Role of Calcium-Activated Potassium Channels with Small Conductance in Bradykinin-Induced Vasodilation of Porcine Retinal Arterioles

Abstract

purpose. Endothelial dysfunction and impaired vasodilation may be involved in the pathogenesis of retinal vascular diseases. In the present study, the mechanisms underlying bradykinin vasodilation were examined and whether calcium-activated potassium channels of small (SK_Ca) and intermediate (IK_Ca) conductance are involved in regulation of endothelium-dependent vasodilation in retinal arterioles was investigated.

methods. Porcine retinal arterioles (diameter ∼ 112 μm, N = 119) were mounted in microvascular myographs for isometric tension recordings. The arterioles were contracted with the thromboxane analogue, U46619, and concentration–response curves were constructed for bradykinin and a novel opener of SK_Ca and IK_Ca channels, NS309.

results. In U46619-contracted arterioles, bradykinin and NS309 induced concentration-dependent relaxations. In vessels without endothelium, bradykinin relaxation was abolished and NS309 relaxation was attenuated. Inhibition of NO synthase with asymmetric dimethylarginine and/or cyclooxygenase with indomethacin markedly reduced bradykinin and NS309 relaxation. NO synthase and cyclooxygenase inhibition together with oxyhemoglobin abolished bradykinin relaxation and attenuated NS309 relaxation. Blocking of SK_Ca and IK_Ca channels with apamin plus charybdotoxin or blocking of SK_Ca channels alone in the absence and the presence of indomethacin markedly reduced bradykinin and NS309 relaxation, whereas blocking of IK_Ca channels had no significant effect. In vessels without endothelium, blocking of SK_Ca channels alone had no effect on sodium nitroprusside-induced relaxation.

conclusions. In porcine retinal arterioles, NO and prostaglandins mediate endothelium-dependent relaxation to bradykinin and NS309. Moreover, these findings suggest that SK_Ca channels contribute to NO-mediated relaxation induced by bradykinin and NS309 and, hence, may play an important role in retinal arterial endothelial function.

Disturbances in the blood supply to the retina and the optic nerve are involved in the pathophysiology of major sight-threatening diseases, such as diabetic retinopathy ¹ and glaucoma. ² However, therapeutic intervention in these diseases is hampered by lack of knowledge of how the retinal blood flow is regulated under normal conditions, which is mainly determined by the diameter of retinal resistance arterioles. ³ This diameter is adjusted by dilating and constricting factors conveyed by the blood stream or released from cells in and around the vascular wall. ³ The vascular endothelial cell plays a central role in vasodilation, and the function of these cells has been shown to be disturbed in patients with metabolic and vascular diseases in general. ⁴ Endothelium-dependent vasodilation can be evaluated indirectly by studying how bradykinin stimulates the endothelium to release vasodilating factors. ⁵ Previous studies have shown that bradykinin-induced vasodilation is mediated by nitric oxide (NO) and prostaglandins in retinal arterioles. ⁶ However, recent findings in mesenteric arterioles suggest that calcium-activated potassium channels with small (SK_Ca) and intermediate (IK_Ca) conductance may both modulate the agonist-induced vasodilation mediated by NO and prostaglandins ⁷ and may also lead to endothelial release of other non-NO, non-prostaglandin substances, which mediate endothelium-derived hyperpolarizing factor (EDHF) type relaxation. ⁴ It is unknown whether SK_Ca and IK_Ca channels contribute to endothelium-dependent vasodilation in retinal arterioles. 

In the present study, we examined the mechanisms underlying bradykinin vasodilation and investigated whether SK_Ca and IK_Ca channels are involved in regulation of endothelium-dependent vasodilation in porcine retinal arterioles. To study their involvement, we investigated bradykinin-induced relaxation of porcine retinal arterioles was investigated after inhibition of NO synthesis with asymmetric dimethylarginine (ADMA) and inhibition of prostanoid synthesis with indomethacin. Bradykinin was replaced by NS309, a novel opener of SK_Ca and IK_Ca channels. NS309 induces endothelium-dependent hyperpolarizations sensitive to inhibitors of SK_Ca and IK_Ca channels in isolated rat mesenteric arteries and guinea-pig carotid arteries. ^{8 9} Therefore, additional experiments were conducted with bradykinin and NS309, where ADMA and indomethacin were replaced by apamin and charybdotoxin that block SK_Ca and IK_Ca channels. 

