ARTICLE Optimization of Enzyme Complexes for Lignocellulose Hydrolysis Alex Berlin, Vera Maximenko, Neil Gilkes, Jack Saddler Forest Products Biotechnology, Department of Wood Science, Faculty of Forestry, University of British Columbia, Vancouver, British Columbia, Canada V6T 1Z4; telephone: (604) 827-5005; fax: (604) 822-9159; e-mail: alex.berlin@ubc.ca Received 7 April 2006; accepted 2 October 2006 Published online 20 October 2006 in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/bit.21238 ABSTRACT: The ability of a commercial Trichoderma reesei cellulase preparation (Celluclast 1.5L), to hydrolyze the cellulose and xylan components of pretreated corn stover (PCS) was significantly improved by supplementation with three types of crude commercial enzyme preparations nom- inally enriched in xylanase, pectinase, and b-glucosidase activity. Although the well-documented relief of product inhibition by b-glucosidase contributed to the observed improvement in cellulase performance, significant benefits could also be attributed to enzymes components that hydro- lyze non-cellulosic polysaccharides. It is suggested that so- called ‘‘accessory’’ enzymes such as xylanase and pectinase stimulate cellulose hydrolysis by removing non-cellulosic polysaccharides that coat cellulose fibers. A high-throughput microassay, in combination with response surface metho- dology, enabled production of an optimally supplemented enzyme mixture. This mixture allowed for a twofold reduction in the total protein required to reach glucan to glucose and xylan to xylose hydrolysis targets (99% and 88% conversion, respectively), thereby validating this approach towards enzyme improvement and process cost reduction for lignocellulose hydrolysis. Biotechnol. Bioeng. 2007;97: 287–296. ß 2006 Wiley Periodicals, Inc. KEYWORDS: bioconversion; lignocellulose; cellulose; hemi- cellulose; xylan; cellulase Introduction Plant biomass contains large amounts of cellulose and other polysaccharides that can be hydrolyzed to glucose and various other simple sugars for subsequent fermentation to fuel ethanol or used in the production of other industrial chemicals. In nature, the biodegradation of plant biomass is a slow process because lignin and substrate crystallinity greatly restrict the access of hydrolytic enzymes to the polysaccharide components. However, raw biomass can be pretreated and partially fractionated, using processes that typically involve elevated temperature and pressure com- bined with acid or base catalysis, to yield lignocellulosic materials that are much more susceptible to enzyme attack (Kim and Lee, 2005; Mais et al., 2002; Palonen et al., 2004). Nevertheless, although the economics of ethanol production from lignocellulose continues to improve (Genencor International, Inc., 2003), the costs for biomass pretreat- ment and enzyme-catalyzed hydrolysis continue to deter widespread commercialization of this process. Hydrolysis of the pretreated lignocellulose to simple sugars typically uses a complex of secreted enzymes derived from filamentous fungi, particularly Trichoderma sp. Such enzyme complexes contain high levels of cellulases (endoglucanases and cellobiohydrolases), together with lower amounts of enzymes that attack non-cellulosic polysaccharides such as hemicellulose and pectin. Attempts to improve the hydrolytic efficiency of such enzyme complexes have traditionally focused on their component cellulases because cellulose is the most abundant poly- saccharide component in lignocellulose. However, it is now recognized that the hydrolytic efficiency of fungal cellulase complexes determined using a model cellulosic substrate does not provide a reliable indication of its performance on pretreated lignocellulose (Berlin et al., 2005a, 2006; Kabel et al., 2005). Evidently, other components in pretreated biomass, particularly hemicellulose and lignin, exert significant restraints on cellulose hydrolysis. For example, one mechanism whereby lignin seems to reduce hydrolytic performance is by binding enzyme components non- productively. Consequently enzyme mixtures with similar cellulase activity may show differences in performance on lignocellulose if they differ in affinity for lignin (Berlin et al., 2005). Similarly, it is probable that hemicelluloses restrict Correspondence to: A. Berlin Contract grant sponsor: Natural Science and Engineering Research Council of Canada (NSERC) Contract grant sponsor: Natural Resources Canada (NRCAN) ß 2006 Wiley Periodicals, Inc. Biotechnology and Bioengineering, Vol. 97, No. 2, June 1, 2007 287