3-오정석( ).hwp

한국생물공학회지제 23 권제 6 호 Korean J. Biotechnol. Bioeng. Vol. 23, No. 6, 474-478(2008) 대장균의 UDP-glucose regeneration 시스템을이용한이당류합성에관한연구 오정석조지아공대화학생명공학과 ( 접수 : 2008. 9. 29., 게재승인 : 2008. 11. 3.) Disaccharide Synthesis using E. coli UDP-glucose regeneration system Jeong Seok Oh School of Chemical and Biomolecular Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332-0100 (Received : 2008. 9. 29., Accepted : 2008. 11. 3.) UDP-glucose regeneration system using metabolic engineeringis unique and efficient strategy for oligosaccharide synthesis. To exploit the efficient UDP-glucose regeneration system, we introduced four enzymes, which would be important in partitioning the flux of UDP regenerationsuch as UDP-glucose pyrophosphorylase, UDP-Kinase gene, UDP-galactose 4-epimerase, and ß-1, 4-galactasyltrasnsferase, into E. coli AD202. To determine the optimal expression level for UDP-regeneration, LacNAc concentration was compared depending on IPTG concentration. 0.5 mm IPTG induction showed the higher oligosaccharides synthesis. Using metabolic engineering under optimal IPTG induction, LacNAc synthesis of AD202/pQNGLU increased until 16 h and showed the 1.34 mm. This concentration is 10 times higher than that of control strain at same reaction time. Lactose of AD202/pQNGLU showed the maximum synthesis of 0.39 mm at 16 h and showed the 2.6 times higher than that of control strain. Key Words : Metabolic engineering UDP-glucose regeneration; Oligosaccharide synthesis; E. coli whole-cell biocatalytic synthesis 1) 서론 올리고당은 antibody 나 protein 의 cofactor 로서, 박테리아와바이러스 receptor 와같은다양한종류의생화학적인합성에중요한역할을해왔고 (1, 2), 이를산업적으로합성하기위해서화학적또는생촉매적합성방법이고안되었다. 화학적인합성방법은여러단계의 protection 과 deprotection 을거치기때문에생산성과합성수율이낮아산업화를하기에는비효율적이다 (3). 반면에효소나 whole-cell 시스템을이용하는생촉매방법은화학적인방법에비해서단계가단순화되어산업적으로더실용적이고, 효과적인방법으로인식되고있다 (4, 5, 6, 7). 대사공학은유전자재조합기술과분자생물학및화학공학적기술을이용하여새로운대사회로를도입하거나기존의대사회로를제거, 증폭, 변경시켜세포나균주의대사특성을원하는방향으로바꾸는일련의기술이다 (8). 대사공학을사용한 whole-cell Corresponding Author : School of Chemical and Biomolecular Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332-0100 Tel : +82-10-5763-1005, Fax : +82-2-3660-0782 E-mail : ojseok@gmail.com 시스템은올리고당합성에있어서큰장점을가지고있다 (9, 10, 11, 12). whole-cell 시스템을이용한효율적인올리고당합성방법에사용되는제한인자는 NAD 나 NADH 와달리 multi-pathway 에의해서공급되는인자이다. 따라서원활한 cofactor 공급시스템개발이필요하게되었다. 가장잘알려진대장균시스템 (13, 14) 은자연적으로 UDP-glucose regeneration 시스템을가지고있다 (15). 하지만자연적인대장균의 UDP-glucose regeneration 효율은높지않기때문에 regeneration 효율을높이기위해서대사공학적인접근이필요하다 (16). 본논문에서는올리고당합성능력을향상시키기위해서대사공학적인방법을사용하여 UDP-glucose regeneration 효율을증가시켜연구를수행하였다. 재료및방법 균주및배양조건 Biology reagents 들은 Qiagen, Invitrogen, Promega, 그리고 BioRad 에서구입하였고, 시약들은 Sigma-Aldrich 와 Fisher 에서구입하여사용하였다. 사용한 bacterial strains 과 plasmid 는 Table 1 에정리해놓았다배양조건은다음과같다. LB 배지를사용하여 30, 250 rpm 474

