Platelet surface glutathione reductase-like activity

Glutathione, an important modulator of the cellular redox environment, is found in blood where it could modulate the redox environment of platelets. Cells contain millimolar concentrations of glutathione, mostly in the reduced form. Plasma contains only 8 to 25 μM glutathione,^1-6  also predominantly in the reduced form. This contrasts with other low-molecular-weight thiols that are mostly in disulfide forms.^2,6  Plasma glutathione levels and GSH/glutathione disulfide (GSSG) ratios in plasma are altered in disease states.^4,5,7  However, little is known about the role of plasma glutathione in the regulation of redox reactions. We recently found that physiologic concentrations of GSH generate thiols from disulfide bonds in α_IIbβ₃ and potentiate platelet aggregation.⁷ 

Recent reports have also raised the possibility that platelets have a transplasma membrane oxidoreductase that, like GSH, can reduce extracellular disulfide bonds in redox-sensitive membrane proteins.^7-9  Although there is little information about such systems in cells, a form of a cell membrane nicotinamide adenine dinucleotide (NADH) oxidase^10,11  in plant cells has been reported to have disulfide reductase activity.¹²  Cells and platelets contain flavoprotein-electron transport systems such as nicotinamide adenine dinucleotide phosphate (NADPH) oxidase on neutrophils; however, the electron acceptor is molecular oxygen (not disulfide bonds) and the product is superoxide anion (O^–₂).^13-15  There are no reports of transplasma membrane electron transport systems that reduce disulfide bonds in mammalian cells. We here report that nonstimulated platelets have the ability to reduce the disulfide bond in extracellular GSSG. This results in the production of GSH, the generation of thiols in the α_IIbβ₃ integrin, and the facilitation of platelet aggregation.

Platelet aggregation, measurement of extracellular sulfhydryls in platelet samples, and sulfhydryl labeling of extracellular of platelets were performed as described elsewhere.⁷ 

We previously found that GSH or a mixture of GSH/GSSG potentiated platelet aggregation stimulated by agonists such as collagen.⁷  When GSSG (5 μM) was used alone as a control for the collagen-induced aggregation studies, surprisingly, it also enhanced platelet aggregation (Figure 1). As expected, GSSG alone did not contain any GSH by the 5,5′-dithiobis-(2-nitrobenzoid) acid (DTNB) assay (not shown). When the concentration of agonist gave partial or submaximal aggregation, a low concentration of GSSG (2 μM) further stimulated aggregation (Figure 1B). Thus, under appropriate conditions low concentrations of glutathione, similar to those found in blood, substantially potentiate aggregation.

We hypothesized that the GSSG was being converted to GSH by the platelets. To test this we incubated GSSG with gel-filtered platelets and measured total thiols in the sample. The combination of GSSG with platelets gave substantially more free thiols as detected by DTNB than could be accounted for by either platelets or GSSG alone (Figure 1C). The amount of sulfhydryls on platelets alone corresponded to about 9.0 × 10^–18 mol thiol/platelet (similar to what has been reported).^16,17  Platelets with GSH (10 μM) gave the additive results of platelets alone plus GSH alone (column 3). Platelets with GSSG (5 μM) gave about double what was expected from platelets alone (about 6.5 μM additional SH; P < .05; n = 5). This suggests that 3 to 4 μM of the GSSG added is being converted to GSH by platelets. Therefore, the addition of GSSG to platelets would result in a ratio of GSH/GSSG of approximately 3:1 to 4:1. This ratio is similar to that found in blood and to the GSH/GSSG ratio that provides maximal potentiation of platelet activation.⁷  In contrast to the effect of platelets on GSSG, when cystine (the disulfide form of cysteine) was incubated with platelets no increase in sulfhydryls was found (Figure 1D). This indicates some specificity of the reaction.

The conversion of GSSG to GSH requires the addition of 2 electrons to reduce the disulfide bond. In cells NADPH serves as the electron donor for GSSG in the following reaction: GSSG + NADPH + H⁺ → 2 GSH + NADP⁺. Therefore, the generation of GSH by platelets in the absence of external reducing agents implicates a transmembrane electron transport system that uses a cytoplasmic electron donor, presumably NAD(P)H. NADPH oxidase^13,14  and NADH oxidase^10-12,18,19  are plasma membrane oxidoreductase systems found in cells. To begin to explore these possibilities, we tested the effect of diphenyleneiodonium (DPI), an inhibitor of flavin-containing enzymes, as well as a structurally unrelated inhibitor of NADPH oxidase, apocynin.²⁰  Both DPI (10 μM) and apocynin inhibited the GSSG reductase activity (Figure 1E). Because the reaction is inhibited by relatively low concentrations of DPI, the mechanism likely involves a flavoprotein.

Time-course studies on the conversion of GSSG to GSH were performed at 37° C, the conditions under which GSSG enhanced platelet aggregation. The platelet reaction with GSSG was complete by 1 minute (Figure 2A). This suggests that the effect of GSSG on platelet aggregation occurs after some GSSG is converted to GSH, and thus after an appropriate redox potential for platelet activation is reached. To test for this possibility we compared the effect of GSSG added 1 minute before collagen to GSSG added simultaneously with collagen. Figure 2B shows that GSSG (5 μM) added simultaneously with collagen did not potentiate aggregation (the 0-second tracing). In contrast to the lack of effect of GSSG added with collagen, GSSG incubated for 1 minute with platelets as expected facilitated aggregation (the 60-second tracing). As expected from previous studies,⁷  controls of GSH (10 μM) or a mixture of GSH (10 μM) and GSSG (2 μM) potentiated aggregation when added simultaneously with collagen (not shown). These results imply that platelets convert GSSG to GSH and this can be part of the stimulus for platelet activation.

The effect of GSH (10 μM) and GSSG (5 μM) on sulfhydryl generation in α_IIbβ₃ is shown in Figure 2C. Using immunoprecipitation the bands shown were previously identified as α_IIb and β₃.⁷  GSH, previously found to generate sulfhydryls in the β₃ subunit, increased sulfhydryl labeling in β₃ by 2-2.5 fold. A concentration of GSSG (5 μM) that potentiated platelet aggregation increased sulfhydryl labeling in the β₃ subunit by almost 3-fold (Figure 2D). This is similar to the effect of GSH (10 μM) and confirms a mechanism in α_IIbβ₃. A 1.6- to 1.7-fold increase in labeling of α_IIb was also found with both GSH and GSSG (not shown). Because GSH by itself does not cause platelet aggregation,⁷  the sulfhydryl generation in β₃ apparently does not by itself activate the receptor. An additional stimulus is necessary.

In summary, the conversion of GSSG to GSH by platelets implicates a novel surface flavoprotein GSH reductase activity that may be a marker for a more general disulfide reductase activity. This activity appears to provide the appropriate redox potential for platelet activation. This reaction may modulate the local redox environment at sites of vascular injury where platelets are found in high concentrations. Redox-sensitive protein on platelets whose function may be modulated by the redox environment include α_IIbβ₃ and protein disulfide isomerase.²¹