Abstract
Purpose:
Purpose: The contribution of epigenetic processes to the normal development and pathophysiology of the visual system is an area of intense interest. DNA methylation in particular is of interest as a potential master epigenetic regulator of gene expression. However, tools to study DNA methylation in a genome-wide manner in rodent model systems are lacking. Rat model systems are a preferred rodent model in a number of biomedical research areas including vision and neuroscience. Epigenetic analyses of rat models have been limited due to the lack of available species-specific research tools. To enable DNA methylation studies in rats we have developed a capture-probe approach (MethylCap-Seq) that targets the promoter regions of RefSeq genes as well as CpG islands not present in repeat regions. As the accuracy of 5-mC analysis is dependent on sequencing depth, a capture oligonucleotide based approach allows for greater depth sequencing to be targeted to the regions of interest.<br />
Methods:
Methods: Our approach targets 18,363 CpG islands, including their shores and shelves. Additionally 4kB upstream and 1 kB downstream of the transcription start site of 18,971 RefSeq genes is captured. The non-overlapping set comprises approximately 170 million bases of the rat genome. Use of a targeting approach provides for genome-wide analysis while also greatly increasing accuracy of 5-mC quantitation, reducing sequencing costs, and increasing sample throughput compared to whole genome bisulfite sequencing methods. In conjunction we developed a targeted methylation analysis workflow to confirm regions of interest from targeted genomic regions by combining genome-wide Multiple Displacement Amplification (MDA) with targeted Bisulfite Amplicon Sequencing (BSAS) to focus on selected genomic regions (1-10Kb) from extremely limited samples (<1ng). <br />
Results:
Results: Using whole genome methylation standards and example in vivo differences we demonstrate the accuracy of MethylCap-Seq and MDA-BSAS for methylation quantitation and provide a bioinformatics workflow for analysis of sequencing data.<br />
Conclusions:
Conclusions: The combination of these approaches allows for discovery of differentially methylated regions in one of the most commonly used rodent models and confirmation/extension of findings into larger sets of limited samples. <br />