2004; Lang et al., 2002; Newman-Casey et al., 2014; Patel et al., 2018; Sator et al., 1998; Sator et al., 1997; Sorrentino et al., 1998; Treister and Mannor, 1970; Uncu et al., 2006; Wojtowiec et al., 2016)
IOP decreases during hyper-estrogenic phases of pregnancy.	(Bahadir Kilavuzoglu et al., 2018; Green et al., 1988; Paterson and Miller, 1963; Phillips and Gore, 1985; Qureshi, 1995; Qureshi et al., 1996)
IOP and outflow facility changes during the menstrual cycle.	(Akar et al., 2004; Dalton, 1967; Paterson and Miller, 1963; Yucel et al., 2005)
Polymorphisms in the estrogen pathway are associated with high IOP and glaucoma risk.	(de Voogd et al., 2008; Jansson et al., 2001; Kosior-Jarecka et al., 2019; Liu and Allingham, 2017; Mabuchi et al., 2010; Pasquale et al., 2013)
Variants in genes known to interact with estrogen signaling have female-biased association with POAG.	(Chen et al., 1999; Kang et al., 2010; Loomis et al., 2014; Razandi et al., 2002; Russell et al., 2000; Wiggs et al., 2011)
Estrogen depleted animal models have evidenced elevated IOP, reduced outflow facility, RGC loss, and reduced visual function.	(Chen et al., 2018; Feola et al., 2019; Feola et al., 2020; Zhou et al., 2007)

4.1. Epidemiological and Clinical Studies

Most of the support for a role of estrogen signaling in glaucoma comes from epidemiological and clinical studies. First, these studies have identified that men have a higher risk of developing glaucoma than women (Budenz et al., 2013; Doshi et al., 2008; Lee and Mackey, 2022; Rudnicka et al., 2006; Tham et al., 2014). For example, Rudnicka et al. showed in a multi-ethnic cohort that men had a 1.37 odds ratio of developing POAG compared to women (Rudnicka et al., 2006; Tham et al., 2014). The Los Angeles Latino Eye Study and the Tema Eye Survey in a West African population published similar findings, suggesting this trend exists in different ethnic populations, an important finding considering ethnicity-related disparity in glaucoma risk (Budenz et al., 2013; Doshi et al., 2008). This sex difference appears to be age-affected with risk in men being higher before age 50, around the time of menopause in women (Paterson and Miller, 1963). Along these same lines, Bankes et al. found a trending lower IOP in females under 40 compared to males under 40, but a trending higher IOP in females over 60 compared to males over 60, suggesting that loss of estrogen signaling after menopause reverses the estrogenic protection in females before menopause (Bankes et al., 1968). Hollows et al. reported a similar finding with females over 50 showing a trend of higher IOP compared to age-matched males (Hollows and Graham, 1966).

Supporting these age-related gender disparities for glaucoma and IOP, a second line of evidence is that menopause, especially early menopause increases glaucoma risk and IOP (Bankes et al., 1968; Hulsman et al., 2001; Panchami et al., 2013; Patel et al., 2018; Qureshi, 1995; Vajaranant and Pasquale, 2012). Similar to menopause, bilateral ovariectomy (i.e., removal of both ovaries) also increases IOP and glaucoma risk (Vajaranant et al., 2014). Aside from effects on IOP, menopause also affects ocular blood flow with statistical differences between male and female ocular blood flow disappearing after menopause (Centofanti et al., 2000; Harris et al., 2000).

Third, late menarche is also correlated with increased POAG risk and IOP (Lee and Mackey, 2022). The reasoning for this is that a late menarche, like an early menopause, reduces lifetime exposure to estrogen (Dewundara et al., 2016; Harris et al., 2000; Hulsman et al., 2001; Lee and Mackey, 2022; Vajaranant et al., 2014). The subsequent chronic estrogen deprivation is considered to have more significant effects than the hormonal transitions of menarche and menopause alone (Harris et al., 2000; Hulsman et al., 2001; Vajaranant et al., 2014).

