Human Galactose 1-Phosphate Uridyltransferase

Abstract

Abstract Human galactose 1-phosphate uridyltransferase was purified approximately 1000-fold from post-mortem liver and from erythrocytes. Studies of pH optimum, Km values for UDP-glucose and galactose 1-phosphate, and the effect of dithiothreitol on enzyme activity gave results that do not differ significantly from those obtained with crude enzyme preparations. Molecular weight determinations by Sephadex G-200 gel filtration showed that the enzyme from individuals homozygous for the wild type, Duarte variant, and galactosemic alleles each have the same molecular size in the range of 90,000. Enzyme purified to homogeneity from the erythrocytes of wild type individuals and analyzed by sodium dodecyl sulfate polyacrylamide disc gel electrophoresis revealed a single molecular species of about 30,000 suggesting that the functional enzyme may have a trimeric structure. Rabbit antibody to the human enzyme was produced by immunization with transferase purified to electrophoretic homogeneity from liver. Liver and erythrocyte samples tested against the antibody by double immunodiffusion with the Ouchterlony technique gave precipitins of complete identity. Hemoglobin-free erythrocyte extracts from wild type, Duarte variant, and galactosemic individuals gave precipitin bands of complete identity against the antibody. The same antiserum used for double immunodiffusion experiments was found to precipitate quantitatively transferase activity from solution. The antitransferase precipitation curves for equal amounts of liver and erythrocyte transferase activity were similar. Duarte variant erythrocyte transferase, however, required twice as much antiserum for complete precipitation when compared with the wild type enzyme. Since Duarte variant erythrocytes contain half of the transferase activity found in the wild type, this finding suggests that the Duarte molecule is catalytically less efficient than the wild type. Galactosemic erythrocytes were found to contain an antigen that quantitatively absorbs the same antibody that precipitates the wild type enzyme. These observations support the hypothesis that the three alleles at the transferase locus in man code for immunologically identical proteins of similar molecular size. From this it may be concluded that the less active Duarte variant and catalytically inactive galactosemic proteins are the consequence of small structural gene mutations.