(L)euconostoc mesenteroides B-512FMC, a constitutive mutant for dextransucrase, was grown on glucose, fructose, or sucrose. The amount of cell-associated dextransucrase was about the same for the three sugars at different concentrations (0.6% and 3%). Enzyme produced in glucose medium was adsorbed on Sephadex G-100 and G-200 but much less enzyme was adsorbed when it was produced in sucrose medium. Sephadex adsorption decreased when the glucose-produced enzyme was preincubated with dextrans of molecular size greater than 10 kDa. The release of dextransucrase activity from Sephadex by buffer (20 mM acetate, pH 5.2) was the highest at 28 degrees-30 degrees C. The addition of dextran to the enzyme stimulated dextran synthesis but had very little effect on the temperature or pH stability Dextransucrase purified by ammonium sulfate precipitation, hydroxyapatite chromatography, and Sephadex G-200 adsorption did not contain any carbohydrate, and it synthesized dextran, showing that primers are not necessary to initiate dextran synthesis. The purified enzyme had a molecular size of 184 kDa on SDS-PAGE. On standing at 4 degrees C for 30 days, the native enzyme was dissociated into three inactive proteins of 65, 62, and 57 kDa. However, two protein bands of 63 and 59 kDa were obtained on SDS-PAGE after heal denaturation of the 184-kDa active enzyme at 100 degrees C. The amount of 63-kDa protein was about twice that of 59-kDa protein. The native enzyme is believed to be a trimer of two 63-kDa and one 59-kDa monomers