Enzyme and Microbial Technology 46 (2010) 170176

Contents lists available at ScienceDirect

Enzyme and Microbial Technology

journal homepage: www.elsevier.com/locate/emt

Inhibition of cellulases by phenols

Eduardo Ximenes a,b , Youngmi Kim a,b , Nathan Mosier a,b , Bruce Dien d , Michael Ladisch a,b,c,
Laboratory of Renewable Resources Engineering, Purdue University, West Lafayette, IN 47907-2032, United States Department of Agricultural and Biological Engineering, Purdue University, West Lafayette, IN 47907-2032, United States c Weldon School of Biomedical Engineering, Purdue University, West Lafayette, IN 47907-2032, United States d Fermentation Biotechnology Research Unit, National Center for Agricultural Utilization Research, USDA, Agricultural Research Service, 1815 N. University Street, Peoria, IL 61604, United States
b a

a r t i c l e

i n f o

a b s t r a c t
Enzyme hydrolysis of pretreated cellulosic materials slows as the concentration of solid biomass material increases, even though the ratio of enzyme to cellulose is kept constant. This form of inhibition is distinct from substrate and product inhibition, and has been noted for lignocellulosic materials including wood, corn stover, switch grass, and corn wet cake at solids concentrations greater than 10 g/L. Identication of enzyme inhibitors and moderation of their effects is of considerable practical importance since favorable ethanol production economics require that at least 200 g/L of cellulosic substrates be used to enable monosaccharide concentrations of 100 g/L, which result in ethanol titers of 50 g/L. Below about 45 g/L ethanol, distillation becomes energy inefcient. This work conrms that the phenols: vanillin, syringaldehyde, trans-cinnamic acid, and hydroxybenzoic acid, inhibit cellulose hydrolysis in wet cake by endo- and exo-cellulases, and cellobiose hydrolysis by -glucosidase. A ratio of 4 mg of vanillin to 1 mg protein (0.5 FPU) reduces the rate of cellulose hydrolysis by 50%. -Glucosidases from Trichoderma reesei and Aspergillus niger are less susceptible to inhibition and require about 10 and 100 higher concentrations of phenols for the same levels of inhibition. Phenols introduced with pretreated cellulose must be removed to maximize enzyme activity. 2010 Elsevier Inc. All rights reserved.

Article history: Received 4 June 2009 Received in revised form 30 October 2009 Accepted 5 November 2009 Keywords: Cellulose Cellobiose Cellulases -Glucosidase Enzyme inhibition Cellulase inhibitors Cellulose hydrolysis T. Reesei A. niger Xylan Aromatic acids Tannins Phenols

1. Introduction The inhibition of cellulolytic enzymes is broad-based. Causes include substrate and product inhibition, mass transfer resistance, particle size effects, and non-productive enzyme binding with lignin [111]. In addition phenolic molecules produced by plants also inhibit enzyme hydrolysis at M to mM concentrations [1219]. Plants deploy such inhibitors to help to protect themselves against pathogens that utilize cellulases to gain entry into plant cells [20,21]. This paper characterizes enzyme inhibition by phenolics and oligosaccharides that are released during pretreatment and hydrolysis of wet cake and lignocellulosic biomass [2226]. Wet cake is a prevalent co-product of the corn to ethanol industry and contains cellulose, hemicellulose, starch, protein, and oil [27,28]. It contains soluble phenolic compounds but little true lignin. Wet cake is derived from pericarp, germ, and insoluble protein, i.e., solids left after corn is fermented to ethanol. In a dry grind

ethanol plant, wet cake is mixed with evaporated light stillage, dried to produce DDGS, and sold as an animal feed [27]. Hydrolysis and fermentation of the cellulosic components in wet cake increases ethanol yields from a bushel of corn by 814%, with the residue being a potentially high-value DDGS that has increased protein content. This approach provides an attractive pathway for introducing cellulose conversion technology into existing corn to ethanol plants [2732]. The low lignin content of wet cake makes it susceptible to enzyme hydrolysis, even without pretreatment. Rates and yields of glucose formation are enhanced compared to lignocellulosic plant matter at equivalent conditions. Unlike heavily lignied substrates such as wood, corn stover, and switchgrass, wet cake enables the differentiation of effects of recalcitrance (a property of the biomass substrate) from inhibition (a characteristic of enzymes). We report on inhibitors present in or released from wet cake and lignocellulose.
2. Materials and methods

Corresponding author at: Department of Agricultural and Biological Engineering, Purdue University, West Lafayette, IN 47907-2032, United States. Tel.: +1 765 494 7022; fax: +1 765 494 7023. E-mail address: [email protected] (M. Ladisch). 0141-0229/$ see front matter 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.enzmictec.2009.11.001

2.1. Materials Wet distillers grains (WDG or wet cake) and thin stillage were obtained from Big River Resources, LLC, a dry grind ethanol plant located in West Burlington, IA (com-

E. Ximenes et al. / Enzyme and Microbial Technology 46 (2010) 170176 Table 1 Compositional analysis of wet cake before and after pretreatment. Solids composition dry weight basis (%). Initial Water extractives Ether extractives Crude protein Glucan (total) Cellulose Starch Xylan and arabinan Xylan Arabinan Ash Total dry matter mass closure 8.8 9.6 36.6 17.6 12.6 5.0 14.9 5.5 2.0 96.4 After pretreatment 2.4. Protein content 13.6 11.1 4.3

171

sis was determined by a d-glucose (GOPOD Format) assay kit (Megazyme, Wicklow, Ireland). Experiments were done in triplicate and results reported as an average with an indicated standard deviation.

