Continuous L-lactic acid production from defatted rice bran hydrolysate using corn stover bagasse immobilized carrier

Abstract

In this paper, L-lactic acid (LLA) was produced using defatted rice bran hydrolysate. Agricultural waste corn stover bagasse (CSB), under the conditions of good mechanical and chemical stabilities, was used as the immobilized carrier. As a result, a maximal concentration of 88 g L⁻¹ of LLA with an average yield of 0.95 g g⁻¹ and productivity of 5.20 g L⁻¹ h⁻¹ were achieved in a single-stage immobilized repeated-batch fermentation. In addition, when single-stage immobilized continuous fermentation was carried out, a maximal LLA concentration of 84 g L⁻¹ was obtained with an average yield and productivity of 0.97 g g⁻¹ and 5.73 g L⁻¹ h⁻¹, respectively. To balance the increases in the productivity, concentration and yield of LLA, two-stage immobilized continuous fermentation was performed. LLA productivities of 6.20 g L⁻¹ h⁻¹ in the first stage and 2.18 g L⁻¹ h⁻¹ in the second stage were achieved. The whole immobilized fermentation process has potential in L-lactic acid biorefineries and industrialization.

---

I. Introduction

With the deep concerns regarding unsustainable development, L-lactic acid (LLA) has attracted interest. LLA has versatile applications in the food, pharmaceutical, cosmetic, textile, leather, and other chemical industries.^1–4 More importantly, LA is the precursor of PLA, a kind of renewable and biodegradable plastic. Hence, the demand for LLA has increased continuously for years.^5,6 Compared to other routes of LLA production, fermentation has advantages in mild production conditions, low energy consumption,⁷ excellent optical purity of the product and safety.⁸ Unfortunately, the high costs of the food and starchy feedstock has limited the industrialization and large-scale applications of most LLA fermentation processes.⁹

The utilization of agricultural residuals for LLA production is attractive due to the low prices of feedstock.¹⁰ Among the different types of agricultural residuals, lignocellulosic biomass is relatively difficult to use because of the toxicity of the hydrolysate and the uneconomical competitiveness of pretreatment.⁸ In contrast, defatted rice bran, the residue from brown rice, is one of the most abundant agricultural by-products in South East Asia.¹¹ After squeezing the rice bran oil, the residual powder of defatted rice bran contains abundant starch to make it feasible for LLA fermentation according to a previous study.¹²

Cell immobilization is a technique to improve the cell density and to maintain the catalytic activity in repeated-use by physical or chemical means to position or limit free cells to a specific region of space. Therefore, with increasing cell concentration during immobilized fermentation, a higher productivity of LLA can be generated. In addition, cell immobilization has the ability to improve the fermentation system stability, decrease the requirement of nutrients, and reduce the difficulties of LLA recovery to make production more efficient.¹³

Traditional immobilized carrier materials include sodium alginate and polyvinyl alcohol.^14,15 However, the immobilized carriers have shortcomings of higher costs, damage to cells in the cumbersome immobilization process and carrier softening. To overcome the disadvantages of the traditional immobilized carriers, various natural materials, such as fruit pieces¹⁶ and lignocellulosic materials,¹⁷ have been applied successfully as immobilized carriers in LLA fermentation.

Corn is an abundant agricultural crop¹⁸ with an extended planting area and good adaptation of margin land. Corn stover bagasse (CSB), residue of corn with commercial value, has been recognized as an excellent immobilized carrier according to previous research,¹⁹ which had not yet been applied and studied in immobilized LLA fermentation. As in the case in China, the main growing area of corn stover is coincident with the rice growing areas. Therefore, under the concept of a biorefinery, the defatted rice bran and CSB, two types of agricultural residuals, would be applied together in a certain immobilized fermentation process of LLA that retains the superiority of each.²⁰

The purpose of this research were to obtain efficient and inexpensive LLA with a high LLA yield and productivity. Defatted rice bran was used as carbon source and the CSB was used directly as the immobilized carrier. Both the single-stage immobilized repeated-batch and continuous LLA fermentation were performed. Besides, aiming to further increase the productivity and yield of LLA, the two-stage immobilized continuous fermentation was also carried out.