Oh, J. S., Disaccharide Synthesis using E. coli UDP-glucose regeneration system 475 에서 OD 600nm 가 0.6-0.7 될때까지배양을수행한후 IPTG 를첨가하고, 충분한양의효소를발현시키기위해서 5 시간더배양하였다. 5 시간배양후 centrifuge 를통해서 cells 을수집하고 PBS buffer 로세척한후 reaction buffer {1M Tris-HCl (7.5) 20 mm K 2HPO 4, 2 mm MnCl 24H 2O, 200 mm glucose, 5 g/l MgSO 47H 2O, 50 mm GluNAc} 로 resuspend 시켜 30, 250 rpm 에서 LacNAc 합성배양을수행하였다. Table 1. Bacterial strains and plasmid used in this work Strain/plasmid Description Source/reference Bacterial strains: E. coli JM109 F trad36 proa+b+laciq (lacz)m15/ (lac -proab) glnv44 e14- gyra96reca1 rela1 [20] enda1 thi hsdr17 E. coli AD202 ompt::kan Dr. Wakarchuk [17] Plasmids: pgem-t Easy vector amp r Promega pqe80l Commercial expressional vector containing T5 promoter and laci q Qiagen pmuel pqe80l containing pgm, galu and gale: lgtb fustion gene Mao et al. pt-ndk pgem-t easy vector containing ndk gene This study pqu pqe80l containing galu gene This study pqnu pqe80l containing ndk and galu gene This study pqnglu pqe80l containing ndk, gale: lgtb fustion and galu gene This study 플라스미드제작발현시킬 gene 들은 ribosome binding site 와 stop cordon 를가지고, 하나의 promoter 하에서발현이될수있도록플라스미드를제작하였다. E. coli UDP-glucose pyrophosphorylase gene (galu) 은 JM109 genomic DNA 에서 forward primer (ATATCTGCAGTGCGATACAGAAATATGAACAC, PstI) 와 reverser primer (TACTAAGC TTGATTGCTCAACGCCG, HindIII) 를사용하여증폭시켰다. 증폭된 galu gene (1 kb) 은 pqe80l 벡터에삽입이되어 pqu 벡터가되었다. E. coli UDP-kinase gene (ndk) 은 JM109 genomic DNA 에서 forward primer (5 -ATATAGATCTATGGCTATTGAACGTACTTTTTCC-3, BglII) 와 reverse primer (5 -TAGGATCC TTATTAACGGGTGCGC, BamHI ) 를사용하여증폭되었다. 증폭된 ndk (428 bp) gene 은 pgem-t Easy vector 에삽입이되어 pt-ndk 벡터가되었고. pt-ndk 벡터를 restriction 효소 BglII 와 SacI 으로처리한후 BglII 과 BamHI-SacI 을가진 ndk gene 얻었다. 이 ndk gene 를 pqu 벡터에삽입하여 pqnu 벡터를만들었다. gale:lgtb fusion gene (2 kb) 은 pmuel 벡터에서 forward primer (5 -AGAGA GATCTCACACAGAATTCATTAAAGAGGAG-3, BglII) 와 reverseprimer (5 -CTTCTCGAGTCTATCAACAGGAGTCCAAGCT, XhoI) 를사용하여증폭하고, pqnu 에삽입하여 pqnglu 벡터를만들었다. 분석방법 UDP-kinase (Ndk), UDP-glucose pyrophosphorylase (GalU) 와 UDP-galactose 4 -epimerase fused with ß-1, 4-galactosyltrasnsferase (GalE:LgtB) 의발현은 SDS-PAGE gel 를통해서확인하였다. Carbohydrate 농도는 CarboPac PA20 analytical column 을가진 Dionex BioLC system (Sunnyvale, CA) 을사용하여측정하였다. 측정방법은다음과같다. Dionex ED50 electrochemical detector 의 pulsed amperometry (waveform: t=0.41 s, p=-2.00 V; t=0.42 s, p=-2.00 V; t=0.43 s, p=0.60 V; t=0.44 s, p=-0.10 V; t=0.50 s, p=-0.10 V) 를사용하였는데, 200 mm sodium hydroxide (A) 와 18 MΩ-cm water 로구성된 mobile phase 을사용하였고, linear gradient 는 t=0 min, 5:95 (A:B); t=5 min, 5:95; t=10 min, 20:80; t=20 min, 20:80; t=21 min, 100:0; t=30 min, 100:0; t=35 min, 5:95; t=50 min, 5:95 이다 Lactose and LacNAc 농도는 standard 용액으로측정한 calibration curves 를사용하여결정되었다. 결과 대사공학을이용한 UDP-glucose regeneration 시스템 Fig. 1 에서보듯이 UDP-glucose regeneration 시스템을이용하여올리고당을합성하는데있어서출발물질은 UDP-glucose 이고, regeneration flux 에관여하는효소는 4 가지이다. 본연구에서는효율적인 UDP-glucose regeneration system 을위해서 regeneration flux 에관련된 4 가지효소 (UDP-galactose-4 -epimerase, β-1,4- galactosyltransferase, UDP-kinase, UDP-glucose pyrophosphorylase) 들이한벡터에서각각발현되는시스템을사용하였다 (Fig. 2). 