Materials and Methods

Solutions

Physiological saline solution (PSS) with the following composition (in mM) was used: 4.8 KCl, 1.14 MgSO₄, 118 NaCl, 25 NaHCO₃, 5 HEPES, 5.5 glucose, and 1.6 CaCl₂. PSS without CaCl₂was termed PSS^0.0. In KPSS, equimolar NaCl was replaced with KCl in the PSS giving a concentration of K⁺ of 60 or 125 mM. The solution was equilibrated with bioair of the following composition: 5% CO₂, 21% O₂, and 74% N₂. 

Bradykinin (vasodilator), 9,11-dideoxy-11a,9a-epoxymetho-prostaglandin F_2a (U46619; thromboxane analogue), asymmetric dimethylarginine (ADMA; NO synthase inhibitor), indomethacin (cyclooxygenase inhibitor), hemoglobin (NO scavenger) and sodium nitroprusside (SNP; NO donor) were purchased from Sigma-Aldrich (St. Louis, MO). Apamin (SK_Ca channel blocker) and charybdotoxin (IK_Ca channel blocker) were purchased from Latoxan (Valence, France). The NS309 (6,7-dichloro-1H-indole-2,3-dione 3-oxime; SK_Ca and IK_Ca channel opener) was kindly donated by Søren-Peter Olesen (Neurosearch A/S, Ballerup, Denmark). 

U46619 was dissolved in ethanol, NS309 in DMSO (99%), apamin and bradykinin in distilled water, and charybdotoxin and indomethacin in PSS^0.0. Apamin, bradykinin, and charybdotoxin were prepared in 2% albumin-coated tubes (Eppendorf, Fremont, CA). 

Tissue

Porcine eyes were obtained from a local abattoir. The pigs were approximately 6 months of age and weighed 85 to 90 kg. One eye from each animal was removed immediately after the pig had been anesthetized with CO₂ and killed by exsanguination. During transport to the laboratory, the eyes were kept in cold PSS. The eyes were dissected and the retinas isolated as described previously. ⁶ In brief, the eyes were bisected by frontal sectioning, the vitreous was removed, and the retina was detached from the underlying pigment epithelium. An arteriolar segment located approximately 1 to 2 mm from the optic disc with a length of ∼2 mm with surrounding retinal tissue attached on each side of the segment was dissected from the retina. The surrounding retinal tissue was removed after the arteriolar segment was mounted in the wire myograph. 

Mounting and Normalization

The arteriolar segment was transferred to a chamber of a dual wire myograph system (model 410A; Danish Myo Technology, Aarhus, Denmark) for isometric tension recordings (Chart5 software program; ADInstruments, Oxfordshire, UK). Each arteriole was mounted in the wire myograph with two 25-μm tungsten wires and was allowed to equilibrate for 30 minutes at 37°C. Subsequently, it was normalized to an internal circumference corresponding to 94% of the tone at a transmural pressure of 70 mm Hg in PSS^0.0, ¹⁰ where the segments had an internal diameter of 112 ± 2 μm, N = 119. After normalization PSS^0.0 was replaced with PSS, and the segment was allowed to equilibrate in bioair for 30 minutes. The vessels developed a myogenic tone after replacing PSS^0.0 with PSS. 

Control of Viability

After equilibration, the viability of the smooth muscle cells was tested by adding 60 mM KPSS and U46619 (10⁻⁷ M) to test the contraction induced by depolarization and by the thromboxane A₂ receptor, followed by U46619 (10⁻⁷ M) to test the contraction induced by the thromboxane A₂ receptor alone. The endothelium-dependent relaxation was tested by addition of bradykinin (3 × 10⁻⁸ M) after contraction with U46619 (10⁻⁷ M). The arterioles were discarded if U46619-induced contraction was less than 0.25 N/m or if bradykinin relaxation was less than 50%. 

Experimental Protocol

The experimental procedure consisted of three steps: addition of inhibitor or blocker, precontraction, and a concentration–response experiment with bradykinin, NS309, or SNP. 

Addition of Inhibitor or Blocker.

