Endothelin-1–Induced Proliferation Is Reduced and Ca²⁺ Signaling Is Enhanced in Endothelin B–Deficient Optic Nerve Head Astrocytes

Abstract

Purpose.: To characterize the influence of endothelin-1 (ET-1) on optic nerve head astrocyte (ONHA) proliferation and Ca²⁺ signaling in ONHAs lacking functional endothelin B (ETB) receptors.

Methods.: ONHAs were isolated from adult wild type (WT) and transgenic spotting lethal (TSL) rats, lacking functional ETB receptors. ONHA specificity was confirmed by positive glial fibrillary acidic protein (GFAP), negative A2B5 (a marker for type II astrocytes located outside the optic nerve head) and myelin basic protein (MBP) labeling. The mitogenic effects of 10⁻⁷ or 10⁻⁹ M ET-1, or vehicle were investigated for 48 or 72 hours in WT and TSL ONHAs. Intracellular calcium levels ([Ca²⁺]_i) were assessed in ONHAs loaded with fura-2 calcium indicator dye.

Results.: ET-1–induced proliferation of TSL ONHAs was blunted at 48 hours (by 37% at 10⁻⁷ M and by 33% at 10⁻⁹ M) and 72 hours (by 117% at 10⁻⁷ M and by 100% at 10⁻⁹ M) compared with WT cells. ET-1–induced ONHA fura-2 ratio increases were significantly greater in TSL ONHAs (by 20% at 10⁻⁷ M and by 48% at 10⁻⁹ M) compared with WT ONHAs. ET-1–induced fura-2 ratio increases were blocked after pretreatment with BQ-610 (ETA antagonist) in WT and TSL ONHAs, but not by BQ-788 (ETB antagonist) in WT ONHAs.

Conclusions.: ET-1–induced ONHA proliferation is reduced in cells lacking functional ETB receptors, ET-1–induced [Ca²⁺]_i increases are enhanced in the absence of functional ETB receptors, and ETA, but not ETB, is required for ET-1–induced [Ca²⁺]_i elevation.

Optic neuropathies, such as glaucoma, are characterized by a progressive loss of retinal ganglion cells (RGCs) leading to visual loss and irreversible blindness. ¹ Intraocular pressure (IOP) is the most widely recognized risk factor associated with glaucoma, ^2,3 however, RGC loss can occur at all levels of IOP and can continue even after IOP is lowered and therefore other mechanisms may be involved in the pathophysiology of glaucoma. ⁴ Several structural changes occur in the glaucomatous optic nerve head (ONH), specifically, reorganization of the connective tissues in and around the lamina cribrosa. ⁵ Changes in cellular components, such as astrocyte proliferation and migration into nerve bundles in the lamina cribrosa, occur concurrently with this ONH remodeling. ⁶ Optic nerve head astrocyte (ONHA) proliferation has been demonstrated in human glaucoma, ^6,7 experimental glaucoma, ^7,8 and in in vitro studies. ^9,10 Whether the presence of reactive astrocytes is protective or harmful in glaucoma and whether their effects are the same across the spectrum of disease is not known. However, reactive astrocytes have been associated with impaired neurite outgrowth in other brain regions, ¹¹ the progression of Alzheimer's disease, ¹² schizophrenia, ¹³ and acute injury. ¹⁴  

Endothelin (ET) is a potent vasoconstrictive peptide that occurs as three isotypes (ET-1, ET-2, and ET-3), which are synthesized in the vascular endothelium as preproendothelins, first cleaved as propeptides and finally into fully active ETs. ¹⁵ ET-1 has a high affinity for, and signals through, two G-protein coupled receptors, endothelin A (ETA) and endothelin B (ETB). ^15,16 Both ETA and ETB receptors are coupled to phosphoinositol signaling with consequent increase of inositol trisphosphate (IP₃) and intracellular calcium concentration ([Ca²⁺]_i). ¹⁷ ET-1 is found throughout the eye, ¹⁸ but its actions are paradoxical with respect to glaucoma. On the one hand, increased ET-1 may have a beneficial effect in the anterior segment by decreasing IOP via decreased aqueous production and increased outflow, ^19,20 yet on the other, increased ET-1 levels in the optic nerve and retina have detrimental effects as chronic ET-1 administration leads to loss of RGCs. ^21,22 Two studies have reported significantly higher ET-1 plasma levels in patients with glaucoma, ^23,24 however, other studies have reported no significant differences. ^{25 –27} Regardless, increased ETB expression has been observed in optic nerves of glaucomatous patients, ⁷ in rabbit nerves after an acute crush injury, ²⁸ or transection, ²⁹ and in rat optic nerves with elevated IOP. ³⁰ Therefore, even if ET-1 levels are normal, the potentially harmful action of ET-1 may be enhanced from increased ETB receptor levels in the optic nerve. 

