One-pot four-enzyme synthesis of ketoses with fructose 1,6-bisphosphate aldolases from Staphylococcus carnosus and rabbit muscle

Article abstract of DOI:10.1016/j.carres.2012.05.007

By the action of d-fructose 1,6-bisphosphate aldolases (FruA) from rabbit muscle and Staphylococcus carnosus, various ketoses were synthesized from glyceraldehydes or other aliphatic aldehydes as acceptors in a one-pot four-enzyme system.

Full text of DOI:10.1016/j.carres.2012.05.007

Carbohydrate Research 357 (2012) 143–146
Contents lists available at SciVerse ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
Note
One-pot four-enzyme synthesis of ketoses with fructose 1,6-bisphosphate
aldolases from Staphylococcus carnosus and rabbit muscle
Zijie Li a,c, Li Cai b, , Mohui Wei , Peng George Wang
⇑
c
c,
⇑
a
National Glycoengineering Research Center, The State Key Laboratory of Microbial Technology, Shandong University, Jinan, Shandong 250100, China
Department of Chemistry, University of South Carolina Salkehatchie, 807 Hampton St., Walterboro, SC 29488, USA
Department of Chemistry, Georgia State University, Atlanta, GA 30302, USA
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
By the action of D-fructose 1,6-bisphosphate aldolases (FruA) from rabbit muscle and Staphylococcus
Received 29 February 2012
Received in revised form 4 May 2012
Accepted 7 May 2012
carnosus, various ketoses were synthesized from glyceraldehydes or other aliphatic aldehydes as
acceptors in a one-pot four-enzyme system.
Ó 2012 Elsevier Ltd. All rights reserved.
Available online 15 May 2012
Keywords:
Enzymatic synthesis
Aldolase
Ketose
FruA
Staphylococcus carnosus
D
-Fructose 1,6-bisphosphate aldolase, FruA, catalyzes the
serve as feasible routes to DHAP on a large scale.11 Among the
existing routes toward DHAP, the most elegant one is to combine
the DHAP synthesis steps with the aldol reaction catalyzed by
aldolase. Fessner and Sinerius first developed a one-pot system
in which DHAP was generated from the oxidation of relatively
pivotal step of glycolysis in vivo, which is the reversible
degradation of
-phosphate and dihydroxyacetone phosphate (DHAP). However,
it is worth noting that the equilibrium constant for this reaction
strongly favors the synthesis direction, which makes FruA
synthetically useful as a biocatalyst. The FruA isolated from rabbit
muscle (RAMA) is the most widely investigated DHAP-dependent
aldolase for preparative synthesis, due to its commercial availability,
high activity, and relaxed aldehyde substrate specificity.
Unfortunately, it was reported that RAMA suffered from low
D-fructose 1,6-bisphosphate to D-glyceraldehyde
3
cheap L-glycerol 3-phosphate (L-GP) by glycerol phosphate oxidase
1
12
(GPO). In our previous work, we utilized this strategy and
developed a one-pot four-enzyme system which can produce
1
3,14
sugar molecules directly.
the action of -rhamnulose 1-phosphate aldolase (RhaD) and
-fuculose 1-phosphate aldolase (FucA), we successfully synthe-
sized four important rare sugars from a much cheaper starting
Employing this system and through
2
–4
L
L
5
operational stability and loss of activity during synthesis. In
contrast, FruA from bacterial sources, such as FruA from Staphylococcus
1
3,14
material, racemic DL-glycerol 3-phosphate (DL-GP).
In the pres-
5,6
carnosus (FruAS.car), is exceptionally temperature and pH stable. In
Sheldon’s work, the activities and stabilities of two bacterial FruA from
S. carnosus and Staphylococcus aureus were compared with RAMA to
investigate their scope of synthetic application.7
While FruA aldolases can tolerate different aldehydes as
acceptors, they, like other DHAP-dependent aldolases, are very
specific for the donor substrate DHAP. Unfortunately, DHAP is
ent study, both RAMA and FruAS.car were studied in the preparative
synthesis of ketoses using the one-pot four-enzyme system
(Scheme 1).
To obtain the enzyme FruAS.car, the gene fda encoding FruA from
S. carnosus was amplified by PCR using plasmid pKKfda as the
template. The fda gene amplified was digested with BamHI and
XhoI then ligated into the pET-28a plasmid with the same enzymes
digested. The recombinant plasmid pET28a-fda was then
transformed into Escherichia coli BL21 (DE3) to express FruAS.car
rather expensive and unstable, which greatly hampers their
application in large-scale synthesis.8
–10
A number of chemical
2
+
and enzymatic approaches have been developed for the
preparation of DHAP, however, most of the approaches do not
and the target protein was purified by Ni -NTA column (Fig. 1)
(see Supplementary data for detailed procedures of cloning and
purification of FruAS.car).
To investigate the stereoselectivity of FruAS.car toward
-glyceraldehydes, -fructose and -sorbose were synthesized
accordingly through the one-pot system. Unlike RhaD and FucA,
D- and
⇑
Corresponding authors. Tel.: +1 843 549 6314x381; fax: +1 843 549 6007 (L.C.);
tel.: +1 404 413 3591 (P.G.W.).
L
D
L
E-mail addresses: CAILI@mailbox.sc.edu (L. Cai), pwang11@gsu.edu (P.G. Wang).
0
008-6215/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.carres.2012.05.007

