Supplemental data
Supplemental Figures
Files in this Data Supplement:
- Supplemental Fig. 1 - Supplementary Figure 1. Effect of modifying rate constants for blocked states in the trapping block model on current time course. Simulations of voltage step responses (?70 to +40 mV) is shown using the ATB model with varying rate constants to observe the effect on the shape of the response. A, Effect of variation of blocked rate constants that lead away from the open state. The original values of rate constants are multiplied by the factor indicated by color at left. Kd = 1 sec-1; koff = 85 sec-1; ? = 91.6 sec-1; kr = 1.8; kon = 5000 mM/sec; ? = 46.1 sec-1. B, As in A, but modifying blocked rate constants that lead towards the open state. A, B, Current scale bars are equivalent to 15 pA, assuming a population of 100 channels. C, For comparison, an experimental current response to a voltage step (?70 to +40) fitted to the ATB model with parameters as shown. D, Zooms of response in C to show the fit to fast responses of unblock and block. Note that C and D are the same as in Fig. 6, A and B, in our study. This figure shows that only variation of ?� produces quantitatively correct behavior, reproducing the amplitude and on and off time course of unblock.
- Supplemental Fig. 2 - Supplementary Figure 2. Tail currents after repolarization to various potentials during stationary activation of NMDARs are also fitted well by the ATB model. A, Repolarizing voltage steps from +40 mV were applied during stationary activation of NMDARs evoked with continuous perfusion of 30 �M NMDA. Top, Voltage commands. Bottom, Corresponding current responses. B, Same currents as in A, at an expanded time scale. C, Corresponding responses of ATB model, with ?� = 300, ?� = 40, kd = 1, kon = 2000 mM/sec, koff = 32.8 sec-1, compared with B. D, Tail currents are shown as conductance changes for a reversal potential of ?5 mV and the voltage command. E, Conductance change predicted by ATB model, compared with D. This figure shows that the ATB model also accounts well for very slow relaxations during reblock, including overshoots (slow tail currents) at intermediate membrane potentials.
- Supplemental Fig. 3 - Supplementary Figure 3. ATB model is consistent with previous studies involving voltage steps. A and B show simulations of the perfusion and voltage step protocol used by Benveniste and Mayer (1995) (compare directly to their Fig. 3C), where 200 �M glutamate was applied for 20 msec (as indicated by the black bar) either alone (black trace) or together with 50 �M Mg2+ (as indicated by the red bar, red current response). Cells were voltage clamped at ?100 mV, and voltage steps to +60 mV were applied 200 msec after glutamate/Mg2+ perfusion (kd = 8.4 sec-1). Simulations for STB are shown in A and for ATB in B. C and D show simulations of Mg2+ blocking and recovery kinetics in the constant presence of agonist as studied by Sobolevsky and Yelshansky (2000) (compare directly to their Fig. 8). Perfusion of 100 �M aspartate as indicated by the top black bar and block?recovery response to application of 100 �M Mg2+ (indicated by the red bar) during stationary activation of current (kon = 5000 mM/sec; koff = 37.5 sec-1; kd = 3 sec-1). C, Results for the STB model. D, ATB model predictions. The inset shows a comparison of the unblock portion of the two traces, indicating that the ATB recovery is somewhat slowed. E, F, Predictions for the protocol used by Spruston et al. (1995) (compare directly with their Fig. 13C) after a synaptic-like pulse of 1 mM glutamate. Black traces indicate results for constant depolarization to +40 mV, and red traces show results for intermittent depolarization (5 msec) from ?80 to +40 mV, with the timing indicated above. Response to a longer depolarization is shown in gray. kd = 2 sec-1. E, Predictions for the STB mechanism. F, Prediction for the ATB mechanism. For all ATB simulations, ?� was set to 270 sec-1, n = 300, and Erev = ?5 mV. This figure, together with a comparison to the figures cited from the literature, shows that the ATB model fits previously published results from other groups better than the STB model. In the case of Spruston et al.�s (1995) experiment (E, F), the difference between STB and ATB is particularly striking because of their voltage protocol in constant physiological magnesium concentration. In Benveniste and Mayer�s (1995) experiment (A, B), there is almost no difference in the response to the voltage step, because magnesium block is mostly removed by the time of the depolarization. In Sobolevsky and Yelshansky�s (2000) experiment (C, D), there is a difference in the speed of unblock (see inset in D), but it is small and its timing is comparable with their stated time constant of perfusion (30 msec). The appearance of the unblock in D seems consistent with their printed trace.