Recent studies have indicated that ET-1 may mediate its actions by promoting activation of ONHAs. ETB expression in ONHAs is increased in monkeys after experimentally induced glaucoma, ⁷ in rabbits ²⁸ and rats ³¹ after infusion of ET-1 into the optic nerve in vivo, and in humans with glaucoma. ⁷ Additionally, human ONHA reactivity is enhanced through both ETB and ETA. ¹⁰ Recently, we showed that ET-1–induced proliferation of rat ONHAs was dependent on both ETB and ETA receptors. ³² The aim of this work was to characterize the effect of ET-1 on proliferation and calcium signaling in rat ONHAs lacking functional ETB receptors. 

Materials and Methods

Animals

Adult transgenic spotting lethal rats (TSL) of the Wistar-Kyoto strain (4–6 months) were originally bred as previously described ³³ and were a gift from Dr. Thomas Yorio (University of North Texas Health Science Center, Fort Worth, TX). Spotting lethal rats possess a naturally occurring deletion in the ETB gene that yields aberrantly spliced ETB mRNA that yields an ETB receptor lacking the first and second putative transmembrane domains. This receptor is nonfunctional and results in precocious death in the neonatal period due to aganglionic megacolon. ³⁴ Transgenic expression of functional ETB receptors onto enteric precursors prevents neonatal demise and allows use of adult TSL animals. Receptor binding studies show no functional ETB receptors in the eyes of TSL rats. ³⁵ Wild type, heterozygous, and homozygous ETB-deficient rats were determined by PCR analysis. All procedures complied with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and ethics board approval was obtained from the Dalhousie University Committee on Laboratory Animals. 

Isolation and Purification of Optic Nerve Head Astrocytes

ONHAs were isolated and purified as previously described. ³² Unless otherwise stated, all reagents were obtained from a single source (Sigma Chemical Co., St Louis, MO). Briefly, the ONH was dissected from two rats (four ONHs) into wash media (50:1 of Hank's balanced salt solution (HBSS): penicillin/streptomycin (P/S)), transferred into growth media (GM; Dulbecco's modified Eagle's medium (DMEM) containing 4 mM l-glutamine, 4500 mg/L glucose, 10% fetal bovine serum (FBS), and 1%–2% P/S), sectioned into explants, centrifuged (1 minute), resuspended in GM, and transferred into a 50 mL tissue culture flask, where they were grown in a 37°C, 5% CO₂ incubator. Cells were left for a week after which time half of the GM was replaced with fresh GM every 3 to 6 days. Cells were grown to approximately 50% confluence and passaged with 0.25% trypsin (diluted in Earle's balanced salt solution) into a 250-mL tissue culture flask. In the absence of serum, nonastrocytes deadhere and yield a purified culture of ONHAs. ¹⁰ At approximately 50% confluence, the GM was replaced with DMEM containing P/S for 24 hours which was then replaced with GM. Cells were passaged at approximately 80% confluence. All experiments were performed on cells between passage number three and six. These procedures were repeated to produce four separate cultures of cells, each from different rats. 

