Experimental and computational investigation of the uncatalyzed rearrangement and elimination reactions of isochorismate

Article abstract of DOI:10.1021/ja056714x

The versatile biosynthetic intermediate isochorismate decomposes in aqueous buffer by two competitive pathways, one leading to isoprephenate by a facile Claisen rearrangement and the other to salicylate via elimination of the enolpyruvyl side chain. Computation suggests that both processes are concerted but asynchronous pericyclic reactions, with considerable C-O cleavage in the transition state but relatively little C-C bond formation (rearrangement) or hydrogen atom transfer to the enolpyruvyl side chain (elimination). Kinetic experiments show that rearrangement is roughly 8-times more favorable than elimination. Moreover, transfer of the C2 hydrogen atom to C9 was verified by monitoring the decomposition of [2-²H]isochorismate, which was prepared chemoenzymatically from labeled shikimate, by ²H NMR spectroscopy and observing the appearance of [3-²H]pyruvate. Finally, the isotope effects obtained with the C2 deuterated substrate are in good agreement with calculations assuming pericyclic reaction mechanisms. These results provide a benchmark for mechanistic investigations of isochorismate mutase and isochorismate pyruvate lyase, the enzymes that respectively catalyze the rearrangement and elimination reactions in plants and bacteria.

Full text of DOI:10.1021/ja056714x

Published on Web 01/25/2006
Experimental and Computational Investigation of the
Uncatalyzed Rearrangement and Elimination Reactions of
Isochorismate
Michael S. DeClue,† Kim K. Baldridge,‡ Peter Kast,† and Donald Hilvert*,†
Contribution from the Laboratory of Organic Chemistry, ETH Zu¨rich,
CH-8093 Zu¨rich, Switzerland, and Organisch-Chemisches Institut, UniVersita¨t Zu¨rich,
Winterthurerstrasse 190, CH-8057 Zu¨rich, Switzerland
Received September 30, 2005; E-mail: hilvert@org.chem.ethz.ch
Abstract: The versatile biosynthetic intermediate isochorismate decomposes in aqueous buffer by two
competitive pathways, one leading to isoprephenate by a facile Claisen rearrangement and the other to
salicylate via elimination of the enolpyruvyl side chain. Computation suggests that both processes are
concerted but asynchronous pericyclic reactions, with considerable C-O cleavage in the transition state
but relatively little C-C bond formation (rearrangement) or hydrogen atom transfer to the enolpyruvyl side
chain (elimination). Kinetic experiments show that rearrangement is roughly 8-times more favorable than
elimination. Moreover, transfer of the C2 hydrogen atom to C9 was verified by monitoring the decomposition
of [2-2H]isochorismate, which was prepared chemoenzymatically from labeled shikimate, by 2H NMR
spectroscopy and observing the appearance of [3-2H]pyruvate. Finally, the isotope effects obtained with
the C2 deuterated substrate are in good agreement with calculations assuming pericyclic reaction
mechanisms. These results provide a benchmark for mechanistic investigations of isochorismate mutase
and isochorismate pyruvate lyase, the enzymes that respectively catalyze the rearrangement and elimination
reactions in plants and bacteria.
benzophenone ketyl. Methylene chloride was dried by distillation from
CaH2. All other chemicals and solvents were reagent grade and used
as received. Chromatographic purifications were performed with
Kieselgel 60, 230-400 mesh silica gel (Fluka). Eluent mixtures are
reported as v/v percentages of the minor constituent in the major
A wide range of aromatic metabolitessincluding 6-hydroxy-
anthranilic acid, naphthoquinones, siderophores, 3-carboxy
aromatic amino acids, and various plant epoxidess can be traced
to isochorismate (1), a key branchpoint intermediate in the
shikimate biosynthetic pathway.1 The uncatalyzed decomposi-
tion of isochorismate has been previously investigated2 and
occurs via two competitive pathways, one leading to 3-carboxy-
phenylpyruvate (3), presumably via the intermediate isoprephen-
ate (2), and the other to salicylate (4) and pyruvate (5) (Scheme
1). Nevertheless, scant attention has been paid to the mechanistic
features of these transformations.
