We are using a positional cloning approach to clone
smg-3, one of 7 smg genes required for the degradation of nonsense mutant mRNAs in C. elegans. Several lines of evidence indicate that
smg-3 maps to a cosmid gap on LGIV. Our current strategy for cloning
smg-3 employs a genomic walk into the cosmid gap described below.
smg-3 was localized to a gap between the cosmids T13A10 and F15B10 in the
unc-44 to
deb-1 physical map interval using an RFLP mapping approach (see Cali and Anderson, abstract #71, 1993 International Worm Meeting). The T13A10/F15B10 cosmid gap is bridged by a single 600 kb YAC, Y73B6. Positioning of
smg-3 to the T13A10/F15B10 cosmid gap was confirmed by transformation rescue experiments. Y73B6, which spans and extends to the right of the T13A10/F15B10 gap has
smg-3(+) activity. A circularized derivative of Y73B6 containing only the C. elegans genomic DNA from T13A10 to F15B10 was isolated. This derivative, TR#242, is approximately 250 kb and has
smg-3(+) activity. Y48F8, which extends into but not across the cosmid gap does not rescue
smg-3 mutants. Thus,
smg-3 is not wholly contained on Y48F8. Since
smg-3 is in the cosmid gap to the right of the Y48F8 right endpoint, a chromosome walk from the right end of Y48F8 was initiated. For this walk, we are using an unamplified lambda library constructed from yeast strains carrying Y48F8 and TR#242 (A. Davies, pers. comm.). Genomic clones obtained from our walk were shown by restriction and Southern blot analyses to overlap with a lambda clone contig isolated by Andrew Davies as part of a genomic walk from T13A10 rightward. Thus, these clones define the right end of a new lambda contig which extends from T13A10 rightward for approximately 55 kb (8 kb beyond the right end of Y48F8). Both
gut-4 and
bli-6 , which are separated by 0.08 map units (A. Davies and J. Shaw, abstract #177, 1995 International Worm Meeting) are contained on this 55 kb lambda contig (A. Davies, pers. comm.). 3-factor mapping data suggests that
smg-3 is tightly linked to
bli-6 (of 13 recombinants in the
bli-6 to
unc-24 interval, no recombinants between
smg-3 and
bli-6 were isolated). These data suggest that
smg-3 is close to the right end of Y48F8. We are testing clones isolated from this walk in transformation rescue experiments as well as in Southern blot analyses to detect potential allele-specific polymorphisms of
smg-3 mutants.