Immunohistochemistry

Cells were characterized as previously described. ^32,36 Briefly, they were plated on 12 mm glass circular coverslips (Fisher Scientific, Ottawa, ON, Canada) in GM for 1 to 2 days, washed (0.1 M phosphate buffer [PB] for 5 minutes), incubated in paraformaldehyde (4% in 0.1 M PB; 10 minutes), washed three times, and incubated in blocking solution (0.3% Triton X-100, 10% FBS in 0.1 M PB; 30 minutes) at room temperature. Single primary antibodies were added for rabbit anti-ETB (corresponding to amino acid residues 298 to 313 of rat ETB which are located within the third intracellular loop of the receptor) or rabbit anti-ETA (both 1:200; Alomone, Jerusalem, Israel); polyclonal rabbit anti-glial fibrillary acidic protein (GFAP; 1:100; Dako Cytomation, Glostrup, Denmark), a marker of astrocytes; monoclonal mouse anti-A2B5 (1:250; mab312, Millipore, Billerica, MA), a marker of oligodendrocyte precursors and type II astrocytes, known to be located exclusively in regions of the optic nerve outside the ONH, ^37,38 or monoclonal mouse anti-myelin basic protein (MBP; 1:50; Chemicon, Temecula, CA), a marker of oligodendrocytes, each in blocking solution overnight at 4°C (n = three per label). Cells were then washed three times, incubated in secondary antibody solutions containing Cy3-conjugated goat anti-rabbit IgG, Cy3-conjugated goat anti-mouse IgG (Jackson Laboratories, West Grove, PA), or rhodamine-conjugated goat anti-mouse IgM (AP128R, Chemicon). Secondary antibodies (1:200 in 0.1 M PB) were used at room temperature for 1 hour. After incubation, cells were washed three times, incubated in TO-PRO-3 iodide (1:1000 in 0.1 M PB, T-3605, (Invitrogen, Burlington, ON, Canada) for 15 minutes at room temperature, washed two times, air-dried overnight and mounted on 1 mm microscope slides with mounting medium. Cells were visualized with a confocal microscope (Nikon C1, Nikon Canada Inc., Toronto, ON, Canada) for double labeling experiments or an inverted Zeiss microscope for differential interference contrast imaging. 

Cell Treatments and Proliferation Assays

Cells (either wild type [WT] or TSL) were passaged, centrifuged (10 minutes), resuspended into single cell solution in GM, and plated on a 96-well plate (Falcon, Franklin Lakes, NJ) at 5000 cells per well in GM (100 μL per well). The following day, GM was removed and replaced with DMEM containing P/S. For proliferation experiments cells were incubated in ET-1 (10⁻⁷, or 10⁻⁹ M; Peptides International, Louisville, KY) or vehicle (0.1% acetic acid at 10⁻⁷ M) for 48 or 72 hours (n = 10–12 per group). For antagonist experiments, cells were pretreated with an ETB antagonist (BQ-788, Peptides International), an ETA antagonist (BQ-610, Peptides International), or both together at concentrations of either 1 or 5 μM for 30 minutes. Control ONHAs were treated with appropriate vehicles (1 or 5 μM DMSO with H₂O). After pretreatment, cells were incubated in ET-1 (10⁻⁷ or 10⁻⁹ M) with either or both antagonists for 72 hours (n = 10–11 per group for WT and n = 12 per group for TSL at 1 μM concentrations; and n = 3 per group for WT and n = 5–6 per group for TSL at 5 μM concentrations). After the appropriate incubation period, proliferation was assessed using a formazan kit (CellTiter 96 Non-Radioactive Cell Proliferation Assay; Promega, Madison, WI). ³⁹ The formazan cell proliferation assay was previously found to yield proliferation results of human ONHAs which were compatible with results using a [³H] thymidine uptake assay. ¹⁰ Optical densities were converted to number of cells based on a standard curve. For ET-1 only and ET-1 with antagonist proliferation experiments, data were converted to a percentage of cells exposed to only DMEM with P/S. Comparisons within cell type (WT or TSL) were made relative to their appropriate vehicles. Comparisons between cell types were made relative to the percent of ONHA at each ET-1 concentration. 