The vessel was incubated with one of the following inhibitors or blockers:

1.  Inhibiting the effect of prostanoids and NO  
   •  
     a. Cyclooxygenase inhibition
      
      i. Indomethacin (3 × 10⁻⁶ M) ⁷
   •  
     b. NO synthase inhibition
      
      i. Asymmetric dimethylarginine (ADMA) (3 ×10⁻⁴ M) ⁷
   •  
     c. Inhibitors a and b together
   •  
     d. Inhibitors a and b together with the NO-scavenger, oxygenated hemoglobin (2.5 × 10⁻⁵ M). ¹¹
2.  Blocking calcium-activated potassium channels  
   •  
     a. IK_Ca channel blockade
      
      i. Charybdotoxin (10⁻⁷ M) ¹²
   •  
     b. SK_Ca channel blockade
      
      i. Apamin (5 × 10⁻⁷ M) ¹³
   •  
     c. Blockers a and b together.
3.  Inhibiting the effect of prostanoids and blocking of calcium-activated potassium channels  
   •  
     a. Indomethacin together with blockers 2a and/or 2b.
4.  
   No inhibitor or blocker (control).

Precontraction.

To induce a stable contraction of the arteriolar segments, U46619 (10⁻⁷ M) was added to the myograph chamber, after incubation with inhibitors or blockers for 30 minutes U46619 (10⁻⁷ M) induced a contraction of 1.01 ± 0.06 mN/mm (N = 119). None of the inhibitors or blockers used influenced U46619 contraction. 

Concentration–Response Experiment.

Bradykinin was added successively in the following concentrations (in M): (10⁻¹⁰, 3 × 10⁻¹⁰, 10⁻⁹, 3 × 10⁻⁹, 10⁻⁸, 3 × 10⁻⁸, 10⁻⁷, and 3 × 10⁻⁷), NS309, in the following concentrations (in M): (10⁻⁷, 2 × 10⁻⁷, 3 × 10⁻⁷, 6 × 10⁻⁷, 10⁻⁶, 2 × 10⁻⁶, 3 × 10⁻⁶, 6 × 10⁻⁶, and 10⁻⁵), and finally SNP in the following concentrations (in M): (10⁻⁸, 3 × 10⁻⁸, 10⁻⁷, 3 × 10⁻⁷, 10⁻⁶, 3 × 10⁻⁶, 10⁻⁵, 3 × 10⁻⁵, and 10⁻⁴). 

Control Experiments

Stability of the Effect of U46619.

To study the stability of the tone produced by U46619, time-control experiments and concentration–response curves for bradykinin and NS309 were performed simultaneously. The contraction induced by U46619 was stable during the period in which the concentration–response experiments for bradykinin and NS309 were performed. 

Removal of the Endothelium.

To exclude the effect of the endothelium on bradykinin- and NS309-induced relaxation, a 25-μm tungsten wire was introduced into the lumen of the mounted vessel to mechanically remove the endothelium. Removal of the endothelium was considered successful, if U46619 (10⁻⁷ M) induced a contraction of more than 0.25 N/m and bradykinin (3 × 10⁻⁸ M) subsequently induced a relaxation of less than 10%. This procedure completely abolished bradykinin relaxation and significantly attenuated NS309 relaxation. 

Contraction with High Concentration of Potassium (KPSS).

To investigate the relaxing effect of NS309 when the effects on potassium channels were excluded the arteriolar segments were contracted with 125 mM KPSS. In KPSS-contracted segments, only the highest concentration of NS309 (10⁻⁵ M) induced relaxation (27.9% ± 4.0%, n = 7). 

Analysis of Data

The myogenic tone was defined as the level of contraction after addition of PSS subtracted from the level of contraction at PSS^0.0 or from the level of contraction after addition of PSS and inhibitor and/or blocker subtracted from the level of contraction at PSS^0.0. The active tone was defined as the level of contraction after addition of U46619 subtracted by the level of contraction after the 30 minutes of equilibration in PSS after the normalization procedure. For further analysis, the tension in the peak response after each addition of bradykinin, and the stable tension after each addition of NS309 and SNP were expressed as a percentage of the active tension. 

To calculate the EC₅₀, concentration–response curves were fitted to the Hill equation, logT/(1 − T) versus log[BK] or log[NS309], where T is the relative tension. EC₅₀ could not be calculated from experiments where bradykinin was added, since relaxation was too weak for the concentration–response curves to reach an asymptotic minimum except for control and charybdotoxin experiments. 

