Bioresource Technology 145 (2013) 302–306

Contents lists available at SciVerse ScienceDirect

Bioresource Technology

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

Exploring novel ultrafine Eri silk bioscaffold for enzyme stabilisationin cellobiose hydrolysis

0960-8524/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.biortech.2013.01.065

⇑ Corresponding author. Address: Bioprocessing Lab (CCB), Deakin University,Victoria, Australia. Tel.: +61 3 5227 2325; fax: +61 3 5227 2170.

E-mail address: munish.puri@deakin.edu.au (M. Puri).

Madan L. Verma a, Rangam Rajkhowa b, Xungai Wang b, Colin J. Barrow a, Munish Puri a,⇑a Centre for Chemistry and Biotechnology (CCB), Geelong Technology Precinct, Deakin University, Geelong, Victoria 3217, Australiab Australian Future Fibres Research and Innovation Centre (AFFRIC), Geelong Technology Precinct, Deakin University, Geelong, Victoria 3217, Australia

h i g h l i g h t s

" Ultrafine Eri silk particles (size 5 lm) were prepared in short milling time without pre-treatment." Environment friendly Eri silk scaffold is used for immobilisation of b-glucosidase." Eri-silk bioscaffold protects the enzyme by increasing its rigidity and stability." Immobilised enzyme retained more than 50% of initial activity for up to eight cycles.

a r t i c l e i n f o

Article history:Available online 1 February 2013

Keywords:Lignocelluloseb-GucosidaseNanobiotechnologyBiofuelImmobilisation

a b s t r a c t

The suitability of optimised ultrafine Eri silk microparticles as novel enzyme supports was studied forpotential application in biofuel production. b-glucosidase (BGL) from Aspergillus niger was immobilisedon Eri silk fibrion particles via an adsorption method resulting in a 62% immobilisation yield. Solubleand immobilised enzymes exhibited pH-optima at pH 4.0 and 5.0, respectively with optimum activityat 60 �C. The Michaelis constant (KM) was 0.16 and 0.27 mM for soluble and immobilised BGL respec-tively. The immobilisation support has a protective effect on the enzyme by increasing rigidity; this isreflected by an increase in stability under thermal denaturation at 70 �C. Immobilised enzyme retainedmore than 50% of initial activity for up to eight cycles. Maximum cellobiose hydrolysis by immobilisedBGL was achieved at 20 h. Crystalline ultrafine Eri silk particles were found to be a promising viable, envi-ronmentally sound and stable matrix for binding BGL for cellobiose hydrolysis.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Exploitation of lignocellulosic (LC) biomass for the productionof biofuels may provide a promising alternative to diminishing glo-bal fossil energy reserves. Enzymatic hydrolysis of cellulosic poly-mers can produce glucose, which readily ferments to ethanol (Puriet al., 2012). This enzyme assisted hydrolysis of biomass is there-fore of interest due to its potential for producing sugars for biofuelproduction. The complete hydrolysis of cellulosic polymers re-quires a combination of different enzyme types such as endoglu-canase, exocellobiohydrolase and b-glucosidase (Xue andWoodley, 2012; Sorensen et al., 2011). Endoglucanase randomlyhydrolyses the b-1,4 bonds in the cellulose backbone, and, in mostobserved cases, exocellobiohydrolases release a cellobiose unitexhibiting recurrent reactions. The cellobiose is finally convertedto glucose by b-glucosidase (BGL). Commercial cellulase prepara-

tions are typically deficient in BGL and an accumulation of cellobi-ose leads to product inhibition. BGL is frequently used tosupplement cellulase preparations for hydrolysis of LC substratesin order to accelerate the conversion of cellobiose to glucose andrelieve product inhibition of cellobiose in the cellulosic ethanolproduction (Lee et al., 2010). Thus, BGL activity plays an essentialrole in the efficient and complete hydrolysis of lignocellulosicbiomass.