Cell Treatments and Calcium Imaging

Calcium imaging experiments were conducted on ONHAs as previously described for RGCs. ⁴⁰ Coverslips were plated with cells, removed from the culture medium and incubated at 37°C for 30 minutes in a solution containing 5 μM fura-2 AM (Invitrogen). Fura-2 (kD = 140 nM) dye is a ratiometric dye, with image pairs collected at 340 and 380 nm excitation and 510 nm emission. The fura-2 dye was first dissolved in dimethyl sulfoxide (0.1% final concentration) and then solubilized in HBSS with 0.1% pluronic acid F-127 (Invitrogen). The microscope chamber was superfused at a rate of 1 mL per minute with 100% oxygenated HBSS, 10 mM HEPES (pH 7.4), 2.6 mM calcium chloride, and warmed to 33°C to 35°C. For antagonist experiments, cells were pretreated with an ETB antagonist (BQ-788), or an ETA antagonist (BQ-610). All drugs (ET-1, BQ-610, and BQ-788) were dissolved in HBSS on the day of the experiment. After loading with fura-2, coverslips were transferred to the superfusion chamber. All drugs were bath-applied by switching from the control superfusate to one containing the appropriate treatment. Image pairs were captured every 20–40 seconds in the absence of drugs to limit possible photo-bleaching, and every 3–10 seconds during drug application to ensure that the peak response was captured. Increases in the fura-2 ratios have been previously shown to correspond to increases in [Ca²⁺]_i, therefore we inferred [Ca²⁺]_i from measured fura-2 ratios. ⁴⁰ Fluorescence images (8-bit processing, 4 × 4 binning) were converted into ratiometric data (340 nm and/or 380 nm) using software (Imaging Workbench 2.2; Axon Instruments, Foster City, CA). The imaging equipment has been previously described by our laboratory. ⁴⁰  

To analyze cells, circular regions of interest were selected on individual cells that responded to treatments. Changes in fura-2 ratio were calculated as the difference between baseline fura and the peak fura ratio. Baseline ratios were calculated from an average of three images before drug treatment. Changes in fura-2 ratios are presented as arbitrary units (a.u. ± SD). ONHAs that had a negligible response to ET-1 (Δ in fura-2 ratio that was < 0.1 a.u.) to ET-1 were excluded from analysis. ⁴¹ The percentage of ONHAs that responded was calculated for all ONHAs except for one set, TSL cells (5 cells) after 10⁻⁸ ET-1 treatment, which were analyzed but the cell images were not recorded. Therefore, while the data from these cells were included in the statistical analysis, it is not known how many cells from this set did not respond. For ET-1 experiments, cells were exposed to ET-1 (10⁻⁷, 10⁻⁸, or 10⁻⁹ M) by bath application for 1 minute, washed for 10 to 15 minutes and then re-exposed to the same concentration of ET-1. For antagonist studies, cells were exposed to antagonist (BQ-788, or BQ-610 at 2 μM) for 2 minutes by bath application, washed for 1 minute, treated with 10⁻⁸ M ET-1, washed for 10 minutes and then re-exposed to 10⁻⁸ M ET-1. ET-1 at 10⁻⁸ M was used because it was the lowest concentration that yielded a consistent and repeatable response. 

Statistical Analysis

Statistical significance was assessed with a one-way ANOVA (SPSS 9.0 software; SPSS Inc., Chicago, IL). The Fisher's least significant difference post hoc test was used for parametric statistics and the Games-Howell post hoc test was used for nonparametric statistics. Independent samples t-tests were performed for comparisons between animal groups. Differences were considered statistically significant if P < 0.05. 

Results

Characterization of ONHAs

The morphology of cells isolated from WT and TSL animals appeared similar (Figs. 1A and 1B). Cell type was identified by double immunolabeling for the nuclear stain TO-PRO-3 with GFAP, A2B5, or MBP. All cells isolated immunolabeled for GFAP (Figs. 1C and 1D) indicating that these cells were astrocytes. Additionally, no immunolabeling was observed for either A2B5 (Figs. 1E and 1F), indicating that no astrocytes were isolated from outside the ONH, or for MBP (Figs. 1G and 1H), indicating oligodendrocytes were not present. Virtually all cells immunolabeled for both ETB (Figs. 1I and 1J) and ETA (Figs. 1K and 1L). No differences were observed between WT and TSL cells for any of the immunomarkers. 

