Ambient light regulates sodium channel activity to dynamically control retinal signaling

Abstract

The retinal network increases its sensitivity in low-light conditions to detect small visual inputs and decreases its sensitivity in bright-light conditions to prevent saturation. However, the cellular mechanisms that adjust visual signaling in the retinal network are not known. Here, we show that voltage-gated sodium channels in bipolar cells dynamically control retinal light sensitivity. In dim conditions, sodium channels amplified light-evoked synaptic responses mediated by cone pathways. Conversely, in bright conditions, sodium channels were inactivated by dopamine released from amacrine cells, and they did not amplify synaptic inputs, minimizing signal saturation. Our findings demonstrate that bipolar cell sodium channels mediate light adaptation by controlling retinal signaling gain.

Figure 1.

Schematic of retinal neural network and sites of adaptation. Voltage-gated sodium channels in bipolar cells contribute to network adaptation in dim (left) but not in bright (right) light conditions. C, Cone photoreceptors; R, rod photoreceptors; B, bipolar cells; G, ganglion cells; H, horizontal cells; A, amacrine cells.

Figure 2.

Voltage-gated sodium channels amplified light-evoked synaptic responses only in dim-light conditions. A, B, L-EPSPs recorded from ON bipolar cells in dim conditions (A) were reduced by TTX, but those recorded in bright conditions (B) were not affected by TTX. C, TTX shifted and compressed the intensity–response relationship in dim (p < 0.01; n = 5) but not in bright conditions (p > 0.1; n = 5). D, The intensity–response relationship in ON sustained bipolar cells, which do not possess sodium channels, were shifted to the right compared with the curve in transient bipolar cells in dim-light conditions (curves for the transient cells are the same as C). The shift of the sustained bipolar cell curves from dim to bright light is attributable to cone and OPL adaptation.

Figure 3.

Bright background illumination attenuated bipolar cell sodium currents by changing the channel voltage dependence. A, TTX-sensitive sodium currents recorded from bipolar cells at different background illuminations were obtained by subtracting the current responses to voltage steps in TTX from those obtained in control conditions. Each record was obtained from a different cell. B, C, The peak amplitude (B) and the charge transfer (measured at 0–50 ms; C) of sodium currents were decreased with background light in an intensity-dependent manner (p < 0.01; no background, n = 7; −4 log, n = 2; −2 log, n = 3; 0 log, n = 4). D, Sodium channel inactivation curves in dim (solid circles) and bright (open circles) conditions. Bright light shifted the curve to the left (half-maximum inactivation voltage, bright, −51.6 mV, n = 6; dim, −36.4 mV, n = 5; p < 0.05). E, Conductances in each condition was normalized to the maximum dim condition value and plotted. Bright light (open triangles) also shifted and compressed the activation curves for sodium currents in bipolar cells compared with dim light (solid triangles; half-maximum activation voltage, bright, −2.95 ± 4.7 mV; dim, −19.2 ± 5.9 mV; p < 0.05). Dotted line represents scaled bright-light curve.

Figure 4.

Dopamine D₁ receptors mediated the effects of light adaptation on voltage-gated sodium currents. A, A dopamine agonist, ADTN (20 μm), reduced the sodium currents in dim conditions (n = 5) compared with control current in dim light (dotted line, same as Fig. 3A, top). B, A D₁ receptor blocker, SCH23390 (SCH; 50 μm), blocked the ADTN effect (n = 3). The dotted line is control current in dim light from Figure 3A (top). C, Bright illumination did not reduce the sodium current when D₁ receptors were blocked (50 μm SCH23390; n = 3). The dotted line is the control current in bright conditions from Figure 3A (bottom). D, E, Effects of agonists and antagonists on peak amplitude (D) and charge transfer (E). *Significant differences from control (p < 0.05); ^#significant differences from ADTN (dim; p < 0.05).

Figure 5.

D₁ dopamine receptor agonist and antagonist mimic light and darkness, respectively. A, In dim conditions, a D₁ agonist, SKF38393 (20 μm), reduced the peak amplitude of L-EPSPs, mimicking bright conditions. The D₁ agonist occluded the TTX (0.5 μm) effect on peak amplitude. B, Graph summarizes effects of D₁ agonist and TTX on L-EPSP peak amplitude (n = 6). C, Cells were pretreated with D₁ receptor antagonist [50 μm SCH23390 (SCH)], and then the preparation was exposed to bright light for 10 min. TTX reduced the L-EPSP, indicating that the D₁ receptor mediated the suppressive effect of background illumination on sodium currents in bipolar cells. D, Summary of TTX effects on L-EPSPs in bright conditions, pretreated with D₁ antagonist (n = 4). *p < 0.01, different from control level.

Figure 6.

Bipolar cell (BC) sodium currents enhanced ganglion cell light sensitivity. A–C, Ganglion cell (GC) L-EPSCs (black lines) were reduced by TTX (red lines) in dim (B) but not bright (C) or scotopic (A) conditions. D, Intensity–response curves in dim light but not in scotopic or bright conditions were shifted (p < 0.05) and compressed (p < 0.01) by TTX, suggesting that BC sodium currents only boosted GC responses in dim conditions. E–H, The sensitivities for mean intensity (E) and contrast (F–H) were enhanced by bipolar cell sodium currents in dim but not scotopic or bright conditions [mean sensitivity, p < 0.01 (E); contrast sensitivity, p < 0.05 (F–H); scotopic, n = 6; dim, n = 11; bright, n = 11]. ctrl, Control.

Figure 7.

The enhancement of ganglion cell (GC) light sensitivity is decreased by increasing adaptation levels. The enhancement of ganglion cell sensitivity to mean illumination attributed to bipolar cell (BC) sodium channels was plotted as a function of background light (same dataset as in Fig. 6E). In scotopic conditions, light sensitivity in ganglion cells was not affected by bipolar cell sodium channels. In mesopic, dim-light-adapted conditions, light sensitivity was dramatically enhanced. The enhancement was decreased by increased ambient light levels and was not observed in photopic conditions. The decrease in light sensitivity was attributed to decreased bipolar cell sodium channel contributions and may be correlated with dopamine release by light adaptation (see Discussion).