Constriction of Retinal Venules to Endothelin-1: Obligatory Roles of ET_A Receptors, Extracellular Calcium Entry, and Rho Kinase

Abstract

Purpose: Endothelin-1 (ET-1) is a potent vasoconstrictor peptide implicated in retinal venous pathologies such as diabetic retinopathy and retinal vein occlusion. However, underlying mechanisms contributing to venular constriction remain unknown. Thus, we examined the roles of ET-1 receptors, extracellular calcium (Ca²⁺), L-type voltage-operated calcium channels (L-VOCCs), Rho kinase (ROCK), and protein kinase C (PKC) in ET-1-induced constriction of retinal venules.

Methods: Porcine retinal venules were isolated and pressurized for vasoreactivity study using videomicroscopic techniques. Protein and mRNA were analyzed using molecular tools.

Results: Retinal venules developed basal tone and constricted concentration-dependently to ET-1. The ET_A receptor (ET_AR) antagonist BQ123 abolished venular constriction to ET-1, but ET_B receptor (ET_BR) antagonist BQ788 had no effect on vasoconstriction. The ET_BR agonist sarafotoxin S6c did not elicit vasomotor activity. In the absence of extracellular Ca²⁺, venules lost basal tone and ET-1–induced constriction was nearly abolished. Although L-VOCC inhibitor nifedipine also reduced basal tone and blocked vasoconstriction to L-VOCC activator Bay K8644, constriction of venules to ET-1 remained. The ROCK inhibitor H-1152 but not PKC inhibitor Gö 6983 prevented ET-1-induced vasoconstriction. Protein and mRNA expressions of ET_ARs and ET_BRs, along with ROCK1 and ROCK2 isoforms, were detected in retinal venules.

Conclusions: Extracellular Ca²⁺ entry via L-VOCCs is essential for developing and maintaining basal tone of porcine retinal venules. ET-1 causes significant constriction of retinal venules by activating ET_ARs and extracellular Ca²⁺ entry independent of L-VOCCs. Activation of ROCK signaling, without involvement of PKC, appears to mediate venular constriction to ET-1 in the porcine retina.

The light-sensitive retinal tissue exhibits a high metabolic rate, which requires the retinal circulation to provide sufficient oxygen and nutrients to maintain its proper function.¹ In the microcirculation, two factors that impact optimal tissue perfusion are the arteriolar resistance for controlling blood flow and the venular resistance for regulating hydrostatic pressure and fluid homeostasis at the level of the capillaries.^2,3 Because the retinal circulation lacks autonomic innervation,⁴ local vascular control mechanisms appear to dominate the regulation of retinal vascular resistance.⁵ Although the importance of the intrinsic ability of retinal arterioles to regulate their resting diameter is well established,⁵ there is a paucity of information on vasomotor function of retinal venules. In vivo studies have reported an increase in retinal venular diameter during flickering light stimulation⁶ or a reduction in retinal venular diameter in response to hyperoxia,⁷ suggesting these vessels play an active role in flow regulation. However, it is difficult to distinguish the direct influence on retinal venular tone from a response secondary to the upstream changes in pressure and/or flow⁸ using a whole-organ approach or during systemic intervention. A recent report has shown that isolated porcine retinal veins are reactive to local vasoactive factors such as vasodilator adenosine and vasoconstrictor endothelin-1 (ET-1).⁹ It is unclear, however, whether smaller upstream venules exhibit similar reactivity, especially vasoconstriction, which dictates not only blood drainage but also capillary pressure and fluid exchange. 

