Expression and Putative Role of 11β-Hydroxysteroid Dehydrogenase Isozymes within the Human Eye

Abstract

purpose. The human eye is an important target tissue for steroid hormones, and glucocorticoids have been implicated in the pathogenesis of ocular disease, including glaucoma. In peripheral tissues, corticosteroid hormone action is regulated at a prereceptor level through the activity of the 11β-hydroxysteroid dehydrogenase (11β-HSD) isozymes: an oxo-reductase (11β-HSD1) that activates cortisol (F) from cortisone (E) and a dehydrogenase (11β-HSD2) that inactivates F to E. The purpose of this study was to analyze the expression and putative role of 11β-HSD within the human eye.

methods. Immunohistochemical and reverse transcription–polymerase chain reaction (RT-PCR) studies were performed on sections of human ocular tissues, surgical trabecular meshwork (TM) specimens and a ciliary nonpigmented epithelial (NPE) cell-line. Free F and E concentrations in aqueous humor were determined by gas chromatography-mass spectrometry (GC/MS). IOP was measured in eight male volunteers before and after oral ingestion of carbenoxolone (CBX), a known inhibitor of 11β-HSD.

results. 11β-HSD1 was expressed in the basal cells of the corneal epithelium and the NPE. 11β-HSD2 was restricted to the corneal endothelium. RT-PCR revealed mRNA for only the glucocorticoid receptor (GR) in the TM specimens, whereas GR, mineralocorticoid receptor and 11β-HSD1 mRNAs were all present in the NPE cell line. The demonstration of free F in excess of E (F/E 14:1) in the aqueous humor suggested predominant 11β-HSD1 activity. Compared with baseline (14.7 ± 1.06 mm Hg, mean ± SD), the IOP decreased significantly on both the third and seventh days of CBX ingestion (12.48 ± 1.11 mm Hg, P < 0.0001 and 11.78 ± 1.50 mm Hg, P < 0.0001, respectively).

conclusions. These results suggest that the 11β-HSD1 isozyme may modulate steroid-regulated sodium transport across the NPE, thereby influencing IOP.

Aqueous humor is produced by the ciliary epithelium, a complex bilayer consisting of an inner nonpigmented epithelial (NPE) layer in direct contact with the aqueous humor and an outer pigmented epithelial (PE) layer adjacent to the highly vascularized connective tissue stroma of the ciliary body. Secretion is dependent on several mechanisms including sodium-potassium adenosine triphosphatase (Na⁺-K⁺-ATPase) active transport, ¹ the carbonic anhydrase enzyme system, diffusion, and ultrafiltration. Aqueous circulates from the posterior chamber into the anterior chamber and is drained predominantly through the trabecular meshwork (TM) into Schlemm’s canal. 

In classic target tissues, such as the kidney, colon, and salivary gland, epithelial sodium transport is regulated, in part, by corticosteroids through stimulation of both the apical epithelial sodium channel and the basolateral Na⁺-K⁺-ATPase pump. ^{2 3 4} At a prereceptor level, the activity of 11β-hydroxysteroid dehydrogenase (11β-HSD), responsible for the interconversion of hormonally active cortisol (F) and inactive cortisone (E), must be considered. ⁵ Two isozymes have been characterized ^{6 7} : an oxo-reductase (11β-HSD1) that regulates F exposure to the glucocorticoid receptor (GR) at several sites, including liver ⁸ and adipose tissue, ⁹ and a dehydrogenase (11β-HSD2) that protects the mineralocorticoid receptor (MR) from F by inactivating it to E. ^{10 11} Deficiency of 11β-HSD2 in the inherited form of hypertension, in the syndrome of apparent mineralocorticoid excess, ^{12 13} or after liquorice or carbenoxolone ingestion ^{14 15 16} results in cortisol-mediated, mineralocorticoid hypertension. 