Enzyme immobilisation is the most promising tool in the fieldof enzyme biotechnology for achieving efficient utilisation of en-zymes (Mateo et al., 2007). Such an approach could minimise thehigh costs typically associated with the use of enzymes for variousapplications. Repeated usability and improved enzyme stability areimportant features of any immobilisation technology (Verma et al.,2012). A variety of supports can be used for enzyme immobilisa-tion, including inorganic and organic substances (Mateo et al.,2007). Among existing organic supports, silk fibroin macromole-cules present a relatively new biomaterial for enzyme immobilisa-tion (Lu et al., 2009). The use of sub-micron silk particles is likely tobe advantageous due to the high inherent surface area. Amongst

M.L. Verma et al. / Bioresource Technology 145 (2013) 302–306 303

immobilisation methods, adsorption, although susceptible to pHvariation, is still the most suitable technique due to its simplicity,and cost-effectiveness as well as the fact that it induces fewermodifications in the active conformation of the enzyme (Mateoet al., 2007).

In the present study, ultrafine Eri silk particles (ca. 5 lm) wereprepared in our laboratory without using chemicals. The millingdifficulty of viscoelastic silk fibres was overcome through a novelwet milling and spray drying process. The ultrafine particles wereproduced in a short period and without degradation. The synthesisof a stable BGL-particle composite, its characterisation, stabilisa-tion and application in cellobiose hydrolysis were examined. Tothe best of our knowledge, this represents the first documented in-stance that ultrafine microparticles powders of Eri silk were usedas a matrix to immobilise BGL from Aspergillus niger.

2. Methods

2.1. Materials

Aspergillus niger b-glucosidase (60 U/g) was procured from Sig-ma (Australia). Cellobiose, bovine serum albumin, glucose oxidase(GOD–POD) assay kit, p-nitrophenol (pNP), para-nitrophenol-b-glucopyranoside (pNPG) were procured from Sigma–Aldrich Aus-tralia. Protein assay dye reagent concentrate from Bio-Rad and sul-phuric acid were procured from Merck, Australia. Eri silk ultrafineparticle was prepared in our laboratory (Rajkhowa et al., 2009).

2.2. Conjugation BGL onto Eri-silk particle

BGL was immobilised on the silk particle using an adsorptionmethod. The particles were washed 3 times with deionized waterfollowed by phosphate buffer (20 mM, pH 7.0). The particles werethen recovered by centrifugation (5000g for 5 min at 20 �C). Thepurified enzyme preparation (100 mg enzyme per gram of support)was mixed with silk fibroin particles in a phosphate buffer. Enzymeimmobilisation was carried in a shaker (50 rpm for 20 h at 37 �C).The immobilised support was removed by centrifugation, washedat least 3 times with 30 mL of deionized water to remove freeBGL from the matrix, and subsequently washed with enzyme assaybuffer. The supernatant was used for protein analysis by the Brad-ford method (Bradford, 1976). The immobilisation efficiency andimmobilisation yield were calculated based on the method opti-mised in our earlier studies (Verma et al., 2012). The washed car-rier was used directly for the determination of activity andstability.

2.3. Characterisation of matrix and enzyme bound matrix

A TriStar 3000 gas adsorption analyser (Micromeritics, USA)was used to measure the surface area of silk particles by the Bru-nauer–Emmett–Teller (BET) technique. Eri silk particles were char-acterised using scanning electron microscopy (SEM, Gemini Supra55 VP) at 5 kV. Fourier transform infrared spectroscopy (FTIR)spectra of desiccated silk particles were recorded with an attenu-ated total reflectance (ATR) mode using a spectrophotometer (Ver-tex 70, Bruker Biosciences Pty Ltd., Australia). To measure differentconformations, average spectrum in the amide I mode was decon-voluted and curve fitted (1595–1705 cm�1) using the OPUS 5.5software. Wide angle X-ray scattering (WAXS) was performed ona diffractometer (PANalytical-Xpert PRO) with Cu Ka radiation(k = 0.154 nm) operated at 40 kV and 30 mA and analysed usingthe Xpert High Score Plus software.

