Bioresource Technology 101 (2010) 4472–4478

Contents lists available at ScienceDirect

Bioresource Technology

journal homepage: www.elsevier .com/locate /bior tech

Effect of lime pre-treatment on the synergistic hydrolysisof sugarcane bagasse by hemicellulases

Natasha Beukes, Brett I. Pletschke *

Department of Biochemistry, Microbiology and Biotechnology, Rhodes University, Grahamstown 6140, South Africa

a r t i c l e i n f o a b s t r a c t

Article history:Received 11 November 2009Received in revised form 14 January 2010Accepted 19 January 2010Available online 13 February 2010

Keywords:HemicellulasesLignocellulosesL-arabinofuranosidase (ArfA)Mannanase (ManA)Xylanase (XynA)

0960-8524/$ - see front matter � 2010 Elsevier Ltd. Adoi:10.1016/j.biortech.2010.01.081

* Corresponding author. Tel.: +27 46 6038081; fax:E-mail address: b.pletschke@ru.ac.za (B.I. Pletschk

Agricultural crop wastes are typically lignocellulosic in composition and thus partially recalcitrant toenzymatic degradation. The recalcitrant nature of plant biomass and the inability to obtain completeenzymatic hydrolysis has led to the establishment of various pre-treatment strategies. Alkaline pre-treat-ments increase the accessibility of the exposed surface to enzymatic hydrolysis through the removal ofacetyl and uronic acid substituents on hemicelluloses. Unlike the use of steam and acid pre-treatments,alkaline pre-treatments (e.g. lime) solubilise lignin and a small percentage of the hemicelluloses. Themost common alkaline pre-treatments that are employed make use of sodium hydroxide and lime. Thisstudy compared the synergistic degradation of un-treated and lime pre-treated sugarcane bagasse usingcellulosomal and non-cellulosomal hemicellulases as free enzymes. The enzyme combination of 37.5%ArfA and 62.5% ManA produced the highest amount of reducing sugar of 91.834 lmol/min for the degra-dation of un-treated bagasse. This enzyme combination produced a degree of synergy of 1.87. The freeenzymes displayed an approximately 6-fold increase in the enzyme activity, i.e. the total amount ofreducing sugar released (593.65 lmol/min) with the enzyme combination of 37.5% ArfA, 25% ManAand 37.5% XynA for the lime pre-treated substrate and a degree of synergy of 2.14. To conclude, this studyindicated that pre-treating the sugarcane bagasse is essential, in order to increase the efficiency of ligno-cellulose enzymatic hydrolysis by disruption of the lignin sheath, that the lime pre-treatment did nothave any dramatic effect on the synergistic relationship between the free enzymes, and that time mayplay an important role in the establishment of synergistic relationships between enzymes.

� 2010 Elsevier Ltd. All rights reserved.

1. Introduction

The possible depletion of fossil fuel resources has sparked a re-newed interest in the potential use of renewable sources such asagricultural waste products for the production of a variety of bio-fuels. Sugarcane bagasse (SCB) is well established as an importantagricultural waste product. Sugarcane bagasse is the solid lignocel-lulosic residue that remains after the sugarcane liquor has been ex-tracted and is commonly burnt as a fuel source; however, SCB mayserve as a valuable renewable source of biomass for the productionof biofuel, as SCB is produced in large quantities in sugar mills andethanol plants (Frollini et al., 2004; Monterio et al., 1998; Paiva andFrollini, 2002; Simkovic et al., 1990; Zarate et al., 2002). Aside frombeing used as an energy source in sugar mills, SCB has been used asthe raw material to generate electricity, and the hydrolysed SCBproducts have also been used in different fermentation processes(Pandey et al., 2000).

ll rights reserved.

+27 46 6223984.e).

Synergism with complex lignocellulose biomass has been ob-served in several enzyme systems, and facilitates an enhancedhydrolytic activity which is greater than the collective sum ofthe activities obtained by individual enzymes (Din et al., 1994;Teeri, 1997). Unfortunately, the synergistic relationships betweencellulosomal enzymes and other cellulosomal and/or non-cellu-losomal enzymes are not well established. These synergy studiesoccurred between a variety of proteins/enzymes such as: scaffol-din proteins, xylanasases, acetyl xylan esterase, a-arabinofura-nosidase, b-galactosidase and mannanases (Beukes et al., 2008;Blum et al., 2000; Ciruela et al., 1998; Kosugi et al., 2002a; Kou-kiekolo et al., 2005; Murashima et al., 2003). Here, synergy wasobserved between cellulosomal cellulases and xylanase fromClostridium cellulovorans when a substrate, such as corn stempowder, was used. Minicellulosomes containing either XynA,EngE, EngH or ExgS were used to degrade corn cell walls. Besidesthe synergistic action between cellulosomal enzymes, synergywas found to occur between specific cellulosomal enzymes andnon-cellulosomal enzymes. This was exemplified by the actionof cellulosomal hemicellulase XynA and non-cellulosomal hemi-cellulases ArfA and BgaA.

