eNOS Activity in CAV1 Knockout Mouse Eyes

Abstract

Purpose: To investigate endothelial nitric oxide synthase (eNOS) activity and the response of conventional outflow facility to nitric oxide donors and a nitric oxide synthase (NOS) inhibitor in caveolin-1 (CAV1) knockout (KO) mice.

Methods: Intraocular pressure (IOP) was measured in both CAV1 KO and wild-type (WT) mice by rebound tonometry. The expressions of caveolin-2 (CAV2), eNOS, eNOS-phospho Ser¹¹⁷⁷, eNOS-phospho Thr⁴⁹⁵, Akt, Akt-phospho Ser⁴⁷³, and nitrotyrosin were measured by Western blot analysis. Nitric oxide donor sodium nitroprusside (SNP), S-nitroso-N-acetyl-D,L-penicillamine (SNAP), or the NOS inhibitor L-NG-Nitroarginine Methyl Ester (L-NAME) were administered topically. The outflow facility was measured by perfusing enucleated mouse eyes at multiple pressure steps.

Results: CAV1 KO mice have elevated IOP and reduced conventional outflow facility when compared with WT mice. CAV2 expression was absent in CAV1 KO mice, but we observed increased expressions of eNOS, eNOS-phospho Ser¹¹⁷⁷, Akt, Akt-phospho Ser⁴⁷³, and nitrotyrosin and reduced expression of eNOS-phospho Thr⁴⁹⁵. Topical application of SNP significantly reduced IOP in WT and KO mice by 1.6 fold (n = 6, P < 0.05), but SNAP did not change IOP significantly (n = 6, P > 0.05). In comparison, the NOS inhibitor L-NAME significantly increased IOP by 50% in KO mice (n = 6, P < 0.05). SNP and SNAP significantly increased, whereas L-NAME significantly reduced pressure-dependent drainage in KO animals.

Conclusions: Although CAV1 KO mice had elevated IOP and decreased outflow facility, CAV1 deficiency (and possibly the loss of CAV2) resulted in increased eNOS activity. The pressure elevation may be a result of increased tyrosine nitration of protein kinase K and impairment of its activity in KO mice.

Primary open-angle glaucoma (POAG) is a complex, genetically heterogeneous disease. At least 20 genetic loci have been identified that are associated with POAG.¹ A genome-wide association study identified a single nucleotide polymorphism, rs4236601, located between caveolin-1 (CAV1) and caveolin-2 (CAV2) on chromosome 7q31,² which was significantly associated with elevated intraocular pressure (IOP). Further studies investigated the link between this polymorphism and POAG with divergent results: some studies confirmed the original conclusion, whereas others did not find a connection between this polymorphism and POAG.^3–6 Although the findings are not consistent, CAV1 has been implicated in the pathology of glaucoma. 

Caveolins are the principal structural components of caveolae. Caveolae are 60- to 80-nm diameter invaginations in the plasma membrane that are essential for endocytosis.^7,8 CAV1 is expressed in trabecular meshwork (TM) cells⁹ and Schlemm's canal (SC) endothelial cells.^10,11 In TM cells, caveolin mediates endocytosis by which TM cells can alter the physiological catabolism of extracellular matrix and regulate aqueous outflow resistance.¹² In SC inner wall endothelial cells, altered CAV1 levels might contribute to aqueous humor outflow resistance.^10,11 Furthermore, CAV1 is involved in the retina physiology and pathophysiology, and the dysregulation of or abnormalities in Cav-1 have been linked to diabetic retinopathy¹³ and inflamed uveitic retinas.¹⁴ 

CAV1 interacts with a large number of different client proteins¹⁵ via its structural scaffolding domain. One of the most important ones is endothelial nitric-oxide synthase (eNOS, Fig. 1),¹⁶ which is a pressure-dependent regulator of IOP.¹⁷ In resting endothelial cells, the binding of CAV1 to eNOS has already been deciphered in that eNOS binds to CAV1 and is inhibited by CAV1 binding. This leads to decreased nitric oxide (NO) production. However, upon stimulation, the inhibitory clamp of CAV1 is relieved and NO production occurs.¹⁶ In a CAV1 knockout (KO) mouse model, it was shown that CAV1 deficiency resulted in IOP elevation and aqueous humor outflow reduction (Elliott MH, et al. IOVS 2014;55:ARVO E-Abstract 2888). However, it is not known if or how CAV1 deficiency affects eNOS signaling. 

