Kuznetsov K. Modulation of tonic firing and intracellular calcium dynamics by Kv3 channels in the rat retinal ganglion cells

Thesis is devoted to investigation of calcium influx dynamics and sustained firing modulation by voltage-gated Kv3 potassium channels in the rat retinal ganglion cells (RGC). Expression of non-inactivating Kv3.1/Kv3.2 potassium channels determines fast- spiking phenotype of many types of neurons including retinal ganglion cells (RGCs); furthermore Kv3 channels regulate neurotransmitter release from presynaptic terminals. Experiments were performed on the whole-mount retinal preparations from 4 to 6 weeks old Wistar rats using simultaneous whole cell patch clamp and intracellular Ca2+ measurements in combination with single-cell RT-PCR. In response to 500-ms depolarization step the RGCs demonstrated fast firing tonic behavior with a mean frequency of spiking 61 ± 5 Hz (n = 68). All of the tonic cells tested (n = 9) expressed specific mRNA for either Kv3.1 or Kv3.2 or for both channels. Bath applications of 4-AP (200 µM) and TEA (250 µM, 500 µM and 1 mM) modified firing patterns dose- dependently as follows: firing frequency was decreased, mean action potential (AP) half- width increased and mean amplitude of afterhyperpolarization was reduced. Neither bath applications of alfa-DTX (100 nM) nor of CdCl2 (200 µM) significantly affected firing properties of RGCs. Bath applications of 4-AP (200 µM) and TEA (500 µM) reduced potassium current amplitude of the RGCs about 25%. The residual currents were similar to Kv3.1/Kv3.2 potassium currents that were described in literature previously. Voltage-gated calcium currents isolation in the RGCs revealed LVA calcium currents in the 10 of 22 tested cells (45%). Activation of the LVA calcium channels didn't lead to changes in the intracellular calcium concentrations. It indicates that LVA calcium channels are weakly expressed in rat RGC somas because we recorded ion currents from whole cell whereas calcium signals only from somas. Amplitude of calcium signals was depended on HVA calcium current duration in linear manner with slope of 2.76±0.76 nM/ms. Such dependence can suggests about the absence of calcium-induced calcium release from intracellular stores in rat RGCs Bath application of CdCl2(200 µM) blocked completely either calcium currents or calcium signal. The amplitude of the Ca2+ signals induced by the cells firing was linearly dependent on number of APs with a mean slope of 7.3 ± 0.9 nM per one AP (n = 8). APs widening by TEA increased the slope of the amplitude vs. AP number plots in a dose-dependent manner: 250 µM of TEA increased the mean slope value to 9.5 ± 1.2 nM/AP (n = 6), 500 µM to 12.4 ± 2.4 nM/AP (n = 6) and 1 mM to 13.2 ± 2.9 nM/AP (n = 6). All these parameters, as well as the cells firing properties, were significantly different from controls and from each other except between 500 µM and 1 mM. This is consistent with the pharmacological properties of Kv3.1/Kv3.2 channels: the TEA IC50 is in the range 150- 300 µM with almost complete block at 1 mM. This suggests that Kv3.1/Kv3.2 channels modulate calcium influx during the fast tonic firing of the rat RGCs and prevent the substantial growth of intracellular calcium concentration at a given firing frequency, which may have an effect of cytotoxicity.