Skip to main content
. 2014 Jan 15;136(6):2529–2537. doi: 10.1021/ja411338t

Figure 5.

Figure 5

Kinetics of bright-state formation during the write pulse for (a,b) Arch(D95H) and (c,d) Arch(D95Q). Voltage was held at either +100 or −100 mV during the write interval and at −100 mV during the dark and read intervals. The length of the write flash, twrite, was varied, keeping its end coincident with the step in voltage from +100 mV to −100 mV. Representative fluorescence traces are shown for (a) Arch(D95H) and (c) Arch(D95Q). (b,d) Plot of memory effect, ΔM, as a function of twrite. In Arch(D95H) the rise in memory was fit by a double exponential with τfast = 0.14 ms (57%) and τslow = 12 ms (43%); a write flash with twrite = 1 ms was sufficient to elicit more than half of the maximal response. (d) The dependence of ΔM on twrite in Arch(D95Q) was dominated by a slow component. A fit to a double exponential yielded τfast = 5 ms (8%) and τslow = 180 ms (92%).