C. elegans mutant identification with a one-step whole-genome-sequencing and SNP mapping strategy

Abstract

Whole-genome sequencing (WGS) is becoming a fast and cost-effective method to pinpoint molecular lesions in mutagenized genetic model systems, such as Caenorhabditis elegans. As mutagenized strains contain a significant mutational load, it is often still necessary to map mutations to a chromosomal interval to elucidate which of the WGS-identified sequence variants is the phenotype-causing one. We describe here our experience in setting up and testing a simple strategy that incorporates a rapid SNP-based mapping step into the WGS procedure. In this strategy, a mutant retrieved from a genetic screen is crossed with a polymorphic C. elegans strain, individual F2 progeny from this cross is selected for the mutant phenotype, the progeny of these F2 animals are pooled and then whole-genome-sequenced. The density of polymorphic SNP markers is decreased in the region of the phenotype-causing sequence variant and therefore enables its identification in the WGS data. As a proof of principle, we use this strategy to identify the molecular lesion in a mutant strain that produces an excess of dopaminergic neurons. We find that the molecular lesion resides in the Pax-6/Eyeless ortholog vab-3. The strategy described here will further reduce the time between mutant isolation and identification of the molecular lesion.

Figure 1. Principle of the WGS-SNP strategy.

The red diamond indicates the mutation of interest. The sequence ‘pileup’, generated by processing the WGS sequencing data, is a representation of the number of reads that are mapped to a specific position in the genome. The relative number of N2 vs. Hawaiian nucleotides is a reflection of the relative distribution of recombinants in the pool.

Figure 2. ot266 mutant animals show ectopic expression of a dopaminergic cell fate marker.

A: The head region of an adult worm is shown. 4 CEP and 2 ADE neurons express gfp in wild-type animals. The marker is vtIs1 (dat-1::gfp). See Table 1 for quantification. B: VAB-3 protein structure and location of alleles.

Figure 3. Application of the WGS-SNP strategy for mapping and cloning ot266.

Red arrow indicates the phenotype-causing mutation in vab-3(ot266). Numbers next to the chromosomes depict how many SNP loci show occurrence of Hawaiian SNPs in the 0.2–0.6 range (see Material and Methods for explanation of this range). Note that the mapping interval obtained from the 50 recombinants, lane 1 dataset is smaller than the one defined from 50 recombinants, two lanes dataset (see also Table 2 ). We believe that this is due to random variability in sample representation in the flow cell. Also note that this dataset does not allow to easily infer the position of the vtIs1 transgene since we did not select for homozygous vtIs1 animals in the F2 generation.

Figure 4. Alternative graphic representation of the WGS-SNP data.

The WGS SNP data is shown for Chromosome X, 50 recombinants, 2 sequencing lanes. The positions of SNP loci are depicted as a XY scatter plot, where the ratio ‘Hawaiian/total number of reads for each SNP is represented. The red line indicates the position of the phenotype-causing variant. 3 different filtering criteria were used, as indicated in the figure. Note that the non-linear appearance of the right arm of the scatter plot reflects the increased recombination rate on that part of the chromosome (as can be visualized by Marey map for CHR X –data not shown). Using 0.1–0.8 filtering criteria, narrowed down the mapping interval even further (1.57 Mb).