Modeling of continuously and directly analyzed biphasic reaction courses of ribulose 1,5-bisphosphate carboxylase/oxygenase.

This paper aims at clarifying the cause of the time-dependent, partial loss of the activity during reaction (so-called fallover) of ribulose 1,5-bisphosphate carboxylase/oxygenase (RuBisCO) from plant sources. This was done by comparing the reaction courses calculated using the reaction models constructed here based on the present conflicting two ideas on fallover with directly measured courses obtained with RuBisCO purified from spinach leaves. Since the ordinary methods with 14CO2 and indicator enzymes were not adequate for analyzing the progress of fallover, we followed the reaction by measuring the change of the light absorbance of ribulose 1,5-bisphosphate (RuBP) at 280 nm. Direct measurements of the reaction course showed that RuBisCO lost its activity with a rate constant of 6.1 to 6.5 x 10(-3) s-1 at both 0.5 and 2 mM RuBP. The rate constant of the recovery of the enzyme to show the original fallover was determined as 1.2 to 1.3 x 10(-3) s-1 with RuBisCO that had just experienced fallover. These constants were used in the models. Calculation with a model assuming the binding of xylulose 1,5-bisphosphate (XuBP) to the catalytic sites of the enzyme as the cause of fallover and using the reported dissociation constant of XuBP in the binding and the reported rate of the formation of XuBP from RuBP gave a rather linear reaction course. The minimum requirements for the model to be valid were that the rate of XuBP formation was more than once for every 600 turnovers, the dissociation constant of XuBP for the catalytic sites was less than 0.1 nM, and the binding of XuBP to the sites showed a strong negative cooperativity. Inclusion of non-catalytic RuBP-binding sites in the model was essential to elucidate the course at higher RuBP concentrations. The model constructed assuming that hysteresis was the cause of fallover could calculate the measured reaction courses for the initial 20 min of reaction of both 0.5 and 2mM RuBP. The rate constants of the hysteretic conformational changes of the predicted enzyme forms to others were given. The direct measurement of the long-term reaction course revealed the two phases in the decay of the activity; fast decay for the initial several minutes and subsequent slow decrease. Although the fast decay could be predicted by the hysteresis model, the slower one required the participation of inhibition by XuBP. We reasoned from these results and the reported characteristics of the binding of other sugar phosphates to the catalytic sites that the initial fast decay of the activity in fallover was due to the hysteretic property of the enzyme and the slower phase of fallover was due to the inhibition of XuBP.