Construction and expression of a polycistronic plasmid encoding N-acetylglucosamine 2-epimerase and N-acetylneuraminic acid lyase simultaneously for production of N-acetylneuraminic acid Wujin Sun, Wenyan Ji, Nan Li, Peng Tong, Jian Cheng, Ying He, Yong Chen, Xiaochun Chen, Jinglan Wu, Pingkai Ouyang, Jingjing Xie ⇑ , Hanjie Ying ⇑ State Key Laboratory of Materials-Oriented Chemical Engineering, National Engineering Technique Research Center for Biotechnology, College of Biotechnology and Pharmaceutical Engineering, Nanjing University of Technology, Nanjing, China highlights " Encoding enzymes for two reactions in one plasmid simplified fermentation process. " Activities of two types of N-acetylglucosamine 2-epimerase were compared. " Intact E. coli cells avoid adding ATP to activate N-acetylglucosamine 2-epimerase. " Lowering N-acetylglucosamine 2-epimerase expression level improved its solubility. " High production of 61.3 g/l N-acetylneuraminic acid. article info Article history: Received 25 June 2012 Received in revised form 4 December 2012 Accepted 7 December 2012 Available online 19 December 2012 Keywords: Biocatalysis N-acetylneuraminic acid Inclusion bodies N-acetylneuraminic acid lyase N-acetylglucosamine 2-epimerase abstract Synthesis of N-acetylneuraminic acid (Neu5Ac) from N-acetylglucosamine (GlcNAc) and pyruvate was carried out by constructing and expressing a polycistronic plasmid encoding an N-acetylglucosamine 2-epimerase (AGE) gene and an N-acetylneuraminic acid lyase (Nal) gene simultaneously. Nal from Escherichia coli K12 and AGEs from Synechocystis sp. PCC 6803 (snAGE) and Anabaena sp. CH1 (anAGE) were used. And four polycistronic plasmids were constructed in which the positions of AGE gene differed with respect to Nal gene. Among these plasmids, pET-28a-Nal-anAGE with anAGE gene located next to Nal gene caused the production of the highest amount of Neu5Ac, generating 61.3 g/L in 60 h by whole-cell catalysis without the addition of ATP as AGE activator. And pET-28a-Nal-anAGE lowered anAGE’s expression level, allowing it to fold properly. Thus, an inclusion-body-free E. coli strain capable of producing Neu5Ac by whole-cell catalysis with high yield and low cost was constructed in the present study. Ó 2012 Elsevier Ltd. All rights reserved. 1. Introduction Mammalian cells are covered with sugar chains which are terminated by a family of 9-carbon amino sugars called sialic acids. These sugar derivatives are often part of the recognition sites where pathogens attach (Varki and Varki, 2007). Sialic acids are involved in the modulation of various biological processes, such as virus invasion, cell differentiation, fertilization, cell adhesion, inflammation, and tumorigenesis (Hu et al., 2010). More than 40 types of sialic acids have been discovered in nature, and Neu5Ac is the most ubiquitous and the biosynthetic precursor for all other sialic acids (Maru et al., 2002; Ogura, 2011). Neu5Ac is a potential raw material in the synthesis of zanamivir, which prevents both influenza type A and B infections (Tao et al., 2010). Neu5Ac is also an important additive in dairy products, as it is able to strengthen the immunity of infants (Oriquat et al., 2011). In addition, Neu5Ac is of great diagnostic value as an important indicator for many diseases (Gopaul and Crook, 2006). Traditionally, Neu5Ac has been prepared by extraction from natural sources (such as milk or eggs), or by hydrolyzing capsular 0960-8524/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biortech.2012.12.042 Abbreviations: Nal, N-acetylneuraminic acid lyase; AGE, N-acetylglucosamine 2-epimerase; snAGE, N-acetylglucosamine 2-epimerase from Synechocystis sp. PCC 6803; anAGE, N-acetylglucosamine 2-epimerase from Anabaena sp. CH1; GlcNAc, N-acetylglucosamine; ManNAc, N-acetyl-D-mannosamine; Neu5Ac, N-acetylneuraminic acid. ⇑ Corresponding author. Tel.: +86 25 86990001; fax: +86 25 58133398. E-mail addresses: [email protected](J. Xie), [email protected], [email protected](H. Ying). Bioresource Technology 130 (2013) 23–29 Contents lists available at SciVerse ScienceDirect Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Transcript
Bioresource Technology 130 (2013) 23–29
Contents lists available at SciVerse ScienceDirect
Construction and expression of a polycistronic plasmid encodingN-acetylglucosamine 2-epimerase and N-acetylneuraminic acid lyasesimultaneously for production of N-acetylneuraminic acid
0960-8524/$ - see front matter � 2012 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.biortech.2012.12.042
Wujin Sun, Wenyan Ji, Nan Li, Peng Tong, Jian Cheng, Ying He, Yong Chen, Xiaochun Chen, Jinglan Wu,Pingkai Ouyang, Jingjing Xie ⇑, Hanjie Ying ⇑State Key Laboratory of Materials-Oriented Chemical Engineering, National Engineering Technique Research Center for Biotechnology, College of Biotechnology andPharmaceutical Engineering, Nanjing University of Technology, Nanjing, China
h i g h l i g h t s
" Encoding enzymes for two reactions in one plasmid simplified fermentation process." Activities of two types of N-acetylglucosamine 2-epimerase were compared." Intact E. coli cells avoid adding ATP to activate N-acetylglucosamine 2-epimerase." Lowering N-acetylglucosamine 2-epimerase expression level improved its solubility." High production of 61.3 g/l N-acetylneuraminic acid.
