Abstract
Purpose:
The Cngb1 locus encoded β-subunit of rod cGMP-gated cation channel and associated glutamic acid rich proteins (GARP1, GARP2) are necessary for phototransduction, disk morphogenesis, and rod structural integrity (Zhang et al. 2009, J Cell Sci). To further dissect individual protein structure/function of the GARPs in rods, we have characterized several new transgenic mouse models that selectively restore the missing proteins.
Methods:
Transgenic mice were made that express GARP1 (G1, 100 fold overexpressed) or GARP2 (G2) or both (G1G2) on an exon 1/promoter knockout (X1 KO) background. A linear mixed model was used to assess differences within each group over time and between groups for both outer nuclear layer (ONL) thickness and full retinal thickness (FRT). An interaction term was added to assess if changes over time differed by group. Established procedures for SD-OCT, and epoxy processed light level histology were used to analyze the genotypes at 3 weeks, 10 weeks.
Results:
Consistent with progressive degeneration, OCT analysis showed significant reduction in ONL thickness of X1 KO (p<0.01), G1 (p<0.001), G2 (p<0.0001), and G1G2 (p<0.0001) mice at 10 weeks compared to 3 weeks. There was also reduction in FRT of G2 (p<0.001) and G1G2 (p<0.001), but not G1 (p>0.05) mice at 10 weeks compared to 3 weeks. G2 ONL thickness reduced faster than X1 KO (p<0.001), G1 (p<0.0001), and G1G2 ONL (p<0.05). G1G2 ONL reduced faster than X1 KO (p<0.05) and G1 (p<0.05). X1 KO, G2, and G1G2 FRT reduced faster than G1 FRT (p<0.05; p<0.0001). Using light microscopy, at 3 weeks 10-12 rows of ONL nuclei were observed in all genotypes. At 10 weeks 3 rows of nuclei were observed in G2, 6-7 rows in G1G2, and 7-8 rows in G1 and X1 KO confirming an ongoing retinal degeneration. Photoreceptors are less uniform and more disorganized in appearance and nuclei appear pyknotic in G1G2 and are almost completely lost in G2, but G1 nuclei and outer segments appear more like WT.
Conclusions:
The rods of mice expressing G2 on an X1 KO background degenerates the most rapidly of all the genotypes. G1 overexpression on X1 KO also shows degeneration, but is much slower than G2. Coexpression of G1 and G2 slows degeneration compared to G2 alone, but degenerates faster than G1. These results indicate that GARP1 and GARP2 are not interchangeable and thus, have different roles in the photoreceptor.