1
constituent. H NMR spectra were referenced to dioxane (C4H8O2) at
3.53 ppm in D2O. 2H NMR spectra were referenced to dioxane-d8
(C4D8O2) at 3.53 ppm in H2O. All other spectra were referenced to the
solvent. KA12/pKAD50/pKS3-02 was obtained by transforming the
Escherichia coli strain KA123,4 with the plasmids pKAD505 and pKS3-
02,6 which confer choramphenicol and kanamycin resistance, respec-
tively.
Here we describe a detailed experimental and computational
study of the thermal decomposition of isochorismate. In addition
to characterizing the reaction manifold kinetically, we have
exploited selectively labeled [2-2H]isochorismate (1a) to probe
the nature of the competing transition states. Together with
theoretical calculations, our data suggest that both elimination
and rearrangement are concerted pericyclic processes with
chairlike geometries. These findings provide a basis for
understanding the mechanisms of the biosynthetic enzymes that
act on isochorismate.
(3aR,4S,6S,6aS)-6-Benzyloxy-2,2-dimethyl-tetrahydro-furo[3,4-
d][1,3]dioxole-4-carboxylic Acid (7). A solution of silver nitrate (20.4
g, 120 mmol) in H2O (29.0 mL) was combined with benzyl 2,3-O-
isopropylidene-R-D-lyxo-pentodialdo-1,4-furanoside (6)7 (14.6 g, 52.3
mmol) dissolved in EtOH (290 mL). A solution of KOH (13.5 g, 240
mmol) in H2O (290 mL) was added dropwise to the resulting mixture
over 1.0 h. After the addition was complete, the reaction mixture was
stirred at ambient temperature for an additional 1.5 h and then filtered
to remove insoluble Ag2O. The mother liquor was concentrated in vacuo
to ca. 250 mL and then extracted with Et2O (3 × 50 mL). The aqueous
layer was separated, acidified to pH 2.0, and extracted again with Et2O
(3 × 150 mL). The combined Et2O layers were dried over MgSO4 and
Materials and Methods
General Methods and Materials. All reactions were carried out in
flame-dried glassware. THF was dried by distillation from sodium
(3) Kast, P.; Asif-Ullah, M.; Hilvert, D. Tetrahedron Lett. 1996, 37, 2691-
2694.
(4) Grisostomi, C.; Kast, P.; Pulido, R.; Huynh, J.; Hilvert, D. Bioorg. Chem.
1997, 25, 297-305.
(5) Dell, K. A.; Frost, J. W. J. Am. Chem. Soc. 1993, 115, 11581-11589.
(6) Schmidt, K.; Leistner, E. Biotechnol. Bioeng. 1995, 45, 285-291.
(7) Brimacombe, J. S.; Hunedy, F.; Tucker, L. C. N. J. Chem. Soc. C 1968,
1381-1384.
† ETH Zu¨rich.
‡ Universita¨t Zu¨rich.
(1) Bentley, R. Crit. ReV. Biochem. Mol. Biol. 1990, 25, 307-384.
(2) Young, I. G.; Batterham, T. J.; Gibson, F. Biochim. Biophys. Acta 1969,
177, 389-400.
9
10.1021/ja056714x CCC: $33.50 © 2006 American Chemical Society
J. AM. CHEM. SOC. 2006, 128, 2043-2051
2043

A R T I C L E S
DeClue et al.
