Relationship between Vitamin B6, DNA damage and diabetes

The active form of vitamin B6, the pyridoxal-5’ phosphate (PLP) is a cofactor for more than 150 reactions involved in protein, carbohydrate and lipid metabolism. In addition, it is able to counteract Reactive Oxygen Species (ROS) and Advanced Glycation End products (AGEs). In eucaryotes PLP is produced, in the salvage pathway, by the concerted action of pyridoxal kinase (PDXK) and pyridoxamine/pyridoxine oxidase (PNPO) which recycle PLP precursor from food. PLP has been associated to different pathologies including diabetes and cancer although underlying mechanisms remain in large part still unknown. It has been previously demonstrated that mutations in Drosophila Pdxkgene (dPdxk1) cause diabetes and chromosome aberrations (CABs) and also that these phenotypes are linked by a causeeffect relationship. The first aim of this thesis has been to verify whether also the inactivation of the other gene of the salvage pathway, PNPO, encoded by sgll gene, produced the same phenotypes observed in dPdxk1 mutants. To this purpose we silenced sgll gene by RNA interference and characterized the resulting phenotypes. This analysis revealed a significant frequency of CABs and diabetic phenotypes such as hyperglycemia, small body size, impaired lipid storage and accumulation of AGEs associated to Sgll depletion. These results allowed us to confirm the hypothesis that PLP deficiency produces CABs through the genotoxic effect of AGEs in turn triggered by high glucose. Our second aim has been to investigate whether human PDXK variants present in the population can impact on DNA integrity and can be considered predictive of cancer risk. For this purpose, we expressed four human PDXK variants (carrying missense mutations) into dPdxk1 mutant flies and tested them for CABs as well as for diabetic phenotypes, finding that none of them was able to completely rescue the CAB phenotype, hyperglycemia nor AGE accumulation. Biochemical analysis of these variants revealed a compromised catalytic activity and/or a reduced affinity for their substrates, which explained their “loss of function” behaviour. These results suggested that mutations in PDXK human gene can impact on genome integrity via AGEs and predispose to cancer. Our third purpose was to test whether low PLP levels can impact on cancer in Drosophila. Thus we tested the effects of the PLP inhibitor 4-deoxypyridoxine (4-DP) on Ras and Ras/Scr cancer models generated by mosaic analysis with repressible marker (MARCM) strategy. This analysis showed that 4-DP caused enlargement of primary tumors as well as appearance of secondary tumors in both cancer models. Taken together all the results collected in this work have contributed to confirm and elucidate the relationship between vitamin B6 diabetes and cancer.