Stability Of Gsno In Vitro

Stability of RSNO is compromised by thermal decomposition, photolysis and the catalytic decomposition by trace metal ions like iron and copper. Although the various RSNO share the common C S N=O motif, they show a big variation in intrinsic stability when in solution.

The reason for this variation is not well understood. Given this instability, aqueous solutions of RSNO tend to release a certain quantity of gaseous NO and accumulate typical decomposition products like thiols, disul des and reaction products of thiyl radicals. It was recently discovered that, in presence of oxygen, a certain quantity of disul de S-oxides (GS(O)SG) and disul de S-dioxides (GS(O2)SG) [37,38] is formed as well. These metabolites of GSH are effective agents for S-glutathiolation of sulfhydryl groups. They can inactivate glycer-aldehyde 3-phosphate and alcohol dehydrogenases, and release zinc from metallothionin and zinc nger proteins. Such mixtures of decomposition products result from a combination of pathways operating simultaneously. In practical applications usually more than just a single pathway contributes signi cantly. The basic decomposition reactions will be discussed below.

In solution, RSNO undergo slow thermal decomposition into a disul de and free NO according to

The rate of thermal decomposition depends on the concentration and type of thiol [39,13]. At starting concentrations of 50 mM, the thermal decomposition of GSNO and CysNO exceeds 10% after ca 6 and 1 h, respectively. At low concentrations in absence of UV or blue light, the stability of RSNO is quite high. In presence of the metal chelator DTPA, even CysNO achieves a half life of 11 h [40]. Therefore, Eq. (5) is not signi cant unless catalyzed by trace metal ions (see below).

Illumination with UV or blue light causes homolytic cleavage of the S N bond. The probability for photolysis is highest at wavelengths near the absorption maximum of RSNO at 336 nM. The quantum yield for photolysis is quite high and exceeds that of photodissociation of the nitrosyl ligand from heme. Fig. 4. shows the photolysis of GSNO by successive sets



300 400 500 600 700

Fig. 4. Photolysis of GSNO by pulsed UV irradiation at 355 nM at room temperature in oxygenated PBS buffer. The successive curves differ by 10 pulses of 50 mJ/pulse. The inset shows the simultaneous photolytic release of free NO radicals as measured with an NO electrode. (From Ref. [41].)

300 400 500 600 700

Wavelength (nm)

Fig. 4. Photolysis of GSNO by pulsed UV irradiation at 355 nM at room temperature in oxygenated PBS buffer. The successive curves differ by 10 pulses of 50 mJ/pulse. The inset shows the simultaneous photolytic release of free NO radicals as measured with an NO electrode. (From Ref. [41].)

of 10 laserpulses (355 nM, 50 mJ/pulse). The photolytic decomposition of S NO bonds has raised interest in the application of S-nitrosated complexes for photodynamic therapy of cancers [41]. It should be mentioned that S-nitrosothiols are not the only endogenous species showing such photolytic release of NO: Alongside endogenous S-nitrosothiols, also nitrite was shown [42] to contribute signi cantly to the light-induced release of NO from vascular tissues of rats.

Reduced trace metal ions like Fe2+ and Cu+ are ef cient catalysts for the decomposition of RSNO [43,44]. Fig. 5 shows the effect of copper and iron on the kinetics of CysNO. The decomposition of GSNO is slower, but qualitatively similar.

The reaction mechanism [46] is given by the following sequence

The reaction releases disul des and free NO. Fe2+ also can catalyze a similar sequence.

The above reactions might leave the erroneous impression that iron and copper act on S-nitrosothiols in similar ways. In reality the iron-catalyzed decomposition of RSNO is complicated by a competing reaction mechanism involving iron complexes carrying two nitrosyl ligands. In principle, copper also may form (di)nitrosyl species like EPR silent

Fig. 5. Effect of copper and iron on the decomposition of 0.9 mM CysNO in HEPES buffer (150 mM, pH = 7.4) as monitored by the optical absorption at 340 nM. The decomposition was studied in the presence of 1 mM free cysteine. (From Ref. [45].)

