Process Development for the Production of (R)-(-)-Mandelic Acid by Recombinant Escherichia coli Cells Harboring Nitrilase from Burkholderia cenocepacia J2315

Article abstract of DOI:10.1021/acs.oprd.5b00269

(R)-(-)-Mandelic acid is an important chiral building block that is widely used in pharmacy and the production of fine chemicals. A more advanced method for obtaining (R)-(-)-mandelic acid is direct hydrolysis of the corresponding racemic mandelonitrile. In order to develop a cost-effective process, a highly efficient enantioselective nitrilase BCJ2315 from Burkholderia cenocepacia J2315 was used for the biotransformation of mandelonitrile to (R)-(-)-mandelic acid. The recombinant Escherichia coli M15/BCJ2315 showed high substrate tolerance and could completely hydrolyze up to 250 mM of mandelonitrile. A fed-batch reaction was performed by periodically or continuously dosing the substrate into the reactor to alleviate substrate inhibition in a monophasic buffer system. Finally, the highest substrate loading (2.9 M) was achieved in the continuous fed batch reaction mode, giving (R)-(-)-mandelic acid at the highest concentration (2.3 M, 350 g/L) with 97.4% ee ever reported. The hydrolysis process was easily scaled up to 2 and 10 L, indicating the potential for the industrial production of optically pure (R)-(-)-mandelic acid.

Full text of DOI:10.1021/acs.oprd.5b00269

Article
pubs.acs.org/OPRD
Process Development for the Production of (R)‑(−)-Mandelic Acid by
Recombinant Escherichia coli Cells Harboring Nitrilase from
Burkholderia cenocepacia J2315
Hualei Wang, Haiyang Fan, Huihui Sun, Li Zhao, and Dongzhi Wei*
State Key Laboratory of Bioreactor Engineering, East China University of Science and Technology, Shanghai 200237, People’s
Republic of China
ABSTRACT: (R)-(−)-Mandelic acid is an important chiral building block that is widely used in pharmacy and the production of
fine chemicals. A more advanced method for obtaining (R)-(−)-mandelic acid is direct hydrolysis of the corresponding racemic
mandelonitrile. In order to develop a cost-effective process, a highly efficient enantioselective nitrilase BCJ2315 from
Burkholderia cenocepacia J2315 was used for the biotransformation of mandelonitrile to (R)-(−)-mandelic acid. The recombinant
Escherichia coli M15/BCJ2315 showed high substrate tolerance and could completely hydrolyze up to 250 mM of mandelonitrile.
A fed-batch reaction was performed by periodically or continuously dosing the substrate into the reactor to alleviate substrate
inhibition in a monophasic buffer system. Finally, the highest substrate loading (2.9 M) was achieved in the continuous fed batch
reaction mode, giving (R)-(−)-mandelic acid at the highest concentration (2.3 M, 350 g/L) with 97.4% ee ever reported. The
hydrolysis process was easily scaled up to 2 and 10 L, indicating the potential for the industrial production of optically pure (R)-
(−)-mandelic acid.
loading and to produce (R)-(−)-mandelic acid at high
concentration by alleviating the inhibition of the highly toxic
mandelonitrile.
INTRODUCTION
■
(R)-(−)-Mandelic acid, which is widely used in pharmacy and
the production of fine chemicals, has attracted more attention
recently.1 It is widely used in the synthesis of antibiotics,2
antiobesity drugs,3 and antitumor agents.4 Its application in the
synthesis of cephalosporins antibiotics has been conducted on
an industrial scale. This acid also acts as an important chiral
resolving agent to resolve racemic alcohols and amines in the
industry.5 In the past decade, many enzymatic approaches have
been developed to produce optically pure (R)-(−)-mandelic
acid.1a Among them, nitrilase-mediated enantioselective hydrol-
ysis of racemic mandelonitrile is a more promising method for
producing optically pure (R)-(−)-mandelic acid, due to the
excellent enantioselectivity, inexpensive starting material
(mandelonitrile), noninvolvement of expensive cofactors, and
100% theoretical product yield (Scheme 1). The nitrilase-
mediated pathway has been successfully applied in industry by
Lonza, BASF, and Mitsubishi Rayon to produce (R)-
(−)-mandelic acid from mandelonitrile in amounts of at least
several tons per year.
