NC State
BioResources
Geng, A., Wu, J., Xie, R., Li, X., Chang, F., and Sun, J. (2015). "Construction of a bacterial cellulase cocktail for saccharification of regenerated cellulose and pretreated corn stover," BioRes. 10(4), 7681-7692

Abstract

To apply bacterial cellulases for efficient saccharification of biomass, three Clostridium thermocellum cellulases and a Thermoanaerobacter brockii β-1,4-glucosidase were synthesized in Escherichia coli, and the proportions among them were optimized. When the activities of CelD, CBHA, CBH48Y, and CglT were set at 554, 0.91, 0.91, and 856 mU per assay, respectively, the percent conversion of regenerated cellulose (0.92 g/L) reached 80.9% within 24 h at 60 °C without shaking. Meanwhile, the percent conversion of pretreated corn stover (0.62 g/L) reached 70.1%. Gradually raising the loads of regenerated cellulose from 0.92 to 4.58 g/L resulted in a linear increase in glucose production from 870 to 3208 μg (R2=0.997), as well as a decrease in the percent conversion from 80.9% to 59.6%. These findings suggested that the cellulase cocktail is efficient in saccharification of regenerated cellulose, as well as pretreated corn stover, and has potential applications in the biofuels industry.


Download PDF

Full Article

Construction of a Bacterial Cellulase Cocktail for Saccharification of Regenerated Cellulose and Pretreated Corn Stover

Alei Geng,* Jian Wu, Rongrong Xie, Xia Li, Fuxiang Chang, and Jianzhong Sun

To apply bacterial cellulases for efficient saccharification of biomass, three Clostridium thermocellum cellulases and a Thermoanaerobacter brockii β-1,4-glucosidase were synthesized in Escherichia coli, and the proportions among them were optimized. When the activities of CelD, CBHA, CBH48Y, and CglT were set at 554, 0.91, 0.91, and 856 mU per assay, respectively, the percent conversion of regenerated cellulose (0.92 g/L) reached 80.9% within 24 h at 60 °C without shaking. Meanwhile, the percent conversion of pretreated corn stover (0.62 g/L) reached 70.1%. Gradually raising the loads of regenerated cellulose from 0.92 to 4.58 g/L resulted in a linear increase in glucose production from 870 to 3208 μg (R2=0.997), as well as a decrease in the percent conversion from 80.9% to 59.6%. These findings suggested that the cellulase cocktail is efficient in saccharification of regenerated cellulose, as well as pretreated corn stover, and has potential applications in the biofuels industry.

Keywords: Cellulases; Saccharification; Clostridium thermocellum; Thermoanaerobacter brockii; Biomass

Contact information: Biofuels Institute, School of the Environment and Safety Engineering, Jiangsu University, Zhenjiang 212013, Jiangsu, China; *Corresponding author: galxj@mail.ujs.edu.cn

INTRODUCTION

Currently, cellulosic biofuel technologies are confronted with two primary problems: the high cost of cellulases and the lack of an efficient pretreatment technology for biomass (Berlin et al. 2007; Taherzadeh and Karimi 2007; Agbor et al. 2011). To cut down on the cost of cellulases, efficient cellulases have been recruited to improve the performance of cellulase systems. Cumulative studies have suggested that cellulases from both filamentous fungi and bacteria hold great potential for future industry applications, although the latter is currently less widely applied in the cellulosic biofuel industry than the former (Maki et al. 2009). Most importantly, many bacteria thrive in extreme environmental niches and produce stress-resistant cellulases, which is of great value for industrial applications.