A fourth line of evidence suggests that the use of exogenous estrogens may decrease IOP. Individuals who use postmenopausal hormones (PMH) have lower IOP and less risk of angle closure than women who do not use PMH (Altintas et al., 2004; Lang et al., 2002; Newman-Casey et al., 2014; Patel et al., 2018; Sator et al., 1998; Sator et al., 1997; Sorrentino et al., 1998; Treister and Mannor, 1970; Uncu et al., 2006; Wojtowiec et al., 2016). In fact, Sator et al. observed a 1-5 mmHg decrease in IOP after 12 weeks of PMH use (Sator et al., 1998; Sator et al., 1997). Furthermore, Newman-Casey, et al. found a 0.4% decrease in risk of developing POAG for each month of using estrogen-only PMH (Newman-Casey et al., 2014). The literature suggests that PMH use has beneficial effects in other ocular tissues and processes (e.g., increased tear production, decreased cataract formation, and improved ocular blood flow) (Lang et al., 2002; Sorrentino et al., 1998).

A fifth support for the role of estrogen signaling in IOP regulation is that IOP decreases while outflow facility increases when circulating estrogen levels are high during pregnancy (Bahadir Kilavuzoglu et al., 2018; Green et al., 1988; Paterson and Miller, 1963; Phillips and Gore, 1985; Qureshi, 1995; Qureshi et al., 1996). Centofanti et al. found a 1 mmHg decrease in IOP and an increase in blood flow to the eye between the first and second trimesters of pregnancy (Centofanti et al., 2002). Kilavuzoglu et al. reported similar findings with IOP levels being negatively correlated with blood estrogen and progesterone levels (Bahadir Kilavuzoglu et al., 2018). Other hormones, including progesterone and relaxin, may be involved in these changes during pregnancy (Paterson and Miller, 1963).

Lastly, changes in IOP levels have been noted during the menstrual cycle with lower IOP being observed in the hyperestrogenic luteal phase and higher IOP around the time of menses (Dalton, 1967; Yucel et al., 2005). However, increases in optic nerve cupping also occur during the hyperestrogenic luteal phase, complicating our understanding of how estrogen levels affect the eye (Akar et al., 2004; Yucel et al., 2005). The changes in IOP levels may correspond to changes in outflow facility, which decreases at the beginning and middle of the menstrual cycle, points where estrogen levels are also decreasing (Akar et al., 2004; Paterson and Miller, 1963). In addition, IOP levels have been shown to be higher in women with polycystic ovarian syndrome, which is characterized by hormone dysfunction that often leads to irregular menstrual cycles (Balıkçı et al., 2022).

Despite this collection of evidence supporting a potential role of estrogen signaling in IOP regulation, it is important to note that other studies failed to suggest this role (Abramov et al., 2005; Costanian et al., 2020; Ebeigbe and Ebeigbe, 2013; Ersoz et al., 2017; Feldman et al., 1978; Gierek et al., 1976; Guaschino et al., 2003; Kang et al., 2018; Kang et al., 2010; Lamble and Lamble, 1978; Na et al., 2014; Nirmalan et al., 2004; Ofri et al., 1999; Paganini-Hill and Clark, 2000; Qureshi et al., 1997; Rudnicka et al., 2006; Toker et al., 2003). These different observations may result from varying assessment criteria and sample sizes (Rudnicka et al., 2006). Although these discrepancies need to be resolved, a plethora of approaches have identified a potentially important role for estrogen signaling in reducing IOP and glaucoma risk as outlined in Table 2.

4.2. Genetic Studies

Genetic studies offer additional support for the role of estrogen signaling in regulating IOP and POAG risk. First, a panel of SNPs in/near estrogen metabolic genes have been significantly associated in female, but not male, POAG patients (Pasquale et al., 2013). For example, sequence variants within or near the ESR1 and ESR2 genes (i.e., genes that encode for ERα and ERβ) have been associated with clinical characteristics, including IOP, visual acuity, visual field, and other ocular parameters altered in glaucoma (de Voogd et al., 2008; Kosior-Jarecka et al., 2019; Mabuchi et al., 2010). Moreover, mutations in CYP1B1, the cytochrome P450 enzyme that metabolizes E2 among other steroid and non-steroid molecules, have been shown to cause juvenile open-angle glaucoma (JOAG) and primary congenital glaucoma (PCG) (Jansson et al., 2001; Liu and Allingham, 2017). CYP1B1 mutations identified in two PCG families showed lowered activity of this enzyme, suggesting dysregulation of this pathway (Jansson et al., 2001). In addition, β-estradiol was identified as the top key upstream regulator of differentially correlated genes from a panel of genes that were differentially expressed in control/POAG TM tissue and/or associated with IOP (Youngblood et al., 2021). ESR1 mediated β-estradiol interactions with 25 genes within the panel, including 12 genes associated with IOP, several of which were also regulated by dexamethasone (Youngblood et al., 2021).

