Creatine phosphate-creatine kinase in enzymatic synthesis of glycoconjugates

Article abstract of DOI:10.1021/ol034319a

Enzymatic production of glycoconjugates is hampered by expensive phosphagens such as acetyl phosphate (AcP) and phosphoenolpyruvate (PEP). Here, we introduce creatine phosphate-creatine kinase system as a novel and practical energy source in carbohydrate

Full text of DOI:10.1021/ol034319a

ORGANIC
LETTERS
2003
Vol. 5, No. 15
2583-2586
Creatine Phosphate−Creatine Kinase in
Enzymatic Synthesis of Glycoconjugates
Jianbo Zhang, Bingyuan Wu, Yingxin Zhang, Przemyslaw Kowal, and
Peng George Wang*
Department of Chemistry, Wayne State UniVersity, Detroit, Michigan 48202
pwang@chem.wayne.edu
Received February 21, 2003
ABSTRACT
Enzymatic production of glycoconjugates is hampered by expensive phosphagens such as acetyl phosphate (AcP) and phosphoenolpyruvate
(PEP). Here, we introduce creatine phosphate−creatine kinase system as a novel and practical energy source in carbohydrate synthesis. This
system was successfully demonstrated in the production of bioactive oligosaccharides with different sugar nucleotide regeneration systems.
The past twenty years have seen a dramatic intensification
in the interest in carbohydrates due to new findings on their
biological roles. Glycosylation not only prolongs the half-
lives of associated proteins and hormones but also helps
many important cellular recognition events in biological
systems.1 Biochemical and biophysical studies of glycocon-
jugates apparently require large-scale preparations of oli-
gosaccharides of interest.2 However, the low efficiencies of
oligosaccharide synthesis systems have seriously limited
progress in glycoscience, especially in the area of glycome-
dicine. Many attempts to improve the productivity have been
made, mainly by exploitation of new glycosyltransferases
and novel biosynthetic pathways.3 Despite that the sugar-
nucleotide regeneration systems solved the major obstacle
in the carbohydrate production, the expensive sugar nucle-
otide pool, the technologies are still greatly dependent on
high-energy phosphate donors.4 For example, in our previous
studies of “superbeads” technology, the major cost was from
phosphoenolpyruvate (PEP).3f
Creatine phosphate (phosphocreatine, CP, PCr) is an
organic phosphate compound in cellular energy metabolism,
in addition to ATP and other nucleoside triphosphates (Figure
1).5 Because the standard free energy of hydrolysis of CP
(-43.1 kJ/mol) is higher than that of ATP (-30.6 kJ/mol),
ADP is phosphorylated into ATP in mammalian brain, heart
and muscle, via the CP-CK (creatine kinase, creatine
phosphokinase, E.C. 2.7.3.2) shuttle for intracellular energy
transfer.6 The CP-CK system has been utilized to regenerate
ATP for de noVo protein syntheses.7 Though it may seem
(1) (a) Varki, A. Glycobiology 1993, 3, 97-130. (b) Rudd, P. M.;
Wormald, M. R.; Stanfield, R. L.; Huang, M.; Mattsson, N.; Speir, J. A.;
DiGennaro, J. A.; Fetrow, J. S.; Dwek R. A.; Wilson, I. A. J. Mol. Biol.
1999, 293, 351-366. (c) Lowe, J. B. Cell 2001 104, 809-812. (d) Helenius,
A.; Aebi, M. Science 2001, 291, 2364-2369.
(2) (a) Danishefsky, S. J.; Allen, J. R. Angew. Chem., Int. Ed. 2000, 39,
836-863. (b) Sears, P.; Wong, C.-H. Science 2001, 291, 2344-2350.
(3) (a) Palcic, M. M. Curr. Opin. Biotechnol. 1999, 10, 616-624. (b)
Izumi, M.; Shen, G.-J.; Wacowich-Sgarbi, S.; Nakatani, T.; Plettenburg,
O.; Wong, C.-H. J. Am. Chem. Soc. 2001, 123, 10909-10918. (c)
DeAngelis, P. L. Glycobiology 2002, 12, 9R-16R. (d) Wong, C.-H.; Haynie,
S. L.; Whitesides, G. M. J. Org. Chem. 1982, 47, 5418-5420. (e) Gilbert,
M.; Bayer, R.; Cunningham, A.-M.; DeFrees, S.; Gao, Y.; Watson, D. C.;
Young, N. M.; Wakarchuk, W. W. Nature Biotechnol. 1998, 16, 769-772.
