We originally isolated
elt-3 by RT-PCR with degenerate primers matching the highly conserved GATA-type zinc finger DNA binding domain in an attempt to identify additional GATA factors expressed in the hypodermis. Northern and RT-PCR analysis shows
elt-3 to be expressed most highly in embryos and at a lower level in post-embryonic stages. We have performed a detailed analysis of the
elt-3 expression pattern using a variety of gfp and lac-Z reporter gene constructs and an affinity purified polyclonal antiserum. Expression is first detected at approximately 240 minutes post first cleavage in 8 cells on the posterior dorsal surface of the embryo that correspond to the position of the granddaughters of Cpaa and Caaa. Soon after this, expression is seen all the major hypodermal cells on the dorsal surface of the embryo and in the ventral hypodermal cells. By the comma stage, expression is seen in all dorsal hypodermal cells (
hyp-7), ventral hypodermal cells (
p1/2-
p11/12), the major hypodermal cells of the head (
hyp-4,
hyp-5 and
hyp-6 ) and of the tail (
hyp-8,
hyp-9,
hyp-10 and
hyp11). No expression is seen in the lateral hypodermal cells (seam cells; V1-V6, H0, H1, H2 and T) or the minor hypodermal cells of the head (H1, H2 H3). An apparently constant level of expression is maintained in the hypodermis throughout embryonic development up to the three-fold stage. In three-fold embryos and in post-embryonic stages a much lower level of reporter gene expression is seen and endogenous ELT-3 cannot be detected by the antibody. ELT-3 is localized to the nuclei of all cells in which it is expressed and the onset of expression is immediately after the final embryonic cell division that gives rise to hypodermal cells. Hence we believe
elt-3 is likely to have an important regulatory function in hypodermal cell differentiation and lineage-specific gene expression. We have identified an
elt-3 homologue in C.briggsae suggesting that this gene is of functional importance in Rhabditid nematodes. Although there are two alternatively spliced forms in C.elegans, this may not be critical to
elt-3 function because only one form can be detected in C.briggsae. A number of features of the gene are conserved between C.elegans and C.briggsae including a long 3'UTR containing a highly conserved 100bp stretch of sequence similarity. We are currently trying to isolate a deletion of
elt-3 by screening for an imperfect excision of a Tc1 insertion 3' to the gene. We are also studying
elt-3 function using an ectopic expression approach. Ectopic expression of ELT-3 in early embryos, under the influence of the heat shock promoter, causes embryos to arrest development with approximately 250 350 cells and no visible sign of morphogenesis (thanks to T.Fukushige for the heat shock lines). Preliminary experiments suggest that ectopic expression of ELT-3 may be sufficient to activate expression of
lin-26 and we are investigating a possible role for
elt-3 in the maintenance of
lin-26 expression. Preliminary experiments also suggest that
elt-3 is not expressed in
elt-1 loss of function mutants although ectopic expression of
elt-1 does not seem sufficient to activate
elt-3 expression.