ET-1 is a potent vasoactive peptide composed of 21 amino acids and plays important roles in many biological functions, including vasoconstriction, cardiovascular remodeling, tissue inflammation, and cell proliferation.^10–17 In the eye, ET-1 is distributed locally in the retina,^18,19 causes retinal vasoconstriction in vivo,^20–23 and reduces retinal blood flow.^23–25 Our previous studies have demonstrated the cell signaling mechanisms contributing to ET-1–induced constriction of porcine and human retinal arterioles.^26–29 Although others have shown that isolated porcine retinal veins are capable of constricting to ET-1,⁹ the underlying mechanisms contributing to this response have not been established. Accumulating evidence has shown that vitreous humor, plasma, and ocular levels of ET-1 are elevated in diabetes,^23,24,30,31 especially in patients with diabetic retinopathy.^31,32 The increased plasma ET-1 level also has been implicated in the pathogenesis of retinal vein occlusion (RVO).^33,34 Because alteration in function of the retinal microcirculation, including tissue edema, is regarded as a key event contributing to visual impairment and blindness with diabetes and RVO, greater insight into the mechanisms controlling retinal venular vasomotor activity would advance our understanding of these retinal pathologies. 

Previous studies in different organ systems and vascular beds have suggested that vasoconstriction to ET-1 can be mediated by the activation of its receptors (ET_A and/or ET_B),^11,12,26 elevation of intracellular calcium (Ca²⁺),^29,35–37 and signaling through Rho kinase (ROCK) and/or protein kinase C (PKC).^12,29,38 However, the contributions of specific ET-1 receptors, extracellular Ca²⁺ entry, and intracellular signaling pathways to ET-1–induced constriction of retinal venules remain unknown. To address these issues directly without the influence of metabolic, hemodynamic, and neuronal/glial factors commonly inherent in in-vivo preparations, porcine retinal venules were isolated and pressurized for in vitro study. The relative functional roles of ET_ARs, ET_BRs, ROCK, PKC, and extracellular Ca²⁺ entry through L-type voltage-operated calcium channels (L-VOCCs) in the ET-1–induced constriction of porcine retinal venules were investigated. We also assessed ET_AR, ET_BR, and ROCK isoform (ROCK1 and ROCK2) expression in porcine retinal venules to assist functional interpretation. 

Methods

Animal Preparation

All animal procedures were performed in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and were approved by the Baylor Scott & White Health Institutional Animal Care and Use Committee. Pigs (Sus scrofa domesticus, Yorkshire) of either sex (age range, 8–12 weeks; weight range, 10–20 kg) purchased from Real Farms (San Antonio, TX, USA) were sedated with Telazol (4–8 mg/kg, intramuscularly), anesthetized with 2% to 5% isoflurane, and intubated. The procedure used for harvesting eyes has been described previously.^39,40 

Isolation and Cannulation of Retinal Venules

We previously described the techniques for isolating and cannulating porcine retinal arterioles, which were applied to retinal venules in the current study.^39–41 The ocular tissue was placed in a cooled dissection chamber (∼8°C) containing a physiological salt solution (PSS; 145.0 mM NaCl, 4.7 mM KCl, 2.0 mM CaCl₂, 1.17 mM MgSO₄, 1.2 mM NaH₂PO₄, 5.0 mM glucose, 2.0 mM pyruvate, 0.02 mM EDTA, and 3.0 mM MOPS) with 0.1% albumin. The retinal venules (1 to 1.5 mm in length without side branches) were identified based on the dark-red deoxygenated blood in the lumen and their thin vascular wall compared with the parallel arterioles containing bright-red oxygenated blood and a thick vascular wall. The second-order venules were carefully dissected with the aid of a stereomicroscope (model SZX12; Olympus, Melville, NY, USA). After removal of any remaining neural/connective tissues, the venule was transferred with a capillary micropipette (Wiretrol II; Drummond Scientific Company, Broomall, PA, USA) to a polymethylmethacrylate vessel chamber containing PSS-albumin solution equilibrated with room air at ambient temperature. Both ends of the venule were cannulated using glass micropipettes filled with PSS containing 1% albumin, and the outside of the venules were securely tied to the pipettes with 11-0 ophthalmic suture (Alcon, Fort Worth, TX, USA). Vessels were pressurized to 5 cm H₂O (4 mm Hg) intraluminal pressure by two independent pressure reservoirs and allowed to develop basal tone before study. This intraluminal pressure was chosen based on a study reporting porcine retinal transmural venous pressure of 0 to 4 mm Hg.⁴² Vasomotor activity of isolated venules was recorded using videomicroscopic techniques throughout the experiments.³⁹ 