(f) Chen, X.; Fang, J.; Zhang, J.; Liu, Z.; Shao, J.; Kowal, P.; Andreana,
P.; Wang P. G. J. Am. Chem. Soc. 2001, 123, 2081-2082.
(4) Introduction to Glycobiology; Taylor, M. E., Drickamer, K., Eds.;
Oxford University Press: New York, NY, 2003; p 9-11.
(5) (a) Jencks, W. P. In Handbook of Biochemistry and Molecular
Biology, 3rd ed.; Fasman, G. D., Ed.; Chemical Rubber Publishing Co.:
Cleveland, OH, 1975; Vol. 6, p 302. (b) Creatine and Creatine Phosphate:
Scientific and Clinical PerspectiVes; Conway, M. A., Clark, J. F., Eds.;
Academic Press: San Diego, CA, 1996.
10.1021/ol034319a CCC: $25.00 © 2003 American Chemical Society
Published on Web 07/03/2003

Our multienzyme system toward sialyllactosides is based
on a new ex ViVo pathway with the CP-CK energy system.13
As shown in Scheme 1, the biosynthetic pathway consists
of sialyltransferase (SiaT) and five other enzymes for the
Scheme 1. Sialylation Cycle with CP-CK Regeneration
System
Figure 1. Creatine kinase reaction.
that it is a logical extension to apply this energy supply
system to oligosaccharide synthesis, this work is, to our
knowledge, the first demonstration of its applicability to
systems outside the realm of in Vitro protein synthesis. To
demonstrate its efficiency in carbohydrate synthesis, we
applied this phosphagen first in the synthesis of sialyloli-
gosaccharides.
Sialyloligosaccharides play important roles in a variety of
cellular functions.8 For instance, sialyllactose (SL) can
neutralize enterotoxins of various pathogenic microbes,
including Escherichia coli, Vibrio cholerae, and Salmonella
typhimurium.9 It has also been reported that 3′-sialyllactose
(Neu5AcR2, 3Lac, 3′-SL) interferes with colonization of
Helicobacter pylori in human stomach and thereby prevents
or inhibits gastric and duodenal ulcers.10 In addition, it has
been proposed to inhibit immune complex formation for
arthritis treatment by blocking the carbohydrate-binding sites
on IgG Fc fragments.11 To date, commercially available
sialyloligosaccharides are still very expensive due to their
low abundance in natural sources. For example, 3′-sialyl-
lactose and 6′-sialyllactose isolated from bovine colostrum
are sold for $54.75 and $71.50 per 0.5 mg, respectively
(Sigma Chemical Co., 2003). Therefore, great efforts have
been focused on synthesizing sialyloligosaccharides from
cheap and simple starting materials.3e,12
recycling of cytidine 5′-monophospho-N-acetylneuraminic
acid (CMP-Neu5Ac). It starts with N-acetylmannosamine
(ManNAc), pyruvate (Pyr), lactose, CP, and catalytic amounts
of ATP and CMP. CMP is initially converted to CDP by
myokinase (E.C. 2.7.4.3) in the presence of ATP, which is
regenerated from its byproduct ADP in a reaction catalyzed
by CK with CP as the phosphate donor. At the same time,
N-acetylneuraminic acid (Neu5Ac) is synthesized in Vitro
by an aldolase (NanA, E.C. 4.1.3.3)-catalyzed condensation
of ManNAc and pyruvate.14 Neu5Ac is then coupled with
CTP to produce CMP-Neu5Ac by NeuA (CMP-sialic acid
synthatase, E.C. 2.7.7.43).15 To drive this reaction forward,
inorganic phosphatase (PPA, E.C. 3.6.1.1) is incorporated
to hydrolyze the byproduct, pyrophosphate (PPi).16 An R2,3-
SiaT (E.C. 2.4.99.4) will then transfer the sialyl residue from
CMP-Neu5Ac to the acceptor.17 The released CMP can be
rephosphorylated by myokinase and creatine kinase with the
consumption of 2 equiv of CP. Overall, the production of 1
equiv of sialyloligosaccharide requires 1 equiv each of
ManNAc, pyruvate, and acceptor plus 2 equiv of CP.