N. Beukes, B.I. Pletschke / Bioresource Technology 101 (2010) 4472–4478 4473

Despite the enhanced enzymatic hydrolysis of lignocelluloseresulting from synergistic relationships between the different en-zymes in non-complexed and complex systems, the completehydrolysis of lignocellulose is never achieved due to the recalci-trant nature of plant biomass. The recalcitrant nature of biomassis due to the complexity of the biomass structure. The recalcitrantnature of plant biomass and the inability to obtain complete enzy-matic hydrolysis led to the establishment of various pre-treatmentstrategies. Alkali pre-treatments proceed at ambient temperaturesand at low pressures, which is advantageous as it eliminates thecost of maintaining the high temperatures and pressures that areusually required in other pre-treatments. Alkali pre-treatments in-crease the accessibility of the surface exposed to enzymatic hydro-lysis through the removal of acetyl and uronic acid substituents onhemicellulose (Chang and Holtzapple, 2000). Unlike the use ofsteam and acid pre-treatments, alkali pre-treatments solubilise lig-nin and a small percentage of the hemicellulose (Chang, 2007; Kaarand Holtzapple, 2000).

Previously, the synergistic associations between two hemicellu-lases and EngE from C. cellulovorans on various substrates wereinvestigated (Beukes et al., 2008). Novel synergistic associationswere established between the three recombinant cellulosomal en-zymes XynA, EngE and ManA, in the degradation of SCB, xylan, car-boxymethylcellulose and locust bean gum. The combination withthe molar ratio of 75% XynA to 25% EngE produced the largest de-gree of synergy (4.65) with SCB.

The aim of this study was to extend the previous studyand to determine if synergistic relationships exist between therecombinant C. cellulovorans cellulosomal enzymes (XynA andManA), and the non-cellulosomal enzyme (ArfA) with SCB assubstrate, in the presence or absence of a lime pre-treatmentstep; and to establish the ideal molar ratio of the three recombi-nant enzymes (as free enzymes) to effectively degrade SCB underthese conditions. Arabinofuranosidases are important enzymesand act in synergy with other glycosyl hydrolases to degradearabinose containing polysaccharides. They also exhibit widesubstrate specificity. SCB was chosen as a substrate due to theimportance of the sugarcane industry to the South African agri-cultural sector, and the fact that this feedstock is also importantfor a number of other countries. Once the ideal molar ratio ofcellulases and hemicellulases has been established, it will allowfor the improvement of existing cellulase and hemicellulase pro-ducing strains and also allow for the design of an enzyme cock-tail for the optimal hydrolysis of SCB.

2. Methods

2.1. Substrate

The chemical characterisation of the SCB was performedaccording to established National Renewable Energy Laboratory(NREL) protocols (Hames et al., 2005). The SCB was finely ground,autoclaved for 20 min at 121 �C, then washed to remove residualsugar from the milling process and to remove any potential bacte-rial and fungal spores that were present on the bagasse, and thenfinally air dried.

2.2. Lime pre-treatment

The SCB was milled with a bench top Waring blender to in-crease the surface area of the bagasse. The bagasse was treatedto a solution of calcium hydroxide (lime) (reagent grade lime-Merck, cat. No. 1020470500) in a ratio of 0.4 g lime per gram drybagasse and incubated at 70 �C for 36 h, shaking at 100 rpm. Thepre-treated bagasse was filtered, and washed with distilled water

until a neutral pH of 7.0 was reached. The pre-treated bagassewas air dried and stored in an air tight container.