대사공학적으로구축된 UDP-glucose regeneration system 의효율성을확인하기위해서 E. coli AD202 에 vector 를 transformation 시켰고, 0.5 mm 의 IPTG 를첨가하여 regeneration flux 에관련된효소들의발현은 SDS-PAGE gel 상에서확인하였다. Fig. 3 에서보듯이 UDP-galactose-4 -epimerase 와 β-1,4-galactosyltransferase 의 fusion protein 은 68 KD, UDP-glucose pyrophosphorylase 은 32 KD 그리고 UDP-kinase 15.3 KD 로정확한크기의발현을확인할수있었다. UDP-glucose regeneration system 이 disaccharide 합성에미치는영향새롭게구축된 UDP-glucose regeneration system 에서의 LacNAc 합성에필요한최적효소의발현량을정하기위해서다양한 IPTG 농도 (0.1, 0.3, 0.5, 1mM) 에서 LacNAc 합성정도를비교해보았다. 샘플은다양한농도의 IPTG 첨가하고 5 시간동안효소를발현시킨후 reaction buffer 에서 20 시간배양을하였고, 일정량을취하여분석하였다. 다양한 IPTG 농도실험에서가장높은 LacNAc 합성농도는 0.5 mm 이다. Disaccharide 합성정도를최적의 IPTG 농도에서시간에따라비교해보았다. Fig. 5A 에서보듯이 E. coli AD202/pQNGLU 의 LacNAc 농도는 16 시간까지증가하는경향을보였고, 최대 LacNAc 농도는 1.34 mm 이다. 이농도는대조구인 AD202/pQE80L 에비해서 10 배정도높은생산성을보이고있다. Lactose 농도를비교해보면 AD202/pQNGLU 의최대 lactose 농도는 16 시간에 0.39 mm 로대조구보다 2.6 배높은농도를나타내었다 (Fig. 5B). 결론적으로 AD202/pQNGLU 의총 disaccharide 농도는 1.73 mm 이며, AD202/pQE80L 보다 6.5 배높은생산성을보였다 (Fig. 5C).

476 Korean J. Biotechnol. Bioeng., Vol. 23, No. 6 고찰 본논문에서는 oligosaccharide 합성에있어서 glucose-6-phosphate 에서 UDP-glucose 로가는 carbon flux 보다 UDP 에서 UTP 로전환시킨후 UDP-glucose 를재순환시키는것이더효율적인 flux 로가정하였다 (Fig. 1). 왜냐하면 oligosaccharide 합성은에너지의존적인시스템으로원활한 cofactor 의공급이제한인자이기때문이다. UDP-glucose regeneration 을간략하게설명하면다음과같다. UDP-kinase 에의해서 UDP 는 UTP 전환되고, UTP 는 UDP-glucose phosphorylase 에의해서 glucose-1-phosphate 에서 UDP-glucose 를전환된다. 이때만들어진 UDP-glucose 는 UDP-galactose-4-epimerase 와 β-1,4-galactosyltransferase 에의해서 N-acetyllactosamine (LacNAc) 이나 lactose 로전환되면서 UDP 를방출하게된다. 이때방출된 UDP 는 UDP-kinase 에의해서다시 UTP 가된다 (Fig. 1). Figure 2. Maps of pqnglu. Abbreviations are as follows: Amp R : ampicillin resistance gene; ColEI: ColEI origin of replication; gale:lgtb: fusion gene of UDP-galactose-4-epimerase and β-1,4-galactosyltransferase; galu: UDP-glucose pyrophosphorylase gene; ndk: UDP-kinase RBS: ribosome binding site; T5: T5 promoter. Figure. 1. Metabolic network of UDP-glucose (UDP-Glc) synthesis in E. coli. ACoA: acetyl-coa; ATP: adenosine triphosphate; Glc: glucose; G1P: glucose-1-phosphate; G6P: glucose-6-phosphate; GlcNAc: N-acetylglucosamine; G3P: glyceraldehyde-3-phosphate; KDGP: 2-keto-3-deoxygluconate-6-phosphate; LacNAc: N-acetyl lactosamine; PEP: phosphoenol-pyruvate; 6PG: 6-phosphogluconate; PRPP: 5-phosphoribosyl-1-pyrophosphate; PYR: pyruvate; R5P: ribose-5-phosphate; TCA cycle: tricarboxylic acid cycle; UDP-Gal: UDP-galactose; UDP-Glc: UDPglucose; UDP: uridine diphosphate; UMP: uridine monophsphate; UTP: uridine triphosphate. Enzymes/Proteins: 1: glucose-binding-protein for glucose uptake; 2: hexokinase; 3: phosphoglucomutase; 4: UDP-glucose pyrophosphorylase 5: UDPgalactose-4-epimerase; 6: β-1,4-galactosyltransferase 7: glucose-6-phosphate dehydrogenase; 8: 6-phosphogluconate dehydrogenase; 9: ribose-phosphate diphosphokinase; 10: orotate phosphoribosyltransferase and orotidine- 50-phosphate decarboxylase; 11: uridylate kinase; 12: UDP kinase; 13: phosphoglucose isomerase; 14: fructose-phosphate kinase and fructose diphosphate aldolase;15: glyceraldehyde-3-phosphate dehydrogenase, phosphoglycerate mutase, and enolase; 16: pyruvate kinase; 17: pyruvate dehydrogenase multienzyme complex; 18: citrate synthase. 