Figure 1.

 

Purification of optic nerve head astrocytes from WT and TSL rats. Cells isolated from optic nerve heads of WT (A) and TSL (B) rats lacking functional ETB receptors appear morphologically similar by differential interference contrast (DIC) imaging. Cells double-labeled for TO-PRO-3 (blue) and glial fibrillary acidic protein (GFAP; magenta-red, C, D). Virtually all TO-PRO-3 positive cells were also GFAP-positive, indicating that they were astrocytes. These cells were negative for A2B5 (E, F) and myelin basic protein (MBP; G, H) indicating that they were type I astrocytes located exclusively in the optic nerve head and that no oligodendrocytes were present. TO-PRO-3 co-localized with ETB (I, J) and ETA (K, L), indicating these cells expressed both ETB and ETA receptors. Scale bar, 50 μm.

Together, these findings indicate that the cultures were pure for astrocytes located within the ONH, that these cells expressed both ETB and ETA receptors, and that no immunolabeling differences existed between WT and TSL ONHAs. 

Reduced ET-1–Induced Proliferation of ETB-Deficient ONHAs

At 48 hours after 10⁻⁷ or 10⁻⁹ M ET-1 exposure, the number of WT ONHAs increased by 72% or 57% respectively while the number of TSL ONHAs increased by 37% or 25% respectively (Fig. 2) all relative to their appropriate vehicles. In all cases, ET-1 caused a significant proliferation of ONHAs compared with vehicle (P < 0.01). Proliferation of TSL ONHAs was reduced at 48 hours after 10⁻⁷ M (37% difference between WT and TSL ONHAs, P = 0.072) and 10⁻⁹ M (33% difference between WT and TSL ONHAs, P = 0.039) ET-1 exposure compared with WT ONHAs (Fig. 2). At 72 hours after 10⁻⁷ and 10⁻⁹ M ET-1 exposure, the number of WT ONHAs increased significantly by 133% and 115% (P < 0.025) respectively while the number of TSL ONHAs increased by 30% and 30% (P < 0.03) respectively, all relative to their appropriate vehicles. Proliferation of TSL ONHAs was reduced at 72 hours after 10⁻⁷ M (118% difference between WT and TSL ONHAs, P = 0.004) and 10⁻⁹ M (100% difference between WT and TSL ONHAs, P = 0.012) ET-1 exposure compared with WT ONHAs (Fig. 2). Therefore, ONHAs proliferated with and without functional ETB receptors; however, this proliferation was blunted in TSL ONHAs compared with ONHAs possessing both ETB and ETA receptors. 

Figure 2.

 

WT and TSL ONHA proliferation at 48 (A) and 72 (B) hours of exposure to 10⁻⁷ or 10⁻⁹ M ET-1. ONHA proliferation was significantly increased above the respective controls at both ET-1 concentrations for WT and TSL ONHAs at both time points. Proliferation of TSL ONHAs was significantly blunted compared with WT ONHAs at 48 hours exposure to 10⁻⁹ M ET-1 and at 72 hours exposure to both concentrations of ET-1. C, control (media only); V, vehicle; *P < 0.05, relative to vehicle within groups; ^δ P < 0.05, comparison between WT and TSL ONHAs. Error bars show SEM.