Figure 1

 

A schematic figure showing how CAV1 interacts with eNOS to regulate eNOS activation. At a resting state, eNOS is in an inactivated state binding to CAV1. When the cells are subjected to stress or other stimulation, the intracellular calcium level then increases, which leads to the disruption of CAV1 and eNOS. At the same time, calmodulin attaches to eNOS in the presence of calcium ions, and Akt is activated through phosphatidylinositol 3-kinase (PI3K). eNOS catalyzes the synthesis of NO from L-arginine, which has a vital role in cell permeability. NO can further stimulate production of cyclic guanosine monophosphate (cGMP) by catalyzing enzyme soluble guanylate cyclase (sGC). sGC converts guanosine-5'-triphosphate (GTP) to cGMP. Once produced, cGMP can have a number of effects in cells, but many of those effects are mediated through the activation of protein kinase G (PKG). SNP and SNAP are NO donors, and L-NAME is a NOS inhibitor.

In view of the presence of CAV1, eNOS, and their interaction in the aqueous humor outflow pathway, we aim to investigate eNOS activity and the response of the conventional outflow facility to NO donors and the NOS inhibitor in CAV1 KO mice. 

Methods

Animals

All experiments were in compliance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. CAV1 KO mice were obtained from Jackson Laboratory through the Model Animal Research Center at Nanjing University. Wild-type (WT) C57BL/6J littermates were used as controls. Mice were bred and housed in clear cages covered loosely with air filters and contained white pine shavings for bedding. In this study, the mice aged 8 to 10 weeks were used. 

IOP Measurements in Mice

We measured IOP in both mouse eyes using rebound tonometery in live animals (TonoLab; ICare, Espoo, Finland). IOP was measured without anesthesia at the same time of day (10:00 to 11:00 AM). The pressures of both eyes were measured three times, and the average was counted as a single measurement. 

Mouse Eye Perfusion

Outflow facility was measured by perfusing enucleated mouse eyes. Mice were killed by neck dislocation and the eyes were enucleated. Perfusions were performed in male and female mice that were 8 to 10 weeks old at the time of the experiments. After IOP measurements, the mice were sacrificed. The eyes were enucleated within 5 minutes of death and were kept in Dulbecco's modified Eagle's medium (Hyclone, Shanghai, China). The experimental setup was developed by our laboratory and is described in detail elsewhere.¹⁸ Briefly, enucleated mouse eyes were cannulated with a 33-gauge bevelled-tip needle (Nanofil; World Precision Instruments, Shanghai, China). The eye was kept moist with gauze underneath the eye and drops of phosphate-buffered saline (PBS) applied between perfusions on top of the cornea. The outflow rate measurement method was as previously described.¹⁹ Briefly, the eyes were cannulated and the eye pressure was stabilized using a perfusion reservoir containing PBS for 30 minutes, and the total infusion volume was recorded after a 10-minute interval. At each pressure, the measurement was repeated twice. The eye was perfused at four different levels of IOPs (6, 13, 19, and 27 mm Hg). We assume that at equilibrium, the total inflow rate equals the total outflow rate. Conventional outflow facility (C_con) and unconventional flow rate (F_u) were calculated according Goldman's equation:  

 

where F_u is the pressure-independent (unconventional) outflow rate, EVP is episcleral venous pressure, and C_con is the conventional (pressure dependent) outflow facility. Because the mouse eyes were enucleated at the time of the experiment, EVP equals zero. A regression was fit in a flow rate − pressure response graph, and the slope of the regression line was C_con.  

