Creation of a shikimate pathway variant

Article abstract of DOI:10.1021/ja049730n

The competition between the Escherichia coli carbohydrate phosphotransferase system and 3-deoxy-d-arabino-heptulosonate 7-phosphate (DAHP) synthase for phosphoenolpyruvate limits the concentration and yield of natural products microbially synthesized via the shikimate pathway. To circumvent this competition for phosphoenolpyruvate, a shikimate pathway variant has been created. 2-Keto-3-deoxy-6-phosphogalactonate (KDPGal) aldolases encoded by Escherichia coli dgoA and Klebsiella pneumoniae dgoA are subjected to directed evolution. The evolved KDPGal aldolase isozymes exhibit 4-8-fold higher specific activities relative to that for native KDPGal aldolase with respect to catalyzing the condensation of pyruvate and d-erythrose 4-phosphate to produce DAHP. To probe the ability of the created shikimate pathway variant to support microbial growth and metabolism, growth rates and synthesis of 3-dehydroshikimate are examined for E. coli constructs that lack phosphoenolpruvate-based DAHP synthase activity and rely on evolved KDPGal aldolase for biosynthesis of shikimate pathway intermediates and products. Copyright

Full text of DOI:10.1021/ja049730n

Published on Web 05/12/2004
Creation of a Shikimate Pathway Variant
Ningqing Ran, K. M. Draths, and J. W. Frost*
Department of Chemistry, Michigan State UniVersity, East Lansing, Michigan 48824
Received January 15, 2004; E-mail: frostjw@cem.msu.edu
Scheme 1 a,b
Phosphoenolpyruvate is a substrate for the first committed step
in the shikimate pathway (Scheme 1) and is also used by the
carbohydrate phosphotransferase (PTS) system for microbial trans-
port and phosphorylation of glucose.1 The resulting competition
between the shikimate pathway and PTS-mediated glucose transport
for cytoplasmic supplies of phosphoenolpyruvate limits the con-
centrations and yields of natural products microbially synthesized
by way of the shikimate pathway. This account explores whether
pyruvate can replace phosphoenolpyruvate in an enzyme-catalyzed
condensation with D-erythrose 4-phosphate to form 3-deoxy-D-
arabino-heptulosonate 7-phosphate (DAHP, Scheme 1). The cen-
terpiece of the successful creation of this shikimate pathway variant
is the directed evolution of 2-keto-3-deoxy-6-phosphogalactonate
(KDPGal, Scheme 1) aldolase.
By catalyzing the reversible cleavage of KDPGal to pyruvate
and D-glyceraldehyde 3-phosphate (G3P, Scheme 1), KDPGal
aldolase enables Escherichia coli to use D-galactonate as a sole
carbon source.2 KDPGal aldolase from Pseudomonas cepacia has
been reported to catalyze the condensation of pyruvate with various
aldehydes including D-erythrose.3 To explore the catalytic activity
of KDPGal aldolase toward phosphorylated D-erythrose 4-phosphate
(E4P, Scheme 1), E. coli dgoA-encoded KDPGal aldolase was
overexpresssed, partially purified, and incubated with pyruvate,
D-erythrose 4-phosphate, 3-dehydroquinate synthase, and 3-dehy-
droquinate dehydratase. Formation of 3-dehydroshikimate in 90%
yield established the ability of KDPGal aldolase to catalyze the
reaction of pyruvate with D-erythrose 4-phosphate as well as the
ability of 3-dehydroquinate synthase to drive this reaction nearly
to completion. Dehydratase-catalyzed dehydration of 3-dehydro-
quinate provides in product 3-dehydroshikimate a chromophore
suitable for continuous spectrophotometric assay.
