Various genome-editing technologies provide a powerful means to manipulate a DNA sequence in its native genomic context. Such manipulations are generally carried out in a low-throughput, gene-by-gene manner, and even emerging high-throughput gene-editing libraries typically target each of many genes once. A new study instead demonstrates high-throughput editing to generate up to thousands of variants of single genes.
Findlay et al. targeted the human BRCA1 gene, for which many germline variants naturally exist in humans, and some of these are linked to breast and ovarian cancer risk. They co-transfected a population of human HEK293T cells with a CRISPR–Cas9 construct and a complex library of repair templates to engineer >4,000 possible hexameric nucleotide combinations in a single region of exon 18 of BRCA1. Simultaneously, they introduced an adjacent universal mutation to facilitate the subsequent PCR-based amplification of edited genomic loci and transcripts. In a related application, the investigators engineered all possible single-nucleotide variants across the 78-bp length of BRCA1 exon 18.
As the engineered sites were in a transcribed region, the team used targeted high-throughput RNA sequencing to identify the resultant variant transcripts and to compare their expression levels. As proof that this approach can identify functional effects of gene variants, the researchers found that nonsense variants resulted in a reduced transcript abundance (through nonsense-mediated decay), and that variants with predicted effects on splicing led to the expected increase or decrease of transcript levels.
As an additional application, the investigators engineered almost 400 variants across a well-conserved region of the essential DBR1 gene in human Hap1 haploid cells to unmask the cellular effects of each variant allele created. To test effects on cell viability, the team took cell samples across an 11-day time course post-transfection and carried out targeted genomic DNA sequencing across the edited region. Deleterious gene variants would lead to slower proliferation or cell death, thus resulting in depletion of these cells (and their genomic copies of variants) over time. Indeed, nonsense, frameshift and enzyme active site mutations were the most depleted, thus validating the approach.
This high-throughput genome-editing method holds great promise for dissecting the functional implications of gene variants, particularly for clinically discovered variants of unconfirmed consequence. A key future direction will be to expand the range and relevance of high-throughput assays to test the physiological effects of these variants.
Findlay, G. M. et al. Saturation editing of genomic regions by multiplex homology-directed repair. Nature http://dx.doi.org/10.1038/nature13695 (2014)
About this article
Cite this article
Burgess, D. Testing all possibilities. Nat Rev Genet 15, 646 (2014). https://doi.org/10.1038/nrg3822