The maximum relaxation was calculated for each concentration–response experiment for bradykinin and NS309. In the experiments using NS309, maximum relaxation was reached in all cases. 

The concentration–response curves for bradykinin and NS309 were evaluated by calculation of the area under curve followed by one-way ANOVA. One-way ANOVA was also used to test for differences in myogenic tone, in maximum relaxation for bradykinin, and in EC₅₀ for NS309. In case of significance, the deviant values were tested by Wilcoxon signed-rank test. 

Results

Development of myogenic tone, EC₅₀, and maximum relaxation for bradykinin and NS309 concentration–response curves are shown in Tables 1 and 2 . Original traces for bradykinin and NS309 with and without endothelium are shown in Figure 1 . When the endothelium was removed, bradykinin relaxation was completely abolished (n = 10), whereas this treatment attenuated NS309 relaxation (n = 10). 

The effects on bradykinin relaxation after inhibition of prostaglandin (n = 10) and NO (n = 10) synthesis are shown in Figure 2 . In the control (n = 10) and the indomethacin treatment groups, ADMA and oxygenated hemoglobin (n = 7) were significantly different from those in the other treatments. 

The effects on NS309 relaxation after inhibition of prostaglandin (n = 10) and NO synthesis (n = 10) are illustrated in Figure 3 . The control (n = 10) was significantly different from the other treatments. 

The effects on bradykinin and NS309 relaxation after blocking the SK_Ca and IK_Ca channels with and without inhibition of prostaglandin synthesis are shown in Figures 4 and 5 , respectively. Blocking of SK_Ca channels alone with apamin (n = 8) or blocking of both SK_Ca and IK_Ca channels with, respectively, apamin and charybdotoxin (n = 8) significantly reduced the relaxation response to bradykinin and NS309. Blocking of IK_Ca channels alone with charybdotoxin (n = 7) had no significant effect on the relaxation response to bradykinin or NS309. After inhibition of prostaglandin synthesis, blocking of SK_Ca channels with apamin (n = 8) or blocking of SK_Ca and IK_Ca channels with, respectively, apamin and charybdotoxin (n = 7 to 8), significantly reduced the relaxing effects of bradykinin and NS309, whereas blocking of IK_Ca channels alone with charybdotoxin (n = 8–9) had no significant effect on the relaxation induced by bradykinin or NS309. 

The effects on SNP after blocking SK_Ca channels with apamin (n = 9) in arterioles without endothelium are shown in Figure 6 . Blocking SK_Ca channels had no significant effect on SNP relaxation. 

Discussion

The present study suggests that in porcine retinal arterioles, NO and prostaglandins are involved in bradykinin- and NS309-induced relaxation. No involvement of EDHF-type relaxation was found. Moreover, SK_Ca channels are involved the NO-mediated relaxation. These findings suggest that SK_Ca channels may have an essential role in retinal arterial endothelial function. 