The eye represents an important target tissue for corticosteroids that express both the MR ¹⁷ and GR. ¹⁸ Corticosteroids have been implicated in the natural diurnal variation of IOP, and increased IOP may also occur in patients with endogenous or exogenous corticosteroid excess. ^{19 20 21 22 23} Despite this, only a few studies have addressed the role of corticosteroids in the regulation of aqueous humor secretion and reabsorption. This may be of considerable relevance due to the widespread use of topical and systemic glucocorticoids in a variety of conditions in clinical ophthalmology, where one of the most important complications is corticosteroid-induced glaucoma. This condition is characterized by increased IOP secondary to increased outflow resistance ^{19 20 21 22 23} and resembles the more common primary open-angle glaucoma (POAG), a leading cause of blindness due to irreversible optic neuropathy associated with uncontrolled IOP. Patients with POAG are at a higher risk of development of corticosteroid-induced glaucoma, and the two conditions appear to be linked by a common genetic defect in the trabecular meshwork-induced glucocorticoid response (TIGR)/myocilin gene, which is located on the long arm of chromosome 1 and mediates outflow resistance by the deposition of an extracellular protein. ²⁴ Other ocular complications associated with corticosteroid excess include posterior subcapsular lens opacities and florid ulceration of the cornea after injudicious use of topical steroids for the treatment of herpes simplex keratitis. 

Our hypothesis was that corticosteroid regulation of aqueous humor production and drainage would be mediated at a prereceptor level through 11β-HSD expression within ocular tissues. Furthermore, because of the established role of corticosteroids in other ocular tissues, we conducted a detailed analysis of the 11β-HSD isozyme expression within the human eye. 

Methods

Immunohistochemistry

Human eye sections were obtained from the Academic Unit of Ophthalmology, University of Birmingham, UK. Immunohistochemical analyses were performed on 5-μm formalin-fixed, paraffin-embedded sections of six human eyes (five males; mean age, 54.1 ± 16.1[ SD] years) acquired at surgical enucleation. In all cases, the underlying diagnosis was choroidal malignant melanoma, and only adjacent normal structures were studied. 

Immunoperoxidase and immunofluorescence studies were performed using antisera raised in sheep against human 11β-HSD1 (amino acids 18-33) and 11β-HSD2 (amino acids 137-160 and 334-358), as previously reported. ^{25 26} Antibody dilutions were 1:200 for 11β-HSD1 and 1:100 for 11β-HSD2. Control sections included the omission of primary antibody and use of antibody pretreated with the immunizing peptides. Secondary antibodies comprised donkey anti-sheep peroxidase conjugate (1:200) or donkey anti-sheep alkaline phosphatase conjugate (1:400; Binding Site, Birmingham, UK). Sections were developed with the peroxidase substrate 3,3′-diaminobenzidine, or an alkaline phosphatase substrate containing the alkaline phosphatase blocking agent levamisole (Vector Red; Vector Laboratories, Peterborough, UK). 

Reverse Transcription–Polymerase Chain Reaction

TM specimens with associated drainage angle tissue, were obtained after ethics committee approval, from eight patients undergoing glaucoma filtration surgery. Specimens were snap frozen in liquid nitrogen and then stored at −70°C until further analysis. The presence of TM and the canal of Schlemm was confirmed histologically on formalin-fixed tissue fragments. Human ciliary NPE cells (ODM-2), were cultured as previously described. ²⁷ RNA was prepared from tissue and cultured cells using a single-step extraction method (RNAzol B RNA isolation kit; AMS Biotechnology, Oxon, UK) according to the manufacturer’s protocol. 

Reverse transcription of RNA was performed using a commercial reverse transcription system (Promega, Southampton, UK). Briefly, 1 μg total RNA and 0.75 μg random hexamers were preannealed by incubation at 70°C for 5 minutes. Primer extension was then performed at 37°C for 60 minutes after the addition of reaction buffer, 1 mM of each dNTP, 80 U rRNasin RNase inhibitor, and 50 U avian myeloblastosis virus (AMV) reverse transcriptase. An aliquot of this reaction was taken for subsequent PCR reactions, by using a previously published method and primer pairs for 11β-HSD1, 11β-HSD2, GR, and MR, ^{28 29} and transcript sizes were generated of 571, 477, 693, and 450 bp, respectively. Human hepatocyte cDNA was used as a positive control for 11β-HSD1 and GR, whereas human placental cDNA was used for 11β-HSD2 and MR. 18S ribosomal RT-PCR was performed to confirm the presence and integrity of RNA in all samples. 