To optimize the reaction conditions and explore the
impacts of different factors on the reaction system, small-
scale (0.25 mL) reactions for 3′-SL were carried out at 37
°C for 48 h.18 The formation of trisaccharide was monitored
by high-performance capillary electrophoresis (HPCE) using
(6) (a) Bessman S. P.; Carpenter, C. L. Annu. ReV. Biochem. 1985, 54,
831-62. (b) Saks, V. A.; Ventura-Clapier, R.; Aliev, M. K. Biochim.
Biophys. Acta 1996, 1274, 81-88. (c) Bessman, S. P.; Luo, J. Med. Sci.
Symp. Ser. 2000, 14, 17-24.
(7) (a) Adamson, S. D.; Herbert, E.; Godchaux, W. Arch. Biochem.
Biophys. 1968, 125, 671-83. (b) Kigawa, T.; Yabuki, T.; Yoshida, Y.;
Tsutsui, M.; Ito, Y.; Shibata, T.; Yokoyama, S. FEBS Lett. 1999, 442, 15-
19. (c) Szostak, J. W.; Bartel, D. P.; Luisi, P. L. Nature 2001, 409, 387-
390.
(8) (a) Comb, D. G.; Watson, D. R.; Roseman, S. In Sialic Acids:
Chemistry, Metabolism, and Function; Schauer, R., Ed.; Springer-Verlag:
New York, 1982; pp 195-261. (b) Schauer, R.; Kamerling, J. P. In New
ComprehensiVe Biochemistry; 1997; Vol. 29b, pp 243-402.
(9) (a) Nilsson, U. J.; Heerze, L. D.; Liu, Y. C.; Armstrong, G. D.; Palcic,
M. M.; Hindsgaul, O. Bioconjugate Chem. 1997, 8, 466-471. (b) Smith,
D. F. Methods Enzymol. 1982, 83, 241-248. (c) Kruse, S.; Pommerencke,
J.; Kleineidam, R. G.; Roggentin, P.; Schauer, R. Glycoconjugate J. 1998,
15, 769-775.
(13) We tried the cycle with ATP as the direct energy source. However,
the cycle suffers from a low yield due to the product inhibitions of ATP/
ADP and low phosphorylating potential of ATP.
(14) Lilley, G. G.; Barbosa, J. A.; Pearce, L. A. Protein Expression. Purif.
1998, 12, 295-304.
(15) Uchida, Y.; Tsukada, Y.; Sugimori, T. J. Biochem. (Tokyo) 1984,
96, 507-522.
(16) (a) Swartz, J. Nature Biotechnol. 2001, 19, 732-733. (b) Liu, Z.;
Zhang, J.; Chen, X.; Wang, P. G. ChemBioChem 2002, 3, 348-355.
(17) (a) Gilbert, M.; Watson, D. C.; Cunningham, A. M.; Jennings, M.
P.; Young, N. M.; Wakarchuk, W. W. J. Biol. Chem. 1996, 271, 28271-
28276. (b) Gilbert, M.; Cunningham, A.-M.; Watson, D. C.; Martin, A.;
Richards, J. C.; Wakarchuk, W. W. Eur. J. Biochem. 1997, 249, 187-194.
(10) Simon, P. M.; Goode, P. L.; Mobasseri, A.; Zopf, D. Infect. Immun.
1997, 65, 750-757.
(11) (a) Teneberg, S.; Jurstrand, M.; Karlsson, K.-A.; Danielsson, D.
Glycobiology 2000, 10, 1171-1181. (b) Dwek, R. A.; Rademacher, T. W.
PCT Int. Appl. 1992, 31 pp.
(12) Endo, T.; Koizumi, S.; Tabata, K.; Ozaki, A. Appl. Microbiol.
Biotechnol. 2000, 53, 257-261.