Scheme 1. Thermal Decomposition of Isochorismate
evaporated to yield 13.8 g (90%) of 7 as a white solid: mp 101-103
°C. 1H NMR (300 MHz, CDCl3): δ 1.32 (s, 3H), 1.46 (s, 3H), 4.53 (d,
J ) 11.7 Hz, 1H), 4.69-4.75 (m, 3H), 5.08 (dd, J ) 5.9, 4.2 Hz, 1H),
5.30 (s, 1H), 7.27-7.39 (m, 5H), 9.29 (br s, 1H). 13C NMR (75 MHz,
CDCl3): δ 24.8, 25.8, 69.4, 79.4, 80.4, 84.1, 105.7, 113.6, 128.0, 128.1,
128.5, 136.6, 171.8. HRMS-MALDI+ (m/z): [M + Na]+ calcd for
C15H18NaO6, 317.1001; found, 317.1001.
15-crown-5 (5 drops), the mixture was stirred overnight at room
temperature to furnish a dark solution. The reaction was subsequently
cooled to 0 °C, quenched with aqueous NaH2PO4 (1.0 M, 100 mL),
and then extracted with CHCl3 (3 × 300 mL). The combined CHCl3
layers were washed with H2O (25 mL), dried over MgSO4, and
evaporated to yield a red oil. Purification by flash chromatography with
silica absorbent (EtOAc/hexanes, 1:1) afforded 21.0 g (99%) 10 as a
mixture of diastereomers. 1H NMR (300 MHz, CDCl3): δ 1.27-1.37
(m, 18H), 1.42 (s, 3H), 1.43 (s, 3H), 3.16 (d, J ) 22.8 Hz, 1H), 3.38
(d, J ) 22.8 Hz, 1H), 3.72 (s, 3H), 3.76 (s, 3H), 4.07-4.24 (m, 10H),
4.34-4.68 (m, 8H), 5.00 (s, 1H), 5.01 (s, 1H), 7.27-7.40 (m, 10H).
13C NMR (75 MHz, CDCl3): δ 16.1, 16.2, 24.8, 24.9, 25.9 (br s), 42.1
(d, J ) 131 Hz), 42.6 (d, J ) 131 Hz), 52.4 (br s), 62.6-62.9 (m),
68.3, 68.6, 78.1, 78.3, 79.8, 80.2, 85.2, 85.4, 104.4, 104.8, 112.5 (br
s), 127.7, 127.9, 128.0, 128.1, 128.4, 128.5, 137.1 (br s), 169.4, 169.5.
31P NMR (121.5 MHz, CDCl3): δ 22.69, 22.95. HRMS-ESI+ (m/z):
[M + Na]+ calcd for C22H31D2NaO9P, 497.1883; found, 497.1871.
2-(Diethoxy-phosphoryl)-[3,3-2H2]-3-((3aS,4R,6S,6aS)-6-hydroxy-
2,2-dimethyl-tetrahydro-furo[3,4-d][1,3]dioxol-4-yl)-propionic Acid
Methyl Ester (11). Compound 10 (21.0 g, 44.2 mmol) was dissolved
in methanol (630 mL), and stirred with H2O (32 mL), ammonium
formate (26.5 g, 420 mmol), and 10% Pd/C (14.0 g) at 50 °C for 1 h.
After cooling to room temperature and filtering off the Pd/C, the
solution was evaporated under reduced pressure to yield an oily white
solid. The residue was suspended in EtOAc (400 mL) and filtered to
remove insoluble contaminants. The EtOAc solution was then dried
over MgSO4 and evaporated to afford 16.8 g (99%) of 11 as a mixture
of diastereomers. This material was used in the next step without further
purification. 1H NMR (300 MHz, CDCl3): δ 1.19-1.27 (m, 18H), 1.33
(s, 3H), 1.34 (s, 3H), 3.07 (d, J ) 22.8 Hz, 1H), 3.14 (d, J ) 22.8 Hz,
1H), 3.64 (s, 6H), 3.96-4.12 (m, 10H), 4.45-4.55 (m, 4H), 5.19 (s,
2H). 13C NMR (75 MHz, CDCl3): δ 16.1 (br s), 24.6, 24.7, 25.7, 25.8,
41.8 (d, J ) 131 Hz), 52.3 (br s), 62.5-62.8 (m), 79.6 (br s), 80.3 (br
s), 85.7, 85.9, 100.3, 100.4, 112.1 (br s), 169.2 (br s). 31P NMR (121.5
MHz, CDCl3): δ 22.72, 23.17. HRMS-ESI+ (m/z): [M + Na]+ calcd
for C15H25D2NaO9P, 407.1414; found, 407.1405.