Many high enantioselective nitrilases have been discovered
and characterized to give potential excellent biocatalysts for the
production of optically pure (R)-(−)-mandelic acid. Nitrilase
from an uncultured organism,1c,6 Alcaligenes sp.,7 Alcaligenes
faecalis ATCC8750,8 A. faecalis ZJUTB10,9 Alcaligenes sp.
MTCC 10675,10 and Pseudomonas putida11 produced (R)-
(−)-mandelic acid with ee values of 98%, 97%, 100%, 99%,
99%, and 99%, respectively. However, the low solubility in
aqueous environments and the high toxicity of mandelonitrile
significantly constrains the industrial application of this
biocatalytic process. To overcome these limitations, an
aqueous−organic biphasic reaction system,12 enzyme immobi-
lization,13 fed-batch reaction,14 or a combination of these
methods15 has been successfully used to achieve high substrate
In our previous study, a highly efficient enantioselective
nitrilase BCJ2315 toward mandelonitrile was identified and
characterized in detail.16 Using a combination of immobiliza-
tion and the ethyl acetate−water biphasic reaction system, as
high a concentration as 1 M of (R)-(−)-mandelic acid was
achieved with recombinant Escherichia coli M15/BCJ2315,15
demonstrating the potential to produce optically pure (R)-
(−)-mandelic acid in industry. However, the addition of the
organic solvent to the reaction mixture negatively affected the
enantioselectivity of nitrilase, as the ee value further reduced to
90% when ethyl acetate content reached 50%. To develop a
more cost-effective biocatalytic process suitable for use in
industry, we conducted a fed-batch reaction by periodically or
continuously dosing substrate into the reactor to alleviate
substrate inhibition in a monophasic buffer system using the
recombinant E. coli M15/BCJ2315 in this study. High substrate
loading (2.9 M) was achieved, and up to 2.3 M of (R)-
(−)-mandelic acid with 97.4% ee was obtained because of the
dilution effect. Scale-up of the developed process was carried
out to verify its potential in industry application.
RESULTS AND DISCUSSION
■
Cultivation of the Recombinant E. coli M15/BCJ2315.
The use of easily available and low-cost whole cell biocatalysts
is a much more economical alternative compared to the use of
cell-free extracts, purified enzyme, immobilized cells, or
enzyme. In order to obtain a low-cost and efficient biocatalyst,
we cultured recombinant E. coli M15/BCJ2315 to produce
Received: August 20, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.oprd.5b00269
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 1. Enantioselective Hydrolysis of Racemic Mandelonitrile to Produce Optically Pure (R)-(−)-Mandelic Acid through
Nitrilase
whole cell biocatalysts with high density and high expression
levels in a stirred bioreactor (7 L). The cultures were induced
with isopropyl-β-D-thiogalactopyranoside (IPTG) (final con-
centration of 0.1 mM) when the OD600 reached 20 and the cells
were harvested when the specific activity reached a maximum of
3418 U/g DCW at an OD600 of 70 after over 24 h of
fermentation. A biomass of 22.8 g DCW/L and a total nitrilase
activity of 77 930 U/L were finally obtained, which are higher
than those of the A. faecalis ATCC8750,8 A. faecalis MTCC
10757,17 and A. faecalis ZJUTB10.18 This culture process offers
an easily available, highly active, and low-cost catalyst that could
be used to produce (R)-(−)-mandelic acid in industry.
Effect of Mandelonitrile Concentrations on Nitrilase
Activity and Enantioselectivity. High substrate loadings are
generally highly desirable for practical applications in
biocatalytic processes. Mandelonitrile is well-known to be
highly detrimental to nitrilase, leading to a low substrate
concentration for biotransformation.14 To study the effect of
mandelonitrile concentration on activity and enantioselectivity
of the recombinant E. coli M15/BCJ2315, we performed the
hydrolysis under different concentrations of mandelonitrile
(30−300 mM) with 10% methanol as a cosolvent to increase
the solubility of the substrate in the aqueous reaction mixture.