a r t i c l e i n f o
Article history:Received 25 June 2012Received in revised form 4 December 2012Accepted 7 December 2012Available online 19 December 2012
Synthesis of N-acetylneuraminic acid (Neu5Ac) from N-acetylglucosamine (GlcNAc) and pyruvate wascarried out by constructing and expressing a polycistronic plasmid encoding an N-acetylglucosamine2-epimerase (AGE) gene and an N-acetylneuraminic acid lyase (Nal) gene simultaneously. Nal fromEscherichia coli K12 and AGEs from Synechocystis sp. PCC 6803 (snAGE) and Anabaena sp. CH1 (anAGE)were used. And four polycistronic plasmids were constructed in which the positions of AGE gene differedwith respect to Nal gene. Among these plasmids, pET-28a-Nal-anAGE with anAGE gene located next toNal gene caused the production of the highest amount of Neu5Ac, generating 61.3 g/L in 60 h bywhole-cell catalysis without the addition of ATP as AGE activator. And pET-28a-Nal-anAGE loweredanAGE’s expression level, allowing it to fold properly. Thus, an inclusion-body-free E. coli strain capableof producing Neu5Ac by whole-cell catalysis with high yield and low cost was constructed in the presentstudy.
� 2012 Elsevier Ltd. All rights reserved.
1. Introduction
Mammalian cells are covered with sugar chains which areterminated by a family of 9-carbon amino sugars called sialic acids.These sugar derivatives are often part of the recognition siteswhere pathogens attach (Varki and Varki, 2007). Sialic acids are
involved in the modulation of various biological processes, suchas virus invasion, cell differentiation, fertilization, cell adhesion,inflammation, and tumorigenesis (Hu et al., 2010). More than40 types of sialic acids have been discovered in nature, and Neu5Acis the most ubiquitous and the biosynthetic precursor for all othersialic acids (Maru et al., 2002; Ogura, 2011). Neu5Ac is a potentialraw material in the synthesis of zanamivir, which prevents bothinfluenza type A and B infections (Tao et al., 2010). Neu5Ac is alsoan important additive in dairy products, as it is able to strengthenthe immunity of infants (Oriquat et al., 2011). In addition, Neu5Acis of great diagnostic value as an important indicator for manydiseases (Gopaul and Crook, 2006).
Traditionally, Neu5Ac has been prepared by extraction fromnatural sources (such as milk or eggs), or by hydrolyzing capsular
Fig. 2. Diagram of the two-step reaction of this study. GlcNAc was transportedinside the cells through the cell membrane, and it was converted into ManNAc byAGE. The resulting ManNAc was condensed with pyruvate by Nal to produceNeu5Ac.
24 W. Sun et al. / Bioresource Technology 130 (2013) 23–29
polysaccharides from Escherichia coli (Hu et al., 2010; Tabata et al.,2002). Current enzymatic production methods are based on theE. coli K1 sialic acid synthesis pathway (Ferrero and Aparicio,2010; Plumbridge and Vimr, 1999) (Fig. 1). N-acetylneuraminicacid lyase (Nal, NanA, EC 4.1.3.3) from E. coli (Hu et al., 2010; Mah-moudian et al., 1997; Maru et al., 1998; Wang et al., 2009; Zhanget al., 2010) and many other organisms (Krüger et al., 2001; Liet al., 2008; Nahalka et al., 2008; Sanchez-Carron et al., 2011) areutilized to condense N-acetyl-D-mannosamine (ManNAc) andpyruvate into Neu5Ac through a reversible reaction. N-acetylneu-raminic acid synthases (NeuB, EC 4.1.3.19) from E. coli (Ishikawaand Koizumi, 2010; Tabata et al., 2002) and other sources (Haoet al., 2005) are used to condense ManNAc and phosphoenolpyr-uvate (PEP) into Neu5Ac by an irreversible reaction.