Experimental Protocols

Cannulated venules were bathed in PSS-albumin at 36° to 37°C to allow development (∼90 minutes) of basal tone. In one series of studies, the vasomotor response to cumulative administration of ET-1 (1 pM to 10 nM; Bachem, Torrance, CA, USA),²⁶ selective ET_BR agonist sarafotoxin S6c (10 pM to 0.1 μM; Tocris Cookson, Ellisville, MO, USA),²⁶ or PKC activator phorbol-12,13-dibutyrate (PDBu; 0.1 μM to 10 μM)²⁹ was evaluated. Retinal venules were exposed to each concentration of ET-1, sarafotoxin, or PDBu for 10 minutes until a stable diameter was established. Because the vasoconstrictor action of ET-1 was maintained after washout, only one concentration-response curve was constructed in each vessel for this drug. The relative roles of ET_ARs and ET_BRs in the retinal venule responses to ET-1 were evaluated after treatment of the vessel with respective antagonists BQ123 (1 μM)²⁶ and BQ788 (0.1 μM).²⁶ To assess the contribution of extracellular Ca²⁺ entry to ET-1–induced vasoconstriction, vasomotor activity was examined in Ca²⁺-free PSS-albumin (with 1 mM EDTA).²⁶ The role of L-VOCCs in ET-1–induced vasoconstriction was examined in the presence of L-type calcium channel antagonist nifedipine (1 μM).^29,37 In another set of vessels, the specificity and efficacy of nifedipine were assessed by examining its effectiveness on vasoconstriction to L-VOCC activator Bay K8644 (6 μM).⁴³ To determine the functional importance of endogenous ROCK and PKC in ET-1–induced vasoconstriction, experiments were performed in the presence of ROCK inhibitor H-1152 (3 μM)^27,29 or broad-spectrum PKC inhibitor Gö 6983 (10 μM).²⁹ Vessels were pretreated with antagonists or inhibitors extraluminally for at least 30 minutes. 

Chemicals

Drugs for the functional studies were purchased from MilliporeSigma (St. Louis, MO, USA) except where specifically stated otherwise. ET-1, sarafotoxin, BQ788, and H-1152 were dissolved in water; BQ123, nifedipine, and Bay K8644 were dissolved in ethanol; and PDBu and Gö 6983 were dissolved in dimethyl sulfoxide. All subsequent concentrations of these drugs were diluted in PSS.²⁶ The final concentration of dimethyl sulfoxide or ethanol in the vessel bath was less than 0.1% by volume. Vehicle control studies indicated that this final concentration of solvent had no effect on vessel viability, vasoconstrictor responses, or maintenance of basal tone (data not shown). 

mRNA Isolation and Real-Time PCR Analysis

Total RNA was isolated from retinal venules (sample pooled from both eyes) and neural retina tissue via RNeasy mini kit (QIAGEN, Crawley, UK), as described in our previous study.⁴⁴ The same amount of mRNA for each sample was used to synthesize cDNA with the SuperScript III First-Strand Synthesis System (Thermo Fisher Scientific, Waltham, MA, USA). RNA and cDNA were quantified using a Nanodrop spectrophotometer (BioTek, Winooski, VT, USA), and sequentially processed using real-time PCR reagents (PowerUp SYBR Green Master Mix, Thermo Fisher Scientific). To perform the real-time PCR experiments, we used specific primer sets for ET_AR (forward primer: 5′-AAGCAGGACAGCCCATTAAG-3′, reverse primer: 5′-CTGCTAGCTGAAGTACTCGGAAC-3′), ET_BR (forward primer: 5′-GGAGTTGAGATGTGTAAGCTGGTG-3′, reverse primer: 5′-TCTGAGTAGGATGGAGCAAGCAGA-3′), and glyceraldehyde-3-phosphate dehydrogenase (GAPDH; forward primer: 5′-CCACCCACGGCAAGTTCCACGGCA-3′, reverse primer: 5′-GGTGGTGCAGGAGGCATTGCTGAC-3′). Real-time PCR was performed by monitoring the increase in SYBR Green fluorescence using the Applied Biosystems QuantStudio 6 Flex Real-Time PCR System (Life Technologies Corporation, Carlsbad, CA, USA). After an initial denaturing at 95°C for 2 minutes, reactions were carried out at 95°C for 1 second and 60°C for 30 seconds for 40 cycles. Relative fold change of each targeted mRNA was determined by the 2^−ΔΔCt method.⁴⁵ 