To test the effect of the NO donor sodium nitroprusside (SNP), S-nitroso-N-acetyl-D,L-penicillamine (SNAP), and NOS-inhibitor L-NG-Nitroarginine Methyl Ester (L-NAME), the mouse eyes were perfused with these drug solutions for 60 minutes before the measurements were taken to exchange the aqueous humor and ensure that the drug concentration was uniform throughout the experiment. The drug concentrations used were 10⁻³ M, 100 μM, and 100 μM for SNP, SNAP, and L-NAME respectively, which were similar to previous publications with monkeys and mice.^17,20,21 The contralateral control eyes were treated in the same way but perfused with a drug vehicle. The SNP solution was kept from light exposure because it is known to degrade when exposed to light. 

Western Blot

Outflow tissue was dissected according to the established method. Eyes from 8 CAV1 KO mice and 8 WT control mice were enucleated. After trimming extraocular tissues, the eyes were bisected sagittally at the equator under a stereomicroscope. The zonule was cut with fine scissors, and the lens was removed. The ciliary body was grasped and gently pulled away from the sclera while its attachments to the sclera were cut with a fine pair of curved scissors. After removing the ciliary body and the attached iris, the outflow tissue, which could be easily distinguished against the white background of the sclera as a pigmented circumferentially oriented band, was cut free. The dissected tissue contained iris TM, SC, and possibly some iris root. Outflow tissue of CAV1 KO and WT mice were prepared using a RIPA solution (Sigma-Aldrich), 50 ug of protein was loaded into each lane of gel, and proteins were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (10% or 12.5% acrylamide). The resolved proteins were transferred by electrophoresis to nitrocellulose membranes. The membranes were blocked with 5% nonfat dry milk in Tris-buffered saline with 0.05% tween-20 for 2 hours. Membranes were then probed with antibodies that specifically recognize eNOS (1:1000; Abcam, Shanghai, China), eNOS-P Ser¹¹⁷⁷ (1:1000; Cell Signaling Technology, Shanghai, China), eNOS-P Ser⁴⁹⁵ (1:1000; Cell Signaling Technology), myocilin (1:500; Abcam), Akt (1:1000; Cell Signaling Technology), and Akt-P Ser⁴⁷³ (1:1000; Cell Signaling Technology) followed by incubation with peroxidase-linked secondary antibodies. Glyceraldehyde 3-phosphate dehydrogenase was used as a loading control. Signals in the linear range of the x-ray film were captured digitally and densitometry performed using Kodak Molecular Imaging Software (Kodak, Shinkawa, Japan). 

Immunohistochemistry

Paraffin sections of the enucleated CAV1 KO and WT mouse eyes were stained for CAV2, endothelial nitric oxide synthase ser1177 (eNOS ser1177), and endothelial nitric oxide synthase Thr495 (eNOS Thr495). After deparaffinization of the sections, endogenous peroxidase activity was neutralized by incubating sections in 1% H₂O₂ in methanol for 20 minutes and washing with PBS. Nonspecific binding sites on cells were blocked in 1% bovine serum albumin solution for 1 hour at room temperature and permeablized with 0.2% Triton-X100 for 5 minutes. Sections were then incubated with CAV2 (rabbit polyclonal, 1:200; Abcam), eNOS (rabbit polyclonal, 1:100; Abcam), eNOS-P Ser¹¹⁷⁷ (rabbit polyconal, 1:100; Santa Cruz, Shanghai, China), and eNOS-P Thr⁴⁹⁵ (rabbit monoclonal, 1:100; Santa Cruz), AKT (1:50; Cell Signaling Technology), Akt-phospho Ser⁴⁷³ (1:50; Cell Signaling Technology) immunoglobulin G (IgG) overnight at 4°C. Sections were washed three times in PBS (10 minutes each) then incubated with rabbit antigoat biotin conjugated secondary antibodies (1:1000 dilution; Abcam) in PBS for 30 minutes at room temperature. Sections were rinsed in PBS three times (10 minutes each) and incubated in streptavidin-biotin solution (1:500 dilution; Invitrogen, Shanghai, China) in PBS for 30 minutes at room temperature. After rinsing in PBS three times (5 minutes each), the sections were incubated in a diaminobenzidine-enhanced substrate system (Sigma-Aldrich, Shanghai, China) for 5 minutes and rinsed three times again in PBS. 

