Glycinergic Inhibition Targets Specific Off Cone Bipolar Cells in Primate Retina

Spontaneous inhibition is present in specific Off-CBC types. A, Spontaneous inhibitory events in representative FMB, DB1, DB2, DB3a, and DB3b cells. The stimulus was a series of depolarizing voltage steps from a holding potential of −70 mV (+5-mV increments, 16 steps). The red traces show currents at – 5 mV near the calculated excitatory reversal potential of −7 mV. B, Trend-subtracted currents during the step to –5 mV from the same cells shown in A. C, Bar graph showing average current variance in the different Off-CBCs during a voltage step to −5 mV. Number of cells for each type is indicated above bars. Data are mean ± SEM; *p < 0.01, **p < 0.01, ***p < 0.001 with Tukey’s post hoc tests after one-way ANOVA.

Spontaneous inhibition in FMB cells is glycinergic and arises at the axon terminals. A, Example traces from an FMB cell showing the frequency of sIPSCs before, during, and after bath application of STR (0.5 μm). Events were recorded during a voltage step to −20 mV from a holding potential of −70 mV. B, Effect of STR on sIPSC frequency for a group of FMB cells (Ctrl and STR; n = 7 cells, washout; n = 2 cells, **p = 0.004, Wilcoxon signed-rank test). Gray points indicate individual cells, black points indicate mean ± SEM. C, top panel, Example trace showing a lack of spontaneous events in an axotomized FMB cell during a voltage step to −20 mV from −70 mV. Bottom panel, Current response to a 20-ms puff application (red bar) of L-glutamate (1 mm) applied to the same cell. The presence of a glutamate-evoked inward current confirms the Off-type physiology of the axotomized midget cell. Holding potential is −70 mV.

Properties of sIPSCs in FMB and DB1 bipolar cells. Quantification of sIPSC events in FMB (A–E) and DB1 (F–J) bipolar cells. A, F, Example traces showing sIPSCs in an FMB (A) and DB1 (F) cell at the holding potentials indicated to the right of traces. B, G, Average sIPSC events extracted from the same bipolar cells shown in A, F using a matched template filter (average of 64–126 events per voltage for FMB, average of 70–130 events per voltage for DB1 cells). Gray shading shows SEM. Red traces show single exponential fits to event decay at −10 mV. C, H, I-V plots showing sIPSC amplitude as a function of voltage for a group of FMB (n = 20 cells; C) and DB1 cells (n = 7 cells; H). Note that events reverse close to the inhibitory reversal potential. D, I, Average sIPSC rise-time for the group of FMB (D) and DB1 (I) cells. E, J, Average sIPSC decay time constants for FMB (E) and DB1 (J) cells. All graphs show mean ± SEM.

Inhibitory input to Off-CBCs is driven by the On-pathway. A, Example traces showing sIPSCs in an FMB cell before, during, and after bath-application of L-AP4 (10 μm). B, Effect of L-AP4 on sIPSC frequency for a group of FMB cells (Ctrl and L-AP4; n = 4 cells, washout; n = 1 cell, *p = 0.028, Wilcoxon test). Gray points represent individual cells, black points show mean ± SEM. C, Example traces showing effect of puffing CPPG (600 μm) in the OPL in the presence of bath-applied L-AP4. CPPG puffs increase the frequency of sIPSCs in FMB, DB1, and DB3a cells but not in DB2 and DB3b cells. D, Plots showing current variance before and after puff application of CPPG in the presence of L-AP4 for FMB/DB1 cells (left panel) and DB2/DB3a/DB3b cells (right panel). Data are from n = 3 FMB (gray), n = 2 DB1 (blue), n = 3 DB2 (red), n = 2 DB3a (green), and n = 2 DB3b (magenta); *p < 0.05 Wilcoxon rank test. Current traces in A, C are at a step to 0 mV. E, Example traces showing reduced frequency of sIPSCs in an FMB cell after bath application of the AMPA receptor antagonist, GYKI-53655 (10 μm). Holding potential is −60 mV. F, Effect of GYKI on sIPSC frequency for a group of five FMB cells (gray points) and one DB1 cell (blue point). Partial washout was obtained in n = 2 FMB cells; *p = 0.016 with Wilcoxon rank test. Black points show mean ± SEM.

Voltage-gated currents in DB1 and DB2 bipolar cells. A–C, Average leak-subtracted voltage-gated currents in DB1 and DB2 cells. Voltage step protocols are shown in top panels, and resulting currents from DB1 and DB2 cells are shown in the middle and bottom panels, respectively. Timing scale bar in lower panels applies to both cell types. Note prominent A-type potassium current, T-type calcium current, and K_IR and I_h currents in DB2 cells that is absent in DB1 cells. Data are normalized to max amplitudes. Gray shading in A–C shows ±1 SEM. D, Comparison of peak outward potassium current in DB1 (open circles, n = 11) and DB2 (solid circles, n = 19) cells measured at a fixed time point 1.8 ms after step onset. Measurement time points are indicated with corresponding symbols in A. E, Comparison of inward current in DB1 (open circles, n = 4) and DB2 (solid circles, n = 13) cells at timepoints indicated by same symbols shown above traces in B (4 ms after step onset). F, Comparison of average instantaneous (open triangle) and time-dependent inward current (black triangles) in DB2 cells (n = 13) and time-dependent current in DB1 cells (red triangles, n = 9). Instantaneous current was measured at a fixed time point (5.7 ms after step onset), time-dependent current is the difference between the minimum current amplitude and instantaneous current. Data are mean ± SEM.

Voltage-gated ion channel subunits in primate DB1 and DB2 cells. A–F, Localization of the T-type calcium channel subunit, Ca_V3.1, in primate bipolar cells. Ca_V3.1 is absent from SCGN+ DB1 cells (A–C, arrows). D–F, A subset of GLT-1+ Off-bipolar cells express Ca_V3.1, consistent with expression in DB2 cells (arrows). G–L, The inward-rectifying potassium channel (K_IR) subunit, Kir2.1, is absent from DB1 cells (G–I, arrows) but present in the soma, dendrites, and axon terminal boutons of GLT-1+ DB2 cells (J–L). Scale bar: 20 μm.