2.3. Expression and purification of recombinant enzymes

Escherichia coli BL21 (DE3) was grown in 5 ml 2 � YT broth(Yeast, Tryptone and sodium chloride – see Bergey, 1974) for12 h at 37 �C, shaking on a Labcon bench shaker at 200 rpm. Threecultures of the E. coli BL21 (DE3) cells that had been transformedwith one of the recombinant plasmids pET29-arfA, pET 29b-manAand pET 29b-xynA were used as pre-inocula. The pre-inocula wereincubated at 37 �C for 14 h. The pre-inocula were used to inoculate500 ml 2 � YT broth, which was incubated at 37 �C, shaking at200 rpm, until an optical density of 0.8 at 600 nm for ArfA and0.6 at 600 nm was obtained for ManA and XynA. The expressionof the recombinant enzymes was induced by the addition of1 mM isopropyl-b-D-thiogalactoside (IPTG) for ArfA, ManA andXynA. The cultures were incubated at 18 �C for 16 h (Murashimaet al., 2002). E. coli BL21 (DE3) cells harbouring the recombinantplasmids were harvested by centrifugation at 8000g for 20 min at4 �C and resuspended in lysis buffer (50 mM NaH2PO4, 300 mMNaCl, 1 mg lysozyme/ml, pH 8). The solution was incubated onice for 30 min, and the soluble recombinant proteins were subse-quently extracted from the cells by sonication (3 times 10 s with10 s intervals) and centrifugation at 10,000g for 20 min at 4 �C.The clear lysates were applied to Protino� Ni-TED (Macherey–Na-gel) columns. The flow-through was collected and the unboundproteins were removed with two 10 ml washing steps using thewashing buffer (50 mM NaH2PO4, 300 mM NaCl, pH 7), prior tothe elution of the bound proteins with 10 ml elution buffer(50 mM NaH2PO4, 300 mM NaCl, 250 mM imidazole, pH 7). Thepurified enzymes were desalted with 50 mM Tris–HCl buffer (pH7.5) and concentrated with PEG 20,000 (Murashima et al., 2002).The samples were resolved on a 12% reducing SDS–PAGE gel (Lae-mmli, 1970).

2.4. Protein determination

The quantity of protein present was measured using Bradford’sreagent (Bradford, 1976). Bovine serum albumin was used as astandard. The molar quantity of each enzyme was subsequentlycalculated using the theoretical molecular weight.

2.5. Enzyme assays

The enzyme concentrations in the reaction mixture were0.35 mM of ArfA, 0.42 mM for ManA and 0.33 mM for XynA; theseconcentrations being the lowest identical protein concentration inmg/ml of the individual enzymes that were active with the SCB. Awide range of different enzyme combinations, ranging from 0% to100% of the individual enzymes were tested with the SCB as previ-ously described (Beukes et al., 2008). Once the optimal enzymecombination was established, the effect of time on the degree ofsynergy and the production of reducing sugar was determinedfor the optimal enzyme combinations obtained with both the un-treated and lime pre-treated bagasse. The amount of sugar releasedinto solution from the pre-treatment was also determined. The lib-erated reducing sugars were quantified using a dinitrosalicylic acid(DNS) reagent (Miller, 1959). A single standard curve for the DNSassay was prepared using xylose as a suitable standard andexpressing the activity in terms of reducing sugars released as D-xylose equivalents. This standard curve was used to measure indi-vidual activities as well as the combined activity. The enzyme as-says were performed in triplicate, and the activities wereexpressed in units (U), where 1 unit was defined as the quantityof enzyme required to release 1 lmol of reducing sugar per min.

Table 1Chemical characterisation of the sugarcane bagasse samples as a percentage of dry mass.

Total extractives Total lignin Acid insoluble Acid soluble Total sugar Arabinan Glucan Xylan Acetate

Un-treated 4.5 42.4 33.7 8.5 44.0 1.1 28.8 14.2 7.5Lime pre-treated 2.1 32.8 24.1 8.7 50.0 1.0 34.0 15.0 5.6

Results are shown as a percentage of dry mass and had a mass balance greater than 90%. The acetate fraction represents the acetyl groups present on the arabinose andglucose in the hemicelluloses chains during the acid hydrolysis step.

4474 N. Beukes, B.I. Pletschke / Bioresource Technology 101 (2010) 4472–4478

The degree of synergy that was obtained in the synergy studies wascalculated by dividing the sum of the actual activities of the recom-binant enzymes obtained with the enzyme assays with the ba-gasse, by the theoretical sum of the recombinant enzyme activities.

Fig. 1. SDS–PAGE of the purified enzymes. Lane 1 contains the protein marker pEQ-GOLD II, lane 2 contains the purified sample of ArfA (56 kDa), lane 3 contains thepurified sample of ManA (47 kDa) and lane 4 contains the purified sample of XynA(57 kDa).