Cu+(NO)2 or paramagnetic Cu+(NO) or Cu2+(NO)2. However, these species have so far only been reported as adsorbants in dry anoxic porous solids [47] but not in aqueous solutions. Therefore, in solutions, iron alone can interact with S-nitrosothiols to form such DNICs. In the presence of iron, a pool of DNIC will form at the expense of S-nitrosothiols [43,45]. These reactions were well studied in vitro and are described in Chapter 11.

Transnitrosation to other types of thiols may have signi cant effect on the lifetime of S-nitrosothiols. The basic transnitrosation reaction involves the transfer of a nitrosonium NO+ moiety

Here Ri—S-, R2—S- can be various deprotonated thiols, from small cysteine anions to sulfhydryl groups on macromolecular proteins. The rates for this reversible transfer vary considerably with the type of thiol, pH and temperature. Transnitrosation rates for cysteine and glutathione were reported [14] at around 80 (Ms)-1 (37°C, pH 7.4). Lower rates of 3 9 (Ms)-1 apply for transnitrosation from S-nitrosoalbumin to cysteine and glutathione. The transfer rates increase signi cantly when pH is raised and a larger fraction of thiols is deprotonated. It was shown that 50 ^M CysNO and S-nitrosoalbumin establish their equilibrium by transnitrosation on a timescale of several minutes in vitro (cf Fig. 6) [48]. Transnitrosation from GSNO is slower due to a combination of smaller transfer rate and orders of magnitude lower concentration of GSNO (Table 1). Therefore, the characteristic timescale for equilibration with GSNO should be many minutes. In human plasma, albumin was con rmed to be the dominant target for transnitrosation from CysNO and GSNO [49], with metal chelators having only insigni cant effect on the extent and rate of S-nitrosation of the albumin target. In whole blood, the reaction balance is shifted considerably because Hb may be S-nitrosated or nitrosylated at the heme. In addition, oxyHb is an ef cient scavenger

0 225 450 675 900

Fig. 6. Kinetics of transnitrosation from CysNO to albumin in 20 mM Tris buffer (pH 7.4, 23°C) in presence of 1 mM metal chelator EDTA. The curves show kinetics for [CysNO] = 50 ^M (bottom), 100, 150 and 200 ^M (top). The transnitrosation reaction was monitored at 412 nM by photometric detection of released Cys- with the DTNB reagent. (From Ref. [48].)

0 225 450 675 900

Time (sec)

Fig. 6. Kinetics of transnitrosation from CysNO to albumin in 20 mM Tris buffer (pH 7.4, 23°C) in presence of 1 mM metal chelator EDTA. The curves show kinetics for [CysNO] = 50 ^M (bottom), 100, 150 and 200 ^M (top). The transnitrosation reaction was monitored at 412 nM by photometric detection of released Cys- with the DTNB reagent. (From Ref. [48].)

of free NO (but not for S-nitrosothiols, see below). In whole blood, S-transnitrosation from GSNO to S-nitrosohemoglobin was reported [49] to remain below a few percent.

The previous in-vitro results should not leave the impression that S-transnitrosation be an exclusively passive process with timescales largely determined by the chemical characteristics of the exchanging thiols. In biological systems, the crossing of cellular membranes will act as a signi cant barrier for S-nitrosation and affect the rate of exchange between the intra- and extracellular compartments. This effect has been shown in vivo: Protein disul de isomerase (PDI) is a protein in cellular membranes that normally catalyzes thiol-disul de exchange. The activity of the PDI protein was found to enhance the intracellular pool of S-nitrosothiols in cultured human erythroleukemia cells [50]. The mechanism of membrane crossing of S-nitrosothiols will be discussed later in this chapter.

The process of transnitrosation from low to high molecular weight has been demonstrated in vivo in rabbits [51]. A collection of transnitrosation rates between various thiols has been collected in Ref. [14]. Analogous results were obtained with transnitrosation from S-nitrosoglutathionyl-sepharose beads [52] to free thiols in solution. The transnitrosation to cysteine and glutathione was rapid and accelerated by an order of magnitude when pH was raised from 5 to 9. In contrast, transnitrosation from the beads towards bovine serum albumin was negligible.