Meyer et al. (2009) found that a minimal enzyme cocktail had a better performance than native Trichoderma reesei enzymes for hydrolysis of pretreated barley straw samples and pointed out the necessity of constructing a minimal cellulase cocktail for efficient biomass processing that could also be applicable to bacteria. Clostridium thermocellum is one of the most famous anaerobic cellulolytic bacteria because of its efficient cellulase system, which consists of both cellulosomal and non-cellulosomal cellulases (Gold and Martin 2007). Many C. thermocellum cellulases have been characterized, and most of them have functioned well, with good thermostability. Thus, it is necessary to construct a basic cellulase cocktail, mainly from C. thermocellum, for the gradual establishment of an efficient cellulase system for biomass saccharification. A core cellulase system consists of the following: (1) an endo-1,4-β-D-glucanase, which scissors the long cellulose chains; (2) two exo-1,4-β-D-glucanases, one of which acts on the reducing side of cellulose chains, and the other on the non-reducing side, to produce cellobiose; and (3) a β-1,4-glucosidase, which acts on cellobiose to produce glucose. Fortunately, all of these enzymes could be found in C. thermocellum.

The final step of saccharification is accomplished by β-1,4-glucosidase, which will decrease the inhibitory effect of cellobiose on both endo-1,4-β-D-glucanases and exo-1,4-β-D-glucanases, which would have resulted in decreases in yield and reaction rate (Shen et al. 2008). However, C. thermocellum has not yet evolved an efficient β-1,4-glucosidase because there is no selection pressure, i.e., it can directly assimilate cellodextrins (Zhang and Lynd 2005). Recently, Prawitwong et al. (2013) found that the percent conversion of cellulose greatly increased from about 30% to more than 90% by supplementing the crude enzymes of C. thermocellum S14 with a thermostable β-1,4-glucosidase from Thermoanaerobacter brockii at 60 °C for 7 days. This enzyme could possibly be employed in the core cellulase system mentioned above.

In this study, a core cellulase cocktail, consisting of an endo-1,4-β-D-glucanase (CelD) (Joliff et al. 1986), a reducing-end-acting exo-1,4-β-D-glucanase (CBH48Y) (Liu et al. 2012), a non-reducing-end-acting exo-1,4-β-D-glucanase (CBHA) (Kataeva et al. 2002) from C. thermocellum DSM 1237, and a β-1,4-glucosidase (CglT) from T. brockii (Breves et al. 1997), was established for the hydrolysis of regenerated cellulose and pretreated corn stover. By using the cellulase cocktail, more efficient members of cellulases and hemicellulases could be introduced for better biomass saccharification in the future.

EXPERIMENTAL

Materials

Enzymes for molecular cloning were purchased from Takara, Dalian, China. pEasy E2 expression vector and Escherichia coli Transetta (F omphsdSB (rBmBgal dcm (DE3) pRARE (argU, argW, ileX, glyT, leuW, proL) (CamR)) were both from TransGen, Beijing, China. Avicel PH101 (Sigma-Aldrich, St. Louis, USA) was used to produce regenerated cellulose. The glucose quantification kit was from Senbeijia, Nanjing, China. p-Nitrophenyl-β-D-glucoside (pNPG) was from Sangon Biotech, Shanghai, China. Other chemical reagents were analytically pure and obtained from Sinopharm Chemical, Shanghai, China.

Methods

Gene clone

The celdcbh48y, and truncated cbha gene were PCR cloned from the genomic DNA of C. thermocellum DSM 1237 (corresponding to ATCC 27405, DSMZ, Germany) using LA taq and primers, as listed in Table 1. The cglt gene was PCR cloned from an artificial synthetic gene with codons optimized (GenBank accession: KP401766), which was synthesized by the BGI company (Shenzhen, China). The gene fragments of celd and cglt were double digested by EcoRV and XhoI, and cbha was double digested by SmaI and XhoI, which were introduced to the EcoRV/XhoI sites of expression vector pET 22r (Geng et al. 2012) separately. The PCR product, cbh48y, was directly T-cloned into a pEasy E2 expression vector. The resulting plasmids carrying celd or cbha were transformed in E. coli BL21 separately. The resulted plasmids carrying cbh48y or cglt were transformed into E. coli Transetta for better gene expression. All the resulting plasmids were verified by sequencing.

