Cross-talk of hydrogen sulfide and nitric oxide in vascular endothelial cells

Abstract

Gasotransmitters, like hydrogen sulfide (H2S) and nitric oxide (NO), are small gaseous molecules that can be generated in different types of mammalian cells by enzymatic catalyzation. Cystathionine y-lyase (CSE) and endothelial NO synthase (eNOS) are responsible for the majority of endogenous production of H2S and NO in vascular endothelium, respectively. H2S and NO maintain different vascular functions. Here we show that H2S interacts with eNOS to increase NO release from endothelial cells (ECs). Two mechanisms are involved in this interaction. Firstly, H2S indirectly induces eNOS phosphorylation. Secondly, H2S directly modifies one cysteine residue of eNOS through S-sulfhydration. Stimulation of eNOS phosphorylation and S-sulfhydration by H2S subsequently increases NO release. The phosphorylation of eNOS by H2S is p38 and Akt-dependent. eNOS S-sulfhydration is partially affected by S-nitrosylation but not by phosphorylation. We further found that knockdown of CSE gene by siRNA technique, or blockage of CSE enzyme activity by PPG (dl-propargylglycine), attenuates NO production. CSE overexpression or L-cysteine (a substrate of H2S) supplementation stimulates NO production. The level of eNOS S-sulfhydration in aortic tissue from CSE knockout (CSE-KO) mice was lower than that from wild type (WT) mice. L-cysteine treatment increases S-sulfhydration of eNOS in ECs isolated from WT mice, but not in ECs isolated from CSE-KO mice. GSNO (a NO donor) induces, but NaHS reduces, eNOS S-nitrosylation. However, GSNO does not alter eNOS S-sulfhydration whereas NaHS alters S-nitrosylation. Site-directed mutagenesis of one cysteine residue Cys-443 in eNOS (Cys-443-eNOS) completely eliminates eNOS S-sulfhydration and partially decreases eNOS S-nitrosylation. Although the mutation of serine 1179 (Ser-1179) completely abolishes eNOS phosphorylation, it does not affect eNOS S-sulfhydration. The dominant configuration of vascular eNOS proteins purified from WT mice is dimer, whereas in CSE-KO mice it is monomer. In the presence of GSNO, more monomers are found with WTeNOS, which is reversed by a subsequent treatment with NaHS. Cys-443-eNOS manifests itself as monomers, which is not changed by either GSNO or NaHS treatments. The production of NO is decreased but superoxide is increased in CSE-KO ECs in comparison with WT ECs. H2S treatment increases EC proliferation, tube formation, angiogenesis, and accelerates wound healing. With an in vitro aortic ring angiogenesis assay, we found a reduction in the number of microvessels formed by culturing aortic rings from CSE-KO mice, even in the presence of VEGF (vascular endothelial growth factor). We further found that wound healing is faster in WT mice when compared with CSE-KO mice, and H2S promotes wound healing recovery. Blockade of NO production by eNOS-specific siRNA or L-NAME (L-NGnitroarginine methyl ester) reverses, but eNOS overexpression potentiates the proliferative effect of H2S. In contrast, CSE knockdown attenuates the pro-proliferative effect of NO. Overall, our studies demonstrate that H2S increases NO release in ECs through phosphorylation and S-sulfliydration of eNOS. Thus, H2S and NO are required for the physiological control of angiogenesis and superoxide production. Mechanistic understanding of H2S-NO interaction in vascular endothelium will help advance novel therapeutic strategies for EC dysfunction related vascular diseases.