Topical Drug Application

Mouse eyes were treated topically with NO donors and an NOS inhibitor following an established method with slight modifications.^20,21 One eye was treated with SNP (4 × 2 μL drops, total dose 160 μg), SNAP (4 × 2 μL drops, total dose 160 μg), and L-NAME (4 × 2 μL drops, total dose 400 μg; Sigma-Aldrich) by topical application at 0, 0.5, 1, and 1.5 hours. The contralateral eye was treated with the drug vehicle. For all eye drop administrations, the drug was administered to the central cornea. The contralateral eyes were treated with the drug vehicle diluted in PBS. IOP was measured in both eyes before drug treatment and three hours after the last drug treatment. 

Statistics

Data normality was tested first. For IOP measurement in the WT and KO mice, the data were not normally distributed, so it was analyzed by the Mann-Whitney U test (SPSS 16 for Windows; SPSS, Chicago, IL, USA). In the experiments involving drug treatments, IOP and outflow data were normally distributed, and they were analyzed by paired-sample t-tests. In all cases, differences were considered significant at P < 0.05. 

Results

The aim of this study was to determine the impact of CAV1 deficiency on eNOS activity and drug response to NOS donor and inhibitor. To accomplish our aim, we studied CAV1 KO mice as an experimental model. We first examined IOP in CAV1 KO mice and WT littermates using rebound tonometry. It was confirmed that IOP was significantly higher in the KO mice (19.6 [1.5] mm Hg, mean [standard deviation, SD], n = 27) when compared with the WT group (12.0 [0.5] mm Hg, mean [SD], n = 19, P < 0.05, Fig. 2). 

Figure 2

 

IOP of CAV1 KO and WT mice. (A) Individual IOP readings in KO and WT mice. (B) Comparison of mean IOP in KO and WT mice. Error bars are standard deviation of the mean. CAV1 KO, n = 27; WT, n = 19. *P < 0.05.

Outflow Facility in CAV1 KO Mice

We then examined if and how CAV1 deletion impacted aqueous humor outflow profile when compared with WT mice.^17–19 The flow rate data was plotted as a function of IOP, and the data were compared between WT and KO mice (Fig. 3A). In KO mice, the outflow rate was 0.060 (0.030), 0.118 (0.028), 0.177 (0.035), and 0.233 (0.046) μL per minute at 5, 12, 19, and 27 mm Hg, respectively (mean [SD], n = 9). In WT mice, the outflow rate was 0.056 (0.026), 0.169 (0.032), 0.252 (0.049), and 0.350 (0.036) μL per minute at 5, 12, 19, and 27 mm Hg, respectively (n = 9). The outflow rate of KO mice was significantly lower than controls at 19 mm Hg and 27 mm Hg (P < 0.05). 

Figure 3

 

Relation between the mean flow rate and IOP in all perfused WT and CAV1 KO mouse eyes. Mouse eyes were perfused at sequential pressures of 5, 12, 19, and 27 mm Hg. (A) The outflow rate and pressure relationship of WT and CAV1 KO mice. The slope of the regression line (solid line) is the conventional outflow facility (C_con). (B) Comparison of C_con of WT and KO mice that C_con of KO mice is significantly lower than WT mice. CAV1 KO, n = 9; WT, n = 9. *P < 0.05.

To determine the conventional outflow facility of the CAV1 KO mouse eyes, a linear regression was performed on the flow rate versus the pressure data. The conventional outflow facility (C_con, the slope of the flow rate versus the pressure graph in Fig. 3A) for the KO and WT mice was 0.010 (0.001) μL per min/mm Hg and 0.015 (0.002) μL per min/mm Hg (mean [SD]), respectively. C_con of KO mice was 33% lower than WT controls (P < 0.05; Fig. 3B). 

CAV2 Expression

To assess the expression of CAV2 on CAV1 KO mice, we stained mouse eye sections with a CAV2 antibody visualized by bright field microscopy (Fig. 4). Staining with the omission of primary antibody resulted in no labeling (negative control [NC], Fig. 4A). The positive control was WT mouse retina, which expresses CAV2 abundantly (Fig. 4B). In CAV1 KO mice, the CAV2 protein was absent (Fig. 4C). In comparison, the WT mice expressed CAV2 in the TM and SC tissues (Fig. 4D). 

