The CRAL/TRIO and GOLD domain protein CGR-1 promotes induction of vulval cell fates in Caenorhabditis elegans and interacts genetically with the Ras signaling pathway

Abstract

Ras-mediated signaling is necessary for the induction of vulval cell fates during Caenorhabditis elegans development. We identified cgr-1 by screening for suppressors of the ectopic vulval cell fates caused by a gain-of-function mutation of the let-60 ras gene. Analysis of two cgr-1 loss-of-function mutations indicates that cgr-1 positively regulates induction of vulval cell fates. cgr-1 is likely to function at a step in the Ras signaling pathway that is downstream of let-60, which encodes Ras, and upstream of lin-1, which encodes a transcription factor, if these genes function in a linear signaling pathway. These genetic studies are also consistent with the model that cgr-1 functions in a parallel pathway that promotes vulval cell fates. Localized expression studies suggest that cgr-1 functions cell autonomously to affect vulval cell fates. cgr-1 also functions early in development, since cgr-1 is necessary for larval viability. CGR-1 contains a CRAL/TRIO domain likely to bind a small hydrophobic ligand and a GOLD domain that may mediate interactions with proteins. A bioinformatic analysis revealed that there is a conserved family of CRAL/TRIO and GOLD domain-containing proteins that includes members from vertebrates and Drosophila. The analysis of cgr-1 identifies a novel in vivo function for a member of this family and a potential new regulator of Ras-mediated signaling.

Figure 1.

Cloning cgr-1. (A) A portion of the physical map on LGX. Polymorphisms are differences between the nucleotide sequences of CB4856 and wild-type N2 and are designated amPX. A total of 130 Lon non-Unc and 69 Unc non-Lon progeny were selected from let-60(n1046gf); unc-78 n2528 lon-2 / CB4856 hermaphrodites. We determined the interval that includes the recombination event by scoring polymorphisms and the presence of n2528 by scoring suppression of the let-60(n1046gf) Muv phenotype. These data positioned n2528 between amP97 and amP94. This interval is expanded below; exons (boxes) and introns (thick lines) of Genefinder predicted open reading frames are shown. The G-to-A change in n2528 animals is shown. (B) Plasmids containing wild-type (pJG1) or mutant T27A10.7 (pJG2, pJG3) were used to generate transgenic strains. The chromosomal genotype was let-60(n1046); n2528. Each rescue result represents an independently derived transgenic strain. We defined full rescue as >80% Muv, partial rescue as 40–80% Muv, and no rescue as <40% Muv.

Figure 2.

cgr-1 encodes a protein with CRAL/TRIO and GOLD domains. (A) Full-length C. elegans CGR-1 protein (Ce CGR-1) is aligned with the predicted full-length C. briggsae CGR-1 protein (Cb CGR-1), with amino acids 239–715 of the predicted human SEC14-1 protein (Hs SEC14-1; Chinen et al. 1996), and with the full-length predicted human SPF protein (Hs SPF; Shibata et al. 2001). Amino acid numbers refer to Ce CGR-1. Shaded residues are identical to Ce CGR-1. The CRAL/TRIO domain is boxed with a solid line and the GOLD domain is boxed with a hatched line. Triangles indicate intron positions. Arrows indicate the position of the cgr-1(n2528) mutation (P131L) and the coding region removed by the cgr-1(am114) deletion (amino acids 121–334). (B) Diagram of the cgr-1 locus. Boxes represent exons. Shaded boxes and the solid box show coding and untranslated regions, respectively. The thin line represents introns. An SL1 leader sequence is spliced 2 nucleotides upstream of the start codon. A poly(A) tail is attached 145 nucleotides downstream of the stop codon. Arrows indicate the position of the n2528 mutation and the endpoints of the am114 deletion. (C) Neighbor-Joining tree illustrating phylogenetic relationships of predicted CRAL/TRIO and GOLD domain-containing proteins. Boldface type indicates proteins that have been biochemically or genetically characterized. Branch lengths are proportional to divergence (scale represents 10% divergence). Numbers at each node indicate bootstrap support out of 1000 replicates. All nodes were well supported. Mm, mus musculus; Dm, Drosophila melanogaster.

Figure 3.

Expression pattern of CGR-1. The genotype of all hermaphrodites was let-60(n1046); cgr-1(n2528); amEx82. The amEx82 array contains pDG125 (Pcgr-1:cgr-1 genomic:GFP) and pRF4. (A and B) Nomarski and fluorescent images (100×) of adult hermaphrodites. CGR-1:GFP expression is visible in intestinal cells. (C and D) Nomarski and fluorescent images (400×) of an L4 hermaphrodite. The arrow marks the vulva. The asterisk marks an ectopic vulval invagination. CGR-1:GFP expression is visible in intestinal cells but not in vulval cells. (E and F) Nomarski and fluorescent images (630×) of adult intestinal cells. CGR-1:GFP exhibits cytoplasmic localization and appears to be largely excluded from nuclei.

Figure 4.

Structure/function analysis of CGR-1. Transgenic strains that expressed the diagrammed CGR-1 protein and had the chromosomal genotype let-60(n1046); cgr-1(n2528) were assayed for rescue of the cgr-1 suppression of Muv phenotype. Extrachromosomal arrays are described in Table 4 and criteria for rescue are described in Figure 1.