Endogenous Regulation of Human Schlemm's Canal Cell Volume by Nitric Oxide Signaling

Abstract

Purpose.: There is a time-course correlation between nitric oxide (NO)-induced decreases in trabecular meshwork (TM) cell volume and NO-induced increases in outflow facility. The Schlemm's canal (SC) cells may also provide resistance to aqueous humor outflow; therefore, this study tests the involvement of the nitric oxide synthase (NOS) and NO signaling pathway and the BK_Ca-channel in mediating SC cell volume decreases.

Methods.: Cell volume was measured in low-passage human SC cells using calcein AM fluorescent dye; images were captured with a confocal microscope, and data were quantified using NIH ImageJ software.

Results.: Inhibition of endogenous NOS resulted in a 7% increase in SC cell volume. Exposure of SC cells to DETA-NO resulted in a 12% to 16% decrease in cell volume that was abolished by the soluble guanylyl cyclase (sGC) inhibitor 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one (ODQ) (5 μM), the protein kinase G (PKG) inhibitor (RP)-8-Br-PET-cGMP-S (50 μM), and the high-conductance calcium-activated potassium channel (BK_Ca channel) inhibitor iberiotoxin (50 nM). Hypertonic media significantly decreased SC cell volume by 14%, whereas hypotonic media significantly increased cell volume by 11.2%.

Conclusions.: These data suggest that endogenous NOS regulates steady state cell volume and the involvement of the NOS/NO/sGC/cGMP/PKG pathway and the BK_Ca-channel in mediating NO-induced reductions in SC cell volume. These decreases in cell volume correlated with the time-course for NO-induced increases in outflow facility, suggesting that the NO-induced reduction in SC cell volume may also influence outflow facility.

The major route for the outflow of aqueous humor is the trabecular meshwork (TM), ^{1 –3} composed of uveal and corneoscleral TM ¹ and juxtacanalicular cells, ⁴ in conjunction with Schlemm's canal (SC). Aqueous humor also exits the eye by way of the uveoscleral outflow pathway, ⁵ moving in a pressure-independent manner through the spaces between ciliary muscle bundles and to the ocular venous system. The SC is composed of the endothelial monolayer of the inner and outer walls. The inner wall of the SC is characterized by giant vacuoles ^{6 –8} that create small spaces between the extracellular material of the juxtacanalicular TM and the inner wall of the SC ⁹ and by transendothelial pores, both of which are the main pathways for the passage of aqueous humor. ^8,10 SC endothelial cells are joined together by tight junctions that allow the joining together of the endothelial cells with their basal laminae and with juxtacanalicular TM cells. ^{11 –13} Both the juxtacanalicular TM region and the inner wall of the SC are thought to be the areas of resistance to the outflow of aqueous humor (see Ref. 14 for review), though the degree to which resistance is generated in each area is controversial. 

Intraocular pressure (IOP) is tightly controlled and maintained by a balance between aqueous humor production by the ciliary body and the egress of fluid from the anterior chamber through the TM/SC and through the uveoscleral pathway. Proper regulation of IOP in the eye has important physiological implications, whereas improper regulation can lead to pathologic consequences. Increased resistance to aqueous humor outflow results in elevation of IOP, which, if left untreated, can lead to progressive retinal ganglion cell damage, resulting in the blinding disease glaucoma. 

The human SC is enriched with the NO-producing enzyme NOS-III (eNOS), ¹⁵ and NADPHd-positive nerve terminals densely innervate this region. ¹⁶ When NO is applied topically to rabbit eyes, IOP is reduced without systemic effects. ¹⁷ NO is thought to cause decreases in IOP by increasing the outflow of aqueous humor, ^{18 –20} thus decreasing resistance to outflow; however, the cellular and biochemical mechanisms are unknown. 