Table 1. Gene Information

Gene expression and protein purification

Each transformant carrying one of the four genes was inoculated into 4 mL of Luria–Bertani (LB) media supplemented with a proper antibiotic and grown at 37 °C overnight. Then, the overnight culture was inoculated into 500 mL of LB, grown to OD 1.2 at 37 °C, and induced at 50 to 100 μM isopropyl-β-D-thio-galactoside (IPTG) at 25 °C to 30 °C, 120 rpm, for 6 to 12 h.

Cells were harvested and disrupted using a supersonic-wave disrupting method described by Yan et al. (2013) and centrifuged at 20,000 g for 15 min. Subsequently, the supernatant was heat treated at 60 °C for 25 min and centrifuged at 8,000 g for 10 min. The supernatant was loaded onto a nickel column (HisTrap FF column, GE Healthcare, Piscataway, NJ), washed with 40 mM imidazole, and eluted with 500 mM imidazole. Finally, the purified enzymes were buffer-exchanged into a solution (5 mM Na3PO4 and 50 mM NaCl, pH 7.4) using a HiTrap desalting column (GE Healthcare) and stored at -20 °C for use. The purity of proteins was determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) using a Mini-PROTEAN II electrophoresis cell (Bio-Rad, Shanghai, China), followed by Coomassie Brilliant Blue R-250 staining.

Enzymatic activity assay

The enzymatic activities of CelD, CBHA, and CBH48Y were assayed using the DNS method (Miller 1959). A standard assay contains 10 μL of properly diluted enzymes, 200 μL of regenerated amorphous cellulose (5.38 g/L glucose equivalent) (Zhang et al. 2006), 100 μL of buffer (100 mM NaAc-HAc, 20 mM CaCl2, pH 6.0), and 90 μL of water. The enzymatic activity of CelD was determined at 60 °C for 1 h, while the activities of CBHA and CBH48Y were determined at 60 °C for 12 h. The activity of CglT was determined by employing the above reaction system, except using pNPG as a substrate at 60 °C for 25 min, as described by Magalhães et al. (2006). A unit of enzymatic activity was defined as the number of micromoles of reducing sugar or p-nitrophenol released per minute per milligram of protein. For thermostability tests, suitable amounts of enzymes were subjected to the standard enzymatic assay after having been incubated at 60 °C for 24 h.

Cellulose saccharification

Regenerated cellulose and corn stalk pretreated by cellulose-solvent-based lignocellulose fractionation (Zhu et al. 2009), were used as substrate to investigate the effects of various compositions of the four enzymes on the saccharification efficiency. Unless otherwise stated, the saccharification was carried out at 60 °C for 24 h without shaking in a total volume of 1170 μL, with a proper content of cellulose and cellulases, as well as 200 μL of acetate buffer. Glucose content was determined using a glucose oxidase kit (Senbeijia, Nanjing, China). The content of cellulose fraction in the pretreated corn stalk was calculated as described by Zhu et al. (2009). Unless otherwise stated, the percentage production of total reducing sugar divided by total cellulose (in glucose equivalents) was used to represent saccharification efficiency. All results are the average of three replicates, and Student’s t-test was performed with Excel 2007 (Microsoft Corp., Redmond, WA) to analyze the significance of the differences.

RESULTS AND DISCUSSION

Expression and Purification of the Four Cellulases

The four purified cellulases each presented a strong band, as analyzed by SDS-PAGE (Fig. 1). The sizes of CelD, CBHA, CBH48Y, and CglT were all as expected (72.4, 104.9, 105.7, and 51.7 kDa, respectively). Although some proteins partially formed inclusion bodies during the expression process, the soluble fractions of the four cellulases were successfully purified using the four steps of centrifuging, heat treatment, nickel-affinity purification, and desalting, as described above. The productivities of these cellulases were currently not considered in this study. Of course, some efficient strategies could be employed to decrease the inclusion bodies and increase the protein production (Esposito and Chatterjee 2006).

