Influence of xylan on the enzymatic hydrolysis of steam-pretreated corn stover and hybrid poplar

Authors


Abstract

The focus of this study was to alter the xylan content of corn stover and poplar using SO2-catalyzed steam pretreatment to determine the effect on subsequent hydrolysis by commercial cellulase preparations supplemented with or without xylanases. Steam pretreated solids with xylan contents ranging from ∼1 to 19% (w/w) were produced. Higher xylan contents and improved hemicellulose recoveries were obtained with solids pretreated at lower severities or without SO2-addition prior to pretreatment. The pretreated solids with low xylan content (<4% (w/w)) were characterized by fast and complete cellulose to glucose conversion when utilizing cellulases. Commercial cellulases required xylanase supplementation for effective hydrolysis of pretreated substrates containing higher amounts of xylan. It was apparent that the xylan content influenced both the enzyme requirements for hydrolysis and the recovery of sugars during the pretreatment process. © 2009 American Institute of Chemical Engineers Biotechnol. Prog., 2009

Introduction

Lignocellulosic biomass has the potential to become a major source of fermentable sugars for the production of bioethanol. It has been estimated that in the United States alone, more than one billion tons per year of biomass could be sustainably harvested in the form of crops and forest residues, which has the potential to replace as much as 30% of total U.S. gasoline consumption.1 To satisfy the biomass requirements for future commercial bioethanol operations, a diverse range of pretreated plant material from agricultural, hardwood, and softwood biomass which contain different amounts and types of hemicellulosic carbohydrates will most likely be required. The bioconversion of lignocellulosic biomass to ethanol consists of three main steps. Pretreatment, to improve accessibility of the substrate to subsequent hydrolysis by cellulases and finally fermentation of the resulting sugars to ethanol. Depending on the conditions employed during pretreatments, such as steam (SP), hemicellulosic sugars can either be preserved in a polymeric form in the solid substrate or as the severity of the pretreatment is increased, the hemicellulose derived sugar are primarily recovered as monomers in the water-soluble stream. Under certain conditions, some of the sugars can be degraded2 resulting in decreased sugar and ethanol yields.3 Variations in the type of hemicellulosic material left in the insoluble, predominantly, cellulosic substrate will likely influence the hydrolysis step, and hemicellulosic hydrolytic activities would also likely aid in achieving good overall hydrolysis of the insoluble component obtained after steam pretreatment. It has been shown that the hemicellulose preserved in the substrate after pretreatment hinders access of cellulases to the cellulosic substrate, and thus decreases overall hydrolysis yields.4, 5 Therefore, in steam pretreatment, a good compromise would be to balance the pretreatment severity so as to solubilize the hemicellulose component without degradation of the component sugars, thus leaving them in a fermentable form. However, as has been shown, this compromise remains a significant challenge especially when pretreating diverse biomass sources.5, 6

Steam pretreatment has been proposed as an efficient pretreatment method for lignocellulosic materials because of its limited use of chemicals, low energy consumption, and short reaction time.6, 7 Steam pretreatment (explosion) combines both physical and chemical elements, causing the rupture of the wood cell wall structure, hydrolysis, and solubilization of the biomass components.8 Previous work has shown that steam pretreatment can successfully fractionate agricultural, hardwood, and softwood biomass, resulting in the recovery of most of the original hemicellulose- and cellulose-derived sugars in a hydrolyzable and fermentable form.9–11 Sulfuric acid and sulfur dioxide (SO2) can be utilized as pretreatment catalysts when necessary. The impregnation of biomass with SO2 prior to pretreatment has been shown to reduce the reaction time and temperature and consequently reduce the formation of sugar degradation products such as furfurals and hydroxymethyl furfurals (HMFs).12 The use of SO2 as a catalyst during steam pretreatment of Douglas-fir enhanced both subsequent cellulose hydrolysis and recovery of hemicellulose-derived sugars.13, 14 In addition, adding SO2 prior to steam pretreatment not only increased the hydrolysis rate but also reduced the degree of polymerization of the oligomers and increased the proportion of monomers in the water soluble stream.15 Although SO2-catalyzed steam pretreatment has been shown to be effective for the pretreatment of agricultural and woody biomass, it is recognized that different pretreatment conditions are required to treat each type of biomass.11, 16, 17

