The
lin-26 gene, which codes for a zinc-finger transcription factor, is involved in the maintenance of all epithelial cell fates in the ectoderm and mesoderm (the somatic gonad) and can induce epithelial markers when overexpressed. In order to place this gene in a differentiation pathway and to look at upstream events, we decided to perform a promoter study. By comparing C. elegans
lin-26 cis-sequences to those of C. briggsae , we identified seven conserved regions. To test their enhancer activity, we cloned promoter fragments in front of a minimal promoter linked to GFP. Moreover we also tested their role by looking at the expression pattern of deleted
lin-26 ::GFP fusion constructs, and at the rescuing activity of deleted constructs in a
lin-26 null background. To date four major cis-elements (from 5' to 3') have been indentified that drive the expression in specific cells: A rep cell element (3.5 kb) : Only defined by enhancer properties, turns on GFP in the rectal rep cells. The expression of
lin-26 in these cells is not essential for viability. A gonad element (445 bp) : This element drives GFP expression in Z1 and Z4, the precursors of the somatic gonad in which
lin-26 is normally expressed, and later in the uterus. In addition, when the gonad element is deleted, animals are sterile and show gonad defects. A glial cell and minor hypodermal cell element (530bp) : This element drives the GFP in about 20 cells in the head (mostly glial-like cells and few cells belonging to the minor hypodermis), 6 cells in the tail (PHshL/R, Hyp8,9,10), and also in the excretory cell. Expression in the head and in the tail is driven separately by two constructs of about 270 bp, which overlap by 130 bp. Because of the division of the pattern into two different regions, we tested the possible regulation of the 'tail element' by two Hox genes belonging to the posterior-group,
nob-1 and
php-3 , by silencing these genes using RNAi techniques. The number of GFP positive cells remained unchanged in RNAi animals, carrying the integrated 'tail element'. A hypodermal element (2.1 kb) : Sufficient to obtain a GFP expression in cells of the major hypodermis, but not necessary for
lin-26 expression in that tissue. This suggests the existence of a functionally redundant regulator for the expression in the hypodermis, located either upstream, or in the 3'UTR. The enhancer activity of a larger fragment (4 kb, comprising this 2.1kb element) requires the activity of
elt-1, since in
elt-1 (
zu180) mutants the number of GFP-expressing cells is dramatically reduced. Moreover, three GATA sites perfectly conserved between C. briggsae and C. elegans are found in that fragment, suggesting a direct regulation which we will investigate. The 530 bp element, in both orientations, is also able to act synergistically with sequences highly conserved in C.briggsae (contained within the 4 kb hypodermal element) which are not able to drive any expression on their own, thus giving rise to new expression patterns including rectal cells (K, K', U, B, F and Y), and the P cells. These data suggest that
lin-26 expression is controlled by redundant elements in a tissue-dependent manner rather than by lineage-dependent mechanisms. We are now focusing on regulators of
lin-26 expression in different tissues, starting with the gonad. A sequence of 100 bp, conserved between C. briggsae and C. elegans , was analysed by the MatInspector software. We selected four transcription factors (Sox, GKLF, Creb, GFI-1) which have homologues in C. elegans . RNAi techniques are currently being used in order to test the influence of these candidates in an integrated line carrying a
lin-26 gonad specific enhancer coupled to GFP. Up to date six candidates have been tested separatly (C32E12.3, F40E10.2: Sox homologues, Y38H8A.5, Y55F3AM.14, C55B7.12: GFI-1 homologues, and the GKLF homologue F54H5.4) without having any influence on the reporter expression. This work, including the promoter study, is still in progress.