1
44
Z. Li et al. / Carbohydrate Research 357 (2012) 143–146
OH
O
O
GPO
2
-
O PO
OH
2-
3
O PO
OH
3
H
R
DL-glycerol 3-phosphate
DL-GP
DHAP
aldehydes
O2
H2O2
catalase
Oxygen supply
1
2
FruA
rt., 22 h
pH = 7.0
H O
2
O
OH
O
OH
R
AP
HO
H O PO
2 3
R
3
7 °C, 22 h
pH = 4.7
OH
Ketose
OH
GPO = glycerol phosphate oxidase;
AP = acid phosphatase
Scheme 1. One-pot four-enzyme synthesis of ketoses with FruA (FruAS.car and RAMA).
temperature stability as FruAS.car under the reaction conditions.
However, when pentanaldehyde was used as the acceptor, a
decreased yield was observed because of its poor water solubility.
Thus a cosolvent DMSO at 10% (v/v) concentration was added
into the reaction mixture to enhance its solubility. It is worth noting
that the generally low yields for all aliphatic aldehydes are
probably owing to the absence of hydroxyl groups compared with
glyceraldehydes. Less favorable interactions resulted in lower yields
because it was reported earlier that aldehydes which carry a
hydroxyl function had higher affinity to and correct recognition
by the biocatalyst, and formed thermodynamically stable cyclic
1
5
products.
Furthermore, the stereoselectivities of FruAS.car and RAMA are
both very high, as the configuration of vicinal diols produced from
the aldol addition with aliphatic aldehydes follows that of the
glyceraldehydes: only one diastereomer was isolated for each
reaction and the products were verified by NMR (see
7
,16–18
Supplementary data).
These results were consistent with
Sheldon’s work, although they directly used DHAP as the donor
and preparative scale synthesis was only achieved with butanal.7
In summary, we proceeded with the one-pot four-enzyme
system and synthesized different ketoses with FruAS.car and RAMA
through the use of glyceraldehydes and a series of aliphatic
aldehydes as acceptors. RAMA demonstrated similar activity as
FruAS.car under the reaction conditions. The absence of hydroxyl
functions for aliphatic aldehydes explains relatively low yields.
Nevertheless, the one-pot system avoids the use of expensive
DHAP and, cheap starting material DL-glycerol 3-phosphate was
adopted to greatly decrease the synthetic cost. Considering the
broad applications of FruA in synthetic chemistry, we believe this
approach could ultimately contribute to the synthesis of other
carbohydrates and their derivatives with biological functions.
Figure 1. SDS–PAGE analysis of FruAS.car expression and purification. Lanes: 1,
whole cells not induced; 2, whole cells induced for 20 h; 3, cell lysate; 4, purified
FruAS.car; 5, protein marker.
which generate a mixture of diastereomers in our previous study
and the stereoselectivities of which were influenced by the
configuration of acceptors,1
stereoselectivity toward
3,14
FruAS.car demonstrated excellent
-glyceraldehyde. Only a single
product was isolated for each reaction ( -fructose and -sorbose,
respectively). Notably, FruAS.car seems to display a discrimination
toward - and -glyceraldehydes since a much higher yield was
obtained for -glyceraldehyde reaction (60%) compared with the
-isomer (28%) (Table 1). It is highly possible that the 2(R)-hydroxy
of the aldehyde acceptor is favorable for enzyme recognition and
D- and L
D
L
D
L
D
L
5
thus lower yield was observed for
L
-glyceraldehyde.1 This
1. Experimental section
hypothesis could be further evidenced by the low yields obtained
when aliphatic aldehydes (absence of 2-hydroxy) were used as
acceptors (Table 1).
1.1. Bacterial strains, plasmids, and materials
As previously reported, FruA has a flexible substrate specificity
and accepts various aldehydes. We thus proceeded to utilize this
property and successfully synthesized a number of ketoses using
simple aliphatic aldehydes as acceptors in the one-pot reaction
system (Table 1). The results clearly indicate that the yields are
quite similar for reactions of which acetaldehyde, propionaldehyde,
and butyraldehyde are acceptors. Meanwhile, it demonstrates that
FruAS.car and RAMA have very similar catalytic activities. A possible
explanation could be that RAMA displays comparable pH and
Oligonucleotide primers were synthesized at Invitrogen (CA,
USA). Pfx DNA polymerase was purchased from Invitrogen.
Restriction enzymes and T4 ligase were purchased from Fermentas
(MBI, Canada). Escherichia coli DH5a [lacZDM15 hsdR recA] was
purchased from Gibco-BRL (Gaithersburg, MD). E. coli BL21 (DE3)
À
À
À
B
[F ompT hsdSB(r
B
m ) gal dcm (DE3)] and plasmid pET-28a were
purchased from Novagen (Carlsbad, CA). Plasmid pKKfda containing
fda gene was a kind gift from Professor Wolf-Dieter Fessner.
DL-Glycerol 3-phosphate magnesium salt,
D-glyceraldehyde,

