To determine whether the differentiating effects of SS treatment also involves changes in the electrophysiological properties of RGC-5 cells, we examined membrane currents using whole-cell voltage-clamp recording. Untreated cells had an input resistance of 55 ± 10 MΩ (mean ± SEM), an apparent capacitance of 28 ± 2 pF, and a resting potential of −34 ± 0.5 mV (
n = 4). After subtracting passive components, a small residual outward current was present at depolarized potentials, yielding a maximum conductance of 3.1 ± 0.5 nS at +90 mV
(Fig. 4A) . Three SS treatment protocols were tested (given as hours in SS with duration of the following incubation in medium alone in parentheses): 1 hour (24 hours), 1 hour (48 hours), and 22 hours (2 hours). SS induced some changes in passive properties, as judged by ANOVA and Dunnett’s post hoc test. After treatment, input resistances were 397 ± 171 MΩ (1 hour [24 hours],
n = 5, not significant, NS), 322 ± 89 MΩ (1 hour [48 hours],
n = 4, NS), and 843 ± 201 (22 hours [2 hours];
n = 3,
P < 0.05), apparent capacitances were 31 ± 5 pF (1 hour [24 hours], NS), 44 ± 3 pF (1 hour [48 hours],
P < 0.05), and 38 ± 3 (22 hours [2 hours]; NS), and resting potentials were −27 ± 9 mV (1 hour [24 hours],
n = 5, NS), −9 ± 6 mV (1 hour [48 hours], NS), and −11 ± 3 (22 hours [2 hours]; NS). The larger capacitance in some cells is consistent with the elaboration of neuronlike processes. In addition, although not significant by ANOVA, SS induced increases in voltage-gated conductance in all groups of treated cells. After leak subtraction, outward currents had conductances at +90 mV of 14 ± 5 nS (1 hour [24 hours], NS), 7 ± 3 nS (1 hour [48 hours], NS), and 5 ± 3 nS (22 hours [2 hours], NS). Inspection of the spread in the data
(Fig. 4B)reveals that some treated cells had small conductance similar to untreated cells, whereas others had much larger conductance. This bimodal distribution probably accounts for the inability to detect significant differences. We therefore further examined cells with conductances greater than the mean of the untreated group plus twice its SD. The probability of finding such values should be extremely low if there were truly no differences between groups. However, half of the treated cells had conductances in this range, suggesting that SS induces the expression of voltage-gated current in a large fraction of cells. In these cells, the current increased monotonically as a function of voltage
(Fig. 4C)and could be fit with a Boltzmann function in the most robust treatment group (i.e., 1 hour [24 hours]), yielding a maximum conductance of 26 nS and an apparent half-activation voltage of +48 mV. Because treated cells displayed neuronlike processes, we cannot be assured of there being an adequate space clamp. Thus, the strongly positive activation range and shallow slope factor we observed may not accurately reflect the biophysical characteristics of the expressed channels. Nonetheless, it seems clear that SS altered the electrical properties of a subset of treated cells.