Cellular mechanisms, including swelling, shrinkage, or both of the cell volume of the TM cells, ^{18,21 –29} have been suggested to regulate aqueous humor outflow. In fact, the time-course for NO-induced increases in aqueous humor outflow facility correlate with NO-induced decreases in TM cell volume. This suggests a functional role for changes in cell volume in regulating aqueous humor outflow facility and that the NO-induced response results in decreased resistance to aqueous humor outflow. We have recently reported functional interactions between the NO system and the BK_Ca channel; that activation of the BK_Ca channel is obligatory for the NO-induced reduction in TM cell volume. ²⁸  

Many of the cell volume studies were performed in TM cells, in part because isolation of SC cells is tedious and time consuming and because it is difficult to predict the isolation yield of cells. Recent studies using electron probe x-ray microanalysis in outflow pathway tissue determined that cells from the inner wall of the SC experienced a regulatory volume decrease in response to hypotonic challenge. ³⁰ Because resistance to aqueous humor outflow also resides in the inner wall of the SC, we hypothesized that endogenous NO regulates SC cell volume. Additionally, we tested the involvement of soluble guanylyl cyclase (sGC), cGMP, protein kinase G (PKG), and the BK_Ca channel in the NO-induced decreases in SC cell volume. We also tested SC cell responses to changes in osmolarity. 

Methods

Cell Culture

SC cells were obtained from whole human donor eyes purchased from the Lions Eye Institute (Tampa, FL) within 24 to 30 hours of death or from corneal scleral rims that were stored in corneal storage medium (Optisol; Chiron Ophthalmics, Irvine, CA) at 4°C. All human tissue was handled in accordance to the tenets of the Declaration of Helsinki. Standard ophthalmic microsurgery instruments were used to bisect the eyes and remove the cornea, iris, lens, and ciliary body. SC cells were cultured by the cannula-derived method. ³¹ Briefly, corneal scleral rims were cut into six symmetrical pieces. Using a dissecting microscope and fine tweezers, gelatin-coated sutures were placed in SC through the lumen of the SC and were maintained in culture medium containing low-glucose (1 g/L) Dulbecco's modified Eagle medium (DMEM) in the presence of 10% fetal bovine serum, 100 U/mL penicillin, and 100 μg/mL streptomycin for 4 weeks in a tissue culture incubator at 37°C in 5% CO₂. Before the sutures were removed from the tissue, the tips of the sutures extending beyond the wedges were cut to remove any contaminants. Sutures from each eye wedge were removed and transferred to one well of a six-well plate and allowed to grow to confluence in culture medium containing low-glucose (1 g/L) DMEM in the presence of 10% fetal bovine serum, 100 U/mL penicillin, and 100 μg/mL streptomycin in tissue culture incubator at 37°C in 5% CO₂. Cells were then passaged in culture flasks, and low-passage cells (2–4) were used in our experimental protocols. SC cells are distinguished from cultures of TM cells by the absence of induction of myocilin after dexamethasone treatment (Fig. 1A), their morphologic differences (Fig. 1B), and the generation of transendothelial electrical resistance of ≥10 Ωcm² after 1 week at confluence ^31,32 (Table 1). Human SC cells and TM cells (numbers represent the ages of the human donors) were grown on filters (Nalge Nunc International, Rochester, NY), as described, and transendothelial resistance was measured in accordance with the manufacturers' protocol (EVOM² Voltohmeter; World Precision Instruments, Sarasota, FL). Data are expressed as Ωcm² and represent the difference (Difference) between the resistance of the cells (Readings) and the resistance of the blank (Blank) filter (Table 1). 

Figure 1.

 

(A) Dexamethasone-induced myocilin expression in TM cells. TM and SC cells were incubated with or without dexamethasone (100 nM) for 7 days, after which they were incubated with anti-myocilin antibody and appropriate secondary antibody. This figure is representative of the TM cell lines developed and demonstrates that exposure to dexamethasone results in increased myocilin expression (+Dex). TM cells not treated with dexamethasone did not express myocilin (−Dex). Additionally, cells incubated with dexamethasone were exposed to secondary antibody only (−Ab) to control for nonspecific binding. Myocilin expression was not observed in SC cells (data not shown). (B) Morphologic differences between TM and SC cells. Arrows: morphologically different cells.

Table 1.