Western Blot Analysis

Western blot analysis was performed as described in our previous study of retinal arterioles with slight modification.^26,46 Retinal venules (sample pooled from both eyes per pig) and neural retina tissue were isolated and sonicated in lysis buffer. The protein content of each lysate was determined with the BCA protein assay kit (Pierce, Rockford, IL, USA). Equal amounts of protein (2.5 μg for ET-1 receptors and 5 μg for ROCK isoforms and p38) were separated by Tris-glycine SDS-PAGE (4%–15% Tris-HCl Ready Gels; Bio-Rad, Hercules, CA, USA), transferred onto a nitrocellulose membrane, and incubated with rabbit anti-ET_AR or anti-ET_BR polyclonal antibody (1:250 dilution; Catalog nos. sc-33535 and sc-33537; Santa Cruz Biotechnology, Santa Cruz, CA, USA),²⁶ or mouse anti-ROCK1 or anti-ROCK2 monoclonal antibody (1:250 dilution; Catalog nos. sc-17794 and sc-398519; Santa Cruz Biotechnology). Membranes were stripped and reprobed with rabbit anti-p38 antibody (1:1000; Catalog no. sc-535; Santa Cruz Biotechnology), which we previously have shown is highly expressed in neural retina tissue.²⁶ After incubation with an appropriate secondary antibody (anti-rabbit or anti-mouse IgG, 1:1000; Catalog nos. 7074S and 7076S; Cell Signaling Technology, Danvers, MA, USA), the membranes were washed and developed by enhanced chemiluminescence (Pierce). 

Data Analysis

The maximal diameter of retinal venules was obtained at the end of each functional experiment by relaxing vessels with 0.1 mM sodium nitroprusside in EDTA (1 mM)-Ca²⁺-free PSS³⁹ at 5 cm H₂O intraluminal pressure. Diameter changes in response to agonists were normalized to the resting diameter and expressed as percentage changes in diameter.²⁶ Data are reported as mean ± SEM, and n represents the number of animals (1–2 vessels per pig per treatment group for functional studies). Student's t-test or repeated measures 2-way ANOVA followed by Bonferroni multiple-range test was used to determine the significance of experimental interventions, as appropriate (GraphPad Prism, Version 6.0; GraphPad Software, La Jolla, CA, USA). P < 0.05 was considered significant. 

Results

Retinal Venular Constriction to ET-1 and the Role of ET-1 Receptors

Porcine retinal venules (total 50 vessels, 1–2 per pig) were isolated and pressurized at 5 cm H₂O with average maximum diameter of 133 ± 2 μm. These vessels developed stable basal tone by constricting to approximately 92% of maximum diameter within 30 minutes at 36°C to 37°C. In one cohort, administration of ET-1 caused constriction of retinal venules in a concentration-dependent manner (Fig. 1A). The threshold concentration of ET-1 for venular constriction was approximately 1 pM and the vessels constricted to approximately 65% of their resting diameters at 10 nM, the highest ET-1 concentration tested. Images of the retinal venular constriction in response to 0.1 nM ET-1 are shown in Figure 1B. 