It has been shown that by employing a low pretreatment severity, defined by the severity factor Ro {Ro = texp (T − 100)/14.75} which links the effects of time (t, min) and temperature (T, °C)18}, only limited chemical and physical changes occur in the pretreated biomass.3, 10, 17, 19 As the severity of pretreatment increases, first the hydrolysis of hemicellulose occurs, releasing oligomeric and monomeric sugars. Eventually, as the severity is raised, the cellulose will begin to hydrolyze. As sugars are hydrolyzed during the steam pretreatment process, there is a concomitant increase in the concentration of lignin in the solid fraction.3 One of the main issues associated with optimizing pretreatment conditions for any feedstock is finding the optimum conditions at which hydrolysis of oligomeric sugars to monomers occurs, with a minimal formation of sugar degradation products. This is even further complicated by the fact that the optimum treatment severities for maximum recovery of hemicellulose, lignin, and cellulose are all different.20 By better understanding the fractionation of lignocellulosic biomass process during steam pretreatment, we should be able to modify the chemical composition of the pretreated substrates, thus altering the ease of hydrolysis of the resulting substrates by cellulases.

The hemicellulosic component of lignocellulosic biomass, xylan in particular, has been frequently cited as a physical barrier around cellulose within the lignocellulosic matrix, restricting access of cellulase to cellulose.2, 5, 21 In the past, significant efforts have been made to improve the efficiency of known enzymes, identifying new, more active enzymes and cofactors, minimizing enzyme production cost, or creating enzyme mixtures optimized for selected pretreated substrates.22 Since the hemicellulosic sugar, composition, in particular the xylan content of hardwood and agricultural biomass ranges from 10 to 30%, the development of improved enzymes for hydrolysis of hemicellulose have been targeted toward these substrates.22 Xylanase supplementation to commercial cellulase preparations has been shown to yield xylooligomers and xylose, as the xylanase activities detected in most of the commercial enzyme preparations are not often sufficient to obtain a complete conversion of xylan to xylose.4 In this context, rather than customizing the enzyme mixture for a specific substrate, we have taken the complementary approach of changing the chemical composition of pretreated poplar and corn stover substrates to increase the likelihood of successful hydrolysis by commercial enzyme preparations by varying the severity of the steam pretreatment process. We first examined the changes in the xylan content of pretreated corn stover and hybrid poplar that were imparted by varying pretreatment conditions. We subsequently study the behavior of the pretreated substrates during enzymatic hydrolysis with cellulases, β-glucosidases, and xylanases.

Materials and Methods

Raw material

Corn stover and hybrid poplar were kindly supplied by National Renewable Energy Laboratory (NREL). The moisture content of corn stover and hybrid poplar was ∼10% (w/w). The corn stover was ground, screened and the fraction collected between 9 and 35 mesh was used in all the steam pretreatment experiments. Hybrid poplar chips were screen trough 1/4 inch mesh. Composition of corn stover and hybrid poplar is detailed in Table 1.

Table 1. Chemical Composition of Corn Stover and Poplar
 % Total Dry Weight (w/w)
Corn StoverHybrid Poplar
Arabinan (%)4.20.6
Galactan (%)1.41.0
Glucan (%)34.443.8
Xylan (%)22.814.9
Mannan (%)0.63.9
Total lignin (%)17.229.1

