Shikimic acid and quinic acid: Replacing isolation from plant sources with recombinant microbial biocatalysis [4]

Full text of DOI:10.1021/ja9830243

J. Am. Chem. Soc. 1999, 121, 1603-1604
1603
Scheme 1a
Shikimic Acid and Quinic Acid: Replacing Isolation
from Plant Sources with Recombinant Microbial
Biocatalysis
K. M. Draths,† David R. Knop,‡ and J. W. Frost*,†
Departments of Chemistry and Chemical Engineering
Michigan State UniVersity
East Lansing, Michigan 48824-1322
ReceiVed August 24, 1998
Shikimic acid and quinic acid, hydroaromatics metabolically
linked by the common pathway of aromatic amino acid biosyn-
thesis (Scheme 1),1,2 have emerged as essential chiral starting
materials in the synthesis of neuraminidase inhibitors effective
in the treatment of influenza.3 Current isolation of shikimic acid
from the fruit of Illicium plants4 precludes its use in kilogram-
level syntheses required for clinical evaluation of neuraminidase
inhibitors. Quinic acid is more readily available although it is
unclear whether its isolation from Cinchona bark5 constitutes a
dependable source of the multiton quantities of chiral starting
material required for manufacture of a prescription drug. To
improve shikimic acid’s availability, a shikimate-synthesizing
Escherichia coli biocatalyst has been constructed. Unexpectedly,
quinic acid was synthesized in addition to shikimic acid. Elaborat-
ing the mechanism responsible for this contamination has led to
the discovery of a new route for quinic acid biosynthesis.
The genomic portion of shikimate-synthesizing SP1.1/pKD12.112
was constructed by insertion of aroB into the serA locus of E.
coli and disruption of the aroL and aroK loci via successive P1
phage-mediated transductions of aroL478::Tn10 and aroK17::
CmR.6 Shikimic acid accumulates due to the absence of the aroL-
and aroK-encoded isozymes of shikimate kinase while the second
copy of aroB increases the catalytic activity of rate-limiting
3-dehydroquinate (DHQ) synthase.7 Disruption of L-serine bio-
synthesis due to loss of genomic, serA-encoded phosphoglycerate
dehydrogenase ensures that serA-, aroFFBR-, and aroE-encoding
plasmid pKD12.112 is retained by the cell. Carbon flow into the
common pathway is increased due to overexpression of aroFFBR
which encodes a mutant isozyme of DAHP synthase insensitive
to feedback inhibition by aromatic amino acids. Overexpression
of aroE-encoded shikimate dehydrogenase compensates for this
enzyme’s feedback inhibition by shikimic acid.7
a (a) 3-Deoxy-D-arabino-heptulosonic acid 7-phosphate synthase
(aroFFBR); (b) 3-dehydroquinate synthase (aroB); (c) 3-dehydroquinate
dehydratase (aroD); (d) shikimate dehydrogenase (aroE); (e,f) shikimate
kinase (aroK, aroL).
protocatechuic acid on activated carbon during decolorization.
Unfortunately, quinic acid contamination was in excess of what
could be purified away from shikimic acid by crystallization.
Quinic acid formation had to be reduced.
Quinic acid biosynthesis, while widespread in plants, has only
been observed in a single microbe, E. coli AB2848aroD/pKD136/
pTW8090A.8 This heterologous construct expresses quinate
dehydrogenase encoded by the qad locus isolated from Klebsiella
pneumoniae.8 Quinate dehydrogenase-catalyzed oxidation of
quinic acid is driven by catabolic consumption of the resulting
3-dehydroquinic acid via the â-ketoadipate pathway in K.
pneumoniae and other microbes. Reduction of 3-dehydroquinic
acid by quinate dehydrogenase dominates in E. coli AB2848aroD/
pKD136/pTW8090A because of the absence of 3-dehydroquinic
acid catabolism. Quinic acid synthesis in E. coli SP1.1/pKD12.112
thus implicates the existence of an oxidoreductase which reduces
3-dehydroquinic acid.
Because of structural similarities between 3-dehydroshikimic
and 3-dehydroquinic acids (Scheme 1), purified E. coli shikimate
dehydrogenase was incubated with 3-dehydroquinic acid. Quinic
acid formation was observed. The Michaelis constant, Km ) 1.2
mM, and maximum velocity, Vmax ) 0.096 mmol L-1 min-1, for
shikimate dehydrogenase-catalyzed reduction of 3-dehydroquinic
Fed-batch fermentor cultivation of SP1.1/pKD12.112 for 42 h
resulted in the synthesis of 27.2 g/L of shikimic acid, 12.6 g/L
of quinic acid, and 4.4 g/L of 3-dehydroshikimic acid (DHS).
