Abstract
Purpose:
To investigate the dependence of the pattern electroretinogram (PERG) on spatial contrast in mice of different strains without or with retinal ganglion cell (RGC) pathology.
Methods:
PERGs were recorded in 24 mice under ketamine/xylazine anesthesia in response to gratings of 0.05 cy/deg with different contrasts (98, 75, 50, and 25%) reversing 2 times/s. Four groups of 6 mice each were tested, D2_1: DBA/2J 2-4 months old; D2_2: DBA/2J 7-8 months old; B6_1: C57BL/6J; B6_2: C57BL/6J with lesion of the contralateral superior colliculus 1 months before. PERG amplitude was measured between the main positive wave (P1) and the following negative wave (N2). PERG latency was the time-to-peak of the P1 wave.
Results:
In all mice, PERG signals above the noise level were recordable at all contrasts. With increasing contrast, the PERG amplitude increased, while the PERG latency decreased. Latency changes were approximately linear and were expressed by a linear regression slope (ms/100% contrast). Amplitude changes were non-linear and were expressed by a Contrast Saturation Index CSI = (100-C½)/C½, where C½ is the contrast that elicits ½ amplitude of the PERG at 98% contrast. CSI >1 indicate compressive non-linearity; CSI<1 indicate expansive non-linearity. Mean amplitude CSI (SEM) were: D2_1: 0.42 (0.1), D2_2: 1.71 (0.82), diff. P<0.05; B6_1: 0.74 (0.24), B6_2: 4.8 (0.61), diff. P<0.001. Mean latency slopes (SEM) were: D2_1: -35.2 (7.3), D2_2: -60 (11), diff. P<0.05; B6_1: -60.2 (5.3), B6_2: -39.6 (7.8), diff. P<0.05. Latency slopes of D2_1 and B6_1 were different (P<0.01).
Conclusions:
The mouse PERG exhibits strong gain control mechanisms demonstrated by latency shortening with increasing contrast associated with non-linear amplitude characteristics. Both D2_1 and B6_1 control strains display expansive response non-linearity, however the latency slope is steeper in B6_1. Both D2_2 and B6_2 mice are known to have abnormal PERG induced by glaucoma and deficit of target-derived factors respectively, however without measurable RGC death. In both strains, pathology converted expansive response non-linearity to compressive, whereas latency slope became steeper in D2_2 and shallower in B6_2. Integrative processing of spatial contrast in the inner retina, strain differences, and pathology-induced adaptive changes of function can be demonstrated with PERG.