Increased Calcium Signaling in ETB-Deficient ONHAs

Bath application (1 minute) of ET-1 increased fura-2 ratio above baseline in 54 of 87 (62%) WT ONHAs and 63 of 77 (82%) TSL ONHAs after 10⁻⁷ M exposure; and in 21 of 49 (43%) WT ONHAs and 36 of 59 (61%) TSL ONHAs after 10⁻⁸ M exposure. In WT ONHAs ET-1 increased the fura-2 ratio above baseline at 10⁻⁷ M (0.68 ± 0.21 a.u., n = 54 cells) or 10⁻⁸ M (0.39 ± 0.24 a.u., n = 21 cells; Fig. 3C with a representative trace shown in Fig. 3A). No ONHAs exposed to 10⁻⁹ M ET-1 met the minimum threshold of fura-2 ratio change of 0.1 a.u. In TSL ONHAs, ET-1 increased fura-2 ratio above baseline at 10⁻⁷ M (0.81 ± 0.21 a.u., n = 63 cells) and 10⁻⁸ M (0.57 ± 0.15 a.u., n = 36 cells; Fig. 3C, representative trace shown in Fig. 3B). The increase in fura-2 ratio in response to ET-1 was significantly greater in TSL ONHAs compared with WT ONHAs at both 10⁻⁷ M ET-1 (0.13 a.u., P = 0.001) and at 10⁻⁸ M ET-1 (0.18 a.u., P = 0.001; Fig. 3C). 

Figure 3.

 

ET-1–induced [Ca²⁺]_i is enhanced in ONHAs cultured from TSL animals lacking functional ETB receptors. Representative example traces from WT (A) and TSL (B) ONHAs show increased fura-2 ratio above baseline to bath application of 10⁻⁸ M ET-1. The increase in [Ca²⁺]_i as determined by change in fura-2 ratio (peak − baseline) was significantly greater in TSL ONHAs compared with WT ONHAs at both 10⁻⁷ M and 10⁻⁸ M ET-1 (C). *P < 0.05 relative to WT. Error bars show SEM.

ONHA Calcium Signaling Occurs Through ETA but Not ETB

To investigate the roles of acute ETB and ETA blockade on ET-1–induced [Ca²⁺]_i increase, the effect of 10⁻⁸ M ET-1 on fura-2 ratio was measured after treatments with either ETB (BQ-788) or ETA (BQ-610) antagonist (Fig. 4), both at 2 μM concentration. Pretreatment of WT cells with BQ-610 prevented ET-1 induced fura-2 ratio elevation (0.06 ± 0.02 a.u. increase above baseline) but the effect of ET-1 recovered after wash (0.33 ± 0.21 a.u.; P < 0.001; n = 9, Figs. 4A and 4D). Similar results were obtained with TSL ONHAs (0.07 ± 0.06 a.u. increase after BQ-610, 0.56 ± 0.18 a.u increase after wash.; P < 0.001; n = 15, Figs. 4C and 4D). Pretreatment of WT cells with BQ-788 did not prevent ET-1 induced increases of fura-2 ratio (0.33 ± 0.21 a.u.; n = 12; Figs. 4B and 4D) and a similar (P = 0.189) response was obtained after wash (0.45 ± 0.22 a.u.; n = 12; Figs. 4B and 4D). 

Figure 4.

 

Blocking ETA, but not ETB, prevented ET-1–induced [Ca²⁺]_i increases. ONHAs were exposed to an antagonist by bath application, washed for 1 minute, exposed to 10⁻⁸ M ET-1, washed for 10 minutes and then re-exposed to 10⁻⁸ M ET-1. BQ-610 (ETA antagonist; 2 μM) prevented ET-1–induced [Ca²⁺]_i increases in ONHAs cultured from WT (A) and TSL (C) animals. BQ-788 (ETB antagonist; 2 μM) did not block the ET-1–induced increase in [Ca²⁺]_i (B). Quantification of Δ fura-2 ratios demonstrated significant increases in [Ca²⁺]_i in WT and TSL ONHAs when BQ-610 was washed away compared with its presence (D). No significant difference was present in Δ fura-2 ratios for WT ONHAs when BQ-788 was present compared with when it was washed away. *P < 0.05 relative to data with antagonist. Error bars show SEM.