Z. Li et al. / Carbohydrate Research 357 (2012) 143–146
145
Table 1
One-pot four-enzyme synthesis of ketoses with FruAS.car and RAMA
Enzyme
Acceptor
Product
Yield (%)
O
OH
HO
OH
OH
FruAS.car
D
-Glyceraldehyde
60
28
OH OH
D-fructose
O
OH
HO
FruAS.car
L-Glyceraldehyde
OH OH
L-sorbose
FruAS.car
RAMA
Acetaldehyde
Acetaldehyde
O
OH
33
29
HO
OH
-deoxy-D-threo-2-pentulose
OH
5
5
5
FruAS.car
RAMA
Propionaldehyde
Propionaldehyde
O
37
34
HO
OH
-threo-2-hexulose
OH
,6-dideoxy-
D
FruAS.car
RAMA
Butyraldehyde
Butyraldehyde
O
40
38
HO
OH
,6,7-trideoxy-D-threo-2-heptulose
OH
FruAS.car
RAMA
Pentanaldehyde
Pentanaldehyde
O
23
13
HO
OH
,6,7,8-tetradeoxy-D-threo-2-octulose
5
acetaldehyde, propionaldehyde, butyraldehyde, pentanaldehyde,
glycerol phosphate oxidase, catalase, aldolase from rabbit muscle
yellow syrup which was further purified by Bio gel P-2 column. After
purification, 108 mg -fructose (60% yield based on
-glyceraldehyde) and 50.9 mg -sorbose (28% yield based on
-glyceraldehyde) were obtained, respectively.
D
(
RAMA), acid phosphatase from sweet potato, isopropyl-1-thio-
b- -galactopyranoside (IPTG), imidazole, kanamycin from
Sigma–Aldrich (St. Louis, MO). -Glyceraldehyde was purchased
D
L
D
L
L
2+
from Capot Chemical Co. (Hangzhou, China). Ni -NTA column was
purchased from Qiagen (Hilden, Deutschland). Amicon Ultra
centrifugal filter (10 K) was purchased from Millipore (Billerica,
MA). Bio gel P-2 gel was purchased from Bio-Rad Laboratories, Inc.
1.3. One-pot four-enzyme synthesis of ketoses with FruAS.car
and RAMA using aliphatic aldehydes
To a solution of DL-glycerol 3-phosphate magnesium salt
(
Hercules, CA). Pierce BCA Protein Assay Kit was purchased from
(457.32 mg, 2 mmol) in 6.86 mL ddH
aliphatic aldehyde (1.5 mmol) at pH 7.0, glycerol phosphate
oxidase (70 U, 2 mg), catalase (1000 U, 1.18 L), and FruA (FruAS.car
or RAMA) (final concentration 0.5 mg/mL). For pentanaldehyde,
1 mL DMSO was added to the reaction mixture. ddH O was added
2
O were added corresponding
Thermo Scientific (IL, USA).
l
1
.2. One-pot four-enzyme synthesis of
D-fructose (or
L-sorbose)
with FruAS.car using
D
-glyceraldehyde (or -glyceraldehyde) as
L
2
the acceptor
to bring the total volume to 10 mL if necessary. The mixture was
shaken at rt for 22 h and the reaction was monitored by TLC
To a solution of DL-glycerol 3-phosphate magnesium salt
548.78 mg, 2.4 mmol) in 6.86 mL ddH O were added -glyceralde-
hyde (or -glyceraldehyde) (2 mL, 0.5 M, 1.0 mmol) at pH 7.0,
glycerol phosphate oxidase (70 U, 2 mg), catalase (1000 U,
.18 L), and FruAS.car (final concentration 0.5 mg/mL). ddH O was
(developed by nBuOH/AcOH/H
anisaldehyde sugar stain). The pH was then adjusted to pH ꢀ5 with
6 M HCl and 11 L acid phosphatase (18 U) was added and the
mixture was shaken at 37 °C for 24 h. After cooling to rt, the pH
wasadjusted to 7.0 with1 M NaOHand themixture wasdiluted with
methanol. The solution was filtered through Celite and washed with
methanol. The filtrate was concentrated under reduced pressure and
the residue was purified by silica gel column chromatography
(dichloromethane/methanol: 20/1 (v/v)) to afford each pure
2
O: 2/1/1 (v/v/v) and stained with
(
2
D
L
l
1
l
2
added to bring the total volume to 10 mL if necessary. The mixture
was shaken at rt for 22 h and the reaction was monitored by TLC
(
developed by nBuOH/AcOH/H
anisaldehyde sugar stain). The pH was then adjusted to pH ꢀ5 with
M HCl and 11 L acid phosphatase (18 U) was added and the
2
O: 2/1/1 (v/v/v) and stained with
6
l
product. Yields were calculated based on L-glycerol 3-phosphate
mixture was shaken at 37 °C for 24 h. After cooling to rt, the pH
was adjusted to 7.0 with 1 M NaOH and the mixture was diluted
with methanol. The solution was filtered through Celite and washed
with methanol. The filtrate was concentrated under reduced
pressure and the residue was purified by silica gel column
(1 mmol) since aliphatic aldehydes were used in excess.
Acknowledgements
P.G.W. acknowledges NIH (R01 HD061935 and R01 GM085267)
for financial support. P.G.W. acknowledges Professor Wolf-Dieter
2
chromatography (EtOAc/iPrOH/H O: 9/3/1 (v/v/v)) to afford a pale