 

Transendothelial Electrical Resistance (Ωcm²)

Cells	Readings (Ω)	Blank (Ω)	Difference (Ω)
SC14	84	72	12
SC25	91	70	21
SC24	84	70	14
TM84	77	72	5

Human TM cells were validated by the presence of dexamethasone-induced myocilin expression (anti-myocilin antibody generously provided by Daniel Stamer, University of Arizona). TM and SC cells were incubated with or without dexamethasone (100 nM) for 7 days, after which they were incubated with anti-myocilin antibody and appropriate secondary antibody. Figure 1A demonstrates that exposure to dexamethasone results in increased myocilin expression. TM cells not treated with dexamethasone did not express myocilin. SC cells that were exposed to dexamethasone did not express myocilin (data not shown). 

For experimental protocols, SC cells (SC36 generously provided by Daniel Stamer; SC14, SC25, and SC24 developed in our laboratory) were grown on chambered coverglasses (Laboratory-Tek II; Nalge Nunc International) in low-glucose DMEM, as described. To eliminate drug effects that might have resulted from growth factors in the serum, cells were exposed to serum-free medium for 1 day before the experiments. 

Measurement of Cell Volume

Cell volume measurements were performed as previously described ¹⁸ using the modified protocols of Mitchell et al. ²⁷ and Bush et al. ^33,34 Cells were imaged using calcein AM, and confocal microscopy was used to calculate the number of voxels. Cell volume was quantified using ImageJ software (developed by Wayne Rasband, National Institutes of Health, Bethesda, MD; available at http://rsb.info.nih.gov/ij/index.html). Before any drug treatments, the cells were loaded with the fluorescent dye calcein AM in DMEM at 37°C, in 5% CO₂ incubator for 60 minutes to ensure a stable baseline. Coverslips containing the cells were subjected to confocal microscopy with a Leica (Wetzlar, Germany) confocal microscope. 

We developed a technique of drug delivery to the SC cells on the coverslips to ensure that the slides did not shift during imaging and that images would be taken of the same cells. Specifically, several ports were drilled in the covers of the glass chambers. Tubes attached to syringes were inserted into each port allowing for the exchange of media and drugs. Images were taken without drug treatment (the 0-minute time point); this served as the experimental control. Drugs were then added to the cells, with care taken not to shift the coverslip. Images were taken of the same cells at varying time periods. Additionally, images were taken of cells that were not exposed to drugs for the time periods indicated; this served as a control for evaluating the stability of the dye. 

For our experiments, the microscope captured a 1024 × 1024 pixel image with 8 bits of resolution. The confocal microscope captures images in z-stacks, which are a series of images captured at different equally spaced depths. A pixel represents a two-dimensional area, whereas a voxel represents a three-dimensional volume. The size of a voxel is determined for each image stack by the area of the pixel (in μm², as given by the Leica confocal software) multiplied by the separation of the images in the stack (in μm; in these experiments, z-section separation =1 μM). ImageJ software was then used to determine which images in the image stack represented the topmost and bottommost section of the cell. All images that represented areas above and below the cell were discarded. The images were converted from 8-bit to 1-bit or binary. A region of interest was then selected, and the ImageJ software was used to calculate the number of voxels in the region of interest in the image stack. Cell volume was determined by multiplying the number of voxels by the size of 1 voxel. We determined threshold levels using fluorescent latex beads (Fluoresbrite; Polysciences Inc., Warrington, PA) of known diameter and volume that were imaged under conditions identical with those used for TM cells (Fig. 2). 

Figure 2.

 

NO decreases SC cell volume. Threshold z-stack images of an SC cell. At 0 minutes (without drug), the threshold voxels were qualitatively and quantitatively greater than at 20 minutes after DETA-NO (100 μM) treatment. Voxel count for cell: 0 minute: 11,606; 20 minutes: 7,435. Scale bar, 20 μm.

To assess the functioning of SC cell volume regulatory mechanisms, the osmolarity of the medium was changed. Hypertonic medium was made by the addition of 150 mM mannitol to DMEM (∼469 mOsm/kg), and hypotonic medium was made by the addition of 30% deionized water to DMEM water (∼208 mOsm/kg). 