Pretreatment

The conditions for steam pretreatment for corn stover and hybrid polar shown in Table 2 were chosen from five different sets of conditions for corn stover and four different sets of conditions for hybrid poplar (not shown). The range of severities for corn stover and poplar were selected to produce water-soluble and insoluble fractions at a range of xylan levels. The names for each represent severity of the pretreatment conditions in terms of the applied severity factor log Ro. The severity range for corn stover ranged from log Ro 3.0 to 4.2 and for hybrid poplar from log Ro 3.4 to 4.1. Corn stover pretreated at M condition was also pretreated without impregnation with SO2. Prior to steam explosion, corn stover and hybrid poplar were impregnated with sulfur dioxide in the amount as shown in Table 2 by adding SO2 to plastic bags containing 300 g dry weight of biomass. The bags were weighed and left at room temperature overnight. The impregnated biomass was added to the reactor of a 2-L Stake Tech II gun in 50 g aliquots that were treated at the specified temperature and time shown in Table 2. After 300 g dry weight of biomass had been discharged to the collecting vessel, the resulting slurry was removed and stored at 4°C. The water soluble fraction (WSF) was separated from water insoluble fraction (WIF) using vacuum filtration. The WIF was then washed with the volume of water equivalent to 20 times the dry weight of the sample. Monomeric and oligomeric sugar concentrations were determined for the wash liquid, WIF, and WSF to calculate the overall sugar recovery.

Table 2. Steam Pretreatment Conditions for Corn Stover and Hybrid Poplar
PretreatmentCorn StoverHybrid Poplar
Severity (Ro)ConditionsSeverity (Ro)Conditions
Low (L)3.0170°C, 9 min, 3% SO23.4190°C, 5 min, 3% SO2
Medium (M)3.4190°C, 5 min, 3% SO2 (190°C, 5 min, 0% SO2)3.6200°C, 5 min, 3% SO2
High (H)4.2210°C, 7.8 min, 3% SO24.1215°C, 5 min, 3% SO2

Enzymatic hydrolysis

All experiments were carried out in 125-mL Erlenmeyer flasks in duplicate and the range reported. Total solution volume was 50 mL. Washed solids were diluted to 8% (w/v) consistency (unless specified otherwise) with acetate buffer (50 mM, pH 5) at 50°C and 150 rpm. Enzymes were added in the form of cellulase at 10 FPU/g cellulose (Spezyme-CP, Genencor-Danisco, Palo Alto, CA) with the protein content of 123 mg/mL, and β-glucosidase at 20 CBU/g of cellulose (unless specified otherwise) (Novozymes 188, Bagsværd, Denmark) with the protein content of 12.5 mg/mL. Samples (500 μL) were taken periodically over 72 h, boiled for 5 min and stored at −20°C. To examine the effect of xylanase supplementation on saccharification of pretreated corn stover and poplar, Multifect® Xylanase (Genencor-Danisco) with a protein content of 43.7 mg/mL (8.000 GXU) mL was used at a concentration of 0.06 g protein/g cellulose. Multifect Xylanase is a commercial xylanase derived from a genetically modified strain of T. reesei. The series of controls for testing the effects of Multifect Xylanase on hydrolysis progress included addition of bovine serum albumin (BSA) or cellulases (Spezyme-CP) at the same protein loading to ensure that the increased digestibility was due to the xylanase activity in the Multifect Xylanase, rather from additional cellulases activity or a nonspecific binding protein effect. Cellulase enzyme (Spezyme-CP, 59 FPU/mL-filter paper activity) and xylanase cocktail (Multifect Xylanase) were kindly provided by Genencor-Danisco. Novozymes-188 β-glucosidase (780 CBU/mL-specific activity) was a kind gift of Novozymes, Denmark.

Sample analysis

The concentration of monomeric sugars (arabinose, galactose, glucose, xylose, and mannose) was determined by HPLC analysis. The HPLC system (Dionex DX-2500, Dionex Corp., Sunnyvale, CA) was equipped with an ion-exchange Carbopac PA-1 column (4 mm × 250 mm) equilibrated with 0.25 M NaOH and eluted with nanopure water at a flow rate of 1 mL/min (Dionex Corp.), an ED40 electrochemical detector (gold electrode), AD20 absorbance detector, and autosampler (Dionex Corp.). Sodium hydroxide (0.2 M) was added postcolumn to enhance the detection. Before injection, samples were filtered through 0.45-μm NYL filters (Millipore, U.S.) and a volume of 20 μL was injected. Analytical-grade standards of L-arabinose, D-galactose, D-glucose, D-xylose, and D-mannose (Sigma) were used to quantify the concentration of sugars. In addition, L-fucose (Sigma) was used as an internal standard.