ETB or ETA Receptor Antagonism Reduced ONHA Proliferation

After pretreatment of WT ONHAs with BQ-788 (1 μM), ET-1 caused a significant proliferation of ONHAs at both 10⁻⁷ M (104% increase; P = 0.010) and 10⁻⁹ M (70% increase; P = 0.009) concentrations relative to vehicle (Fig. 5); however, in the presence of a higher concentration BQ-788 (5 μM), ET-1–induced proliferation was prevented. Similarly, WT ONHAs proliferated significantly with 10⁻⁷ M ET-1 exposure after pretreatment with 1 μm BQ-610 (120% increase, P = 0.002), but not with 10⁻⁹ M ET-1. Proliferation was blocked with 5 μM BQ-610 for both ET-1 concentrations (P > 0.060). In the presence of both antagonists, the number of WT ONHAs increased significantly only with the higher ET-1 and lower antagonist concentrations (113% increase, P = 0.026). 

Figure 5.

 

Proliferation of WT and TSL ONHAs after ET-1 (10⁻⁷ M and 10⁻⁹ M) exposure for 72 hours in the presence of BQ-788, an ETB antagonist, BQ-610, an ETA antagonist, and both antagonists at 1 μM and 5 μM concentrations. V, vehicle only; A, antagonist only; −7, 10⁻⁷ M ET-1; −9, 10⁻⁹ M ET-1. *P < 0.05 relative to V. Error bars show SEM.

In TSL ONHAs there was significant proliferation with 10⁻⁷ M ET-1 in the presence of both concentrations of BQ-788 (34% increase, P = 0.004; and 41% increase, P < 0.001; Fig. 5) but not 10⁻⁹ M ET-1. With BQ-610, there was significant proliferation only with the higher ET-1 and lower antagonist concentration (31% increase, P = 0.001). When pretreated with both antagonists, there was significant proliferation with 10⁻⁷ M ET-1 and 1 μM of the antagonists (17% increase, P = 0.023) and a significant decrease with 10⁻⁹ M ET-1 and 5 μM of the antagonists (27% decrease, P = 0.002; Fig. 5) Interestingly, there was a decrease in TSL ONHAs with exposure to the antagonists alone (without ET-1) at either 1 μM (19% decrease; P = 0.012) or 5 μM (22% decrease, P = 0.007) compared with vehicle (Fig. 5). 

Discussion

The results of this study indicate that ET-1 signaling promotes ONHA proliferation through both ETB and ETA receptors. This conclusion is based on our observations that ET-1 induces proliferation of TSL ONHAs; however, this proliferation was blunted compared with ONHAs from WT animals. Furthermore, ETB receptor antagonism (BQ-788 at 5 μM) blunted ET-1–induced proliferation of WT ONHAs and ETA receptor antagonism (BQ-610 at 5 μM) reduced ET-1's mitogenic effects in both WT and TSL ONHAs. Because ONHA [Ca²⁺]_i was blocked by ETA, but not ETB receptor antagonism, our data indicate that elevation of [Ca²⁺]_i occurs in response to ET-1 signaling through ETA but not ETB receptors. This study is the first to investigate ONHAs from transgenic animals lacking functional ETB receptors. 

ONHA proliferation has been demonstrated in early stages of human glaucoma, ⁴² as well as in experimental models of the disease. ^10,30,43 We recently demonstrated that ET-1 caused ONHA proliferation in cultures from Brown Norway rats by 15%–20% at 48 hours and by 21%–29% at 72 hours. ³² In the present study, we observed a larger mitogenic effect (57%–72% at 48 hours and 115%–133% at 72 hours) of ET-1 on ONHAs isolated from TSL Wistar-Kyoto strain rats. In human ONHAs, 10⁻⁷ M ET-1 increased the number of ONHA by 12% at 48 hours and 32% at 96 hours. ¹⁰ Also of interest, we previously demonstrated that at 48 hours and 72 hours after ET-1 exposure, either ETB or ETA receptor blockade significantly blunted proliferation. ³² In this study, ETB and/or ETA antagonists at higher concentrations significantly blunted ONHA proliferation at 72 hours at both concentrations of ET-1. A lower concentration of antagonist blunted ONHA proliferation at 10⁻⁹ M, but not 10⁻⁷ M ET-1. These observations are consistent with studies of brain astrocytes in which ET-1–induced proliferation occurs independently through either ETB or ETA signaling. ⁴⁴  

