Differential effects of antipsychotic drugs on contrast response functions of retinal ganglion cells in wild-type Sprague-Dawley rats and P23H retinitis pigmentosa rats

Abstract

Antipsychotic drugs haloperidol and clozapine have been reported to increase the sensitivity of retinal ganglion cells (RGCs) to flashes of light in the P23H rat model of retinitis pigmentosa. In order to better understand the effects of these antipsychotic drugs on the visual responses of P23H rat RGCs, I examined the responses of RGCs to a drifting sinusoidal grating of various contrasts. In-vitro multielectrode array recordings were made from P23H rat RGCs and healthy Sprague-Dawley (SD) rat RGCs. Retinas were stimulated with a drifting sinusoidal grating with eight values of contrast (0, 4, 6, 8.5, 13, 26, 51, and 83%). Contrast response functions based on response amplitudes were fitted with a hyperbolic ratio function and contrast thresholds were determined from the fitted curves. SD rat RGCs were divided into two categories, saturating and non-saturating cells, based on whether they showed saturation of responses at high contrast levels. Most SD rat RGCs (58%) were saturating cells. Haloperidol and clozapine decreased the responses of saturating SD rat RGCs to all grating contrasts, except for the highest contrast tested. Clozapine also decreased the responses of non-saturating SD rat RGCs to all grating contrasts, except for the highest contrast tested. Haloperidol did not however significantly affect the responses of non-saturating SD rat RGCs. Haloperidol and clozapine increased the contrast thresholds of both saturating and non-saturating cells in SD rat retinas. Most (73%) P23H rat RGCs could be categorized as either saturating or non-saturating cells. The remaining 'uncategorized' cells were poorly responsive to the drifting grating and were analyzed separately. Haloperidol and clozapine increased the responses of non-saturating and uncategorized P23H rat RGCs to most grating contrasts, including the highest contrast tested. Haloperidol and clozapine also increased the responses of saturating P23H rat RGCs to most grating contrasts but these increases were not statistically significant. Haloperidol and clozapine decreased the contrast thresholds of saturating cells, non-saturating cells and uncategorized cells in P23H rat retinas, although the decrease in contrast thresholds of saturating cells was not found to be statistically significant. Overall, the findings show that haloperidol and clozapine have differential effects on the contrast response functions of SD and P23H rat RGCs. In contrast to the effects observed on SD rat RGCs, both haloperidol and clozapine increased the responsiveness of P23H rat RGCs to both low and high contrast visual stimuli and decreased contrast thresholds.

Fig 1. MEA recordings from two RGCs.

Upper traces in (A) and (B) show the responses of two RGCs to a full-field flash (500 ms) of light. Cell recorded by electrode 58 was an OFF cell; cell recorded by electrode 18 was an ON cell. Lower traces in (A) and (B) show the responses of the two RGCs to the drifting sinusoidal grating. Whereas the cell recorded by electrode 58 clearly responded to the drifting grating, no response was elicited from the cell recorded by electrode 18.

Fig 2. Effects of haloperidol on responses of SD rat RGCs to drifting sinusoidal grating of various contrasts.

(A1 –A2) RGC contrast response functions. (A1) Contrast response function from saturating RGCs (n = 39) before and after application of haloperidol. (A2) Contrast response function from non-saturating RGCs (n = 25) before and after application of haloperidol. Data points in (A1) and (A2) are the mean ± SEM (errors smaller than the symbol size are not visible). ‡ P < 0.001 (Holm-Bonferroni multiple correction). (B1 –B2) RGC contrast thresholds. (B1) Contrast thresholds for saturating RGCs (n = 39) before and after application of haloperidol. (B2) Contrast thresholds for non-saturating RGCs (n = 25) before and after application of haloperidol. In (B1) and (B2), boxes represent the interquartile range (IQR) between first and third quartiles and the line inside represents the median. Whiskers denote the lowest and highest values within 1.5 x IQR from the first and third quartiles. Circles represent all data points. † P < 0.01, ‡ P < 0.001 (Wilcoxon signed-rank test).

