Titration experiments were employed to measure the binding stoichiometry of alpha 2M for trypsin at high and low concentrations of reactants. These titration experiments were performed by measuring the SBTI-resistant trypsin activity and by direct binding measurements using 125I-labeled trypsin. The binding stoichiometry displayed a marked dependence upon protein concentration. At high alpha 2M concentrations (micromolar), 2 mol of trypsin are bound/mol of inhibitor. However, at low alpha 2M concentrations (e.g., 0.5 nM), only 1.3 mol of trypsin were bound/mol of inhibitor. Sequential additions of subsaturating amounts of trypsin to a single aliquot of alpha 2M also resulted in a reduction in the final binding ratio. A model has been formulated to account for these observations. A key element of this model is the observation that purified 1:1 alpha 2M-proteinase complexes are not capable of binding a full mole of additional proteinase [Strickland et al. (1988) Biochemistry 27, 1458-1466]. The model predicts that once the 1:1 alpha 2M-proteinase complex forms, this species undergoes a time-dependent conformational rearrangement to yield a complex with greatly reduced proteinase binding ability. According to this model, the ability of alpha 2M to bind 2 mol of proteinase depends upon the association rate of the second enzyme molecule with the binary (1:1) complex, the enzyme concentration, and the rate of the conformational alteration that occurs once the initial complex forms. Modeling experiments suggest that the magnitude of the rate constant for this conformational change is in the order of 1-2 s-1.