Materials and Reagents

Routine reagents and iberiotoxin (IBTX) were purchased from Sigma (St. Louis, MO). Other reagents were obtained as follows: sodium nitroprusside (SNP), 8-Br-cGMP sodium salt, 1H-[1,2,4] oxadiazolo[4,3-a]quinoxalin-1-one (ODQ), and diethylenetriamine nitric oxide (DETA-NO) from σ-RBI (Natick, MA); L-N^G-nitroarginine methyl ester (hydrochloride) (L-NAME) Cayman Chemicals (Ann Arbor, MI); and (RP)-8-Br-PET-cGMP-S from Calbiochem (La Jolla, CA). 

Statistical Analysis

Statistical analysis was performed using ANOVA, followed by Holm-Sidak and Fisher LSD method for comparison of significant difference among different means. 

Results

Inhibition of Endogenous NOS Increases SC Cell Volume

SC14, SC24, and SC25 cells were incubated with calcein AM and then exposed to L-NAME (50 μM), the NOS nonspecific inhibitor. Cells were imaged, and changes in cell volume were recorded. In all cell lines, voxel counts revealed that L-NAME caused statistically significant increases in SC cell volume. Figure 3 demonstrates that there was a 7% increase in cell volume 10 minutes after L-NAME treatment. 

Figure 3.

 

Inhibition of endogenous NOS increases SC cell volume. L-NAME increases SC cell volume. Confocal images of the same cells were acquired with a 20× objective lens at 1-μm z-step intervals to a depth of 15 μm. Data shown represent the mean ± SEM for 68 cells (derived from SC14, SC24, and SC25), expressed as percentage of initial volume at the 0-minute time point without drugs. *P < 0.05, significantly different from control.

NO Decreases SC Cell Volume

Previous studies demonstrated that the NO donor DETA-NO (100 μM) increased outflow facility in porcine anterior eye organ perfusion in a time course that correlated with DETA-NO–induced decreases in TM cell volume. ¹⁸ Aqueous humor exits the eye through the TM and SC, and both areas are thought to create resistance to the exit of aqueous humor from the eye. Therefore, because we used DETA-NO as a NO donor in the previous studies, we initially used DETA-NO to test the effects of NO on SC cell volume in these studies. SC14, SC24, and SC25 cells were incubated in calcein AM dye and imaged at the 0-minute time point (control). The cells were then exposed to varying concentrations of DETA-NO (1–100 μM), and images of the same cells were captured to determine the saturating concentration(s) of the agent. Figure 4A demonstrates that though 1 and 10 μM DETA-NO had no effect on SC14 and SC24 cell volume, 50 and 100 μM were saturating. Although SC14 and SC24 cells responded to DETA-NO treatment, SC25 cells did not. We were surprised at the lack of response of SC25 cells to DETA-NO and wanted to determine whether this was due to donor variability. We then tested whether SC25 cells would respond to another NO donor and used the NO donor sodium nitroprusside (SNP). ³⁵ Cells were preincubated with calcein AM, then exposed to SNP (10 and 100 μM), and images were captured as previously described. Figure 4B demonstrates that SNP caused a statistically significant decrease in SC25 cell volume. 

Figure 4.

 

NO-induced decreases in cell volume are concentration dependent. (A) SC cells were exposed to varying concentrations of DETA-NO (10–100 μM). Images were captured at the 0- and 20-minute time points. Data shown for control, 10, 50, and 100 μM DETA-NO represent the mean ± SEM of 18, 19, 21, and 18 cells, respectively, and are expressed as percentage of initial volume at the 0-minute time point without drugs. Average voxel counts for control, 10, 50, and 100 μM DETA-NO are 9454 ± 926, 9634 ± 777, 6860 ± 545, and 8645 ± 834, respectively. *P < 0.05; significantly different from control ANOVA and the Holm-Sidak method. (B) SC cells were exposed to SNP (10 and 100 μM). Images were captured at the 0- and 20-minute time points. Data shown for control, 10, and 100 μM SNP represent the mean ± SEM of 30, 22, and 19 cells, respectively, and are expressed as percentage of initial volume at the 0-minute time point without drugs. Average voxel counts for control, 10, and 100 μM SNP are 17,881 ± 2052, 41,635 ± 8608, and 21,993 ± 1722, respectively. *P < 0.05; significantly different from control. ANOVA and the Holm-Sidak method.

