We are interested in how individual cells decide to activate the cell death program during development. To analyze this process we, in collaboration with other members of the laboratory, have been studying two genes thought to be involved in this process,
ces-1 and
ces-2(1). As reported at the last worm meeting(2), both
ces-1 and
ces-2 encode putative transcription factors, a zinc-finger protein similar to members of the family defined by the snail gene of Drosophila in the case of
ces-1, and a protein of the bZip family in the case of
ces-2. These findings suggest that programmed cell death, like many other cell fates, is regulated at the level of differential gene transcription. There exist three dominant, gain-of-function alleles of
ces-1,
n703,
n1895 and
n18961, which behave similarly, causing survival of four pharyngeal cells that normally undergo programmed cell death. To identify the molecular lesions in these alleles, we determined the sequences of
ces-1 genomic DNA. For all three alleles the coding sequence, 3' UTR, and all introns examined (four of five) were identical to wild-type. We then used a mutation-detection technique (Chemical Cleavage of Mismatch; thanks to Giovanni Lesa and Paul Sternberg for their version of this protocol) to scan larger regions around the
ces-1 coding sequence. We identified a lesion 5' to the start of the coding sequence in the allele
n703. Subsequent studies of this region in all three gain-of-function alleles showed that they contain identical changes: a G-to-A transition ~600 bp before the translation start site. The most interesting feature about this change is that it alters the sequence TAGGTA (mutated sequence TAGATA), which contains the core of the binding consensus of the Snail protein, CAGGTG(3), suggesting that
ces-1 might regulate its own transcription. This hypothesis is consistent with
ces-1 genetics, since snail family members have been shown to function primarily as repressors and a mutation in regulatory DNA that leads to a gain-of-function results from disrupted negative regulatory sites. To determine whether this region is involved in
ces-1 regulation, we have been conducting two sets of experiments. First, we have cloned the C. briggsae homolog of
ces-1 (thanks to David Baillie for his library) and are comparing the regions upstream of coding sequences for conservation. We have found that the upstream regions between the two species are entirely unconserved except for a 130 bp region that is 82% identical (107/130 bases). This region surrounds the site of the gain-of-function mutations. This comparison has also revealed that there are five sites in this region close or identical to snail consensus binding sites (see figure). Second, we are using bacterially produced CES-1 protein in electrophoretic mobility-shift assays to determine the binding-site specificity of CES-1. Preliminary results show that CES-1 is capable of binding to snail consensus binding sites. We hope to use this assay in the future to determine the in vivo targets of
ces-1 and whether any of the ced genes are transcriptionally regulated by CES-1 to control programmed cell death. {See WBG for Figure.} Figure: Alignment between the regions of C. elegans and C. briggsae DNA around the site of the
ces-1 gain-of-function mutations. Shaded residues indicate identities. Boxed regions conform or are close to snail consensus binding sites. The arrow indicates the base altered in all three
ces-1 gain-of-function mutations. 1. Ellis, R.E. & Horvitz, H.R. Development 112, 591-603 (1991). 2. Metzstein, M.M., Tsung, N., Hengartner, M.O., Ellis, R.E. & Horvitz, H.R. 10th International C. elegans Meeting,
p374 (1995) . 3. Mauhin, V., Lutz, Y., Dennefeld, C. & Alberga, A. Nucleic Acids Res 21, 3951-7 (1993).