Nitrilase BCJ2315 could tolerate as high as 250 mM
mandelonitrile (Table 1). Only 75% conversion was achieved
hydrolyzing 500 mM of mandelonitrile with three fungal
nitrilases, including NitAn, NitAb, and NitAc.19 The optical
purity of mandelic aicd correlated inversely with the
mandelonitrile concentration for nitrilase BCJ2315. The ee
value decreased with increasing substrate concentration. The
highest ee value (99.1%) was observed for the lowest
concentration of substrate (30 mM) in the reaction medium
(Table 1). Therefore, high concentration and high enantiose-
lectivity were inversely correlated. To solve this contradiction
problem, the reaction mode must be well-designed.
Fed-Batch Production of (R)-(−)-Mandelic Acid (100
mL Scale). To implement a cost-effective process, a high
product concentration (gproduct/L) with a simple downstream
process is required to meet the process economic requirements
of the industry.20 Based on the negative effect of high
concentration of mandelonitrile on BCJ2315 activity and
enantioselectivity, a fed-batch mode was designed for
BCJ2315 that included periodically or continuously dosing of
substrate into the reactor to alleviate substrate inhibition.
Controlling mandelonitrile concentration below the inhibitory
concentration in the biocatalytic medium is essential for
maintaining enzyme activity. Additionally, a relatively low
substrate concentration is beneficial for improving the
enantiomeric purity of the products.21
In the periodical fed-batch process, 100 mM of mandeloni-
trile was fed into the reaction system when the substrate was
completely hydrolyzed at each batch. Finally, the reaction was
performed for 11 batches, and as high as 1.1 M mandelonitrile
were completely hydrolyzed in 23 h (see Figure 1). However,
during the 12th feeding of mandelonitrile, only 54 mM of
mandelonitrile was hydrolyzed after a long incubation time of 6
h. Finally, as high as 0.96 M of (R)-(−)-mandelic acid was
obtained because of the increased reaction volume caused by
serial mandelonitrile feeding for 11 batches, affording a
concentration of 146 g/L product. The optical purity of the
(R)-(−)-mandelic acid was determined to be 97.6% ee by chiral
HPLC analysis. However, when 200 mM of mandelonitrile was
fed at once, the substrate could not be completely hydrolyzed
even with an extended reaction time at the fourth feeding. This
may be because of enzyme inactivation over the period of
hydrolysis.
In the continuous fed-batch process, mandelonitrile was kept
at optimal concentrations by adjusting the feeding rate, which
could avoid the inactivation of the enzyme at high substrate
concentration conditions (see Figure 2). At the first feeding
stage, the mandelonitrile was fed at a rate of 24 g/L·h, which
could maintain relatively high hydrolysis rate. This process
lasted for 11 h of reaction time, and up to 1.67 M (R)-
Table 1. Effect of Mandelonitrile Concentrations on (R)-
(−)-Mandelic Acid Production
product
concentration
(mM)
substrate
concentration (mM)
reaction time
(min)
ee value of
acid (%)
30
15
30
99.1
98.8
98.1
97.5
96.7
95.3
92.1
50
25
50
100
150
200
250
300
45
100
150
200
250
225
70
100
180
300
for the hydrolysis of 300 mM mandelonitrile after 5 h of
biotransformation. The tolerance of BCJ2315 toward man-
delonitrile was slightly higher than that of nitrilase from
Alcaligenes sp., which could hydrolyze as high as 200 mM
mandelonitrile.14 Both nitrilases were unable to transform
mandelonitrile at concentrations of 300 mM or higher.
Compared with bacterial nitrilases, fungal nitrilases performed
better in the hydrolysis of high concentrations of mandeloni-
trile. High conversions (75−93%) were achieved when
B
DOI: 10.1021/acs.oprd.5b00269
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Figure 1. Time course of the hydrolysis reaction in periodical fed-
batch mode (100 mL scale).
Figure 3. Time course of the hydrolysis reaction in continuous fed-
batch mode (10 L scale).
