Rationale: Several drugs used to treat bipolar disorder (lithium and carbamazepine), when administered chronically to rats, reduce the turnover of arachidonic acid, but not docosahexaenoic acid, in brain phospholipids by decreasing the activity of an arachidonic acid-selective phospholipase A(2). Although chronic valproic acid produces similar effects on brain arachidonic acid and docosahexaenoic acid turnover, it does not alter phospholipase A(2) activity, suggesting that it targets a different enzyme in the turnover pathway.
Materials and methods/results: By isolating rat brain microsomal long-chain fatty acyl-CoA synthetases (Acsl), we show in vitro that valproic acid is a non-competitive inhibitor of Acsl, as it reduces the maximal velocity of the reaction without changing the affinity of the substrate for the enzyme. While valproic acid inhibited the synthesis of arachidonoyl-CoA, palmitoyl-CoA, and docosahexaenoyl-CoA, the K (i )for inhibition of arachidonoyl-CoA synthesis (14.1 mM) was approximately one fifth the K (i) for inhibiting palmitoyl-CoA (85.4 mM) and docosahexaenoyl-CoA (78.2 mM) synthesis. As chronic administration of valproic acid in bipolar disorder achieves whole-brain levels of 1.0 to 1.5 mM, inhibition of arachidonoyl-CoA formation can occur at brain concentrations that are therapeutically relevant to this disease. Furthermore, brain microsomal Acsl did not produce valproyl-CoA.
Conclusions: This study shows that valproic acid acts as a non-competitive inhibitor of brain microsomal Acsl, and that inhibition is substrate-selective. The study supports the hypothesis that valproic acid acts in bipolar disorder by reducing the brain arachidonic acid cascade, by inhibiting arachidonoyl-CoA formation.