Amplification-free target enrichment for native-strand sequencing using CRISPR/Cas9

It is an advantage in many situations to enrich for regions of interest (ROI) prior to sequencing. Here, we introduce a PCR-free enrichment method for nanopore sequencing, using Cas9 (Fig. 1). Because native strands are sequenced, fragment length and modifications are preserved. In the method, sample DNA is initially dephosphorylated to prevent subsequent off-target ligation. Cas9 is then used to cleave the DNA at predetermined sites, exposing ligatable on-target ends. All 3’ ends are dA-tailed and sequencing adapters only ligate to the cleaved ends. The entire library is then added to the flow cell. In this way the fraction of reads corresponding to the ROI is enriched several thousand-fold, enabling many samples to be run on the same flow cell, or a lower-cost flow cell to be used.

The enzyme CYP2D6 metabolises ~25% of common drugs and its activity between individuals varies widely, due to high levels of polymorphism. To determine an individual’s optimal drug dosage, we need to accurately detect all variants that influence the enzyme’s activity. Long reads help to distinguish between CYP2D6 and its close paralogue, CYP2D7 (Fig. 2a) and between their gene conversion products. We enriched the region from NA18256, previously shown to have one copy of allele *1 and three copies of allele *10. Previous tests did not distinguish between the *10 and *36 variants, the latter being a gene conversion variant. We identified the expected alleles and obtained reads that captured three copies of CYP2D6 (Fig. 2b). Analysis indicates that the haplotype contains one copy of *10 allele and two copies of *36 (Fig. 2c).

Friedreich’s Ataxia is an autosomal recessive neurodegenerative disorder, affecting approximately one in 50,000 individuals, with symptoms including progressive limb ataxia and dysarthia. The disorder is caused by reduced levels of frataxin, a protein encoded by the FXN gene on Chromosome 9q (Fig. 3a) that is involved in the assembly of iron-sulphur clusters in mitochondria. The first intron of the gene contains a GAA repeat, which reaches 70–1,000 copies in both alleles of patients, compared to 5–30 copies in unaffected individuals. The longer the repeat, the more severe the symptoms and the earlier the age of onset. Repeat expansion is thought to alter the structure of the DNA at the locus and to lead to hypermethylation. This, in turn, suppresses production of frataxin, leading to disease symptoms (Fig. 3b). The presence of a large repeat and the involvement of methylation mean that PCR is best avoided when investigating the genetics of the disease. We enriched the region from carrier parents and their affected child using Cas9. We aligned reads to hg38 using minimap2 and haplotyped with MarginPhase. We counted repeat length in base-space (Fig. 3c). Sequencing results agree well with the values derived from Southern blotting. A high level of somatic instability has been reported in expanded FXN alleles, which may account for the difference in repeat size between the father and child. Next we called CpG methylation with Nanopolish. In the parent samples, reads from the repeat-expanded haplotype were hypermethylated in the ~2 kb region between exon 1 and the repeat expansion. In the patient sample, both haplotypes were expanded, and both were hypermethylated. The haplotype with the largest expansion has a slightly higher level of methylation (Fig. 3d).

Download the PDF