Correlation between the molecular effects of mutations at the dimer interface of alanine-glyoxylate aminotransferase leading to primary hyperoxaluria type I and the cellular response to vitamin B6.

Primary hyperoxaluria type I (PH1) is a rare disease caused by the deficit of liver alanine-glyoxylate aminotransferase (AGT). AGT prevents oxalate formation by converting peroxisomal glyoxylate to glycine. When the enzyme is deficient, progressive calcium oxalate stones deposit first in the urinary tract and then at the systemic level. Pyridoxal 5'-phosphate (PLP), the AGT coenzyme, exerts a chaperone role by promoting dimerization, as demonstrated by studies at protein and cellular level. Thus, variants showing a destabilized dimeric structure should, in principle, be responsive to vitamin B6, a precursor of PLP. However, models to predict the extent of responsiveness of each variant are missing. We examined the effects of pathogenic interfacial mutations by combining bioinformatic predictions with molecular and cellular studies on selected variants (R36H, G42E, I56N, G63R, and G216R), in both their holo- (i.e., with bound PLP) and apo- (i.e., without bound PLP) form. We found that all variants displayed structural alterations mainly related to the apoform and consisting of an altered tertiary and quaternary structure. G216R also shows a strongly reduced catalytic efficiency. Moreover, all but G216R respond to vitamin B6, as shown by their increased specific activity and expression level in a cellular disease model. A global analysis of data unraveled a possible inverse correlation between the degree of destabilization/misfolding induced by a mutation and the extent of B6 responsiveness. These results provide a first explanation of factors influencing B6 response in PH1, a model possibly valuable for other rare diseases caused by protein deficits.