Aqueous Humor Cortisol and Cortisone Assays

Aqueous humor F and E concentrations were determined by gas chromatography-mass spectrometry (GC/MS) by pooling aqueous humor specimens from 16 patients who had given informed consent (seven males, nine females; age 72.8 ± 12.8 years) undergoing routine phakoemulsification of cataract at the Birmingham and Midland Eye Centre. Patients with a history of glaucoma, diabetes, uveitis, or underlying endocrine disease and those on topical or systemic corticosteroids were excluded. GC/MS was performed, using an adaptation of a previously reported method. ³⁰ Samples were pooled to provide a total volume of 1.15 ml, diluted to 4-ml and 1-ml portions (25%), for analysis. Isotope-labeled internal standards (24 ng ²H₂-cortisone and 32 ng ²H₄-cortisol) were added to each 1-ml aliquot, diluted to 4 ml with water, and extracted by a solid-phase extraction cartridge (Sep-Pak; Waters, Milford, MA). The steroid extract was derivatized to make methyloxime-trimethylsilyl ethers, which were then analyzed by selected-ion–monitoring GC/MS. The ions—mass-to-charge ratio (m/z) 531 cortisone, m/z 533-labeled cortisone, m/z 605 cortisol, and m/z 609-labeled cortisol—were monitored, and the E and F concentrations determined from the 531:533 and 605:609 peak area ratios. 

Clinical Study

A pilot observational clinical study was performed by recruiting eight healthy male volunteers (age 21.5 ± 1.3 years) who were not receiving any systemic or topical medications and had no family history of glaucoma. The study protocol followed the tenets of the Declaration of Helsinki and was approved by the local ethics committee. Informed consent was obtained from all volunteers. Baseline IOP readings were measured by a single observer using the same Goldmann applanation tonometer at 8 AM and 12, 4, and 8 PM on two consecutive days. Using this method, intraindividual variability in IOP was less than 0.5% at any given time point. Systolic and diastolic blood pressures, recorded with an automated digital blood pressure monitor, (HEM-705CP; Omron Healthcare, Inc., Vernon Hills, IL) were measured at each time point. Urine was collected for cortisol (tetrahydrocortisol [THF], alloTHF, urinary free F [UFF]) and cortisone (tetrahydrocortisone [THE], urinary free E [UFE]) metabolites. 

The UFF-to-UFE ratio was used as an index of renal 11β-HSD2 activity, and the THF+alloTHF-to-THE ratio as an index of global 11β-HSD activity (i.e., 11β-HSD1 and -2), as previously validated by our group. ³⁰ Subjects then received carbenoxolone (CBX; 100 mg) treatment three times a day for seven consecutive days. IOP measurements and blood pressure recordings were repeated at each time point on the third and seventh days of CBX ingestion. A further 24-hour urine collection was performed on the last day of CBX ingestion. Statistical analysis was performed by computer (Minitab 13.1 for Windows; University Park, PA). A combination of multiple linear regression and balanced analysis of variance was used to analyze IOP, and a paired t-test was used to evaluate the urinary steroid metabolites before and after CBX treatment. The association between changes in IOP and urinary steroid metabolites was assessed by linear regression and Spearman rank correlation. 

Results

Immunohistochemistry

11β-HSD1 was localized to the basal cells of the corneal epithelium (Fig. 1A ; negative control Fig. 1B ), whereas 11β-HSD2 immunoreactivity was restricted to the corneal endothelium (Fig. 1C ; negative control Fig. 1D ). In the ciliary body, intense staining for 11β-HSD1 was seen, most notably in the ciliary epithelium (Fig. 2A ; negative control Fig. 2B ), but there was no staining for 11β-HSD2 at this site (Fig. 2C ; negative control Fig. 2D ). To examine the PE in more detail, immunofluorescence studies were performed, which confirmed that 11β-HSD1 immunoreactivity was localized only to the NPE (Fig. 2E) and that there was no expression of either isozyme in the PE. No staining was observed in the TM, canal of Schlemm (Fig. 2F) , or lens (data not shown). 

Reverse Transcription–Polymerase Chain Reaction

RT-PCR analysis of TM specimens revealed mRNA for the GR, but not for the MR or either 11β-HSD isozyme (Fig. 3A) , consistent with the negative 11β-HSD immunohistochemical data. In contrast, in the NPE ODM-2 cells, mRNAs for MR, GR, and 11β-HSD1, but not for 11β-HSD2 (Fig. 3B) , were identified. 

Gas Chromatography–Mass Spectrometry

GC/MS of the aqueous humor specimens demonstrated high levels of free F (3.6 ng/ml) compared with free E (lower limit of detection, 0.25 ng/ml), giving an F-to-E ratio of 14:1. 