Involvement of NO and Prostaglandins in Endothelium-Dependent Relaxation

Bradykinin induced concentration-dependent relaxation, which was completely abolished by removal of the endothelium. Moreover, inhibition of NO synthase and cyclooxygenase together with an NO scavenger completely abolished bradykinin-induced relaxation. These findings suggest that endothelium-dependent bradykinin-induced relaxation was mediated by release of the vasodilating agents NO and prostanoids. NO has been reported to play an important role in regulating ocular blood flow. ^{14 15 16 17 18} In porcine and human ophthalmic arteries, NO synthase inhibition increases basal myogenic tone and decreases acetylcholine- and bradykinin-induced vasodilation, indicating that NO regulates both basal tone and agonist-induced vasodilation. ^{14 15} Furthermore, altered NO bioavailability has been associated with ischemia–reperfusion injury, ^{19 20 21} and the diseases primary open-angle glaucoma ^{22 23} and diabetic retinopathy. ^{24 25 26 27} In ischemia–reperfusion injury, administration of either NO donors or l-arginine at the time of retinal ischemia improves retinal function after ischemia, presumably due to enhanced retinal blood flow, ^{19 20} whereas postischemic treatment with NO synthase inhibitors seems to improve retinal function. ²¹ In patients with normal-pressure glaucoma, acetylcholine-induced, NO-mediated forearm vasodilation is reduced compared with that in healthy volunteers, suggesting a general endothelial dysfunction in these patients, ²² which is suggested to be due to impaired NO formation. ²³ Moreover, in eyes from patients with primary open-angle glaucoma, during NO synthase inhibition the decrease of optic nerve head blood flow is less pronounced than that in healthy control eyes. ²⁸ However, in subcutaneous resistance arteries of patients with normal-pressure glaucoma, the endothelium-dependent relaxation is not altered compared with that in subcutaneous resistance arteries of healthy subjects. ²⁹ In patients with type I diabetes mellitus, NO synthase inhibition decreases blood flow velocity in the ophthalmic artery and increases blood pressure less than in healthy subjects. These results suggest that either basal NO release is decreased or NO sensitivity is impaired in patients with type I diabetes mellitus, leading the authors to propose that the l-arginine-NO system is involved in the pathophysiology of diabetic retinopathy. ²⁴ Moreover, NO formation is decreased in insulin resistant states, ²⁶ which may relate to a decreased activity due to inactivation of NO by free radicals. ²⁷ However, no difference in the ocular and systemic effects of NO synthase inhibition was observed between patients with type I diabetes and control subjects, indicating that responsiveness to NO of the choroidal vasculature and the ophthalmic artery is not altered in early type I diabetes. ²⁵  

Previous studies have suggested different involvement of prostaglandins in regulating ocular blood flow. ^{30 31 32 33 34 35 36 37 38 39} In segments of human posterior ciliary arteries with endothelium, spontaneous myogenic tone increases in response to indomethacin. ³⁰ In agreement with these findings, indomethacin also increased myogenic tone in the present study, hence providing additional support for a basal release of prostaglandins in retinal arterioles. 

In isolated bovine retinal small arteries, acetylcholine relaxation is not inhibited by indomethacin, suggesting release of NO and no involvement of prostaglandins. ³¹ However, prostaglandins were found to be involved in vasodilator responses in rat, ^{32 38 39} canine, ^{33 34} bovine, ³⁶ and porcine ^{35 37} eyes. In previous studies of pressurized porcine retinal arteries, we observed that inhibition of cyclooxygenase induced a significant right shift of the bradykinin concentration response, ⁶ and in the present study the effect of the cyclooxygenase inhibitor, indomethacin, on bradykinin relaxation was even more pronounced. We have observed that indomethacin at high concentrations (>10 μM) may block calcium channels (Simonsen et al., unpublished observation, 1995), and it cannot be excluded that this may underestimate the contribution of prostaglandins to bradykinin relaxation. Perivascular tissue was observed to inhibit vasoconstriction in porcine retinal arterioles by formation of vasodilatory prostaglandins, ⁴⁰ and although we carefully removed adhering tissue in the present study, we cannot exclude that formation of prostaglandins in residual perivascular tissue influences the vascular response. However, it is more likely that basal tone rather than bradykinin-induced vasodilation is influenced by perivascular tissue, since bradykinin-induced responses were abolished in arterioles without endothelium. Moreover, the difference in contribution from prostaglandins to vasodilation observed in previous studies may be due to species variation. However, as the porcine retinal vessels are largely similar to human retinal vessels, ^{41 42 43} results obtained from porcine tissue may be extrapolated to human conditions. This study showed no significant difference between NO synthase and/or cyclooxygenase inhibition, suggesting that prostanoids may be involved in the regulation of the NO pathway in the retinal circulation. Supporting this possibility, prostaglandins, such as prostaglandin E₂, have been demonstrated to increase eNOS activity, ^{44 45} thereby increasing the importance of a functional NO pathway. However, further studies are needed to address whether the interaction of prostaglandins and NO takes place in the endothelial or smooth muscle cell layer of retinal arterioles. 

Involvement of Calcium-Activated Potassium Channels in Bradykinin- and NS309-Induced Relaxation

Both SK_Ca and IK_Ca channels are expressed in endothelial cells, ^{46 47 48} and in agreement with these observations, removal of the endothelial cell layer markedly reduced NS309 relaxation. The remaining NS309 relaxation observed in segments without endothelium may indicate that SK_Ca and IK_Ca channels are not expressed only in endothelial cells or that NS309 at high concentrations activates other channels in porcine retinal arterioles. At high concentrations, NS309 (>10 μM) blocks L-type voltage-dependent calcium channels. ⁴⁹ However, in 125 mM K⁺-contracted segments, excluding the effects of potassium channels, NS309 induced only slight relaxation, indicating that the compound is specific for SK_Ca and IK_Ca channels at the concentrations used in the present study. 