Genetic studies have sought to examine sex-biased association by analyzing the influence of sex chromosomes. Interestingly, a discovery genome-wide association study (GWAS) identified three loci on the X-chromosome that were associated with IOPcc (corneal compensated IOP, i.e., an IOP measurement that accounts for the effects of corneal hysteresis on IOP) (Simcoe et al., 2020). Two of these loci had significant or nominal associations with POAG (Simcoe et al., 2020).

In addition, three genes with a sex-biased association for POAG have been identified in autosomes. First, sequence variants near or within the CAV1/CAV2 locus have been associated with POAG and IOP in multiple studies (Khawaja et al., 2018; Liu and Allingham, 2017; MacGregor et al., 2018). Interestingly, these variants are associated with POAG predominantly in women (Loomis et al., 2014; Wiggs et al., 2011). Secondly, many NOS3 variants are associated with POAG only in women (Kang et al., 2011). Thirdly, the POAG association with SNPs in guanylate cyclase genes GUCY1A3/GUCY1B3 was stronger in women (Buys et al., 2013). CAV1/CAV2 and NOS3 have been shown to interact with ERα, with CAV1 translocating ERα to the membrane and ERα stimulating NOS3 production of nitric oxide (NO) through a non-genomic mechanism (Chen et al., 1999; Razandi et al., 2002; Russell et al., 2000). Meanwhile, soluble guanylate cyclase is the intracellular receptor for NO (Buys et al., 2013). The sex-specific association of these genes provides additional support for the role of estrogen signaling in IOP regulation and POAG pathogenesis.

4.3. Animal- and Cell-based Studies

Although many studies have examined the neuroprotective effects of E2 (Nakazawa et al., 2006; Russo et al., 2008; Simpkins et al., 2005), the number of studies examining the effects of estrogen on IOP regulation in vivo or in vitro is limited. However, a handful of animal studies have offered pivotal findings. Of note, Chen et al. observed a significant increase in IOP and a concomitant decrease in RGC number in female mice lacking aromatase, the enzyme required for E2 production (Figure 4A–B (Chen et al., 2018). Likewise, ovariectomy-induced menopause in female rats has been shown to reduce outflow facility by as much as 34% (Figure 4C) (Feola et al., 2020). In an ocular hypertension (OHT) model, ovariectomy leads to reduced spatial frequency, a measure of visual acuity assessed by the optomotor response (Feola et al., 2019). In addition, ovariectomy appeared to exacerbate the increase in IOP observed in DBA/2J mice whereas systemic E2 significantly reduced IOP in the ovariectomized mice (Zhou et al., 2007). Elevated IOP in ovariectomized DBA/2J mice was accompanied by RGC loss while lower IOP in E2-treated, ovariectomized mice was accompanied by abrogated RGC loss compared to ovariectomized animals that did not receive E2 treatment (Zhou et al., 2007).

Figure 4. Effects of loss of estrogen on outflow facility and intraocular pressure.

(A-B) Eliminating estrogen production through aromatase knockout significantly elevated IOP in 12- and 24-week-old female knockout mice. (C) Ablating estrogen production by ovariectomy significantly reduced outflow facility in female rats. A-B adopted from (Chen et al.) according to the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/). C adopted from (Feola et al.) according to the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by-nc-nd/4.0/).