With KDPGal aldolase-catalyzed condensation of pyruvate and
D-erythrose 4-phosphate established in vitro, attention turned to
gauging the impact of this activity in E. coli CB734, which lacks
all isozymes of DAHP synthase.4 Growth of E. coli CB734 on
glucose-containing minimal salts medium required supplementation
with L-phenylalanine, L-tyrosine, L-tryptophan, and aromatic vita-
mins (entry 1, Table 1). E. coli CB734/pNR7.088 with its plasmid-
encoded E. coli dgoA was able to biosynthesize its own aromatic
vitamins (entry 2, Table 1). Plasmid-encoded Klebsiella pneumoniae
dgoA afforded a 4-fold higher KDPGal aldolase specific activity
in E. coli CB734/pNR6.252 relative to plasmid-encoded E. coli
dgoA in E. coli CB734/pNR7.088 (Table 2). E. coli CB734/
pNR6.252 was able to provide for its own aromatic vitamin and
L-tryptophan requirements (entry 3, Table 1).
a Metabolites: G3P, D-glyceraldehyde 3-phosphate; E4P, D-erythrose
4-phosphate; KDPGal, 2-keto-3-deoxy-6-phosphogalactonate; DAHP, 3-deoxy-
D-arabino-heptulosonate 7-phosphate. bEnzymes (genes): (a) KDPGal
aldolase (dgoA); (b) DAHP synthase (aroF, aroG, aroH); (c) 3-dehydro-
quinate synthase (aroB); (d) 3-dehydroquinate dehydratase (aroD); (e)
shikimate dehydrogenase (aroE).
Table 1. Directed Evolution of KDPGal Aldolase (DgoA)
c
c
c
entry
constructf
M9a
M9b
F
FY
FYW
FYWvitc
d
e
1
2
3
4
5
6
E. coli CB734
-
-
-
-
-
+
+
-
-
-
+
+
+
-
-
+
+
+
+
-
+
+
+
+
+
+
CB734/pNR7.088
CB734/pNR6.252
CB734/pKP01
CB734/pKP02
CB734/pKP03
-
-
-
-
+
+
+
+
+
+
a Contained 0.05 mM IPTG. b Contained 0.2 mM IPTG. c Supplements
added to M9 medium containing 0.2 mM IPTG included: F, L-phenylala-
nine; Y, L-tyrosine; W, L-tryptophan; vit, p-aminobenzoate, p-hydroxyben-
zoate, 2,3-dihydroxybenzoate. d No growth (-). e Growth (+). f All native
and evolved K. pneumoniae dgoA genes were inserted into plasmid
pJF118EH with transcription controlled by a Ptac promoter. All native and
evolved E. coli dgoA genes were inserted into plasmid pTrc99A with
transcription controlled by a Ptrc promoter.
to the mutant dgoA plasmid insert in CB734/pKP02 (entry 5, Table
1) that enabled this construct to grow in the absence of aromatic
amino acid supplements. Subsequent gene shuffling gave the
evolved dgoA plasmid insert in CB734/pKP03 (entry 6, Table 1)
that facilitated growth in the absence of aromatic amino acid
supplements when evolved KDPGal aldolase expression was
reduced by lowering IPTG concentrations.
The seven most active K. pneumoniae mutants and the seven
most active E. coli mutants were selected for characterization. Each
mutant contained 4-9 amino acid substitutions. No insertion or
deletion mutants were found. Two amino acid substitutions (V85A,
V154F) were observed in all of the examined K. pneumoniae dgoA
and E. coli dgoA mutants. KP03-3, the most active evolved K.
pneumoniae KDPGal aldolase, showed a 4-fold higher DAHP
formation specific activity and a 30-fold lower KDPGal cleavage
K. pneumoniae dgoA and E. coli dgoA were subjected to two
rounds of error-prone PCR mutagenesis5 followed by one round
of DNA shuffling.6 CB734/pKP01 (entry 4, Table 1), which carried
a mutant dgoA plasmid insert resulting from the first round of PCR
mutagenesis performed on the K. pneumoniae dgoA plasmid insert
in E. coli CB734/pNR6.252, required only L-phenylalanine supple-
mentation for growth. The second round of PCR mutagenesis led
9
6856
J. AM. CHEM. SOC. 2004, 126, 6856-6857
10.1021/ja049730n CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
Table 2. Specific Activities of Native and Evolved DgoA Isozymes
g/L of 3-dehydroshikimate in 48 h in 5% yield from glucose. Only
a trace amount of 3-dehydroshikimate was synthesized by NR7/
pNR8.074, which expressed plasmid-encoded, native K. pneumoniae
dgoA. E. coli NR7/pEC03-1serA synthesized 12 g/L of 3-dehy-
droshikimate in 5% yield from glucose. For comparison, 2.0 g/L
of 3-dehydroshikimate was synthesized in 0.9% yield by E. coli
NR7/pNR8.075, which expressed plasmid-encoded, native E. coli
dgoA.
