Proline dehydrogenase regulation regulation by the P53 family and the regulatiory circuit with HIF-1.

Proline differs from the other amino acids because its á-nitrogen is contained within a pyrrolidine ring. Therefore, it cannot be metabolized by the general transaminases and decarboxylases acting on other amino acids. Proline dehydrogenase (PRODH) is a stress-inducible, key enzyme in proline metabolism, catalyzing its conversion into Ä1-pyrroline-5-carboxylate, a crucial compound interconnecting proline metabolism with glutamate and á- ketoglutarate (á-KG) synthesis and with the Tricarboxylic Acids (TCA) and Urea cycles. Consequently, PRODH can influence various cellular pathways, including glutamatergic transmission, glutathione levels as well as the activity of a number of enzymes using á-KG as a substrate. Proline can also be regarded as an emergency substrate, as abundant stores are released during degradation of intracellular or extracellular matrix proteins (especially collagens). PRODH is localized in the inner membrane of mitochondria and after reduction of the FAD cofactor bound to form the holoenzyme, it can directly transfer electrons to cytochrome C to generate ATP or it can oxidize O2 to generate reactive oxygen species (ROS). Thus when cells are under stress, PRODH has been proposed to act either as a survival factor, favouring maintenance of “survival energy levels”, or as a cell death effector, inducing ROS-dependent apoptosis. Alterations in PRODH protein levels and catalytic activity have been implicated in diseases such as hyperprolinemia, DiGeorge syndrome, schizophrenia and cancer. For cancer in particular, several lines of evidence suggest a central role of PRODH as a mitochondrial tumor suppressor: 1) expression of PRODH is reduced in diverse colorectal and renal cancer cells as compared to normal counterparts; 2) restoration of PRODH expression in human hypo-expressing colon cancer cell lines suppresses their ability to form tumours when injected into SCID mice; 3) PRODH expression is regulated transcriptionally and posttranscriptionally by several cellular sensors of cell health and homeostasis, whose functions are deregulated during carcinogenesis, including p53, PPAR and mTOR (mammalian target of rapamycin). However, the exact mechanisms by which these proteins control PRODH function have been only partially elucidated. Understanding transcriptional and post-transcriptional regulation of a gene and its product is clue to understanding its function. In the first two years of my PhD work we identified and characterized the p53 Response Elements (REs) in the PRODH gene, responsible for p53 binding and transactivation of this target. We confirmed p53-dependent induction of endogenous PRODH in response to genotoxic damage in cell lines of different histological origin and we established that overexpression of p73  or p63  is sufficient to induce PRODH expression in p53-null cells. The p53 family-dependent transcriptional activation of PRODH was linked to specific intronic response elements (REs), among those predicted by bioinformatics tools and experimentally validated by a yeast-based transactivation assay upon modulated expression of p53, p63 and p73 and by p53 occupancy measurements in HCT116 human cells by ChIP. Based on the following pieces of evidence i) it has been proposed that during nutrient stress extracellular matrix (ECM) proteins may be degraded to provide substrates for energy production (ecophagy), ii) an abundant protein in ECM is collagen, that is very rich in proline and hydroxyproline, iii) the key enzyme in hydroxyproline metabolism is hydroxyproline dehydrogenase, homologous to PRODH, whose gene (PRODH2) was also shown, although less convincingly, to be a p53 target, we decided to characterize the p53 REs present in this gene as well. We demonstrated that the PRODH2 gene was not responsive to p63 nor p73 and was at best a weak p53 target, based on minimal levels of PRODH2 transcript induction by genotoxic stress observed only in one of four p53 wildtype cell lines tested. Consistently, all predicted p53 REs in PRODH2 were poor matches to the p53 RE consensus and showed limited responsiveness, only to p53, in the functional assay. Taken together, our results highlight that PRODH but not PRODH2 expression is likely under control of the entire p53 family members, supporting a deeper link between p53 proteins and metabolic pathways, as PRODH functions in modulating the balance of proline and glutamate levels and of their derivative alpha-keto-glutarate in the metabolism under normal and pathological (tumor) conditions. Another important transcription factor that we considered for a possible role in regulation of PRODH, is the Hypoxia Inducible Factor 1 (HIF-1), whose function influences cellular metabolism and is altered during the tumourigenic process. HIF proteins are composed of two subunits,  and , both constitutively expressed in cells. However, the á subunits are rapidly degraded by the proteasome at normal oxygen concentrations found in tissues. Key to HIF- degradation is its oxygen-dependent hydroxylation at specific residues (prolines 402 and 564) by Prolyl Hydroxylases (PHD), that target the protein for ubiquitylation and proteasomal degradation in presence of molecular oxygen, -KG and vitamin C. During hypoxia, HIF- subunits become stabilized, which enables them to form heterodimers with HIF-, that activate numerous cell survival pathways. HIF-1 has been shown to control the expression of more than a hundred genes, either by direct transcriptional activation of protein coding genes and microRNAs (miRs), or by interacting or interfering with other transcription factors. HIF-1 activation results in profound alterations in tumour cell behaviour, which include triggering the angiogenic switch, shifting glucose metabolism towards glycolysis, promoting epithelial-to-mesenchymal transition and acquisition of an invasive phenotype, as well as increasing chemo- and radio-resistance. For this reason, tumour cells often maintain HIF-1 overexpression after they return to a normoxic environment. We tested the hypothesis that an increase in PRODH activity, by increasing -KG, would provide substrate for the hydroxylation reaction catalyzed by PHDs, thus leading to a decrease in HIF-1  levels. indeed, ectopic expression of PRODH led to down-regulation of HIF- and VEGF protein levels in the U87glioblastoma cell line. This finding confirmed what was already reported to occur for colon cancer cell lines. In addition to a role of PRODH in regulating HIF-1 stability in normoxia, we hypothesized that a regulatory circuit between PRODH and HIF-1 could exist. PRODH was found to be downregulated 2-fold in a transcriptomics analysis of genes regulated following induction of focal brain ischemia in rat. On the other hand, however, very recently PRODH was shown to be induced by hypoxia and this induction was AMPK-dependent and HIF-independent. Therefore, the question was still open for investigation. Our expectations are unbiased, because PRODH possesses the ability to promote either cell survival, in conditions in which energy levels are low, by producing ATP or inducing ROS dependent autophagy, or ROS induced apoptotic cell death. Of course a different outcome depending on the cell lines tested as well as on other types of stress acting on the cells concomitantly with the hypoxic stress may be expected. In a first attempt to verify if PRODH transcript levels were modified by hypoxia, we exposed cancer cell lines of different histological origin (HCT116, colon; MCF7, breast; U87MG, glia; SHSY-5Y, neural crest) to 1% hypoxia, anoxia or to treatment with CoCl2, a PHDs inhibitor, and compared the levels of expression with those obtained in the same untreated cell lines, by using real time RTqPCR. All cell lines showed a marked decrease in PRODH transcript, in particular after treatment with CoCl2, and a reduction also in protein levels, although of minor entity compared to transcript decrease. Preliminary results obtained during my PhD work confirm that PRODH-HIF-1 regulatory circuit does indeed exist, and lays the foundations for further investigations, to clarify the relationship between these two proteins to increase knowledge about PRODH regulation and its possible downregulation during the tumourigenic process