Although ovariectomy exacerbated IOP elevation in the DBA/2J genetic glaucoma model, ovariectomy did not exacerbate IOP in an ocular hypertension mouse model, in which a hypertonic saline injection into the episcleral vein is used to sclerose the AH outflow pathway (Feola et al., 2019). The different findings between these studies likely arise from variations in the models they use and may point to the location of estrogen’s effect. While the DBA/2J mice mimic pigment dispersion syndrome, in which pigment particles from the iris accumulate in the trabecular meshwork outflow pathway, injecting hypertonic saline into the episcleral vein causes sclerosis of the outflow pathway, leading to more dramatic effects (Fernandes et al., 2015; Morrison et al., 2015; Pang and Clark, 2020). As a result, the hypertonic saline model saw a large elevation of IOP (i.e., into the 40-mmHg range) within weeks while the elevation in the DBA/2J mice was much more modest (i.e., ≤20 mmHg) (Feola et al., 2019; Zhou et al., 2007). Because IOP in the hypertonic saline model was already drastically elevated, the smaller effects of ovariectomy would not be as noticeable as they would be in the DBA/2J mice, which had only moderately elevated IOP. Furthermore, by sclerosing the outflow pathway, the hypertonic saline model may eliminate the outflow tissues (i.e., the TM and SC) on which estrogen signaling loss would have had an effect (Morrison et al., 2015). Although the TM would be under stress due to pigment dispersion in the DBA/2J model, the effects of impaired estrogen signaling may manifest themselves more clearly. Interestingly, IOP was higher in DBA/2J females than males (Zhou et al., 2007).

To date, we are aware of only two studies examining the treatment of TM cells with E2. Russell-Randall et al. employed E2 as a control for NO agonism and demonstrated that E2 induced NO production in an immortalized human TM cell line (Russell-Randall and Dortch-Carnes, 2011). In a separate study, Mookherjee et al. showed that E2 treatment of TM cells increased the expression of glaucoma-associated MYOC through a mechanism mediated by EREs in the MYOC promoter (Mookherjee et al., 2012). Interestingly, mutations in the CYP1B1 gene lowered CYP1B1 metabolism of E2 and resulted in MYOC upregulation (Mookherjee et al., 2012). Although overexpression of wild-type myocilin (encoded by MYOC) is not pathogenic, these findings could help to explain the digenic effects of CYP1B1 and MYOC in which these two genes interact with each other in IOP regulation (Kim et al., 2001; Liu and Allingham, 2017; Resch and Fautsch, 2009; Youngblood and Liu, 2020).

4.4. Possible Mechanisms

Several mechanisms may explain the effects of estrogen signaling on IOP regulation and POAG risk (Figure 5). One possible mechanism involves the effects of estrogen signaling on cells that turnover ECM components; TM turnover of ECM plays a vital role in IOP homeostasis (Acott et al., 2021). The role of estrogen signaling in ECM modulation has been well-established, especially as pertaining to cancer metastasis and epithelial to mesenchymal transition (Di Zazzo et al., 2019; Gelissen and Huang, 2022; Jeon et al., 2016; Lephart and Naftolin, 2021; Piperigkou and Karamanos, 2020; Schuler and Murdoch, 2021). In addition, the loss of estrogen in menopause leads to changes in skin ECM during the normal process of aging, affecting ECM molecules such as collagen, elastin, and matrix metalloproteinases (Lephart and Naftolin, 2021). At a more mechanistic level, mechanosensitive transcriptional regulators YAP and TAZ are known to upregulate ECM-related genes such as TGM2, SERPINE1, and CCN2 in response to TGFβ2 treatment of TM cells (Dupont et al., 2011; Li et al., 2022; Liu et al., 2021; Thomasy et al., 2013) and in response to being grown on stiffer substrates (Dupont et al., 2011; Raghunathan et al., 2013; Thomasy et al., 2013; Yemanyi et al., 2020). Interestingly, treating myometrium cells with E2 causes the nuclear localization of YAP/TAZ, presumably affecting their transcriptional activity (Purdy et al., 2020). As additional support for interactions between the estrogen receptors and YAP/TAZ, GPR30 has been shown to regulate YAP/TAZ in cancer cells (Zhou et al., 2015), and YAP/TAZ has been shown to reduce ESR1 expression in breast cancer cells (Ma et al., 2022). However, an interaction between YAP/TAZ and components of the estrogen signaling pathway has not yet been established in TM cells.