Figure 1

 

Vasomotor response of isolated and pressurized porcine retinal venules to ET-1. (A) Venular diameters were recorded before (R: resting diameter) and after 30-minute treatment (T) with an ET-1 receptor antagonist. In the absence of receptor antagonist, retinal venules constricted to ET-1 in a concentration-dependent manner (Control; n = 9). ET_AR antagonist BQ123 (1 μM; n = 5), but not ET_BR antagonist BQ788 (0.1 μM; n = 5), inhibited venular constriction to ET-1. Both antagonists did not alter resting basal tone. *P < 0.05 versus Control. (B) Representative images of a porcine retinal venule at resting diameter and after constriction in response to ET-1 (0.1 nM). (C) The response of retinal venules to ET_BR agonist sarafotoxin S6c was examined (n = 6).

Incubation of the retinal venules with the ET_AR antagonist BQ123 did not alter the resting vascular diameter but the vasoconstriction elicited by ET-1 was abolished, except for approximately 10% constriction remaining at the highest concentration (10 nM) of ET-1 (Fig. 1A). In the presence of ET_BR antagonist BQ788, neither the resting diameter nor the vasoconstriction to ET-1 was altered (Fig. 1A). The retinal venules failed to respond to the challenge of ET_BR agonist sarafotoxin S6c throughout the concentrations tested (Fig. 1C). 

ET-1 Receptor mRNA and Protein Expressions in Porcine Retinal Venules

ET_AR and ET_BR mRNAs were detected in retinal venules and in neural retina tissue devoid of retinal vessels (Fig. 2A). The neural retina tissue expressed a relatively low amount of ET-1 receptor mRNA compared with the venular tissue. At the protein level, immunoblotting showed that ET_ARs and ET_BRs were strongly expressed in retinal venules, whereas only modest expression of both ET-1 receptors was detected in neural retina tissue (Fig. 2B). 

Figure 2

 

Molecular analyses of ET-1 receptors in porcine retinal venules. (A) Equal amounts of total RNA isolated from porcine retinal venules (RV) and neural retina tissue (RT) were reverse transcribed and then analyzed by real-time PCR for detection of ET_AR, ET_BR, and GAPDH mRNAs. The ET_AR and ET_BR transcripts were normalized to GAPDH expression and presented as relative mRNA expression. n = number of pigs studied. (B) Equal amount of protein was loaded for Western blot analyses of ET_AR and ET_BR in RVs and neural RT from pigs. Data represent four independent experiments.

Roles of Extracellular Ca²⁺ Entry and L-VOCCs in ET-1-Induced Venular Constriction

In the presence of a Ca²⁺-free solution, retinal venules completely lost basal tone, and constriction of these vessels to ET-1 was nearly abolished (Fig. 3A). At 10 nM ET-1, the retinal venules constricted to approximately 88% of their resting diameter in the absence of extracellular Ca²⁺ versus 65% in the presence of extravascular Ca²⁺. The L-VOCC inhibitor nifedipine also reduced venular basal tone to the same level as in the Ca²⁺-free solution, but it did not affect the constriction of these vessels to ET-1 (Fig. 3A). Furthermore, the L-VOCC agonist Bay K8644 elicited a modest but significant constriction (8% reduction in resting diameter) of retinal venules (Fig. 3B), which was prevented in the presence of nifedipine (Fig. 3B). 

Figure 3

 

Roles of extracellular calcium and L-VOCCs in vasoconstriction of isolated retinal venules to ET-1. (A) Venular diameters were recorded before (R: resting diameter) and after 30-minute treatment (T) of the vessel with Ca²⁺-free solution (n = 10) or L-VOCC inhibitor nifedipine (1 μM; n = 7). In the absence of treatment (Control; n = 12), retinal venules constricted to ET-1 in a concentration-dependent manner. Both Ca²⁺-free solution and nifedipine abolished resting tone, and the venular constriction to ET-1 was attenuated in the Ca²⁺-free solution. The vasoconstrictor response to ET-1 remained intact in the presence of nifedipine. #P < 0.05 versus R; *P < 0.05 versus Control. (B) The effect of L-VOCC agonist Bay K8644 (6 μM) on resting diameter (R) was examined in the absence (n = 3) or presence of nifedipine (1 μM; n = 3). *P < 0.05 versus R.