Recent studies have proposed that the relative contribution from SK_Ca and IK_Ca channels to EDHF-type vasodilation may be related to their connection to different vasodilating mechanisms, possibly by special clustering. ^{50 51} Thus, in rat mesenteric resistance arteries, the IK_Ca channels are located within endothelial cell projections that form myoendothelial gap junctions and the Na⁺/K⁺-ATPase is localized close to microdomains that lead to smooth muscle hyperpolarization. ⁵¹ In contrast, in rat superior mesenteric arteries, where SK_Ca and IK_Ca channels are found to be involved in the release of NO, it is not possible to link a certain subtype of calcium-activated potassium channels to release of NO, ⁷ and in HUVECs blockade of both SK_Ca and IK_Ca channels is necessary for inhibition of agonist-stimulated NO production. ⁵² However, in lamb coronary resistance arteries, blocking of only SK_Ca channels with apamin inhibits NO-mediated vasodilation, ¹³ suggesting a difference in large versus small arteries in the coupling of a certain subtype of calcium-activated potassium channels to a vasodilator pathway. In the present study, blocking of SK_Ca channels alone reduced the relaxation caused by bradykinin and NS309. Inhibition of NO synthase and/or cyclooxygenase and inhibition of NO synthase and cyclooxygenase together with an NO-scavenger, attenuated the NS309 relaxation to the same degree. Moreover, during inhibition of cyclooxygenase, blocking of SK_Ca channels reduced bradykinin- and NS309-induced relaxation. Finally, in arterioles without endothelium, blocking SK_Ca channels had no effect on SNP relaxation. These results suggest that SK_Ca channels are located in the endothelial cells upstream of NO synthase and coupled to NO-mediated relaxation in porcine retinal arterioles. 

Blocking of IK_Ca channels failed to reduce bradykinin and NS309 relaxation. However, blocking of both SK_Ca and IK_Ca channels reduced the bradykinin and NS309 relaxation even further compared with blocking of SK_Ca channels alone, but this was not the case in the presence of indomethacin. 

The present results imply that in retinal arterioles, SK_Ca and IK_Ca channels are essential for the release of, respectively, NO and prostaglandins, and hence endothelial function. As a dysfunctional endothelium is found in patients with glaucoma ^{22 23} and diabetic retinopathy, ²⁴ the perspective is that specific activation of SK_Ca and IK_Ca channels may improve endothelium-dependent vasodilation and retinal blood flow in these patients. 

In summary, in porcine retinal arterioles, NO and prostaglandins mediate bradykinin- and NS309-induced relaxation. Moreover, our findings suggest that apamin-sensitive SK_Ca channels contribute to NO-mediated relaxation induced by bradykinin and NS309 and, may play an important role in retinal arterial endothelial function. 

 Supported by the Lundbeck Foundation and the Danish Cardiovascular Research Academy.

 Submitted for publication November 17, 2008; revised January 20, 2009; accepted May 4, 2009.

 Disclosure: T. Dalsgaard, None; C. Kroigaard, None; T. Bek, None; U. Simonsen, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Myogenic Tone, EC₅₀, and Maximum Relaxation for Bradykinin and NS309, an Opener of SK_Ca and IK_Ca Channels

Treatment	n	Myogenic Tone (N/m)	Bradykinin			NS309
						EC50 (×10−9 M)	Max Relax (%)	EC50 (×10−6 M)	Max Relax (%)
Control	10	0.32 ± 0.05	5.5 ± 1.3	92.4 ± 2.4	1.1 ± 0.3	93.2 ± 3.2
Indomethacin	10	0.43 ± 0.14*	—	70.1 ± 8.5*	2.6 ± 0.5*	90.1 ± 3.6
ADMA	10	0.58 ± 0.15*	—	61.5 ± 11.1*	2.1 ± 0.3*	92.4 ± 3.2
Indo/ADMA	10	0.79 ± 0.17*	—	31.3 ± 10.7*	2.7 ± 0.5*	92.1 ± 2.9
ADMA/OxyHb	8	0.67 ± 0.19*	—	19.8 ± 8.0*	3.2 ± 0.4*	92.7 ± 2.6
Indo/ADMA/OxyHb	7–9	0.71 ± 0.12*	—	5.7 ± 3.2*	3.1 ± 0.4*	94.4 ± 2.4

 Values are mean ± SEM. Differences were evaluated by one-way ANOVA and the Wilcoxon signed-rank test. Indo, indomethacin; OxyHb, oxygenated hemoglobin.