The protein content of the commercial preparations was determined using a Pierce BCA Protein Assay kit (Thermo Scientic, Rockford, IL). 2.5. Determination of phenolic compounds Phenolic compounds were assayed by the Prussian blue method [47]. Three milliliters of diluted sample liquid were transferred to a 1-cm cuvette and 200 L of 0.008 M K2 Fe(CN)6 was added, followed immediately by the addition of 200 L of 0.1 M FeCl3 in 0.1 M HCl. Absorbance was read after 5 min at 700 nm at room temperature, against a gallic acid standard. 2.6. Inhibition assays For cellulose hydrolysis, the inhibitory effects of the lyophilized solids or purchased pure components were tested using hydrolysis procedures based on NREL LAP009 [48]. The nal reaction volume consisted of 5 mL of 0.1 M citrate buffer (pH 4.8) with the following components added: a weight of wet cake containing the equivalent of 100 mg glucan, cellulase activity equivalent to 15 FPU/g glucan (Spezyme CP), and -glucosidase of 40 CBU/g glucan (Novozyme 188). The mixture was incubated in a New Brunswick shaker at 50 C, 200 rpm for 24 h. Glucose formed was analyzed by HPLC (Aminex HPX-87H column) or by Megazyme kit. Lyophilized solids from pretreated wet cake were obtained from the liquid recovered by centrifugation (5000 rpm, 20 min). The liquid contained 20% (w/v) (17%, w/w) soluble components of which 85% (7.2 out of 8.5 g in 50 mL) were recovered. The difference was due to volatile components. Oat spelt xylan contains 10% glucose, 15% arabinose and the remainder xylan. It has arabinose side chains with O3 linked sugars, and is representative of an arabinoxylan substrate (or inhibitor) [49,50]. Xylan, pectin (from citrus) and starch hydrolysis products were obtained using Multifect Xylanase, Multifect Pectinase or Spezyme Fred, respectively. The enzymes were mixed with 10 g/L oat spelt xylan, pectin or starch, respectively, at a volume ratio of 1:4 enzyme: substrate, incubated for 10 min at 50 C, and boiled for 10 min to deactivate the enzyme. Reducing sugars were then determined [46]. In order to evaluate inhibitory effects, measured quantities of the phenols, or xylan, pectin or starch or their hydrolyzates were added to 100 mM sodium citrate, pH 4.8, containing 150 g/L untreated wet cake or 5 g/L (14.6 mM) cellobiose. Both washed and unwashed substrates were used. Washing consisted of passing 100 mL water at 50 C over 0.51 g untreated substrate. Hydrolysis was carried out with the indicated enzymes. Rates and extents of hydrolysis in the presence and absence of added inhibitors were then compared. 2.7. HPLC analysis HPLC analysis of liquid samples was performed on a system consisting of a Varian 9010 Solvent Delivery System, Waters 717 plus Auto sampler, Aminex HPX87H column (Bio-Rad, Hercules, CA), Waters 2414 Refractive Index Detector, Waters 2487 Dual Absorbance Detector, and a Hewlett Packard HP3396G Integrator. The mobile phase was 5 mM H2 SO4 ltered through 0.2 m nylon lter (Millipore) and degassed. The mobile phase ow rate was 0.6 mL/min. Column temperature was maintained at 60 C by an Eppendorf CH-30 Column Heater controlled by an Eppendorf TC-50.

positions in Tables 1 and 2). Spezyme CP (cellulase), Multifect Xylanase (xylanase), Multifect Pectinase (pectinase) and Spezyme Fred (thermostable -amylase from a genetically modied Bacillus licheniformes) were from Genencor, Danisco Division (Palo Alto, CA). Novozyme 188 (also spelled Novozym 188) was from Sigma (St. Louis, MO, Cat. No. C6150). Other reagents and chemicals were purchased from SigmaAldrich (St. Louis, MO). 2.2. Liquid hot water (LHW) pretreatment Wet cake was slurried in thin stillage to give 150 g/L total solids, dry basis, and a total soluble and insoluble solids concentration of 200 g/L. The slurry was loaded into a 316 stainless steel tubing (2.54 cm diameter 2.1 mm wall thickness 11.4 cm long) tted with a pair of 1 in. (2.54 cm) Swagelok tube end ttings (Swagelok, Solon, OH). Each tube had a volume of 45 mL and was lled with 33.7 mL slurry to give about 25% free space for liquid expansion. The slurry is heated at 160 C for 24 min by placing the tube in a Tecam SBL-1 uidized sand bath (Cole-Parmer, Vernon Hills, IL) at 160 C for 24 min (4 min heat-up time and 20 min hold time). The pressure within the tube was held at the saturation vapor pressure of water in order to keep the water in a liquid state [3337]. After 24 min, the tube was cooled by placing it in water and then in an ice-water slurry. The pretreated material was stored frozen until use [27,36,3841]. During pretreatment, the pH was maintained at 47 in order to minimize hydrolysis of the cellulose and hemicellulose to monosaccharides which rapidly degrade to aldehydes and acids [3843]. Stillage was used in place of water, since it self-buffers at the desired pH range for liquid hot water pretreatment and is added to wet cake in corn to ethanol plants. Liquid from pretreated wet cake contained 140 g/L of dissolved solids when recovered by centrifugation of pretreated wet cake at 5000 rpm for 20 min. The liquid was lyophilized. Lyophilized solids were redissolved in sodium citrate (100 mM, pH 4.8) at concentrations of 5 and 10 g/L, and evaluated for inhibition of enzyme activities. 2.3. Enzyme activity assays Cellulase and cellobiase activities were measured by procedures of the International Union of Pure and Applied Chemistry (IUPAC). Action against a specied substrate denes enzyme activity whether the enzyme is puried or exists in a background of other enzymes [44,45]. In these assays, release of reducing sugars from lter paper was determined per Miller [46], while glucose from cellobiose hydroly-