II. Experimental section

A. Microorganism and culture

Lactobacillus rhamnosus LA-04-01 stored in our lab was used in this work. The semi-synthetic medium consisted of 20 g L⁻¹ of glucose, 15 g L⁻¹ of yeast extract, 10 g L⁻¹ of soya peptone, 5 g L⁻¹ of sodium acetate, 2 g L⁻¹ of ammonium citrate, 10 g L⁻¹ of NaCl, 0.2 g L⁻¹ of MgSO₄·7H₂O, 0.05 g L⁻¹ of MnSO₄·7H₂O, 20 g L⁻¹ of agar. It was grown on the semisynthetic medium at 42 °C for 36 h and the stock cultures were maintained at 4 °C. The medium for cell growth contained 40 g L⁻¹ of glucose, 10 g L⁻¹ of yeast extract, 10 g L⁻¹ of soya peptone, 0.5 g L⁻¹ of sodium acetate, 0.2 g L⁻¹ of ammonium citrate, 0.01 g L⁻¹ of NaCl, 0.2 g L⁻¹ of KH₂PO₄, 0.2 g L⁻¹ of MgSO₄·7H₂O, 0.05 g L⁻¹ of MnSO₄·7H₂O and 40 g L⁻¹ of CaCO₃. The initial concentration of the carbon source and yeast extract in the medium for LLA production were 120 g L⁻¹ and 15 g L⁻¹, respectively, and the remaining compositions were the same as those of the medium for cell growth.

B. Defatted rice bran hydrolysis

The hydrolysis of defatted rice bran was performed according to the method reported by Wang.¹² An appropriate amount of defatted rice bran was formulated into a 200–250 g L⁻¹ suspension of defatted rice bran with water. The pH of the defatted rice bran suspension was adjusted to 6.0 with a 20% sodium hydroxide solution. An appropriate amount of amylase (Liquozyme® Supra/2.2X) was then added to the defatted rice bran suspension at a ratio of 1 mL kg⁻¹ (v/w) in a high-pressure steam sterilization pot, the temperature was kept at 108 °C for 5 min and 95 °C for 60 min to complete the liquefaction of defatted rice bran. When the temperature of the defatted rice bran suspension was cooled to 60 °C, the pH of the medium was adjusted to 4.2. An appropriate amount of glucoamylase (Dextrozyme DX/1.5X) was then added to the medium at a ratio of 1 mL kg⁻¹ (v/w) at 60 °C for 24 h of hydrolysis. After hydrolysis, the defatted rice bran hydrolysate supernatant was obtained by centrifugation, and used as a substrate for fermentation.

C. Pretreatment of corn stover bagasse as a carrier

According to the method reported by Yu,²¹ the CSB were collected and peeled off. The spongy tissue of the stem was then cut into suitable particle size, and placed in hot water to soak repeatedly 4–8 times until the sugar in the stem had been fully released. Subsequently, the CSB was dried at 60 °C. The particle size and the addition of carriers were adjusted to meet the different requirements of the experiments for further study.

D. Cell immobilization fermentation

The initial sugar concentration of the defatted rice bran hydrolysate supernatant medium was adjusted to 120 g L⁻¹ 250 mL flasks contained 100 mL medium of the defatted rice bran hydrolysate supernatant and 5 g of calcium carbonate in flask fermentation, while a 1 L fermenter contained 250 mL medium of the defatted rice bran hydrolysate supernatant and 7.5 g of calcium carbonate. According to the specific requirements of the experiments, different amounts of CSB carriers were added to the fermenter, and the pH of the medium was adjusted to 6.0. The fermenter with the medium inside was kept at 121 °C for 20 min of sterilization. The medium was then cultivated at 42 °C, 180 rpm with the inoculum of 20% (v/v). For repeated-batch fermentation, batch fermentation was carried out first in a 1 L fermenter, and after the end of batch fermentation, the suspend liquor was pumped out by a peristaltic pump (Baoding Chuangrui Precision Pump Co., Ltd). The sterilized defatted rice bran hydrolysate was pumped into the fermentation system under a certain liquid to solid ratio for another batch of LLA fermentation. Therefore, fermentation could operate repeatedly and the immobilized cells on the surface of the CSB carrier could maintain their activity because of the fresh substrate pumped into the fermenter periodically. For the single-stage continuous fermentation, fresh defatted rice bran hydrolysate was pumped into the fermenter continuously at different dilution rates, and the fermentation broth was pumped out continuously under the same rate of dilution to maintain stable fermentation. As for the two-stage immobilized continuous fermentation, two 1 L fermenters in series were used. The fermentation broth with a high concentration of residual sugar after the first stage of fermentation was pumped into the second stage as the substrate. The immobilized cells further use the sugar in the fermentation broth to produce LLA. The entire dilution rate of each stage was the same to maintain a stable two-stage system.