DHS accumulation reflected the expected feedback inhibition of
shikimate dehydrogenase.7 By contrast, quinic acid biosynthesis
was surprising, given the absence in E. coli of quinate dehydro-
genase, an oxidoreductase which interconverts 3-dehydroquinic
and quinic acids. 3-Dehydroshikimic acid was removed from the
fermentation broth by heating followed by adsorbing the resulting
acid to quinic acid compares with Km ) 0.11 mM and Vmax
)
0.11 mmol L-1 min-1 for shikimate dehydrogenase-catalyzed
reduction of 3-dehydroshikimic acid to shikimic acid. To further
explore the role of aroE-encoded shikimate dehydrogenase in the
formation of quinic acid in an intact microbe, aroB was site-
specifically inserted into the serA locus of an E. coli aroD mutant
lacking 3-dehydroquinate dehydratase. The resulting E. coli
QP1.1 was then transformed with plasmid pKD12.112. With the
use of fed-batch fermentor conditions identical to those used for
SP1.1/pKD12.112, QP1.1/pKD12.112 synthesized 60 g/L of
quinic acid along with 2.6 g of 3-dehydroquinic acid (Figure 1)
in 60 h.
† Department of Chemistry.
‡ Department of Chemical Engineering.
(1) Haslam, E. Shikimic Acid: Metabolism and Metabolites; Wiley &
Sons: New York, 1993.
(2) Pittard, A. J. In Escherichia coli and Salmonella: Cellular and
Molecular Biology; Neidhardt, F. C., Ed.; ASM Press: Washington, DC, 1996;
Chapter 28.
(3) (a) Kim, C. U.; Lew, W.; Williams, M. A.; Liu, H.; Zhang, L.;
Swaminathan, S.; Bischofberger, N.; Chen, M. S.; Mendel, D. B.; Tai, C. Y.;
Laver, W. G.; Stevens, R. C. J. Am. Chem. Soc. 1997, 119, 681. (b) Rohloff,
J. C.; Kent, K. M.; Postich, M. J.; Becker, M. W.; Chapman, H. H.; Kelly, D.
E.; Lew, W.; Louie, M. S.; McGee, L. R.; Prisbe, E. J.; Schultze, L. M.; Yu,
R. H.; Zhang, L. J. Org. Chem. 1998, 63, 4545.
(4) Reference 1, pp 40-42.
(5) Reference 1, p 56.
(6) Løbner-Olesen, A.; Marinus, M. G. J. Bacteriol. 1992, 174, 525.
(7) Dell, K. A.; Frost, J. W. J. Am. Chem. Soc. 1993, 115, 11581.
(8) Draths, K. M.; Ward, T. L.; Frost, J. W. J. Am. Chem. Soc. 1992, 114,
9725.
10.1021/ja9830243 CCC: $18.00 © 1999 American Chemical Society
Published on Web 02/05/1999

1604 J. Am. Chem. Soc., Vol. 121, No. 7, 1999
Communications to the Editor
Table 1. Products Synthesized by E. coli SP1.1/pKD12.112 as a
Function of Time and D-Glucose Addition Parameters
Legend:
Kc ) 0.1
Kc ) 0.8
(QA)
SAa
QA
DHS
SA
QA
DHS
12 h
18 h
24 h
30 h
36 h
42 h
1.1
5.3
11.4
17.1
23.1
27.2
0.0
2.5
5.7
8.3
10.8
12.6
0.3
1.2
2.2
2.7
4.2
4.4
1.0
3.1
6.4
10.9
15.7
20.2
0.0
0.0
0.8
1.3
1.8
1.9
0.2
0.6
1.2
2.2
3.5
4.6
Figure 1. Cultivation of QP1.1/pKD12.112 under fed-batch fermentor
conditions.
a Concentrations in g/L of shikimic acid (SA), quinic acid (QA),
and 3-dehydroshikimic acid (DHS).
Legend:
The rate of D-glucose addition, and thus D-glucose availability,
in all fermentation runs was controlled by the proportional-
integral-derivative (PID) setting gain (Kc). For example, synthesis
by E. coli SP1.1/pKD12.112 of the mixture of shikimic and quinic
acids (Table 1) employed a PID setting of Kc ) 0.1. Increasing
D-glucose availability by increasing the PID setting to Kc ) 0.8
resulted (Table 1) in a drastic reduction in the formation of quinic
acid throughout the entire fermentation. After 42 h of cultivation,
E. coli SP1.1/pKD12.112 synthesized 20.2 g/L of shikimic acid,
4.6 g/L of DHS, and only 1.9 g/L of quinic acid. The decrease in
the synthesized titers of shikimic acid is consistent with the known
impact of increased D-glucose availability on the concentration
and yield of L-phenylalanine synthesized by E. coli.11 More
importantly, improvement of the shikimate:quinate molar ratio
from 2.4:1 (Kc ) 0.1) to 11.8:1 (Kc ) 0.8) allowed quinic acid to
be completely removed during crystallization of shikimic acid.
