Next Generation Sequencing of Mobile Genetic Elements in Dry Age-Related Macular Degeneration

Abstract

 

Purpose

 

Previously, we have demonstrated increased levels of non-coding RNA derived from Alu sequences in human eyes with advanced dry age-related macular degeneration (geographic atrophy, GA) using multiple techniques including adaptor-ligation PCR and in-situ immunolocalization. Here, we developed a next-generation Alu RNA sequencing pipeline to quantify all known Alu sub-family sequences from tissues and cells.

 

Methods

 

Primary human RPE isolates (n=3) were treated with Dicer1 antisense oligo-nucleotide (nt) or control. RNA was purified (RIN>9) and separated into total or 300-600 nt size fractions. Specific Alu sequences were added to hRPE RNA samples as a positive control. Macular RPE/choroid samples were obtained from normal age-matched and GA eyes (harvest<8 hours, RIN>8). Sequencing libraries were created (200ng, Truseq/ Ribominus) and 40 million paired-end 100-bp reads were acquired (Illumnia HiSeq 2000). A bioinformatics pipeline allowing for the alignment of Alu sequences (mismatch rate < 3 bp, UCSC hg19) was optimized, and Alu sequences were identified in RPE samples.

 

Results

 

The Alu-Seq pipeline designed was capable of detection of Alu sub-family RNA sequences in cells and tissues. Down-regulation of Dicer1 expression in hRPE led to over a 5-fold increase in Alu RNA sequences compared to control, and we were able to simulatneously study levels of 37 Alu sequence sub-families found in our samples. Size selection of sample RNA for 300-600 nt RNA significantly increased the fold-increase in the Dicer1 deficient cells. Macular RPE/Choroid from human eyes with GA harbored 30% higher levels of Alu RNA compared to age-matched controls with a similar distribution of subfamily Alu sequence expression to the in vitro model.

 

Conclusions

 

We have successfully designed a next-generation sequencing sample preparation protocol and bioinformatics pipeline for the identification of Alu sub-family sequences. Similar to our previous data, Alu RNA levels were increased in GA eyes and in cell culture models of Dicer1 deficiency. Moreover, we have identified a specific pattern of Alu sub-family gene expression that reveals more significant upregulation of Alu Y family expression compared to others. Size fractionation of total RNA samples resulted in significantly improved detection of elevated Alu RNA expression.