E. Preparation of SEM sample of carriers

According to the literature,²¹ The CSB with the bacteria cells was first kept in the 3.5% glutaraldehyde solution and soaked for 6 h, then washed successively with 50%, 60%, 70%, 90%, 95%, and 100% ethanol solutions for dehydration, and then placed in a vacuum freeze dryer overnight to dry.

F. Analysis

1 mL of fermentation broth was collected and centrifuged at 8000 rpm for 5 min. After the appropriate dilution, the concentrations of LLA and glucose in the supernatant liquid were detected using an SBA-40C biosensor analyzer (Institute of Biology, Shandong Province Academy of Sciences, P. R. China).^22,23

III. Results and discussion

A. The effect of carrier addition on fermentation

As the immobilized repeated batch fermentation time increased, the concentration and productivity of LLA increased. The concentration and productivity of LLA gradually became stable after 5 batches of fermentation (Fig. 1(a) and (b)). The concentration and productivity of LLA did not increase with increasing carrier loading until the carrier addition was above 4% (w/v). On the contrary, the concentration and productivity of LLA were lower than those of 3–4% (w/v). This might be caused by the low level of the liquid phase in the medium, leading to non-uniform mixing and mass-transfer with the calcium carbonate promptly, resulting in a lower pH, which is harmful to the metabolism of the bacteria.^13,24,25 Therefore, the optimized immobilized carrier loading was 3–4% (w/v). After 4 batches of fermentation, the LLA productivity could be maintained at about 3.00 g L⁻¹ h⁻¹.

Fig. 1 The effect of carrier addition on the immobilized fermentation: (a) the effect of carrier addition on the LLA concentration in immobilized fermentation (2% (w/v): the 20 g carrier was added into the 1 L medium, the rest were the same); (b) the effect of carrier addition on the LLA productivity in immobilized fermentation (2% (w/v): the 20 g carrier was added into the 1 L medium, the rest were the same).

SEM of the cross-section and the vertical-section of the carriers with immobilization showed that a large number of Lactobacillus rhamnosus cells were embedded in the cavities among the structure of the carrier. The structure of the carrier was a rough surface, and the empty cavities would provide a stable mini-environment which contributed to the metabolism of the bacteria (Fig. 2).¹⁹ Therefore, cell immobilization using CSB, a type of lignocellulosic biomass, had a significant effect on cell enrichment, and good reproducibility for batch fermentation (Fig. 1(a) and (b)). That is, natural CSB had excellent biocompatibility for cell immobilization.

Fig. 2 Carrier with Lactobacillus rhamnosus embedded inside and the original carrier under SEM: (a) transverse section of the original carrier showed plant stem cell cavities; (b) transverse section of the carrier showed a large number of Lactobacillus rhamnosus entrapped in the plant stem cell cavities; (c) the center enlargement of (b); (d) longitudinal section of the original carrier showed plant stem cell cavities; (e) longitudinal section of the carrier indicated a large number of Lactobacillus rhamnosus entrapped in the plant stem cell cavities.

B. The effect of carrier particle size on fermentation

As shown in Fig. 3(a) and (b), the different particle size of the CSB carrier had a slight impact on the concentration and productivity of LLA. However, with increasing carrier particle size, the increasing range of the LLA yield was lower. Therefore, the efficient surface areas of the different carrier particle sizes were similar. As shown in Fig. 3(b), the carrier particle size of 5 mm was the optimal choice. After 4 batches of fermentation, the average LLA productivity was stable at about 2.90 g L⁻¹ h⁻¹; the highest productivity of LLA was 2.96 g L⁻¹ h⁻¹. In addition, because CSB has porous characteristics, the immobilized carrier had a wide range of applications, and can be pre-treated using various methods to control and change the particle size and porosity of the immobilization carriers, which would significantly improve the capacities of the immobilization carriers, including high cell density and high activity.

Fig. 3 The effect of carrier particle sizes on the immobilized fermentation: (a) the effect of carrier particle sizes on the LLA concentration in immobilized fermentation; (b) the effect of carrier particle sizes on the LLA productivity in immobilized fermentation.

