The kinetics of the forward ATP sulfurylase-catalyzed reaction were examined using a new assay based on 32PPi released from [gamma-32P]MgATP in the presence of inorganic sulfate. Replots yielded Vmaxf = 6.6 units mg protein-1, KmA = 0.13 mM, Kia = 0.33 mM, and KmB = 0.55 mM, where A = MgATP and B = SO2-4. Thiosulfate, a dead-end inhibitor of the reaction, was competitive with sulfate and noncompetitive with respect to MgATP. The ratio kcat/KmA was determined for several alternative inorganic substrates, B, where A = MgATP and B = SO2-4, SeO2-4, MoO2-4, WO2-4, or CrO2-4. For SO2-4 and SeO2-4, the ratio was 5-6.5 X 10(4) M-1 S-1; for the others, the ratio was 5.8-7.3 X 10(5) M-1 S-1. The results support a random addition of MgATP and inorganic substrate. The kinetics of the reverse reaction were examined using a new assay based on 35SO2-4 release from [35S]APS (adenosine 5'-phosphosulfate) in the presence of MgPPi. Reciprocal plots were linear, intersecting below the horizontal axis. Replots yielded Vmaxr = 50 units mg protein-1, KmQ = 0.3 microM, Kiq = 0.04 microM, and KmP = 4 microM, where Q = APS and P = PPi (total of all species). MgATP and SO2-4 were both competitive with APS and noncompetitive with respect to MgPPi. Taken together with earlier results suggesting that APS is competitive with both MgATP and SO2-4 and that MgPPi is noncompetitive with respect to both substrates, the qualitative results point to a random A-B, ordered P-Q kinetic mechanism. The Scatchard plot for [35S]APS binding was curved, indicating either negative cooperativity or more than a single class of sites. [gamma-32P]MgATP displayed half-site saturation in the presence of saturating FSO-3.