Dilation of Porcine Retinal Arterioles to Cilostazol: Roles of eNOS Phosphorylation via cAMP/Protein Kinase A and AMP-Activated Protein Kinase and Potassium Channels

Abstract

Purpose.: Cilostazol, a selective inhibitor of phosphodiesterase 3, has antiplatelet aggregation and peripheral vasodilation effects. We examined the effects of cilostazol on the retinal microvascular diameter to determine its dependence on the endothelium and/or smooth muscle to reveal the signaling mechanisms involved in this vasomotor activity.

Methods.: Porcine retinal arterioles were isolated, cannulated, and pressurized without flow in vitro. Video microscopic techniques recorded the diametric responses to cilostazol.

Results.: The retinal arterioles dilated in response to cilostazol in a dose-dependent (100 pM–10 μM) manner; the dilation decreased by 60% after endothelial removal. The nitric oxide (NO) synthase inhibitor, NG-nitro-L-arginine methyl ester (L-NAME), inhibited cilostazol-induced vasodilation comparable to denudation. Inhibition of soluble guanylyl cyclase and blockade of protein kinase A (PKA) were comparable to L-NAME. Compound C, an AMP-activated protein kinase (AMPK) inhibitor, partially inhibited cilostazol-induced vasodilation, which exhibited a weaker inhibitory effect on cilostazol-induced vasodilation than blockade of PKA. The large-conductance Ca²⁺-activated K channel (BK_Ca channel) blocker, iberiotoxin, also inhibited cilostazol-induced vasodilation. The residual vasodilation decreased further with co-administration of L-NAME and iberiotoxin.

Conclusions.: Cilostazol elicits endothelium-dependent and -independent dilation of the retinal arterioles mediated by NO release and BK_Ca channel activation, respectively. Endothelial nitric oxide synthase (eNOS) phosphorylation via the cAMP/PKA and AMPK pathways and consequent activation of the soluble guanylyl cyclase/cyclic guanosine monophosphate pathway might play an important role in cilostazol-induced vasodilation of the retinal arterioles.

Introduction

Cilostazol, a selective inhibitor of phosphodiesterase 3 (PDE3), is an antiplatelet drug with antiplatelet aggregation and peripheral vasodilatory effects. ¹ It is used to treat ischemic symptoms related to chronic peripheral arterial obstruction and for secondary prevention of cerebral infarction. ^1,2 Cilostazol also has the potential to prevent cardiovascular events in patients with type 2 diabetes, ^3,4 indicating that cilostazol might protect against diabetes-related vascular dysfunction including that in the retinal vessels. Because we recently reported that the retinal blood flow (RBF) is impaired in patients with type 2 diabetes mellitus with mild and no diabetic retinopathy (DR), ⁵ improvement of the impaired retinal circulation could lead to new therapeutic modalities for DR in patients with type 2 diabetes. In addition, a few in vivo animal studies have reported that cilostazol increased the retinal, choroidal, and optic nerve head blood flow, ^6,7 suggesting that cilostazol has the potential to improve the ocular circulation. However, the underlying mechanisms of the effect of cilostazol on the retinal circulation remain unclear. Based on those findings, we examined the effect of cilostazol on the retinal microvascular diameter using an isolated vessel technique to exclude the confounding effects of metabolic, hemodynamic, humoral, and glial/neuronal factors associated with in vivo experiments. 

Materials and Methods

Animal Preparation

The Animal Care Committee of Asahikawa Medical University approved all animal procedures, which were performed in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. The eyes were enucleated immediately from pigs of either sex (age, 16–24 weeks; weight, 10–15 kg) after they were killed in a local slaughterhouse and the eyes were transported to the laboratory in a moist chamber on ice. 

Isolation and Cannulation of Microvessels

The techniques for identification and isolation of retinal microvessels have been described previously. ^8–11 Briefly, the isolated retinal arterioles (90–130 μm in situ) were cannulated with a pair of glass micropipettes and pressurized to 55 cm H₂O (∼40 mm Hg) intraluminal pressure without flow using two independent pressure reservoir systems. ¹² The vasomotor activity of the isolated vessels was recorded continuously using video microscopic techniques ⁸ throughout the experiments. 

