The human red blood cell enzyme glutathione reductase (GR) was reporte
d to be inhibited by the anticonvulsant drug valproic acid (VPA) [Cota
riu et al., Biochem Pharmacol 43: 425-429, 1992]. When attempting to r
eproduce and extend these experiments, we could not detect any signifi
cant effect of VPA on glutathione reductase in haemolysates from 20 he
althy children and 10 children under VPA therapy, no matter which conc
entration of the drug (0.9 or 1.8 mM in a haemolysate diluted by a fac
tor of 50 or 1.8 mM directly in the assay), which incubation time (0-6
0 min) and which assay system were chosen. An influence of VPA on FAD-
free apoglutathione reductase was also excluded. GR-activities of 10 c
hildren under VPA therapy (1.08 +/- 0.14 U/mL blood or 7.57 +/- 0.94 U
/g Hb) were almost identical with the activities of age- and sex-match
ed controls (1.04 +/- 0.17 U/mL or 7.79 +/- 1.32 U/g Hb). No correlati
on between erythrocyte GR activity and serum levels of VPA was observe
d. Finally, incubation of crystalline human GR with VPA did not lead t
o enzyme inhibition; rather, in most experiments the enzyme was stabil
ized by incubation with VPA. Possible explanations for the discrepanci
es between the results of Cotariu et al. and our data are discussed.