A Small Update on TGF-beta Signalling Cathy Savage, Scott Townsend, Scott Bairdl, and Richard Padgett. Waksman Institute and Department of Molecular Biology and Biochemistry, Rutgers University, Piscataway, NJ 08855. University of Pittsburgh, Department of Biological Sciences, Pittsburgh, PA 15260. We are interested in characterizing the genes
sma-2,
sma-3, and
sma-4 because they may participate in a TGF-beta signal transduction pathway that includes
daf-4, a TGF-beta superfamily receptor. All four of these genes are required for proper body size, and for the proper development of male sensory rays and spicules. We have used mosaic analysis to determine the requirements for
daf-4 and
sma-2 in establishing sensory ray identity.
daf-4 is required cell autonomously in each ray neuroblast to determine the identity of the corresponding ray. This result is consistent with temperature shift experiments which showed that
daf4 activity is required during the lifetime of the ray neuroblasts (WBG 13(2), p.39). Similarly,
sma-2 mosaics reveal a cell autonomous requirement for
sma-2 activity in the ray neuroblasts. This result implies two conclusions: (1) the
sma-2 gene product does not act as a diffusible factor influencing the identities of multiple neuroblasts, and (2)
sma-2 does not act in neighboring cells that may signal to the neuroblasts. The requirement for
sma-2 and
daf-4 activities in the same cells suggests that
sma-2 may be necessary to transduce the signal received by the
daf-4 receptor. Similar mosaic studies of
sma-3 and
sma-4 are ongoing. In the last gazette, we reported preliminary evidence of the cloning of
sma-2: the identification of a missense mutation in
sma-2(
e502). This result has now been confirmed by the presence of a small deletion in a nearby exon in sma- 2
(el491). The predicted
sma-2 product is a novel protein of about 400 amino acids. SMA-2 does not contain any significant hydrophobic stretches, suggesting that the protein is not secreted; this is consistent with its cell autonomous requirement in ray neuroblasts. Since the protein contains no recognizable protein motifs, we do not know what its biochemical function may be. Although SMA-2 is not related to any other proteins in the database, the genome sequencing consortium has now identified two additional predicted ORFs with homology to
sma-2 These proteins contain two conserved domains, an N-terminal region of ~120 residues and a C-terminal region of ~200 residues, that are separated by a less well conserved proline-rich domain. The first
sma-2 homolog that was found maps near the predicted location of
sma-3. We are therefore testing the hypothesis that
sma-3 encodes this homolog. Finally, the most recently identified
sma-2 homolog maps to a cosmid overlapping one which we had already shown could rescue a
sma-4 mutant. Deletions that disrupt the coding region of this homolog also eliminate rescuing activity, showing that
sma-4 does in fact encode a
sma-2 homolog. These results suggest the intriguing hypothesis that
sma-2,
sma-3, and sma- 4 may define a novel family of signalling proteins that act downstream of a TGF-beta superfamily receptor.