2584
Org. Lett., Vol. 5, No. 15, 2003

a UV detector (λ ) 245 nm) after PMP (1-phenyl-3-methyl-
5-pyrazolone) derivatization. The PMP method constitutes
a great advantage over the common reductive-amination
derivatization method, where it avoids acidic desialylation
during the process.19 Although the reported pH optimum of
sialyltransferase is 6.517 and it is suggested that acidic media
is favorable for the creatine phosphate-cleavage reaction
(Figure 1),20 the multienzyme sialylation reaction was
performed at pH 7.5 to prevent a decomposition of activated
neuraminic acid at lower pH. This pH was well tolerated by
NanA (pH optimum 7.7),14 NeuA (pH optimum 9.5),15
myokinase (pH optimum 8.0),21 and PPA (pH optimum
9.1).22 DTT was used to stabilize NanA, NeuA, and SiaT in
the reaction. The ManNAc can be epimerized chemically
from cheaper GlcNAc to reduce the cost in large-scale
production.23
Scheme 3. CUPE Cycle with CP-CK Regeneration Systema
a Abbreviation: GalE, UDP-Gal 4′-epimerase.
PEP was replaced with CP, the galactoside yields were
comparable with the cycles supplied with the old PEP-PK
system (Table 1). These new cycles, especially the CUPE
Besides the sialylation discussed above, the new energy
system can be utilized in other sugar nucleotide regeneration
systems.24 Herein, we coupled this energy source to the
known UDP-Gal regeneration cycles, the KTCU cycle
(Scheme 2) and the CUPE cycle (Scheme 3), for the synthesis
Table 1. Yields of Glycoconjugate Synthesis with Different
Sugar Nucleotide Regeneration Systems
Scheme 2. KTCU Cycle with CP-CK Regeneration Systema
a Abbreviations: GalK, galactokinase, GalPUT, galactose-1-
phosphate uridylyltransferase; GalU, glucose-1-phosphate uridyl-
yltransferase; GalT, galactosyltransferase; Glc-1-P, glucose-1-
phosphate; Gal-1-P, galactose-1-phosphate; UDP-Glc, uridine 5′-
diphospho-R-D-glucose; UDP-Gal, uridine 5′-diphospho-R-D-
galactose.
a Yields in parentheses are isolated yields based on 100 mg-scale
preparative synthesis. Others are milligram-scale reactions and were obtained
by HPCE or HPLC analyses.
of biomedically important galactosides such as R-Gal epitope
and globotriose.25 In our previous studies, PEP was coupled
as a direct energy source to drive the glycosylation. When
cycle, are much more cost effective, because only 1 equiv
of CP was needed in the reaction and another phosphate can
be provided by inexpensive Glc-1-P. Optimization of reaction
conditions was conducted with a chromophore-labeled
lactose [(4,5-dimethoxy-2-nitrophenyl)methyl O-(â-D-galac-
(18) Optimal condition for the multienzyme sialylation was as follows:
acceptor (25 mM), ManNAc (25 mM), pyruvate (125 mM), CP (50 mM),
MnCl2 (10 mM), MgCl2 (5mM), DTT (2mM), NanA (4 U/mL), NeuA (4
U/mL), PPA (4 U/mL), SiaT (0.4 U/mL), CK (20 U/mL), myokinase (20
U/mL), catalytic amounts (2.5 mM) of ATP and CMP in 50 mM HEPES
buffer (pH 7.5).
(23) (a) Muhmaudian, M.; Noble, D.; Drake, C. S.; Middleton, R. F.;
Montgomery, D. S.; Piercey, J. E.; Ramlakhan, D.; Todd, M.; Dawson, M.
J. Enzyme Microb. Technol. 1997, 20, 393-400. (b) Maru, I.; Ohnishi, J.;
Ohta, Y.; Tsukada, Y. Carbohydr. Res. 1998, 306, 575-8.
(24) Freeze, H. H. In Essentials of Glycobiology; Varki, A., Cummings,
R., Esko, J., Freeze, H. H., Hart, G., Marth, J., Eds.; Cold Spring Harbor
Laboratory Press: Cold Spring Harbor, NY, 1999; Chapter 6, p 69-70.
