Identification of three RGC types in adult RD retina. Mean age (± SD) P44 ± 4. A. Cross sections of fixed tissue show gross structural differences of WT and RD retinas. Note the lack of OPL, ONL and OS in RD retina. GCL, ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; OS, photoreceptor outer segments. B. Two-photon z-projections (above, larger images) and side-projections (below) of three identified types of WT (left) and RD (right) RGCs. Cells were filled with 125 μM OGB-1 during patch-clamp recordings. C. Dendritic stratification within the IPL for populations of ON (n=7 WT, 9 RD), OFF T (n=12 WT, 12 RD), and OFF S (n=5 WT, 6 RD) RGCs from WT (filled symbols) and RD (open symbols) retina. See text for statistical comparisons. WT data in C from (Margolis and Detwiler, 2007). Scale bars are 50 μm.

Rhythmic resting synaptic currents in identified RD RGCs. A. Whole-cell voltage clamp recordings of resting synaptic currents in ON, OFF T and OFF S RGCs from WT (left, grey) and RD (right, black) retinas. Vhold=−70 mV. B. Currents on an expanded time scale. Note strong periodicity in RD currents from all three cell types and different dynamics of currents in ON and OFF cells. C. Power spectra of RD (black) and WT (gray) currents (same cells as in B). The multiple peaks in RD spectra reflect harmonics of the fundamental ‘beating’ frequency around 10 Hz.

Distinct balance of excitatory and inhibitory synaptic currents in ON and OFF RD RGCs. A. Comparison of excitatory (Vhold=−70 mV) (black, lower traces) and inhibitory (Vhold=0 mV) (grey, upper traces) synaptic currents from individual cells. B. Power spectra of excitatory (solid line) and inhibitory (dotted line) currents from the cells in A. C. (top) Correlation between frequency of inhibitory and excitatory currents from individual ON, OFF T and OFF S cells (n=12). Frequency measured from power spectra as in Fig. 4. (bottom) Ratio of inhibitory to excitatory power for ON, OFF T and OFF S cells. Note that ratio is low for ON cells and high for OFF cells, reflecting a distinct balance of excitatory and inhibitory synaptic currents. Recordings made with Cs-MeSO4 internal solution (see Materials and Methods).

Similarity of spike- and rebound-evoked Ca2+ signals in WT and RD RGCs. A. Ca2+ signals (Ca2+) recorded from proximal dendrites during voltage responses (Vm) to injected current (Iinj) in ON WT (top) and RD (bottom) RGCs. B. Same as A for OFF T RGCs. Recordings made in whole-cell current clamp in the presence of synaptic blockers (in μM: 50 L-APB, 20 CNQX, 50 APV, 1 strychnine, and 50 picrotoxin). C. Mean Ca2+ signal amplitude per spike (ΔF/F/# spikes) for ON (top; n=4,3) and OFF T (bottom; n=4[5],3) RGCs. Note that WT and RD cells display similar Ca2+ signaling profiles for both spiking and rebound firing.

Rhythmic resting spike activity in identified RD RGCs. A. Whole-cell current clamp recordings from ON, OFF T and OFF S RGCs in RD retina. B. Activity on an expanded time scale. Note underlying oscillatory membrane depolarizations present in RD (left, black) but not WT (right, grey) cells. C. Inter-spike interval histograms from cells in B. RD solid line, WT dotted line. Insets show power spectra for RD RGCs (same cells in B). Note clear presence of peak frequencies at 12–14 Hz in these cells.

Correlation between spiking and synaptic currents in individual RD RGCs. A. Whole-cell voltage clamp (top; Vhold=−70 mV) and current clamp (bottom) recordings from individual ON (above) and OFF T (below) cells. B. Power spectra of current (dotted line) and voltage (solid line) from the cells in A. Inset shows Gaussian fits to spectra in the range of 3–20 Hz. C. Correlation of peak frequencies of spiking and synaptic currents in individual cells (n=18), as measured from power spectra.

Similarity of intrinsic electrical properties in WT and RD RGCs. A. Voltage (Vm) responses to injected current (Iinj) for ON (top) and OFF T (bottom) cells recorded in whole-cell current clamp mode in the presence of synaptic blockers (in μM: 50 L-APB, 20 CNQX, 50 APV, 1 strychnine, and 50 picrotoxin). Arrows indicate presence of rebound firing in both WT and RD OFF T cells. B. Four superimposed responses to negative current injection. Strength of responses is proportional to the level of hyperpolarization. C. Average maximum firing frequency and ADP strength for ON (left; n=6,3) and OFF T cells (right; n=5,4).

Calcium dynamics during resting activity in RD RGCs. A. Simultaneously recorded calcium indicator fluorescence (Ca2+, red) and membrane potential (Vm, black) from cells show at left. Red boxes on proximal dendrites indicate scanned regions. B. Ca2+ signals (red) and spike rate (black; bin width, 150 ms) during resting activity (injected current, Iinj=0) from two different cells than in A. Note the longer time scale. In ON cell (left), Iinj reduced spike rate and resting Ca2+. Baseline membrane potentials for ON cell at rest (during pulse) were −71, −70 (−94), −70 (−108) mV for top, middle and bottom panels, respectively. Baseline membrane potentials for OFF T cell at rest (stim) were −70, −70 (−87), −70 (−95) mV for top, middle and bottom panels, respectively. C. Blow-ups of regions corresponding to bars in B (middle panel at ~t=9 s for ON cell; bottom panel at ~t=7 s for OFF T cell). Note that both spikes and subthreshold regenerative-like events are associated with Ca2+ signals during hyperpolarization in OFF T cell (arrows). Plot at bottom right shoes rebound-evoked Ca2+ signal during resting activity (Iinj=−400 pA for 0.5 s). Superimposed grey trace is the mean of 3 responses to −550, −400 and −300 pA. Hyperpolarizing portion of mean response is shown in inset at right to emphasize Ca2+ decrease; scale bar is 10% ΔF/F.