The rate of GTP hydrolysis on p21ras is accelerated by similar to 10(5) times by the catalytic domains of GTPase-activating proteins (GAPs), p120-GAP (GAP-344) or neurofibromin (NFL-334). The kinetic mechanism of this activation has been investigated by following the release of inorganic phosphate (P-i), using a fluorescent probe that is sensitive to P-i [Brune, M., Hunter, J., Corrie, J. E. T., & Webb, M. R. (1994) Biochemistry 33, 8262-8271]. Measurements were made in real time with a stopped-flow apparatus, in which the p21ras complex with the 2',3'-methanthraniloyl analogue of GTP (mantGTP) was mixed with the GAP in the presence of this P-i probe. The results show that P-i release is fast and that the overall hydrolysis is contrlled by the cleavage itself or a conformational change preceding the cleavage. The time courses were single exponentials over a range of [GAP-344] and were modeled to show that a single step controlled P-i release. The maximum rate constant was 15 s(-1) (all data at 30 degrees C, pH 7.6, low ionic strength) in experiments in which GAP-344 underwent a single turnover, compared with 5 s(-1) for multiple-turnover experiments, and possible causes of this discrepancy were investigated and discussed. With NF1-334 the time courses were more complex, showing a lag prior to rapid release of P-i. The results were consistent with a K-d of 0.04 mu M for NF1-334 binding to the p21ras . mantGTP complex, and two steps partially contributing to the overall rate. This NF1-344 affinity is some 3 orders of magnitude tighter than that of GAP-344. The cleavage rate at saturating NF1-334 increases from similar to 10 to 30 s(-1) as the ionic strength increases to physiological, although the interaction between NF1-334 and p21ras weakens with increasing ionic strength. The data are interpreted with GTP hydrolysis occurring uncoupled from any transduction, such that once the GAP interacts with p21ras, the hydrolysis (deactivation) occurs at the maximum rate possible due to the catalytic mechanism.