2.4. Enzyme assay

The enzyme activity for soluble and immobilised BGL was mea-sured using a modification of the method of Takahashi et al. (2011).pNPG, (0.09 mL), and enzyme solution (2 U) were incubated at60 �C for 10 min before being stopped by addition of 0.5 M Na2CO3

(0.5 mL). The liberated 4-nitrophenol was measured at 405 nm andits concentration calculated from a plot constructed for standard p-nitrophenol. One unit of enzyme activity is defined as 1 lmol of p-nitrophenol liberated per minute at pH 4.0, 60 �C. Activity ofimmobilised BGL (10 mg) was measured with pNPG (equalamounts corresponding to soluble enzyme) at pH 5.0, 60 �C. All en-zyme assays were performed in triplicate with reporting of meanvalues ± SD.

2.5. Effects of pH on the catalytic activity of the soluble andimmobilised BGL

The optimum pH for immobilised and soluble BGL activity wasdetermined using standard assay conditions in sodium acetate buf-fer at varying pH (pH 3.0–8.0, 10 mM). Relative activities at varyingpH values were quantified relative to the control activities, and ex-pressed as fractions thereof. These control activities, those repre-senting 100% relative activity, were taken to be at pH 4.0 and 5.0for soluble and immobilised BGL, respectively.

2.6. Effects of temperature on the catalytic activity of the soluble andimmobilised BGL

The effect of temperature on immobilised and soluble BGLactivity was determined by measuring relative enzyme activitywhile varying the temperature in the 30–80 �C range. Other reac-tion conditions adhered to the standard assay protocol. The activ-ities at 60 �C for soluble and immobilised BGL were taken ascontrol for the calculation of relative activity.

2.7. Determination of kinetic parameters

The effect of substrate concentration in the immobilised andsoluble BGL activity was tested in different concentrations of pNPG(2–10 mM). The assays were performed under optimal pH andtemperature. The KM and Vmax values were determined by aLineweaver–Burk plot.

2.8. Kinetics of thermal stabilities of soluble and immobilised BGL

Thermal stability of the soluble and immobilised BGL at a se-lected temperature (70 �C) was quantified in terms of the loss inenzyme activity when incubated in the absence of a substrate.The activity of soluble and immobilised BGL was observed after acertain period (10 h). The relative activity of the soluble and immo-bilised BGL without incubation was defined as control and arbi-trarily attributed 100% relative activity for each of theirrespective reactions.

2.9. Reusability study of immobilised BGL

Immobilised preparation was evaluated at 60 �C by carrying outthe hydrolysis of pNPG under standard assay conditions to explorereusability. After each cycle, the immobilised enzyme was re-moved by centrifugation at 2000g for 5 min. The immobilisedBGL was collected and washed simultaneously with deionizedwater and assay buffer. In running the second cycle, the immobi-lised enzyme was redissolved in fresh buffer and added to freshpNPG. The activity of the immobilised enzyme after the first cyclewas defined as the control and attributed a relative activity of

Table 1Percentage of different conformations in the amide I region (average ± standarddeviations).

Feature Side chain b-sheet Random coil/a-helix

b-turn

Soluble BGL 5 ± 0.05 20.62 ± 0.01 64.54 ± 0.05 9.84 ± 0.01Matrix 1.04 ± 0.78 59.87 ± 4.02 21.69 ± 1.50 21.51 ± 1.58Immobilized

BGL5.48 ± 0.36 55.82 ± 0.65 16.80 ± 0.36 21.89 ± 0.08

304 M.L. Verma et al. / Bioresource Technology 145 (2013) 302–306

100%. Each cycle in this study is the complete hydrolysis of thesubstrate present in a reaction mixture.

