Messenger RNAs are rapidly degraded in wild type cells if they contain upstream nonsense codons or are incorrectly processed. If such mRNAs persist and are translated, truncated or aberrant protein products might be generated, which can be detrimental to cell function. The smg genes in C. elegans mediate the process of recognition and destabilization of aberrant mRNAs, termed mRNA surveillance (1,2). In order to understand the molecular mechanisms underlying recognition of aberrant mRNAs, we are cloning
smg-4, which was defined by the allele
ma116. Genetically,
smg-4(
ma116) maps quite close and to the right of
unc-42(
e270). Correlation of the genetic and physical map by southern blots of deficiency strains for the region showed that
smg-4 and
unc-42 are located in part of the C. elegans physical map covered only by clones in yeast artificial chromosomes (YACs), and not cosmids, a "YAC bridge." Indeed, mutants in both genes are rescued by YACs spanning the bridge, Y1D1 and Y42F5, but not by pools of cosmids at either end of the bridge. We have now isolated new mutations in
smg-4 with psoralen/UV at a frequency of 1/10,000, but the EMS target size was found to be small (<1/20,000). These new psoralen/UV mutants are viable as homozygotes, as is a DEO-induced allele isolated by S. Kuchma in P. Anderson's laboratory. To narrow the rescuing region, circular deletion derivatives of the rescuing YACs were generated (reagents from S. Kim). However, clones deleting the left end of Y42F5 can rescue
unc-42(
e270), but not
smg-4 (
ma116), a surprising result given the alleles' genetic map positions. Work is in progress to determine whether a domain required for
smg-4 expression might lie to the left of
unc-42, i.e. the two genes may reside in a complex locus. For instance, some of the newly generated
smg-4 alleles might map to the left of
e270. Alternatively, the
smg-4 gene might have been disabled during homologous recombination of the YAC and fragmentation vector in yeast. A genomic phage walk is also in progress to identify other clones which will rescue
smg-4 mutants. 1. J. Hodgkin, et al. Genetics 23(2) 301-313 (1989). 2. R. Pulak and P. Anderson. Genes and Dev. 1885-1897(1993).