Worm Breeder's Gazette 15(4): 16 (October 1, 1998)
These abstracts should not be cited in bibliographies. Material contained herein should be treated as personal communication and should be cited as such only with the consent of the author.
Department of Molecular Biology and Pharmacology, Washington University School of Medicine, St. Louis, MO 63110
DNA sequence polymorphisms (DSPs) have been used extensively to position mutations on the C. elegans physical map. DSPs typically have been identified using restriction enzyme digests and Southern blotting to detect RFLPs, or PCR and gel electrophoresis to detect differences in product size (Ruvkun, G. et al. Genetics 121, 501-516, 1989; Williams, B. D. et al. Genetics 131, 609-629, 1992). These approaches are limited because: (1) they identify only a subset of the existing DSPs, and (2) the precise position of the base change(s) responsible for an RFLP is unknown. Here we describe the use of DNA sequencing to identify DSPs. This approach is simple, efficient, and it largely overcomes these limitations. The sar-5(n2527) mutation affects vulval development and was genetically mapped between lon-2 and unc-6, cloned genes separated by about 1900 kb on chromosome X. To identify DSPs that could be used to precisely position n2527, we chose RC301 to compare with our N2-derived strain, since we previously had success finding RFLPs in RC301. (We are grateful to Tom Barnes for suggesting RC301.) We generated let-60(gf); lon-2 sar-5(n2527) unc-6/ RC301 animals, selected 201 Lon non-Unc recombinants, and selected self-progeny homozygous for the recombinant chromosome. We scored the sar-5 phenotype and prepared DNA from these animals. To identify DSPs, we used N2 genomic sequence to design 21 pairs of oligo- nucleotide primers that amplify predicted intergenic regions, reasoning that DSPs are more frequent in noncoding sequence. Primers typically contained 20 bases and 50% G/C. Products were successfully amplified using 19 of these primer pairs, purified from agarose gels, and directly sequenced using both amplification primers and an ABI automated sequencer. Primers were separated by ~1.2 kb to exploit the capability of the sequencer - up to 600 bases/primer. We sequenced DNA from RC301 and lon-2 sar-5(n2527) unc-6 worms. Thirteen differences between the sequence of RC301 and N2 were detected in 11 of the 19 products: nine substitutions of a single base pair (bp) and one substitution of two adjacent bps; one deletion of two bps; one insertion of a single bp and one insertion of 300 bp. The 300 bp insertion was detectable by PCR and gel electrophoresis, whereas the other DSPs were detectable only by sequencing. Although no DSPs were detected in the N2-derived strain, these ABI traces helped verify changes detected in RC301. In practice, we alternated between identifying DSPs and mapping sar-5. For example, the first two DSPs subdivided the 1900 kb interval into three parts, and sar-5 mapped to the central interval. We then searched for DSPs in that interval. After five rounds of identifying DSPs and mapping, sar-5(n2527) was positioned between two DSPs separated by 9.6 kb. This interval contains only two predicted open reading frames, which we are sequencing to identify the n2527 mutation. Based on our results, we conclude that DSPs are abundant in RC301. In the 17 kb of intergenic sequence analyzed, we detected an average of one DSP per 1.4 kb. It is likely that DSPs are abundant elsewhere in the RC301 genome. For example, the accompanying article shows that they can be readily identified on chromosome II R. Given this abundance, DNA sequencing is an efficient method for identifying DSPs. In addition, DNA sequencing overcomes both of the major limitations of previous detection methods, since it can (1) identify most or all existing DSPs and (2) establish the precise position of DSPs. We conclude that this method can be used to position most mutations to an interval containing only a few genes. This is important, because it makes it possible to identify the mutation by sequencing the interval without using injection rescue. This might be particularly useful for gain-of-function mutations that cannot be rescued by injection of wild-type DNA. Given the abundance of DSPs in RC301, a systematic effort to identify DSPs throughout the genome might yield a high density map that could be used for initial mapping efforts, like the STS mapping approach described by Williams et al. (1992).