The potential of enzyme recycling during the hydrolysis of a mixed softwood feedstock Maobing Tu a, * , Xiao Zhang b , Mike Paice b , Paul MacFarlane c , Jack N. Saddler c a School of Forestry and Wildlife Sciences, Auburn University, Auburn, AL 36849-5418, USA b FPInnovations, Pulp and Paper Research Institute of Canada (Paprican), 570 Boul. Saint Jean, Pointe-Claire, Quebec, Canada H9R 3J9 c Department of Wood Science, University of British Columbia, Vancouver, BC, Canada V6T 1Z4 article info Article history: Received 3 November 2008 Received in revised form 19 March 2009 Accepted 27 June 2009 Available online 25 July 2009 Keywords: Cellulase Enzyme recycling Response surface methodology Lignocellulosic hydrolysis abstract Despite recent improvement in cellulase enzymes properties, the high cost associated with the hydrolysis step remains a major impediment to the commercialization of full-scale lignocellulose-to-ethanol bio- conversion process. As part of a research effort to develop a commercial process for bioconversion of soft- wood residues, we have examined the potential for recycling enzymes during the hydrolysis of mixed softwood substrate pretreated by organosolv process. We have used response surface methodology to determine the optimal temperature, pH, ionic strength, and surfactant (Tween 80) concentration for max- imizing the recovery of bound protein and enzyme activity from the residual substrates after hydrolysis. Data analysis showed that the temperature, pH and surfactant concentration were the major factors gov- erning enzyme desorption from residual substrate. The optimized conditions were temperature 44.4 °C, pH 5.3 and 0.5% Tween 80. The optimal conditions significantly increased the hydrolysis yield by 25% after three rounds of hydrolysis. This bound enzyme desorption combining with free enzyme re-adsorp- tion is a potential method to recover cellulase enzymes and reduce the cost of enzymatic hydrolysis. Ó 2009 Elsevier Ltd. All rights reserved. 1. Introduction Bioconversion of lignocellulosics to produce fuel-grade ethanol presents an attractive opportunity for the production of renewable, environmentally friendly biofuels (Farrell et al., 2006). Although there have been recent advances towards improving the efficiency of the bioconversion process (Cherry, 2003), the high cost of hydro- lytic enzymes required for the hydrolysis step remains a significant obstacle to the application of this bioconversion technology at a commercial scale. Strategies to reduce enzyme cost include increasing enzyme production efficiency, increasing enzyme spe- cific activity and recycling cellulase enzymes to be used in subse- quent hydrolysis (Cherry, 2003; Tu et al., 2007a,b). As part of a research effort to bring the bioconversion of lignocelluloses closer to commercialization, we assessed the potential of enzyme recy- cling during the hydrolysis of a mixed softwood feedstock (spruce, pine and Douglas-fir) pretreated by an ethanol organosolv process. Cellulase recycling provides a practical strategy to reduce the cost of enzyme for lignocellulose hydrolysis in bioconversion process (Gregg et al., 1998; Lu et al., 2002; Mabee et al., 2004). During hydrolysis, cellulase enzymes typically are present in the system in two different forms (Tu et al., 2007a,b). Some cellulase enzymes remain free in the solution (free enzyme), while others are bound to the residual substrates (both cellulose and lignin). An efficient enzyme recycling requires effective recovery of both free and ad- sorbed cellulase. Previous studies have demonstrated that free en- zyme could be recovered potentially by membrane filtration (Tjerneld, 1994) and affinity re-adsorption with fresh substrates (Ramos et al., 1993; Tu et al., 2007a,b). However, there is a lack of effective strategy to recover bound enzyme after lignocellulose hydrolysis (Deshpande and Eriksson, 1984). This is primarily due to the non-productive binding between enzyme and residual lig- nin. Although surfactants have been found to be able to detach ac- tive cellulase from residual lignin (Alkasrawi et al., 2003; Castanon and Wilke, 1981; Eriksson et al., 2002; Tu et al., 2007a,b), a practi- cal method to apply surfactants for recovering bound cellulase is yet to be developed. Desorption of cellulase from lignin is influ- enced by a number of factors including temperature, pH, ionic strength, and the amount of the surfactant (Andreaus et al., 1999; Otter et al., 1989). In recent years, response surface method- ology (RSM) has been shown as a promising tool for optimizing biotechnology process (Ahuja et al., 2004; Eon-Duval et al., 2003; Ismail et al., 1998; Vazquez and Martin, 1998). In this study, RSM was applied to investigate and optimize the cellulase desorption and recovery of active enzymes. Four independent variables were selected for the RSM: temperature, pH, ionic strength and surfac- tant concentration. 0960-8524/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2009.06.108 * Corresponding author. Tel.: +1 334 844 8829; fax: +1 334 844 1084. E-mail address: mtu@auburn.edu (M. Tu). Bioresource Technology 100 (2009) 6407–6415 Contents lists available at ScienceDirect Bioresource Technology journal homepage: www.elsevier.com/locate/biortech