Control Experiment

Cannulated, pressurized arterioles were bathed in physiologic salt solution (PSS) with albumin (0.1%) at 36° to 37°C to allow development of stable basal tone; after approximately 30 to 40 minutes, concentration-dependent vasodilation in response to cilostazol (100 pM–10 μM) developed. After measurement of the control dose response of cilostazol without drugs, the vessels were washed with PSS to allow redevelopment of the basal tone. In some studies, the vasodilation elicited by cilostazol was re-examined after 30 minutes to confirm the reproducibility of the response (n = 6). 

Endothelial Denudation

The role of the endothelium in cilostazol-induced dilation was evaluated by comparing the responses before and after endothelial removal, which we reported previously. ¹¹ Briefly, the vessel was perfused intraluminally with a nonionic detergent, CHAPS (0.4%), for 1 to 2 minutes to remove the endothelial cells, ^10,13 which we confirmed previously by immunohistochemical staining. ¹¹  

Mechanistic Studies of Cilostazol-Induced Dilation

In the first series of studies, we assessed the involvement of endothelium-derived vasodilators (i.e., prostaglandins, nitric oxide [NO], and cytochrome P450 metabolites) in mediating the vascular response in the presence of known effective concentrations of specific enzyme inhibitors indomethacin (10 μM), ^9,14 N^G-nitro-L-arginine methyl ester (L-NAME, 10 μM), ^8,9 and sulfaphenazole (10 μM), ¹⁵ respectively. We also assessed the effects of charybdotoxin (0.1 μM), an inhibitor of intermediate-conductance calcium (Ca)-dependent K channels (IK_Ca) plus apamin (0.1 μM), an inhibitor of small-conductance Ca-activated K channels (SK_Ca). These channels are required for activation of endothelium-derived hyperpolarizing factor (EDHF)-type relaxation. ^16–18 The role of guanylyl cyclase/cyclic guanosine monophosphate (cGMP) signaling was assessed by treating the vessels with the soluble guanylyl cyclase inhibitor, 1H-1,2,4-oxadiazolo[4,3-a]quinoxalin-1-one (ODQ, 0.1 μM). ^9,10 In the second series of studies, to examine the involvement of protein kinase A (PKA) and AMP-activated protein kinase (AMPK), we studied the cilostazol-induced response after incubation with the PKA inhibitor, Rp-8-Br-cAMPS (100 μM), ¹⁹ and the AMPK inhibitor, compound C (10 μM). ²⁰  

Roles of Potassium Channels in Cilostazol-Induced Dilation

To study the involvement of the K channels, we examined these pathways by treating the vessels with various K channel inhibitors: the nonselective K channel blocker, tetraethylammonium (TEA, 10 mM) ²¹ ; the large-conductance Ca²⁺-activated K channel (BK_Ca channel) blocker, iberiotoxin (0.1 μM) ^11,22–24 ; and the adenosine triphosphate-sensitive K channel blocker, glibenclamide (5 μM). ⁹  

Role of PKA and AMPK

To examine the direct contribution of PKA to the retinal arterioles, we studied the vasodilatory effect of the PKA agonist, forskolin (0.1 μM), ¹⁹ before and after incubation with Rp-8-Br-cAMPS, L-NAME, iberiotoxin, or endothelial denudation. We also examined the vasodilatory effect of the AMPK agonist, 5-aminoimidazole-4-carboxamide 1-β-D-ribofuranoside (AICAR; 3 mM), ^25,26 before and after incubation with compound C, L-NAME, endothelial denudation, or iberiotoxin. These experimental protocols were examined by a method similar to that of the cilostazol trials. 

Response to Sodium Nitroprusside

Sodium nitroprusside (SNP; 0.1 μM–100 μM) was used to probe endothelium-independent NO-mediated vasodilation. The vascular response to SNP was examined in the presence of various interventions (Table). 

Table. 