Methyl 3,4-O-Isopropylidene[6,6-2H2]shikimate (12). A solution
of 11 (6.63 g, 17.3 mmol) in dry methanol (100 mL) was added
dropwise to a precooled solution of sodium methoxide, which was
prepared by adding dry methanol (200 mL) to sodium (1.3 g, 56.2
mmol, washed once with hexanes) over 30 min under a flow of nitrogen
while maintaining the temperature at 0 °C. The mixture was then
warmed to room temperature and stirred for 2 h. The reaction was
quenched by addition of a saturated aqueous solution of NaHCO3 (350
mL) and extracted with Et2O (3 × 500 mL). The combined Et2O layers
were washed with saturated aqueous NaCl (50 mL), dried over MgSO4,
and evaporated to provide 2.4 g of a light yellow oil. Purification by
flash chromatography with silica absorbent (EtOAc/hexanes, 3:7)
yielded 1.2 g (31%) of 12 as a clear oil. 1H NMR (300 MHz, CDCl3):
δ 1.41 (s, 3H), 1.45 (s, 3H), 2.22 (br s, 1H), 3.78 (s, 3H), 3.89 (d, J )
7.5 Hz, 1H), 4.09 (apparent t, J ) 7.5 Hz, 1H), 4.75 (dd, J ) 6.2, 3.6
Hz, 1H), 6.93 (d, J ) 3.6 Hz, 1H). 13C NMR (75 MHz, CDCl3): δ
25.7, 27.9, 28.6 (pent, J ) 20.0 Hz), 52.1, 68.5, 72.2, 77.8, 109.6,
130.4, 134.0, 166.6. HRMS-ESI+ (m/z): [M + Na]+ calcd for C11H14D2-
NaO5, 253.1019; found, 253.1013.
(3aR,4S,6S,6aS)-6-Benzyloxy-2,2-dimethyl-tetrahydro-furo[3,4-
d][1,3]dioxole-4-carboxylic Acid Methyl Ester (8). Compound 7 (13.8
g, 47.1 mmol) was stirred with anhydrous MeOH (300 mL) and
chlorotrimethylsilane (8.8 mL, 69.6 mmol) under nitrogen at ambient
temperature for 3 h. The reaction was then treated with a dilute solution
of ammonium hydroxide (1.0 mM, 50 mL) and concentrated in vacuo
to ca. 50 mL. The residue was diluted with H2O (50 mL) and extracted
using Et2O (3 × 125 mL). The combined Et2O layers were dried over
MgSO4 and evaporated to yield 12.5 g of a light yellow oil. Purification
by flash chromatography with silica absorbent (EtOAc/hexanes, 1:1)
1
yielded 10.3 g (71%) of 8 as an off-white solid: mp 54-55 °C. H
NMR (300 MHz, CDCl3): δ 1.29 (s, 3H), 1.43 (s, 3H), 3.82 (s, 3H),
4.50 (d, J ) 11.7 Hz, 1H), 4.65-4.73 (m, 3H), 5.01 (dd, J ) 6.0, 4.5
Hz, 1H), 5.28 (s, 1H), 7.27-7.37 (m, 5H). 13C NMR (75 MHz,
CDCl3): δ 25.0, 25.9, 52.2, 69.3, 79.6, 80.5, 84.0, 105.5, 113.2, 127.8,
127.9, 128.3, 136.7, 167.6. HRMS-ESI+ (m/z): [M + Na]+ calcd for
C16H20NaO6, 331.1158; found, 331.1148.