Figure 4

 

Expression of CAV2 in CAV1 KO mice. NC is the antibody negative control where the primary antibody was omitted (A), and PC is positive control of the mouse retina (B). CAV2 is expressed abundantly in WT mice (C), whereas the expression of CAV2 is absent in CAV1 KO mice (D). Western blot confirmed this finding. NC is the antibody negative control where the primary antibody was omitted, and PC is positive control of the mouse retina. Scale bar: 30 μm.

eNOS Activity

Next, the expression of eNOS and its phosphorylated proteins, Akt, and its phosphorylated proteins and tyrosine nitration were investigated using Western blot (Fig. 5). The outflow tissue dissected expressed myocilin, which is a marker for TM cells (Fig. 5A). There was no significant difference in myocilin expression in WT and CAV1 KO mice. In CAV1 KO and WT mice, the expressions of eNOS, eNOS-phospho Ser¹¹⁷⁷, Akt, and Akt-phospho Ser⁴⁷³ were higher in CAV1 KO mice than in WT mice (Figs. 5B, 5C). In contrast, the expression of eNOS-phospho Thr⁴⁹⁵ was significantly reduced in CAV1 KO mice. Immunohistochemistry confirmed the expression of these proteins, including eNOS, eNOS-phospho Ser¹¹⁷⁷, eNOS-phospho Thr⁴⁹⁵, Akt, and Akt-phospho Ser⁴⁷³ (Fig. 6). 

Figure 5

 

Western blot analysis of eNOS, Akt, and their phosphorylated protein levels in CAV1 knockout mice. (A) Outflow tissue expresses TM marker myocilin in WT and CAV1 KO mice. (B) Representative blots showing comparison of protein and phosphorylation levels in WT and KO mice. (C) Densitometric analyses of Western blots. eNOS-P ser¹¹⁷⁷ and eNOS Thr⁴⁹⁵ (normalized to total eNOS), Akt-P (normalized to Akt) from WT, and CAV1 KO mouse aqueous humor outflow tissue. Error bar shows SD. *P < 0.05, n = 6.

Figure 6

 

Immunofluorescence staining of eNOS, AKT, and their phosphorylated proteins in CAV1 KO mice. eNOS (A, B), eNOS-Ser1177(C, D), eNOS-Thr495 (E, F), Akt (G, H), Akt-P (I, J) are expressed in KO and WT mouse TM and Schlemm's canal endothelium. NC is antibody negative control where the primary antibody was omitted (n = 6). Scale bar: 30 μm.

Tyrosine Nitration

Western blot analysis and immune staining demonstrated increased tyrosine nitration of proteins in the CAV1 KO mouse outflow tissue (Fig. 7). Nitrotyrosine is a surrogate measure of peroxynitrite, and increased tyrosine nitration indicated the formation of peroxynitrite in Cav1 KO mouse outflow tissue. 

Figure 7

 

Increased tyrosine nitration of proteins in Cav1 KO mice. (A, B) Representative blots and densitometry analyses of the mouse outflow tissue in WT and CAV1 KO mice. (C) Immunohistochemistry staining of nitrotyrosine. Scale bar: 30 μm.

IOP Response to NO Donors and NOS Inhibitor

We then examined if and how CAV1 deletion affected the IOP response to the NO donors SNP and SNAP. For SNP-treated eyes, IOP was significantly lower than contralateral vehicle-treated eyes by 12% (mean [SD], 16.7 [1.0] mm Hg versus 14.7 [1.1] mm Hg, n = 6, P < 0.05; Fig. 8A). The magnitude of reduction by SNP application was similar to the WT control mice, which was 17% (mean [SD], 11.2 [1.2] mm Hg and 9.3 [1.0] mm Hg, n = 6, P < 0.05; Fig. 8B). SNAP did not have a significant effect on mouse eye IOP either in CAV1 KO mice (mean [SD], 16.0 [1.0] mm Hg versus 14.7 [1.4] mm Hg in vehicle-treated and SNAP-treated eyes, respectively, n = 6, P > 0.05; Fig. 8C) or in WT control mice (10.7 [0.7] mm Hg versus 10.4 [0.7] mm Hg, n = 6, P > 0.05; Fig. 8D). 