Figure 5. Possible mechanisms by which estrogen signaling may affect intraocular pressure.

By modulating extracellular matrix turnover, actin stress fiber formation, and focal adhesion assembly, estrogen signaling may impact the stiffness of trabecular meshwork cells and their ability to sense mechanical changes in their environment. Estrogen signaling likely produces nitric oxide in the aqueous humor outflow tissues, which would lead to the relaxation of the trabecular meshwork, Schlemm’s canal, and the distal vessels, thereby increasing outflow. The effects of estrogen on these processes may explain estrogen’s possible promotion of IOP homeostasis, which is affected when estrogen levels decline below physiologic levels.

In addition to the ECM, estrogen may regulate the actin cytoskeleton and cell adhesions that are involved in TM cell response to mechanical stretch (Lakk and Križaj, 2021; Liton and Gonzalez, 2008; Tumminia et al., 1998; WuDunn, 2009) and TM cell regulation of IOP homeostasis (Acott and Kelley, 2008; Acott et al., 2021; Gasiorowski and Russell, 2009; Stamer, 2012; Stamer and Acott, 2012; Stamer and Clark, 2017; Zhong et al., 2013; Zhou et al., 1996). Estrogen signaling is known to impact the actin cytoskeleton (Babayan and Kramár, 2013; Giretti and Simoncini, 2008; Kramár et al., 2013; Sanchez et al., 2012; Sanchez and Simoncini, 2010) and to be involved in mechanotransduction (Imai et al., 2008; Wang et al., 2023). For example, estrogen signaling has been shown to impact neuronal morphology through effects on the actin cytoskeleton (Babayan and Kramár, 2013; Kramár et al., 2013; Sanchez et al., 2012; Sanchez and Simoncini, 2010) and has been implicated in the mechanoresponse of osteocytes to mechanical loading (Imai et al., 2008). Furthermore, in osteocytes, the formation of focal adhesions by αvβ3-integrin in response to shear stress is at least partially regulated by estrogen (Geoghegan et al., 2019). In the eye, IOP is elevated upon αvβ3-integrin activation (Faralli et al., 2019). In TM cells in particular, αvβ3-integrin activation by dexamethasone (Dickerson et al., 1998; Faralli et al., 2013; Yemanyi et al., 2020) induces cross-linked actin network (CLAN) formation (Filla et al., 2011; Filla et al., 2009; Filla et al., 2006) and promotes fibronectin fibril formation (Filla et al., 2019), hallmarks of glaucomatous TM. Taken together, a focal adhesion protein (i.e., αvβ3-integrin) important for IOP regulation is regulated by estrogen in other tissues, thereby suggesting that estrogen may influence IOP by regulating focal adhesion proteins such as αvβ3-integrin.

In addition, these findings also highlight that estrogen and dexamethasone may affect similar pathways. Both dexamethasone and E2 are sterols, thereby sharing similar, but distinct chemical structures. Both β-estradiol and dexamethasone were identified as being key upstream regulators of a group of differentially correlated genes that were differentially expressed in control/POAG TM and/or associated with IOP (Youngblood et al., 2021). β-estradiol and dexamethasone regulated several of the same genes (e.g., CAV2, DLL1, IGF1, SPTBN1, and VEGFC) (Youngblood et al., 2021). This suggests that the two sterols have overlapping downstream targets, which they may affect in opposing way with dexamethasone inducing ocular hypertensive effects and estrogen exerting normotensive effects.