효율적인효소발현을위해서하나의 vector 시스템을사용하였다. 각효소들은 ribosome binding site 와 stop cordon 을가지고 T5 promoter 에의해서발현을시켰고, SDS-PAGE 로확인을하였다 (Fig. 2, 3). Figure 3. Analysis of expression of Ndk, GalE:LgtB, GalU of AD202/pQNGLU by SDS-PAGE. The 68, 32, and 15.3 kd bands correspond to the fusion protein, GalU and Ndk proteins. The samples were taken 5 h after induction with 0.5 mm IPTG. UDP-glucose regeneration 의효율은각효소들의발현량에의존적이다. 따라서 IPTG 농도에따른 LacNAc 합성정도를비교해보았다 (Fig. 4). 0.5 mm IPTG 농도에서최대의 LacNAc 농도를나타내었다. 하지만 1 mm IPTG 에서는 0.5 mm IPTG 에서보다낮은 LacNAc 농도를나타내었다. 이것은각효소들의 overexpression 에의한 metabolic burden(18, 19) 으로추측된다. 대사공학적으로재구성된 metabolic flux regeneration 을가진 AD202/pQNGLU 은 Fig. 5 에서보듯이원활한 UTP 공급과효율적인 UDP-glucose 전환에의해서대조구인 AD202/pQE80L 보다높은 LacNAc 합성정도를보였다 (Fig. 5).

Oh, J. S., Disaccharide Synthesis using E. coli UDP-glucose regeneration system 477 본논문에서 metabolic flux regeneration 으로 disaccharides 합성을증가시킬수있다는것을보여주었다. 이전략은또한다른 oligosaccharides 합성에도적용시킬수있다. 왜냐하면 UDP-glucose 는 UDP-galactose 나 UDP-glucuronic acid 와같은다른 UDP-sugars 의 starting material 이기때문이다. 요약 Figure 4. LacNAc production depending on IPTG concentrations. The samples were taken the more 20 h culture after 5 h expression with different IPTG induction, centrifuge cells and resuspend with reaction buffer. 효율적인 UDP-glucose regeneration system 을구축하기위해서재순환시스템에관여하는 4 가지효소 (UDP-glucose pyrophosphorylase, UDP-Kinase gene, UDP-galactose 4-epimerase, and ß-1, 4-galactasyltrasnsferase) 들을 E. coli AD202 에서발현시켜 Disaccharide 합성정도를보았다. Disaccharide 는 0.5 mm IPTG 농도에서가장높은농도를나타내었다. 대조구와비교한결과 LacNAc 농도는 1.34 mm 로 10 배정도정가하였고, lactose 농도는 0.39 mm 로대조구보다 2.6 배증가하였다. 총 disaccharide 농도는 1.73 mm 이며, 대조구보다 6.5 배높은생산성을보였다. 본논문은결과는 metabolic flux regeneration 으로 disaccharides 합성을증가시킬수있다는것을보여주었다. Acknowledgment (A) 이논문은 2005 년도정부재원 ( 교육인적자원부학술연구조성사업비 ) 으로한국학술진흥재단의지원을받아연구되었음 (KRF -2005-214- D00255). REFERENCES (B) (C) Figure 5. Time course of disaccharide synthesis: (A) LacNAc, (B) Lactose, (C) total disaccharides, ( )AD202/pQE80L, ( )AD202/pQNGLU. 1. Wong, C.-H. and G. M. Whitesides (1994), Enzymes in Synthetic Organic Chemistry 12. 1st edition. New York: Oxford. 2. Gabius, H.-J. and S. Gabius (1997), Glycoscience: status and perspectives. Champman and Hall Gmbh. Weinheim, Germany. 3. Flowers, H. M. (1978), Chemical synthesis of oligosaccharides. Methods Enzymol. 50, 93-121. 4. Flitsch, S. L. (2000), Chemical and enzymatic synthesis of glycopolymers. Curr. Opin. Chem. Biol. 4(6), 619-625. 