3. Results and discussion

3.1. Chemical characterisation of SCB

The chemical characterisation of the SCB was kindly performedby Mr. S. Walford according to established National Renewable En-ergy Laboratory (NREL) protocols (Hames et al., 2005). The SCBsamples were dewaxed through three extraction steps using hex-ane, acetone and ethanol with a Soxhlet apparatus. The extractedmaterial was dried and the samples were hydrolysed with 72%(v/v) sulphuric acid and autoclaved to remove the sugar from thesamples. After the hydrolysis the fractions were filtered to removethe insoluble lignin from the sugar solutions. The acid soluble lig-nin present in the sugar solutions were detected by UV–Vis spec-troscopy (Hyman et al., 2007).

Chang and Holtzapple indicated that lime and other alkalinepre-treatments increase the accessibility of lignocellulose to en-zyme hydrolysis by removing the acetyl and uronic acid substitu-ents that may be present on the hemicellulose fraction oflignocellulose (Chang and Holtzapple, 2000). The lignocellulosicpre-treatment with lime is an effective means to increase the sur-face area of the lignocelluloses, exposing the cellulose structure toenzymes (Mosier et al., 2005). The increase in the accessibility tothe cellulose and hemicellulose is believed to be due to the limealtering the structure of lignin, so that the lignin is solubilisedand can be removed from the lignocellulose. This was confirmedwhen comparing the chemical composition of the un-treated andlime pre-treated bagasse (Table 1).

Unlike the use of steam and acid pre-treatments, alkali pre-treatments e.g. lime has been found to solubilise lignin and a smallpercentage of the hemicellulose (Chang, 2007; Kaar and Holtzap-ple, 2000). The lime pre-treated bagasse has approximately 10%less lignin and a higher quantity of sugar relative to percentageof dry mass than that of the un-treated lignin (Table 1). As depictedin Table 1, the percentage of acetate present in the pre-treated ba-gasse was less than that found in the un-treated bagasse. Table 1potentially illustrates that the lime pre-treatment partially solubi-lised and removed some of the lignin and the acetate substituentpresent on the hemicellulose backbone, thus increasing the expo-sure of the lignocellulose structure to enzymatic hydrolysis. The in-crease in the solubilisation of the lignin moiety and the removal ofthe acetate substituent from the bagasse was expected, since sim-ilar results using lime pre-treatment were obtained by Chang(2007) and Kaar and Holtzapple (2000).

3.2. Expression and purification of the recombinant proteins

The expression of the 1,476 nucteotide arfA gene in the pET29vector produces a protein consisting of 492 amino acids with amolecular weight of approximately 56 kDA (Kosugi et al., 2002a).The manA gene encoding the 425 amino acid ManA has a molecular

weight of 47 kDa (Mosier et al., 2005), and XynA had an expectedmolecular weight of 57 kDA (Beukes et al., 2008; Tamaru et al.,2000). The recombinant proteins were expressed in the E. coliBL21 (DE3) cells and the purification of the clear lysates obtainedfrom the IPTG induced cultures harbouring the AfrA, ManA andXynA constructs were carried out using nickel affinity chromatog-raphy as described previously (Beukes et al., 2008). The efficiencyof the purification of the different proteins was assessed on 12%reducing SDS–PAGE, from which it was seen that the proteins thatwere expressed at the expected molecular weights of ArfA(56 kDa), ManA (47 kDa) and XynA (57 kDa) (Fig. 1).

3.3. Enzyme assays

With the increasing interest in the use of renewable energysources, lignocellulosic biomass has attracted a lot of attention asa potential source for biofuel. However, the complete hydrolysisof plant biomass has yet to be achieved, primarily due to the het-erogeneic complexity and composition that contributes to the re-calcitrant nature of lignocellulose (Tamaru et al., 2000). As aresult, a lot of research has revolved around two questions; firstly,what can be used to decrease the recalcitrance of lignocellulose,and secondly, what is needed to facilitate the complete and effi-cient hydrolysis of lignocellulose? This report focused on thehydrolysis of un-treated and lime pre-treated SCB through synergystudies using the hemicellulases: ArfA, ManA and XynA in order toestablish the optimal combination of hemicellulases which willfacilitate the most effective means to hydrolyse the hemicellulosicfraction of the lignocellulose.

Some polysaccharides, especially xylan, may have several typesof sugar substituents attached to the polysaccharide backbone,

0

0.5

1

1.5

2

2.5

0

20

40

60

80

100

120

140

M12

.5 X

87.5

M25

X75

M37

.5 X

62.5

M50

X50

M62

.5 X

37.5

M75

X25

M87

.5 X

12.5

A12.

5 M

87.5

A25

M75

A37.

5 M

62.5

A50

M50

A62.