*  Significantly different from control (P < 0.05).

Table 2.

 

Myogenic Tone, EC₅₀, and Maximum Relaxation for Bradykinin and NS309, an Opener of SK_Ca and IK_Ca Channels

Treatment	n	Myogenic Tone (N/m)	Bradykinin			NS309
						EC50 (×10−9 M)	Max Relax (%)	EC50 (×10−6 M)	Max Relax (%)
−Indomethacin
Control	10	0.32 ± 0.05	5.5 ± 1.3	92.4 ± 2.4	1.1 ± 0.3	93.2 ± 3.2
ChTx	7	0.40 ± 0.17	4.9 ± 0.9	86.9 ± 3.0	2.3 ± 0.9	95.0 ± 4.5
Apamin	8	0.48 ± 0.11	—	66.9 ± 11.1*	4.2 ± 1.0*	93.8 ± 4.1
ChTx/apamin	8	0.52 ± 0.13	—	38.1 ± 8.9*	8.3 ± 1.5*	80.0 ± 9.7
+Indomethacin
Control	10	0.43 ± 0.14	—	70.1 ± 8.5	2.6 ± 0.5	90.1 ± 3.6
ChTx	8–9	0.58 ± 0.09	—	70.8 ± 5.5	2.8 ± 0.3	95.4 ± 1.7
Apamin	8	0.59 ± 0.10	—	40.6 ± 11.0*	8.6 ± 1.9*	70.5 ± 13.0*
ChTx/apamin	7–8	0.60 ± 0.09	—	33.7 ± 5.7*	11.0 ± 0.8*	43.6 ± 5.6*

 Data are the mean ± SEM. Differences were evaluated by one-way ANOVA and the Wilcoxon signed-rank test. ChTx, charybdotoxin.

*  Significantly different from their respective controls (P < 0.05).

Figure 1.

 

Original traces showing the response to increasing concentrations of bradykinin (A, B) and NS309, an opener of SK_Ca and IK_Ca channels (C, D) in porcine retinal arterioles, with endothelium (A, C) and without endothelium (B, D). A high myogenic tone was developed in the vessels in (B), (C), and (D), before addition of U46619.

Figure 2.

 

Bradykinin (BK)-induced vasorelaxation in the presence of indomethacin (indo), ADMA, indo+ADMA, ADMA+oxygenated hemoglobin (OxyHb), and indo+ADMA+OxyHb. Mean ± SEM. One-way ANOVA; *P < 0.05 versus the control.

Figure 3.

 

Vasorelaxation induced by NS309, an opener of SK_Ca and IK_Ca channels, in the presence of indomethacin (indo), ADMA, indo+ADMA, and indo+ADMA+oxygenated hemoglobin (OxyHb). Mean ± SEM. One-way ANOVA; *P < 0.05 versus the control.

Figure 4.

 

Bradykinin (BK)-induced vasorelaxation in the presence of apamin, charybdotoxin (ChTx), and apamin+ChTx (A), and in the presence of indomethacin (indo) (B). Mean ± SEM (n = 7–11). One-way ANOVA; *P < 0.05 versus the control.

Figure 5.

 

Vasorelaxation induced by NS309, an opener of SK_Ca and IK_Ca channels (controls), in the presence of apamin, charybdotoxin (ChTx), and apamin+ChTx (A) and indomethacin (indo) (B). Mean ± SEM (n = 7–11). One-way ANOVA; *P < 0.05 versus the control.

Figure 6.

 

Sodium nitroprusside (SNP)-induced vasorelaxation in vessels without endothelium in the absence or presence of apamin. Mean ± SEM.

The authors thank Henriette Johanson and Helle Zibrandtsen for excellent technical assistance.