25
25
follows: [α]D = −153.1 (c 1.0, H2O) (literature: [α]D
=
−155 (c 1.0, H2O)8); H NMR (400 MHz, DMSO): δ12.62
(brs, 1H), 7.45−7.39 (m, 2H), 7.36−7.31 (m, 2H), 7.31−7.26
(m, 1H), 5.84 (brs, 1H), and 5.01 (s, 1H) ppm. The success of
the scale-up made this process a preferable route for large-scale
production of (R)-(−)-mandelic acid in industry.
1
́
Vesela et al. compared the yield (%), product concentration
(g/L), volume productivity (g/L·d), and ee value (%) of the
product and catalyst productivity (g/g·dcw) of nitrilase from
different sources in various biocatalytic process types (batch,
fed-batch, and biphasic system).19 The potential of nitrilases
from Alcaligenes sp.,14,22 A. faecalis,23 B. cenocepacia J2315,15 and
Aspergillus niger19 as biocatalysts in the form of recombinant
whole cells or immobilized recombinant cells were compared in
detail. All compared nitrilases showed an acceptable enantio-
selectivity of ee > 94.5%. The highest volumetric productivity
and catalyst productivity were observed to be 982 g/L·d and
156 g/g·dcw, respectively, observed from the immobilized
nitrilase from B. cenocepacia J2315 after 6 rounds of recycling in
batch hydrolysis of 500 mM mandelonitrile in our previous
study.15 Additionally, the highest product concentration was
observed to be 150 g/L by completely hydrolyzing 1 M
mandelonitrile in a biphasic system using the immobilized
recombinant E. coli M15/BCJ2315 in the same study.15
Recently, a higher product concentration of (R)-(−)-mandelic
acid was reported by Zhang et al., using immobilized
recombinant E. coli cells expressing Alcaligenes sp. nitrilaseas
biocatalyst.24 In a repeated fed batch hydrolysis, a total of 1.6 M
mandelonitrile was completely hydrolyzed after 15 rounds of
substrate dosing and 7 rounds of biocatalyst recovery, affording
a product concentration of 243 g/L. However, these above-
mentioned higher parameters were only obtained when the
catalyst was repeatedly recycled. The immobilization of nitrilase
and repeated recovery of the catalysts are complex and time-
consuming, limiting industrial application. Compared with
these processes, our continuous fed-batch process using
recombinant E. coli M15/BCJ2315 is simple and cost-effective
and shows high-volume productivity (350 g/L·d), a high ee
value (97.4%) of the product, the highest concentration of
product (350 g/L), and high catalyst productivity (115 g/g·
dcw, the biocatalysts loading was 3.85 g dcw/L, equal to 10 g/L
of wet cells)), demonstrating the potential of this method for
industrial production of (R)-(−)-mandelic acid.
Figure 2. Time course of the hydrolysis reaction in continuous fed-
batch mode (100 mL scale).
(−)-mandelic acid was obtained by that time. At the second
feeding stage, the hydrolysis rate decreased greatly, probably
due to the leakage of the E. coli cells and inactivation of the
enzyme over the period of hydrolysis. Thus, the mandelonitrile
feeding rate was adjusted to 12 g/L·h and this process lasted for
another 11 h of reaction time. At the final stage, the substrate
feeding was stopped and an additional 4 h of hydrolysis was
performed to guarantee a complete hydrolysis of all the added
substrate. Finally, a total of 2.9 M of mandelonitrile was
completely hydrolyzed to produce as high as 2.3 M of (R)-
(−)-mandelic acid because of the increased reaction volume
caused by continuous mandelonitrile feeding and pH control,
affording 350 g/L product with 97.4% ee.
Scale-up of Continuous Fed-Batch Reaction to 2 and
10 L. To explore the potential of the continuous fed-batch
method for industrial large-scale production of (R)-(−)-man-
delic acid, this hydrolysis process was scaled up to 2 and 10 L.
Both processes were very stable during scale-up, and
approximately 100% of the added mandelonitrile was
completely hydrolyzed to form (R)-(−)-mandelic acid within
24 h. The time course of the hydrolysis on the 10-L scale was
shown in Figure 3. The product was finally obtained as white
powdery solid in 93% isolate yield and 99.5% ee after
recrystallization in toluene. The product was characterized as
C
DOI: 10.1021/acs.oprd.5b00269
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
continuously supply the substrate, mandelonitrile was fed into
the system after 10 min of hydrolysis at a feeding rate of 24 g/
L·h. Real-time conversion rate was monitored by HPLC, and
the feeding rate was adjusted to keep the conversion above
80%. To achieve a complete hydrolysis of the added
mandelonitrile, the reaction lasted for an additional 4 h after
the feeding was stopped.