2.10. Kinetics of cellobiose hydrolysis

Hydrolysis of cellobiose (1%, w/v) was studied using soluble andimmobilised BGL. Activity of immobilised BGL (10 mg) was mea-sured with cellobiose at 60 �C. The aliquots were removed at inter-vals to measure the degree of cellobiose hydrolysis. Maximumconversion was studied at regular 4 h intervals (up to 20 h) at60 �C in a shaking water bath. Glucose (GO) assay kit was usedfor measuring glucose concentration during cellobiose hydrolysis.From the glucose concentration, residual cellobiose levels duringhydrolysis were calculated based on the stoichiometry of thereaction.

3. Results and discussion

Enzyme-immobilised silk fibroin membranes have been usedsuccessfully for the determinations of glucose, hydrogen peroxideand uric acid. However, those studies were conducted using Bom-byx mori silk fibroin, whereas the current study utilised Eri silkwhich has a different amino acid composition. More importantly,mechanically produced silk particles were utilised which, unlikeregenerated silk fibroin, retain the crystalline structure of parent fi-bres. Structural difference may provide new challenges and oppor-tunities in using the milled ultrafine particles.

3.1. Preparation and characterisation of silk particles

Compared to particles fabricated by self-assembly from a silksolution, milling produces a relatively broad size distribution. Thisparticle size distribution is narrowed down as the milling pro-gresses but after a certain time, a stable particle size distributionand size reduction reaches its limit. At that stage milling wasstopped and particles after spray drying were used for enzymeimmobilisation. The BET surface area of the particles was19.89 m2/g. Theoretical calculations show that such a BET surfacearea comes from a solid particle of ca. 300 nm size. This indicatesthat the silk particles with a median size of 5 lm used in this studyhave porous architecture. The SEM images of the particles werehighly porous and appear to be aggregates of finer particles, be-sides the diversity of amino acids and availability of binding citesis evident from its composition. The high surface area togetherwith the presence of functional amino acids may assists in goodadsorption of enzymes and fast kinetics.

The Saturniidae family silk fibres show characteristics b-sheetpeaks at 2h positions of 16.5, 20.4, and 24.3, corresponding to crys-talline plane spacings of 5.30 Å, 4.32 Å, and 3.67 Å, respectively.The crystalline structure is important for providing thermal stabil-ity and insolubility of the particles, both of which are importantrequirements for an enzyme support matrix. Silk shows a broadamorphous halo with a centre around 2h = 20.30 due to the shortrange order related to chain–chain correlation constrained byinterchain hydrogen bonding resulting in the broad nature of thepeaks (Martel et al., 2007). The halo of the WAXS from the enzymefalls within the same region and thus differences in the pattern dueto enzyme binding are not reflected in the XRD results.

The FTIR absorption positions, particularly in the amide I region,are a good indication of the backbone conformations and cantherefore be used to determine molecular conformations of a pro-tein. The amide I peak around 1628 cm�1 and amide II peak around1523 cm�1 indicates the presence of the b-sheet structure in Erisilk particles. The BGL has an amide I peak around 1650 cm�1

and an amide II peak around 1640 cm�1, which are typical

positions for helical conformations in a protein. Quantitative anal-ysis of the different conformations was done by curve fitting theFSD spectra of the amide I region (Table 1). The fraction of b-sheetin the Eri silk powder matrix is nearly 3 times that of the BGL en-zyme. Upon immobilisation, the b-sheet fraction marginallydropped but the matrix structure remained essentially crystalline.It is likely that the porous nature of the silk particles resulted inmigration of enzymes into the silk pores, resulting in only smallchanges in structure of the outer surface of the particles afterimmobilisation.

3.2. Enzyme immobilisation and characterisation studies

Following immobilisation of enzymes onto the porous surfacesof silk particles, the binding efficiency was investigated using theBradford method by calculating the protein content in the superna-tant, with subsequent washing until no protein was leached.Immobilisation yield was found to be 62% with immobilisationefficiency of 64%. Immobilisation of BGL from Pyrococcus furiosusin gelatine gel by cross-linking with transglutaminase allowedimmobilisation yields in the range 25–39%. However, when BGLfrom almonds was immobilised, the observed yield was just 5%(Nagatomo et al., 2005).

