THIAMIN AND DERIVATIVES AS MODULATORS OF RAT-BRAIN CHLORIDE CHANNELS

Several membrane fractions were prepared from rat brain by differential and sucrose density gradient centrifugation. Most fractions took up Cl-36- rapidly at a rate linear with time during the first 30-60 s, then the rate progressively slowed down. The lowest rate of uptake was found in the mitochondrial fraction. Oxythiamin partially inhibited Cl-36- uptake in all fractions. In P2 (crude synaptosomal fraction), oxythiamin decreased the initial rate of uptake by 32%, the apparent K(i) being 1.5 mM. Thiamin and amprolium were less effective as inhibitors. 4,4'-Diisothiocyanostilbene-2,2'-disulfonic acid (0.1-1 mM) inhibited Cl-36- uptake by 40 50%. In the presence of this compound at a concentration greater-than-or-equal-to 5 x 10(-4) M, oxythiamin became ineffective. Cl-36- uptake was increased by GABA (0.1 mM) and this effect was antagonized by picrotoxin as expected, but not by oxythiamin. The rate of Cl-36- uptake did not appreciably depend on the external chloride concentration and was unaffected by bumetanide or by replacement of external Na+ by choline. Taken together, these data suggest that the oxythiamin-sensitive Cl-36- influx is essentially diffusional and is not related to the GABA receptor or the Na:K:2Cl co-transport. Partial replacement of external Na+ by K+ or treatment with 0.1 mM veratridine (which should both result in membrane depolarization) increased Cl-36- uptake by 50 and 30% respectively; the inhibitory effect of oxythiamin was enhanced to the same proportion. When a brain homogenate was preincubated in the presence of 1 mM thiamin diphosphate and a synaptoneurosomal fraction was subsequently prepared from the homogenate, the vesicles took up more Cl-36- than control vesicles; the thiamin triphosphate content of the vesicles was also increased by pretreatment with thiamin diphosphate. When Cl-36- uptake was measured in vesicles depolarized by K+ or veratridine, the difference between thiamin diphosphate-treated and control vesicles disappeared. The difference also disappeared in the presence of oxythiamin. A possible interpretation of these data is that membrane depolarization favors the opening of a chloride channel which is modulated by thiamin derivatives.