 

Resting Diameters and Diametric Responses of Retinal Arterioles to SNP

		Resting Diameter	Sodium Nitroprusside, μM
			0.1	1	10	100
Control	24	62.1 ± 2.9	4.2 ± 1.4	16.1 ± 2.2	51.5 ± 6.1	86.1 ± 3.5
Denudation	4	57.7 ± 7.0	6.0 ± 1.5	25.2 ± 2.9	57.0 ± 6.3	87.4 ± 3.4
L-NAME	4	58.5 ± 7.3	4.6 ± 2.3	28.6 ± 6.0	59.1 ± 4.9	88.7 ± 2.9
Rp-8-Br-cAMPS	4	60.3 ± 8.0	8.9 ± 4.1	23.3 ± 5.0	47.3 ± 8.2	80.1 ± 3.0
Compound C	4	60.5 ± 6.3	4.5 ± 1.9	19.2 ± 4.0	40.7 ± 8.8	85.9 ± 1.5
Compound C + Rp-8-Br-cAMPS	4	62.0 ± 4.3	6.9 ± 1.8	21.1 ± 4.1	49.1 ± 3.5	81.3 ± 1.9
Compound C + Rp-8-Br-cAMPS + L-NAME	4	64.2 ± 7.0	5.5 ± 1.4	16.6 ± 1.8	53.3 ± 7.7	87.3 ± 2.3
TEA	4	65.0 ± 7.4	5.2 ± 2.9	18.6 ± 2.1	55.7 ± 6.8	91.3 ± 1.2
TEA + L-NAME	4	59.8 ± 6.6	6.0 ± 2.1	25.9 ± 4.3	49.8 ± 2.3	86.4 ± 2.2
Iberiotoxin	4	58.5 ± 5.7	6.3 ± 3.5	19.8 ± 3.2	45.9 ± 4.2	82.7 ± 3.3
Iberiotoxin + L-NAME	4	59.2 ± 6.8	5.8 ± 2.3	23.4 ± 2.5	50.5 ± 5.1	88.2 ± 4.1

 Data are expressed as the mean percentage of maximal dilation ± SE. n, number of vessels. There are no significant changes in the resting diameters. Based on two-way ANOVA, the responses to SNP are unaffected by any perturbations compared with the control.

All drugs were administered extraluminally unless otherwise stated. Each pharmacologic inhibitor was incubated with the vessels for at least 30 minutes. 

Chemicals

All chemicals were obtained from Sigma-Aldrich (St. Louis, MO), except for compound C, which was obtained from Calbiochem (San Diego, CA). L-NAME, apamin, charybdotoxin, TEA, iberiotoxin, Rp-8-Br-cAMPS, AICAR, and SNP were dissolved in PSS. Indomethacin, sulfaphenazole, and ODQ were dissolved in ethanol. Cilostazol, glibenclamide, forskolin, and compound C were dissolved in dimethyl sulfoxide (DMSO). Subsequent dilutions of these drugs were prepared in PSS. The final concentration of ethanol or DMSO in the vessel bath was 0.1%. ⁹ Vehicle control studies indicated that this final concentration of solvent had no effect on the arteriolar function. 

Data Analysis

At the end of each experiment, the vessels were relaxed in ethylenediaminetetraacetic acid (1 mM) calcium-free PSS to obtain the maximal diameter at 55-cm H₂O intraluminal pressure. ^8,10 The diametric changes in response to cilostazol and SNP were normalized to the resting diameters and expressed as the percent changes in diameter compared with the maximal dilation. ^8,10 The data are expressed as the mean ± SE, and n represents the number of vessels studied. Statistical comparisons of the changes in resting tone caused by antagonists were performed using the Student's t-test. Two-way ANOVA, followed by the Bonferroni multiple-range test, was used to determine the significance of the difference between the first control experiments, and one-way analysis was used to determine the significance of the difference between control and experimental interventions. P less than 0.05 were considered significant. 

Results

Dilation of Retinal Arterioles Induced by Cilostazol

The basal tone in all vessels (n = 102) ranged from 55% to 70% (average, ∼61% ± 3%) of their maximal diameter (Table). The average resting and maximal vessel diameters were 60 ± 4 and 99 ± 4 μm, respectively. There were no significant changes in the resting tone after any interventions. Cilostazol induced consistent concentration-dependent dilation of the retinal arterioles. The highest concentration (10 μM) elicited approximately 40% of the maximal dilation (Fig. 1). Further study showed that cilostazol-induced dilation was reproducible and did not deteriorate after repeated application (first trial, resting diameter, 61 ± 3 μm; maximal diameter, 96 ± 5 μm; second trial, resting diameter, 60 ± 5 μm; maximal diameter, 96 ± 5 μm) (Fig. 1). 