3.3. Effects of pH on catalytic activity of the soluble and immobilisedBGL

The optimum pH values for soluble and immobilised enzymeswere found to be 4.0 and 5.0, respectively (Fig. S1a). The immobi-lised enzyme derivative was more stable at higher pH values. Pre-vious research found that the optimum profile of immobilised BGLfrom A. niger was shifted by approximately 0.7 pH units when com-pared to the optimum pH of its soluble form (Jung et al., 2012).Zhou and Zhang (2011) also reported a change in pH optimum ofa 1.5 unit towards the basic side upon immobilising BGL on silk fi-broin nanoparticles. This change in pH was possibly due to interac-tions between the charged groups of the enzyme molecule andstationary charges on the carrier (Gomes et al., 2010).

3.4. Effects of temperature on catalytic activity of the soluble andimmobilised BGL

The optimum temperature for soluble and immobilised enzymewas found to be 60 �C (Fig. S1b). The immobilised enzyme retained70% of its activity in the temperature range 60–70 �C, while thefree enzyme retained 50% of its activity at the same temperature.In a previous study it was reported that the optimum temperatureof the immobilised and soluble BGL was 60 �C (Zhou and Zhang,2011). Temperature optima (at 65 �C) were the same for solubleand immobilised cellobiase on pore silica particles (Calsavaraet al., 2001). In addition, the immobilised enzyme may more read-ily contact the substrate with an increase in temperature (Mateoet al., 2007).

M.L. Verma et al. / Bioresource Technology 145 (2013) 302–306 305

3.5. Kinetic parameters

The values of KM and Vmax for soluble and immobilised BGLwere estimated from the double reciprocal plots of the initial ratesof pNPG hydrolysis. The KM value of the soluble enzyme (0.16 mM)was almost half that of the immobilised enzyme (0.27 mM), whileVmax of the immobilised BGL (18.65 U/mg) was higher than that ofsoluble BGL (8.15 U/mg). The KM of the soluble BGL was 1.1 mM.The apparent KM value for the immobilised enzyme was about10-fold higher than that for the soluble enzyme (Tu et al., 2006).The KM value for the soluble enzyme A. niger BGL was reportedto be 0.57 mM using pNPG (Chauve et al., 2010). The KM valuefor soluble BGL from Agaricus arvensis was 2.5 mM, and 3.8 mMfor immobilised BGL (Singh et al., 2011). These changes in thekinetic parameters suggest that slightly decreased access of the

Fig. 1. Thermostability studies (a) and cellobiose hydrolysis (b) of soluble and immo

substrate to the active site of enzyme in the immobilised BGL sug-gesting diffusion limitations.

3.6. Thermal stability of soluble and immobilised BGL

The improved thermal stability of the BGL immobilised on silkparticles as compared to the soluble enzyme is presented inFig. 1a. The soluble enzyme lost its activity after 2 h of incubationat 70 �C, whereas the silk immobilised BGL retained 55% of initialactivity after 6 h incubation at the same temperature. The solubleBGL from A. arvensis completely lost its initial activity after55 min when incubated at 50 �C (Singh et al., 2011). The commer-cial enzyme preparation (Novozym 188 from A. niger) showed noactivity after 2 h of incubation at 67 �C (Sorensen et al., 2011).The increased stability observed for the immobilised enzyme

bilised BGL. Bars indicate the standard deviation from triplicate determinations.

306 M.L. Verma et al. / Bioresource Technology 145 (2013) 302–306

may be attributed to a reduction in the protein structure mobility,due to anchorage to the support promoted by the physical interac-tion and subsequent translation of the rigidity at each anchoragepoint to the entire enzyme structure, thus shielding it from damag-ing effects of the environment (Taqieddin and Amiji, 2004). Goodenzyme stabilisation of three enzymes (glucose oxidase, lipase,and horseradish peroxidase) when entrapped in B. mori silk filmshas also been reported (Lu et al., 2009). The present study confirmsthat enhanced thermal stability of enzymes can also be achievedby immobilising on milled silk particles, which have competitiveadvantages over analogous products from silk solutions in termsof a preferable manufacturing environment (i.e. chemical-free pro-cess), the feasibility of translating the process to an industrial scale,and enhanced stability (high crystallinity).