SC Cells Respond to Changes in Osmolarity

To validate that the reductions in cell volume observed in SC cells in response to DETA-NO were not artifactual, the cell response to osmotic changes was assessed. SC14 and SC25 cells were incubated with calcein AM, as previously described, and stable baselines were established. Images were captured at the 0-minute time point in isotonic medium (control), after which the osmolarity of the medium was altered. Images of the same cells were then captured at 1-minute intervals for up to 10 minutes (the microscope captured 512 × 512-pixel images that allowed for an image to be captured within half a minute). Cell volume was altered in response to hypotonic and hypertonic media (Fig. 5). There was a 4.8% increase in SC cell volume 2 minutes after the application of hypotonic medium, with a maximum increase of 16% in SC cell volume 10 minutes after the extracellular environment was changed from isotonic to hypotonic. Exposure of SC cells to hypertonic medium resulted in a significant decrease (15%) in cell volume 1 minute after isotonic changes (Fig. 5). 

Figure 5.

 

Changes in osmolarity effect changes in SC cell volume. SC14 and SC25 cells were used. Hypotonic DMEM increased cell volume (mean ± SEM; n = 38 cells), whereas hypertonic DMEM decreased cell volume (mean ± SEM; n = 45 cells). There were no changes in cell volume in cells incubated in isotonic medium (mean + SEM; n = 59 cells, representing mean controls for hypotonic and hypertonic treatments). Data are expressed as percentage volume at the 0-minute time point; for hypotonic, hypertonic, and isotonic DMEM, they are 6994 ± 776, 3287 ± 245, and 5437 ± 279, respectively. *P < 0.05, significantly different from control (0-minute time point).

DETA-NO– and SNP-Induced Decreases in SC Cell Volume Involve sGC and PKG

Pharmacologic manipulations allowed us to determine the signaling pathway involved in the NO-induced decreases in SC cell volume. SC14, SC24, and SC25 cells were incubated in calcein AM, and a stable baseline was achieved. To determine the involvement of sGC, images of SC14 and SC24 cells were captured at the 0-minute time point (control) without drugs, then with DETA-NO (100 μM) in the presence or absence of ODQ (1 and 5 μM), an sGC inhibitor. To determine the involvement of sGC in the SNP-induced decreases in SC cell volume, images of SC25 cells were captured at the 0-minute time point (control) without drugs, then with SNP (100 μM) in the presence or absence of ODQ (5 μM). Figure 6A demonstrates that 1 μM ODQ partially inhibited the DETA-NO–induced decreases in SC cell volume, whereas 5 μM ODQ completely abolished the DETA-NO–induced decreases in SC cell volume. As with DETA-NO, ODQ abolished the SNP-induced decreases in cell volume (Figure 6C). The PKG-specific inhibitor (RP)-8-Br-PET-cGMP-S (25 μM) was used to assess the involvement of PKG. Figure 6B demonstrates that (RP)-8-Br-PET-cGMP-S abolished the NO-induced decreases in SC cell volume. As with DETA-NO, (RP)-8-Br-PET-cGMP-S (25 μM) abolished the SNP-induced decreases in SC cell volume (Fig. 6C). 

Figure 6.