In addition to the neuraminidase inhibitor GS4101, quinic acid
has been used in the synthesis of antitumor agent esperamicin-
A, immunosuppressant FK506, and a variety of other molecules.12
Shikimic acid has recently been used as the starting point for
synthesis of a large combinatorial library of molecules.13 Microbial
syntheses of shikimic acid and quinic acid reported in this account
are already capable of supplanting isolation from plant sources.
As for the future, comparison of the theoretical maximum yield14
(43%) for microbial synthesis of shikimic and quinic acids from
D-glucose with the yields achieved thus far for microbial synthesis
of shikimic acid (14%) and quinic acid (23%) suggests that sizable
increases in yields and titers are possible. Such improvements
could lead to a pronounced expansion in the synthetic utilization
of the highly functionalized, six-membered carbocyclic rings and
multiple asymmetric centers found in shikimic and quinic acids.
(SA)
(GA)
Figure 2. Equilibration of shikimic acid and quinic acid catalyzed by
SP1.1/pKD12.112.
Minimizing the cytosolic concentration of 3-dehydroquinic acid
appeared to be a reasonable strategy for reducing quinic acid
contamination of the shikimic acid synthesized by E. coli SP1.1/
pKD12.112. The aroD gene encoding 3-dehydroquinate dehy-
dratase was consequently localized on plasmid pKD12.152A along
with aroE, aroFFBR, and serA. However, attendant amplified
expression of 3-dehydroquinate dehydratase did not reduce the
levels of quinic acid contamination in the shikimic acid synthe-
sized by SP1.1/pKD12.152A under fed-batch fermentor conditions
identical to those employed for SP1.1/pKD12.112. This suggested
that quinic acid formation may not result from de novo biosyn-
thesis but rather from equilibration of initially synthesized
shikimic acid.
Equilibration of quinic and shikimic acids has previously been
examined in cell-free extracts of K. pneumoniae.9 To be relevant
to quinic acid formation in E. coli SP1.1/pKD12.112, the common
pathway must be able to operate in vivo in the reverse of its
normal biosynthetic direction. To test this possibility, SP1.1/
pKD12.112 cells collected from the fermentor after 24 h were
washed, resuspended in fresh minimal salts medium containing
shikimic acid, and then shaken. Formation of quinic acid and
3-dehydroshikimic acid along with a corresponding decrease in
shikimic acid concentration (Figure 2) indicated that SP1.1/
pKD12.112 can catalyze formation of quinic acid from initially
synthesized shikimic acid.
Acknowledgment. Research was supported by the National Science
Foundation and United States Department of Agriculture. A construct
carrying aroL478::Tn10 and aroK17::CmR was provided by Professor
M. G. Marinus.
The possible role of shikimic acid transport from the culture
medium into the microbial cytoplasm during observed equilibra-
tion pointed to a strategy for minimizing quinic acid contamina-
tion. Shikimic acid transport10 in E. coli may be an evolutionary
vestige of a previous ability to catabolize shikimic and quinic
acids as sole sources of carbon for growth and metabolism. Since
utilization of non-glucose carbon sources is often subject to
catabolite repression, increasing D-glucose availability might
repress shikimic acid transport, thereby minimizing formation of
quinic acid.
Supporting Information Available: Fermentation and purification
methods (PDF). This material is available free of charge via the Internet
at http://pubs.acs.org.
JA9830243
(11) (a) Konstantinov, K. B.; Nishio, N.; Yoshida, T. J. Ferment. Bioeng.
1990, 70, 253. (b) Konstantinov, K. B.; Nishio, N.; Seki, T.; Yoshida, T. J.
Ferment. Bioeng. 1991, 71, 350.
(12) Barco, A.; Benetti, S.; De Risi, C.; Marchetti, P.; Pollini, G. P.;
Zanirato, V. Tetrahedron Asymmetry 1997, 8, 3515.
(9) Mitsuhashi, S.; Davis, B. D. Biochim. Biophys. Acta 1954, 15, 268.
(10) (a) Pittard, J.; Wallace, B. J. J. Bacteriol. 1966, 92, 1070. (b) Brown,
K. D.; Doy, C. H. Biochim. Biophys. Acta 1976, 428, 550.
(13) Tan, D. S.; Foley, M. A.; Shair, M. D.; Schreiber, S. L. J. Am. Chem.
Soc. 1998, 120, 8565.
(14) Draths, K. M.; Frost, J. W. J. Am. Chem. Soc. 1995, 117, 2395.