Another possible mechanism for estrogen regulation of IOP involves the NO-mediated increase in outflow facility (Becquet et al., 1997; Sator et al., 1998). High pressure in the eye leads to movement of the TM and SC inner wall towards its outer wall, leading to compression of SC, which in turn, results in increased shear stress (Aliancy et al., 2017; Ashpole et al., 2014; Dismuke et al., 2014; Dismuke et al., 2008; McDonnell et al., 2020; Reina-Torres et al., 2021; Stamer et al., 2011). This shear-stress induces the synthesis of NO by endothelial nitric oxide synthase (NOS3) (McDonnell et al., 2020; Reina-Torres et al., 2021; Stamer et al., 2011). NO production then relaxes the TM, subsequently increases AH outflow, and is thought to dilate the distal vessels and improve permeability of the inner wall of SC (Aliancy et al., 2017; Dismuke et al., 2014; Dismuke et al., 2008; McDonnell et al., 2018; Reina-Torres et al., 2021; Stamer et al., 2011). Consistent with this proposed scheme of estrogen action, estradiol is known to regulate NO production through NOS3 in other tissues (Kim et al., 2008; Kim et al., 2014; Kypreos et al., 2014; Mantione, 2008; Ndzie Noah et al., 2021; Teoh et al., 2020). Because estrogen stimulates NO production, estrogen also could be involved in NO production in the AH outflow pathway (Reina-Torres et al., 2021). As mentioned above, NOS3 is associated with POAG in females, further supporting a connection between NOS3, estrogen, and IOP elevation/POAG risk (Kang et al., 2011). In addition to NOS3, the POAG association with SNPs in guanylate cyclase genes GUCY1A3/GUCY1B3 was stronger in women (Buys et al., 2013). Soluble guanylate cyclase is the intracellular receptor for NO (Buys et al., 2013). In addition, CAV1/CAV2 also has a female-biased association with POAG (Wiggs et al., 2011). This is interesting because CAV1 has been shown to translocate ERα to the nucleus, where ERα stimulates NOS3 production of NO through a non-genomic mechanism (Chen et al., 1999; Razandi et al., 2002; Russell et al., 2000). Thus, further studies are needed to clarify the exact mechanisms (genomic and non-genomic) of ERα in NO production. Overall, the female-specific association of CAV1, CAV2, and NOS3 and evidence for their interaction with ERα make this pathway a strong candidate for future study of how estrogen signaling influences AH outflow and IOP.

4.5. Estrogen Derivatives

Certain estrogen derivatives and estrogen receptor modulators may also influence IOP homeostasis and other glaucoma-related processes. For example, treatment with synthetic estrogen mestranol alone and combined with synthetic progestin ethynodiol caused a decrease in IOP by ~2 mmHg over 6 months, likely due to an accompanying increase in outflow facility (Treister and Mannor, 1970).

5. Future Directions

Although there has been significant progress in our understanding of the role of estrogen in IOP regulation, several significant questions remain (Table 3). First, it will be important to decipher which of the estrogen receptors is responsible for the estrogenic effects on IOP regulation or whether the estradiol ligand may be interacting with other steroid receptors. In addition, it will be important to identify the primary site of estrogen signaling in the outflow pathway (i.e., TM, SC, and/or distal vessels). Next, although we have posited several possible mechanisms, future efforts should be directed towards identifying and confirming which estrogen-regulated pathways regulate IOP homeostasis. This step will be essential for identifying targets for therapeutic intervention. Mechanistic questions include: what are the effects of estrogen treatment/deprivation on TM cell ECM; does estrogen play a role in regulating the stiffness of TM cells by modulating their actin cytoskeleton; is TM cell mechanosensation altered by estrogen signaling; and can estrogen improve AH outflow through NOS3-mediated NO production. Lastly, small molecule agonists and antagonists that modify these targets while causing minimal off-target effects will need to be identified to transform our understanding into practical application.

Table 3.

Several questions surrounding the role of estrogen signaling in intraocular pressure homeostasis require future study.

Significant Questions for Future Study
1	Which estrogen receptor is responsible for estrogenic effects on intraocular pressure?
2	Which cells and tissue types are targeted by estrogen signaling?
3	What are the effects of estrogen on the extracellular matrix of trabecular meshwork cells?
4	Does estrogen affect the actin cytoskeleton and subsequently the stiffness of trabecular meshwork cells?
5	Does estrogen signaling alter trabecular meshwork cell mechanosensation?
6	Can estrogen improve aqueous humor outflow by inducing NOS3-mediated nitric oxide production?
7	How can we therapeutically target these pathways and what are the best drug delivery methods?

