Mutations in
unc-84 and
unc-83 cause defects in
hyp7 and P cell nuclear migrations that normally occur as a component of cell migration(1,2).
unc-84 and
unc-83 appear to be the only genes specifically required for these nuclear migrations because over sixty alleles of these genes, but no others, have been identified in various screens that would identify P-cell migration defects. Strong loss-of-function mutations in
unc-84 and
unc-83 cause no defects in other nuclei, indicating that these genes play a role unique to nuclear migration. Thus, the cloning of these genes should allow a direct study of nuclear migration during development. We defined a physical region for
unc-84 and determined that an 8kb subclone of cosmid W01D6 rescued the Unc and Egl phenotypes of
unc-84(
e1410) and
unc-84(
e1174). This subclone hybridizes to transcripts of 3.5 and 2.5 kb, both present in early embryonic and L1 RNA preparations, which is consistent with
unc-84 function in
hyp7 and P cell migration. Using this subclone as a probe, we identified over 30 positive cDNA clones in a screen of approximately 7.5 x 105 plaques of Okkema's embryonic cDNA library. Of the five cDNA clones that have been characterized, three are approximately 3.5kb and two are approximately 2.5kb. The difference between the two classes is an additional 1 kb of sequence on the 3' end of the 3.5kb cDNAs. Fixsen and Horvitz have shown that some alleles of
unc-84 intragenically complement(2). To determine if the genetically separable functions correspond to physical domains of UNC-84, we have sequenced the coding region and intron/exon boundaries of the
unc-84 alleles characterized by Fixsen. A mutation was identified in all 17 alleles sequenced. The mutations clustered into domains consistent with the complementation data. In addition, a severe allele which fails to complement all other alleles has a single mutation that introduces a stop codon in the sequence unique to the long transcript. This suggests that, with regards to the assays of penetrance and complementation, the 2.5 kb transcript has no function. The 3.5kb transcript contains a long open reading frame that encodes a predicted protein of 1111 amino acids. This predicted protein has a C-terminal region that is similar to the C-terminal region of the fission yeast protein Sad1. In addition, both proteins contain a predicted transmembrane domain.
sad1+ is an essential gene whose product associates with the spindle pole body throughout mitotic and meiotic cycles(3). The Sad1 domain appears essential to
unc-84 function because missense mutations in the domain cause severe defects in nuclear migration when homozygous. We have defined a small physical region for
unc-83 and are currently attempting to rescue the gene. [1]Horvitz, H. R. and J. E. Sulston. 1980. Genetics. 96:435-454. [2]Fixsen, W. D. 1985. Ph.D. Thesis. MIT. [3]Hagan, I. and M. Yanagida. 1995. J. Cell Biol. 129:1033-1047.