The vasorelaxant effect of H(2)S as a novel endogenous gaseous K(ATP) channel opener

Abstract

Hydrogen sulfide (H(2)S) has been traditionally viewed as a toxic gas. It is also, however, endogenously generated from cysteine metabolism. We attempted to assess the physiological role of H(2)S in the regulation of vascular contractility, the modulation of H(2)S production in vascular tissues, and the underlying mechanisms. Intravenous bolus injection of H(2)S transiently decreased blood pressure of rats by 12- 30 mmHg, which was antagonized by prior blockade of K(ATP) channels. H(2)S relaxed rat aortic tissues in vitro in a K(ATP) channel-dependent manner. In isolated vascular smooth muscle cells (SMCs), H(2)S directly increased K(ATP) channel currents and hyperpolarized membrane. The expression of H(2)S-generating enzyme was identified in vascular SMCs, but not in endothelium. The endogenous production of H(2)S from different vascular tissues was also directly measured with the abundant level in the order of tail artery, aorta and mesenteric artery. Most importantly, H(2)S production from vascular tissues was enhanced by nitric oxide. Our results demonstrate that H(2)S is an important endogenous vasoactive factor and the first identified gaseous opener of K(ATP) channels in vascular SMCs.

Fig. 1. The effect of H₂S in vivo on mean arterial blood pressure (BP) and heart rate of rats. (A) Intravenous injection of H₂S induced significant decrease in BP. This effect was mimicked by intravenous injection of pinacidil (2.8 µmol/kg) and antagonized by a prior intravenous injection of glibenclamide (2.8 µmol/kg). (B) Effect of H₂S on rat heart rate. Heart rate was recorded 30 s after intravenous injection of PBS (control), H₂S or pinacidil. *P < 0.05 compared with control.

Fig. 2. The H₂S-induced relaxation of rat aortic rings and the underlying mechanisms. (A) Relaxation of the PHE-precontracted tissues by H₂S in the form of either standard NaHS solution (square) or H₂S gas-saturated solution (circle). (B) Inhibitory effect of l-NAME (100 µM, 20 min, circle) on the H₂S-induced relaxation (control, square). (C) The effects of H₂S (180 µM) on the endothelium-free or endothelium-intact aortic tissues pretreated with l-NAME or charybdotoxin (ChTX)/apamin. (D) The relaxant effect of H₂S was not affected by pretreating the tissues with SQ22536, SOD or catalase, respectively. (E) The effect of ODQ treatment (10 µM for 10 min) on the relaxant effects of SNP (0.1 µM) or H₂S (600 µM). n = 8 for each data point. *P < 0.05 compared with control.

Fig. 3. The K⁺ channel-mediated vascular effects of H₂S. (A) The relaxant effect of H₂S on the aortic tissues precontracted with 20 or 100 mM KCl. (B) Inhibitory effect of TEA on the H₂S-induced vasorelaxation. The concentration-dependent vasorelaxant effects of H₂S with or without TEA (10 mM) pretreatment of aortic rings were determined. (C) The H₂S (600 µM)-induced vasorelaxation was not affected by pretreatment of aortic tissues with either 10 µM iberiotoxin (IbTX) or 2.5 mM 4-aminopyridine (4-AP). *P < 0.05 compared with control, n = 8. (D) The vasoactive effects of pinacidil (left) and H₂S (right) on the precontracted aortic tissues. The vasorelaxant effect of H₂S (600 µM) was examined in the presence of glibenclamide at different concentrations. *P < 0.05 compared with control, **P < 0.05 compared with the H₂S group in the absence of glibenclamide, n = 8.

Fig. 4. The effect of H₂S and pinacidil on K_ATP channel currents in rat aortic SMCs. (A) The effect of H₂S on K_ATP channel currents. The original records from one cell (left) and the mean current–voltage relationship of six cells (right) in the absence and then presence of H₂S (300 µM) are shown. (B) The effect of pinacidil (5 µM) on K_ATP channel currents. The original records from one cell (left) and the mean current–voltage relationship of five cells (right) in the absence and then presence of pinacidil are shown. Dashed line indicates zero current level. (C) The antagonistic effect of glibenclamide (Gli, 5 µM) on the effect of H₂S or pinacidil (Pin) on K_ATP channel currents (holding potential, –60 mV; test potential, +40 mV). (D) The membrane hyperpolarization induced by H₂S (300 µM) or pinacidil (5 µM), respectively. *P < 0.05 compared with control; n = 3–7 for each group.

Fig. 5. The modulation of K_ATP channel currents by H₂S with different intracellular ATP concentrations and by glibenclamide (Gli) in rat aortic SMCs. (A) The ATP-dependence of the H₂S (300 µM)-induced increases in K_ATP channel currents. Holding potential, –60 mV; test potential, +40 mV. *P < 0.05 compared with control. (B) The time course of the inhibition of K_ATP currents by glibenclamide. Aortic SMCs were dialysed with 0.2 mM ATP and 1 mM GDP. The voltage pulses of +30 mV were applied from a holding potential of –60 mV. Inset shows the representative of original current traces before and after the application of 10 µM glibenclamide.

Fig. 6. Differential expression of CSE in rat vascular tissues. (A) RT–PCR analysis of the expression of CSE (234 bp) in rat liver, mesenteric artery, tail artery, pulmonary artery and aorta. (B) Quantitative comparison of CSE mRNA levels in rat tail artery, mesenteric artery, pulmonary artery and aorta with RPA. This is representative of three experiments. *P < 0.05 compared with mesenteric artery; **P < 0.01 compared with tail artery, mesenteric artery and aorta. A, artery. (C) The transcriptional expressions of CSE and β-actin in cultured SMCs and EC (endothelial cell) detected by RT–PCR. (D) In situ hybridization showing the location of CSE mRNA in rat aorta wall by antisense probe on the left and sense probe as control on the right.

Fig. 7. Regulation of the endogenous H₂S production in different rat tissues. (A) Accumulated endogenous H₂S levels in rat tail artery (TA), mesenteric artery (MA), aorta and ileum. (B) H₂S production rate of aorta tissues was stimulated by SNP in a concentration-dependent manner. *P < 0.05 compared with control, n = 3. (C) The SNAP-induced concentration-dependent upregulation of CSE transcriptional expression in cultured aortic SMCs, determined using northern blotting. The 28S ribosome RNA was assayed as the housekeeping control. n = 3, *P < 0.05 compared with control.