Figure 8

 

IOP of CAV1 KO (A, C, E) and WT (B, D, F) mice treated with topical applications of NO donors SNP and SNAP, and NOS inhibitor L-NAME. SNP significantly reduced IOP when compared with vehicle-treated eyes in both KO (A) and WT mice (B); however, SNAP did not significantly affect IOP (C, D). Eyes treated with the NOS inhibitor L-NAME showed significantly higher IOP when compared with vehicle-treated eyes in both KO (E) and WT (F) mice.

NOS inhibitor L-NAME significantly increased IOP in CAV1 KO mice by 136% (14.5 [1.2] mm Hg versus 19.7 [3.0] mm Hg, n = 10; Fig. 8E). In WT mice, L-NAME significantly increased IOP by 120% (from 10.3 [1.0] mm Hg to 12.0 [1.0] mm Hg, n = 6, P < 0.05; Fig. 8F). 

The Effect of NO Donor and NOS Inhibitor on Aqueous Outflow

Following IOP measurements, we further investigated the effects of NO donors on outflow rate and facility in CAV1 KO and WT mice. With the NO donor SNP, the outflow rates at 5, 12, 19, and 27 mm Hg were 0.143 (0.046), 0.303 (0.105), 0.457 (0.123), and 0.663 (0.053) μL per minute in CAV1 KO mice, respectively. SNP-treated eyes showed a linear relationship in the flow-pressure graphs (Fig. 9A). The conventional outflow facility (C_con) in SNP-treated and control eyes was 0.013 (0.006) and 0.021 (0.007) μL per min/mm Hg, respectively, which was an increase by 1.6 folds (Fig. 9B, P < 0.05). Similarly, SNAP significantly increased C_con from 0.013 (0.004) to 0.021 (0.005) μL per min/mm Hg, which was by 1.5 folds (n = 7, P < 0.05; Figs. 9C, 9D). 

Figure 9

 

CAV1 KO eyes perfused with NO donor SNP, SNAP, and NOS inhibitor L-NAME. (A) Comparison of flow rate in SNP-treated eyes and paired control eyes perfused with the drug vehicle (PBS). The outflow rate was significantly higher in SNP-perfused eyes (n = 11, *P < 0.05, error bars are mean ± SD). (B) Conventional outflow facility was significantly higher in SNP-treated CAV1 KO mice when compared with vehicle controls (*P < 0.05, n = 11). (C, D; E, F) Similarly shown outflow rate data and conventional outflow facility comparison for SNAP and L-NAME. In the box plots, the bottom and top of the box is the first and third quartiles, and the band inside the box is the second quartile (the median). The ends of the whiskers are the minimum and maximum values.

The outflow data of each pair of eyes treated with L-NAME and the vehicle control is illustrated in Figures 9E and 9F. C_con was significantly lower in L-NAME–treated eyes (0.006 [0.002] μL per min/mm Hg) when compared with PBS-treated controls (0.012 [0.003] μL per min/mm Hg, P < 0.05, n = 5; Figs. 9E, 9F), which was a reduction of 49%. 

Discussion

Our goal was to investigate eNOS activity and the response of conventional outflow facility to NO donors and a NOS inhibitor in CAV1 KO mice. We found that eNOS, its phosphorylated form eNOS-P Ser1177, and Akt were significantly upregulated in CAV1 KO mice when compared with WT control mice. However, a CAV1 deficiency did not affect the capacity to respond to NO donors SNP and SNAP when compared with WT mice. 

Our data confirmed that CAV1 KO mice had significantly increased IOP and lower conventional outflow facility when compared with the WT mice (Figs. 2, 3). This is consistent with a study performed earlier in which Elliot et al. observed a significant elevation in IOP in Cav-1 KO mice at 5 weeks of age and increased IOP sustained at 3 and 6 months (Elliott MH, et al. IOVS 2014;55:ARVO E-Abstract 2888). However, we cannot conclude the contribution of unconventional outflow to IOP elevation in KO mice from our data because this was not measured in the study. 

