Introduction
A diffusing factor, which is released from vascular endothelium by a stimulation with acetylcholine to relax vascular smooth muscle, was discovered and called endothelium-derived relaxation factor (EDRF) (Furchgott and Zawadzki, 1980). It was later identified as nitric oxide (NO) (Ignarro et al., 1987; Palmer et al., 1987), which activates soluble guanylate cyclase and subsequently protein kinase G through cyclic GMP (Arnold et al., 1977). In the brain an excitatory neurotransmitter glutamate had been recognized to induce cGMP (Mao et al., 1974), and the activation of N-methyl-D-aspartate (NMDA) receptors by glutamate was found to induce a release of a diffusible factor which had similar properties to EDRF in a Ca2+ -dependent manner (Garthwaite et al., 1988). NO producing activity was identified as an enzymatic reaction with arginine as a substrate in nicotinamide adenine dinucleotide phosphate (NADPH) and Ca2+/calmoduline-dependent manner (Bredt and Snyder, 1990), and the activity is localized to neurons as well as vascular endothelial cells (Bredt et al., 1991).
Hydrogen sulfide (H2S), which was identified in the brain (Warenycia et al., 1989; Goodwin et al., 1990; Savage and Gould, 1990), changes the levels of neurotransmitters when administered to animals (Warenycia et al., 1989), facilitates the induction of hippocampal long-term potentiation, a synaptic model of memory formation in the brain (Abe and Kimura, 1996), and relaxes vascular smooth muscle in synergy with NO (Hosoki et al., 1997; Zhao et al., 2001). H2S is produced by three enzymes cystathionine β-synthase (CBS), cystathionine γ-lyase (CSE) and 3-mercaptopyruvate sulfurtransferase (MPST) along with cysteine aminotransferase (CAT) or D-amino acid oxidase (DAO), from D/L-cysteine as a source of sulfur (Stipanuk and Beck, 1982; Abe and Kimura, 1996; Hosoki et al., 1997; Nagahara, 2007; Shibuya et al., 2009; Chiku et al., 2009; Mikami et al., 2011; Shibuya et al., 2013).
H2S induces Ca2+ influx in astrocytes by activating transient receptor potential ankyrin 1 (TRPA1) channels, and we later found hydrogen polysulfide (H2Sn, n > 2) more effectively activate the channels than H2S does (Nagai et al., 2004; Nagai et al., 2006; Oosumi et al., 2010; Kimura et al., 2013). H2S2 and H2S3 were identified in the brain, and MPST produces H2Sn, and other persulfurated molecules such as cysteine persulfide, glutathione persulfide, and persulfurated proteins (Kimura et al., 2015; Kimura et al., 2017; Koike et al., 2017; Nagahara et al., 2018a; Nagahara et al., 2018b; see also Kimura, 2020).
Two modes of action have been proposed for both H2S and NO signaling. One is that both molecules diffuse to the heme for redox reaction (Vitvitsky et al., 2015; Ruetz et al., 2017) as well as to cysteine residues of the target proteins to S-sulfurate or S-nitrosylate them (Mustafa et al., 2009; Lancaster, 2017). The other is that sulfur and NO are transferred from respective enzymes to the target proteins for S-sulfuration and S-nitrosylation. An example of the first mode of action of NO is EDRF, which diffuses from endothelium to vascular smooth muscle (Furchgott and Zawadsky, 1980). In the nervous system NO diffuses from postsynapse to presynapse as a retrograde transmitter to induce a release of neurotransmitter glutamate (Garthwaite et al., 1988; Garthwaite, 1991; Zhuo et al., 1993). On the other hand, the enzyme-mediated S-nitrosylation proceeds by clusters of enzymes which generate NO, synthesize S-nitrosylated proteins, and transnitrosylate it (Seth et al., 2018).
H2S diffuses to heme of target proteins to regulate their activity or to the target cysteine disulfide to reduce it (Aizenman et al., 1989; Abe and Kimura, 1996; Matsui et al., 2018). H2S and H2Sn diffuse to the targets to S-sulfurate (S-sulfhydrate) cysteine residues to modify their activity (Mustafa et al., 2009; Kimura et al., 2015). As to enzyme-mediated S-sulfuration, MPST transfers sulfur from 3MP to cysteine residues of target proteins (Nagahara et al., 2012; Kimura et al., 2017).
H2S exerts regulatory, beneficial and protective effects at physiological concentrations, while it is toxic at higher concentrations. In Down’s syndrome (DS) and ethylmalonyl encephalopathy (EE), the levels of H2S and/or H2Sn are increased and cause damage to the brain (Tiranti et al., 2009; Panagaki et al., 2019). For DS and EE a decrease in the levels of H2S has been proposed to have therapeutic potential. In contrast, in Huntington’s disease and Alzheimer’s disease H2S levels are not enough to properly function (Paul et al., 2014; Sbodio et al., 2016; Sbodio et al., 2018; Cao et al., 2018; Vandini et al., 2019). For these diseases a supplementation of H2S may have a benefit. In schizophrenia both beneficial and toxic effects of H2S and H2Sn have been reported (Koike et al., 2016; Topcuoglu et al., 2017; Unal et al., 2018; Xiong et al., 2018; Ide et al., 2019). The balance of H2S, H2Sn and NO as well as S-sulfuration together with S-nitrosylation plays an important role for the pathogenesis of these neuronal diseases.