To address these questions, both SC and TM cells will be of interest due to their regulatory role in AH outflow. In addition, several animal models are available for in vivo studies of estrogen signaling effects in the eye, including natural estropause models (Koebele and Bimonte-Nelson, 2016), surgical menopause (i.e., ovariectomy) models (Acosta et al., 2009; Koebele and Bimonte-Nelson, 2016), chemical menopause models (National Toxicology Program 1989; Carolino et al., 2019; Dhillon and Von Burg, 1996; Flaws et al., 1994; Hoyer et al., 2001; Hu et al., 2001a; Hu et al., 2001b; Kappeler and Hoyer, 2012; Koebele and Bimonte-Nelson, 2016; Mayer et al., 2004; Smith et al., 1990; Springer et al., 1996), and genetic knockout models. Both age and sex should be included in the experimental design to allow for assessment of sex-specific outcomes and age-related effects on hormone levels.

6. Limitations and Factors to Consider

The eventual goal is to manage elevated IOP and glaucoma by targeting the estrogen pathway. However, there are many possible issues that limit a systemic or localized approach to treating glaucoma. First, systemic estrogen therapy could lead to several unintended side effects, including an increased risk of breast cancer. Second, additional deleterious side effects could develop with systemic estrogen therapy, including the development of female secondary sex characteristics in male users. Third, estrogen therapy could lead to unwanted changes in corneal thickness in patients affected by disorders like keratoconus, a corneal thinning disease that is thought to be affected by hormone changes (Escandon et al., 2022; Karamichos et al., 2022; Torres-Netto et al., 2019; Yuksel et al., 2016).

As a result of these unintended consequences, it will be vital to develop therapies that locally target only IOP-relevant portions of the estrogen signaling pathway. Due to AH outflow patterns, the conventional outflow pathway is the ideal system to do this. Although even localized treatments could still have unintended local or systemic effects, it vastly improves upon using a systemic treatment. This strategy requires the development of an appropriate delivery system as well as deciphering the mechanism by which estrogen signaling regulates IOP homeostasis. Both questions will require in vivo testing in animal models. Although animal models are an essential part of the drug development process, models are limited by their very nature and do not represent the complexity of heterogenous human populations and complex human diseases. As a result, in vivo and ex vivo experiments using cells and tissues derived from human organ donors will continue to be important. Cells derived from individuals with glaucoma will be especially valuable as inducing glaucomatous changes in normal human cells and tissue cannot completely mimic the complexities of the disease. However, human cells and tissues are a limited resource, with cells and tissues from individuals with glaucoma being especially rare. Hopefully, in the future the availability of these critical resources will improve.

Aside from these logistical limitations, the overlapping crosstalk of the different estrogen signaling pathways poses a significant hurdle for elucidating IOP- and POAG-relevant parts of the estrogen signaling pathway. The promiscuous nature of both estrogen ligands and receptors further complicates this deciphering process. Despite these limitations and difficulties, the strong evidence summarized in this review supports the estrogen signaling pathway to be a promising, as-of-yet-unused target for reducing glaucomatous ocular hypertension.

7. Conclusion

In conclusion, strong evidence from epidemiological studies, genetic associations, in vitro/ ex vivo cell and tissue-based data, and animal models support a strong, protective, regulatory role for estrogen signaling in IOP homeostasis and glaucoma development. Estrogen signaling is known to target ECM turnover, actin cytoskeletal assembly, mechanotransduction, and NO-induced vessel relaxation, each of which could affect AH outflow through the TM, SC, and distal vessels. Because estrogen signaling comprises several distinct and overlapping pathways, significant attention will need to be directed towards identifying which estrogen receptor pathway and what downstream targets are responsible for regulating IOP homeostasis and glaucoma development.

Highlights:

• Elevated intraocular pressure (IOP) is the only treatable risk factor for glaucoma.

• Estrogen signaling is inversely correlated with glaucoma risk and high IOP.

• Sequence variants in the estrogen pathway are associated with high IOP and glaucoma risk.

• Estrogen signaling-related variants have female-biased association with POAG.

• Estrogen-depleted animal models have shown elevated IOP levels.