Fig. 1. Purified cellulases on a 10% SDS-PAGE gel stained with Coomassie Brilliant Blue R-250: Lane 1, CglT; Lane 2, CBH48Y; Lane 3, protein marker; Lane 4, CBHA; Lane 5, CelD

Effect of CelD Load on Saccharification

The exoglucanase activities (CBHA plus CBH48Y at a 1:1 ratio) and β-glucosidase activities were set at 0.28 and 11.6 U/g cellulose, respectively, which were situated at medium levels as were found in our preliminary tests (data not shown). As the load of the endoglucanase, CelD, increased from 0 to 12 μL (85.9 U/g cellulose), the production of total reducing sugar remarkably increased from 290 to 436 μg (percent conversion from 27.0% to 40.5%). However, as the CelD load further increased 6.5 times (from 12 to 75 μL), the production of total reducing sugar only slightly increased, to 47.1% (Fig. 2).

Fig. 2. Effect of CelD load on saccharification. The activity of CelD was 7.7 mU/μL. The exoglucanase activities (CBHA plus CBH48Y at a 1:1 ratio) and β-glucosidase activities were set at 0.28 and 11.6 U/g cellulose, respectively. The load of regenerated cellulose was 2.69 g/L, and the total reaction volume was 400 μL. Saccharification levels are presented in the form of total reducing sugar and percent conversion.

The production of reducing sugar at a CelD load of 50 μL was not significantly different from that at 75 μL (p=0.057). Apparently, endo-1,4-β-D-glucanase can provide new ends for exo-1,4-β-D-glucanase and thus is important for saccharification.

Effect of CglT Load on Saccharification

The endoglucanase and exoglucanase activities (CBHA plus CBH48Y at a 1:1 ratio) were set at 85.9 and 0.28 U/g cellulose, respectively. As the load of the β-glucosidase, CglT, was increased from 0 to 10 μL (16.6 U/g cellulose), the production of total reducing sugar greatly increased, from 190 to 468 μg (percent conversion from 17.6% to 43.5%). As the load of CglT was further increased 2.1 times (from 10 to 21 μL), the production of total reducing sugar only slightly increased, to 44.6% (Fig. 3). The production of reducing sugar at a CglT load of 10 μL was not significantly different from that at 21 μL (p=0.22). Usually, the middle product, cellobiose, has an inhibitory effect on both endo-1,4-β-D-glucanases and exo-1,4-β-D-glucanases (Shen et al. 2008); therefore, increasing the load of CglT greatly increased the production of reducing sugar.

Fig. 3. Effect of CglT load on saccharification. The activity of CglT was 1.79 mU/μL. The endoglucanase activity and exoglucanase activities (CBHA plus CBH48Y at a 1:1 ratio) were set at 85.9 and 0.28 U/g cellulose, respectively. The load of regenerated cellulose was 2.69 g/L, and the total reaction volume was 400 μL. Saccharification levels are presented in the form of total reducing sugar and percent conversion.

Effect of CBHA and CBH48Y Proportion on Saccharification

The endoglucanase activity, exoglucanase activities (CBHA plus CBH48Y), and β-glucosidase activities were set at 85.9, 0.28, and 16.6 U/g cellulose, respectively. As the fraction of CBHA increased from 0% to 100% (i.e., CBH48Y decreased from 100% to 0%), the production of total reducing sugar underwent an inverse-U fluctuation (Fig. 4). The highest production of reducing sugar was 462 μg (42.9% conversion) at 50% CBHA, in contrast to only 385 and 245 μg reducing sugar at 100% CBH48Y and 100% CBHA, respectively. Hence, the optimal proportion of CBHA:CBH48Y was 1:1.