((3aS,4R,6S,6aS)-6-Benzyloxy-2,2-dimethyl-tetrahydro-furo[3,4-
d][1,3]dioxol-4-yl)-[di-2H]methanol (9). A solution of 8 (15.7 g, 51
mmol) in dry THF (120 mL) was added dropwise over 30 min to a
stirred suspension of lithium aluminum deuteride (2.14 g, 51 mmol)
and THF (410 mL) at 0 °C under a flow of nitrogen. After the addition
was complete the mixture was warmed to room temperature and stirred
overnight. The reaction was then cooled to 0 °C, treated with a saturated
aqueous solution of MgSO4 (20 mL), and extracted using Et2O (3 ×
250 mL). The combined Et2O layers were dried over MgSO4 and
evaporated to yield 13.3 g (93%) of 9 as a white solid. This material
was used in the next step without further purification: mp 84-85 °C.
1H NMR (300 MHz, CDCl3): δ 1.31 (s, 3H), 1.47 (s, 3H), 2.25 (br s,
1H), 4.12 (d, J ) 3.6 Hz, 1H), 4.50 (d, J ) 12.0 Hz, 1H), 4.68 (d,
J ) 6.0 Hz, 1H), 4.70 (d, J ) 11.4 Hz, 1H), 4.81 (dd, J ) 6.2, 3.6 Hz,
1H), 5.15 (s, 1H), 7.27-7.39 (m, 5H). 13C NMR (75 MHz, CDCl3): δ
24.4, 25.8, 60.2 (pent, J ) 21.8 Hz), 68.9, 79.4, 80.2, 85.1, 105.1,
112.6, 127.8, 128.0, 128.3, 137.2. HRMS-ESI+ (m/z): [M + Na]+ calcd
for C15H18D2NaO5, 305.1332; found, 305.1326.
3-((3aS,4R,6S,6aS)-6-Benzyloxy-2,2-dimethyl-tetrahydro-furo-
[3,4-d][1,3]dioxol-4-yl)-2-(diethoxy-phosphoryl)-[3,3-2H2]propion-
ic Acid Methyl Ester (10). Triflic anhydride (9.8 mL, 58.2 mmol)
was added dropwise over 10 min to a stirred solution of 9 (12.6 g,
44.6 mmol) in dry CH2Cl2 (125 mL) and pyridine (7.9 mL, 97.8 mmol)
at -30 °C under a flow of nitrogen. After 15 min at -30 °C the reaction
was quenched by the addition of H2O (20 mL) and diluted with CH2-
Cl2 (150 mL). The CH2Cl2 layer was separated, washed with aqueous
NaH2PO4 (1.0 M, 25 mL), dried over MgSO4, and evaporated to afford
18.5 g (100%) of the triflate as a light pink oil. This intermediate is
highly reactive and should be used as quickly as possibly. The crude
triflate (18.5 g, 44.6 mmol) was dissolved in dry DMF (105 mL) and
added to the ylid of methyl diethylphosphonoacetate, prepared in
advance by adding methyl diethylphosphonoacetate (14.2 mL, 77.6
mmol) in dry DMF (70 mL) to a stirred suspension of sodium hydride
(2.8 g, 69.7 mmol) in dry DMF (170 mL) at 0 °C. After addition of
Methyl [6,6-2H2]Shikimate (13). Compound 12 (950 mg, 4.1 mmol)
was mixed with Dowex-50WX8-200 (1.6 g, washed with methanol three
times and dried under vacuum for 30 min) in methanol (100 mL)
9
2044 J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006

Thermal Decomposition of Isochorismate
A R T I C L E S
containing 5.8 mL of H2O. The mixture was stirred overnight at room
temperature under an atmosphere of nitrogen. The solution was then
filtered and the resin washed with methanol (3 × 5 mL). The combined