5 M

37.5

A75

M25

A87.

5 M

12.5

A87.

5 X

12.5

A75

X25

A62.

5 X

37.5

A50

X50

A37.

5 X

62.5

A25

X75

A12.

5 X

87.5

degr

ee o

f syn

ergy

umol

/min

sug

ar p

rodu

ced

Enzyme combinations

Activity Synergy

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

0

20

40

60

80

100

120

A12.

5 M

12.5

X75

A12.

5 M

25 X

62.5

A12.

5 M

37.5

X50

A12.

5 M

50 X

37.5

A12.

5 M

62.5

X25

A12.

5 X

12.5

M75

A25

M62

.5 X

12.5

A37.

5 M

50 X

12.5

A50

M37

.5 X

12.5

A62.

5 M

25 X

12.5

A75

M12

.5 X

12.5

A62.

5 M

12.5

X25

A50

M12

.5 X

37.5

A37.

5 M

12.5

X50

A37.

5 M

12.5

X62

.5

A25

M25

X50

A25

M37

.5 X

37.5

A25

M50

X25

A37.

5 M

37.5

X25

A50

M25

X25

A37.

5 M

25 X

37.5

degr

ee o

f syn

ergy

umol

/min

sug

ar p

rodu

ced

Enzyme combinations

Activity Synergy

a

b

Fig. 2. Enzyme activity with native (un-treated) sugarcane bagasse and the corresponding degrees of synergy. The enzyme assays were set up using enzyme combinations ofArfA (A), ManA (M) and XynA (X) in different molar percentages, where data points represent means ± SD (n = 3). Fig. 2a represents the assays set up with variouscombinations of two enzymes and Fig. 2b represents the assays set up with all three enzymes.

N. Beukes, B.I. Pletschke / Bioresource Technology 101 (2010) 4472–4478 4475

increasing the complexity of the polysaccharide and may be widelydistributed within the plant cell walls (Kosugi et al., 2002b), thusseveral enzymes are required to effectively degrade xylan. Xylan-ases, mannanases and pectinases will act on the hemicellulosebackbone; however, side chain-cleaving enzymes such as L-arabin-ofuranosidase (ArfA) and feruloyl esterases would also be required(Aspinall, 1980; Koukiekolo et al., 2005; Thomson, 1993).

The synergistic associations between the recombinant enzymes(ArfA, ManA and XynA) were determined through the quantifica-tion of the reducing sugars released during the degradation ofpre-treated and un-treated SCB. The synergistic relationships be-tween ArfA, ManA and XynA were also determined by varyingthe molar ratios of the enzymes used in the hydrolysis of thepre-treated and un-treated bagasse (Beukes et al., 2008). For boththe present study and the previous study conducted by Beukeset al. (2008), the various enzymes were standardised to the sameprotein concentration (mg/ml) and the assay volume remained

constant throughout. The trends in activity (specific activity oractivity in lmol/min) for both studies are therefore comparable.

The most effective combination of enzymes was determined byquantifying the amount of reduced sugar present in solution, andthe greatest degrees of synergy were determined by dividing thesum of the actual activities of the recombinant enzymes obtainedwith the enzyme assays with the different bagasse samples, bythe theoretical sum of the recombinant enzyme activities (Figs. 2and 3).

The highest amount of sugar was expected to be released fromthe un-treated bagasse using a combination of all three enzymes;however, the highest amount of sugar quantity was released usingonly two enzymes (i.e. ArfA and ManA) (Fig. 2). This potentiallyindicates that the mannan component of the hemicelluloses maybe juxtaposed to the lignin sheath. These results also indicate thatthe xylan backbone may be buried and thus protected from hydro-lysis under the mannan.

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

0

50

100

150

200

250

300

350

400

450

M12

.5 X

87.5

M25

X75

M37

.5 X

62.5

M50

X50

M62

.5 X

37.5

M75

X25

M87

.5 X

12.5

A12.

5 M

87.5

A25

M75

A37.

5 M

62.5

A50

M50

A62.

5 M

37.5

A75

M25

A87.

5 M

12.5

A87.

5 X

12.5

A75

X25

A62.

5 X

37.5

A50

X50

A37.

5 X

62.5

A25

X75

A12.

5 X

87.5

degr

ee o

f syn

ergy

umol

/min

sug

ar p

rodu

ced

Enzyme combinations

Activity Synergy

0

0.5

1

1.5

2

2.5

0

100

200

300

400

500

600

700

A12.

5 M

12.5

X75

A12.

5 M

25 X

62.5

A12.