The reducing-end-acting and non-reducing-end-acting exo-1,4-β-D-glucanases are two oppositely functioning enzymes and usually function synergistically in cellulose hydrolysis. Hence, the cooperation of CBHA and CBH48Y maximally produced 120% reducing sugar compared with CBH48Y alone, and produced 188% reducing sugar compared with CBHA alone. Similarly, Liu et al (2012) reported 134% enhancement in reducing sugar using a combination of recombinant CBHA and CBH48Y (expressed in Bacillus subtilis) compared with that of the sum of each single enzyme. Barr et al. (1996) also found synergistic effects between these two classes of exocellulases from Thermomonospora fusca and T. reesei. These two types of exoglucanase worked simultaneously from the two ends of a cellulose chain without competition on the substrate, showing increased efficiency in cellulose decomposition.

Fig. 4. Effect of CBHA and CBH48Y proportion on saccharification. The endoglucanase activity, exoglucanase activities (CBHA plus CBH48Y), and β-glucosidase activities were set at 85.9, 0.28, and 16.6 U/g cellulose, respectively. The load of regenerated cellulose was 2.69 g/L, and the total reaction volume was 400 μL. Saccharification levels are presented in the form of total reducing sugar and percent conversion.

Effect of Exoglucanase Load on Saccharification

The endoglucanase activity and β-glucosidase activities were still set at 85.9 and 16.6 U/g cellulose. The exoglucanase activities of 0.28 U/g cellulose were set as one-fold, and the proportion of CBHA:CBH48Y was fixed at 1:1 (Fig. 5). As the load of exoglucanases was increased from 0 to 3.0 fold, i.e., from 0 to 0.84 U/g cellulose, the production of total reducing sugar remarkably increased from 80 to 624 μg (7.4% to 58% conversion), suggesting a great influence of exoglucanases on the saccharification of regenerated cellulose. In addition, the exoglucanase activities were much less than that of the endoglucanase and β-glucosidase, which was consistent with other reports. Gao et al. (2010) recently optimized purified commercial fungal cellulases and found that the optimal activities of endo-1,4-β-D-glucanase, exo-1,4-β-D-glucanase, and β-D-glucosidase activities were 60.3, 0.32, and 372 U/g glucan, respectively (or 9.0, 14.6, and 3 mg/g glucan, respectively). Obviously, the optimization of both fungal and bacterial cellulases suggests the usage of a heavy β-1,4-glucosidase load (to effectively decompose cellobiose), followed by endo-1,4-β-D-glucanase (to provide new ends for exo-1,4-β-D-glucanase) and exo-1,4-β-D-glucanase, the last to produce cellobiose.

Fig. 5. Effect of exoglucanase load on saccharification. One-fold activity of CBHA and CBH48Y was set at 0.30 mU (0.28 U/g cellulose). The endoglucanase activity and β-glucosidase activities were still set at 85.9 and 16.6 U/g cellulose. The load of regenerated cellulose was 2.69 g/L, and the total reaction volume was 400 μL. Saccharification levels are presented in the form of total reducing sugar and percent conversion.

Effect of β-glucosidase in the High Enzyme-Load Situation

When all four enzymes were set at the optimized proportion and at double activities (CelD at 515 U/g cellulose, the two exoglucanases together at 1.68 U/g cellulose at a 1:1 ratio, and CglT at 99.6 U/g cellulose), the production of total reducing sugar reached 742 μg (68.9% conversion) (Fig. 6).

Fig. 6. Effect of β-glucosidase in the high enzyme-load situation. The activity of CglT was 1.79 mU/μL. The activity of CelD was 515 U/g cellulose, and the activity of the two exoglucanases together was 1.68 U/g cellulose at a 1:1 ratio. The load of regenerated cellulose was 0.92 g/L, and the total reaction volume was 1170 μL. Saccharification levels are presented in the form of total reducing sugar and percent conversion.

To further reduce the middle product inhibition by cellobiose, additional CglT was added. As the CglT increased from 60 to 480 μL (from 99.6 to 796 U/g cellulose), the production of total reducing sugar kept increasing, from 742 to 860 μg (percent conversion from 68.9% to 80.0%). Subsequently doubling the activities of all the enzymes, however, the production of total reducing sugar only slightly increased, from 80.0% to 81.5%. Obviously, higher β-1,4-glucosidase load further decreased the inhibitory effect of cellobiose and increased the production of glucose.