3. Results
Table 2 Compositional analysis of thin stillage (g/L). Initial Glucose Cellobiose Glucan (oligosaccharide, DP > 2) Xylose Xylan (oligosaccharide) Arabinose Arabinan (oligosaccharide) Succinic acid Lactic acid Glycerol Acetic acid Butanediol Ethanol 0.9 0.0 12.4 0.7 3.7 0.4 0.5 0.0 16.8 14.4 0.3 1.9 0.6 After pretreatment 0.8 3.1 14.6 2.5 8.2 2.9 4.3 0.3 1.0 13.5 1.4 0.0

3.1. Inhibitors formed during pretreatment of wet cake Inhibition doubled when the concentration of the lyophilized pretreatment liquid solids was doubled (Table 3). This behavior motivated LC analysis of the redissolved solids in order to identify potential inhibitory molecules, and to better understand possible inhibitory action of specic molecules on cellulase and -glucosidase. It is notable that washing wet cake that was not pretreated had no effect on enzyme hydrolysis by cellulase and -glucosidase; washed and unwashed both yielded 30% hydrolysis after 24 h. This shows that pretreatment releases enzyme inhibitors, while dissolved solids present in the stillage, a significant component of wet cake, are not a source. Untreated wet cake was subsequently used to gauge the effect of added inhibitors on enzyme hydrolysis.

Moisture of wet cake before drying is 64.7% and thin stillage is 92.3% [27].

172

E. Ximenes et al. / Enzyme and Microbial Technology 46 (2010) 170176

Table 3 Inhibition of glucan hydrolysis in untreated wet cake by the mixed component action of cellulases (Spezyme 15 FPU/g glucan) and -glucosidase (40 CBU/g glucan). %Redissolved solids 0.0 0.5 1.0 %Inhibition after 24 h 0 38 3.0 77 1.1 %Glucan conversion 30 18.6 6.9 Hydrolysis rate (g glucose/g glucan/h) 0.012 0.007 0.003

3.2. Inhibitors of cellulose hydrolysis Potential inhibitors present in pretreated wet cake were assayed for their effects on the activity of commercial cellulase (Spezyme CP) and -glucosidase (Novozyme 188) preparations. The compounds assayed included corn oil, starch and its hydrolysis products, hemicellulose and its hydrolysis products (xylose, xylooligomers, arabinose, d-glucuronic acid), pectin and its hydrolysis products (uronic acids) and phenolic compounds (trans-cinnamic acid, 4-hydroyxbenzoic acid, syringaldehyde and vanillin). Usually, when most biomasses are pretreated, the phenolic compounds originate from Klasson lignin. While wet cake does not contain signicant amounts of Klasson lignin (less that 0.2%) [27], it does contain soluble lignin related molecules from which lower molecular weight phenol compounds may be generated. The concentration of soluble phenolics was measured colorimetrically in pretreated wet cake with stillage at 20% (w/w) solids. It was determined to be in the range of 1 g/L, which is equivalent to about 5 g/L phenolics in solid phase. Wet cake contains signicant amounts of starch, which pretreatment would partially hydrolyze. Starch added at 10 g/L inhibited hydrolysis of glucan (in untreated wet cake) by a mixture of cellulases and -glucosidase. To evaluate the inhibitory potential of starch hydrolysis products on cellulases, starch was partially hydrolyzed to a mixture of glucose, maltose, and maltodextrins using a commercial amylase preparation (Spezyme Fred). In contrast to unhydrolyzed starch, this mixture was not inhibitory (Fig. 1A and B). Hemicellulose (oat spelt xylan) and pectin also inhibits glucan hydrolysis. However, when xylan is hydrolyzed into soluble products, including xylo-oligosaccharides and xylose, the products are signicantly more inhibitory to glucan hydrolysis (Fig. 1A and B) and is consistent with a recent report by Kumar and Wyman [26]. While the hydrolysates contain d-glucuronic acid and arabinose, neither of these was inhibitory to cellulases (Fig. 2A). Of the phenolic compounds screened, vanillin was the most signicant inhibitor of hydrolysis of glucan in wet cake by a mixture of commercial cellulase and -glucosidase. Syringaldehyde and transcinnamic acid were moderately inhibitory and 4-hydroxybenzoic acid was the least inhibitory (Fig. 2B). The inhibitory effect of 4hydroxybenzoic acid is about the same as the inhibition effects caused by an equivalent molar concentration of xylose. Overall, Figs. 1 and 2 show between 2 and 50% inhibition for hydrolysates, unhydrolyzed polysaccharides and added monosaccharides and phenols. Absence of inhibition corresponds to a glucose formation rate of 0.012 g/g glucan/h for untreated wet cake (Table 3).