C. The immobilized repeated-batch fermentation

The CSB carriers had excellent immobilization property in LLA fermentation with good reproducibility. In this section, fermentation was carried out for 20 batches, of which the initial four batches were at the stable growth phase of the immobilization fermentation system (Fig. 4). From the fifth batch, the LLA production from fermentation was gradually stabilized with a total of about 16 batches of repeated-batch fermentation. As a result, the LLA concentration remained at about 84 g L⁻¹ in each batch. The LLA productivity was about 5.2 g L⁻¹ h⁻¹, which was almost twice as much as that of fermentation in flasks. The yield of LLA was basically about 0.95 g g⁻¹ at the same time. Owing to the electrostatic interaction between the cells and carrier, the immobilized cell system had a very strong impact of resistance and abrasion resistance. It could protect cells and reduce the activity loss of the cells in long term of fermentation.^26,27 The structure of the CSB carrier played an important role in maintaining the stability of the intracellular pH, preventing acidification, and protecting the microorganisms from the impact of high shear forces.²⁸ Consequently, the CSB fully met the needs of LLA fermentation with cell immobilization.

Fig. 4 The trend of LLA concentration, sugar concentration and LLA productivity in each batch of the immobilized repeated-batch fermentation.

D. The single-stage immobilized continuous fermentation

A total of 600 h of the single-stage immobilized continuous fermentation in 1 L fermenter was then performed, the dilution rates of 0.025 h⁻¹, 0.050 h⁻¹, 0.075 h⁻¹, 0.100 h⁻¹, and 0.125 h⁻¹ were chosen to examine the immobilized continuous fermentation, respectively. As shown in Fig. 5, in general, the LLA concentration of 84 g L⁻¹ and the productivity of 2.10 g L⁻¹ h⁻¹ in fermentation came with a dilution rate of 0.025 h⁻¹. With increasing dilution rate, the concentration of LLA decreased markedly, while the productivity of LLA increased significantly. When the dilution rate rose to 0.125 h⁻¹, the LLA concentration fell to about 44 g L⁻¹ with a productivity of 5.80 g L⁻¹ h⁻¹. The significant decrease in the LLA yield could be due to the high concentration of the residual sugar in the medium. Despite the inhibitory effect of the initial sugar, the inhibition of LLA from the fermentation also had a dramatic impact on the production of LLA. In addition, the concentration of LLA increased in the fermentation broth, the LLA productivity decreased significantly. Compared to the fermentation at a high dilution rate, a higher concentration of LLA was obtained at a lower dilution rate. Therefore, to maximize the productivity of LLA and make full use of the residual carbon source in the medium, the multi-stage immobilized continuous fermentation was a better choice. The progressive consumption of the carbon source resulted in a low residual sugar concentration to improve the LLA yield.

Fig. 5 The trend of LLA concentration, sugar concentration and LLA productivity in single-stage immobilized continuous fermentation under the different dilution rates.

E. The two-stage immobilized continuous fermentation

Although a higher productivity of LLA was achieved using the single-stage immobilized continuous fermentation under a higher dilution rate, however, the final concentration and yield of LLA were relatively low with a high concentration of residual sugar. To maximize the final concentration of LLA and minimize the residual carbon source concentration, two 1 L fermenters in a series connection were performed according to the previous study²⁸ to break the dilution limitations. The optimized parameters above were used in the multiple stage of the immobilized fermentation process.

A total 700 h of the two-stage immobilized continuous fermentation was performed at dilution rates of 0.050 h⁻¹, 0.075 h⁻¹, 0.100 h⁻¹, and 0.125 h⁻¹, respectively. This is because the concentration of residual sugar in the first stage was almost 0 g L⁻¹ at a dilution rate of 0.025 h⁻¹. Substantially, the basic need of the LLA fermentation of the second stage was limited. Therefore the dilution rate of fermentation liquid chosen was 0.050 h⁻¹. As apparent from Table 1, Fig. 6(a) and (b), in the first stage of fermentation, increasing the dilution rate would result in a significant decrease in LLA production, while the productivity of LLA increased significantly, which was the phenomenon of the single-stage immobilized continuous fermentation. In general, a LLA concentration of 78 g L⁻¹ with a productivity of 3.90 g L⁻¹ h⁻¹ and a yield of 0.93 g g⁻¹ was achieved at a dilution rate of 0.050 h⁻¹. However, when the dilution rate reached 0.100 h⁻¹, the LLA concentration fell from 78 g L⁻¹ to 62 g L⁻¹ with an upward movement of the LLA productivity to 6.20 g L⁻¹ h⁻¹, which was 1.71 times of that reported in a previous study.¹²