5 M

37.5

X50

A12.

5 M

50 X

37.5

A12.

5 M

62.5

X25

A12.

5 M

75 X

12.5

A25

M62

.5 X

12.5

A37.

5 M

50 X

12.5

A50

M37

.5 X

12.5

A62.

5 M

25 X

12.5

A75

M12

.5 X

12.5

A62.

5 M

12.5

X25

A50

M12

.5 X

37.5

A37.

5 M

12.5

X50

A37.

5 M

12.5

X62

.5

A25

M25

X50

A25

M37

.5 X

37.5

A25

M50

X25

A37.

5 M

37.5

X25

A50

M25

X25

A37.

5 M

25 X

37.5

degr

ee o

f syn

ergy

umol

/min

sug

ar p

rodu

ced

Enzyme combinationsActivity Synergy

a

b

Fig. 3. Enzyme activity with the lime pre-treated sugarcane bagasse and the corresponding degrees of synergy. The enzyme assays were set up using enzyme combinations ofArfA (A), ManA (M) and XynA (X) in different molar ratios, where data points represent means ± SD (n = 3). Fig. 3a represents the assays set up with various combinations oftwo enzymes and 3b represents the assays set up with the three enzymes.

4476 N. Beukes, B.I. Pletschke / Bioresource Technology 101 (2010) 4472–4478

The degradation of native (un-treated) bagasse using differentcombinations of two enzymes (Fig. 2a) indicated that the greatestquantity of sugar (91.834 lmol/min) was produced with a molarratio of 37.5% ArfA and 62.5% ManA. This enzyme combination alsoproduced a degree of synergy of 1.87 (Fig. 2a). The greatest degreeof synergy (2.35) was achieved using the enzyme combinationwith the molar ratio 12.5% ArfA and 87.5% XynA, which was ex-pected, since these two enzymes are well known to associate andact synergistically in the degradation of xylan (Kosugi et al.,2002a; Koukiekolo et al., 2005). This enzyme combination pro-duced 71.159 lmol sugar/min, which is approximately 29% less su-gar than the 91.834 lmol/min produced by 37.5% ArfA and 62.5%ManA above. However, this combination increased the degree ofsynergy by approximately 25%. In Fig. 2a, the degree of synergyvalues for the first and last ratios studied (12.5% ManA and 87.5%XynA; and 12.5% ArfA and 87.5% XynA) were large, while the total

amounts of reducing sugar release were relatively low. A high de-gree of synergy is usually associated with a large release of reduc-ing sugar, but this may not always necessarily be the case, as thedegree of synergy is merely a ratio of the sum of the actual enzymeactivities to the theoretical sum of enzyme activities. In this case, alarge degree of synergy was found with little concomitant reducingsugar release. For optimal hydrolysis of SCB (or any other sub-strate) a large degree of synergy coupled to a large release of reduc-ing sugar is ideal.

There was no observed benefit in using combinations contain-ing all three enzymes (Fig. 2b). However; for pre-treated bagasse,both activity and degree of synergy was significantly increasedby using combinations containing all three enzymes (Fig. 3a andb). The pre-treatment drastically increased the total amount ofsugar produced from the enzymatic hydrolysis of the sugarcanebagasse (Fig. 3a and b); however, a maximal activity of

Fig. 4. The effect of time on enzyme synergy. The enzyme reactions were set up using the optimal enzyme combination with the (a) un-treated bagasse and the (b) lime pre-treated bagasse. The effect of time on the degradation of the different bagasse samples using the individual enzymes that correspond to the molar percentages that producedthe optimal enzyme activity. The degrees of synergy were obtained by dividing the sum of the actual activities of the enzymes obtained with the enzyme assays with thebagasse samples, by the theoretical sum of the enzyme activities obtained with the two substrates. The data points represent means ± SD (n = 3).

N. Beukes, B.I. Pletschke / Bioresource Technology 101 (2010) 4472–4478 4477

593.65 lmol sugar produced per min was obtained with the molarratio of 37.5% ArfA, 25% ManA and 37.5% XynA, which is approxi-mately 6.5-fold greater than the optimal activity obtained withthe un-treated SCB (Fig. 3b). This molar ratio also produced a de-gree of synergy of 2.14 which is approximately 9% less than the de-gree of synergy (2.35) obtained between 12.5% ArfA and 87.5%XynA with the un-treated bagasse. The molar ratio of 37.5% ArfA,25% ManA and 37.5% XynA achieved the greatest degree of synergywith the pre-treated bagasse of 2.17 and produced approximately517.495 lmol sugar per min (Fig. 3b).