Figure 1. 

 

Dilatation as a function of the cilostazol concentration. There is no significant difference between the two repeated trials (P = 0.83, n = 6).

Role of the Endothelium

The cilostazol-induced dilation decreased partly and the response to the highest cilostazol concentration significantly (P < 0.001) decreased from 40% to 15% (Fig. 2). 

Figure 2. 

 

Role of the endothelium in retinal arteriolar dilation in response to cilostazol (10 μM). Endothelium removal (n = 4), incubation with L-NAME (10 μM, n = 4) or ODQ (0.1 μM, n = 4), but not indomethacin (10 μM, n = 4), sulfaphenazole (10 μM, n = 4), or the combination of charybdotoxin (0.1 μM) and apamin (0.1 μM, n = 4) significantly reduced vasodilation in response to cilostazol. *P < 0.05 versus control.

Role of Endothelium-Derived Factors

The nitric oxide synthase (NOS) inhibitor L-NAME significantly (P < 0.001, Fig. 2) reduced cilostazol-induced vasodilation, which was comparable to that produced by denudation (L-NAME versus denudation (P > 0.05). Inhibition of cytochrome P450 epoxygenase, prostaglandins, and the combination of IK_Ca and SK_Ca by sulfaphenazole, indomethacin, and apamin plus charybdotoxin did not affect the vasodilatory response to cilostazol. ODQ significantly (P < 0.001) reduced the vasodilatory response to cilostazol in a manner similar to L-NAME. 

Role of K Channels

TEA significantly (P < 0.001) inhibited cilostazol-induced vasodilation of the retinal arterioles (Fig. 3) In addition, iberiotoxin attenuated cilostazol-induced dilation of the retinal arterioles in a manner similar to that of TEA, whereas glibenclamide was ineffective (Fig. 3). The residual vasodilation in the presence of TEA or iberiotoxin decreased further with subsequent L-NAME treatment (Fig. 3). 

Figure 3. 

 

Role of potassium channel in retinal arteriolar dilation in response to cilostazol (10 μM). TEA (10 mM, n = 4) and iberiotoxin (0.1 μM, n = 4), but not glibenclamide (5 μM, n = 4), reduced vasodilation in response to cilostazol (10 μM). Residual vasodilation in the presence of TEA or iberiotoxin decreases further following co-incubation with L-NAME (10 μM, n = 4 each). *P < 0.05 versus control, †P < 0.05 versus TEA, ‡P < 0.05 versus iberiotoxin.

Roles of PKA and AMPK

Cilostazol-induced vasodilation decreased significantly with pretreatment with the PKA blocker Rp-8-Br-cAMPS (P < 0.001) and the AMPK blocker compound C (P < 0.001, Fig. 4). The residual vasodilation in the presence of compound C further significantly (P < 0.001) decreased with subsequent treatment with Rp-8-Br-cAMPS or a combination of Rp-8-Br-cAMPS and L-NAME, whereas there was no significant change between the presence of Rp-8-Br-cAMPS and Rp-8-Br-cAMPS plus iberiotoxin. The combination of Rp-8-Br-cAMPS and iberiotoxin did not cause further reduction of the vasodilatory response to cilostazol compared with Rp-8-Br-cAMPS alone. 

Figure 4. 

 

Rp-8-Br-cAMPS (100 μM, n = 4) and compound C (10 μM, n = 4) decrease vasodilation in response to cilostazol (10 μM). Residual vasodilation in the presence of compound C decreases further after co-incubation with Rp-8-Br-cAMPS or Rp-8-Br-cAMPS and L-NAME. *P < 0.05 versus control, †P < 0.05 versus compound C.

The PKA activator, forskolin, induced dilation of the retinal arterioles (Fig. 5). The dilatory effect of forskolin was reduced partly by pretreatment with iberiotoxin, and the residual vasodilation in the presence of iberiotoxin further significantly (P < 0.001) decreased with subsequent treatment with L-NAME. Pretreatments with L-NAME and endothelial removal also significantly reduced (P < 0.001) the forskolin-induced vasodilation and almost abolished the pretreatment with Rp-8-Br-cAMPS. 