3.7. Reusability study of the immobilised BGL

Reusability studies have shown that the immobilised enzymewas stable for up to eight cycles at 60 �C, and retained approxi-mately 50% of initial activity at the end of the last reaction cycle.It is anticipated that the recycled uses of stable support boundBGL can improve the economics of cellulosic ethanol productionso long as economically viable materials are employed as host forenzyme immobilisation (Lee et al., 2010). It was shown earlier thatimmobilisation of BGL on Eupergit C for LC hydrolysis resulted insix cycles of operational stability (Tu et al., 2006). However, siliconoxide nanoparticle immobilised BGL retained significant activityfor up to 15 cycles (Singh et al., 2011). Enhancement of immobi-lised enzyme activity by pre-treatment of BGL with cellobioseand glucose was reported. BGL from A. niger covalently immobi-lised onto silica gel was used 20 times, maintaining higher than80% of initial activity (Jung et al., 2012).

3.8. Hydrolysis of cellobiose by soluble and immobilised BGL

Soluble BGL and immobilised BGL were used for cellobiosehydrolysis at 60 �C (Fig. 1b). More than 50% of cellobiose hydrolysiswas achieved by immobilised BGL within 4 h, reaching a maximumat 20 h, with hydrolysis rates thereafter remaining constant. Whilethe rate of cellobiose hydrolysis achieved by soluble BGL was rela-tively low at a short reaction incubation of 4 h, maximum hydroly-sis of cellobiose was achieved in a long incubation of 20 h.Cellobiose hydrolysis rate was higher in immobilised BGL. Thismay be due to good thermal stability in silk fibrion particle boundBGL than in the soluble enzyme (Lu et al., 2009). The conversion ofcellulose to glucose in the presence of Eupergit C immobilised BGL(82%) was similar to that seen when free BGL was used (88% con-version) (Tu et al., 2006). This immobilised enzyme preparationwill therefore be a suitable process for obtaining fermentable sug-ars from LC biomass.

4. Conclusion

BGL was immobilised by adsorption to silk fibrion particles.Structural stability of the enzyme immobilised particles was con-firmed using XRD and FTIR. The kinetics of the soluble and immo-bilised enzyme suggested that the enzyme undergoes aconformational change during immobilisation, resulting in achange in pH and kinetic constants. Porous Eri silk fibroin particlesprovide an ideal microenvironment for the immobilised BGL thatexhibited good thermostability and reusability. This study also

demonstrates that BGL immobilised on silk fibrion particles canbe used for the effective hydrolysis of cellobiose.

Acknowledgements

The authors are thankful to the ARC Discovery Grant DP1094979 for supporting this research work. We also thank ElectronMicroscopic facility (IFM) and an award of Alfred Deakin Post-Doc-toral fellowship to one of the authors (MLV).

Appendix A. Supplementary material

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.biortech.2013.01.065.

References

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.

Calsavara, L.P., De Moraes, F.F., Zanin, G.M., 2001. Comparison of catalytic propertiesof free and immobilised cellobiase novozyme 188. Appl. Biochem. Biotechnol.91–93, 615–626.

Chauve, M., Mathis, H., Huc, D., Casanave, D., Monot, F., Ferreira, N., 2010.Comparative kinetic analysis of two fungal beta-glucosidases. Biotechnol.Biofuels 3, 3.

Gomes, J.M., Romero, M.D., Fernandez, T.M., Garcia, S., 2010. Immobilisation andenzymatic activity of b-glucosidase on mesoporous SBA-15 silica. J. PorousMater. 17, 657–662.