Posthydrolysis analysis of all liquid samples allowed quantification of the amount of oligomeric sugars present. 0.7 mL of 70% H2SO4 was added to 15 mL of the liquid samples, and the volume was made up to 20 mL with water. Samples were autoclaved at 121°C for 1 h and analyzed by HPLC as described earlier.

Solid samples were analyzed in triplicate for insoluble (Klason) lignin and sugars using the modified Tappi T-222 on-88 method as previously described.3 The hydrolysate from this analysis was retained and analyzed for soluble lignin by absorbance at 205 nm and sugars using HPLC as described earlier.

Statistical paired T-test was performed to test the effects of Multifect Xylanase supplementation on the cellulose to glucose and xylan to xylose conversion compared to the controls.

Results and Discussion

Chemical composition of solid fraction and liquid fraction

Corn stover and hybrid poplar were utilized in this study because both of these biomass sources possess high xylan contents, 15 and 23% for hybrid poplar and corn stover, respectively. Both the corn stover and hybrid poplar have also been successfully pretreated by SP with xylose yields of 78% and 70% for corn stover and poplar, respectively.16, 17 Moreover, corn stover is an abundant agricultural by-product in North America, and hybrid poplar has been well recognized as an energy crop.23, 24 Corn stover and poplar are characterized by different types of hemicellulose. Arabinoxylans have been identified as the main hemicellulose in corn stover as well in wheat, rye, barley, oat, rice, and sorghum,25, 26 whereas the most abundant hemicellulose constituent of poplar is glucuronoxylan.26 The backbone of this hardwood xylan consists of β-(1→4)-D-xylopyranosyl residues, with, on average, one α-(1→2)-linked 4-O-methylglucuronic acid (MeGlcA) substituent per 10–20 such residues.27 Many of the xylose residues also contain an O-acetyl group at position C-2 and/or C-3. The content of such O-acetyl groups ranges from 9 to 17 wt %, which corresponds to approximately four to seven acetyl groups per 10 xylopyranosyl residues.27 Xylan from corn stover has the same backbone as hardwood xylans, however, it is more branched and contains larger proportions of L-arabinofuranosyl units.26 In general, arabinofuranosyl units are attached to some C-3 positions of the main xylan chain and glucuronic acid and/or its 4-O-methyl ether linked to some xylose units.

Corn stover and hybrid poplar were pretreated at low (L), medium (M), and high (H) severities (see Table 2) to produce WSF and WIF with a range of chemical compositions. In addition, to determine the effect of SO2 addition on the relegation of xylan to either the liquid fraction or the WIF, the corn stover was also pretreated at M severity without SO2 supplementation. The chemical compositions of WIFs after steam pretreatment of corn stover and hybrid poplar are shown in Table 3. Since arabinose, galactose, and mannose make up only a small portion of total sugars in corn stover and poplar, only the two main sugars, glucan and xylan, were considered in this study (Tables 1 and 3). As the severity of the pretreatment was increased, the concentration of glucan in WIF increased from 49.1 to 56.9% and from 54.3 to 57.5% for corn stover and poplar, respectively. The increase in glucan content was most likely due to the fact that at increased pretreatment severity, the xylan content decreased from 18.8 to 5.3% and from 3.6 to 1.3% for corn stover and poplar, respectively (Table 3). The lower xylan content of pretreated hybrid poplar when compared with corn stover could be attributed to a 7.9% lower xylan content in the original biomass (Table 1). The increased glucan and decreased xylan content with increased severity of steam pretreatment correlated well with results described previously for similar pretreatment severities of corn fiber, barley straw, and wheat straw.2, 3, 28 Although the glucan and lignin content in untreated hybrid poplar were 9.4 and 12.1% higher compared with corn stover, the amount of glucan and lignin in pretreated biomass at medium severity were quite similar as follows: 56 and 55%, and 29 and 31% for corn stover and hybrid poplar respectively (Table 3). By varying the severity of SO2-catalyzed steam pretreatment for corn stover and hybrid poplar, solid substrates were produced at a range of xylan contents while maintaining similar levels of glucan and lignin content for M severities for both substrates.

