Several lines of evidence suggest that
e1751 might be a gain-of- function allele of
mab-5. The effect of
e1751 on the Q migrations is directly opposite that of
mab-5 loss-of-function mutations. (Ed Hedgecock, Dev. 100) In addition,
e1751 causes generation of ectopic V ray papillae, whereas
mab-5 loss-of-function alleles prevent V ray production. (Andrew Chisholm, pers. comm.) Mapping experiments by Ed Hedgecock showed that
e1751 could not be easily separated from
mab-5 by recombination, and finally, a restriction length polymorphism associated with
e1751 was observed on Southern blots probed with the
mab-5 containing cosmid ZC102. (R. Hoskins, pers. comm.) We have confirmed this observation and mapped the rearrangement: [See Figure 1] The rearrangement is a large tandem duplication of approximately 100 kb. that includes the entire
mab-5 coding region and about 4 kb. of upstream sequences. As might be expected for a tandem duplication,
e1751 reverts spontaneously to wild-type. These revertants no longer contain the duplication and presumable arise by recombination. We do not believe that the increased number of
mab-5 gene copies explains the gain-of-function phenotype of
e1751 because neither
e1751/nDf16 nor
e1751/mab-5(lf) is wildtype. Instead, the rearrangement may change the activity of one of the promoters. The truncation of one promoter (copy #1) 4 kb upstream of the gene could remove negative control elements or bring new enhancers close to the
mab-5 gene. Some of our earlier genetic experiments suggested to us that
e1751 was not an allele of
mab-5. In these experiments the phenotype of
e1751/nDf16 was compared to the phenotype of
e1751/mab-5
(e1239). Assuming that
e1751 is an allele of
mab-5 one would expect it to look the same over null and over deficiency. Instead, the QR migration in
e1751/nDf16 is fully mutant while
e1751/mab-5
(e1239) has an intermediate phenotype like
e1751/+. We have repeated these experiments using
e2088, another allele of
mab-5 that is likely to be null, and the result is the same. These results imply that something on the wild-type chromosome other than
mab-5 is ameliorating the
e1751 phenotype. We do not know how to interpret these results at present, but have cloned both copies of
mab-5 from
e1751 and plan to test their activities in transgenic lines.