 

Effects of ODQ and (RP)-8-Br-PET-cGMP-S on the NO-induced decreases in SC cell volume; SC14 and SC24 cells were used to test the DETA-NO effects, and SC25 was used to test the SNP effects. (A) ODQ abolishes the NO-induced decreases in SC cell volume. Data are expressed as percentage of initial volume at the 0-minute time point. For SC14 and SC24 cells: DETA-NO–treated group represents the mean ± SEM, n = 45 cells; DETA-NO + 1 μM ODQ-treated group represents the mean ± SEM, n = 40 cells; and DETA-NO + 5 μM ODQ-treated group represents the mean ± SEM, n = 42 cells. Voxel count for the 0-minute time point: DETA-NO, 10,821 ± 661; DETA-NO + 1 μM ODQ, 9686 ± 766; DETA-NO + 5 μM ODQ, 9812 ± 735. *P < 0.05, significantly different from the 0-minute time point. #P < 0.05, significantly different from DETA-NO–treated cells. (B) Effects of PKG inhibitor (RP)-8-Br-PET-cGMP-S (PKGi) on DETA-NO–induced decreases in SC cell volume. Cells were incubated with DETA-NO (100 μM) in the presence or absence of (RP)-8-Br-PET-cGMP-S (25 μM). Images were taken, and cell volume was measured. Data are expressed as percentage of initial volume at the 0-minute time point and represent the mean ± SEM for DETA-NO (n = 30 cells) and DETA-NO +PKGi (n = 29 cells). Voxel count for the 0-minute time point: DETA-NO, 8263 ± 697; DETA-NO + PKGi, 4744 ± 435. *P < 0.05, significantly different from the 0-minute time point. #P < 0.05, significantly different from DETA-NO–treated cells. (C) ODQ and (RP)-8-Br-PET-cGMP-S (PKGi) abolishes the SNP-induced decreases in SC cell volume. SNP-treated group represents the mean ± SEM (n = 40 cells). SNP + 5 μM ODQ-treated group represents the mean ± SEM (n = 15 cells). SNP + (RP)-8-Br-PET-cGMP-S (25 μM)-treated group represents the mean ± SEM (n = 25 cells). Voxel count for the 0-minute time point: SNP, 5307 ± 454; SNP + ODQ, 4758 ± 398; SNP + (RP)-8-Br-PET-cGMP-S, 5567 ± 448. *P < 0.05, significantly different from the 0-minute time point. #P < 0.05, significantly different from SNP-treated cells.

BK_Ca Channel Involvement

SC cells were preincubated with IBTX (100 nM), an inhibitor of the BK_Ca channel, and images were captured at the 0-minute time point (control). DETA-NO (100 μM) or SNP (100 μM) was then added to SC14 and SC25 cells, respectively, and images were captured at 20-minute time periods after DETA-NO or SNP exposure. Figures 7A and 7B demonstrate that IBTX alone had no effect on SC cell volume. However, IBTX abolished the DETA-NO–induced decreases in SC cell volume (Fig. 7A), and SNP-induced decreases in SC cell volume (Fig. 7B). 

Figure 7.

 

BK_Ca channels mediate the NO-induced decreases in SC cell volume. SC14 cells were used to test the DETA-NO effects, and SC25 was used to test the SNP effects. (A) Effects of IBTX on DETA-NO–treated cells. IBTX (100 nM) only: data are expressed as percentage of initial volume at the 0-minute time point and represent the mean ± SEM (n = 23 cells). DETA-NO: data are expressed as percentage of initial volume at the 0-minute time point and represent the mean ± SEM (n = 27 cells). IBTX + DETA-NO: data are expressed as percentage of initial volume at the 0-minute time point and represent the mean ± SEM (n = 11 cells). Voxel count for the 0-minute time point: IBTX, 19,959 ± 2755; DETA-NO 29,664 ± 2641; IBTX + DETA-NO, 16,357 ± 2218. *P < 0.05, significantly different from time 0. #P < 0.05 significantly different from DETA-NO–treated cells. (B) Effects of IBTX on SNP-treated cells. SNP: data are expressed as percentage of initial volume at the 0-minute time point and represent the mean ± SEM (n = 43 cells). IBTX + SNP: data are expressed as percentage of initial volume at the 0-minute time point and represent the mean ± SEM (n = 22 cells). Voxel count for the 0-minute time point: SNP, 13,773 ± 2490; SNP + IBTX, 19,293 ± 1842. *P < 0.05, significantly different from the 0-minute time point. #P < 0.05, significantly different from SNP-treated cells.