With evolved KDPGal aldolase, the first reaction in the shikimate
pathway can consume the pyruvate byproduct instead of competing
for the phosphoenolpyruvate substrate required by PTS-mediated
glucose transport.8 This constitutes a fundamental departure from
all previous strategies employed to increase phosphoenolpyruvate
availability in microbes.9 Beyond increasing the maximum theoreti-
cal yield for 3-dehydroshikimate synthesis from 43 to 86% (mol/
mol),7a a shikimate pathway variant based on condensation of py-
ruvate with D-erythrose 4-phosphate may be important as a theo-
retical construct. Growth environments can be envisioned where
minimizing expenditure of phosphoenolpyruvate by the shikimate
pathway might be a metabolic advantage. The shikimate pathway
variant outlined in this account may thus serve as a model of a
naturally occurring aromatic biosynthetic pathway that remains to
be discovered.
DAHP
assaya
KDPGal
assaya
enzyme
description
K. pneumoniae DgoA
KP03-3
native enzyme
I10V, E71G, V85A,
P106S, V154F, E187D,
Q191H, F196I
native enzyme
F33I, D58N, Q72E,
A75V, V85A, V154F
0.29b
1.30c
77b
2.6c
E. coli DgoA
EC03-1
0.068d
0.56e
6.7d
1.0e
a Specific activity is defined as units of enzyme activity per mg of protein
in crude cell lysate. One unit of activity ) one µmol of DAHP formed or
KDPGal cleaved per minute. See Suporting Information for assay protocols.
Crude cell lysates were prepared from: b E. coli CB734/pNR6.252; c E.
coli CB734/pKP03-3; d E. coli CB734/pNR7.088; e E. coli CB734/pEC03-
1. See Supporting Information for full descriptions of these plasmids.
Acknowledgment. Research was supported by a grant from the
National Institutes of Health and expedited by suggestions made
by Professor Ronald W. Woodard.
Supporting Information Available: Construction of plasmids and
E. coli NR7; enzyme assays; in vitro synthesis of 3-dehydroshikimate;
directed evolution of KDPGal aldolase (PDF). This material is available
free of charge via the Internet at http://pubs.acs.org.
Figure 1. Growth in the absence of aromatic supplements in glucose-
containing minimal salts medium under shake-flask conditions. E. coli
CB734/pNR7.126 expressing native AroFFBR (squares); E. coli CB734/
pEC03-1 expressing evolved E. coli DgoA (circles); E. coli CB734/pKP03-3
expressing evolved K. pneumoniae DgoA (triangles).
References
(1) Postma, P. W.; Lengeler, J. W.; Jacobson, G. R. In Escherichia coli and
Salmonella: Cellular and Molecular Biology, 2nd ed.; Neidhardt, F. C.,
Ed.; ASM Press: Washington, DC, 1996; pp 1149-1174.
specific activity relative to native K. pneumoniae KDPGal aldolase
(Table 2). EC03-1, the most active evolved E. coli KDPGal aldolase,
exhibited an 8-fold higher DAHP formation specific activity and a
7-fold reduced KDPGal cleavage specific activity relative to the
native E. coli KDPGal aldolase (Table 2).
