A statistical approach called response surface methodology (RSM) is used for the prediction of the kinetic constants of glucose oxidase (GOx) as a function of reaction temperature and pH. Lineweaver-Burk transformation of the Michaelis-Menten equation was utilized as the integral part of the RSM algorithm. The effects of variables, namely reciprocal of substrate concentration (0.033-0.5 mM(-1)), reaction temperature (14.9-40.1 degrees C) and reaction pH (pH 4.4-8.5) on the reciprocal of initial reaction rate were evaluated and a second order polynomial model was fitted by a central composite circumscribed design (CCCD). It was observed that optimum reaction temperature and pH for the GOx reaction depended on the substrate concentration and varied between 27.8 degrees C and 6.4 pH and 32.7 degrees C and 6.1 pH in the investigated range of substrate concentration. The maximum reaction rate (V-max) and Michaelis-Menten constant (K,,) of GOx were obtained for each reaction parameter by using the model equation. The maximum reaction rate varied between 3.5 mu mol/min mg enzyme and 29.8 mu mol/min mg enzyme. Michaelis-Menten constant was determined between 1.9 mM and 16.8 mM in the tested reaction parameters. The kinetic constants of GOx were also determined with the conventional method at six reaction parameters and compared with the results of the proposed method. The correlation coefficients (R-2) between the results of two methods were determined as 0.940 and 0.869 for V-max and K-m, respectively. (c) 2005 Elsevier B.V. All rights reserved.