Two Forms of the DNA Ligase of Human Cells.

We have further characterized the polynucleotide ligase purified from cultures of the human heteroploid line EUE [Spadari, Ciarrocchi and Falaschi, Eur. J. Biochem. 22, 75 (1971)]. The 350-fold purified enzyme gives positive response with four different assays; the rates with the different substrates are different, but the variations are identical to those observed with the purified ligase induced by T4-phage. The enzyme can reconstitute the transforming activity of Bacillus subtilis DNA inactivated by pancreatic DNAase. The human cell enzyme is unable to use a hybrid substrate where an interrupted polydeoxynucleotide is annealed to a polyribonucleotide, whereas in the same conditions the T4 enzyme gives appreciable activity. The partial dependence of the purified enzyme on proteins present in a boiled crude extract, reported in the previous paper, seems due to an aspecific protective effect of the soluble proteins of cell extracts. The enzyme does not show any appreciable sequence specificity in the phosphodiester bond it can form. The purified enzyme can be fractionated into two molecular forms, one having a molecular weight of 190000, the other 95000; fresh extracts of EUE cells contain almost exclusively the high molecular-weight form; ageing of the extract or purification lead to the appearance of the second peak, without variations in the total activity. This could correspond to the conversion of a dimer structure into a monomer. The “dimer” has a pH optimum close to 8.1, whereas the “monomer” has its optimum at 7.5; this explains the bimodal pH curve previously described in the purified enzyme. [ABSTRACT FROM AUTHOR]