CAV1 KO diminished CAV2 protein expression in these mice (Fig. 4). The caveolin gene family has three members (CAV1, CAV2, and CAV3). CAV1 and CAV2 are normally coexpressed and form a hetero-oligomeric complex in many cell types, including type I pneumocytes, endothelia, fibroblastic cells, and adipocytes.^22,23 Both of them have been linked to the pathogenesis of glaucoma. The CAV1 and CAV2 hetero-oligomeric dimers were thought to drive caveolae formation in vivo, and CAV1 was known to stabilize CAV2 protein levels.^24,25 The absence of CAV2 in CAV1 KO mice is consistent with other reports that CAV2 was trapped at the Golgi and targeted to the proteasomal degradation pathway in the absence of CAV1.²⁶ The knockout of CAV1 diminished CAV2 expression in mouse eye outflow tissue, which was consistent with previous findings in pulmonary vasculature.²⁷ 

Our results and several other studies lend support to a role for CAV1 and CAV2 in the pathogenesis of elevated IOP. CAV1 and CAV2 genes are expressed in glaucoma and IOP regulation-related tissues, for example, TM and SC.^12,28 Chen et al. found that the CAV1 and CAV2 genes were significantly associated with IOP.²⁹ CAV1 and CAV2 are also known to negatively regulate transforming growth factor beta signaling,^30,31 which has been implicated in the pathogenesis of POAG.³² This is consistent with the hypertensive glaucoma risk-associated allele of rs17588172 and confers decreased expression of CAV1 and CAV2. However, a study reported that CAV1 and CAV2 knockdown have differential effects on aqueous outflow.¹² Outflow rates increased significantly in CAV1-silenced anterior segments, whereas outflow rates decreased in CAV2-silenced anterior segments.¹² Further studies are necessary to understand the mechanisms of CAV1 and CAV2 in regulating aqueous humor dynamics. 

CAV1 KO resulted in an increase in eNOS activity in the eye (Figs. 5, 6). eNOS had multiple phosphorylation sites. Our data showed that CAV1 KO mice had higher expression levels of eNOS, eNOS-P Ser¹¹⁷⁷, Akt, and Akt-P Ser⁴⁷³ when compared with WT control mice, and a reduction of eNOS-P Thr⁴⁹⁵. This is consistent with the mechanism that eNOS and CAV1 interaction results in eNOS activity inhibition.³³ In addition, when the CAV1 gene is absent, eNOS expression was upregulated. 

It is interesting to note that although we saw increased eNOS expression, CAV1 KO had increased IOP and reduced conventional outflow when compared with WT mice. The Western blot results seem to imply that in the absence of CAV1, eNOS maintains an active state. Incidentally, CAV1 KO mice also exhibit pulmonary hypertension.³⁴ These mice had a marked increase in plasma NO levels³⁵ and enhanced eNOS activation in vessels from CAV1 KO mice.^36,37 Surprisingly, mice with genetic deletions of CAV1 and NOS3 double knockout surprisingly did not develop pulmonary hypertension.³⁵ NO reacts with superoxide to form the damaging reactive nitrogen species peroxynitrite, which modifies proteins and may interfere with their function through tyrosine nitration.³⁸ Zhao et al. demonstrated that the activation of eNOS in CAV1^–/– lungs led to the impairment of protein kinase G activity through tyrosine nitration.³⁵ These data may have suggested that persistent eNOS activation secondary to CAV1 deficiency resulted in tyrosine nitration of protein kinase G and impairment of its activity, which thereby induced pulmonary hypertension.³⁵ The increased IOP in CAV1 KO mice could also be a result of a similar mechanism because tyrosin nitration was higher in the mouse outflow tissue (Fig. 7). This is consistent with a study of the expression of nitrotyrosine in the human TM of POAG patients. It demonstrated that the presence of oxidant peroxynitrite in uveal, corneoscleral meshwork, and juxtacanalicular tissues indicated long-term exposure to peroxynitrite in TM from patients with severe POAG.³⁹ Oxidant peroxynitrite can cause cell damage and death by tyrosine nitration and the formation of nitrotyrosine.⁴⁰ 

