Enzyme-free targeted DNA demethylation using CRISPR-dCas9-based steric hindrance to identify DNA methylation marks causal to altered gene expression.

Abstract

DNA methylation involves the enzymatic addition of a methyl group primarily to cytosine residues in DNA. This protocol describes how to produce complete and minimally confounded DNA demethylation of specific sites in the genome of cultured cells by clustered regularly interspaced short palindromic repeats (CRISPR)-dCas9 and without the involvement of an epigenetic-modifying enzyme, the purpose of which is the evaluation of the functional (i.e., gene expression or phenotypic) consequences of DNA demethylation of specific sites that have been previously implicated in particular pathological or physiological contexts. This protocol maximizes the ability of the easily reprogrammable CRISPR-dCas9 system to assess the impact of DNA methylation from a causal rather than correlational perspective: alternative protocols for CRISPR-dCas9-based site-specific DNA methylation or demethylation rely on the recruitment of epigenetic enzymes that exhibit additional nonspecific activities at both the targeted site and throughout the genome, confounding conclusions of causality of DNA methylation. Inhibition or loss of DNA methylation is accomplished by three consecutive lentiviral transductions. The first two lentiviruses establish stable expression of dCas9 and a guide RNA, which will physically obstruct either maintenance or de novo DNA methyltransferase activity at the guide RNA target site. A third lentivirus introduces Cre recombinase to delete the dCas9 transgene, which leads to loss of dCas9 from the target site, allowing transcription factors and/or the transcription machinery to interact with the demethylated target site. This protocol requires 3-8 months to complete owing to prolonged cell passaging times, but there is little hands-on time, and no specific skills beyond basic molecular biology techniques are necessary.