Cancers develop from cells that inappropriately divide. Understanding the control mechanisms that regulate cell divisions would provide insight into how tumors develop and possible ways of treatment. Although a lot is already known about the biochemistry of cell cycle regulation, little is known about how the regulatory genes function in multicellular organisms. Simply put, the dual specificity phosphatase,
cdc25, activates cyclin dependent kinases (such as
cdc2) which in turn drives the cell cycle machinery. Searching the C. elegans' genome sequence database, numerous cell cycle regulator homologues can be found including cdks and cyclins. From the
cdc25 family, four homologues have been identified which we have named after the cosmid they were sequenced from: ZK637, R05H5, F16B4, K06A5. Why are there so many
cdc25s in the worm while only 3 can be found in mammals, 2 in Drosophila, and only 1 in yeast? Perhaps the C. elegans'
cdc25s act in different cells, or at different developmental stages, or perhaps some may regulate meiosis while others regulate mitosis? The protein sequences of all the C. elegans' CDC25s all share remarkable conservation of the catalytic domain. However, this homology diverges outside the phosphatase domain, the predicted intron/exon structures are distinct and all map to separate chromosomes. We subcloned the
cdc25 genes and reintroduced them back into the worm as multicopy extrachromosomal arrays with no noticeable effect on the animals' development. Characterization of the expression patterns of all these genes is continuing. Our data suggest that only three of the four
cdc25s are expressed genes: RT-PCR, RNA in situ experiments, Northern blot analysis, and GFP and LacZ fusion constructs all show that the R05H5
cdc25 is either expressed weakly or not at all. One (ZK637
cdc25 mRNA pattern) has been reported in an earlier report [WBG 14(2): 76]. An interesting expression pattern has also been characterized for the K06A5
cdc25 by indirect immunohistochemistry. We first saw antibody staining in oocyte nuclei. After fertilization, staining in the polar bodies was absent despite still being present in both pronuclei. After the first mitosis, we also saw staining in the embryonic nuclei as well as plasma membranes. This pattern persisted at least until gastrulation. To perturb expression of each of the
cdc25 genes, we injected antisense RNAs corresponding to all four
cdc25s. While R05H5
cdc25 and ZK637
cdc25 antisense RNA injections, singly or in combination, did not create a noticeable effect, both K06A5 and F16B4
cdc25s produced lethal phenotypes. Antisense K06A5
cdc25 RNA generated aneuploidy in the early embryo resulting from meiotic defects. Using videomicroscopy of live embryos treated with K06A5
cdc25 antisense RNA, enlarged polar bodies, multiple pronuclei and unstable cleavage furrows were often observed. Antisense F16B4
cdc25 RNA caused L1 larval lethality. Interestingly, a few embryos that were laid soon after the antisense K06A5 and F16B4
cdc25 treatment matured into sterile adults. Additional experiments demonstrated redundancy between the genes since antisense RNA injections of ZK637 and F16B4
cdc25 together caused a late embryonic lethality. Furthermore, antisense RNA injections of ZK637, K06A5 and F16B4
cdc25 caused embryos to arrest development at the single cell stage. This was a phenotype similar to that of
cdc2 RNA antisense treatment (van den Heuvel, personal communication and our observations). Candidate lethal mutants for the K06A5 and F16B4
cdc25 are now being evaluated. Weak rescue was obtained for a K06A5
cdc25 candidate,
mei-2, after injection of a cocktail of neighboring cosmids including K06A5 (Srayko & Mains, 1997 International Worm Meeting abstract 565). However, sequencing the
cdc25 gene from homozygous
mei-2 (
ct102) animals showed that K06A5
cdc25 is not
mei-2. We are preparing to screen for deletion alleles of each of these
cdc25s. Research sponsored by the National Cancer Institute, DHHS, under contract with ABL.