Table 1 Performance of the two-stage immobilized continuous fermentation at different dilution rates

Dilution rate (h^−1)	The LLA production in first stage			The LLA production in second stage
	Concentration (g L^−1)	Yield (g g^−1)	Productivity (g L^−1 h^−1)	Concentration (g L^−1)	Yield (g g^−1)	Productivity (g L^−1 h^−1)
0.050	78.33	0.93	3.92	5.34	0.98	0.27
0.075	74.22	0.83	5.57	9.56	0.98	0.72
0.100	62.00	0.64	6.20	23.83	0.98	2.38
0.125	50.67	0.54	6.33	20.33	0.83	2.54

---

Fig. 6 The trend of LLA production in two-stage immobilized continuous fermentation: (a) the trend of LLA concentrations and glucose concentrations in two-stage immobilized continuous fermentation; (b) the trend of LLA productivities in two-stage immobilized continuous fermentation under different dilution rates.

However, unlike the trend of LLA production of the first stage, the LLA concentration of the second stage increased from 5.34 g L⁻¹ to 23.83 g L⁻¹ with increasing dilution rate (Table 1). This could be due to the residual sugar concentration of medium outflowed from the first stage at a low dilution rate. As the dilution rate increased, the residual sugar concentration of medium outflowed from the first stage increased slightly. Therefore, a higher LLA concentration could be achieved at a higher dilution rate. In addition, the productivity of LLA produced from the second stage could be increased gradually by increasing the LLA concentration in the second stage. On the contrary, the LLA yield decreased with increasing dilution rate.

According to Fig. 6(b) and Table 1, the optimized dilution rate of two-stage immobilized continuous fermentation was 0.100 h⁻¹. Under this condition, the LLA concentration and yield reached their peaks, which were 86 g L⁻¹ and 0.98 g g⁻¹, respectively, and the productivities of LLA were 6.20 g L⁻¹ h⁻¹ in first stage and 2.18 g L⁻¹ h⁻¹ in second stage, respectively. In addition, the residual sugar concentration of the medium outflowed from the second stage under the dilution rate of 0.100 h⁻¹ was 0 g L⁻¹ (the maximal residual sugar concentration ≤ 2 g L⁻¹), which ensured that the entire cell immobilization fermentation system could be maintain at a higher dilution rate to achieve a higher processing throughput.

As the dilution rate increased, the LLA productivity increased continuously, while the yield and the concentration of LLA significantly decreased (less than 0.90 g g⁻¹ and 80 g L⁻¹), the residual sugar concentration of the fermentation broth from the second stage under a higher dilution rate was more than 10 g L⁻¹, which was not conducive to improving the subsequent separating operation of the LLA.

IV. Conclusions

The utilization of CSB as immobilization carrier material to produce LLA from defatted rice bran hydrolysate was studied. The LLA fermentation with cells immobilized onto CSB carriers could greatly simplify the process (including carrier preparation and cell immobilization) to improve the efficiency. As a result, the optimized size of the carrier was 5 mm, and the addition of carrier was preferable in the range of 3.0–4.0% (w/v). In addition, the phenomenon of the larger particle size having poor mass transfer effects was proven. During the two-stage immobilized continuous fermentation, a maximum LLA concentration of 86 g L⁻¹ with a yield of 0.98 g g⁻¹ was achieved at a dilution rate of 0.100 h⁻¹. Under this condition, the high LLA productivities in the first and second-stages of immobilized continuous fermentation were 6.20 g L⁻¹ h⁻¹ and 2.18 g L⁻¹ h⁻¹, respectively. The biorefinery process based on the defatted rice bran and CSB shows promise in the production of L-lactic acid on an industrial scale.

Acknowledgements

This study was funded by the National Basic Research Program of China (973 program: 2013CB733600, 2012CB725200), the National Nature Science Foundation of China (21390202), and National High-Tech R&D Program of China (2014AA022101, 2014AA021904).