Fig. 1. Effect of (A) polysaccharide hydrolysates (1.2 g/L) vs (B) unhydrolyzed polysaccharides (10 g/L) on cellulose hydrolysis in the presence of mixed component of cellulase activities (Spezyme 15 FPU/g glucan or 26 mg protein/g glucan) and -glucosidase (40 CBU/g glucan or 8.5 mg protein/g glucan). For 2% glucan (w/v) initial concentration, glucose released in the control case (no inhibitor) was 0.3 g glucose/g glucan (0.012 g glucose/g glucan/h), or 30% conversion of glucan (cellulose fraction) in wet cake in 24 h.

Novozyme 188 added to Spezyme CP gave signicantly higher glucose formation than Spezyme CP to which a puried Aspergillus niger -glucosidase had been added (Table 4). Both commercial enzyme preparations were tested at the same enzyme loading. However, Novozyme 188 contains high amylase activity (5300 IU/mL) [51] while puried -glucosidase has only low levels of starch hydrolyzing activity (<0.001 U/mg protein (Megazyme International Ireland Ltd., Co., Wicklow, Ireland)). Therefore the additional glucose is from starch. Further indication of starch

Table 4 Enzyme hydrolysis of pretreated wet cake at 20% solids (w/v) by mixed component cellulase and -glucosidase. Source of added -glucosidase pH of Hydrolysis mixture -Glucosidase loading (CBU/g glucan) Novozyme 188 (A. niger) -Glucosidase (A. Niger) 3.0 3.0 g protein/ g glucan 0.64 0.04 pH 4.0 pH 6.0 pH 7.0

Glucose formed (g/L) 13.5 0.78a 4.42 0.46

Glucan conversion (%) 45 14.7

Glucose formed (g/L) 9.2 0.49a 8.6 0.25

Glucan conversion (%) 31 29

Glucose formed (g/L) 6.08 0.40a 4.16 0.08

Glucan conversion (%) 20 14

Hydrolysis at 50 C for 24 h by cellulase (Spezyme CP-0.58 FPU/mg protein and 15 FPU/g glucan (cellulose and starch) using -glucosidase from Novozyme 188 (not puried, from A. niger) or puried -glucosidase (from A. niger). a May include starch hydrolysate since enzyme contains amylase activity.

E. Ximenes et al. / Enzyme and Microbial Technology 46 (2010) 170176

173

hydrolysis is given by glucan (wet cake) hydrolysis at pH 6.0 where the combined action of Spezyme CP and the puried A. niger -glucosidase gives the lower conversion than the combined action of Spezyme and Novozyme 188 (Table 4), due to different pH optima for -glucosidase and amylase. The xylanase and -arabinofuranosidase activities in Novozyme 188 -glucosidase form xylose and arabinose and explain the range of monosaccharides obtained when wet cake is hydrolyzed by a mixture of t Spezyme CP and Novozyme 188 (Table 5). 3.3. Inhibition of Trichoderma reesei vs. A. niger -glucosidase activities Compared to enzyme from A. niger; -glucosidase from T. reesei was more sensitive to inhibition by starch, pectin and their hydrolysis products (Fig. 3AD). Xylan hydrolysis products inhibited hydrolysis of 5 g/L (15 mM) cellobiose to glucose more severely than xylan itself and arabinose was more inhibitory than xylose and d-glucuronic acid (Fig. 4A and B). All inhibitors tested, and particularly phenols, affected -glucosidase from T. reesei more severely than -glucosidase from A. niger. A 10 g/L of 4-hydroxybenzoic acid, hydrolysis of cellobiose by T. reesei -glucosidase activity was inhibited 92.5% compared to 35% for A. niger -glucosidase (Fig. 4C and D). Absence of inhibition corresponded to a cellobiose hydrolysis rate of 0.5 and 7.5 g glucose/g cellobiose/h for T. reesei and A. niger -glucosidases, respectively.
Fig. 2. The effect of (A) carbohydrates and (B) phenols (10 g/L) on cellulose hydrolysis in the presence of a mixed cellulase activities: Spezyme (15 FPU/g glucan or 26 mg protein/g glucan) and -glucosidase (40 CBU/g glucan or 8.5 mg protein/g glucan). For 2% glucan initial concentration (w/v), glucose released in the control case (no inhibitor) was 0.3 g glucose/g glucan (0.012 g glucose/g glucan/h), or 30% conversion of glucan (cellulose fraction).