However, ManNAc is too expensive to be applied on an indus-trial scale (Hu et al., 2010). It is possible to synthesize E. coli K1ManNAc from UDP-GlcNAc by UDP-GlcNAc 2-epimerase (NeuC,EC 5.1.3.14); or via a 3-step reaction with GlcNAc as a substratethat involves the phosphotransferase system (PTS), ManNAc-6Pepimerase (NanE) and a proposed phosphatase (Ferrero and Apari-cio, 2010; Plumbridge and Vimr, 1999). However, ManNAc couldbe utilized as an intermediate for Neu5Ac synthesis. N-acetylgluco-samine 2-epimerase (AGE, EC 5.1.3.8), identified in mammals as arenin binding protein, catalyzes a single-step reaction to produceManNAc from inexpensive GlcNAc in a reversible process (Maruet al., 1996). However, AGE is not prevalent in nature and no func-tional AGE has been reported in E. coli. The hypothetical proteinYihS from E. coli, which has significant structural similarity toAGE, shows no AGE activity (Itoh et al., 2008). AGEs have been dis-covered in mammals, such as humans (Lee et al., 2004), pigs (Itohet al., 2000) and rats (Van Rinsum et al., 1983); in cyanobacteria,such as Synechocystis sp. PCC 6803 (Tabata et al., 2002) and Ana-baena sp. CH1 (Lee et al., 2007a); and recently in Bacteroides ovatusATCC 8483 (Sola-Carvajal et al., 2012).
A simple and efficient two-step enzymatic catalytic strategyinvolving AGE and Nal has been adopted to produce Neu5Ac(Fig. 2) (Hu et al., 2010; Maru et al., 1998; Wang et al., 2009; Zhanget al., 2010). While Nal from E. coli is soluble when cloned and
Outside Inside
GlcNAcPTS(nagE)
GlcNAc-6-P
ManNAc-6-P
NanE
ManNAcPTS(manXYZ)
ManNAc
NanK
ATP
ADP
Phosphatase ?
Fructose-6-P
UDP-GlcNAc
NeuC
Neu5AcNeu5Ac
NanT
Cell Membrane
NanA NeuB
pyr PEP
Fig. 1. Sialic acid synthesis pathway in E. coli K1 (Ferrero and Aparicio, 2010;Plumbridge and Vimr, 1999). Pyr: pyruvate, PEP: phosphoenolpyruvate, NanT, E. coliNeu5Ac transporter, NeuC, E. coli UDP-GlcNAc 2-epimerase, NeuB, E. coli Neu5Acsynthase, NanA, Neu5Ac aldolase, NanK, ManNAc kinase, NanE, ManNAc-6Pepimerase, PTS, phosphotransferase system.
over-expressed in E. coli, AGE tends to form inclusion bodies whenexpressed heterogeneously (Chien et al., 2007; Lee et al., 2007a).When recombinant plasmids inducible at 42 �C were constructedto produce Neu5Ac (Zhang et al., 2010), the increased temperaturenegatively impacted AGE expression (Hu et al., 2010). Therefore,chemically induced expression would be preferable. Most studieswere designed to carry out the two-step reaction in two separatecells, which requires laborious fermentation of two distinct strains(Hu et al., 2010; Maru et al., 1998; Wang et al., 2009; Zhang et al.,2010). Over-expressing these two enzymes in one single straincould simplify the fermentation process as well as reduce produc-tion costs. In addition, AGE is dependent on adenosine triphos-phate (ATP) as an activator (Takahashi et al., 2001). Adding ATPas a cofactor into a cell-free system significantly increases produc-tion costs, while intact cells contain sufficient ATP to activate thereaction (Ishikawa and Koizumi, 2010).
In the present study, AGEs from Synechocystis sp. PCC 6803 andAnabaena sp. CH1 were used to convert GlcNAc into ManNAc. Nalfrom E. coli K12 was chosen to condense ManNAc and pyruvateinto Neu5Ac because pyruvate is economically more competitivethan PEP. Four chemically-induced polycistronic plasmids co-expressing both genes were introduced into E. coli Rosetta (DE3)pLysS to produce Neu5Ac from GlcNAc. The different positions ofthese genes on the expression plasmid lead to different expressionlevels and catalysis efficiency.
2. Methods
2.1. Chemicals, enzymes and strains
Restriction enzymes, T4 DNA ligase, Taq DNA polymerase, DNAmarkers and protein molecular weight markers were obtainedfrom Takara Biotechnology Co. Ltd (Dalian, China). E. coli DH5aand E. coli Rosetta (DE3) pLysS (Merck, Germany) were used ascloning and expression hosts, respectively. Vectors pMD18-T sim-ple and pET-28a (+) were used as cloning and expression vectors.GlcNAc, ManNAc and Neu5AC were purchased from Sigma Aldrich(Shanghai, China). Oligonucleotide primers were synthesized byGenescript (Nanjing, China). Whole-gene synthesis was done byGeneray (Shanghai, China). All other chemicals were of reagentgrade and obtained from commercial sources.
