Worm Breeder's Gazette 14(2): 49 (February 1, 1996)

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.

<i>ceh-21</i> is a funny cut class homeobox gene in a cluster

Giuseppe Cassata1, Thomas R. Burglin1, Gary Ruvkun2

1 Biocenter, University of Basel, Switzerland
2 Dept. of Molecular Biology, Harvard Medical School, Boston, MA, USA

We originally sequenced a cDNA with sequence similarity to the cut repeat (cm18e7) that had come out of the genome project. This showed that this cDNA also encodes a homeodomain, although this homeodomain is quite divergent from the homeodomain of the Drosophila cut gene and its vertebrate homologues. Furthermore, a highly conserved W residue was changed so that we could not exclude an error due to reverse transcriptase. Hybridization and restriction mapping showed that this cDNA mapped to a cosmid (BBBC5 on X) that we had detected in our original degenerate oligo screen for homeobox genes. However, the hybridization locus was not in the homeodomain of the cDNA, but within a few kb 3' to that cDNA. Since cm18e7 has only 1 cut repeat, we expected that there should be longer cDNAs, and screen several cDNA libraries using cm18e7 as a probe and detected about 15 clones, which we are analyzing. In the meantime, the genome sequence project has now sequenced cosmid T26C11, which extends more towards the 5' region of ceh-21 compared to BBBC5. The genomic sequence presents now quite a puzzling organization: a cut repeat (A), a homeodomain (A) (predicted ORF T26C11.7), another cut repeat (B), another homeodomain (B) (ORF T26C11.6 = ceh-21), and yet another homeodomain (C) (ORF T26C11.5), the latter predicted by hybridization. The analyzed cDNAs of ceh-21 seem to span only cut repeat and homeodomain B (ORF T26C11.6). Further analysis of more cDNAs is necessary, to elucidate if perhaps alternative splicing plays a role, of if they are polycistronic. A blast search with the cut domain revealed an additional gene on another chromosome on cosmid F22D3 that had a similar structure of one cut repeat and one homeodomain. ceh-21 and T26C11.7 are also quite similar on the nucleotide level, suggesting a recent duplication. All four homeodomains ar! e clearly more similar to each other than to other homeodomains. Furthermore, these homeodomains are also quite different from the Drosophila cut homeodomain. A particular noteworthy difference is the fact that in the most conserved area of helix 3, the 4 new homeodomains have a FF instead of the normal WF (denoted by the star in the Figure). The W residue has so far been absolutely conserved in all homeodomains, this seems to be the first exception. Also the cut repeats are more similar to each other than to the Drosophila cut repeats. All this suggests that the 2 to 4 (?) cut class genes found so far in C. elegans might not be the homologues of Drosophila cut, and that a bona-fide cut homologue remains to be found. Likewise, additional cut class genes of this novel type might be found in flies and vertebrates.
                                       cut repeats
T26C11.7 EELDTQLIAHRILDELKEQCIPQASLAVKVLGRSQGTLSDLLRKPKPWGIM-KNGRGTFQRMANWLDLDPVVRRALCFMKKED
ceh-21   EELDTVDIARRILSELKERCIPQTALAEKILARSQGTLSDLLRMPKPWSVM-KNGRATFQRMSNWLGLDPDVRRALCFLPKED
F22D3    IYIDTKDLCKRIAFELKNHSIPQAIFAERILCRSQGTLSDLLRNPKPWNKL-KSGRETFRRMYNWVAQPLATRLAILDMKTED
cut rp1  QALDTLHISRRVRELLSVHNIGQRLFAKYILGLSQGTVSELLSKPKPWDKLTEKGRDSYRKMHAWACDDNAVMLLKSLIPKKD
cut rp2  NNLNTEDIVRRVKEALSQYSISQRLFGESVLGLSQGSVSDLLARPKPWHMLTQKGREPFIRMKMFLEDENAVHKLVASQYKIA
cut rp3  QDLDTHDITTKIKEALLANNIGQKIFGEAVLGLSQGSVSELLSKPKPWHMLSIKGREPFIRMQLWLSDANNVERLQLLKNERR

                                 homeodomain
                                                         *
T26C11.7  ERIRRFTFTQTQLDSLHTVFQQQDRPNREMQQALSATLKLNRSTVGNFFMNARRRLPKAA
ceh-21    VKVIRLTFTETQLKSLQKSFQQNHRPTREMRQKLSATLELDFSTVGNFFMNSRRRLRIDQ
T26C11.5  PKKTKSPFTEHEIAVMMALFEINKSPNHEEVQKLAVQLNLGYRSVANFFMNKRAKERKKL
F22D3     SKRPRLVFTDIQKRTLQAIFKETQRPSREMQQTIAEHLRLDLSTVANFFMNARRRSRLGG
cut       SKKQRVLFSEEQKEALRLAFALDPYPNVGTIEFLANELGLATRTITNWFHNHRMRLKQQV
Ahox-1    QNSIRTNFTTKQLSELEKEFHFNRYLTRARRVEIAATLELNETQVKIWFQNRRMKQKKRE
smox-3    QRRIRTTFTSLQLKELERAFQETHYPDIYTREDLALRIDLTEARVQVWFQNRRAKFRKTE