We studied the kinetics of the use-dependent block of the Na+ current (I(Na)) by disopyramide and lidocaine. I(Na) was recorded from isolated guinea pig ventricular myocytes by using the whole-cell patch-clamp technique. The use-dependent block of I(Na) by disopyramide with 20- and 200-msec depolarizing pulses developed in two exponential functions. The degree of the use-dependent block and the amplitude of the fast (A(f)) and slow (A(s)) components with the short (20-msec) pulse protocol were comparable to those with the long (200-msec) pulse protocol. When pH was raised from 7.3 to 8.0, disopyramide increased A(f) without a change in A(s). At pH 6.5, I(Na) block developed with a single exponential function revealing only the slow component. The fast and slow components of I(Na) block by disopyramide could be explained by binding of the uncharged and charged forms, respectively, to the activated state of the channel. Development of I(Na) block by lidocaine also was expressed by two exponentials at all pulse durations (5-200 msec). As pulse durations were prolonged or holding potentials were depolarized, the degree of the use-dependent block and A(f) increased. When pH was lowered to 6.5, the short pulse produced only the slow component, whereas the long pulse caused two exponentials with decreased A(f) and increased A(s). Internal application of QX-314, a permanently charged lidocaine analogue, produced a single exponential block of I(Na) with a very slow onset rate. Therefore, binding of the charged form of lidocaine to the activated state seemed to be responsible for the slow component and binding of the uncharged form to the inactivated state for the fast component. These results suggest that the charged and uncharged forms of disopyramide and lidocaine are responsible for two components of I(Na) block in pulse trains, but their forms are not a sole determinant of state-dependent binding to the Na+ channel.