Detection of CRISPR-mediated genome modifications through altered methylation patterns of CpG islands.

Background – epigenetics
The field of epigenetics deals with hereditary changes that do not change the DNA sequence itself. Epigenetic markers determine which genes are activated or silenced in certains cells and tissues, so they influence gene expression. Thus, the basic structure of the DNA remains unchanged. Epigenetic markers are biochemical appendages of DNA.

Epigenetics is decisive for the coordinated development of organisms and determines which functions each cell has. Epigenetics also plays an important role during the entire lifespan of an individual: due to external environmental factors, every higher living being is confronted with conditions that make it necessary to adapt quickly to new situations with a change in gene regulation. This is mediated by changes in the epigenetic markers.

At the level of the DNA, small epigenetic markers can be attached to the base cytosine (one of the four letters of DNA). These appendages are known as methyl groups. They are transferred to the bases of the DNA by certain enzymes known as DNA methyltransferases. The most widespread and probably most important DNA methylation in higher living beings is that of cytosine. In mammals there are areas in the genome in which a particularly large number of cytosines and guanines occur in the DNA sequence, which is why such areas are also referred to as CpG islands. These CpG islands are distributed throughout the entire genome and can mainly be found in genes and their regulatory areas. The cytosines in the CpG islands are hypomethylated, which means they are methylated only to a small extent. This characteristic methylation pattern can be changed by environmental influences. The totality of all epigenetic markers in an organism and its cells and their distribution throughout the entire genome is summarized under the term epigenome.

Results of the Farris et al. study
The study carried out by Farris et al. investigated to what extent the methylation patterns change in the area of target sequences when DNA templates are incorporated that are mediated by the use of CRISPR/Cas, i.e. so-called SDN-3 applications. To incorporate a DNA fragment into the genome, a DNA template is first of all introduced into the cell together with CRISPR/Cas. Then, CRISPR/Cas mediates the cut at the corresponding target sequence, thereby activating the cell’s own repair mechanisms. When an external DNA template is introduced, the HDR (i.e. homology directed repair) is activated, and this triggers the integration of the DNA template into the target sequence. The DNA template is developed in such a way that it contains the intended change in the target sequence (small changes in the DNA sequence up to entire gene sequences) as well as a large part of the surrounding DNA region. The sequence of the DNA template corresponds exactly to the DNA sequence of these neighboring areas, and are therefore known as homologous DNA sequences.

First of all, the scientists compared the methylation pattern in the entire genome of CRISPR/Cas modified and unmodified mice. To do this, they used a procedure called ‘whole genome bisulfite sequencing‘ to identify unintended changes in their methylation patterns. They examined two mouse strains that were modified by using CRISPR/Cas and DNA templates at two different target areas (HDR1 and HDR2). They also analyzed another genome-edited mouse strain to which no external DNA template was added, and in which the activation of the NHEJ repair had led to small insertions and deletions at the target site (so-called SDN-1 applications). The target areas of all mice modified by CRISPR/ Cas were within or in the immediate vicinity of CpG islands, i.e. areas in the genome that normally have a low DNA methylation pattern.

The investigations showed that both within the DNA region of the intended change and in the homologous DNA sequences of the HDR1 and HDR2 mouse strains, the original methylation pattern was changed and more methylation groups were present. In the case of the NHEJ strain, the scientists could not detect any change in the methylation pattern at the target sequence or within the adjacent CpG island. The changes seem to be a specific effect of the insertion of DNA templates by the HDR after a CRISPR/Cas application. The results were obtained in mice which were in the tenth generation after the original CRISPR/Cas application. Thus, they can be inherited, last for many generations and are detectable. In their study, the scientists suggest that the change in the methylation pattern could be used to detect DNA insertions in the genome resulting from the application of CRISPR/Cas (i.e. SDN-3 applications).

Relevance for plants
The methylation pattern in plants is different compared to mammals: the cytosines are more evenly distributed over the entire genome and there are no specific CpG islands in plant genomes. However, DNA methylation is also an important epigenetic marker that regulates gene expression in plants and thus many processes, e.g. during their development. Another study (Lee et al., 2020) investigated in Arabidopsis thaliana whether CRISPR/Cas can also mediate unintended changes in the methylation pattern. Here, however, only NHEJ changes, i.e. small changes to the DNA sequence (SDN-1 applications), were investigated. The methylation pattern was examined at four different locations in the genome of plants modified by CRISPR/Cas as well as unchanged control plants. Two target areas contained little DNA methylation in their normal state, the other two more. The result: the DNA methylation pattern did not change in any of the plants modified by CRISPR/Cas compared to the controls. Changes introduced at the target sequence also had no effect on the DNA methylation pattern. The question that remains open here is whether similar effects can also occur with the activation of HDR and the incorporation of DNA templates, i.e. SDN-3 applications in plants.

Open questions
It was not investigated whether the gene expression in the HDR1 and HDR2 mice was influenced by the altered methylation pattern. It can be assumed that gene expression in the immediate vicinity of the intended change may be impaired. Further investigations must investigate this.

In addition, other epigenetic markers should also be examined after applying CRISPR/Cas: besides the methylation of DNA, proteins can also be altered by attaching epigenetic markers that are in direct contact with the DNA and can influence gene expression. For these investigations, methods such as ChIP-seq experiments should be carried out to investigate the distribution of such epigenetic markers over the entire genome.

The results obtained by Farris et al. could also be important for the development and the risk assessment of gene drives. With CRISPR/Cas-based gene drives, the mode of action of the drive is based on the transmission of the genetic information for CRISPR/Cas to all offspring. During embryogenesis, the target sequence is cut by CRISPR/Cas and the genetic information for CRISPR/Cas is incorporated into the wildtype chromosome via HDR repair. During the risk assessment of gene drive organisms, the epigenome should also be taken into account to investigate possible unintended effects on gene expression.

Farris MH, Texter PA, Mora AA, Wiles MV, Mac Garrigle EF, Klaus SA, Rosfjord K (2020) Detection of CRISPR-mediated genome modifications through altered methylation patterns of CpG islands. BMC Genomics 21 (1):856. doi:10.1186/s12864-020-07233-2
Lee JH, Mazarei M, Pfotenhauer AC, Dorrough AB, Poindexter MR, Hewezi T, Lenaghan SC, Graham DE, Stewart CN (2020) Epigenetic Footprints of CRISPR/Cas9-Mediated Genome Editing in Plants. Frontiers in Plant Science 10 (1720). doi:10.3389/fpls.2019.01720