Table 3. Chemical Composition of Corn Stover and Hybrid Poplar WIF Pretreated at Different Severities
PretreatmentCorn StoverHybrid Poplar
Glucan (%)Xylan (%)Total Lignin (%)Glucan (%)Xylan (%)Total Lignin (%)
  • *

    Chemical composition of corn stover pretreated at 190°C, 5 min, 0% SO2.

Low49.118.824.254.33.629.9
Medium56.0 (55.7)*9.5 (18.4)*28.6 (25.4)*55.12.530.5
High56.95.331.157.51.335.7

We have shown that by varying the severity of SP, we can keep the hemicellulose with the WIF or WSF. Since it has been shown that by utilizing SO2 during pretreatment the proportion of monomers in the water-soluble stream can be increased,15 we next tested the steam pretreatment of corn stover with and without addition of SO2 as a catalyst. The steam pretreatments of corn stover were performed at 190°C, 5 min, with and without addition of SO2. Although both in the presence and absence of SO2 the amount of glucan in the WIF was identical (56%), the amount of xylan differed, from 9.5 to 18.4%, in the presence vs. absence of SO2, respectively (Table 3). In the absence of SO2 impregnation, the predominant reaction during the pretreatment of the corn stover was most likely autohydyrolysis. Autohydrolysis is highly effective at hydrolyzing the hemicelluloses of agricultural and hardwood biomass in which, as reported by Nabarlatz et al.,29 the acetic acid side-groups associated with the hemicellulose inherently found in hemicellulose act as a catalyst during pretreatment. When relying on the autohydrolysis process during steam pretreatment of corn stover, we solubilized only 3% of xylose, thus preserving the majority of xylan in the lignocellulosic matrix, while the addition of SO2 resulted in a 42% of xylan being solubilized. In addition, to obtain the composition of the WIF substrates from the steam pretreatment of corn stover and poplar, a better idea of the fate of sugars can be obtained through the analysis of the WSF.

It was apparent that as the pretreatment severity was increased, the concentration of xylose and glucose in WSF increased for both substrates (Table 4). Similar results were reported for wheat straw, corn fiber, and corn stover.2, 3, 17 In addition, more xylose rather than glucose was found in the liquid stream since hemicellulose is more labile and becomes solubilized during the pretreatment. Although the glucose content in WSF obtained at L, M, and H severities was lower for poplar (from 1.5 to 5.3 g/L) when compared with corn stover (from 4.9 to 7.1 g/L), the amounts of xylose released at M and H severities were identical for both substrates. In each of the seven pretreatment conditions, over 60% of hemicellulose-derived sugars were recovered in oligomeric form in WSF (data not shown). Similar results were reported by Allen et al.,30 in which low concentrations of monomeric hemicellulose-derived sugars and high concentrations of oligomers were observed during the pretreatment of corn fiber by hot water and steam fractionation. Although we have shown that steam pretreatment of corn stover at 190°C, 5 min, with and without catalyst generated WIF with differing amounts of residual xylan, the amount of xylose in the WSFs also differed. Not surprisingly, 94% of the xylose recovered in WSF was present in oligomeric form as compared with 77% when the steam pretreatment was performed with the addition of SO2 as the catalyst (data not shown). These results are similar to those observed by Clark et al.,15 We have shown that the chemical composition of the pretreated corn stover and hybrid poplar could be altered by varying the severity of steam pretreatment or through the addition/subtraction of the SO2 catalyst. If a solid substrate with low xylan content is required, we could either increase the severity of pretreatment or use SO2 as a catalyst. However, when altering the chemical composition of pretreated biomass, monitoring overall sugar recovery is required, since it is well known that as pretreatment severity increases, the overall sugar recovery and, in particular, hemicellulose decreases due to sugar degradation.2, 17

Table 4. Chemical Composition of Corn Stover and Hybrid Poplar WSF Pretreated at Different Severities
PretreatmentCorn StoverHybrid Poplar
Glucose (g/L)Xylose (g/L)Glucose (g/L)Xylose (g/L)
  • *

    Chemical composition of corn stover pretreated at 190°C, 5 min, 0% SO2.