Roles of Rho Kinase and PKC in ET-1-Induced Venular Constriction

The Rho kinase inhibitor H-1152 did not significantly alter the resting diameter, but it abolished ET-1–induced constriction of retinal venules (Fig. 4A). In contrast, PKC inhibitor Gö 6983 affected neither resting diameter nor vasoconstriction to ET-1 (Fig. 4A). Cumulative addition of PKC activator PDBu to the vessel bath also did not significantly affect the resting diameter of retinal venules (Fig. 4B). Immunoblotting showed that both ROCK1 and ROCK2 isoforms were strongly expressed in retinal venules, whereas only faint expression of these proteins was detected in neural retina tissue (Fig. 4C). Sufficient neural retina protein was present in these samples because p38 protein expression was strikingly stronger in neural retina tissue than in venules (Fig. 4C). 

Figure 4

 

Roles of ROCK and PKC in vasoconstriction of isolated retinal venules to ET-1. (A) Venular diameters were recorded before (R: resting diameter) and after 30-minute treatment (T) with ROCK inhibitor H-1152 (3 μM; n = 6) or PKC inhibitor Gö 6983 (10 μM; n = 7). In the absence of treatment (Control; n = 12), retinal venules constricted to ET-1 in a concentration-dependent manner. The resting diameter of retinal venules was not altered by either drug treatment. The ET-1–induced constriction was abolished by H-1152 but not Gö 6983. *P < 0.05 versus Control. (B) Venular diameters were recorded under resting conditions (R) and at different concentrations of PKC activator PDBu (n = 6). (C) Equal amount of protein was loaded for Western blot analyses of ROCK isoforms (ROCK1 and ROCK2) and p38 in retinal venules (RV) and neural retina tissue (RT). Data represent three independent experiments.

Discussion

Although venules are known to play a critical role in regulation of postcapillary pressure and capillary fluid exchange,^2,3 there is limited study on the direct vasomotor function and cognate mechanisms of retinal venules. The salient findings of the present study are that pressurized porcine retinal venules exhibit basal tone and constrict to ET-1 through ET_AR activation. The entry of extracellular Ca²⁺ and activation of ROCK signaling, independent of L-VOCC and PKC pathways, appear to mediate the venular constriction to ET-1. We also found that extracellular Ca²⁺ is important in maintenance of basal tone, linking to L-VOCC activity. 

Studies using isolated vessel preparations have shown that ET-1 directly causes constriction of small arterioles²⁶ and veins⁹ isolated from the porcine retina. Our current study corroborates the latter report on porcine retinal veins, as well as extends this earlier finding by demonstrating venular constriction to ET-1 and characterizing the underlying mechanism. Our findings show that porcine retinal venules develop stable basal tone of approximately 8% reduction in diameter from its maximum value. This level of basal tone is slightly less than the nearly 20% tone reported in isolated porcine coronary venules⁸ and significantly less than the 50% to 60% tone developed in porcine and human retinal arterioles.²⁷ Our previous study in isolated retinal arterioles from pigs showed that the threshold concentration for ET-1–induced constriction is 1 pM, and the 10 nM concentration elicits a 75% reduction in resting diameter.²⁶ In the present study, retinal venules exhibited a similar threshold concentration of 1 pM in response to ET-1, but constricted less robustly than the arterioles with only 35% reduction in resting diameter with 10 nM ET-1 (Fig. 1). This venular responsiveness is comparable to that observed in a recent ET-1 study in isolated retinal veins from pigs⁹ and is consistent with the report of the greater amount of smooth muscle in retinal arteries than retinal veins.⁹ It appears that vascular smooth muscle, albeit a thin layer, in the porcine retinal venule is sufficient to exert an active response to ET-1. These vasoconstrictor responses appear pathophysiologically relevant, because the concentrations of ET-1 used in the current study were within the clinical and experimental range reported for vitreous fluid (picomolar range)^47,48 and the estimated level at the local microvasculature (nanomolar range).⁴⁹ The ability of retinal venules to develop basal tone and to constrict in response to ET-1 suggests that these vessels may contribute to the physiological and/or pathophysiological regulation of flow resistance, local pressure, and fluid exchange in the retinal microcirculation. 