2.2. Construction of polycistronic plasmids co-expressing two genes
The temperature-inducible vectors, pBV220 containing AGEfrom Synechocystis sp. PCC 6803 (pBVS) and Nal from E. coli K12(pBVN), were gifts from Dr. P. Xu (Zhang et al., 2010). The AGE genefrom Anabaena sp. CH1 was optimized according to its nucelotidesequence (Genbank accession number DQ661858 (Lee et al.,2007a)) and synthesized chemically by Generay. The synthesizedgene was ligated into the EcoRI/NotI restriction sites of pPIC9K(pPIC9k-anAGE).
Fig. 3. Maps of constructed polycistronic plasmids. One Nal gene and two types of AGE genes (snAGE and anAGE) were used to construct the polycistronic plasmids. (a) Naland snAGE were constructed in the same plasmid with different gene positions relative to the promoter. (b) Nal and anAGE were constructed in the same plasmid withdifferent gene positions relative to the promoter.
W. Sun et al. / Bioresource Technology 130 (2013) 23–29 25
The snAGE gene was amplified from pBVS using primers snAGE-S/snAGE-A (Table 1). The Nal gene was amplified from pBVN usingprimers Nal-S/Nal-A (Table 1). The PCR products were purified andligated into the pMD18-T simple vector. The resulting plasmids,pMD18-T-snAGE and pMD18-T-Nal, were sequenced. Afterdigestion with EcoRI/NotI and NdeI/XhoI the DNA fragments wereinserted into the corresponding restriction sites of pET-28a toconstruct pET-28a-snAGE and pET-28a-Nal. Plasmid pPIC9 K-anAGE was amplified and digested with EcoRI/NotI, and theresulting 1-kb fragment was ligated into pET-28a (+) to obtainpET-28a-anAGE.
The ribosomal binding site (RBS, AAGGAG) and spacer (se-quence between RBS and the starting codon, ATATACC) from the
pET-28a (+) vector were used to link the three genes in four com-binations (Fig. 3). The primers, SD-Nal-S/SD-Nal-A, SD-snAGE-S/SD-snAGE-A, and SD-anAGE-S/SD-anAGE-A (Table 1), were usedto amplify the Nal, the snAGE and the anAGE fragments with theRBS and the spacer sequences added onto their 50 ends. The PCRproducts were ligated into the pMD18-T simple vector to be se-quenced. Plasmids, pMD18-T-SD-Nal, pMD18-T-SD-snAGE andpMD18-T-SD-anAGE, with the correct sequence were digestedwith XhoI and ligated into pET-28a-snAGE, pET-28a-anAGE andpET-28a-Nal, respectively. To ligate the second gene into a singleXhoI restriction site of the expression vectors, the vectors werelinearized with XhoI followed by dephosphorylation with alkalinephosphatase. The sequences of the constructed polycistronic
a200KDa116KDa97.2KDa66.4KDa
44.3KDa
29.0KDa
20.1KDa
Nal
AGE
1 2 3 4 5 6 7 8 9
b200KDa116KDa97.2KDa66.4KDa
44.3KDa
29.0KDa
20.1KDa
Nal
AGE
1 2 3 4 5 6 7 8 9
Fig. 4. (a) SDS–PAGE analysis of expressed proteins. L1, L2, E. coli host without anyplasmid, supernatant and pellet; L3, protein molecular weight marker; L4, L5, pET-28a-Nal, supernatant and pellet; L6, L7, pET28a-snAGE, supernatant and pellet; L8,L9, pET28a-anAGE, supernatant and pellet. (b) SDS–PAGE analysis of co-expressedproteins. L1, L2, pET28a-Nal-snAGE, supernatant and pellet; L3, L4, pET28a-snAGE-Nal, supernatant and pellet; L5, protein molecular weight marker; L6, L7, pET28a-Nal-anAGE, supernatant and pellet; L8, L9, pET28a-anAGE-Nal, supernatant andpellet.
0
1
2
3
4
5
6
7
8
Enz
yme
activ
ity (
U/m
g)
Nal activityAGE activity
Fig. 5. Nal and AGE activities of recombinant strains with E. coli Rosetta (DE3) withpLysS as a control. The reactions were performed in triplicate, and error barsrepresent the standard error of the mean.
26 W. Sun et al. / Bioresource Technology 130 (2013) 23–29
plasmids were determined, and plasmids with open readingframes in the correct orientation were transformed into E. coli Ro-setta (DE3) pLysS.