Low4.911.21.519.3
Medium5.9 (5.1)*22.2 (17.4)*2.421.7
High7.123.15.323.7

As expected, when the severity of pretreatment increased, less xylan remained within WIF, and thus the combined (WSF+WIF), overall xylan recovery decreased for corn stover and poplar (Table 5). This phenomenon was very clearly illustrated during pretreatment of corn stover at M severity with and without SO2 addition, in which keeping the xylan with the solids increased sugar recovery by 34% (Table 5). The residual xylan present in WIF seemed to be a good indication of overall hemicellulose recovery as a function of pretreatment severity, as was shown by Kabel et al.,2 Although keeping the xylan with the solid fraction resulted in an improvement to the overall sugar recovery, both the corn stover and poplar pretreated at L and M severities were characterized by similar overall carbohydrate recovery. However, goals of substrate pretreatment include maximizing sugar recovery and also producing a substrate responsive to subsequent hydrolysis by hydrolytic enzymes. A substrate pretreated with a lower severity may result in a greater overall sugar recovery, but results in a WIF fraction recalcitrant to cellulose hydrolysis. Therefore, ease of hydrolysis of the WIF from both the corn stover and poplar biomass produced with varying xylan contents was assessed.

Table 5. The Combined Glucan and Xylan Recovery of WSF and WIF After Steam Pretreatment of Corn Stover and Hybrid Poplar (assuming 100% as maximum recovery for each sugar)
PretreatmentCorn StoverHybrid Poplar
Glucan Recovery (%)Xylan Recovery (%)Glucan Recovery (%)Xylan Recovery (%)
  • *

    Glucan and xylan recovery for corn stover pretreated at 190°C, 5 min, 0% SO2.

Low99839397
Medium100 (100)*64 (98)*10089
High88458868
Table 6. An Increase in Cellulose to Glucose and Xylan to Xylose Conversion During Supplementation of Hydrolytic Mixture with Cellulases, β-Glucosidases, and Multifect® Xylanase (0.06 g of protein/g of cellulose) at 8% (w/v) Solids Consistency and 10 FPU/g of Cellulose Enzyme Loading and IU:FPU of 2:1
PretreatmentCorn StoverHybrid Poplar
Increase in Cellulose Conversion (%)Increase in Xylan Conversion (%)Increase in Cellulose Conversion (%)Increase in Xylan Conversion (%)
  1. The series of controls for this set of experiments included addition of BSA or cellulases (Spezyme CP) at the same protein loading. Based on the paired T-test, there was no difference between control and tested group for hybrid poplar (P > 0.05).

Low9.09.200
Medium8.418.300
High5.520.500

Enzymatic hydrolysis of corn stover and poplar

An increase in cellulose to glucose conversion was seen with increasing pretreatment severity for corn stover and poplar, which has been previously observed in the case of agricultural, hardwood, and softwood biomass (Figures 1A,B).17, 31, 32 It has been shown that agricultural biomass is more responsive to steam pretreatment and results in a solid substrate that is easier to hydrolyze than hardwood biomass.32, 33 However, somewhat surprisingly, the cellulose to glucose conversion occurred at a faster rate in the case of steam-pretreated poplar compared with corn stover for all the conditions tested. Complete cellulose to glucose conversion could be achieved after 28 h of saccharification for M- and H-pretreated poplar solids, whereas 100% cellulose conversion was observed after 72 h of hydrolysis for H-pretreated corn stover (Figure 1A). Moreover, at all pretreatment conditions, the hydrolysis yields after the first 12 h were higher for all the poplar samples when compared with corn stover. Although glucan content in all the samples was similar (50–58%), the low xylan content in poplar (1–3%) when compared with corn stover (5–20%) might have contributed to the higher saccharification rate of the pretreated poplar, since it has been shown that the percentage of residual xylan appears to be a good indicator concerning cellulose digestibility.2, 5 It has also been shown that xylan when absorbed onto the surface of cellulose can hinder the action of cellulases.2 Since the amount of lignin was similar for all the pretreated samples (Table 3), enhanced cellulose to glucose conversion may be due to xylan removal. Similar results have been shown by Jeoh et al.,5 in which decreasing the xylan content in WIF during dilute acid pretreatment of corn stover improved the extent of cellulose conversion.

