Development and validation of a stochastic molecular model of cellulose hydrolysis by action of multiple cellulase enzymes

Background: Cellulose is hydrolyzed to sugar monomers by the synergistic action of multiple cellulase enzymes: endo-β-1,4-glucanase, exo-β-1,4 cellobiohydrolase, and β-glucosidase. Realistic modeling of this process for various substrates, enzyme combinations, and operating conditions poses severe challenges. A mechanistic hydrolysis model was developed using stochastic molecular modeling approach. Cellulose structure was modeled as a cluster of microfibrils, where each microfibril consisted of several elementary fibrils, and each elementary fibril was represented as three-dimensional matrices of glucose molecules. Using this in-silico model of cellulose substrate, multiple enzyme actions represented by discrete hydrolysis events were modeled using Monte Carlo simulation technique. In this work, the previous model was modified, mainly to incorporate simultaneous action enzymes from multiple classes at any instant of time to account for the enzyme crowding effect, a critical phenomenon during hydrolysis process. Some other modifications were made to capture more realistic expected interactions during hydrolysis. The results were validated with experimental data of pure cellulose (Avicel, filter paper, and cotton) hydrolysis using purified enzymes from Trichoderma reesei for various hydrolysis conditions. Results: Hydrolysis results predicted by model simulations showed a good fit with the experimental data under all hydrolysis conditions. Current model resulted in more accurate predictions of sugar concentrations compared to previous version of the model. Model results also successfully simulated experimentally observed trends, such as product inhibition, low cellobiohydrolase activity on high DP substrates, low endoglucanases activity on a crystalline substrate, and inverse relationship between the degree of synergism and substrate degree of polymerization emerged naturally