Exenatide induces aortic vasodilation increasing hydrogen sulphide, carbon monoxide and nitric oxide production

Abstract

Background: It has been reported that GLP-1 agonist exenatide (exendin-4) decreases blood pressure. The dose-dependent vasodilator effect of exendin-4 has previously been demonstrated, although the precise mechanism is not thoroughly described. Here we have aimed to provide in vitro evidence for the hypothesis that exenatide may decrease central (aortic) blood pressure involving three gasotransmitters, namely nitric oxide (NO) carbon monoxide (CO), and hydrogen sulphide (H2S).

Methods: We determined the vasoactive effect of exenatide on isolated thoracic aortic rings of adult rats. Two millimetre-long vessel segments were placed in a wire myograph and preincubated with inhibitors of the enzymes producing the three gasotransmitters, with inhibitors of reactive oxygen species formation, prostaglandin synthesis, inhibitors of protein kinases, potassium channels or with an inhibitor of the Na+/Ca2+-exchanger.

Results: Exenatide caused dose-dependent relaxation of rat thoracic aorta, which was evoked via the GLP-1 receptor and was mediated mainly by H2S but also by NO and CO. Prostaglandins and superoxide free radical also play a part in the relaxation. Inhibition of soluble guanylyl cyclase significantly diminished vasorelaxation. We found that ATP-sensitive-, voltage-gated- and calcium-activated large-conductance potassium channels are also involved in the vasodilation, but that seemingly the inhibition of the KCNQ-type voltage-gated potassium channels resulted in the most remarkable decrease in the rate of vasorelaxation. Inhibition of the Na+/Ca2+-exchanger abolished most of the vasodilation.

Conclusions: Exenatide induces vasodilation in rat thoracic aorta with the contribution of all three gasotransmitters. We provide in vitro evidence for the potential ability of exenatide to lower central (aortic) blood pressure, which could have relevant clinical importance.

Figure 1

Effect of exenatide on the vasoactivity of rat thoracic aorta. Original records of myography experiments. Time-control of an epinephrine contracted aortic segment (A). Dose-dependent vasodilatory effect of exenatide on rat thoracic aorta following epinephrine contraction. 23.9, 71.7, 310, 788, 1980, 3170 nanomoles of exenatide were used to relax the vessels (B) (n = 5 of each experiment).

Figure 2

Role of GLP-1 receptor and endothelial denudation in the vasodilatation due to exenatide. Exenatide concentration-relaxation curves of vessels treated with exenatide only (▲) and vessels preicubated with GLP-1R anatagonist exendin(9–39) (■) (A). Vasodilation evoked by exenatide in endothelium-intact and endothelium-denuded vessels (B). 23.9, 71.7, 310, 788, 1980, 3170 nanomoles of exenatide were used to relax the vessels (n = 5 of each experiment), *P < 0.01 compared to exenatide only (at respective concentration of exenatide).

Figure 3

Role of gasotransmitters and prostaglandins in the vasodilatory effect of exenatide. Inhibition of eNOS with 300 μM N_ω-Nitro-L-arginine methyl ester hydrochloride (L-NAME) (A). Inhibition of CO production by blocking heme oxygenase enzyme with 10 μM Tin-protoporphyrin IX dichloride (B). Blocking H₂S production by inhibiting cystathionine-γ-lyase with 10 mM DL-Propargylglycine (PPG) (C). Inhibition of prostaglandin production with 3 μM indomethacin (D). 23.9, 71.7, 310, 788, 1980, 3170 nanomoles of exenatide were used to relax the vessels (n = 5 of each experiment), *P < 0.01 compared to the relaxation caused by exenatide only (at respective concentration of exenatide).

Figure 4

Effect of free radicals on the vasodilation due to exenatide. Concentration-relaxation curve of exenatide with/without the addition of 200 U/ml of the free radical scavanger superoxide dismutase (SOD) (A). Concentration-relaxation curve analyzing the possible role of hydrogen peroxide by blocking its formation with 1000 U/ml catalase (B). 23.9, 71.7, 310, 788, 1980, 3170 nanomoles of exenatide were used to relax the vessels (n = 5 of each experiment), *P < 0.01 compared to the relaxation caused by exenatide only (at respective concentration of exenatide).

Figure 5

Concentration-relaxation curves showing the possible effector molecules of the exenatide induced vasodilation. Blocking cAMP-dependent protein kinase A (PKA) with 5 μM H89 hydrochloride (A). Inhibiton of soluble guanylyl cyclase with 3 μM 1H- (1,2,4) (ODQ) (B). 23.9, 71.7, 310, 788, 1980, 3170 nanomoles of exenatide were used to relax the vessels (n = 5 of each experiment), *P < 0.01 compared to the relaxation caused by exenatide only (at respective concentration of exenatide)

Figure 6

Role of potassium channels and the Na⁺/Ca²⁺-exchanger in the vasodilator effect of exenatide. Blockade of large-conductance calcium-activated potassium channels with 2 mM tetraethylammonium (TEA) (A). Inhibition of ATP-sensitive potassium channels with 10 μM glibenclamide (B). KCNQ-type K_v channels blocked by 30 μM XE991 (C). Selective inhibition of the Na⁺/Ca²⁺-exchanger with 4 μM SEA0400 (D). 23.9, 71.7, 310, 788, 1980, 3170 nanomoles of exenatide were used to relax the vessels (n = 5 of each experiment), *P < 0.01 compared to the relaxation caused by exenatide only (at respective concentration of exenatide).

Figure 7

Hypothetical mechanism of the exenatide induced vasodilation according to our data and according to previous examinations [[3]-[7],[10]-[14],[19]-[31]]. HO: heme oxygenase, eNOS endothelial nitric oxide synthase, CSE: cystathionine-γ-lyase, COX: cyclooxygenase, PG: prostaglandin, H₂S: hydrogen sulphide NO: nitric oxide, CO: carbon monoxide, O₂^–•:superoxide anion, PKA: cAMP-dependent protein kinase, PKG: cGMP-dependent protein kinase, BK_Ca: large-conductance calcium activated potassium channel, KCNQ: a type of voltage-gated potassium channel, K_ATP: ATP-sensitive potassium channel.