NEW YORK (GenomeWeb) – A new study published today in Nature Biotechnology shows that on-target editing by CRISPR-Cas9 can result in large deletions and complex genomic rearrangements that extend over many kilobases in the genome, and that this genomic damage may be pathogenic.
The study highlights the problem of off-target effects caused by CRISPR-based genome editing, which has long been a concern for researchers and companies looking to turn the technology into treatments for a variety of human diseases.
As the three authors, led by Allan Bradley of the Wellcome Sanger Institute, noted in their paper, a variety of sensitive detection methods, modified Cas9 enzymes, and improved delivery protocols have been developed to limit the damage that off-target effects can cause. Further, it was thought that the majority of on-target Cas9 activity resulted in indels of less than 20 base pairs, with more complex deletions being rarer.
But the researchers speculated that those assessments "may have missed a substantial proportion of potential genotypes generated by on-target Cas9 cutting and repair, some of which may have potential pathogenic consequences following somatic editing of large populations of mitotically active cells."
In order to test their theory, Bradley and his colleagues looked at allelic diversity induced by Cas9 at the X-linked PigA locus, which is hemizygous in male embryonic stem (ES) cells. They introduced Cas9 and guide RNA constructs targeting intronic and exonic PigA sites into mouse ES cells and found that single gRNAs targeting exons 2 to 4 yielded rates of complete PigA loss at 59 percent to 97 percent. Further, single gRNAs targeting intronic sites also yielded PigA-deficient cells at significant frequencies — 10 different guides located 263 bp to 520 bp from the nearest exon caused 8 percent to 20 percent PigA loss, whereas two guides greater than 2 kb away induced 5 percent to 7 percent loss.
In order to better understand the genetic changes underlying the PigA-deficient cells, they then amplified a 5.7-kilobase region around exon 2 from pools of cells edited with three selected gRNAs and sequenced the PCR products using the PacBio platform. "We observed a depletion of read coverage on a kilobase scale around the cut sites, consistent with the presence of large deletions," the authors wrote. "The most frequent lesions in these cells were deletions extending many kilobases up- or downstream, away from the exon. We conclude that, in most cases, loss of PigA expression was likely caused by loss of the exon, rather than damage to intronic regulatory elements."
When they clustered the PacBio reads, the researchers observed that editing with the three different gRNAs resulted in 183 unique, high-quality alleles, ranging from simple deletions and insertions to complex rearrangements. In order to characterize a variety of the edited PigA loci, they isolated single-cell clones and amplified the PigA loci around the gRNA target site using PCR primer pairs positioned progressively further apart (up to 16 kilobases), until amplicons were generated. They then sequenced the amplicons using Sanger sequencing.
Through this method, they found simple deletions overlapping both the cut site and the exon in almost 75 percent of PigA-deficient alleles generated by single, intronic gRNAs. The largest spanned 9.5 kilobases. The remaining events were deletions combined with large insertions or more complex, multiple-lesion alleles. Further, 23 of 133 recovered alleles contained additional lesions — such as SNPs, indels, large deletions, and insertions — that were non-contiguous with the lesion at the cut site.
In order to investigate whether the observed on-target DNA repair-associated damage was a property of undifferentiated mouse ES cells, Bradley and his colleagues conducted a follow-up experiment in a differentiated human cell line. They found that editing PIGA in this cell line resulted in a loss of the gene at frequencies comparable to those observed in mouse ES cells.
"We show that extensive on-target genomic damage is a common outcome at all loci and in all cell lines tested. Moreover, the genetic consequences observed are not limited to the target locus, as events such as loss-of-heterozygosity will uncover recessive alleles, whereas translocations, inversions, and deletions will elicit long-range transcriptional consequences," the authors concluded. "Given that a target locus would presumably be transcriptionally active, mutations that juxtapose this to one of the hundreds of cancer-driver genes may initiate neoplasia."
In a clinical context, they warned, this could result in one or more edited cells "with an important pathogenic lesion." This means that when editing is done ex vivo, the genome must be carefully examined afterwards, they added. And since genetic damage of this size is undetectable by commonly used short-range PCR assays, comprehensive genomic analysis is needed to identify cells with normal genomes before edited cells are re-injected into patients.