In a recent study posted to the preprint server Research Square* while under review for publication in Epigenetics & Chromatin, researchers develop and characterize a highly specific EpiEditing system using catalytically inactivated Cas9 (dCas9) to achieve allele-specific deoxyribonucleic acid (DNA) methylation (ASM).
Study: Development of super-specific epigenome editing by targeted allele-specific DNA methylation. Image Credit: Natali _ Mis / Shutterstock.com
*Important notice: Research Square publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.
Background
Epigenomic editing involves the precise reprogramming of specific genomic regions using a tool called an EpiEditor. EpiEditor typically consists of components such as dCas9, DNMT3A/3L, and single guide ribonucleic acid (RNA) (sgRNA). By introducing methylation to a particular locus, gene expression can be modulated.
ASM specifically refers to the targeted delivery of methylation to only one allele of a genomic locus. In the context of diseases caused by a dominant mutation, selectively methylating the mutant allele could be utilized to suppress its expression while preserving the functionality of the gene.
About the study
In the present study, a single sgRNA expression vector was constructed by replacing a section of the pMulti-sgRNA-LacZ-DsRed vector backbone with the sgRNA expression cassette from the AIO-mCherry vector.
For targets with single nucleotide polymorphisms (SNPs) in the protospacer adjacent motif (PAM) region, a 20 nucleotide (nt) genomic sequence upstream of the PAM was selected as the sgRNA binding site. Comparatively, for targets with SNPs in the sgRNA seed region, a 20 nt sequence upstream of the nearest PAM containing the SNP was used.
The selected sgRNAs were evaluated for their efficiency and potential off-target effects using CRIPOR and CCTop. Quadruplex DNA formation in the sgRNA region was analyzed using QGRS Mapper, whereas cloning of sgRNA expression vectors was performed using Golden Gate Assembly. Moreover, transfection of HEK293 cells was carried out with a mixture of dCas9-10x SunTag, scFv-DNMT3A/3L, and the multi-sgRNA expression vector.
Fluorescent markers were included in each plasmid to facilitate the sorting of triple-positive cells. Genomic DNA was extracted, bisulfite converted, and subjected to library preparation for next-generation sequencing (NGS).
DNA methylation analysis was also performed using Trim Galore!, PEAR, bwameth, and MethylDackel. RNA isolation and complementary DNA (cDNA) synthesis were followed by expression analysis using gene-specific primers and library generation with two-step polymerase chain reaction (PCR). Allelic expression ratios were determined for genes with SNPs in exons.
Study results
HEK293 cells were selected for this study due to their ease of handling and high transfection efficiency. To gather data on SNPs in the genome, a public database was consulted. Initial estimates of DNA methylation levels were obtained from MBD2-pulldown sequencing data generated in a previous study.
An in silico search was conducted to identify SNPs in unmethylated CGIs (CpG islands) located in gene promoter regions. The search results were filtered based on the presence of SNPs near the potential PAM site or within the seed region of a promising sgRNA. Fourteen target genes with suitable SNPs were identified, with up to three sgRNAs designed to target each gene allele.
The experiments were categorized based on the location of SNP in sgRNA, including the seed region, position two of the PAM sequence, or PAM position three. SNPs in the PAM region leading to guanine (G) to Y change, where Y represents cysteine (C) or taurine (T), were preferred, as these changes exhibited the best discrimination of scFv-DNMT3A/3L-sfGFP, dCas9-10x SunTag-BFP, and sgRNA-DsRed constructs into HEK293 cells. Control experiments were also conducted using a scrambled sgRNA without a binding site in the human genome.
The impact of ASM on allele-specific gene expression ratios was also determined. Genes with SNPs in exons were selected, and allelic expression ratios were determined.
ASM led to significant changes in allelic expression ratios for the ISG15-Seed2 and MRPL52-PAM3 genes. The researchers were also interested in developing and characterizing highly specific EpiEditing systems using dCas9 to achieve ASM. An improved version of DNMT3A/3L with reduced non-specific activity was used as the catalytic part and connected to the sgRNA/dCas9 complex through the SunTag system for signal amplification and dimerization.
Heterozygous SNPs within the seed region of the sgRNA or PAM region of dCas9 were targeted to achieve allelic discrimination. The efficiency of targeted DNA methylation varied among the targets, with most successful targets reaching average DNA methylation of more than 50% at selected CpG sites.
The success of ASM was evaluated based on the overall level of ASM and the ratio of DNA methylation gain at the on- and off-target alleles. The most successful ASM experiments had the SNP placed in the PAM region. Unwanted DNA methylation of the off-target allele could arise from either off-target binding of the sgRNA/dCas9 complex or in the targeted activity of the DNMT3A/3L.
Conclusions
The researchers of the current study successfully achieved targeted allele-specific DNA methylation using sgRNA/dCas9 complexes in HEK293 cells. This novel approach was associated with high specificity, efficiency, and stability of allele-specific methylation that generated changes in the allelic ratio of gene expression.
Taken together, the study findings demonstrate the potential of sgRNA/dCas9 complexes for allele-specific epigenome editing and their applicability in personalized medicine.
*Important notice: Research Square publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.
Revolutionary AI predicts aging and disease from DNA patterns