3.4. Hemicellulose hydrolysis of pretreated wet cake The action of Novozyme 188 on LHW pretreated wet cake (15% solids, w/v, in stillage with a total solids content of 20%) gave xylooligosaccharides, xylose and arabinose in addition to glucose. A greater extent of hydrolysis occurred at pH 4.0, than at pH 6.0 and 7.0 (Table 5). -glucosidase from A. niger contains enzymes with xylanase activity and this accounts for appearance of these prod-

Fig. 3. The effect of polysaccharide hydrolysate on -glucosidase from (A) T. reesei and (B) A. niger and the respective polysaccharides (10 g/L) for (C) T. reesei and (D) A. niger. Enzymes used: Spezyme CP and Novozyme 188 (both 15 CBU/g cellobiose or 17 and 3.4 mg of protein/g cellobiose for Spezyme CP and Novozyme 188, respectively). For 0.5% cellobiose initial concentration (w/v), 0% inhibition of cellobiose activity for Spezyme CP and Novozyme 188 corresponds to 34 and 515 CBU/mL, or 0.5 and 7.5 g glucose/g cellobiose/h, respectively.

174

E. Ximenes et al. / Enzyme and Microbial Technology 46 (2010) 170176

Table 5 Xylose and arabinose release due to enzyme hydrolysis of pretreated wet cake (20% solids, w/v) by mixed component cellulase and -glucosidase. pH of hydrolysis mixture Source of added -glucosidase Novozyme 188 (A. niger) -glucosidase (A. Niger) -Glucosidase (CBU/g glucan) 3.0 3.0 -Glucosidase (g protein/g glucan) 0.64 0.04 pH 4.0 Xylose (g/L) 0.56 0.04 0.16 0.01 Arabinose (g/L) 0.59 0.05 0.30 0.01 pH 6.0 Xylose (g/L) 0.38 0.03 0.16 0.01 Arabinose (g/L) 0.40 0.02 0.29 0.01 pH 7.0 Xylose (g/L) 0.24 0.02 0.08 0.01 Arabinose (g/L) 0.33 0.02 0.28 0.01

Hydrolysis at 50 C for 24 h by (Spezyme CP-0.58 FPU/mg protein) and 15 FPU/g glucan) (cellulose and starch) using -glucosidase from Novozyme 188 (not puried, from A. niger) or puried -glucosidase (from A. niger).

Fig. 4. The effect of (A) carbohydrates and (B) phenols (10 g/L) on cellobiose hydrolysis. Enzymes used: Spezyme CP and Novozyme 188 (both 15 CBU/g cellobiose or 17 and 3.4 mg of protein/g cellobiose for Spezyme CP and Novozyme 188, respectively). For 0.5% cellobiose initial concentration (w/v), 0% inhibition of cellobiose activity for Spezyme CP and Novozyme 188 corresponds to 34 and 515 CB/mL, 0.5 and 7.5 g glucose/g cellobiose/h, respectively.

ucts [44]. These enzymes may be subject to inhibition to phenols, although this was not addressed in this study. 4. Discussion The amounts and types of enzyme inhibitors formed during lignocellulose pretreatments will depend upon biomass composition, pretreatment type, and reaction conditions [2326,52]. Phenols appear to be the strongest inhibitors of enzyme production by different microorganisms as well as the enzyme activities themselves [1215]. Cellulases, -glucosidases and hemicellulases are all affected [1519,24]. Our work shows lignin related compounds are inhibitory to glucan hydrolysis (Fig. 2B), with vanillin being among the most inhibitory (Table 6). Fig. 4C and D shows that inhibition of enzyme from A. niger requires 4 higher concentrations than -glucosidase from T. reesei. In addition to phenolics, xylan hydrolysates also cause major inhibition of the mixed component action of cellulases and glucosidases. Li et al. [51] and Dien et al. [44] reported hemicellulase and other non-cellulolytic activities are found in -glucosidase from A. niger, i.e., Novozyme 188. This helps to explain the seemingly complex interactions between inhibition and mixtures of T. reesei enzymes and A. niger -glucosidase (Novozyme 188) that is able to convert 71% of the xylan from LHW-treated distillers dried grain with solubles (DDGS) to xylose [55]. This -glucosidase preparation may moderate cellobiose inhibition of cellobiohydrolase (a cellulase) by hydrolyzing the cellobiose to glucose, but then release other inhibitors of cellulase and Novozyme 188 due to

