Loss of PILRB gene at age-related macular degeneration associated PILRB/PILRA locus impairs photoreceptor function in mouse retina

Genome-wide association studies have uncovered mostly non-coding variants at over 60 age- related macular degeneration (AMD) susceptibility loci. To ascertain the target gene at the PILRB/PILRA locus, we used a CRISPR strategy to produce germline deletions in the mouse paired immunoglobin-like type 2 receptor (Pilr) genes that encode highly related activating (PILRB) and inhibitory (PILRA) receptors. We show that a combined loss of Pilrb1 and Pilrb2, but not Pilra, leads to an early but relatively stationary defect in photoreceptor function without any apparent change in other retinal cells. PILRB immunostaining is specifically detected at the proximal part of photoreceptor outer segment. Reduced expression of select calcium-regulated phototransduction and synapse-associated proteins, including GCAP1 and 2, PDE6B, AIPL1, PSD95, and CTBP1, indicate dysregulation of calcium homeostasis as a possible mechanism of retinal phenotype in Pilrb1/2 -/- mice. Our studies suggest an association of PILRB, and not PILRA, with AMD pathogenesis. Pilra, and Pilrb1/2 genes were deleted in the mouse genome by CRISPR-mediated editing in C57BL/6J zygotes. F0 founder carrying the deleted region was crossed with C57BL/6J mice. F1 founders carrying the deleted region were crossed to C57BL/6J mice for at least two additional generations before heterozygous/homozygous progeny were used in the following experiments. RNA from Pilrb1/2+/+ and Pilrb1/2 -/- mice retinas were isolated using RNeasy Mini Kit (Qiagen, Valencia, CA), with on-column DNA digestion, following the manufacturer's instructions. RNA quality and quantity were evaluated using Agilent 4200 TapeStation (Agilent Technologies, Santa Clara, CA). RNA-seq data was generated using TruSeq Stranded mRNA Sample Prep Kit (Illumina, San Diego, CA), as previously described, and 125 base pair-end reads were generated on NextSeq 2000 platform (Illumina, San Diego, CA). The alignment and quantification pipeline used human reference genome GRCh38.p7 and Ensembl v102 for annotation. Transcript-level counts were performed by kallisto v0.45.0 package and summarized to gene level using tximport v1.30.0. Gene level counts were converted to count per million (CPM) and then TMM normalized using the edgeR v.4.0.3 package in R. Please note that M12* are samples taken from 12 month old mice and P28* are samples taken from 28 day old mice (28 days post-natal).