Saccharification under Various Substrate Loads

According to the optimized composition, the activities of CelD, CBHA, CBH48Y, and CglT were fixed at 554, 0.91, 0.91, and 856 mU per assay, respectively (i.e., for 0.92 g/L cellulose, the CelD, CBHA, CBH48Y, and CglT were set at 515, 0.84, 0.84, and 796 U/g cellulose, respectively). As the load of regenerated cellulose was gradually increased from 0.92 to 4.58 g/L, the glucose production increased linearly (R2=0.997) from 870 to 3208 μg, each approximate to the amount of production of total reducing sugar, respectively (Fig. 7). However, the percent conversion gradually decreased, from 80.9% to 59.6%. For the saccharification of pretreated corn stover, as the cellulose fraction increased from 0.63 to 1.90 g/L, the glucose production gradually increased, from 444 to 944 μg. However, the percent conversion gradually decreased, from 70.1% to 49.6%.

Fig. 7. Saccharification under various substrate loads. The activity of CelD, CBHA, CBH48Y, and CglT was fixed to 554, 0.91, 0.91, and 856 mU per assay, respectively, and the total reaction volume was 1170 μL. The total reducing sugar production and glucose production are shown. The percent values above the bars indicate the percent conversions, which were calculated based on the glucose production. RAC: regenerated amorphous cellulose. PCS: pretreated corn stover.

The yield of saccharification is affected by several factors, for example, enzyme activities, substrate accessibility and concentration, and product inhibition. All four cellulases remained active throughout the saccharification; however, after incubation at 60 °C for 24 h, CelD remained at about 10% activity, and both CBHA and CBH48Y remained at about 60% activity. The highly thermostable CglT was reported to remain at 97% activity under the same situation (Prawitwong et al. 2013). The gradual decrease in enzyme activity might have affected the saccharification. Although the corn stover had been pretreated, with its structure destroyed, the residual hemi-cellulose chains might still have interfered with the cellulose accessibility to cellulases or might have directly blocked the hydrolysis (Viikari et al. 2012). Therefore, obviously, the percent conversion of pretreated corn stover was lower than that of the regenerated cellulose (Fig. 7). In addition, the continuous decrease in substrate might also have limited the enzymatic activities; that is why prolonging the saccharification to 48 h did not substantially increase glucose production (data not shown). The strong linear relationship (R2 was 0.997) between the glucose production and the substrate load suggested that a higher substrate load would stimulate higher enzyme efficiency. Hence, it could be better to maintain a high substrate load and allow periodic supplementation of substrate and enzymes, as well as the removal of products, for higher saccharification efficiency.

The relationship between cellulases and biomass is subtle. Different kinds of biomass are composed of various components and structures and therefore require various cellulolytic enzymes; in other words, specific biomass might also require specific components of cellulolytic enzymes. Fortunately, there are many highly evolved cellulolytic microbes, whose genomes have been sequenced (Rubin 2008), from which different types of efficient enzymes could be recruited into the enzyme cocktails, which, in turn, could possibly enhance the level of biomass saccharification.

CONCLUSIONS

  1. Four cellulases were successfully synthetized in E. coli and purified. The composition of a cellulase system using these cellulases was optimized for efficient saccharification of both regenerated cellulose and pretreated corn stover.
  2. All four cellulases were essential for saccharification, although the proportion among them was very different.
  3. Glucose production was positively correlated with the concentration of regenerated cellulose, implying the importance of a high substrate load in saccharification.

ACKNOWLEDGMENTS

This work was financially supported by the National Natural Science Foundation of China (31201752), the Startup Foundation of Jiangsu University (11JDG109), and the Priority Academic Program Development of Jiangsu Higher Education Institution. Jian Wu acknowledges the support of the Natural Science Fund for Colleges and Universities in Jiangsu Province, China (13KJB530003).