Once the optimal molar enzyme ratios were determined forboth un-treated and pre-treated bagasse, the effect of time onthe synergistic relationship was determined. As expected, the totalamount of reducing sugar produced (after 120 h) with the un-trea-ted bagasse was substantially less than the pre-treated bagasse(Fig. 4). The theoretical sum of the individual recombinant enzymeactivities produced 2053 lmol of sugar, with the degradation ofthe un-treated bagasse; however, the enzymatic degradation ofthe un-treated bagasse produced a total sugar of 3535 lmol(Fig. 4a). The incubation time for the assay did not drastically affectthe degree of synergy observed at the various intervals, as the de-gree of synergy varied between 1.20 and 1.72. Thus the degree ofsynergy obtained these assays after 120 h was 1.72, which is sim-ilar to what was previously obtained (Fig. 2a). With regard to thelime pre-treated bagasse, time had a definite effect on the degrees

of synergy obtained. Over the 5 days the degree of synergy in-creased from 1.05 to 1.92 (approximately double); however the de-gree of synergy of 1.92 is slightly lower that the degree of synergythat was previously obtained (Fig. 2b).

4. Conclusions

A molar ratio of 37.5% ArfA, 25% ManA and 37.5% XynA and limepre-treatment resulted in the optimal hydrolysis of lime pre-trea-ted SCB. Results from a previous study (Beukes et al., 2008) indi-cated that a molar ratio of 75% XynA to 25% EngE produced thelargest degree of synergy (4.65) with un-treated SCB. It thereforeappears that lime treatment, together with a enzyme cocktail ofXynA with EngE, ManA and ArfA, will allow for the optimal (i.e.synergistic) hydrolysis of SCB. This information will also allowfor the future improvement of existing hemi/cellulase producingstrains such as C. cellulovorans or Trichoderma reesei.

Acknowledgements

We would like to thank Prof. Roy Doi (University of Davis Cali-fornia, USA) for kindly supplying the three enzyme constructs usedin this study. The SCB was kindly donated by Mr. I. Ramluken fromUshukela Milling (Pty) Ltd., Durban, South Africa. We would also

4478 N. Beukes, B.I. Pletschke / Bioresource Technology 101 (2010) 4472–4478

like to thank Mr. S. Walford from the SMRI for assisting with thechemical characterisation of the SCB. This material is based uponwork supported by the National Research Foundation of South Afri-ca (NRF) and Rhodes University Joint Research Committee (JRC).Any opinion, findings and conclusions or recommendations ex-pressed in this material are those of the author(s) and thereforethe NRF does not accept any liability in regard thereto.

References

Aspinall, G.O., 1980. The Biochemistry of Plants, vol. 3. Academic Press Inc., NewYork.

Bergey, D.H., 1974. Bergey’s Manual of Determinative Bacteriology, eighth ed.Williams and Wilkins Publishers, Baltimore, Maryland, USA, pp. 1860–1937.

Beukes, N., Chan, H., Doi, R.H., Pletschke, B.I., 2008. Synergistic associations betweenClostridium cellulovorans enzymes XynA, ManA and EngE against sugarcanebagasse. Enzyme Microb. Technol. 42, 492–498.

Blum, D.L., Kataeva, I.A., Li, X.-L., Ljungdahl, L.G., 2000. Feruloyl esterase activity ofthe Clostridium thermocellum cellulosome can be attributed to previouslyunknown domains of XynY and XynZ. J. Bacteriol. 182, 1346–1351.

Bradford, M.M., 1976. A rapid and sensitive method for the quantitation ofmicrogram quantities of protein utilizing the principle of protein-dye binding.Anal. Biochem. 72, 248–254.

Chang, M.C.Y., 2007. Harnessing energy from plant biomass. Curr. Opin. Chem. Biol.11, 677–684.

Chang, V.S., Holtzapple, M.T., 2000. Fundamental factors affecting biomassenzymatic reactivity. Appl. Biochem. Biotechnol. 84, 5–37.

Ciruela, A., Gilbert, H.J., Ali, B.R.S., Hazlewood, G.P., 1998. Synergistic interaction ofthe cellulosome integrating protein (CipA) from Clostridium thermocellum with acellulosomal endoglucanase. FEBS Lett. 422, 221–224.

Din, N., Damude, H.G., Gilkes, N.R., Miller, R.C., Warren Jr., R.A., Kilburn, D.G., 1994.C1–Cx revisited: intramolecular synergism in a cellulase. PNAS 91, 11383–11387.

