In this study we have examined the mechanism by which CysNO activates sGC in both the SH-SY5Y neuroblastoma cell line and in HPASMCs. These cell lines were chosen as they both contain sGC, but are of diverse origin. We show that CysNO-dependent cGMP formation exhibits a biphasic response in both cell types. CysNO levels up to 20 μM stimulate cGMP formation, whereas levels above 20 μM diminish cGMP formation. Temporally, 20 μM CysNO results in rapid and sustained activation of sGS, whereas 120 μM CysNO again shows a biphasic response with an initial increase and slow decrease in cGMP level.
The activation of cGMP formation observed at lower levels of CysNO was inhibited by both L-leucine and oxyHb. L-Leucine is a ligand for the L-AT system [
21], and we have previously demonstrated that it will inhibit the intracellular accumulation of S-nitrosothiols in cells exposed to CysNO [
7,
10]. We show in that both cell lines used in this study respond in an analogous way to all other cell lines we have so far investigated in that the uptake of CysNO is inhibited by L-leucine and to a much lesser extent by D-leucine. This indicates that L-AT-dependent CysNO uptake is active in both the SH-SY5Y and HPASMC cells. Recent siRNA studies of both L-AT1 and L-AT2 have confirmed the importance of this transporter in transmembrane S-nitrosothiol transport [
9]. Cysteine is a poor substrate for the L-AT system, but the S-nitrosation of the cysteine thiol appears to confer enough hydrophobicity on the amino acid to make it a strong ligand for this transporter system. The fact that CysNO-dependent cGMP formation was inhibited by L-leucine indicates that the transport of CysNO to the interior of the cell is an absolute requirement for the activation of sGC. This was not the case for the spontaneous NO donor SPER/NO, which liberates NO in the extracellular space. Interestingly, CysNO-mediated cGMP formation was also inhibited by oxyHb suggesting that CysNO does not directly activate cGMP but requires prior metabolism to form NO. As NO is a freely diffusible molecule and extracellular NO generated from SPER/NO is clearly able to activate sGC, these data indicate that very little CysNO spontaneously decays to NO in the extracellular space. It should be noted that these experiments were performed in HBSS in the presence of the metal ion chelator DTPA to specifically minimize metal ion-dependent CysNO decay. Of great interest, these data suggest that cells have the ability to reduce intracellular S-nitrosothiols to form NO allowing the activation of NO-dependent pathways. The mechanisms by which NO activates sGC are relatively well established and involve binding to an open coordination site of the ferrous heme group of sGC, which displaces a proximal histidine from the heme iron resulting in a conformational activation of enzyme activity. Recent evidence suggests heme-NO bound sGC only accounts for tonic activation whereas a second non-heme binding site is present for transient activation in response to acute stimuli [
24]. However, other investigators did not observe such behavior in cellular systems and suggest only a single ligand binding site is physiologically relevant [
25]. The mechanism for intracellular CysNO reduction is currently unknown, but appears to involve the activity of a flavoprotein reductase enzyme as it can be inhibited by the non-specific flavoprotein inhibitor DPI. Previous work has shown that glutathione-dependent formaldehyde dehydrogenase has the ability to reduce S-nitrosoglutathione in cells in an NADH-dependent manner [
26,
27]. More recently it has been suggested that carbonyl reductase 1 can act as an NADPH-dependent GSNO reductase [
28], however, in bovine aortic endothelial cells, we only detected an NADH-dependent activity [
10]. Regardless of the enzyme, these reductases generate hydroxylamine or ammonia but not NO, and therefore cannot be responsible for CysNO-dependent sGC activation. Tissue homogenates have been shown to possess the ability to metabolize S-nitrosothiols to nitrate, suggesting NO as an intermediate, by a mechanism independent of metal ions and cellular thiols, but inhibitable by potassium ferricyanide [
29].
As shown in , the model that best fits our data involves the uptake of CysNO via the L-AT system, followed by intracellular reduction of CysNO, or some other secondary S-nitrosothiol that is formed via transnitrosation, to liberate NO which can then proceed to bind to, and activate sGC.
At higher concentrations of CysNO, we observed diminished cGMP formation, consistent with the recent work of Sayed et. al. [
11,
12], that was due to direct inhibition at the level of sGC. This latter conclusion was drawn from the fact that cGMP formation could not be stimulated by SPER/NO in cells that had been co-treated with CysNO. It has been proposed that CysNO is able to directly inactivate sGC via the S-nitrosation of a protein thiol [
11]. Although our data support such a mechanism, it is crucially important to assess protein S-nitrosation and other thiol modifications in the context of cellular changes in thiol redox state. The higher levels of CysNO that inhibit sGC activity also decrease cellular glutathione levels, and this is seen in both concentration- and time-dependent studies. In addition, sGC inhibition only occurs at total intracellular S-nitrosothiol levels in the nmol/mg range that are 1–2 orders of magnitude higher than those observed under (patho)physiological conditions [
30]. Thus, sGC inactivation takes place under conditions when antioxidant defenses are depleted and additional proteins will be subjected to oxidative modifications that would not typically be modified if these systems were intact. Consequently, the perceived tolerance to CysNO-dependent vasodilation needs to be assessed in the context of total cellular redox status and the total level of thiol modification within the cell or tissue. The L-AT-dependent CysNO uptake mechanism appears avid, and exposure of cells to relatively low levels of CysNO may generate a significant oxidative stress on the cell due to the rapid concentration of extracellular CysNO into the small volume of the cell [
10].
The data presented here give credence to the hypothesis that CysNO may represent a paracrine modulator of vascular tone as mechanisms of uptake and sGC activation are present within vascular smooth muscle cells. It is likely that the pharmacological activity of S-nitrosothiols is mediated by such processes, and this is likely the basis of the chiral specificity observed with some effects of S-nitrosothiols in vivo [
6]. In addition, the ability of inhaled NO to elicit vascular effects may occur through the stabilization of the vasodilatory activity in the form of an S-nitrosothiol. What remains unclear is whether this represents a physiological process. The formation of S-nitrosothiols in specific locations or cell types, such as in the red blood cell through the nitrite reductase activity of deoxyhemoglobin [
31–
33], may represent a pool of S-nitrosothiols that can be transferred to the vasculature to elicit a vasodilatory response. Of interest in this regard is the increase in red blood cell S-nitrosothiol content in the presence of nitrite [
34] and the observation that the vasodilatory ability of red cells from septicemic animals is correlated to their S-nitrosothiol content [
35].
In conclusion CysNO requires both transmembrane transport and intracellular metabolism in order to release NO and activate sGC. CysNO metabolism occurs via the activity of an as yet unidentified flavoprotein reductase. At higher concentrations CysNO directly inhibits sGC, most likely through the modification/S-nitrosation of essential thiol residues. However, such inhibition only takes place in the context of glutathione depletion and the accumulation of high and non-physiological concentrations of protein S-nitrosothiols. These data provide a mechanism by which extracellular S-nitrosothiols can elicit vasodilation by targeted NO delivery to the smooth muscle cell.