It should be noted that reaction (7) proceeds in the presence of a competing pathway which leads to the formation of disul de bonds and the release of some free NO. It has recently been shown [53] that the sulfhydryl groups of certain proteins can also be modi ed by nucleophilic attack of the protein thiolate on the sulfur of GSNO rather than on the NO+ moiety. This reaction pathway amounts to S-glutathiolation of the sulfhydryl moiety. Several proteins underwent a combination of S-nitrosation and S-glutathiolation when exposed to GSNO. In contrast, bovine serum albumin, actin and alcohol dehydrogenase were only S-nitrosated by fresh GSNO. S-glutathiolation of intracellular proteins has been demonstrated in vivo: upon incubation with exogenous CysNO, NIH-3T3 broblasts underwent combined S-nitrosation and S-glutathiolation of the cysteine residues of H-ras protein [54]. High capacity of S-glutathiolation was attributed [37] to glutathione disul de S-oxide (GS(O)SG) which appears as one of the decomposition products of GSNO itself.

The lifetime of hours for GSNO may be shorted to minutes by addition of other thiols, in particular SNAP and cysteine [55 57]. The effect may be inhibited by the thiol-blocking compound N-ethylmaleimide [58]. Two different mechanisms have been identi ed for this phenomenon: transnitrosation and formation of disul des. In the rst case, the GSNO is depleted by the transnitrosation to the shorter lived CysNO. In the same spirit, transnitrosation to more stable species may stabilize the pool of S-nitrosothiols up to a certain degree. The second mechanism is mediated by the reductive nature of the thiol anions and involves the formation of a disul de bridge under release of a nitroxyl anion [24,59]:

If RS represents glutathione, the reaction releases GS SG disul de at a rather slow rate with a second order rate constant of k = 8.3 x 10-3 (Ms)-1 [24]. Given that tissues contain GSH below mM concentrations (Table 1), this decay channel of GSNO appears insigni cant in vivo. If on the other hand RS represents a sulfhydryl group on a protein, the reaction (8) amounts to S-glutathiolation of the protein. It has been con rmed that the coincubation with a mixture of GSNO and glutathione causes S-glutathiolation of the sulhydryl groups on a wide range of different proteins. [53,60]. More detailed studies [37] have shown that intermediate oxides like GS(O)SG or GS(O2)SG are better S-glutathiolating agents than GSNO itself (see below). Therefore, the process of S-glutathiolation may in fact be more complex than suggested by Eq. (8). S-glutathiolation maybe reversed by dithiothreitol [60].

Release of NO radicals from RSNO can be caused by reductants, exogenous as well as endogenous. Ascorbate was found to have a pronounced effect on the lifetime of RSNO [39]. Without chelation of spurious copper, small quantities of ascorbate reduced Cu2+ to Cu+ and initialized the reaction Eq. (6) with formation of disul des and the release of free NO. After chelation of spurious copper, the true reduction of RSNO became apparent with the release of free NO and thiols instead of disul des. The effective rate constant accounts for the reduction pathway plus a thermal decomposition rate kj ke = k[Asc] + kT (9)

At pH = 7.4, the value of k was 0.25, 0.015 and 0.032 (Ms)-1 for CysNO, GSNO and SNAP, respectively. Careful observations showed that the ascorbate monoanion HA and the dianion A2- have different rates of reduction. Therefore, the rate k was highly dependent on pH and increased over four orders of magnitude when the pH was changed from 3.6 to 11 [39].

It might seem plausible to expect that the stability of nitrosothiols be inversely related to its potency as vascular effector. More concretely, one might expect that the more rapidly decaying nitrosothiols elicit stronger physiological responses for relative short times. When tested for the responses of vasorelaxation and platelet aggregation, no such correlation was found [61]. This result con rms that S-nitrosothiols are potent physiological effectors in their own respect. In particular, they can act without having to release free NO.

+1 0

Post a comment