REFERENCES CITED

Agbor, V. B., Cicek, N., Sparling, R., Berlin, A., and Levin, D. B. (2011). “Biomass pretreatment: Fundamentals toward application,” Biotechnol. Adv. 29(6), 675-685. DOI: 10.1016/j.biotechadv.2011.05.005

Barr, B., Hsieh, Y., Ganem, B., and Wilson, D. B. (1996). “Identification of two functionally different classes of exocellulases,” Biochemistry 35(2), 586-592. DOI: 10.1021/bi9520388

Berlin, A., Maximenko, V., Gilkes, N., and Saddler, J. (2007). “Optimization of enzyme complexes for lignocellulose hydrolysis,” Biotechnol. Bioeng. 97(2), 287-296. DOI: 10.1002/bit.21238

Breves, R., Bronnenmeier, K., Wild, N., Lottspeich, F., Staudenbauer, W. L., and Hofemeister, J. (1997). “Genes encoding two different beta-glucosidases of Thermoanaerobacter brockiiare clustered in a common operon,” Appl. Environ. Microbiol. 63(10), 3902-3910.

Esposito, D., and Chatterjee, D. K. (2006). “Enhancement of soluble protein expression through the use of fusion tags,” Curr. Opin. Biotechnol. 17(4), 353-358. DOI: 10.1016/j.copbio.2006.06.003

Gao, D. H., Chundawat, S. P. S., Krishnan, C., Balan, V., and Dale, B. E. (2010). “Mixture optimization of six core glycosyl hydrolases for maximizing saccharification of ammonia fiber expansion (AFEX) pretreated corn stover,” Bioresource Technol. 101(8), 2770-2781. DOI: 10.1016/j.biortech.2009.10.056

Geng, A. L., Zou, G., Yan, X., Wang, Q. F., Zhang, J., Liu, F. H., Zhu, B. L., and Zhou, Z. H. (2012). “Expression and characterization of a novel metagenome-derived cellulase Exo2b and its application to improve cellulase activity in Trichoderma reesei,” Appl. Microbiol. Biotechnol. 96(4), 951-962. DOI: 10.1007/s00253-012-3873-y

Gold, N. D., and Martin, V. J. J. (2007). “Global view of the Clostridium thermocellum cellulosome revealed by quantitative proteomic analysis,” J. Bacteriol. 189(19), 6787-6795. DOI: 10.1128/JB.00882-07

Joliff, G., Béguin, P., Juy, M., Millet, J., Ryter, A., Poljak, R., and Aubert, J. P. (1986). “Isolation, crystallization and properties of a new cellulase of Clostridium thermocellumoverproduced in Escherichia coli,” Nature Biotechnol. 4, 896-900. DOI: 10.1038/nbt1086-896

Kataeva, I. A., Seidel, R. D., Shah, A., West, L. T., Li, X. L., and Ljungdahl, L. G. (2002). “The fibronectin type 3-like repeat from the Clostridium thermocellum cellobiohydrolase CbhA promotes hydrolysis of cellulose by modifying its surface,” Appl. Environ. Microbiol. 68(9), 4292-4300. DOI: 10.1128/AEM.68.9.4292-4300.2002

Liu, J. M., Xin, X. J., Li, C. X., Xu, J. H., and Bao, J. (2012). “Cloning of thermostable cellulase genes of Clostridium thermocellum and their secretive expression in Bacillus subtilis,” Appl. Biochem. Biotech. 166(3), 652-662. DOI: 10.1007/s12010-011-9456-z

Magalhães, P. O., Ferraz, A., and Milagres, A. F. M. (2006). “Enzymatic properties of two β-glucosidases from Ceriporiopsis subvermispora produced in biopulping conditions,” J. Appl. Microbiol. 101(2), 480-486. DOI: 10.1111/j.1365-2672.2006.02946.x