Scale-up of Continuous Fed-Batch Reaction to 2 and
10 L. The continuous fed-batch process was scaled up to 2 and
10 L at 30 °C under mechanical agination of 250 rpm in 5 L
and 10 L double jacket glass reactors (ShenSheng, Shanghai,
China), respectively. The reaction mixture contained 100 mM
of mandelonitrile, 100 mM of phosphate sodium buffer (pH
8.0), 10 g/L of E. coli M15/BCJ2315 (wet cells), and 10%
methanol. The control of the reaction temperature and pH,
agitation of the reaction mixture, and substrate feeding were
performed as described previously.25 The substrate feeding was
conducted as it is on the 100 mL scale.
Purification of the Product. After all of the mandelonitrile
in the reaction mixture was hydrolyzed completely, the
biocatalysts were removed through centrifugation. Methanol,
which was used as a latent solvent in the reaction, was then
removed from the supernatant using a rotary evaporation
apparatus. The supernatant was then acidified to pH 1.0 with
HCl 11% (v/v), and then the mandelic acid of approximately
54% total mass was separated by crystallization. After crystal
collection by suction filtration, the remaining solution was
condensed to 30% volume by rotary evaporation. The mandelic
acid of 43% total mass was separated by crystallization when
cooled to 25 °C. All crystals were collected and dissolved in
toluene for recrystallization. The refined mandelic acid was
detected by HPLC and alkalimetric titration.
Analytical Methods. TLC was used for monitoring the
mandelonitrile hydrolysis in real time. The developing solvents
were n-hexane/ethyl acetate/acetic acid (5:1:0.1). The
retention factor value (Rf value) was 0.05 (mandelic acid),
0.7 (mandelonitrile), and 0.4 (benzaldehyde) detected at 254
nm wavelength.
CONCLUSIONS
■
In summary, a continuous fed-batch process was developed to
completely hydrolyze as high as 2.9 M mandelonitrile using the
E. coli whole-cell system harboring an efficient nitrilase as
biocatalyst, giving (R)-(−)-mandelic acid at the highest
concentration (350 g/L) with 97.4% ee. The scale-up of this
hydrolysis process was easy and stable, demonstrating great
potential on large-scale production of (R)-(−)-mandelic acid in
industry.
EXPERIMENTAL SECTION
■
Materials. Mandelonitrile was purchased from Guangde
Chemicals Co., Ltd. (Anhui, China). The (R)-(−)-mandelic
acid and (S)-(+)-mandelic acid were purchased from Sigma-
Aldrich (Milwaukee, USA). Thin-layer chromatography (TLC)
silica gel glass sheets used for TLC analysis were purchased
from Yantai Jiangyou Co., Ltd. (Shandong, China). The
recombinant E. coli M15/BCJ2315 harboring the nitrilase
BCJ2315 (GenBank accession number: WP_012492804) from
B. cenocepacia J2315 was reported previously.16
Cultivation of the Recombinant E. coli M15/BCJ2315.
The cultivation of recombinant E. coli M15/BCJ2315 was
performed in a 7-L fermenter (BIOTECH, Shanghai, China) as
described previously,15 with modified fermentation medium
(12 g/L tryptone, 16 g/L yeast extract, 1.26 g/L glycerol, 24.75
g/L K2HPO4, 3.47 g/L KH2PO4, and 1 g/L MgSO4) and
supplementary medium (40 g/L tryptone, 30 g/L yeast extract,
50% glycerol). Supplementary medium was fed at a rate of 1
mL/min after 2 h of fermentation and induction of the cells was
initiated by adding IPTG at a final concentration of 0.1 mM to
the fermenter when the culture density reached an OD600 of 20.
