Worm Breeder's Gazette 14(3): 11 (June 1, 1996)
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.
University of Wisconsin Dept. of Genetics 445 Henry Mall , Madison, WI 53706
DNA molecules injected into the C. elegans hermaphrodite gonad undergo efficient homologous recombination during the formation of transforming arrays (1). We have taken advantage of this phenomenon to rescue unc-68 with overlapping "long range" PCR fragments amplified from N2 genomic DNA. unc-68 (ryr-1) encodes a 5071 amino acid ryanodine receptor gene that spans over 30 kb on LGV. Yasuji Sakube and Hiro Kagawa sequenced the unc-68 coding region, which is contained in cosmid MO4C11(2). However, MO4C11 failed to rescue unc-68 mutants, as did other cosmids that share sequence with MO4C11. Sakube and Kagawa further showed with promoter fusions to LacZ that muscle-specific expression of unc-68 requires sequences outside of MO4C11 that lie in a cosmid gap (2). These experiments suggest that the cosmids that failed to rescue lacked control regions that may include "unclonable" sequences. Rather than attempt rescue with YACs or mixtures of lambda and cosmid clones, the unc-68 gene was directly amplified, including the upstream sequences in the cosmid gap. Using Sakube and Kagawa's genomic sequence, three fragments (about 10 kb each) that together span unc-68 were amplified from N2 genomic DNA with Boeringer Manheim's Expand(TM) Long Template PCR polymerase mixture. The middle fragment overlapped the 5' and 3' fragments by about 450 bp at either junction. The fragments were purified on a spin column and injected into N2 and unc-68(r1161) hermaphrodites, either alone or with the pRF4 rol-6 marker plasmid (aprox. 100ug/ml total DNA conc.). The initial injections of both strains failed to yield non-Unc or Rol transformants, suggesting a toxic effect of unc-68 overexpression. After diluting the fragments 25-fold with inert DNA (BRL 1kb ladder), injected N2 animals segregated Rol progeny, and both wild -type and Rol (non-Unc) lines were segregated from injected unc-68(r1161) hermaphrodites. The wild-type r1161 animals transformed without rol-6 DNA were fully rescued, as judged by motility assays and by their sensitivity to contraction by ryanodine. These transformants segregate wild-type and Unc animals. Single worm PCR of wild-type transformants showed that both the r1161 (deletion) allele and wild-type unc-68 sequences were present. A male stock from a strain carrying a stable array (rEx95), has been used to rescue other unc-68 alleles. We envision other uses for Long Range PCR. First, we hope to use the transformation rescue assay to localize mutations within the 30 kb which comprises unc-68. Mutant alleles will be amplified in 3 (or more) pieces, and each fragment can be mixed with 2 wild type fragments and tested for rescue. Any mutant fragment able to rescue wild-type can be excluded from containing the mutation of interest. Second, amplifying large genes like unc-68 in a "modular" fashion could simplify the creation of expression constructs, and of epitope tagged or mutated alleles. Portions of genes amplified from genomic DNA can be designed to contain convenient restriction sites in the PCR primers, allowing subsequent ligation to cloned DNA. Cloned DNA sequences can be joined to amplified genomic sequences in vivo by providing a homologous overlap at one or both ends. Third, candidate genes (ORFs) can be tested for rescue for the price of a primer pair. Any ORF from sequenced DNA can be amplified from uncloned genomic DNA and tested for rescue. For example, if the sequence of a rescuing cosmid is known, candidate genes can be quickly tested without multiple subcloning steps. There are obvious concerns about the introduction of mutations during PCR, but by pooling several small PCR reactions for each fragment, the contribution of any given mutation is minimized. Futhermore, the multicopy nature of transforming arrays also reduces the effect of random mutations. 1. Mello, C. and Fire, A. 1995 DNA Transformation, in Caenorhabditis elegans, Modern Biological Analysis of an Organism, pp.452-482. Epstein and Shakes, Eds. Academic Press. 2. 10th International Worm Meeting Abstracts p. 452