1) E4F1: a new key regulator of pyruvate metabolism
One of our main research programs aims at understanding the role of E4F1, a multifunctional protein that was initially discovered as a transcription factor targeted by viral oncoproteins and that we later on identified as an atypical ubiquitin E3 ligase for p53. Our detailed analysis of E4F1 functions in vitro and in vivo showed that E4F1 controls metabolism through both p53- dependent and independent mechanisms. Although E4F1 was initially characterized as a transcription factor regulating the E4 promoter of the adenovirus serotype V, its cellular target genes remained poorly characterized until recently. In close collaboration with Dr. C. Sardet's group (IRCM, Montpellier), we identified the direct transcriptional targets of E4F1 at the whole genome level using ChIP-seq and gene expression profiling of E4F1 proficient/deficient cells. Unexpectedly, these analyses showed that E4F1 controls the transcription of several genes (Dlat, Dld, Mpc1, Slc25a19, Pdpr) related to pyruvate metabolism, including genes encoding key subunits and regulators of the pyruvate dehydrogenase (PDH) complex (PDC)20-22, a mitochondrial multiprotein complex that converts pyruvate into Acetyl-Coenzyme A to fuel the Krebs cycle. Moreover, E4f1 inactivation in many different cell types results in impaired PDH activity and leads to the metabolic reprogramming of E4F1-deficient cells characterized by impaired pyruvate decarboxylation and the redirection of the glycolytic flux towards lactate production. Consistent with our findings, a non-synonymous homozygote mutation (K144Q) in the coding region of the human E4F1 gene was recently identified in 2 patients presenting clinical symptoms reminiscent of the Leigh syndrome, an inborn metabolic disorder commonly associated with PDC deficiency. These data indicate that E4F1 is a new key regulator of pyruvate metabolism.
Over the past 15 years, we generated several whole-body and tissue-specific E4f1 conditional KO (cKO) mice to address the impact of E4F1/PDH deficiency in different cell types, including in skeletal muscle cells, adipocytes, hepatocytes, neurons and keratinocytes. Because some E4f1 cKO mice display phenotypes that recapitulate the clinical symptoms of Leigh syndrome patients (chronic lactate acidemia, exercise intolerance, and neuronal defects), we believe these mice are interesting animal models to further understand the molecular defects underlying these symptoms. Using these different models, we are currently using several MS-based proteomic approaches to further characterize the molecular consequences of defective glucose-derived AcCoA production (acetylome) and alterations of the redox status (oxydome) of E4F1/PDH-deficient cells. Interestingly, using skin-specific E4f1 cKO mice, we also unraveled an expected function for E4F1-mediated control of the PDC in epidermal stem cell maintenance and keratinocyte differentiation, linking for the first time pyruvate metabolism to proper skin homeostasis.
2) MDM2 controls amino acid metabolism and mitochondrial respiration
Through its E3 ligase activity, MDM2 is one of the main negative regulators of p53. However, an increasing body of evidence indicates that MDM2 oncogenic activities also rely on p53-independent functions. We recently identified a previously undescribed function of MDM2 in chromatin that controls cancer cell metabolism independently of p53. Briefly, we found by ChIP-seq that chromatin-bound MDM2 controls the transcription of genes implicated in serine/glycine metabolism and redox homeostasis in cancer cells. Furthermore, we also demonstrated that a fraction of the MDM2 protein localizes to the mitochondrial matrix to regulate the transcription of the mitochondrial genome and control mitochondrial respiration both in normal and cancer cells. Finally, our data also indicate that the implication of MDM2 in serine/glycine metabolism contributes to its pro-oncogenic activities in some cancer types. Our findings have important clinical perspectives in particular for cancer patients harboring MDM2 amplification. Altogether, our data highlight a new p53-independent function for the MDM2 oncoprotein in metabolism.