Novel form of adaptation in mouse retinal rods speeds recovery of phototransduction

Abstract

Photoreceptors of the retina adapt to ambient light in a manner that allows them to detect changes in illumination over an enormous range of intensities. We have discovered a novel form of adaptation in mouse rods that persists long after the light has been extinguished and the rod's circulating dark current has returned. Electrophysiological recordings from individual rods showed that the time that a bright flash response remained in saturation was significantly shorter if the rod had been previously exposed to bright light. This persistent adaptation did not decrease the rate of rise of the response and therefore cannot be attributed to a decrease in the gain of transduction. Instead, this adaptation was accompanied by a marked speeding of the recovery of the response, suggesting that the step that rate-limits recovery had been accelerated. Experiments on knockout rods in which the identity of the rate-limiting step is known suggest that this adaptive acceleration results from a speeding of G protein/effector deactivation.

Figure 1.

Previous light exposure causes shortening of saturating responses in darkness. (A) Suction electrode recording of a wild-type mouse rod. Test flashes (ticks in light monitor, lower trace; 2,180 photons/μm²) were delivered to the rod before and after 3 min of steady, saturating light (5,750 photons/μm² s). (B) Responses from a representative cell before (solid trace), immediately after (dashed trace), and several minutes after (solid trace) the adapting light (7,055 photons/μm² s). The initial and final traces superimpose. Dark currents of this cell (in pA) for the initial, post, and final traces were 12.8, 11.7, and 13.0, respectively. Flash strength was 2,076 photons/μm². (C) Time spent in saturation (T_sat) for individual cells before (initial), immediately after (post) and several minutes after (final) three minutes light exposure. Intensities of background light ranged from 2,896–9,672 photons/μm² s, which just saturated the cells.

Figure 2.

Time in saturation of a bright flash response (2,180 photons/μm²) changed over time following the adapting light. Time 0 represents the time the current returned to baseline following adapting light offset. Adapting light (5,750 photons/μm² s) was applied for 3 min.

Figure 3.

Two mechanisms could account for response shortening. (Top) A decrease in transduction gain would reduce light-activated PDE activity and produce a response that remained in saturation for a shorter time than the initial response (thick trace). The maximal response amplitude is attained when all of the channels in the outer segment are closed (represented by dashed line), and beyond this point, differences in PDE activity result in different saturation times. (Bottom) In the second possible mechanism, the amplitude of the light-activated PDE activity is unaffected by adaptation (thin trace), but recovers faster, resulting in a response with a shorter saturation time.

Figure 4.

Adaptation produced no change in the gain of transuction. (A) Average responses (each to five flashes) of a representative cell before (initial, solid trace) and shortly after (post, dashed trace) the adapting light. (B) Expanded time scale, showing rising phases of responses in A. Closed squares, initial; open squares, post. Flash strength was 56.0 photons/μm². The adapting light (7,016 photons/μm² s) was applied for 3 min. Error bars reflect SEM.

Figure 5.

Response saturation times increased with increasing flash strengths. A family of saturating responses from a representative wild-type cell to flashes ranging in strength from 519 photons/μm² to 74,968 photons/μm², in roughly twofold increments. The form of the response is invariant across flash strengths, as described previously (Pepperberg et al., 1992). At flash strengths between 5,000 and 10,000 photons/μm², the form of the response changed (arrow; in this cell 5,630 photons/μm²). This is also the flash strength at which the Tsat relation typically deviates from linearity (see text). The time in saturation was measured from the midpoint of the time of the flash and the time at which the response had recovered to 90% of its maximal amplitude. Traces represent the average of two responses to a particular flash strength. Dark current for this cell was 17 pA.

Figure 6.

Adaptation shortens the dominant time constant of recovery. (A) Average dominant time constant of recovery (τ_D) of 14 rods, before (initial) and immediately after (post) several minutes of the adapting light. Each point is the average of data points from 8–14 cells. Error bars represent SEM. (B) Average dominant time constant of recovery (τ_D) from individual wild-type rods, before (initial) and immediately after (post) adapting light. Adapting light intensities ranged from 2,896–9,672 photons/μm² s, which was sufficiently bright to just saturate the cell.

Figure 7.

Long-lasting adaptation in RGS9 knockout rods. (A) Responses from a representative RGS9 knockout rod before (black), immediately after (gray), and several minutes after (black) the adapting light. In this case, the final trace was slightly faster than the initial trace. Dark current (in pA) was 11.7, 12.7, and 11.3, respectively. Flash strength was 647.6 photons/μm². Adapting light (286.4 photons/μm² s) was applied for 3 min. (B) Time spent in saturation for individual cells before (initial), immediately after (post), and several minutes after (final) a 3-min light exposure. Intensities ranged from 286 to 517 photons/μm² s, which was sufficient to just saturate the cell. (C) Time spent in saturation of each response as a function of the time at which the flash was given. Time 0 is the time at which the current returned to baseline following background light removal.

Figure 8.

Adaptation speeds the rate-limiting recovery step in RGS9 knockout rods. (A) Dominant time constant of recovery (τ_D) of a representative RGS9 knockout rod before (initial, closed squares) and immediately after (post, open squares) several minutes after background light. Adapting light intensity was 286 photons/μm² s and was applied for 3 min. (B) Average τ_D values before (initial; n = 4), immediately after (post; n = 4), and several minutes after (final; n = 3). Error bars represent SEM.