Notes and references

1. C. Gao, C. Q. Ma and P. Xu, Biotechnol. Adv., 2011, 29, 930–939.
2. M. A. Abdel-Rahman, Y. Tashiro and K. Sonomoto, J. Biotechnol., 2011, 156, 286–301.
3. L. J. Chen, A. W. Zeng, H. B. Dong, Q. Li and C. C. Niu, Bioresour. Technol., 2012, 112, 280–284.
4. G. Reddy, M. Altaf, B. J. Naveena, M. Venkateshwar and E. V. Kumar, Biotechnol. Adv., 2008, 26, 22–34.
5. L. M. Wang, B. Zhao, B. Liu, C. Y. Yang, B. Yu, Q. G. Li, C. Q. Ma, P. Xu and Y. H. Ma, Bioresour. Technol., 2010, 101, 7895–7901.
6. T. Tanaka, M. Hoshina, S. Tanabe, K. Sakai, S. Ohtsubo and M. Taniguchi, Bioresour. Technol., 2006, 97, 211–217.
7. R. Datta and M. Henry, J. Chem. Technol. Biotechnol., 2006, 81, 1119–1129.
8. K. Okano, T. Tanaka, C. Ogino, H. Fukuda and A. Kondo, Appl. Microbiol. Biotechnol., 2010, 85, 413–423.
9. H. Oh, Y. J. Wee, J. S. Yun, S. H. Han, S. W. Jung and H. W. Ryu, Bioresour. Technol., 2005, 96, 1492–1498.
10. R. P. John, K. M. Nampoothiri and A. Pandey, Appl. Microbiol. Biotechnol., 2007, 74, 524–534.
11. Y. Peng, H. Shi, X. B. Qi, C. J. Xiao, H. Zhong, R. L. Z. Ma and B. Su, BMC Evol. Biol., 2010, 10, 15.
12. Y. Wang, Z. X. Yang, P. Y. Qin and T. W. Tan, RSC Adv., 2014, 4, 8907–8913.
13. M. A. Abdel-Rahman, Y. Tashiro and K. Sonomoto, Biotechnol. Adv., 2013, 31, 877–902.
14. Y. J. Wang, X. J. Yang, H. Y. Li and W. Tu, Polym. Degrad. Stab., 2006, 91, 2408–2414.
15. K. Y. A. Wu and K. D. Wisecarver, Biotechnol. Bioeng., 1992, 39, 447–449.
16. Y. Kourkoutas, V. Xolias, M. Kallis, E. Bezirtzoglou and M. Kanellaki, Process Biochem., 2005, 40, 411–416.
17. A. K. Chandel, M. L. Narasu, G. Chandrasekhar, A. Manikyam and L. V. Rao, Bioresour. Technol., 2009, 100, 2404–2410.
18. K. Öhgren, A. Rudolf, M. Galbe and G. Zacchi, Biomass Bioenergy, 2006, 30, 863–869.
19. S. B. Yan, X. S. Chen, J. Y. Wu and P. C. Wang, Appl. Microbiol. Biotechnol., 2012, 94, 829–838.
20. J. J. Bozell and G. R. Petersen, Green Chem., 2010, 12, 539–554.
21. J. L. Yu, X. Zhang and T. W. Tan, J. Biotechnol., 2007, 129, 415–420.
22. Z. Li, J. K. Lu, Z. X. Yang, L. Han and T. W. Tan, Biomass Bioenergy, 2012, 39, 53–58.
23. S. F. Ding and T. W. Tan, Process Biochem., 2006, 41, 1451–1454.
24. V. M. Yabannavar and D. I. C. Wang, Biotechnol. Bioeng., 1991, 37, 544–550.
25. S. Savoie, C. P. Champagne, S. Chaisson and P. Audet, J. Appl. Microbiol., 2007, 103, 163–174.
26. L. M. D. Goncalves, M. T. O. Barreto, A. M. R. B. Xavier, M. J. T. Carrondo and J. Klein, Appl. Microbiol. Biotechnol., 1992, 38, 305–311.
27. Z. Shi, P. Wei, X. Zhu, J. Cai, L. Huang and Z. Xu, Enzyme Microb. Technol., 2012, 51, 263–268.
28. J. L. Yu, T. Zhang, J. Zhong, X. Zhang and T. W. Tan, Biotechnol. Adv., 2012, 30, 811–816.

---