2.3. Expression of recombinant expression systems
The E. coli strains carrying the recombinant plasmids, pET-28a-Nal, pET-28a-snAGE, pET-28a-anAGE, pET-28a-Nal-snAGE, pET-28a-Nal-anAGE, pET-28a-snAGE-Nal and pET-28a-anAGE-Nal,were grown on LB agar supplemented with 10 lg/mL kanamycinand 34 lg/mL chloromycetin at 37 �C. A single colony was trans-ferred into 5 mL of LB medium with kanamycin and chloromycetinat the same concentrations as used in the solid medium in a 50 mLflask. The culture was incubated overnight at 37 �C, 200 rpm. Oneml of the overnight culture was transferred to a 1-l flask containing100 mL of LB medium with kanamycin and chloromycetin. Whenthe OD600 reached 0.6–0.8, 0.5 mM isopropyl b-D-1-thiogalactopy-ranoside (IPTG) was added and the culture was incubated at 28 �C,200 rpm for 10 h before harvesting.
2.4. Enzyme assay
E. coli Rosetta (DE3) pLysS without plasmids was cultured underthe same conditions as the recombinant E. coli strains as a controlto analyze expression level and solubility. Cells harvested bycentrifugation at 12,000g for 5 min were resuspended in 0.1 MTris–HCl pH 7.5 followed by sonication. Cell debris was removedby centrifugation at 12,000g for 20 min. The supernatant and pelletwere analyzed by sodium dodecyl sulfate–polyacrylamide gel elec-trophoresis (SDS–PAGE). The SDS–PAGE was performed using 12%polyacrylamide gels and protein stained with Coomassie brightblue. The protein concentration was determined by the Bradfordmethod using bovine serum albumin as standard.
Nal activity was assayed by measuring its ability to condenseManNAc and pyruvate into Neu5Ac. The reaction mixture wascomposed of 0.1 M pyruvate, 0.1 M ManNAc and 0.1 M Tris–HClpH 7.5. An appropriate amount of crude enzyme was added to1 mL reaction mixture and incubated at 37 �C for 20 min. AGEactivity was assayed by measuring its ability to transform GlcNAcinto ManNAc. The reaction mixture contained 0.1 M GlcNAc, 5 mMATP and 0.1 M Tris–HCl pH 7.5. Reactions were terminated by boil-ing the mixture for 5 min. After centrifugation at 12,000g for10 min and filtration through 0.22 lm membrane, the concentra-tions of the substrate and the product were analyzed by highperformance liquid chromatography (HPLC). All tests wereperformed in triplicate and 1 unit of enzyme activity was definedas the amount of enzyme needed to produce 1 lmol product permin. Concentrations of GlcNAc, ManNAc, Neu5Ac, and pyruvatewere analyzed on an Agilent 1200 system, equipped with a Bio-Rad Aminex HPX-87H column (300 � 7.8 mm) using a refractiveindex detector. The mobile phase consisted of 10 mM H2SO4 at0.4 mL/min, 55 �C.
2.5. Production of Neu5Ac from GlcNAc by whole-cell transformation
Four recombinant strains harboring polycistronic plasmidswere cultivated and induced individually in 500 mL LB mediumunder the conditions described above. 2 g cells (wet weight) ofeach strain were harvested for whole-cell catalysis. A 50-mL glassvessel with 20 mL of the reaction mixture was used for the con-version of GlcNAc into Neu5Ac. The reaction mixture contained0.8 M GlcNAc, 1.2 M pyruvate, 10 mM Mgcl2�6H2O and 0.1 MTris–HCl pH 7.5. The harvested cells were washed twice with0.1 M Tris–HCl pH 7.5 and resuspended with the reaction mixtureto catalyze the two continuous reactions (Fig. 2). The reactionsystem was incubated in a 30 �C water bath shaker at 200 rpm
for 60 h. The system pH was adjusted to 7.5 every 2 h. Sampleswere taken at intervals to analyze the concentration of GlcNAc,ManNAc and Neu5Ac by HPLC.
Fig. 6. Concentrations of GlcNAc, ManNAc and Neu5Ac during the catalystic process. ManNAc and Neu5Ac were expressed in units of mg/mL, and GlcNAc was denoted as thepercentage of its starting concentration. Reactions were performed in triplicate, and error bars represent the standard error of the mean. (a) Reaction process for strain pET-28a-Nal-snAGE. (b) Reaction process for strain pET-28a-snAGE-Nal. (c) Reaction process for strain pET-28a-Nal-anAGE. (d) Reaction process for strain pET-28a-anAGE-Nal.