Clinical Study

There was no significant difference in measured IOP between either eye of each subject before or after the ingestion of CBX. The mean IOPs measured at baseline on days 1 and 2 were similar at 15.05 ± 1.19 and 14.31 ± 1.04 mm Hg, respectively. Compared with mean daily baseline levels on days 1 or 2, IOP was lower on the third (12.48 ± 1.11 mm Hg, P < 0.001) and seventh (11.78 ± 1.50 mm Hg, P < 0.001) days of CBX ingestion (Fig. 4) . The difference between days 3 and 7 were not significant (P = 0.14). There was a small reduction in IOP during the course of the day (i.e., from 8 AM to 8 PM; baseline reduction of 0.38 ± 0.58 mm Hg [2.55%], P = 0.30), which became more marked on days 3 and 7 of CBX ingestion (reduction of 0.87 ± 0.74 mm Hg [6.5%], P = 0.01 and 1.69 ± 1.73 mm Hg [13.3%], P = 0.03, respectively). 

Systolic (SBP) and diastolic (DBP) blood pressures (baseline SBP 130.5 ± 10.0 mm Hg, during treatment SBP 125.7 ± 21.5 mm Hg; baseline DBP 75.5 ± 9.4 mm Hg, during treatment DBP 74.4 ± 7.6 mm Hg) and serum electrolytes including potassium did not alter significantly throughout the course of the study. 

The UFF-to-UFE ratio increased significantly after CBX administration (0.50 ± 0.19 vs. 1.14 ± 0.38, P < 0.01) indicating inhibition of 11β-HSD2. Despite this, the urinary THF+alloTHF-to-THE ratio decreased significantly (0.92 ± 0.23 vs. 0.70 ± 0.19, P = 0.001), reflecting concomitant inhibition of 11β-HSD1 activity. There was a significant positive correlation between the reductions in IOP and urinary THF+alloTHF-to-THE ratio (r = 0.83, P = 0.01), but no correlation was seen with the UFF-to-UFE ratio. 

Discussion

The glaucomas constitute a prevalent group of conditions characterized by a distinctive excavating optic neuropathy with corresponding visual field loss. ³¹ In the majority of cases, IOP is elevated, and loss of vision is painless and progressive, often escaping detection until advanced. IOP is maintained by a balance between production and drainage of aqueous humor. Aqueous is produced by the ciliary processes, which consist of a specialized bilayer of neuroendocrine epithelium covering a stromal core. Anatomically, the two apical surfaces of the PE and NPE lie opposed to each other, communicating with each other through numerous gap junctions. Tight junctions exist near the apices of the NPE cells, which, together with the nonfenestrated iris vessels, contribute to the blood–aqueous barrier. Several mechanisms are involved in aqueous production, the most important of which is energy-dependent Na⁺-K⁺-ATPase. The pump has been localized to the basolateral surface of the NPE and is therefore in contact with the epithelial–aqueous interface ^{1 32} (Fig. 5) , in contrast to epithelial cells within corticosteroid target tissues, such as the kidney and colon, where the pump is in contact with the epithelial–stromal vascular interface. 

The underlying regulatory mechanisms for this active secretory process are not known, but corticosteroids are known to stimulate both the apical epithelial sodium channels and basolateral Na⁺-K⁺-ATPase. ^{2 3 4 33} Corticosteroids seem likely to play a role in the regulation of ocular ciliary epithelial Na⁺-K⁺-ATPase and therefore aqueous production, contributing to the maintenance of IOP. This is supported by the demonstration of both the MR and GR in various ocular tissues, endorsed by our studies, and the presence of F and aldosterone within the aqueous humor. ^{17 18 19} IOP also varies throughout the day, with a circadian rhythm similar to that reported for F and one that is accentuated in patients with POAG. ²³ Furthermore, both MR ³⁴ and GR antagonists ³⁵ lower IOP acutely, suggesting an important role for corticosteroids in aqueous humor production. 