NO donor SNP decreased the IOP, whereas the NOS inhibitor L-NAME increased the IOP in CAV1 KO mice (Fig. 8). This finding is consistent with the results from outflow facility, that is, NO donor increased outflow facility, whereas NOS inhibitor decreased outflow facility. With the exogenous NO released by SNP, it may stimulate sGC, which leads to elevation of intracellular cGMP levels and then cGMP interacts with various cyclic-nucleotide–gated channels, protein kinases, and protein phosphodiesterases. The NO-cGMP signaling pathway plays a role in regulating aqueous humor outflow.^41,42 Our data revealed that without the interaction with CAV1, eNOS and its downstream pathways were still intact and functional. NO donors seemed to increase conventional outflow targeting both TM and SC cells. Although both TM⁴³ and SC cells⁴⁴ are responsive to NO, recent evidence seems to point to a primary role of SC cells as the cellular targets for NO in IOP lowering. Ashpole et al.⁴⁵ showed that shear stress increased NO levels by 8-fold after 7 days of perfusion. This is comparable with human umbilical vein endothelial cells, and greater than TM cells, which appeared shear insensitive. 

The magnitude of IOP reduction by SNP was smaller in CAV1 mice than that of WT mice, that is, 12% in CAV1 KO mice versus 33% in WT mice. This is similar to a previous study in the mouse small intestine, in which CAV1 gene knockout caused impaired NO function, and the responsiveness to the exogenous NO donors SNP and SNAP were reduced when compared with control.^46,47 

In our study, the topical application of SNP reduced IOP, whereas SNAP did not have a significant IOP-lowering effect. This is consistent with findings in a monkey study²¹ and our previous findings in eNOS KO and WT mice. However, our data confirmed that both SNP and SNAP induced a greater pressure-dependent drainage in both WT and CAV1 KO mouse eyes when compared with vehicle-treated eyes (Fig. 9). The reason for the discrepancy in IOP and conventional outflow facility results is not known. IOP measurement was conducted in live animals, whereas the outflow facility was measured in enucleated perfused eyes. It could also be a result of poor cornea penetration of SNAP. For topical application of drugs, the epithelium appears to be rate limiting to drug movement for hyrdophilic compounds, whereas the stroma is rate limiting for hydrophobic compounds. SNAP was soluble in dimethyl sulfoxide (57.5 mg/mL) and in water (2.1 mg/mL), with a partition coefficient of around 0.9.⁴⁸ The hydrophilicity of the drug might have prevented it from penetrating the corneal epithelium. 

NO can influence IOP by targeting the conventional and unconventional outflow pathways, although it predominantly acts on the TM and SC. In the conventional outflow pathway, NO decreases the volume of TM^49,50 and SC cells,^44,50 which correlated with increased outflow facility. It can also decrease the contractility of TM cells, a process known to increase outflow facility.⁵¹ Previous studies showed that NO-donating compounds significantly increased conventional outflow and lowered IOP in rabbits,^42,52–54 pigs,⁵⁵ dogs,⁵² monkeys,^21,52 and perfused postmortem human eyes.^56,57 In the uveosclera pathway, isosorbide dinitrate (ISDN), isosorbide mononitrate (ISMN), SNP, and SNAP have been shown to relax isolated bovine ciliary muscles at resting tension, although to a lesser extent than isolated TM strips.⁴³ 

In conclusion, although CAV1 KO mice had elevated IOP and decreased outflow facility, CAV1 deficiency (and possibly the loss of CAV2) resulted in increased eNOS activity. The pressure elevation may be because of increased tyrosine nitration of protein kinase K and impairment of its activity in KO mice. 

Acknowledgments

The authors thank Wang Yiqiang, PhD, for his generosity in sharing animal tissue and Shi Jufang for her excellent assistance with the animal work. We are grateful for the insightful comments of W. D. Stamer, PhD, and C. R. Ethier, PhD, on the manuscript. Supported by the National Science Foundation China (81100662, 81371015), the Shanghai Municipal Health Bureau Young Outstanding Scientist Program (XYQ2013083), the 211 Project of Fudan University (EHF158351), and the Scientific Research Foundation for the Returned Overseas Chinese Scholars (State Education Ministry). 

Disclosure: Y. Lei, None; M. Song, None; J. Wu, None; C. Xing, None; X. Sun, None