Frollini, E., Paiva, J.M.F., Trindade, W.G., Razera, I.A.T., Tita, S.P., 2004. Lignophenolicand phenolic resins as matrix in vegetal fibres reinforced composites. In:Wallenberger, F., Weston, N. (Eds.), Natural Fibres, Polymers and Composites –Recent Advances. Kluwer Academic Publishers, pp. 193–219.

Hames, B., Ruiz,Sluiter, A., Sluiter, J., Zempleton, D., 2005. Preparation of Samples forCompositional Analysis Laboratory Analytical Procedure (LAP). NREL TechnicalReport NREL/TP-510-42620.

Hyman, D, Sluiter, A., Crocker, D., Johnson, D., Sluiter, J., Black, S., Scarlata, C., 2007.Determination of Acid Soluble Lignin Concentration Curve by UV–Vis

Spectroscopy Laboratory Analytical Procedure (LAP). NREL Technical ReportNREL/TP-510-42620.

Kaar, W.E., Holtzapple, M.T., 2000. Using lime pretreatment to facilitate theenzymatic hydrolysis of corn stover. Biomass Bioenerg. 18, 189–199.

Kosugi, A., Murashima, Y., Doi, R.H., 2002a. Characterization of two noncellulosomalsubunits, ArfA and BgaA, from Clostridium cellulovorans that cooperate with thecellulosome in plant cell wall degradation. J. Bacteriol. 184, 6859–6865.

Kosugi, A., Murashima, Y., Doi, R.H., 2002b. Xylanase and acetyl esterase activities ofXynA, a key subunit of the Clostridium cellulovorans cellulosome for xylandegradation. Appl. Environ. Microb. 68, 6399–6402.

Koukiekolo, R., Cho, H.-Y., Kosugi, A., Inui, M., Yukawa, H., Doi, R.H., 2005.Degradation of corn fiber by Clostridium cellulovorans cellulases andhemicellulases and contribution of scaffolding protein CbpA. Appl. Environ.Microb. 71, 3504–3511.

Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of thehead of bacteriophage T4. Nature 277, 680–685.

Miller, G.L., 1959. Use of dinitrosalicylic acid reagent for the determination ofreducing sugars. Anal. Chem. 31, 426–428.

Monterio, S.N., Rodriguez, R.J.S., De Souza, M.V., D’Almeida, J.R.M., 1998. Sugarcanebagasse waste as reinforcement in low cost composites. Adv. Perform. Mater. 5,183–191.

Mosier, N., Wyman, C., Dale, B., Elander, R., Lee, Y.Y., Holtzapple, M., Ladisch, M.,2005. Features of promising technologies for pretreatment of lignocellulosicbiomass. Bioresour. Technol. 96, 673–686.

Murashima, K., Kosugi, A., Doi, R.H., 2002. Synergistic effects on crystalline cellulosedegradation between cellulosomal cellulases from Clostridium cellulovorans. J.Bacteriol. 184, 5088–5094.

Murashima, K., Kosugi, A., Doi, R.H., 2003. Synergistic effects of cellulosomalxylanase and cellulases from Clostridium cellulovorans on plant cell walldegradation. J. Bacteriol. 185, 1518–1524.

Paiva, J.M.F., Frollini, E., 2002. Sugarcane bagasse reinforced phenolic andlignophenolic composites. J. Appl. Polym. Sci. 83, 880–888.

Pandey, A., Soccoi, C.R., Nigam, P., Soccoi, V.T., 2000. Biotechnological potential ofagro-industrial residues I: sugarcane bagasse. Bioresour. Technol. 74, 69–80.

Simkovic, I., Mlynar, J., Alfoldi, J., Micko, M.M., 1990. New aspects in cationization oflignocelluloses materials. XI. Modification of bagasse with quaternaryammonium groups. Holzforschung 44, 113.

Tamaru, Y., Karita, S., Ibrahim, A., Chan, H., Doi, R.H., 2000. A large gene cluster forthe Clostridium cellulovorans cellulosome. J. Bacteriol. 182, 5906–5910.

Teeri, T., 1997. Crystalline cellulose degradation: new insight into the function ofcellobiohydrolases. Trends Biotechnol. 15, 160–167.

Thomson, J.A., 1993. Molecular biology of xylan degradation. FEMS Microbiol. Rev.104, 65–82.

Zarate, C.N., Aranguren, M.I., Reboredo, M.M., 2002. Resol vegetable fibrescomposites. J. Appl. Polym. Sci. 77, 1832–1840.