Much more is known about endogenous or exogenous corticosteroids and their effect on reducing aqueous outflow. ^{36 37} This occurs in approximately 30% of patients taking glucocorticoids, increasing to more than 90% in patients with established POAG. ^{21 38} This ocular hypertensive effect of corticosteroids is thought to have a hereditary component and may be a marker for the subsequent development of glaucoma. ^{38 39} Susceptible individuals may have an increase in IOP within a few hours or as long as months to years after the administration of corticosteroids. Both the acute and chronic forms of corticosteroid-induced glaucoma appear to respond to the cessation of corticosteroid therapy. The underlying pathogenesis is unclear but is thought to be mediated by deposition of an extracellular protein in the trabecular meshwork that is likely to be the product of the TIGR gene, also known as myocilin. ^{24 37}  

Our study demonstrated the expression of predominantly 11β-HSD1 within human ocular tissue, principally the NPE and the corneal epithelium. 11β-HSD2 expression was restricted to the corneal endothelium. Although the presence of the 11β-HSD isozymes in the corneal tissues could imply a role in stromal dehydration and the preservation of corneal transparency, the intense 11β-HSD1 expression seen in the NPE and the ODM-2 cell line and the absence of expression of either 11β-HSD isozyme in the TM suggest that 11β-HSD1 may have a role in aqueous production rather than drainage. These data are supported by a recent in situ hybridization study demonstrating the expression of mRNA for 11β-HSD1 in ciliary epithelial cells, for both MR and GR in the NPE, and for GR in the TM. ⁴⁰ Contrary to our findings, mRNAs for both isozymes were also demonstrated in the TM. 

Our data demonstrating the exclusive presence of 11βHSD1, detected by both RT-PCR and immunohistochemistry of the NPE, is surprising in view of the established autocrine role of 11β-HSD2, but not 11β-HSD1, in modulating corticosteroid-regulated sodium transport within other epithelial cells, notably kidney, colon, and salivary gland. ^{5 7 10 11} Nevertheless, the novel analyses of F and E within aqueous humor samples and our clinical study appear to support the expression of a functional 11β-HSD1 enzyme within the NPE. In urine and saliva, free concentrations of E exceed those of F, giving F-to-E ratios of 0.8 and 0.2, respectively. ^{29 30} This has been attributed to the predominant expression of 11β-HSD2 in the kidney and salivary glands. Conversely, in our study aqueous humor F concentrations exceeded those of E yielding an F-to-E ratio in excess of 14:1. In contrast to urine and saliva, E concentrations were very low in kidney and salivary gland, indicating functional expression of the 11β-HSD1 isozyme. Earlier studies provide evidence of cortisol metabolism within both human and rabbit ocular tissues, but these data mainly refer to the cortisol A-ring metabolism (5α/β-reductase, tetrahydrocortisol) and not the interconversion of F and E by 11β-HSD. ^{41 42}  

In addition, the systemic administration of CBX to healthy volunteers provided further evidence for a potential role of 11β-HSD1 within the eye. Previous in vitro and clinical studies have demonstrated that CBX inhibits both 11β-HSD2 and 11β-HSD1 activities. ^{7 15} Our data supported this concept: The increase in the UFF-to-UFE ratio was indicative of inhibition of 11β-HSD2 activity, whereas the concomitant decrease in the THF+alloTHF-to-THE ratio indicated inhibition of 11β-HSD1 activity. Although our clinical study was single-blind and observational, our finding of a 17.5% decrease in IOP from baseline after 3 to 7 days, suggested an inhibition of 11β-HSD1 activity within the NPE, a reduction in local F concentrations with consequent decreased aqueous production, and a decrease in IOP (Fig. 5) . 

Under normal physiological conditions, activity of 11β-HSD1 may mediate exposure of the GR within the TM to F, which could contribute to aqueous humor outflow resistance and increased IOP. This may account for the acute and chronic changes in IOP observed in steroid-induced glaucoma and certain patients with POAG. The positive correlation between the reduction in IOP after systemic administration of CBX and decrease in the THF+alloTHF-to-THE ratio potentially supported our hypothesis. If, as we had anticipated at the initiation of this study, 11β-HSD2 was the predominant isozyme in human ocular tissues including the NPE, then the reverse would have been observed (i.e., an increase in IOP after CBX-induced enzyme inhibition). Although the type 1 isozyme may exhibit both reductase and dehydrogenase activities, in intact cells, 11β-HSD1 is principally a reductase. Preliminary activity data in ODM-2 cells have confirmed that this is indeed the case (data not shown), consistent with the results from our novel analysis of free F and E concentrations in aqueous humor. Nevertheless, double-blind, placebo-controlled trials incorporating aqueous humor dynamics and outflow facility studies are now required to further evaluate this observation. 