Figure 5. 

 

Vasodilatory responses of isolated porcine retinal arterioles to forskolin (0.1 μM, n = 16). Rp-8-Br-cAMPS (100 μM, n = 4) negated the forskolin-induced vasodilation and L-NAME (10 μM, n = 4), endothelium denudation (n = 4) and iberiotoxin (0.1 μM, n = 4) reduced the forskolin-induced vasodilation. Residual vasodilation in the presence of iberiotoxin decreases further following co-incubation with L-NAME (n = 4). *P < 0.05 versus control.

The AMPK agonist AICAR induced vasodilation of the retinal arterioles, and the dilatory effect was reduced significantly (P < 0.001) by pretreatment with compound C or L-NAME and endothelial removal (Fig. 6). 

Figure 6. 

 

Vasodilatory responses of isolated porcine retinal arterioles to AICAR (3 mM, n = 12). L-NAME (10 μM, n = 4), endothelium denudation (n = 4), and compound C (10 μM, n = 4) almost abolish the AICAR-induced vasodilation. *P < 0.05 versus control.

Response to SNP

Various interventions did not affect the SNP-induced dilation of the retinal arterioles (Table), suggesting that the vascular smooth muscle function was unaltered by these interventions. 

Discussion

The current study showed that cilostazol, a PDE3 inhibitor, induced concentration-dependent vasodilation of the isolated porcine retinal arterioles (Fig. 1). Because the plasma cilostazol concentration reaches 2.1 μM within 3 hours after oral administration of 100 mg in healthy men, ²⁷ the current data clearly showed that cilostazol may have clinical potential to elicit 30% to 40% vasodilation of the retinal arterioles at these concentrations (Fig. 1). It was reported that infusion of cilostazol into the internal carotid artery had approximately a 30% vasodilatory effect on the retinal arterioles in rats. ⁷ Although that study did not report the exact concentration of cilostazol in the retinal arterioles, the results were comparable to the current findings, indicating that such vasodilating agents may be used in retinal diseases in which vasoconstriction is problematic. 

Although previous studies have reported conflicting results that cilostazol-induced vasodilation is endothelium-independent in cerebral and spinal arterioles ^28–30 or both endothelium-dependent and endothelium-independent in the thoracic aorta, ³¹ we observed that removing the vascular endothelium with CHAPS significantly decreased, but did not eliminate, the cilostazol-induced vasodilation (Fig. 2). This suggested that there are both endothelium-dependent and endothelium-independent pathways in the cilostazol-induced vasodilation of the retinal arterioles. 

We also found that the inhibitory effects of NOS blockade by L-NAME on cilostazol-induced vasodilation (Fig. 2) were comparable to those seen with denudation, suggesting that cilostazol caused vasodilation via NO production from the endothelium in the retinal arterioles. Furthermore, cilostazol caused vasodilation of isolated rat thoracic aortas via NO production ³¹ and increased NO production in cultured vascular endothelial cells. ^32,33 Therefore, we speculated that NO greatly contributes to the endothelium-dependent component of cilostazol-induced vasodilation of the retinal arterioles. 

In contrast to L-NAME, the vasodilatory response to cilostazol was unaffected by inhibition of synthesis of prostaglandins (Fig. 2). In addition, cilostazol-induced vasodilation was unaffected by pretreatment with the cytochrome P450 metabolite inhibitor, sulfaphenazole, ¹⁵ and the specific K channel blockers, IK_Ca blocker charybdotoxin and SK_Ca blocker apamin, ^16–18 indicating that EDHF might not be involved in cilostazol-induced vasodilation in the retinal arterioles (Fig. 2). Taken together, NO primarily contributes to the endothelium-dependent component of cilostazol-induced vasodilation of the retinal arterioles independent of prostaglandins and EDHF. 

Among the pathways that are involved in NO vasodilation, the NO/cGMP pathway is considered a major vasodilatory mechanism. ³⁴ Indeed, inhibition of cilostazol-induced vasodilation by ODQ incubation was comparable to L-NAME (Fig. 2), suggesting that cilostazol may dilate the retinal arterioles via the NO/cGMP pathway. 

