Sequencing of gene-rich regions of Caenorhabditis elegans' 100 Mb genome is expected to be completed by the end of 1996 (Consortium report, 1996). Many genes have been predicted, some of which have been implicated by homology in development and disease. Clearly, reverse genetic studies will become increasingly important in C. elegans molecular biology. When a predicted gene of interest is identified, it is first essential to ask whether the gene corresponds to a gene previously identified by mutation, a question that is commonly addressed by DNA microinjection complementation experiments. If a mutation in the gene does not exist, it can be constructed by transposon-generated excision. Unfortunately, these experiments require a fair time investment and a bit of guessing about candidate loci before an answer is reached. We present here that genetic ablation with
mec-4(d) is an efficient means of indicating likely mutant phenotypes of a gene of interest, along with obtaining an idea about its expression pattern, facilitating reverse genetic studies. Dominant mutations in
mec-4 cause degeneration of the six touch neurons. It has been postulated that MEC-4 forms (part of) a multisubunit ion channel, which in aberrant form, causes osmotic imbalance and consequent cell necrosis. Expression of
hsp-16p/mec-4(d) in transgenics induced degeneration of many different cell types all over the worm body, indicating a broad necrotic potency for
mec-4(d). We used regulatory sequences from well characterized worm genes (including
mec-7,
unc-54,
myo-2,
unc-4) to express
mec-4(d) and observed specific ablation or disruption of function of cells in which the toxic allele is expressed. Such a strategy was also employed with
glr-1 by Maricq et al., (Nature 378:78-81). One example of an application of this technique for reverse genetics is our analysis of T28D9_7, a predicted ORF from the genome sequencing project which is also ~34% similar to
mec-4. The putative regulatory sequences of T28D9_7 were fused upstream of
mec-4(d). In integrated transgenics, the ablation fusion killed the same cells in which the T28D9_7/lac-Z fusion was expressed (T28D9_7 shows embryonic and L1 expression). The integrated ablation transgenics were uncoordinated and showed touch abnormalities. In this instance, there were no candidate unc or touch abnormal mutations in LGII to which T28D9_7 maps. In the case of another predicted gene R13A1_4, the ablation phenotype immediately led us to identify
unc-8 as a candidate gene, which was confirmed shortly thereafter (see abstract by Tavernarakis et al.). In conclusion, we find that ablation of cells using
mec-4(d) can greatly aid in determining the potential phenotype of a mutation, in cases where the function of the cell is essential for the expression of the mutant phenotype. This strategy also helps in indentifying the cells expressing the gene since ablation of specific cell types results in pertinent behaviorial abnormalities. This works particularly well for genes expressed in the nervous system. The
mec-4(d) ablation vector, which houses a promoterless cassette and a polylinker to facilitate cloning is available for use.