the hydrolytic action of this multi-component enzyme mixture on xylan (Figs. 1A, 2A and 4A, B and Table 5). Further evidence is given by oat spelt xylan and pectin that inhibited both cellulose and cellobiose hydrolysis, with hydrolysates of these polysaccharides being signicantly more inhibitory (Figs. 1A, B and 3AD). Inhibition by pectin hydrolyzate is a function of the source of enzyme (Fig. 3A and B). This follows the work of Kumar and Wyman [26] who reported that xylobiose and higher xyloligomers inhibited enzymatic hydrolysis of pure glucan, pure xylan, and pretreated corn stover. Xylose causes greater inhibition than xylan but less than xylobiose, and xylotriose. Xylose is known to inhibit hydrolysis of other cellulosic substrates by cellulases [54], as well as cellobiose hydrolysis by A. niger -glucosidase [23]. Arabinose is more inhibitory than xylose or d-glucuronic acid for hydrolysis of cellobiose by both T. reesei and A. niger glucosidases (Fig. 4A and B), with little effect on wet cake hydrolysis (Fig. 2A). Lesser inhibition by glucose, a known product inhibitor of glucosidase from T. reesei, is consistent with the observed inhibitory action of starch hydrolysate (Fig. 3A). Ladisch et al. [1] and Gong et al. [53] showed T. reesei -glucosidase activity was inhibited by 50% at 0.20.4 g/L glucose (i.e., Ki = 1.02 mM), for a ratio of enzyme activity to initial substrate of about 40 CBU/g cellobiose. Dekker [11] showed that 10 g/L cellobiose was completely hydrolyzed by A. niger -glucosidase (Novozyme 188) within 1 h at pH 5.0 and 50 C, but when 51 and 102 g/L glucose was added, hydrolysis required 22 h at 50 C. Starch is moderately inhibitory to cellobiose hydrolysis by the T. reesei -glucosidase as well as cellulase components,

E. Ximenes et al. / Enzyme and Microbial Technology 46 (2010) 170176 Table 6 Ranking of inhibitory effect of phenolic compounds on the hydrolysis of cellulose and cellobiose.

175

Concentration (mM) Cellulase (Spezyme CP) + -glucosidase (Novozyme 188) -Glucosidase (Spezyme CP) -Glucosidase (Novozyme 188)

66 4

55 3

67 2

67 1

3 1

2 3

1 1

4 4

Inhibition follows the designation of 4 > 3 > 2 > 1, with 4 being the most severe inhibition, and 1 being the least. -Glucosidase from Novozyme 188 is not signicantly inhibited, and it has the rating of 1 for two of the phenolic compounds shown. All concentrations equivalent to 1 g/L.

but not for -glucosidase from A. niger (Fig. 3AD). Non-productive adsorption/binding of both cellulases and -glucosidase to starch may cause some apparent inhibition but this effect is small. 5. Conclusions Phenols are major inhibitors for hydrolysis of glucan (in wet cake), cellobiose and for the combinations of enzymes evaluated in this study. -Glucosidase from T. reesei is more susceptible to inhibition than -glucosidase from A. niger with other enzyme activities present in the A. niger preparations generating additional inhibitors, including hemicellulose and starch hydrolysis products, hence resulting in counteracting effects. The qualitative ranking of inhibition effects in Table 6 show opportunities exist for optimizing enzyme formulations by varying the source and amounts of -glucosidase or adding cellulase. In the case of 4-hydroxybenzoic acid, signicantly higher levels of -glucosidase would improve hydrolysis given the strong effect of this phenol. In the case of vanillin, more cellulase is needed. The key message from our experimental results and analysis of the literature is that removal of inhibitors, other than glucose or cellobiose must be achieved during hydrolysis in order to maximize enzyme activity. Statement of competing interest Michael Ladisch is Chief Technology Ofcer at Mascoma Corporation. Acknowledgments The authors wish to thank Xingya (Linda) Liu, Rick Hendrickson and Thomas Kreke, for their excellent technical assistance, and Dr. Mira Sedlak and Dr David Hogsett for their internal review of this paper. We thank Big River Resources for providing wet cake, and Genencor for their gift of enzymes. The material in this work was supported by DOE grant #DE-AC36-99GO10337; DOE BES Project 0012846, DOE Grant #DE-FG02-06ER06-03, #GO12O26-174 DOE grant #DE-FG02-06ER64301 and USDA IFAFS contract #00-521049663, and Mascoma Corporation. References
[1] Ladisch MR, Gong C-S, Tsao GT. Corn crop residues as a potential source of single cell protein: kinetics of Trichoderma viride cellobiase action. Dev Ind Microbiol Ser 1977;18:15768. [2] Ladisch MR, Flickinger MC, Tsao GT. Fuels and chemicals from biomass. Energy 1979;4:26375.

