RNA-seq profiling of RPE reprogramming

The plasticity of the retinal pigment epithelium (RPE) has been observed during proliferative vitreoretinopathy (PVR), a defective repair process in humans. In contrast, in the embryonic chick, the RPE can be efficiently reprogrammed to regenerate a complete neural retina after surgical removal and when supplied an exogenous source of FGF2. Here, we analyzed discrete RPE cell populations during early times of transiently reprogrammed (RPE 6 hours post-retinectomy) and reprogrammed (RPE 6 hours post-retinectomy and FGF2 treatment) cells, using laser capture microdissection followed by RNA sequencing (LCM-seq) and computational analysis. Overall design: RPE from chicken embryos were isolated via the Veritas Laser Capture Microdissection system in triplicate across three conditions: 1) intact embryonic day 4 (E4) RPE, 2) E4 RPE 6 hours post retinectomy (transiently reprogrammed RPE; t-rRPE), and 3) E4 RPE 6 hours post retinectomy and FGF2 treatment (reprogrammed RPE; rRPE). Total RNA extraction was performed using PicoPure RNA Isolation Kit (Arcturus, Applied Biosystems, Foster city, CA), including a treatment with DNase I. Intact, total RNA was reverse-transcribed using an Oligo(dT) primer, and limited cDNA amplification was performed using the SMARTer® Ultra® Low Input RNA Kit for Sequencing – v4 (Takara Bio USA, Inc., Mountain View, CA). The resulting full-length cDNA was fragmented and tagged, followed by limited PCR enrichment to generate the final cDNA sequencing library (Nextera® XT DNA Library Prep, Illumina, San Diego, CA). Libraries were sequenced as single-end 75 base pair reads on an Illumina NextSeq500 to generate 44-68 million single-end 75 bp reads for each sample. Reads were quality trimmed and adapter sequences removed using Trim Galore v0.6.1 and Cutadapt v1.18. Trimmed reads were aligned to Gallus gallus genome GRCg6a with Ensembl annotation 98 using STAR v2.7.3. Assembly and quantification of transcripts was performed with Stringtie v2.0.4 and normalization of gene counts was performed with DESeq2 v1.22.2.