An experimental and modeling approach to describe the deactivation of cellulases at the air-liquid interface.

Understanding the reaction mechanisms involved in the enzymatic hydrolysis of cellulose is important because it is kinetically the most limiting step of the bioethanol production process. The present work focuses on the enzymatic deactivation at the air-liquid interface, which is one of the aspects contributing to this global deactivation. This phenomenon has already been experimentally proven, but this is the first time that a model has been proposed to describe it. Experiments were performed by incubating Celluclast cocktail solutions on an orbital stirring system at different enzyme concentrations and different surface-to-volume ratios. A 5-day follow-up was carried out by measuring the global FPase activity of cellulases for each condition tested. The activity loss was proven to depend on both the air-liquid surface area and the enzyme concentration. Both observations suggest that the loss of activity takes place at the air-liquid surface, the total amount of enzymes varying with volume or enzyme concentration. Furthermore, tests performed using five individual enzymes purified from a Trichoderma reesei cocktail showed that the only cellulase that is deactivated at the air-liquid interface is cellobiohydrolase II. From the experimental data collected by varying the initial enzyme concentration and the ratio surface to volume, it was possible to develop, for the first time, a model that describes the loss of activity at the air-liquid interface for this configuration.

Mechanistic modeling of enzymatic hydrolysis of cellulose integrating substrate morphology and cocktail composition.

A mechanistic model of enzymatic hydrolysis taking into account the morphology of the cellulosic particles and its evolution with time was developed. The individual behavior of the main enzymes involved in the reaction (cellobiohydrolases, endoglucanases, and ß-glucosidases), as well as synergy effects, were also included. A large panel of experimental tests was done to fit and validate the model. This database included different enzymes mixtures and operating conditions and allowed to determine and compare with accuracy the adsorption and kinetic parameters of the different enzymes. Model predictions on short hydrolysis times were very satisfactory. On longer times, a deactivation constant was added to represent the hydrolysis slowdown. The model also allowed to predict the impact of enzymes ratios and initial substrate parameters (chain length distribution, polymerization degree) on hydrolysis, and to follow the evolution of these parameters with time. This model revealed general trends on the impact of cellulose morphology on hydrolysis. It is a useful tool to better understand the mechanisms involved in enzymatic hydrolysis of cellulose and to determine optimal cellulolytic cocktails for process design.