W. Sun et al. / Bioresource Technology 130 (2013) 23–29 27
3. Results and discussion
3.1. Construction and expression of recombinant plasmids
The maps of the four plasmids constructed with the requiredgenes are shown in Fig. 3. SDS–PAGE and enzyme assay resultsfor these recombinant strains are shown in Fig. 4 and Fig. 5 withE. coli Rosetta (DE3) pLysS as the control sample. Nal was mostlyexpressed in a soluble form and the recombinant strain showedapproximately 12-fold higher Nal activity than the control. In con-trast, AGE was expressed mainly as inclusion bodies, and no AGEactivity was detected in the control group. The anAGE activitywas three times that of snAGE in the present study; however, puri-fied anAGE showed 22 times the activity of purified snAGE in thereports, with 525.8 U/mg and 23.2 U/mg, respectively (Lee et al.,2007b; Tabata et al., 2002). Since the activity of anAGE was lowerin the crude extract it is possible that anAGE had a lower expres-sion level or lower solubility compared with that of snAGE in thecrude extract (Fig. 4a). As the original nucleotide sequence encod-ing anAGE was not available as a control, it was not possible to as-sess whether the differences in the expression levels of anAGE wascaused by codon optimization or the protein sequence itself.
3.2. Effects of gene position on protein expression
The order of genes encoding Nal and AGE in the expression vec-tor severely affected their expression. Arranged in the order of pro-moter-Nal-AGE, Nal activity was approximately the same as when
expressed alone (Fig. 5), while the expression level of Nal is muchhigher than that of AGEs (Fig. 4). However, when the orderswitched to promoter-AGE-Nal, Nal activity dropped to approxi-mately 25% of that achieved when expressed alone (Fig. 5). Thesame pattern was also observed for AGEs. It is worth to mentionthat when arranged adjacent to the promoter, AGE formed a largeamount of inclusion bodies, while negligible inclusion bodies wereformed when AGE was right next to the Nal and far away from thepromoter (Fig. 4b).
3.3. Whole-cell catalysis to produce Neu5Ac from GlcNAc
Both of the reactions in the present study were reversible, Nalfavored cleavage of Neu5Ac into ManNAc and pyruvate (Li et al.,2008), and AGEs tended to epimerize ManNAc into GlcNAc (Leeet al., 2007a; Tabata et al., 2002). However, their reverse reactionswere needed to produce Neu5Ac from GlcNAc. This characteristicof these enzymes made production of Neu5Ac from GlcNAC viaManNAc an unfavored reaction, and a large amount of substrateand intermediate existed in the reaction system even when itreached equilibrium.
The concentrations of ManNAc and Neu5Ac (mg/mL), and thepercentage of GlcNAc that remained (starting concentration of176.5 mg/mL) in the reaction system were sampled during thereaction process (Fig. 6). ManNAc reached 80% of its highest con-centration in the first 20 h and then the ratio of Neu5Ac to ManNAcincreased. The delay of Nal to catalyze ManNAc and pyruvate intoNeu5Ac made it the rate-limiting step of the two reactions.
28 W. Sun et al. / Bioresource Technology 130 (2013) 23–29
As for final production of Neu5Ac, the strain harboring pET-28a-Nal-anAGE showed a higher production level than the other threestrains. It produced 61.3 mg/mL Neu5Ac, which was almost twicethe production of the other strains. In total, 52% of GlcNAc wasconverted into Neu5Ac and ManNAc, and 25% of GlcNAc wassuccessfully converted into Neu5Ac. The advantage of Nal-anAGEover anAGE-Nal in the production of Neu5Ac might be explainedfrom the perspectives of both Nal and AGE. For Nal, a higherexpression level alleviated the rate-limiting step whereas a lowerexpression level of AGE allowed proper folding of this enzymeand adequate activity.
The anAGE from Anabaena sp. CH1 showed much higher activityand productivity than snAGE from Synechocystis sp. PCC 6803. Taoet al. (2011) produced 191 mM Neu5Ac (59 g/L) by constructing atemperature-inducible plasmid for simultaneous expression ofNal and snAGE. Disruption of the nagE gene (Fig. 1) and the addi-tion of surfactant were used to reduce side reactions and enhancemass transfer. Even though such measures were not carried out inthe current study, nearly the same level of production wasachieved as in the study by Tao et al.
The construction of the polycistronic expression system in thepresent study resulted in several advantageous features. First, con-structing pET-28a-Nal-anAGE led to high expression of both Naland AGE in soluble forms. Second, two enzymes in one cell avoidedthe need to transfer ManNAc across the cell membrane to link thetwo reactions. Both by constructing snAGE and Nal on individualplasmids, Tabata et al. (2002) and Zhang et al. (2010) produced12.3 g/L and 19.1 g/L Neu5Ac, which are one-fourth and one-thirdof the production in the present study, respectively. Third, it ismuch more convenient to obtain large amounts of the biocatalyticreaction by fermenting a single strain. Xu et al. (2007) obtained18.32 g/L Neu5Ac by culturing E. coli and Pseudomonas stutzeri cellssimultaneously, which was a more complex process than the oneapplied in the present study. Lastly, by whole-cell catalysis, it isunnecessary to lyse cells and cells generated their own ATP. Bothfeatures reduced the cost of Neu5Ac production. Hu et al. (2010)immobilized Nal and AGE, but had to add more than 2.5 mM ATPto the cell-free system to produce Neu5Ac.