[3] Ladisch MR, Hong J, Voloch M, Tsao GT. Cellulase kinetics. In: Hollaender A, Rabson R, Rodgers P, Pietro AS, Valentine R, Wolfe R, editors. Basic life science, vol. 18 (Trends Biol Ferment Fuels and Chem). New York: Plenun Press; 1981. p. 5583. [4] Ladisch MR, Gong C-S, Tsao GT. Cellulose to sugarsnew path gives quantitative yield. Science 1978;201:7435. [5] Yang B, Willies DM, Wyman CE. Changes in the enzymatic hydrolysis rate of avicel cellulose with conversion. Biotechnol Bioeng 2006;94(6):11228. [6] Yang B, Wyman CE. BSA treatment to enhance enzymatic hydrolysis of cellulose in lignin containing substrates. Biotechnol Bioeng 2006;94(4):6117. [7] Peri S, Kara S, Lee YY, Karim MN. Modeling intrinsic kinetics of enzymatic cellulose hydrolysis. Biotechnol Prog 2007;23:62637. [8] Reese ET, Siu RGH, Levinson HL. The biological degradation of soluble cellulose derivatives and its relationship to the mechanism of cellulose hydrolysis. J Bacteriol 1950;59(4):48597. [9] Halliwell G, Griin M. The nature and mode of action of the cellulolytic component C1 of Trichoderma koningii on native cellulose. Biochem J 1973;135:58794. [10] Ladisch MR, Gong C-S, Tsao GT. Cellobiase hydrolysis by endoglucanase (glucanglucanohydrolase) from Trichoderma reesei: kinetics and mechanism. Biotechnol Bioeng 1980;22:1017126. [11] Dekker RFH. Kinetic, inhibition, and stability of a commercial -d-glucosidase (cellobiase) preparation from Aspergillus niger and its suitability in the hydrolysis of lignocellulose. Biotechnol Bioeng 1986;28:143842. [12] Vohra RM, Shirsot CK, Dhawan S, Gupta KG. Effect of lignin and some of its components on the production and activity of cellulase(s) by Trichoderma reesei. Biotechnol Bioeng 1980;22:1497500. [13] Martin S, Akin DE. Effect of phenolic monomers on the growth and -glucosidase activity of Bacteroides ruminicola and on the carboxymethylcellulase, -glucosidase, and xylanase activities of bacteroides succinogenes. Appl Environ Microbiol 1988;54(12):301922. [14] Paul SS, Kamra DN, Sastry VRB, Sahu NP, Kumar A. Effect of phenolic monomers on biomass and hydrolytic enzyme activities of an anaerobic fungus isolated from wild nil gai (Baselophus tragocamelus). Lett Appl Microbiol 2003;36:37781. MJ, Lema JM. Inhibition of cellulase by sunower [15] Sineiro J, Dominguez H, Nnez polyphenols. Biotechnol Lett 1997;19(6):5214. [16] Berlin A, Balakshin M, Gilkes N, Kadla J, Maximenko V, Kubo S, et al. Inhibition of cellulase, xylanase and -glucosidase activities by softwood lignin preparations. J Biotechnol 2006;125(2):198209. [17] Akin DE, Rigsby LL. Corn ber: structure, composition, and response to enzymes for fermentable sugars and co-products. Appl Biochem Biotechnol 2008;144:5968. [18] Pan X. Role of functional groups in lignin inhibition of enzymatic hydrolysis of cellulose to glucose. J Biobased Materials and Bioenergy 2008;2(1): 2532. [19] Ximenes EA, Brandon SK, Doran-Peterson J. Evaluation of a Hypocrea jecorina enzyme preparation for hydrolysis of Tifton 85 bermudagrass. Appl Biochem Biotechnol 2008;146:89100. [20] Cook MT, Basset HF, Thompson F, Taubenhaus JJ. Protective enzymes. Science 1911;33:6249. [21] Mandels M, Reese ET. Inhibition of cellulases. Annu Rev Phytopathol 1965;3:85102. [22] Fan LT, Lee Y, Gharpuray MM. The nature of lignocellulosics and their pretreatment for enzymatic hydrolysis. Adv Biochem Eng 1982;23:15787. [23] Dekker RFH. Inhibitors of Trichoderma reesei -glucosidase activity derived from auto-hydrolysis-exploded eucalyptous-regnans. Appl Microb Biotechnol 1988;29:5938. [24] Excofer G, Toussaint B, Vignon MR. Saccharication of steam-exploded poplar wood. Biotechnol Bioeng 1991;38:130817.