Previous studies have reported involvement of various K channels in the vasodilation of the retinal arterioles. ^{8,9,11,21,22,35,36} The current data showed that the BK_Ca channel blocker, iberiotoxin, inhibited cilostazol-induced vasodilation (Fig. 3) in the same manner as the nonselective K channel blocker, TEA, whereas this vasodilation was unaffected by glibenclamide, suggesting that the BK_Ca channel, which can produce membrane hyperpolarization that results in relaxation of smooth muscle, ³⁷ may be involved primarily in cilostazol-induced vasodilation of the retinal arterioles. Although the endothelial K_Ca channel might regulate endothelial nitric oxide synthase (eNOS) activity, ³⁸ the current finding that the combination of L-NAME and iberiotoxin further reduced dilation in response to cilostazol compared with iberiotoxin (Fig. 2) and L-NAME alone (L-NAME and iberiotoxin versus L-NAME alone; 7.8 ± 2.7% vs. 13.1 ± 3.9%; P < 0.05) suggested that the BK_Ca channel might be involved in the endothelium-independent, smooth muscle-dependent vasodilation of the retinal arterioles in response to cilostazol. 

Endothelial cell-derived NO is synthesized continuously in the endothelium from L-arginine by activation of eNOS. Although regulation of eNOS activity has been previously thought to be simply Ca²⁺-calmodulin dependent, eNOS activity is regulated by protein phosphorylation in an intracellular Ca²⁺-independent manner. ³⁹ PKA is a protein kinase that induces phosphorylation of eNOS and stimulates NO production without an increase in intracellular Ca²⁺. ⁴⁰ The current study showed that the PKA inhibitor, Rp-8-Br-cAMPS, reduced the cilostazol-induced vasodilation of the retinal arterioles (Fig. 4). Because cilostazol induced PKA-dependent eNOS phosphorylation and NO release in human aortic endothelial cells, ³² it is likely that cilostazol induces vasodilation of isolated porcine retinal arterioles by NO production from the retinal vascular endothelium via activation of PKA and phosphorylation of eNOS in the retinal vascular endothelial cells. We also found that the combination of Rp-8-Br-cAMPS and iberiotoxin did not further reduce the cilostazol-induced vasodilation compared with Rp-8-Br-cAMPS alone (Fig. 4), suggesting that Rp-8-Br-cAMPS may block both the endothelial and smooth muscle components of the cilostazol response in the retinal arterioles. Taken together, activation of PKA induced by cilostazol may be associated with not only eNOS phosphorylation in the endothelium, but also activation of the BK_Ca channel in the smooth muscle of the retinal arterioles. 

Because inhibiting PDE3 by cilostazol increases intracellular cAMP, ¹ which can activate PKA, it is important to confirm the role of the cAMP/PKA pathway on cilostazol-induced vasodilation in the retinal arterioles. Therefore, we examined the direct effect of activation of PKA by the PKA activator, forskolin, ¹⁹ and found that forskolin induced endothelium-dependent, NO-mediated vasodilation (Fig. 5), which was comparable to a previous study that reported that forskolin induced endothelium-dependent vasodilation in rat aorta. ^41,42 We also found that the dilation in response to forskolin was inhibited partly after treatment with iberiotoxin (Fig. 5). Because PKA also activated K_Ca channels in smooth muscle, ^43,44 it appears that cilostazol may cause vasodilation by activation of PKA mainly via eNOS phosphorylation in the endothelium, but partly in the BK_Ca channels in the smooth muscle cells in the retinal arterioles. 

In addition to the PKA-related phosphorylation of eNOS, AMPK also plays an important role in phosphorylation of eNOS in the endothelial cells. ⁴⁵ Although AMPK is a modulator of cellular energy status and metabolism, ⁴⁶ it recently has emerged as a potential regulator of vascular function. ^25,26,47 The current study showed that the AMPK inhibitor, compound C, partially reduced cilostazol-induced vasodilation of the retinal arterioles, suggesting that cilostazol elicits vasodilation via an AMPK-dependent mechanism (Fig. 4). We also found that the AMPK activator, AICAR, ^25,26 induced endothelium-dependent, NO-mediated vasodilation of the retinal arterioles (Fig. 6), suggesting that activation of AMPK, per se, may play a role in NO production from the endothelium via phosphorylation of eNOS induced by cilostazol. 