Effect of Mandelonitrile Concentrations on Nitrilase
Activity and Enantioselectivity. To study the effect of
substrate concentration on activity and enantioselectivity of the
nitrilase, the hydrolysis reaction was performed with varied
substrate concentrations (30−300 mM) in a 5 mL reaction
system containing 100 mM of sodium phosphate buffer (pH
8.0), 50 mg of E. coli M15/BCJ2315 wet cells, and 10% of
methanol (v/v). Samples (100 μL) were periodically with-
drawn, and the conversion of the reaction and optical purity of
the product were determined through high-performance liquid
chromatography (HPLC) analysis.
Fed-Batch Production of (R)-(−)-Mandelic Acid (100
mL Scale). The fed-batch process was performed by
periodically or continuously feeding the substrate into the
reaction system. The hydrolysis was performed in a 250 mL
three-necked round-bottom flask on 100 mL scale. The control
of temperature, pH, and agitation of the reaction mixture were
performed as described previously.25 The temperature of the
reaction was set to 30 °C. The pH of the reaction mixture was
maintained at 7.5−8.0 by adding sodium hydroxide (6 M) for
pH adjustment. The agitation speed was set to 250 rpm.
The reaction mixture containing 90 mL of sodium phosphate
buffer (100 mM, pH 8.0) and 1 g of the E. coli M15/BCJ2315
wet cells was first incubated at 30 °C for 30 min. Next, 10 mL
of mandelonitrile (solubilized in methanol, 1 M) was added to
initiate the hydrolysis. Samples (100 μL) were periodically
withdrawn to monitor the reaction process by TLC and HPLC
analysis.
HPLC was used to quantitatively analyze the decrease of
mandelontirle, formation, and optical purity of the (R)-
(−)-mandelic acid in the reaction mixture as described
previously.15
AUTHOR INFORMATION
Corresponding Author
*E-mail: dzhwei@ecust.edu.cn. Tel.: +86-21-64252078. Fax:
+86-21-64250068.
■
Funding
This work was supported by the National Natural Science
Foundation of China (No. 21406068/B060804), China
Postdoctoral Science Foundation funded project (No.
2014M560308), and the National Basic Research Program of
China (No. 2012CB721103).
Notes
The authors declare no competing financial interest.
REFERENCES
■
(1) (a) Groger, H. Adv. Synth. Catal. 2001, 343, 547−558. (b) Kaul,
P.; Banerjee, A.; Mayilraj, S.; Banerjee, U. C. Tetrahedron: Asymmetry
2004, 15, 207−211. (c) DeSantis, G.; Zhu, Z.; Greenberg, W. A.;
Wong, K.; Chaplin, J.; Hanson, S. R.; Farwell, B.; Nicholson, L. W.;
Rand, C. L.; Weiner, D. P.; Robertson, D. E.; Burk, M. J. J. Am. Chem.
Soc. 2002, 124, 9024−9025.
To periodically supply the substrate, mandelonitrile (100
mM) was fed into the system when it was completely
hydrolyzed as revealed by TLC and HPLC analysis. To
D
DOI: 10.1021/acs.oprd.5b00269
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
(2) Terreni, M.; Pagani, G.; Ubiali, D.; Fernandez-Lafuente, R.;
Mateo, C.; Guisan, J. M. Bioorg. Med. Chem. Lett. 2001, 11, 2429−
2432.
(3) Mills, J.; Schmiegel, K. K; Shaw, W. N. U.S. Patent 4391826,
1983.
(4) Surivet, J.-P.; Vatele, J.-M. Tetrahedron 1999, 55, 13011−13028.
(5) Yadav, G. D.; Sajgure, A. D.; Dhoot, S. B. J. Chem. Technol.
Biotechnol. 2008, 83, 1145−1153.
(6) Robertson, D. E.; Chaplin, J. A.; DeSantis, G.; Podar, M.;
Madden, M.; Chi, E.; Richardson, T.; Milan, A.; Miller, M.; Weiner, D.
P.; Wong, K.; McQuaid, J.; Farwell, B.; Preston, L. A.; Tan, X. Q.;
Snead, M. A.; Keller, M.; Mathur, E.; Kretz, P. L.; Burk, M. J.; Short, J.