(25) (a) Fang, J.; Li, J.; Chen, X.; Zhang, Y.; Wang, J.; Guo, Z.; Brew,
K.; Wang, P. G. J. Am. Chem. Soc. 1998, 120, 6635-6638. (b) Chen, X.;
Kowal, P.; Hamad, S.; Fan, H.; Wang, P. G. Biotech. Lett. 1999, 21, 1131-
1135. (c) Zhang, J.; Kowal, P.; Fang, J.; Andreana, P.; Wang, P. G.
Carbohydr. Res. 2002, 337, 969-976. (d) Chen, X.; Liu, Z.; Zhang, J.;
Zheng, W.; Kowal, P.; Wang, P. G. ChemBioChem 2002, 3, 47-53.
(19) (a) Kazuaki, K.; Suzuki, S.; Honda, S.; Lee, Y. C. Anal. Biochem.
1991, 199, 256-268. (b) Honda, S.; Togashi, K.; Taga, A. J. Chromatogr.
A 1997, 791, 307-311. (c) Fu, D.; Zopf, D. Anal. Biochem. 1999, 269,
113-123. (d) Suzuki, S.; Tanaka, R.; Takada, K.; Inoue, N.; Yashima, Y.;
Honda, A.; Honda, S. J. Chromatogr. A 2001, 910, 319-29.
(20) pH optimum is ∼6.5 for CK, see: Wyss, M.; Smeitink, J.; Wevers,
R. A.; Wallimann, T. Biochim. Biophys. Acta 1992, 1102, 119-166.
(21) The Enzymes, 2nd ed.; Boyer, P. D., Lardy, H., Myrba¨ck, K., Eds.;
Academic Press: New York, 1962; Vol. 6, pp 139-149.
(22) Josse, J.; Wong, S. C. K. In The Enzymes, 3rd. ed.; Boyer, P. D.,
Ed.; Academic Press: New York, 1971; Vol. IV, pp 499-527.
Org. Lett., Vol. 5, No. 15, 2003
2585

topyranosyl)-(1f4)-â-D-glucopyranoside, LacChr]26 as the
galactosyl acceptor so that RP-HPLC with UV detection (λ
) 348 nm) could be used to monitor the reaction conve-
niently and quantitatively.27
with CK than with PK or AK.34 (4) The cost of AK is much
higher than that of CK and PK, and PEP is the most
expensive reagent among all three regeneration systems
(Sigma Chemical Co., 2003). (5) Inhibition of PK by
pyruvate requires that more kinase be used in the reaction.35
Therefore, the balance of all these factors, in our experience,
is that CP-CK is a more convenient and useful system than
PEP-PK and AcP-AK systems for carbohydrate synthesis,
either on laboratory or industrial scale (Table 2).
The capacity of the CP-CK energy system was explored
in practical synthesis of glycoconjugates. Using recombinant
enzymes expressed in E. coli, we synthesized 3′-SL and
several galactosides in preparative scale (about 0.1∼1 g). In
large-scale synthesis (2.5 mmol, 100 mL) of R-Gal, only 4%
(mol/mol) (1 mM) UDP was applied to decrease the cost
further without significantly affecting the yield (25%). The
enzymatic product of LacChr with R1,3-GalT is a novel
R-Gal derivative, R-GalChr. It can also be regarded as a
potential assay substrate for other transferases.28 Time course
studies indicated that the reaction reached saturation after
36 h. R-GalChr accumulated to 17 mM (72% yield) in the
reaction mixture. The purified enzymatic products were
characterized by NMR and mass spectrometry and confirmed
by comparison with the authentic compounds.3f,25a,c,29
The acetyl phosphate-acetyl kinase (AcP-AK) system
as a practical energy source was first applied in carbohydrate
synthesis by G. M. Whitesides and his colleagues in 1977.30
PEP-PK system was then introduced in carbohydrate
chemistry by the same group in 1982.31 Our approach based
on CP-CK has both advantages and disadvantages relative
to the schemes based on PEP-PK and AcP-AK systems.32
(1) PEP is a stronger phosphorylating agent than PCr and