1
46
Z. Li et al. / Carbohydrate Research 357 (2012) 143–146
Fessner for the plasmid pKKfda. L.C. thanks University of South
Carolina and USC Salkehatchie for financial support.
5. Brackamp, H. P.; Kula, M. R. Tetrahedron Lett. 1990, 31, 7123–7126.
6
7
.
.
Brackamp, H. P.; Kula, M. R. Appl. Microbiol. Biotechnol. 1990, 34, 287–291.
Schoevaart, R.; Van Rantwijk, F.; Sheldon, R. A. Tetrahedron: Asymmetry 1999,
1
0, 705–711.
8
9
.
.
Breuer, M.; Hauer, B. Curr. Opin. Biotechnol. 2003, 14, 570–576.
Sugiyama, M.; Hong, Z.; Greenberg, W. A.; Wong, C. H. Bioorg. Med. Chem. 2007,
Supplementary data
1
5, 5905–5911.
Supplementary data (cloning, expression and purification of
FruAS.car and NMR spectra of ketoses) associated with this article
can be found, in the online version, at http://dx.doi.org/10.1016/
j.carres.2012.05.007.
10. Sanchez-Moreno, I.; Garcia-Garcia, J. F.; Bastida, A.; Garcia-Junceda, E. Chem.
Commun. (Cambridge, UK) 2004, 1634–1635.
1
1. Schumperli, M.; Pellaux, R.; Panke, S. Appl. Microbiol. Biotechnol. 2007, 75,
3–45.
12. Fessner, W.-D.; Sinerius, G. Angew. Chem., Int. Ed. 1994, 33, 209–212.
3
1
3. Li, Z.; Cai, L.; Qi, Q.; Styslinger, T. J.; Zhao, G.; Wang, P. G. Bioorg. Med. Chem. Lett.
2
011, 21, 5084–5087.
References
1
1
1
4. Li, Z.; Cai, L.; Qi, Q.; Wang, P. G. Bioorg. Med. Chem. Lett. 2011, 21, 7081–7084.
5. Fessner, W.-D.; Walter, C. Top. Curr. Chem. 1997, 184, 97–194.
6. Schoevaart, R.; van Rantwijk, F.; Sheldon, R. A. Chem. Commun. (Cambridge, UK)
1.
2.
3.
Goldberg, R. N.; Tewari, Y. B. J. Phys. Chem. Ref. Data 1995, 24, 1669–1698.
Fessner, W. D.; Helaine, V. Curr. Opin. Biotechnol. 2001, 12, 574–586.
Gefflaut, T.; Blonski, C.; Perie, J.; Willson, M. Prog. Biophys. Mol. Biol. 1995, 63,
1
999, 2465–2466.
1
1
7. Kobori, Y.; Myles, D. C.; Whitesides, G. M. J. Org. Chem. 1992, 57, 5899–5907.
8. van Herk, T.; Hartog, A. F.; Schoemaker, H. E.; Wever, R. J. Org. Chem. 2006, 71,
3
01–340.
4
.
Bednarski, M. D.; Simon, E. S.; Bischofberger, N.; Fessner, W.-D.; Kim, M. J.; Lees,
6
244–6247.
W.; Saito, T.; Waldmann, H.; Whitesides, G. M. J. Am. Chem. Soc. 1989, 111,
6
27–635.