176

E. Ximenes et al. / Enzyme and Microbial Technology 46 (2010) 170176 [40] Mosier N, Hendrickson R, Ho N, Sedlak M, Ladisch MR. Optimization of pH controlled liquid hot water pretreatment of corn stover. Bioresour Technol 2005;96:198693. [41] Mosier NS, Hendrickson R, Brewer M, Ho NWY, Sedlak M, Dreshel R, et al. Industrial scale-up of pH-controlled liquid hot water pretreatment of corn ber for fuel ethanol production. Appl Biochem Biotechnol 2005;125:7785. [42] Weil J, Sarikaya A, Rau S-L, Goetz J, Ladisch C, Brewer M, et al. Pretreatment of yellow poplar sawdust by pressure cooking in water. Appl Biochem Biotechnol 1997;68(12):2140. [43] Lu Y, Mosier NS. Kinetic modeling analysis of maleic acid-catalyzed hemicellulose hydrolysis in corn stover. Biotechnol Bioeng 2008;101(6):117081. [44] Dien BS, Ximenes EA, OBryan PJ, Moniruzzaman M, Li X-L, Balan V, et al. Enzyme characterization for hydrolysis of AFEX and liquid hot-water pretreated distillers grains and their conversion to ethanol. Bioresour Technol 2008;99(12):521625. [45] Ghose TK. Measurement of cellulase activities. Pure Appl Chem 1987;59(2):25768. [46] Miller GL. Use of dinitrosalicylic acid reagent for determination of reducing sugar. Anal Chem 1959;31:4268. [47] Budini R, Tonelli D, Girotti S. Analysis of total phenols using the Prussian blue method. J Agric Food Chem 1980;28:12368. [48] Brown L, Torget R. NREL analytical procedure: LAP009 enzymatic saccharication of lignocellulosic biomass hydrolysis. Golden, CO: National Renewable Energy Laboratory; 1996. [49] Leathers T. Bioconversions of maize residues to value-added co-products using yeast-like fungi. FEMS Yeast Res 2003;3:13340. [50] Bolam DN, Xie H, Pell G, Hogg D, Galbraith G, Henrissat B, et al. X4 modules represent a new family of carbohydrate-binding modules that display novel properties. J Biol Chem 2004;279(22):2295363. [51] Li X-L, Dien BS, Cotta MA, Wu YV, Saha BC. Prole of enzyme production by Trichoderma reesei grown on corn ber fractions. App Biochem Biotechnol 2005;121:32134. [52] Cantarella M, Cantarella L, Gallifuoco A, Spera A, Alfani F. Effect of inhibitors released during steam-explosion treatment of poplar wood on subsequent enzyme hydrolysis and SSF. Biotechnol Prog 2004;20:2006. [53] Gong C-S, Ladisch MR, Tsao GT. Cellobiase from Trichoderma viride: purication, properties, kinetics, and mechanism. Biotechnol Bioeng 1977;19:95981. [54] Xiao Z, Zhang Z, Gregg DJ, Saddler JN. Effects of sugar inhibition on cellulases and -glucosidase during enzymatic hydrolysis of softwood substrates. Appl Biochem Biotechnol 2004;113116:111526. [55] Ximenes EA, Dien BS, Ladisch MR, Mosier N, Cotta MA, Li X-L. Enzyme production by industrially relevant fungi cultured on co-product from corn dry grind ethanol plants. Appl Biochem Biotechnol 2007;137140:17183.

[25] Panagiotou G, Olsson L. Effect of compounds released during pretreatment of wheat straw on microbial growth and enzymatic hydrolysis rates. Biotechnol Bioeng 2007;96(2):2508. [26] Kumar R, Wyman CE. Effect of supplementation at moderate cellulase loadings on initial glucose and xylose release from corn stover solids pretreated by leading technologies. Biotechnol Bioeng 2009;102(2):4567. [27] Kim Y, Mosier NS, Hendrickson R, Ezeji T, Blaschek H, Dien B, et al. Composition of corn dry-grind ethanol by-products: DDGS, wet cake, and thin stillage. Bioresour Technol 2008;99(12):516576. [28] Kim Y, Hendrickson R, Mosier NS, Ladisch MR, Bals B, Balan V, et al. Enzyme hydrolysis and ethanol fermentation of liquid hot water and AFEX pretreated distillers grains at high-solids loadings. Bioresour Technol 2008;99(12):520615. [29] Shell C, Riley C, Petersen GR. Pathways for development of a biorenewable industry. Bioresour Technol 2008;99(12):51604. [30] Ladisch M, Dale B, Tyner W, Mosier N, Kim Y, Cotta M, et al. Cellulose conversion in dry grind ethanol plants. Bioresour Technol 2008;99(12):51579. [31] Perkis D, Tyner W, Dale R. Economic analysis of a modied dry grind ethanol process with recycle of pretreated and enzymatically hydrolyzed distillers grains. Bioresour Technol 2008;99(12):52439. [32] Schwietzke S, Kim Y, Ximenes E, Mosier N, Ladisch M. Ethanol production from maize. In: Kriz AL, Larkins BA, editors. Biotechnology in agriculture and forestry, molecular genetic approaches to maize. Berlin, Heildelberg: Springer; 2009. p. 34764. [33] Ladisch MR, Kohlmann K, Westgate P, Weil J, Yang Y. Process for treating cellulosic material. US Patent 5,846,787; December 8, 1998. [34] Weil JR. Effects of hydrothermal treatment of cellulosic material and structural interactions of enzymes. Ph.D thesis. West Lafayette, IN: Purdue University; 1997. [35] Weil JR, Sarikaya A, Rau S-L, Goetz J, Ladisch CM, Brewer M, et al. Pretreatment of corn ber by pressure cooking in water. Appl Biochem Biotechnol 1998;73:117. [36] Weil JR, Brewer M, Hendrickson R, Sarikaya A, Ladisch MR. Continuous pH monitoring during pretreatment of yellow poplar wood sawdust by pressure cooking. Appl Biochem Biotechnol 1998;7072:99111. [37] Kim Y, Hendrickson R, Mosier NS, Ladisch MR. Liquid hot pretreatment of cellulosic biomass. Methods Mol Biol 2009;581:93102. [38] Kohlmann KL, Westgate P, Velayudhan A, Weil J, Sarikaya A, Brewer MA, et al. Enzyme conversion of lignocellulosic plant materials for resource recovery in a controlled ecological life support system. Adv Space Res 1996;18(12):25165. [39] Mosier N, Wyman C, Dale B, Elander R, Lee YY, Holtzapple M, et al. Features of promising technologies for pretreatment of lignocellulosic biomass. Bioresour Technol 2005;96:67386.