Worm Breeder's Gazette 10(2): 108

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.

A Putative mab-5 Deletion Disrupts a Gene Containing a Homeobox

Michael Costa and Cynthia Kenyon

Figure 1

We are cloning the mab-5 gene to study the molecular mechanism by 
which patterns of cell types arise along the anteroposterior body axis 
of C.  elegans after hatching.  As a brief review, mab-5 is required 
to give cells located in the posterior body region their normal 
identities.  This gene influences the fates of epidermal, neural, and 
mesodermal cells in a cell-autonomous fashion.  It affects sequences 
of cell division, cell death, fates of specific neurons, production of 
mating structures and sensilla, cell fusion, and also directions of 
cell migration.  The simplest way to explain most of the differences 
between mab-5 mutants and the wild type is to postulate that this gene 
allows posterior cells to develop differently than they would if 
located further anteriorly.  Differences between mab-5 and wild type 
males and hermaphrodites also suggest that mab-5 interacts with 
elements of the sex determination pathway to generate much sexual 
dimorphism in the posterior body region.
We identified and cloned a closely linked Tc1 element in the strain 
N62.  We obtained cosmids in this region from John Sulston and Alan 
Coulson (thanks!) and used them to probe Southern blots of mab-5 
mutants.  In the mab-5 (e2088) mutant, we detected a deletion of about 
150 bp located approximately one cosmid length, or 35 Kb, from the Tc1 
element.  To test whether this deletion could be in the mab-5 gene, we 
carried out high resolution mapping experiments.  We found that the 
Mab-5 phenotype and the deletion always cosegregated in 102 
recombinants within the 0.5 mu sma-3 to unc-36 interval.  Since we 
examined 38 crossovers in the approximately 315 Kb interval between 
mab-5 and unc-36 (A.  Coulson, pers.  comm.), we expect about one 
crossover per 8 Kb.  Thus it seems quite likely that this deletion is 
the mab-5 mutation.
We have sequenced wild type genomic DNA in this region and 
determined that part of a large open reading frame is deleted in e2088.
By Northern analysis, we found that this open reading frame is part 
of a polyadenylated RNA present in larvae (we have yet to look at 
other developmental stages).  Currently, we are doing Northern 
analysis of the mab-5 mutants and transformation experiments to 
provide further evidence that we have cloned mab-5.  We have sequenced 
partial cDNAs containing this open reading frame isolated from a 
library kindly given to us by C.  Thacker and M.  Capecchi.  By 
comparing the sequence to known protein sequences, we have found that 
the open reading frame contains a homeobox highly homologous to those 
of Drosophila pattern formation genes.  This putative mab-5 
homeodomain is most similar to that of the Antennapedia gene.  Of the 
60 amino acids in the mab-5 homeodomain, 44 are identical to those of 
Antennapedia, and only 4 of the substitutions are not conservative.  
The figure below shows these homeodomains using dashes to represent 
amino acids identical to those of mab-5 with conservative 
substitutions underlined.  Assuming that this sequence corresponds to 
mab-5, this similarity indicates that homeodomains are used to 
generate global aspects of body pattern in animals other than 
Drosophila.  At a more mechanistic level, since homeodomain proteins 
are thought to be transcriptional regulators, it also suggests that 
mab-5 exerts its many influences on cell division, differentiation, 
and migration by regulating gene expression.
[See Figure 1]

Figure 1