We conclude that by mediating local intraocular cortisol levels, 11β-HSD1 may have a twofold role within the human eye: first, a short-term physiological role, centered around the sodium transporting NPE and the secretion of aqueous humor, maintaining a normotensive, intraocular environment; second, a more long-term pathologic role, through interactions with the GR and TM, contributing to outflow resistance in susceptible individuals. Relative expression and activity of this isozyme, could therefore represent one of the underlying pathogenic mechanisms of POAG, one of the most common causes of visual loss in the Western world. ³⁰ The future pharmacologic manipulation of 11β-HSD activity with topical or systemic derivatives of CBX or more selective 11β-HSD1 inhibitors may provide a novel treatment option for patients with glaucoma. 

⁴  SR and EAW contributed equally to the work and share the role of first author.

 This work forms the basis of an International Patent Application (No. 9914648.2), “Glaucoma Treatment.”

 Presented at the annual meeting of the Association for Research in Vision and Ophthalmology, Fort Lauderdale, Florida, May 2000.

 Supported by the West Midlands National Health Service Executive (Locally Organised Research Scheme) and the Medical Research Council, United Kingdom. PMS is a Medical Research Council Senior Clinical Fellow.

 Submitted for publication December 18, 2000; revised March 14, 2001 accepted April 6, 2001.

 Commercial relationships policy: N.

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U. S. C. §1734 solely to indicate this fact.

 

Figure 1.

 

Expression of 11β-HSD isozymes within the human cornea. (A) Immunoperoxidase staining for 11β-HSD1 revealed immunoreactivity of specific corneal epithelial basal cells (BC). (B) 11β-HSD1 negative control showing absence of staining in the corneal epithelium. (C) Corneal endothelial cells (CE) showing immunoperoxidase staining for 11β-HSD2. (D) 11β-HSD2 negative control showing absence of staining in the corneal endothelial cells. Magnification, ×400.

Figure 2.

 

Expression of 11β-HSD isozymes within the human ciliary epithelium and TM. (A) Immunoperoxidase staining for 11β-HSD1 in the ciliary epithelium, comprising an inner NPE layer and a PE layer, shows staining in the NPE. (B) Ciliary epithelium 11β-HSD1 negative control. (C) Absence of staining for 11β-HSD2 in the ciliary epithelium. (D) 11β-HSD2 negative control. (E) Immunofluorescence staining confirming that 11β-HSD1 expression was confined to the NPE. No staining was observed in the PE. (F) Absence of staining of either isozyme in the TM. Magnification, ×250.

Figure 3.

 

RT-PCR analysis of the (A) TM and (B) NPE ciliary cell line ODM-2. An mRNA species for the GR was identified in the TM, but MR, 11β-HSD1, and 11β-HSD2 mRNAs were undetectable. In contrast, in the ODM2 NPE cells, there was expression of both MR and GR mRNAs as well as 11β-HSD1, but no 11β-HSD2 mRNA was detected. Integrity of the RNA samples was confirmed by the presence of 18S ribosomal RNA (18S). −ve, no DNA template; +ve, human hepatocyte cDNA for GR and 11β-HSD1 and human placental cDNA for MR and 11β-HSD2.

Figure 4.

 

Mean change in IOP after 3 and 7 days of CBX (300 mg/d) treatment. Results represent the mean levels of right and left eye IOP in eight normal subjects measured at 8 AM and 12, 4, and 8 PM by the same observer. Data are mean ± SD. CBX caused a highly significant reduction in mean IOP on days 3 and 7 of CBX ingestion.

Figure 5.

 

Proposed model highlighting the role of 11β-HSD1 expression within the NPE and the control of corticosteroid-regulated IOP. Expression of 11β-HSD1 ensures the autocrine conversion of E to F within the NPE, modulating the concentration of F within the anterior segment of the human eye. In the short term, F induces sodium and concomitant water transport into the posterior chamber, through epithelial sodium channels, including Na⁺-K⁺-ATPase, resulting in aqueous humor production. Inhibition of 11β-HSD1 after systemic administration of CBX lowers F generation and therefore IOP. In the long term, F may interact with the GR within the TM, which in susceptible individuals, may contribute to outflow resistance, thereby increasing IOP.

The authors thank Eamon O’Neill and Ian Cunliffe (Birmingham and Midland Eye Centre, UK) for collection of the trabecular meshwork specimens, Miguel Coca-Prados (Yale University, New Haven, CT) for donating the ODM-2 nonpigmented ciliary epithelial cell line, and Tim Marshall (Department of Public Health and Epidemiology, University of Birmingham, UK) for invaluable help with statistical analysis.