Fig 3. Effects of clozapine on responses of SD rat RGCs to drifting sinusoidal grating of various contrasts.

(A1 –A2) RGC contrast response functions. (A1) Contrast response function from saturating RGCs (n = 33) before and after application of clozapine. (A2) Contrast response function from non-saturating RGCs (n = 27) before and after application of clozapine. Data points in (A1) and (A2) are the mean ± SEM. * P < 0.05, ‡ P < 0.001 (Holm-Bonferroni multiple correction). (B1 –B2) RGC contrast thresholds. (B1) Contrast thresholds for saturating RGCs (n = 33) before and after application of clozapine. (B2) Contrast thresholds for non-saturating RGCs (n = 27) before and after application of clozapine. In (B1) and (B2), boxes represent the interquartile range (IQR) between first and third quartiles and the line inside represents the median. Whiskers denote the lowest and highest values within 1.5 x IQR from the first and third quartiles. Circles represent all data points. Note the contrast threshold value for one cell in (B2) was immeasurable (i.e., exceeded 83%). ‡ P < 0.001 (Wilcoxon signed-rank test).

Fig 4. Effects of haloperidol on responses of P23H rat RGCs to drifting sinusoidal grating of various contrasts.

(A1 –A3) RGC contrast response functions. (A1) Contrast response function from saturating RGCs (n = 5) before and after application of haloperidol. (A2) Contrast response function from non-saturating RGCs (n = 50) before and after application of haloperidol. (A3) Contrast response function from uncategorized RGCs (n = 22) before and after application of haloperidol. Data points in (A1), (A2) and (A3) are the mean ± SEM. * P < 0.05, ‡ P < 0.001 (Holm-Bonferroni multiple correction). Note that the control curve in (A3) is a bad fit, giving the impression that the cells were ‘saturating’ cells. This bad fit is because only two data points have values greater than zero. (B1 –B3) RGC contrast thresholds. (B1) Contrast thresholds for saturating RGCs (n = 5) before and after application of haloperidol. (B2) Contrast thresholds for non-saturating RGCs (n = 50) before and after application of haloperidol. (B3) Contrast thresholds for uncategorized RGCs (n = 22) before and after application of haloperidol. In (B1), (B2) and (B3), boxes represent the interquartile range (IQR) between first and third quartiles and the line inside represents the median. Whiskers denote the lowest and highest values within 1.5 x IQR from the first and third quartiles. Circles represent all data points. ‡ P < 0.001 (Wilcoxon signed-rank test). Note the contrast threshold values for one cell in (B1) and ten cells in (B3) were immeasurable (i.e., exceeded 83%).

Fig 5. Effects of clozapine on responses of P23H rat RGCs to drifting sinusoidal grating of various contrasts.

(A1 –A3) RGC contrast response functions. (A1) Contrast response function from saturating RGCs (n = 5) before and after application of clozapine. (A2) Contrast response function from non-saturating RGCs (n = 58) before and after application of clozapine. (A3) Contrast response function from uncategorized RGCs (n = 21) before and after application of clozapine. Data points in (A1), (A2) and (A3) are the mean ± SEM. * P < 0.05, † P < 0.01, ‡ P < 0.001 (Holm-Bonferroni multiple correction). Note that the control curve in (A3) is a bad fit, giving the impression that the cells were ‘saturating’ cells. This bad fit is because only two data points have values greater than zero. (B1 –B3) RGC contrast thresholds. (B1) Contrast thresholds for saturating RGCs (n = 5) before and after application of clozapine. (B2) Contrast thresholds for non-saturating RGCs (n = 58) before and after application of clozapine. (B3) Contrast thresholds for uncategorized RGCs (n = 21) before and after application of clozapine. In (B1), (B2) and (B3), boxes represent the interquartile range (IQR) between first and third quartiles and the line inside represents the median. Whiskers denote the lowest and highest values within 1.5 x IQR from the first and third quartiles. Circles represent all data points. ‡ P < 0.001 (Wilcoxon signed-rank test). Note the contrast threshold values for one cell in (B2) and nine cells in (B3) were immeasurable (i.e., exceeded 83%).