Figure 1.

Cellulose to glucose (A) and xylan to xylose (B) conversion during enzymatic hydrolysis of corn stover (L, M, and H) and poplar (L, M, and H) at 1% (w/v) solids consistency and 15 FPU/g of cellulose enzyme loading and IU:FPU of 2:1.

Similar to the cellulose conversion, an increase in the xylan to xylose conversion was seen with increasing process severity for corn stover. The xylose digestibility profiles for poplar pretreated at L and M severities were virtually identical, as 50% conversion of xylan to xylose was achieved after 72 h of hydrolysis (Figure 1B). The already low xylan content ranging from 1.3 to 3.6% in the case of the poplar substrates may have contributed to their poor xylose digestibility when compared with the corn stover substrates, where 63, 86, and 98% conversions were achieved for L, M, and H severities, respectively (Figure 1B). Since the commercial cellulase cocktail used in this study had a low xylanase activity (data not shown), we next tested the effects of xylanase supplementation on cellulose to glucose and xylan to xylose conversions of WIF from the steam pretreated corn stover and poplar.

Xylanase supplementation

Although there has been significant progress in hemicellulose hydrolysis research,34 the effects of xylanase loading on cellulose to glucose and xylan to xylose conversion still remains undefined. The effects of Multifect Xylanase loading on cellulose to glucose and xylan to xylose conversion were assessed for steam pretreated corn stover at 190°C, 5 min, both with and without impregnation of the 3% SO2. The series of controls for this set of experiments included the addition of BSA or cellulases (Spezyme CP) at equivalent protein loadings to the supplemented xylanases to ensure that the increased digestibility was due to the activity of the xylanases preparation, rather than from additional cellulase activity or a nonspecific binding protein effect as been shown previously by Yang and Wyman.35 As the Multifect Xylanase loading was increased from 0.006 to 0.06 g of protein/g of cellulose, the cellulose to glucose conversion increased from 4 to 12%, respectively, when compared with the control hydrolysis (Figures 2A,B). However, further increases in xylanase loading beyond the 0.06 g/g cellulose did not improve cellulose digestibility. Interestingly, it was also apparent that at loadings beyond 0.09 g of protein/g of cellulose, the addition of the xylanase preparation decreased cellulose to glucose conversion (Figure 2A). A similar pattern was observed during xylan to xylose conversion, in which high (over 0.06 g of protein/g of cellulose) Multifect Xylanase loading reduced xylose digestibility. However, not surprisingly, Multifect Xylanase loading from 0.006 to 0.06 g of protein/g of cellulose greatly improved xylan conversion from 10 to 30%, respectively, when compared with the control hydrolysis. A possible explanation for the observation might include the competition for cellulosic binding sites between added xylanases and the cellulases in the cellulase preparation, as suggested by Berlin et al.,4 who showed negative effects on cellulose hydrolysis at higher levels of supplementation during enzymatic hydrolysis of dilute acid-pretreated corn stover. The underlying mechanism responsible for this phenomenon is currently unknown, but it is evident that oversupplying the saccharification mixture with xylanases results in a decrease in hydrolysis yields.

Figure 2.

The increase in cellulose to glucose (A) and xylan to xylose (B) conversion during hydrolysis of corn stover pretreated at M severity (1%, w/v), 10 FPU/g of cellulose enzyme loading and IU:FPU of 2:1 with increasing Multifect® Xylanase supplementation from 0 to 0.12 g of protein/g of cellulose.