M. Appl. Environ. Microbiol. 2004, 70, 2429−2436.
(7) Zhang, Z. J.; Xu, J. H.; He, Y. C.; Ouyang, L. M.; Liu, Y. Y.
Bioprocess Biosyst. Eng. 2011, 34, 315−322.
(8) Yamamoto, K.; Oishi, K.; Fujimatsu, I.; Komatsu, K. Appl.
Environ. Microbiol. 1991, 57, 3028−3032.
(9) Liu, Z. Q.; Dong, L. Z.; Cheng, F.; Xue, Y. P.; Wang, Y. S.; Ding,
J. N.; Zheng, Y. G.; Shen, Y. C. J. Agric. Food Chem. 2011, 59, 11560−
11570.
(10) Bhatia, S. K.; Mehta, P. K.; Bhatia, R. K.; Bhalla, T. C. Biotechnol.
Appl. Biochem. 2014, 61, 459−465.
(11) Banerjee, A.; Dubey, S.; Kaul, P.; Barse, B.; Piotrowski, M.;
Banerjee, U. C. Mol. Biotechnol. 2009, 41, 35−41.
(12) Zhang, Z. J.; Pan, J. A.; Liu, J. F.; Xu, J. H.; He, Y. C.; Liu, Y. Y. J.
Biotechnol. 2011, 152, 24−29.
(13) (a) Banerjee, A.; Kaul, P.; Banerjee, U. C. Appl. Microbiol.
Biotechnol. 2006, 72, 77−87. (b) Kaul, P.; Banerjee, A.; Banerjee, U. C.
Biomacromolecules 2006, 7, 1536−1541. (c) Pawar, S. V.; Meena, V. S.;
Kaushik, S.; Kamble, A.; Kumar, S.; Chisti, Y.; Banerjee, U. C. 3
Biotechnol. 2012, 2, 319−326.
(14) Zhang, Z. J.; Xu, J. H.; He, Y. C.; Ouyang, L. M.; Liu, Y. Y.;
Imanaka, T. Process Biochem. 2010, 45, 887−891.
(15) Ni, K.; Wang, H.; Zhao, L.; Zhang, M.; Zhang, S.; Ren, Y.; Wei,
D. J. Biotechnol. 2013, 167, 433−440.
(16) Wang, H. L.; Sun, H. H.; Wei, D. Z. BMC Biotech. 2013, 13, 14.
(17) Nageshwar, Y. V.; Sheelu, G.; Shambhu, R. R.; Muluka, H.;
Mehdi, N.; Malik, M. S.; Kamal, A. Bioprocess Biosyst. Eng. 2011, 34,
515−523.
(18) Xue, Y. P.; Xu, M.; Chen, H. S.; Liu, Z. Q.; Wang, Y. J.; Zheng,
Y. G. Org. Process Res. Dev. 2013, 17, 213−220.
(19) Vesela, A. B.; Krenkova, A.; Martinkova, L. Mol. Biotechnol.
2015, 57, 466−474.
(20) Lima-Ramos, J.; Neto, W.; Woodley, J. M. Top. Catal. 2014, 57,
301−320.
(21) Kim, P. Y.; Pollard, D. J.; Woodley, J. M. Biotechnol. Prog. 2007,
23, 74−82.
(22) Zhang, Z. J.; Pan, J.; Liu, J. F.; Xu, J. H.; He, Y. C.; Liu, Y. Y. J.
Biotechnol. 2011, 152, 24−29.
(23) Liu, Z. Q.; Zhang, X. H.; Xue, Y. P.; Xu, M.; Zheng, Y. G. J.
Agric. Food Chem. 2014, 62, 4685−4694.
(24) Zhang, Z. J.; Pan, J.; Li, C. X.; Yu, H. L.; Zheng, G. W.; Ju, X.;
Xu, J. H. Bioprocess Biosyst. Eng. 2014, 37, 1241−1248.
(25) Wang, H. L.; Sun, H. H.; Gao, W. Y.; Wei, D. Z. Org. Process Res.
Dev. 2014, 18, 767−773.
E
DOI: 10.1021/acs.oprd.5b00269
Org. Process Res. Dev. XXXX, XXX, XXX−XXX