Reduction of function of the gene
dog-1increases the frequency of spontaneous mutations (1).
dog-1 encodes a DEAH helicase (1). Previous work has shown that
dog-1 mutants have an increased frequency of deletions in some genomic regions that have long poly-guanine stretches (more than 18 Gs) (1). It has been suggested that DOG-1 may be necessary for resolving secondary structures of guanine-rich DNA during DNA synthesis (1). There are many poly G stretches in the C. elegans genome and it may be possible to use
dog-1 to enrich for mutations in genes that contain, or are near, poly G stretches. As current protocols for identifying deletions in genes are labor intensive, any procedure that could increase the frequency of deletions in a specific gene would greatly increase the likelihood of recovering a mutation in this gene. We are interested in studying the spectrum of mutations caused by
dog-1 to determine if it is really specific for poly-G sequences or whether it also induces other types of mutations. To this end we are using two screening procedures. Firstly, we are screening for visible phenotypes. In a small-scale screen we have identified 3 Dpy, 7 Unc, 1 Lin, 2 Vab and 8 Ste mutations. We are continuing this screening, trying to place mutations with similar phenotypes into complementation groups and to map selected genes. Our second approach is to do screens that have a known spectrum of targets because they have been done with other mutagens. In particular we are screening for suppressors of
sel-12. Using EMS only 5 genes are recovered at high frequency in screens for strong
sel-12 suppressors (2,3). Mutations in these same five genes have been identified with UV/TMP and mutator strains (2). One of these genes,
spr-3, has a poly C stretch in its promoter. It has previously been shown that deletions in the
spr-3 genomic sequence can be detected by PCR in a
dog-1 strain (2). However, it was not shown whether these mutations were germline or somatic. So far we have identified four
dog-1 induced spr mutations and found that at least three are alleles of
spr-3. This verifies that
dog-1 can induce multiple mutations in the same gene at high frequency and suggests that
dog-1 may indeed be a very specific type of mutator. We are continuing to screen for additional spr mutations and are trying to determine the exact nature of their mutations. 1) Cheung et al. (2002) Nature Genetics, 31 405-409 2) Lakowski et al, Development (in Press) 3) Wen et al. (2000) PNAS 97, 14524-14529