Discussion

This study demonstrated that in SC cells, the inhibition of endogenous NOS activity resulted in the cell volume being increased. Exogenously applied NO, derived from NO donors, decreased SC cell volume, and these decreases were dependent on the activation of sGC, PKG, and the BK_Ca channel (Fig. 8). This provides the first quantitative evidence for NO-induced changes in SC cell volume and the first pharmacological evidence of the NO regulatory pathway in SC cells. We also demonstrated changes in SC cell volume in response to changes in osmolarity. 

Figure 8.

 

Summary diagram of the pathway of NO regulation of SC cell volume. NO donors cause the formation of NO, which then binds to and activates soluble guanylate cyclase (sGC), the synthetic enzyme of cGMP. cGMP and its analogue, 8-Br-cGMP, activate PKG, which may, directly or indirectly, phosphorylate BK_Ca channels, with subsequent K⁺ efflux and decreases in cell volume.

The increases in SC cell volume (and TM cell volume; unpublished data, DZE) in response to the inhibition of NOS activity by L-NAME correlates with L-NAME– induced decreases in outflow facility observed in human anterior segment organ culture perfusion. ³⁶ Previous studies demonstrated that NOS III is localized in the human TM and SC. ¹⁵ The findings that L-NAME–induced inhibition of basal NOS activity in SC cells resulted in increased cell volume suggest a functional role for endogenous NOS III in regulating SC steady state cell volume and aqueous humor outflow facility. It is also noteworthy that exposure of SC cells to IBTX alone did not result in any increase in SC cell volume. This may suggest that the effects of basal NOS in regulating cell volume may not involve BK_Ca. Additionally, in human glaucoma, NOS III expression is reduced, ³⁷ suggesting dysfunction not only in the cell volume steady state regulatory mechanism but also in the cells' ability to produce endogenous NO that may be needed to alter cell volume in response to changes in ion fluxes. 

Conversely, activation of the NO system resulted in decreased cell volume and increased outflow facility. In fact, DETA-NO–induced decreases in SC cell volume are time and concentration dependent. Significant decreases in SC cell volume were observed 10 to 20 minutes after drug treatment. There is a correlation between the time-course for the NO-induced decreases in SC cell volume and the NO-induced increases in aqueous humor outflow facility. In previous studies, we observed significant increases in NO-induced enhancement of outflow facility 20 minutes after bolus application of DETA-NO (100 μM), ¹⁸ suggesting that DETA-NO may influence decreases in resistance to the outflow of aqueous humor, possibly through the modulation of SC cell volume. Similarly, DETA-NO decreased TM cell volume in a time-course that correlated with DETA-NO–induced stimulation of outflow facility. ¹⁸ These data suggest that NO decreases outflow resistance in the outflow pathway, possibly by decreasing cell volume in TM and SC cells. These observations support previous reports suggesting that resistance to aqueous humor outflow may reside in both the juxtacanalicular region of the TM and the inner wall of the SC. Although SC25 cells did not respond to DETA-NO treatment, cell volume was decreased in response to SNP (100 μM). These data suggest human donor variability in response to NO donors. Although we do not know the physiological mechanisms responsible for this variability, the mechanisms by which SNP and DETA-NO release NO are different. SNP belongs to the iron nitrosyl family and, though controversial, ³⁸ it is reported that chemical reduction of SNP is necessary for the release of NO. ³⁹ DETA-NO belongs to the NONOates or NO/nucleophile complexes, anionic groups [N(O)NO] that are chemically active. DETA-NO is able to spontaneously generate two molecules of NO per [N(O)NO] unit. 

