Finally we report that the sumoylation of c-Fos is a dynamic process that can be reversed via multiple mechanisms. This supports the idea that this modification does not constitute a final inactivation step that necessarily precedes protein degradation.

Abstract. Several lines of evidence suggest that hepatocyte growth factor/scatter factor (HGF/SF), a soluble protein secreted by embryo fibroblasts and several fibroblast lines, may elicit morphogenesis in adjacent epithelial cells. We investigated the role of HGF/SF and its membrane receptor, the product of the c-met protooncogene, in the early development of the metanephric kidney. At the inception of the mouse metanephros at embryonic day 11, HGF/SF was expressed in the mesenchyme, while met was expressed in both the ureteric bud and the mesenchyme, as assessed by reverse transcription PCR, in situ hybridization, and immnuohistochemistry. To further investigate the expression of met in renal mesenchyme, we isolated 13 conditionally immortal clonal cell lines from transgenic mice expressing a temperature-sensitive mutant of the SV-40 large T antigen. Five had the HGF/ SF+/met + phenotype and eight had the HGF/SF-/met + phenotype. None had the HGF/SF+/met -nor the HGF/SF-/met-phenotypes. Thus the renal mesenchyme contains cells that express HGF/SF and met or met alone. When metanephric rudiments were grown in serum-free organ culture, anti-HGF/SF antibodies (a) inhibited the differentiation of metanephric mesenchymal cells into the epithelial precursors of the nephron; (b) increased cell death within the renal mesenchyme; and (c) perturbed branching morphogenesis of the ureteric bud. These data provide the first demonstration for coexpression of the HGF/SF and met genes in mesenchymal cells during embryonic development and also imply an autocrine and/or paracrine role for HGF/SF and met in the survival of the renal mesenchyme and in the mesenchymal-epithelial transition that occurs during nephrogenesis. They also confirm the postulated paracrine role of HGF/SF in the branching of the ureteric bud.

c-fos gene is expressed constitutively in a number of tissues as well as in certain tumor cells and is inducible, in general rapidly and transiently, in virtually all other cell types by a variety of stimuli. Its protein product, c-Fos, is a short-lived transcription factor that heterodimerizes with various protein partners within the AP-1 transcription complex via leucine zipper/leucine zipper interactions for binding to specific DNA sequences. It is mostly, if not exclusively, degraded by the proteasome. To localize the determinant(s) responsible for its instability, we have conducted a genetic analysis in which the half-lives of cFos mutants and chimeras made with the stable EGFP reporter protein were compared under two experimental conditions taken as example of continous and inducible expression. Those were constitutive expression in asynchronously growing Balb/C 3T3 mouse embryo fibroblasts and transient induction in the same cells undergoing the G0/G1 phase transition upon stimulation by serum. Our work shows that c-Fos is degraded faster in synchronousthan in asynchronous cells. This difference in turnover is primarily accounted for by several mechanisms. First, in asynchronous cells, a unique C-terminal destabilizer is active whereas, in serum-stimulated cells two destabilizers located at both extremities of the protein are functional. Second, heterodimerization and/or binding to DNA accelerates protein degradation only during the G0/G1 phase transition. Adding another level of complexity to turnover control, phosphorylation at serines 362 and 374, which are c-Fos phosphorylation sites largely modified during the G0/G1 phase transition, stabilizes c-Fos much more efficiently in asynchronous than in serum-stimulated cells. In both cases, the reduced degradation rate is due to inhibition of the activity of the C-terminal destabilizer. However, in serum-stimulated cells, this effect is partially masked by the activation of the N-terminal destabilizer and basic domain/leucine zipper-dependent mechanisms. Taken together, our data show that multiple degradation mechanisms, differing according to the conditions of expression, may operate on c-Fos to ensure a proper level and/or timing of expression. Moreover, they also indicate that the half-life of c-Fos during the G0/G1 phase transition is determined by a delicate balance between opposing stabilizing and destabilizing mechanisms operating at the same time.