5. Koeller, K. M. and C. H. Wong (2000), Complex carbohydrate synthesis tools for glycobiologists: enzyme-based approach and programmable one-pot strategies. Glycobiology 10(11), 1157-1169. 6. Crout, D. H. and G. Vic (1998), Glycosidases and glycosyl transferases in glycoside and oligosaccharide synthesis. Curr. Opin. Chem. Biol. 2(1), 98-111. 7. Meynial-Salles, I and D. Combes. (1996), In vitro glycosylation of proteins: An enzymatic approach. J. Biotechnol. 46(1), 1-14. 8. Bailey, J. E. (1991), Toward a science of metabolic engineering. Science 252, 1668-1675. 9. Bettler, E., E. Samain, V. Chazalet, C. Bosso, A. Heyraud, D. H. Joziasse, W. W. Wakarchuk, A. Imberty, and R. A. Geremia. (1999), The living factory:in vivo production of N-acetyllactosamine containing carbohydrates in E. coli. Glycoconjugate J. 16, 205-212 10. Endo, T. and S. Koizumi (2000), Large-scale production of oligosaccharides using engineered bacteria. Curr. Opin. Struct. Biol. 10, 536 541. 11. Endo, T., S. Koizumi, K. Tabata, S. Kakita, and A. Ozaki. (1999),

478 Korean J. Biotechnol. Bioeng., Vol. 23, No. 6 Largescale production of N-acetylactosamine through bacterial coupling. Carbohydrate Res. 316, 179 183. 12. Ruffing, A. and R. R. Chen. (2006), Metabolic engineering of microbes for oligosaccharide and polysaccharide synthesis. Microbial Cell Factories 5, 25. 13. Jana, S. and J. K. Deb, (2005), Strategies for efficient production of heterologous proteins in E. coli. Appl. Microbiol. Biotechnol. 67, 289 298. 14. Sorensen, H. P. and K. K. Mortenson. (2005), Advanced genetic strategies for recombinant protein expression in E. coli. J. Biotechnol. 115, 113 128. 15. Hokke, C. H., A. Zervosen, L. Ellig, D. H. Joziasse, and D. H. Van den Eijnden. (1996), One-pot enzymatic synthesis of the Gal α13galβ14glcnac sequence with in situ UDP-Gal regeneration. Glycoconj. J. 13, 687-692. 16. Ruffing, A., M. Zichao, and R. R. Chen. (2006), Metabolic engineering of Agrobacterium sp. for UDP-galactose regeneration and oligosaccharide synthesis. Metabolic. Eng. 8, 465 473. 17. Nakano, H., T. Yamazaki, T. Ikeda, H. Masai, S. Miyataki, and L. Saito. (1994), Purification of glutathione S-transferase fusion proteins as a nondegraded form by using a protease-negative E. coli strain, AD202. Nucleic. Acids. Res. 22, 543-544. 18. Rozkov, A. Avignone-rossa, F. ERTL, Jones, D. O'kennedy, J. J. Smith, J. W. Dale, and M. E. Bushell. (2004), Characterization of the Metabolic burden on Escherichia coli DH1 cells imposed by the presence of a plasmid containing a gene therapy sequence. Biotechnol. Bioeng. 88, 909-915. 19. Dave, S.-W. O., P. M. Nissom, R. Philp, S. K.-W. Oh, and M. G.-S. Yap. (2006), Global transcriptional analysis of metabolic burden due to plasmid maintenance in Escherichia coli DH5 during batch fermentation. Enzyme Microb. Technol. 39, 391-398. 20. Yanisch-Perron, C., J. Vieira, and J. Messing. (1985), Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mp18 and puc19 vectors. Gene 33, 103-119.