Constructs expressing evolved dgoA were examined for growth
rates and synthesis of 3-dehydroshikimate. E. coli CB734/pEC03-1
and E. coli CB734/pKP03-3 were completely dependent on plasmid-
encoded, evolved DgoA isozymes EC03-1 and KP03-3, respec-
tively, for the formation of DAHP. E. coli CB734/pNR7.126 relied
on plasmid-encoded, feedback-insensitive AroFFBR for DAHP
synthase activity. When cultured under identical conditions where
growth was dependent on de novo synthesis of aromatic amino
acids and aromatic vitamins, E. coli CB734/pEC03-1 and E. coli
CB734/pKP03-3 entered the logarithmic phases of their growths
12 and 36 h, respectively, later than E. coli CB734/pNR7.126
(Figure 1).
Synthesis of 3-dehydroshikimate employed E. coli NR7, which
was constructed from E. coli KL3 using site-specific chromosomal
insertions to inactivate all DAHP synthase isozymes. E. coli KL3
has been extensively used in studies7 examining the impact of
phosphoenolpyruvate availability on the synthesis of 3-dehydro-
shikimate. Constructs were cultured under identical fermentor-
controlled conditions. E. coli NR7/pKP03-3serA synthesized 8.3
(2) (a) Cooper, R. A. Arch. Microbiol. 1978, 118, 199-206. (b) Deacon, J.;
Cooper, R. A. FEBS Lett. 1977, 77, 201-205.
(3) (a) Henderson, D. P.; Cotterill, I. C.; Shelton, M. C.; Toone, E. J. J. Org.
Chem. 1998, 63, 906-907. (b) Cotterill, I. C.; Henderson, D. P.; Shelton,
M. C.; Toone, E. J. J. Mol. Catal. B: Enzym. 1998, 5, 103-111.
(4) E. coli CB734 was obtained from Professor Ronald Bauerle (University
of Virginia).
(5) (a) Leung, D. W.; Chen, E.; Goeddel, D. V. BioTechniques 1989, 1, 11-
15. (b) Cadwell, R. C.; Joyce, G. F. PCR Methods Appl. 1992, 2, 28-33.
(6) (a) Stemmer, W. P. C. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 10747-
10751. (b) Stemmer, W. P. C. Nature 1994, 370, 389-391. (c) Crameri,
A.; Raillard, S.-A.; Bermudez, E.; Stemmer, W. P. C. Nature 1998, 391,
288-291.
(7) (a) Li, K.; Mikola, M. R.; Draths, K. M.; Worden, R. M.; Frost, J. W.
Biotechnol. Bioeng. 1999, 64, 61-73. (b) Yi, J.; Li, K.; Draths, K. M.;
Frost, J. W. Biotechnol. Prog. 2002, 18, 1141-1148. (c) Yi, J.; Draths,
K. M.; Li, K.; Frost, J. W. Biotechnol. Prog. 2003, 19, 1450-1459.
(8) PTS-mediated glucose transport is found in microbes such as E. coli,
Bacillus subtilis, and Streptomyces coelicolor. Microbes that do not utilize
a PTS system and do not expend phosphoenolpyruvate during glucose
transport include Zymomonas mobilis and Saccharomyces cereVisiae.
(9) (a) Glf-mediated glucose transport: Snoep, J. L.; Arfman, N.; Yomano,
L. P.; Fliege, R. K.; Conway, T.; Ingram, L. O. J. Bacteriol. 1994, 176,
2133-2135. (b) Recycling of pyruvate to phosphoenolpyruvate: Patnaik,
R.; Liao, J. C. Appl. EnViron. Microbiol. 1994, 60, 3903-3908. (c) Use
of non-PTS sugars: Patnaik, R.; Spitzer, R. G.; Liao, J. C. Biotechnol.
Bioeng. 1995, 46, 361-370. (d) GalP-mediated glucose transport: Flores,
N.; Xiao, J.; Berry, A.; Bolivar, F.; Valle, F. Nat. Biotechnol. 1996, 14,
620-623. (e) Glucose adjuncts: Li, K.; Frost, J. W. J. Am. Chem. Soc.
1999, 121, 9461-9462.
JA049730N
9
J. AM. CHEM. SOC. VOL. 126, NO. 22, 2004 6857