In the current study, the functional role of specific ET-1 receptor subtypes was characterized by examining the vascular response to ET-1 in the presence of ET_AR or ET_BR antagonist. The concentration-dependent constriction of retinal venules was nearly abolished by pharmacologic ET_AR blockade (Fig. 1). These results are consistent with the major contribution of ET_ARs in retinal arteriolar constriction to ET-1 reported in our previous studies.²⁶ By contrast, ET_BR blockade (BQ778) did not affect ET-1–induced venular constriction, suggesting that ET_BR does not contribute to this vasomotor response. This contention is supported by the efficacy of BQ778 and the observed unresponsiveness of the vessel to the ET_BR agonist sarafotoxin. Sarafotoxin did not alter resting vascular tone or cause vasoconstriction in the present study (Fig. 1). Moreover, the same concentrations of sarafotoxin were reported in our previous study to elicit significant constriction of porcine retinal arterioles in a manner sensitive to ET_BR blockade (BQ778).²⁶ It appears that ET_BR plays little role in retinal venular constriction to ET-1, a mechanism that differs from their upstream arterioles, in which both ET_ARs and ET_BRs contribute to the vasoconstriction to ET-1,^26,28 at least under normal physiological conditions. At the molecular level, we showed expression of both ET_AR and ET_BR mRNA and protein in isolated porcine retinal venules (Fig. 2). The mRNA and protein expressions of both ET-1 receptor subtypes were also detected in the neural retina tissue but were strikingly lower than those in the retinal venules. Expression of both ET-1 receptor subtypes in the neural retina layers is also supported by the immunohistochemical data in pigs⁵⁰ and rodents,^51–53 but the quantitative comparison with retinal vascular tissue is not available. Although we can detect the mRNA and protein levels in retinal venules, the distinct cellular distribution, that is, endothelial versus smooth muscle cells, of ET_ARs and ET_BRs in the retinal venules requires further investigation. Collectively, our functional and molecular findings support the dominant role of ET_ARs in mediating ET-1–induced constriction of porcine retinal venules. 

Vascular smooth muscle contraction is regulated by changes in intracellular Ca²⁺ with Ca²⁺ entry occurring through several types of channels, including L-VOCCs.^54,55 Although activation of ET_ARs elicits Ca²⁺ signaling in blood vessels, the specific channels involved are diverse, depending on the vascular bed and species.¹² The potential roles of extracellular Ca²⁺ and L-VOCCs in basal tone and ET-1–induced constriction of retinal venules have not been explored previously. In the present study, the resting basal tone was abolished in the absence of extracellular Ca²⁺ or during exposure to nifedipine, suggesting that extracellular Ca²⁺ and L-VOCCs are indispensable for maintenance of basal tone at physiologic intraluminal pressure. The constriction of retinal venules to ET-1 also required Ca²⁺ entry, because ET-1 failed to elicit vasoconstriction in the absence of extracellular Ca²⁺ (Fig. 3). However, the inability of nifedipine to prevent the ET-1–induced vasoconstriction indicates that L-VOCCs do not mediate the extracellular Ca²⁺ entry. The efficacy of nifedipine in blocking L-VOCCs was supported by the obliteration of the venular constriction to L-VOCC activator Bay K8644 (Fig. 3). Our finding with ET-1 is consistent with the lack of an effect of L-VOCC blockade on the ET-1–induced contraction of rat mesenteric veins⁵⁶ and our previous results with porcine retinal arterioles.²⁹ It seems that the pathway of Ca²⁺ entry for vasoconstriction in response to ET-1 in retinal venules is distinct from that used for basal tone maintenance in that the former does not involve L-VOCCs. The specific smooth muscle channels responsible for ET-1–induced constriction in retinal venules remains unclear, and it will be the subject of future investigation. 

