Single nucleotide polymorphisms in wild isolates of Caenorhabditis elegans

Abstract

Caenorhabditis elegans (isolate N2 from Bristol, UK) is the first animal of which the complete genome sequence was available. We sampled genomic DNA of natural isolates of C. elegans from four different locations (Australia, Germany, California, and Wisconsin) and found single nucleotide polymorphisms (SNPs) by comparing with the Bristol strain. SNPs are under-represented in coding regions, and many were found to be third base silent codon mutations. We tested 19 additional natural isolates for the presence and distribution of SNPs originally found in one of the four strains. Most SNPs are present in isolates from around the globe and thus are older than the latest contact between these strains. An exception is formed by an isolate from an island (Hawaii) that contains many unique SNPs, absent in the tested isolates from the rest of the world. It has been noticed previously that conserved genes (as defined by homology to genes in Saccharomyces cerevisiae) cluster in the chromosome centers. We found that the SNP frequency outside these regions is 4.5 times higher, supporting the notion of a higher rate of evolution of genes on the chromosome arms.

Figure 1

Nature of the SNPs. (A) The number of base pairs involved in substitutions, insertions, and deletions is indicated. Note that we find on average one SNP per 2000 bp. After checking the Bristol N2 sequence for 63 potential SNPs, we found 3 to be a sequence error in N2. Thus, we estimate the error rate of the C. elegans genome sequence to be ∼1 in 42,000. (B) The number of SNPs involved in the first, second, or third base of a codon in coding sequences is plotted. It also shows whether the SNP alters the amino acid. Coding sequences were predicted by Genefinder.

Figure 2

Comparison of SNPs among different natural isolates. (A) The presence of SNPs originally identified in one strain was tested in nine other strains and Bristol N2 by RFLP analysis or sequencing. SNP-containing clones are shown vertically along the chromosomes using their physical map position. Some clones contain multiple SNPs. Eleven unique SNPs in clone W08E3* are not shown in this figure. (B) Local variation was tested by comparison of strains isolated in California for SNPs that can be visualized by RFLP analysis. Most were isolated from flower beds on Caltech campus and from a vegetable garden in Altadena. The presence of a SNP is indicated with a shaded square. Isolates from other locations are plotted for reference.

Figure 3

Detailed comparison of SNPs among isolates in SNP-rich regions near clone W08E3 on chromosome I (A) and clone H04J21 on chromosome III (B). Regions of ∼1 kb flanking the highly polymorphic clones were analyzed by directed sequencing at 1, 5, and 50 kb on each side in 10 natural isolates and N2. Polymorphisms within the tracts are shown vertically. The presence of a SNP is marked with a shaded square.

Figure 4

SNP distribution on the chromosomes. (A) DNA repeats and SNPs on chromosome I are found mainly on the chromosome arms, whereas similarity with yeast genes is highest on the autosome centers (figure adapted from The C. elegans Sequencing Consortium 1998, p. 2016). The number of SNPs/Mb (B) and the number of sequenced clones/Mb (C) are plotted for the six chromosomes. Autosomes are divided into genetically defined compartments of the left arm (L), the central cluster (C), and the right arm (R) (Barnes et al. 1995; The C. elegans Sequencing Consortium 1998). SNPs occur more frequently on the arms than on the central regions. The X chromosome does not show a higher SNP frequency on the arms.