AcP, so it can be used to drive more thermodynamically
unfavorable reactions.5a (2) The stabilities of PEP and PCr
in solution are much higher than that of AcP. They can be
added in one portion at the beginning of the reaction, while
the AcP must be added continuously.30,33 (3) Since the Km
of ADP for CK is lower than that for PK and AK, it is
possible to use lower concentration of A(T,D)P to reduce
the product inhibition and achieve higher turnover numbers
Table 2. Comparison of ATP Regeneration Systems:
AcP-AK, PEP-PK, and CP-CK Systems
entry
AcP-AK PEP-PK
CP-CK
-43.1
free energy of reagent
(∆G°′, kJ /mol)
-43.1
-61.9
half-life of reagent
0.34
98
12
(days)
Km value of enzyme
(mM, MgADP)
cost of enzyme/reagent
($/kU and $/mmol)
product inhibition on enzyme
(KI, mM)
1.5
90/7.9
-
0.3
0.051
9.8/9.7
-
11.6/38
1
In summary, we report here a novel and versatile phos-
phate donor in glycoconjugate synthesis and demonstrated
its efficiency in syntheses of natural and unnatural oligosac-
charides. In addition, the CP-CK energy system is cost
effective and of great utility in the large-scale production of
carbohydrates. Work to apply this new energy system in our
superbeads technology is in progress.
Acknowledgment. We thank Dr. Ziye Liu for helpful
discussion. This work was supported by a research grant from
the National Institutes of Health (AI 44040).
(26) Cohen, S. B.; Halcomb, R. L. J. Org. Chem. 2000, 65, 6145-6152.
(27) Optimal condition for the enzymatic galactosylation was as fol-
lows: acceptor (25 mM), Glc-1-P (25 mM), CP (25 mM), MnCl2 (10 mM),
MgCl2 (5mM), DTT (2mM), CK (20 U/mL), GalU (4 U/mL), PPA
(4 U/mL), GalT (0.4 U/mL), myokinase (20 U/mL), catalytic amounts (2.5
mM) of UDP-Glc or UDP in 50 mM HEPES buffer (pH 7.5).
(28) Shao, J.; Zhang, J.; Kowal, P.; Lu, Y.; Wang, P. G. Biochem.
Biophys. Res. Commun. 2002, 295, 1-8.
Supporting Information Available: Abbreviations, ex-
perimental procedures and full characterization for oligosac-
charides 3′-sialyllactose, R-Gal trisaccharide, globotriose, and
R-GalChr and 1H NMR and 13C NMR spectra for the
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
(29) (a) Nakamura, T.; Bubb, W. A.; Saito, T.; Arai, I.; Urashima, T.
Carbohydr. Res. 2000, 329, 471-476. (b) Boons, G.-J.; Demchenko, A.
V. Chem. ReV. 2000, 100, 4539-4565.
OL034319A
(30) Pollak, A.; Baughn, R. L.; Whitesides, G. M. J. Am. Chem. Soc.
1977, 99, 2366-2367.
(34) (a) Rose, J. A.; Grumberg-Manago, M.; Korey, S. R.; Ochoa, S. J.
Biol. Chem. 1954, 211, 737-756. (b) McQuate, J. T.; Utter, M. F. J. Biol.
Chem. 1959, 234, 2151-2157. (c) Saks, V. A.; Chernousova, G. B.;
Gukovsky, D. E.; Smirnov, V. N.; Chazov, E. I. Eur. J. Biochem. 1975,
57, 273-290.
(35) Reybard, A. M.; Hass, L. F.; Jacobson, D. D.; Boyer, P. D. J. Biol.
Chem. 1966, 236, 2277-2283.
(31) Hirschbein, B. L.; Mazenod, F. P.; Whitesides, G. M. J. Org. Chem.
1982, 47, 3765-3766.
(32) Shih, Y.-S.; Whitesides, G. M. J. Org. Chem. 1977, 42, 4165-
4166.
(33) (a) DiSabato, G.; Jencks, W. P. J. Am. Chem. Soc. 1961, 83, 4400-
4405. (b) Benkovic, S. J.; Schray, K. J. Biochemistry 1968, 7, 4090-4096.
2586
Org. Lett., Vol. 5, No. 15, 2003