Fluoride is a slow, tight-binding inhibitor of the calcium ATPase of sarcoplasmic reticulum

Abstract

Addition of sodium fluoride in the millimolar concentration range to a solution containing the sarcoplasmic reticulum CaATPase undergoing turnover in its vesicular or nonionic detergent-solubilized forms produced a slow (time range of minutes) complete loss of enzymatic activity. In the presence of magnesium and the absence of calcium, similar results were obtained under nonturnover conditions. Time courses were adequately fit by a function corresponding to a monophasic transformation with a pseudo first order rate constant kobs. In the absence of Mg2+ (EDTA present) no inhibition developed; kobs depended hyperbolically on the Mg2+ concentration with the half maximal effect occurring near 4 mM. The fluoride concentration dependence of kobs showed no evidence of approaching saturation (highest [F-] used was 40 mM) and corresponded to a rate law which was approximately second-order with respect to fluoride. A number of ligands known to bind to the CaATPase were found to decrease kobs. Calcium prevented onset of fluoride inhibition with a midpoint in the micromolar range, implying an effect due to binding at the high affinity transport sites. ATP also protected with a midpoint in the micromolar range, consistent with an effect caused by active site binding of the nucleotide; protection was only partial, suggesting the ATPase can bind fluoride and ATP simultaneously. Prevention of fluoride inhibition by Pi occurred with a [Pi]1/2 of 12 mM at pH 6.5, a concentration similar to that which produces active site phosphorylation. Finally, protection by orthovanadate was found to be competitive and have a midpoint of 5 microM. These results point to an effect exerted at or near the phosphorylation site. The value of kobs increased from essentially zero above pH 8 to a plateau below pH 6; the transition had a midpoint near pH 7.2. Inhibition persisted after removal (with EGTA present) of unbound fluoride by dialysis. Reversal of fluoride inhibition was very slow, with a t1/2 of 16 h at 37 degrees C. These results suggest that fluoride behaves like a slow, tight-binding inhibitor of the ATPase and that the resulting complex is a stable transition (or intermediate) state analog. Plausible molecular bases for our results are that fluoride acts at the phosphorylation site as an analog of Pi or of hydroxide, which may be considered a substrate in the normal hydrolysis of the phosphorylated enzyme. A role for aluminum was ruled out after finding that the addition of EGTA to 10 mM or aluminum sulfate to 0.2 mM or deferoxamine to 0.5 mM produced no significant change in kobs.