ODQ antagonized the actions of DETA-NO and SNP in decreasing cell volume in SC cells, suggesting that NO binds to sGC and causes the release of cGMP. In SC cells, 1 μM ODQ did not significantly inhibit the DETA-NO–induced decreases in cell volume, whereas 5 μM abolished the DETA-NO–induced decreases. ODQ oxidizes sGC to its ferric form and causes conformational changes in sGC, such that sGC no longer responds to NO. We do not know why higher concentrations of ODQ are needed in SC cells in response to DETA-NO to observe the predicted result. We and others ⁴⁰ have previously inhibited sGC activity with 1 μM ODQ. Our previous studies demonstrated a more generic and consistent response to 1 μM ODQ in bovine choroid plexus tissue slices, ⁴¹ and bovine ciliary processes tissue slices ³⁵ in response to SNP, mouse spinal cord tissue slices ⁴² in response to both SNP and DETA-NO, and bovine TM tissue slices, human TM cultured cells, and porcine eye anterior segment organ perfusion ^18,43 in response to DETA-NO. 

As with human TM cells, the NO-induced reductions in SC cells were abolished when exposed to (RP)-8-Br-PET-cGMP-S, suggesting the involvement of protein phosphorylation of PKG in mediating the NO response. Additionally, inhibition of the BK_Ca channel abolished the ability of NO to reduce SC cell volume. Other studies have demonstrated that PKG phosphorylates the BK_Ca channel ^44,45 ; thus, NO-induced activation of PKG may allow for BK_Ca channel activation and subsequent efflux of K⁺ ions from the cell. Although this study suggests the involvement of the BK_Ca channel and K⁺ efflux in the NO-induced decreases in SC cell volume, other studies have demonstrated that the cell volume decrease is accompanied by K⁺ and Cl⁻ efflux ⁴⁶ induced by the activation of K⁺ and Cl⁻ channels or K⁺ and Cl⁻ symport. This suggests that K⁺ efflux may initiate a parallel Cl⁻ efflux in SC cells. ³⁰  

SC cells responded to changes in osmolarity; exposure to hypertonic medium resulted in cell shrinkage, whereas exposure to hypotonic medium resulted in cell swelling. The cell response to changes in tonicity was rapid and required accelerated imaging that was achieved using 512 × 512-pixel imaging. Significant cell swelling was achieved within 2 minutes of exposure to hypotonic medium. However, unlike TM cells, SC cells did not experience a spontaneous regulatory volume decrease in the presence of hypertonic medium. As with TM cells, ¹⁸ SC cells were treated with hypertonic medium; regulatory volume increase was achieved only after the hypertonic medium was removed from the cells and the cells were allowed to incubate at 37°C in isotonic medium. 

In our current studies we identified a morphologically homogenous population of SC cells that, when analyzed, demonstrated sensitivity to NO and cell volume decreases in response to NO treatment. Although all cells analyzed resulted in decreased cell volume in response to NO treatment, all cell lines did not respond similarly to the same NO donor; the physiological significance of this observation is unknown at this time. 

The morphologic homogeneity observed in SC cells is unlike the morphologic heterogeneity observed in TM cells. We previously identified two distinct cell populations in TM cell culture, consistent with the identified regions of the TM (the cribriform or juxtacanalicular region, the uveal region, and the corneoscleral region). However, unlike the SC, all cells analyzed did not experience decreases in cell volume in response to DETA-NO. Although we cannot determine which regional cell type in the TM culture experienced decreased cell volume, our data are consistent with the structure/function relationship that exists in tissues. SC endothelial cells are joined together by tight junctions that allow the joining together of the endothelial cells with their basal laminae and with juxtacanalicular cells. ^{11 –13} As aqueous humor flows from the uveal region through the corneoscleral region to the juxtacanalicular regions, the paracellular spaces narrow. As the fluid approaches the inner wall of the SC, the paracellular spaces found in the TM no longer exist, and the aqueous humor flows within and between the pores, intercellular junctions, and gaps in the SC cells. ¹³ Thus, the ability of NO to influence cell volume in cells from the juxtacanalicular region and inner wall of the SC underscores the potential role for changes in cell volume in regulating outflow resistance. 

Footnotes

 Supported in part by the University of Florida and Alcon Research, Ltd.

Footnotes

 Disclosure: D.Z. Ellis, Alcon Research, Ltd. (F); N.A. Sharif, Alcon Research, Ltd. (F, E); W.M. Dismuke, Alcon Research, Ltd. (F)