It is generally accepted that the process of smooth muscle contraction is coupled to the level of myosin light chain (MLC) phosphorylation,⁵⁷ which is modulated by Ca²⁺-dependent activity of MLC kinase^38,58,59 and by MLC phosphatase.⁶⁰ Protein kinases such as ROCK and PKC can inhibit MLC phosphatase, leading to enhanced MLC phosphorylation^61–63 and subsequent contraction of vascular smooth muscle.^38,60 Both ROCK and PKC have been shown to be possible signaling molecules modulating contractile myofilament sensitivity to Ca²⁺, thus regulating the force of smooth muscle contraction.^64,65 However, it is unknown whether ET-1 also uses these signaling molecules in the retinal venules to exert its contractile action. We found that specific pharmacological blockade of ROCK prevented constriction of porcine retinal venules to ET-1 without altering basal tone (Fig. 4). In the presence of PKC inhibitor Gö 6983, both resting tone and ET-1–induced vasoconstriction remained intact. Moreover, the retinal venules were not responsive to the PKC activator PDBu (Fig. 4). This observation contrasts with potent PDBu-induced constriction of porcine retinal arterioles, which was blocked by PKC inhibitor Gö 6983 in our previous studies.^28,29 Overall, these results suggest that activation of ROCK, but not PKC, signaling is involved in the venular constriction to ET-1. Furthermore, we detected both ROCK isoforms (ROCK1 and ROCK2) at the protein level in retinal venules. Although the ROCK2 isoform has been suggested to play a major role in smooth muscle contraction,⁶⁶ the cellular distributions and their individual functional linking to extracellular Ca²⁺ entry in retinal venules remain unknown and warrant further investigation. 

A fundamental understanding of vasomotor regulation mechanisms of retinal venules in response to ET-1 is important, because increased local or plasma levels of this peptide in the retina have been implicated in the pathogenesis of RVO and diabetic retinopathy.^32,67,68 A complication of RVO⁶⁹ and late-stage diabetic retinopathy⁷⁰ is the onset of edema and fluid accumulation in the retina contributing to neural retina dysfunction and blindness. Constriction of retinal veins by ET-1 may increase retinal venous pressure, which could promote edema and reduction in retinal perfusion pressure under disease states.⁶⁸ This notion is supported by clinical studies reporting elevation of retinal venous pressure and reduced retinal blood flow in patients with RVO^71,72 and diabetic retinopathy,^73–75 including those with diabetic macular edema.⁷⁶ Taken together, these clinical observations along with our current findings underpin the concept that retinal vein constriction to elevated levels of ET-1 contributes to the pathogenesis of retinal diseases, such as RVO and diabetes.^33,67,68,73 Evaluation of the impact of experimental diabetes⁷⁷ on ET-1–induced constriction of retinal venules could help corroborate this hypothesis. 

In summary, we found that isolated porcine retinal venules develop modest basal tone and constrict markedly to ET-1 in an extracellular Ca²⁺-dependent manner. Although L-VOCCs play a critical role in maintaining basal tone of retinal venules, they do not contribute to the ET-1–induced vasoconstriction. Retinal venules express both ET_ARs and ET_BRs, but ET_ARs play a dominant role in vasoconstriction to ET-1. It appears that activation of ROCK but not PKC signaling mediates the venular constriction to ET-1. This study provides important insight into the mechanisms of ET-1–induced constriction of retinal venules and lays a foundation for future research to better understand vasomotor regulation of these microvessels under physiological and pathophysiological conditions in the retina. 

Acknowledgments

Supported by National Institutes of Health National Eye Institute Grants R01EY023335 and R01EY024624 (TWH), Retina Research Foundation (TWH, LK), the Baylor Scott & White-Central Texas Foundation, Ophthalmic Vascular Research Program of Baylor Scott & White Health (LK), and the Kruse Chair Endowment (LK). 

Disclosure: Y.-L. Chen, None; Y. Ren, None; W. Xu, None; R.H. Rosa Jr, None; L. Kuo, None; T.W. Hein, None