Because we observed slight, but significant, residual vasodilation of the retinal arterioles in response to cilostazol at the highest concentration (10⁻⁵ M) compared with a lower concentration (10⁻¹⁰ M) after incubation with TEA and L-NAME (Fig. 3), it is likely that there may be other possible mechanisms of action of the vasodilating effect of cilostazol. We reported previously that fenofibrate-induced vasodilation of the retinal arterioles is associated with the AMPK and PI3-kinase pathways, ⁴⁸ which can activate phosphorylation of eNOS. ^49,50 However, our preliminary data did not support the hypothesis that the PI3 pathway is involved in the cilostazol-induced vasodilation via phosphorylation of eNOS, because the PI3 kinase inhibitor, wortmannin, did not inhibit vasodilation of the retinal arterioles in response to cilostazol (data not shown). Further study is needed to elucidate the mechanism of the residual vasodilation. 

In addition to the partially reduced cilostazol-induced vasodilation by pretreatment with compound C, residual vasodilation was attenuated further with subsequent treatment with Rp-8-Br-cAMPS or a combination of Rp-8-Br-cAMPS and L-NAME (Fig. 4), suggesting that the impact of AMPK on cilostazol-induced vasodilation of the retinal arterioles might be weaker than that of the cAMP/PKA pathway. It was reported previously that AMPK is unlikely to play a major role in regulating eNOS under normal conditions in which PKA would be expected to dominate regulation of eNOS in murine or porcine aortic endothelial cells, ⁵¹ which may agree with the current findings. In addition, the results that there were no further inhibitory effects on the vasodilation in response to forskolin by compound C and to AICAR by Rp-8-Br-cAMPS (data not shown), suggested that there is no crosstalk between these pathways leading to eNOS phosphorylation. Taken together, AMPK plays at least some role in cilostazol-induced vasodilation, probably via activation of eNOS in the endothelium, of the retinal arterioles. These results might contribute to further studies that focus on the effect of other AMPK activators (e.g., fenofibrate, ⁵² metformin, ⁵³ or resveratrol ²² ) on the retinal microcirculation. In addition, as discussed previously, our data suggested that cilostazol may cause vasodilation by activation of PKA via mainly eNOS phosphorylation in the endothelium, but partly by the BK_Ca channels in the smooth muscle cells in the retinal arterioles, which may be another possible explanation for the differences in the impact of the AMPK and PKA pathways on cilostazol-induced vasodilation of the retinal arterioles. 

In conclusion, we showed that cilostazol, a PDE3 inhibitor, elicits potent dilation of the retinal arterioles, which has two components of endothelium-dependent and endothelium-independent pathways. The overall meaning of our findings is that endothelium-dependent dilation and endothelium-independent dilation in response to cilostazol are mediated by NO induced by eNOS activation via the cAMP/PKA and AMPK pathways and consequent activation of the soluble guanylyl cyclase/cGMP pathway and activation of the BK_Ca channel, partly via PKA activation, in smooth muscle, respectively. Because the RBF is impaired in early-stage DR in patients with type 2 diabetes mellitus, ⁵ cilostazol-induced vasodilation might be a novel potential drug for treating DR. Because instillation of cilostazol eye drops reduced elevated IOP in rabbit eyes, ⁵⁴ topical cilostazol might be a novel future treatment for DR Further clinical study is needed to determine if cilostazol can improve impaired RBF in patients with type 2 diabetes and other retinal vascular disorders. 

Footnotes

 Supported by grants from a Grant-in-Aid for Scientific Research (C; 18591904, TN), Young Scientists (B; 23791956, AI) from the Ministry of Education, Science, and Culture, Tokyo, Japan, the Uehara Memorial Foundation, the Takeda Foundation, and the Akiyama Life Science Foundation (TN).

Footnotes

 Disclosure: I. Tanano, None; T. Nagaoka, None; T. Omae, None; A. Ishibazawa, None; T. Kamiya, None; S. Ono, None; A. Yoshida, None