Mutation rate and predicted phenotypic target sizes in ethylnitrosourea-treated mice

Abstract

Chemical mutagenesis of the mouse is ongoing in several centers around the world, with varying estimates of mutation rate and number of sites mutable to phenotype. To address these questions, we sequenced approximately 9.6 Mb of DNA from G1 progeny of ethylnitrosourea-treated mice in a large, broad-spectrum screen. We identified 10 mutations at eight unique sites, including six nonsynonymous coding substitutions. This calibrates the nucleotide mutation rate for two mutagenesis centers, implies significance criteria for positional cloning efforts, and provides working estimates of effective genetic target sizes for selected phenotypes.

Figure 1.—

Mutagenesis and breeding. C57BL/6J male mice (G₀) were injected with ENU to induce mutations in spermatogonia. Gametes carrying mutations (small asterisks) were recovered by breeding G₀ males to C57BL/6J females (B6). Progeny males (G₁) will be heterozygous for any transmitted mutations (large asterisk). G₁ males were bred to B6 females and their tail biopsies were screened for mutations by DNA sequencing. G₃ progeny derived by backcross of G₂ females with their G₁ sires were ascertained for neurological phenotypes. G₃ litters from a given G₁ include heterozygotes and homozygotes (double asterisk) for transmitted viable mutations.

Figure 2.—

G₁ DNA sequencing. (A) Chromosomal locations of 54 autosomal genes resequenced. Genes are indicated by horizontal lines; genes separated by <0.5 Mb are shown by diagonal lines. Several genes were resequenced in more than one fragment for a total of 84 fragments. Solid stars indicate genes for which we observed nonsynonymous substitutions; open stars, synonymous substitutions. (B) Number of base calls at or above a given Q value (phrap). Base scores Q ≥ 15 were evaluated for potential heterozygosity. Restricting analysis to bases Q ≥ 20 does not substantially alter our results. Ovals indicate the number of mutations identified for a given initial Q value and later confirmed.

Figure 3.—

Mutations identified by resequencing. Heterozygous sequence traces and protein alignments are shown for mutations detected by resequencing. Gene names are indicated at left and the protein coding sequence and reading frame are indicated above each sequence trace. Homozygous mutant is also shown for Pnmt. Phylogenetic analysis using apparent orthologs (black text) or paralogs (gray text) is shown for each nonsynonymous substitution.

Figure 4.—

Probability of no confounding mutations as a function of interval size and mutation rate. Curves are shown for germline mutation rates reported here and by Coghill  et al. (2002) and for the rate in ES cells reported by Vivian  et al. (2002).