Maki, M., Leung, K. T., and Qin, W. S. (2009). “The prospects of cellulase-producing bacteria for the bioconversion of lignocellulosic biomass,” Int. J. Biol. Sci. 5(5), 500-516. DOI: 10.7150/ijbs.5.500

Meyer, A. S., Rosgaard, L., and Sørensen, H. R. (2009). “The minimal enzyme cocktail concept for biomass processing,” J. Cereal Sci. 50(3), 337-344. DOI: 10.1016/j.jcs.2009.01.010

Miller, G. L. (1959) “Use of dinitro-salicylic acid reagent for determination of reducing sugars,” Anal. Chem. 31, 426-428.

Prawitwong, P., Waeonukul, R., Tachaapaikoon, C., Pason, P., Ratanakhanokchai, K., Deng, L., Sermsathanaswadi, J., Septiningrum, K., Mori, Y., and Kosugi, A. (2013). “Direct glucose production from lignocellulose using Clostridium thermocellum cultures supplemented with a thermostable beta-glucosidase,” Biotechnol. Biofuels 6, 184. DOI: 10.1186/1754-6834-6-184

Rubin, E. M. (2008). “Genomics of cellulosic biofuels,” Nature 454(7206), 841-845. DOI: 10.1038/nature07190

Shen, Y., Zhang, Y., Ma, T., Bao, X. M., Du, F. G., Zhuang, G. Q., and Qu, Y. B. (2008). “Simultaneous saccharification and fermentation of acid-pretreated corncobs with a recombinant Saccharomyces cerevisiae expressing beta-glucosidase,” Bioresource Technol. 99(11), 5099-5103. DOI: 10.1016/j.biortech.2007.09.046

Taherzadeh, M. J., and Karimi, K. (2007). “Enzyme-based hydrolysis processes for ethanol from lignocellulosic materials: A review,” BioResources 2(4), 707-738. DOI: 10.15376/biores.2.4.707-732

Viikari, L., Vehmaanpera, J., and Koivula, A. (2012). “Lignocellulosic ethanol: From science to industry,” Biomass Bioenerg. 46, 13-24. DOI: 10.1016/j.biombioe.2012.05.008

Yan, X., Geng, A. L., Zhang, J., Wei, Y. J., Zhang, L., Qian, C. L., Wang, Q. F., Wang, S. Y., and Zhou, Z. H. (2013). “Discovery of (hemi-) cellulase genes in a metagenomic library from a biogas digester using 454 pyrosequencing,” Appl. Microbiol. Biotechnol. 97(18), 8173-8182. DOI: 10.1007/s00253-013-4927-5

Zhang, Y. H. P., and Lynd, L. R. (2005). “Cellulose utilization by Clostridium thermocellum: Bioenergetics and hydrolysis product assimilation,” Proc. Natl. Acad. Sci. USA 102(20), 7321-7325. DOI: 10.1073/pnas.0408734102

Zhang, Y. H. P., Cui, J. B., Lynd, L. R., and Kuang, L. R. (2006). “A transition from cellulose swelling to cellulose dissolution by o-phosphoric acid: Evidence from enzymatic hydrolysis and supramolecular structure,” Biomacromolecules 7(2), 644-648. DOI: 10.1021/bm050799c

Zhu, Z. G., Sathitsuksanoh, N., Vinzant, T., Schell, D. J., McMillan, J. D., and Zhang, Y. H. P. (2009). “Comparative study of corn stover pretreated by dilute acid and cellulose solvent-based lignocellulose fractionation: Enzymatic hydrolysis, supramolecular structure, and substrate accessibility,” Biotechnol. Bioeng. 103(4), 715-724. DOI: 10.1002/bit.22307

Article submitted: July 13, 2015; Peer review completed: September 12, 2015; Revised version received and accepted: September 21, 2015; Published: September 28, 2015.

DOI: 10.15376/biores.10.4.7681-7692