4. Conclusions
Neu5Ac is an important drug precursor and food additive, andthe E. coli strain constructed in this study has a promising futurefor industrial application to synthesize Neu5Ac. The strategy of try-ing more than one AGE in the present study successfully enhancedfinal production, and the access to the genetic materials was con-venient using whole-gene synthesis. It is also worth noting thatthe problem of forming AGE inclusion bodies was successfullysolved by constructing polycistronic plasmids in this study, andthis strategy could be applied in many other protein expressions.
Acknowledgements
This work was supported by the National Basic ResearchProgram of China (973) (Grant No.: 2011CBA00806); the NationalScience Fund for Distinguished Young Scholars (Grant No.:21025625); Program for New Century Excellent Talents at theUniversity of Ministry of Education of China (Grant No.: NCET-11-0987); the Research Fund for the Doctoral Program of HigherEducation of China (RFDP) (Grant No.: 20113221120007); Programfor Changjiang Scholars and Innovative Research Team in theUniversity (Grant No.: IRT1066); the natural science foundationof Jiangsu Province (Grant No.: BK2011031) and the PriorityAcademic Program from Development of Jiangsu Higher EducationInstitutions.
References
Chien, P.S., Tzeng, T.T., Lin, S.C., Hsu, W.H., 2007. Refolding of recombinant N-acetyl-D-glucosamine 2-epimerase by a fed-batch process. J. Chin. Inst. Chem. Eng. 38,215–221.
Ferrero, M.A., Aparicio, L.R., 2010. Biosynthesis and production of polysialic acids inbacteria. Appl. Microbiol. Biotechnol. 86, 1621–1635.
Gopaul, K.P., Crook, M.A., 2006. Sialic acid: a novel marker of cardiovasculardisease? Clin. Biochem. 39, 667–681.
Hao, J., Balagurumoorthy, P., Sarilla, S., Sundaramoorthy, M., 2005. Cloning,expression, and characterization of sialic acid synthases. Biochem. Biophys.Res. Commun. 338, 1507–1514.
Hu, S., Chen, J., Yang, Z., Shao, L., Bai, H., Luo, J., Jiang, W., Yang, Y., 2010. Coupledbioconversion for preparation of N-acetyl-D-neuraminic acid usingimmobilized N-acetyl-D-glucosamine-2-epimerase and N-acetyl-D-neuraminicacid lyase. Appl. Microbiol. Biotechnol. 85, 1383–1391.
Ishikawa, M., Koizumi, S., 2010. Microbial production of N-acetylneuraminic acid bygenetically engineered Escherichia coli. Carbohyd. Res. 345, 2605–2609.
Itoh, T., Mikami, B., Hashimoto, W., Murata, K., 2008. Crystal structure of YihS incomplex with D-mannose: structural annotation of Escherichia coli andSalmonella enterica yihS-encoded proteins to an aldose-ketose isomerase. J.Mol. Biol. 377, 1443–1459.
Itoh, T., Mikami, B., Maru, I., Ohta, Y., Hashimoto, W., Murata, K., 2000. Crystalstructure of N-acyl-D-glucosamine 2-epimerase from porcine kidney at 2.0 aresolution. J. Mol. Biol. 303, 733–744.
Krüger, D., Schauer, R., Traving, C., 2001. Characterization and mutagenesis of therecombinant N-acetylneuraminate lyase from Clostridium perfringens. Eu. J.Biochem. 268, 3831–3839.
Lee, J.O., Yi, J.K., Lee, S.G., Takahashi, S., Kim, B.G., 2004. Production of N-acetylneuraminic acid from N-acetylglucosamine and pyruvate usingrecombinant human renin binding protein and sialic acid aldolase in one pot.Enzyme Microb. Technol. 35, 121–125.
Lee, Y.C., Chien, H.C., Hsu, W.H., 2007a. Production of N-acetyl-D-neuraminic acidby recombinant whole cells expressing Anabaena sp. CH1 N-acetyl-D-glucosamine 2-epimerase and Escherichia coli N-acetyl-D-neuraminic acidlyase. J. Biotechnol. 129, 453–460.