The TSL Wistar-Kyoto strain was used in this study because of the naturally occurring deletion of functional ETB receptors; an experimental model which is not available in Brown Norway rats. The use of littermate wild type animals allowed us to attribute our findings to the absence of ETB, and not to rodent strain. ET-1 induced significant proliferation of TSL ONHAs at 48 and 72 hours after exposure; however, this effect was significantly lower than in ONHAs that possessed both functional ETB and ETA receptors. In contrast, we previously reported ³² that in Brown Norway ONHAs ET-1–induced proliferation was completely prevented at 48 hours with an ETB receptor antagonist. Therefore, acute receptor blockade by an ETB antagonist has a more potent effect on proliferation than the absence of functional ETB receptors. These differences may be due to changes in ETA receptor levels which have been shown to occur in ETB knockout mice, ⁴⁵ to desensitization of the ETA receptors in the absence of ETB, or to heterodimerization of the two receptors. ⁴⁶  

Previous reports have demonstrated that endothelin promotes proliferation of brain astrocytes, ^47,48 vascular smooth muscle cells, ⁴⁹ Swiss 3T3 fibroblasts, ⁵⁰ and type I astrocytes ⁵¹ through Ca²⁺ signaling and that this mitogenic effect occurs through ETA. ^47,48 In cultured human ONHAs, ETA antagonism completely blocked ET-1–induced increases of [Ca²⁺]_i; however, pretreatment of the cells with PD142893, a mixed ETA/ETB antagonist blocked the increase in ET-1–induced intracellular calcium level by only 60% to 75%. ¹⁰ We observed that ETA, but not ETB receptor blockade prevented increased [Ca²⁺]_i in WT ONHAs. Interestingly, in TSL ONHAs we observed enhanced ET-1–induced [Ca²⁺]_i increases compared with wild type ONHAs that possessed both ETB and ETA receptors. The mechanism for this finding is unclear because ETA and ETB receptors are both G-protein coupled receptors linked to phosphoinositol signaling that leads to increased IP₃ and, therefore, release of Ca²⁺ from IP₃-sensitive intracellular stores. ^17,47 ETA receptor upregulation in the TSL ONHAs or desensitization are possible explanations. Furthermore, these results may be due to a novel ET receptor or heterodimerization of the ETB and ETA receptors that results in a unique ligand signaling cascade, as described by Prasanna and colleagues. ^10,20,30 We previously suggested that there may be overlap in the mitogenic function of ETB and ETA receptors. ³² However, our proliferation and calcium-imaging results indicate that ETB and ETA may have separate signaling pathways which independently induce ONHA proliferation. Interestingly, we observed no change in [Ca²⁺]_i with 10⁻⁹ M ET-1; however, significant ONHA proliferation occurred at this concentration. Although it is possible that [Ca²⁺]_i is not coupled with ET-1 induced proliferation in ONHAs, given the extensive evidence from other models identifying an association between [Ca²⁺]_i and proliferation, our findings are likely due to differences in our experimental paradigms. [Ca²⁺]_i changes were observed after only brief (1 minute) exposure to ET-1 whereas proliferation was assessed after continuous ET-1 exposure for 48 or 72 hours. Therefore [Ca²⁺]_i changes may occur with 10⁻⁹ M ET-1 after prolonged exposure. Future studies should investigate whether impairment of ET-1–induced [Ca²⁺]_i elevation prevents proliferation and attempt to elucidate the signaling mechanism by which ETB promotes ONHA proliferation. 

In summary, our findings indicate that ETB and ETA influence ONHA reactivity, but that mitogenic effects via ETB are not due to calcium signaling. Future strategies aimed at blunting ET-1 induced mitogenic effects on ONHAs should target both the ETB and ETA receptors. 

Footnotes

 Supported by Grant MOP-89808 from the Canadian Institutes of Health Research (BCC) and by a Postdoctoral Fellowship Award from the Dalhousie Medical Research Foundation (JAM).

Footnotes

 Disclosure: J.A. Murphy, None; M.L. Archibald, None; W.H. Baldridge, None; B.C. Chauhan, None

The authors thank Thomas Yorio for generously providing TSL breeder animals.