We have shown that by performing the steam pretreatment of corn stover at M severity with and without addition of SO2, we can create a substrate with identical glucan content and a twofold difference in xylan content (Table 3). When employing cellulase and β-glucosidase mixtures without xylanase supplementation, the cellulose to glucose conversion was identical for the two corn stover substrates (Figure 3A). However, by supplementing the saccharification mixture with the xylanase (0.06 g of protein/g of cellulose), the conversion increased by 22 and 32%, for cellulose to glucose and for xylan to xylose cconversion, respectively. It was apparent that xylanases improved the access of cellulases to the substrate that had a higher xylan content as suggested previously by Berlin et al.,4 and Merino and Cherry.22

Figure 3.

The influence of xylanase supplementation on glucan to glucose (A) and xylan to xylose (B) conversion during hydrolysis of corn stover pretreated at M severity with and without SO2 supplementation.

The hydrolytic mixture consisted of cellulases, β-glucosidases and Multifect® Xylanases (0.06 g of protein/g of cellulose), 8% (w/v) solids consistency, and 10 FPU/g of cellulose enzyme loading and IU:FPU of 2:1.

Since we produced a substrate characterized by very low residual xylan content and low xylan to xylose conversion (∼50%) utilizing cellulases and β-glucosidases during steam pretreatment of hybrid poplar (Table 3 and Figure 1B), we next tested the digestibility of poplar pretreated solids at L, M, and H severities with additional Multifect Xylanase supplementation. Although xylanase supplementation increased cellulose to glucose and xylan to xylose conversion for corn stover, from 6 to 9% and from 9 to 21% for glucose and xylose, respectively, based on paired T-test results, no statistically significant improvements in glucose and xylose release were observed for poplar pretreated at all the tested pretreatment severities (Table 6). It has been suggested that xylanases stimulate cellulose hydrolysis by removing noncellulosic polysaccharides that coat cellulose fiber.4, 22 Since 76% of the xylan was solubilized during steam pretreatment of hybrid poplar, which resulted in WIF samples with low (<4%) xylan content, we likely eliminated the hemicellulose derived physical barrier impeding the accessibility of cellulases to cellulose. This is also supported by the fact that already complete cellulose conversions were observed during the hydrolysis of the pretreated poplar with cellulases and β-glucosidases (Figure 1A,B). Thus, our results indicate that similar to a previous work by Jeoh et al.,5 a pretreatment process that removes hemicellulose to allow more cellulase–cellulose interaction would greatly improve the enzyme digestibility of the biomass sample. However, based on the results here, it must be considered that the dissolution of xylan either by increasing process severity or by increasing SO2 impregnation during steam pretreatment may have significant implications for the recovery of carbohydrates during the process.

Conclusions

In this study, we altered the hemicellulose content of corn stover and poplar via steam pretreatment to determine the implications for subsequent hydrolysis by commercial cellulase preparations with and without xylanase supplementation. We have shown that by employing a lower pretreatment severity, we can produce a solid substrate with a higher xylan content, and thus increased polymeric hemicellulose recovery. Similarly, by performing steam pretreatment experiments without addition of SO2, the water-insoluble substrate would be characterized by high xylan content and complete glucan and xylan recoveries could be possible.

The reduction in residual xylan in the substrate improved the cellulose digestibility during saccharification with cellulases. However, the hydrolysis of the substrates (L, M with and without SO2 and H pretreated corn stover) with higher xylan contents was improved by supplementing the cellulases with xylanases. There was no improvement in glucan to glucose and xylan to xylose conversion for low xylan substrates (L-, M-, and H-pretreated poplar). Although the solubilization of xylan during pretreatment slightly decreased xylan recovery, the fast and complete cellulose to glucose conversion during hydrolysis that did not require additional xylanases may offset the sugar losses occurring during pretreatment.

Acknowledgements

The authors acknowledge the financial support for this research from Natural Resources Canada. They also thank Genencor-Danisco (Palo Alto, CA, U.S.) and Novozymes (Bagsværd, Denmark) for providing the enzymes used in this study, and NREL for providing the feedstock.

Ancillary

Advertisement