Lee, Y.C., Wu, H.M., Chang, Y.N., Wang, W.C., Hsu, W.H., 2007b. The central cavityfrom the (Alpha/Alpha)6 barrel structure of Anabaena sp. CH1 N-acetyl-D-glucosamine 2-epimerase contains two key histidine residues for reversibleconversion. J. Mol. Biol. 367, 895–908.
Mahmoudian, M., Noble, D., Drake, C.S., Middleton, R.F., Montgomery, D.S., Piercey,J.E., Ramlakhan, D., Todd, M., Dawson, M.J., 1997. An efficient process forproduction of N-acetylneuraminic acid using N-acetylneuraminic acid aldolase.Enzyme Microb. Technol. 20, 393–400.
Maru, I., Ohnishi, J., Ohta, Y., Tsukada, Y., 1998. Simple and large-scale production ofN-acetylneuraminic acid from N-acetyl-D-glucosamine and pyruvate using N-acyl-D-glucosamine 2-epimerase and N-acetylneuraminate lyase. Carbohyd.Res. 306, 575–578.
Maru, I., Ohnishi, J., Ohta, Y., Tsukada, Y., 2002. Why is sialic acid attracting interestnow? complete enzymatic synthesis of sialic acid with N-acylglucosamine 2-epimerase. J. Biosci. Bioeng. 93, 258–265.
Maru, I., Ohta, Y., Murata, K., Tsukada, Y., 1996. Molecular cloning and identificationof N-acyl-D-glucosamine 2-epimerase from porcine kidney as a renin-bindingprotein. J. Biol. Chem. 271, 16294–16299.
Nahalka, J., Vikartovska, A., Hrabarova, E., 2008. A crosslinked inclusion bodyprocess for sialic acid synthesis. J. Biotechnol. 134, 146–153.
Ogura, H., 2011. Development of miracle medicines from sialic acids.. Proc. Jpn.Acad. Ser. B Phys. Biol. Sci. 87, 328–361.
Oriquat, G.A., Saleem, T.H., Abdullah, S.T., Soliman, G.T., Yousef, R.S., Adel Hameed,A.M., Salim, M.L., 2011. Soluble CD14, sialic acid and L-Fucose in breast milk andtheir role in increasing the immunity of breast-fed infants. Am. J. Biochem.Biotechnol. 7, 21–28.
Plumbridge, J., Vimr, E., 1999. Convergent pathways for utilization of the aminosugars N-acetylglucosamine, N-acetylmannosamine, and N-acetylneuraminicacid by Escherichia coli. J. Bacteriol. 181, 47–54.
Sanchez-Carron, G., Garcia-Garcia, M.I., Lopez-Rodriguez, A.B., Jimenez-Garcia, S.,Sola-Carvajal, A., Garcia-Carmona, F., Sanchez-Ferrer, A., 2011. Molecularcharacterization of a novel N-acetylneuraminate lyase from Lactobacillusplantarum WCFS1. Appl. Environ. Microbiol. 77, 2471–2478.
Sola-Carvajal, A., Sánchez-Carrón, G., García-García, M.I., García-Carmona, F.,Sánchez-Ferrer, Á., 2012. Properties of BoAGE2, a second N-acetyl-D-glucosamine 2-epimerase from Bacteroides ovatus ATCC 8483. Biochimie. 94,222–230.
Tabata, K., Koizumi, S., Endo, T., Ozaki, A., 2002. Production of N-acetyl-neuraminicacid by coupling bacteria expressing N-acetyl-glucosamine 2-epimerase and N-acetyl-neuraminic acid synthetase. Enzyme Microb. Technol. 30, 327–333.
Takahashi, S., Hori, K., Takahashi, K., Ogasawara, H., Tomatsu, M., Saito, K., 2001.Effects of nucleotides on N-acetyl-D-glucosamine 2-epimerases (renin-bindingproteins): comparative biochemical studies. J. Biochem. 130, 815–821.
Tao, F., Zhang, Y., Ma, C., Xu, P., 2010. Biotechnological production and applicationsof N-acetyl-D-neuraminic acid: current state and perspectives. Appl. Microbiol.Biotechnol. 87, 1281–1289.
W. Sun et al. / Bioresource Technology 130 (2013) 23–29 29
Tao, F., Zhang, Y., Ma, C., Xu, P., 2011. One-pot bio-synthesis: N-acetyl-D-neuraminic acid production by a powerful engineered whole-cell catalyst. Sci.Rep. 1, 142.
Van Rinsum, J., Van Dijk, W., Hooghwinkel, G.J., Ferwerda, W., 1983. Subcellularlocalization and tissue distribution of sialic acid precursor-forming enzymes.Biochem. J. 210, 21–28.
Varki, N.M., Varki, A., 2007. Diversity in cell surface sialic acidpresentations: implications for biology and disease. Lab. Invest. 87,851–857.