Jung, Y.R., Shin, H.Y., Song, Y.S., Kim, S.B., Kim, S.W., 2012. Enhancement ofimmobilised enzyme activity by pretreatment of b glucosidase with cellobioseand glucose. J. Ind. Eng. Chem. 18, 702–706.

Lee, S.M., Jin, L.H., Kim, J.H., Han, S.O., Na, H.B., Hyeon, T., Koo, Y.M., Kim, J., Lee, J.H.,2010. Beta-glucosidase coating on polymer nanofibers for improved cellulosicethanol production. Bioprocess. Biosyst. Eng. 33, 141–147.

Lu, S., Wang, X., Lu, Q., Hu, X., Uppal, N., Omenetto, F.G., Kaplan, D.L., 2009.Stabilization of enzymes in silk films. Biomacromolecules 10, 1032–1042.

Martel, A., Burghammer, M., Davies, R.J., Riekel, C., 2007. Thermal behavior ofBombyx mori silk: evolution of crystalline parameters, molecular structure, andmechanical properties. Biomacromolecules 8, 3548–3556.

Mateo, C., Palomo, J.M., Lorente, G.F., Guisan, J.M., Lafuente, R.F., 2007. Improvementof enzyme activity, stability and selectivity via immobilisation techniques.Enzyme Microb. Technol. 40, 1451–1463.

Nagatomo, H., Matsushita, Y., Sugamoto, K., Matsui, T., 2005. Preparation andproperties of gelatin-immobilised b-glucosidase from Pyrococcus furiosus.Biosci. Biotechnol. Biochem. 69, 128–136.

Puri, M., Abraham, R.E., Barrow, C.J., 2012. Biofuel production: prospects, challengesand feedstock in Australia. Ren. Sust. Energy Rev. 16, 6022–6031.

Rajkhowa, R., Wang, L., Kanwar, J.R., Wang, X., 2009. Fabrication of ultrafine powderfrom Eri silk through attritor and jet milling. Powder Technol. 191, 155–163.

Singh, R.K., Zhang, Y.W., Nguyen, N.P.T., Jeya, M., Lee, J.K., 2011. Covalentimmobilisation of b-1,4-glucosidase from Agaricus arvensis ontofunctionalized silicon oxide nanoparticles. Appl. Microbiol. Biotechnol. 89,337–344.

Sorensen, A., Lübeck, P.S., Lübeck, M., Teller, P.J., Ahring, B.K., 2011. b-glucosidasesfrom a new Aspergillus species can substitute commercial b-glucosidases forsaccharification of lignocellulosic biomass. Can. J. Microbiol. 57, 638–650.

Takahashi, M., Konishi, Y., Takeda, T., 2011. Biochemical characterization ofMagnaporthe oryzae b-glucosidases for efficient b-glucan hydrolysis. Appl.Microbiol. Biotechnol. 91, 1073–1082.

Taqieddin, E., Amiji, M., 2004. Enzyme immobilisation in novel alginate–chitosancore-shell microcapsules. Biomaterials 25, 1937–1945.

Tu, M., Zhang, X., Kurabi, A., Gilkes, N., Mabee, W., Saddler, J., 2006. Immobilisationof b-glucosidase on Eupergit C for lignocellulose hydrolysis. Biotechnol. Lett. 28,151–156.

Verma, M.L., Barrow, C.J., Kennedy, J.F., Puri, M., 2012. Immobilisation of b-D-galactosidase from Kluyveromyces lactis on functionalized silicon dioxidenanoparticles: characterization and lactose hydrolysis. Int. J. Biol. Macromol. 50,432–437.

Xue, R., Woodley, J.M., 2012. Process technology for multi-enzymatic reactionsystems. Bioresour. Technol. 115, 183–195.

Zhou, Z., Zhang, Y., 2011. Biosynthesis of b-glucosidase-silk fibrion nanoparticlesconjugates and enzymatic characteristics. Adv. Mater. Res. 175–176, 186–191.