methanol layers were evaporated to afford 870 mg of a light yellow
solid. Purification by flash chromatography with silica absorbent
(MeOH/CH2Cl2, 1:9) yielded 720 mg (92%) of 13 as a white solid:
Nucleosil 100-7 C-18 VP 250/21, CH3CN/H2O containing 0.1% TFA,
linear gradient from 5% to 40% CH3CN over 40 min, total flow rate
of 10 mL/min) to give isochorismic acid (retention time 21.7 min) and
chorismic acid (retention time 23.9 min). The corresponding fractions
were lyophilized, affording isochorismic acid (37 mg, 6%) as a white
1
solid and chorismic acid (55 mg, 9%) as an off-white solid. The H
1
NMR spectra of chorismic acid9 (not reported) and isochorismic acid10
mp 113.5-114.5 °C. H NMR (300 MHz, CD3OD): δ 3.68 (dd, J )
1
7.2, 3.9 Hz, 1H), 3.73 (s, 3H), 3.98 (d, J ) 6.9 Hz, 1H), 4.37 (apparent
t, J ) 3.9 Hz, 1H), 6.78 (d, J ) 3.6 Hz, 1H). 13C NMR (75 MHz,
CD3OD): δ 30.8 (pent, J ) 20.0 Hz), 52.4, 67.2, 68.2, 72.5, 130.1,
139.1, 168.7. HRMS-ESI+ (m/z): [M + Na]+ calcd for C8H10D2NaO5,
213.0706; found, 213.0700.
match published data. H NMR (300 MHz, DMSO-d6): δ 4.47 (bs s,
1H), 4.53 (d, J ) 4.7 Hz, 1H), 4.96 (br s, 1H), 5.31 (br s, 1H), 6.24
(dd, J ) 9.3, 4.7 Hz, 1H), 6.35 (dd, J ) 9.3, 5.4 Hz, 1H), 6.96 (d,
J ) 5.4 Hz, 1H). 1H NMR (300 MHz, D2O, pD 7.5, 60 °C): δ 4.49 (d,
J ) 2.4 Hz, 1H), 4.54 (d, J ) 4.8 Hz, 1H), 4.61 (d, J ) 5.1 Hz, 1H),
5.01 (d, J ) 2.4 Hz, 1H), 5.98 (ddd, J ) 9.6, 4.4, 1.2 Hz, 1H), 6.12
(ddd, J ) 9.6, 5.6, 1.2 Hz, 1H), 6.59 (d, J ) 5.6, 1.2 Hz, 1H).
Preparation of the Crude Cell Extract from E. coli Strain KA12/
pKAD50/pKS3-02. A 500 mL Erlenmeyer flask supplied with Miller’s
LB broth8 (100 mL) was sterilized by autoclaving. After cooling, the
broth was supplemented with solutions of kanamycin sulfate in H2O
(15 mg/mL, 200 µL), chloramphenicol in EtOH (30 mg/mL, 200 µL),
tryptophan in H2O (8 mg/mL, 500 µL), and ammonium ferrous sulfate
in H2O (5 mM, 200 µL), and then inoculated with a single colony of
E. coli strain KA123,4 transformed with plasmids pKAD505 and pKS3-
02.6 This preculture was shaken at 230 rpm for 24 h at 37 °C (final
OD ) 2.2 at 600 nm). Six 3 L Erlenmeyer flasks, each containing
Miller’s LB broth (1.0 L), kanamycin sulfate (30 mg), chloramphenicol
(60 mg), tryptophan (40 mg), and an aliquot of a freshly prepared
aqueous solution of ammonium ferrous sulfate (5 mM, 2.0 mL) were
each inoculated with 15 mL of the KA12/pKAD50/pKS3-02 preculture.
The cultures were shaken at 230 rpm for 24 h at 37 °C (final OD )
3.8 at 600 nm). After cooling to room temperature, cells were collected
by centrifugation at 5000 rpm (4200g; 15 min, 4 °C). The cell pellet
was resuspended in Tris-HCl (50 mM, pH 8.0) containing DL-
dithiothreitol (DTT, 1.0 mM). Recentrifugation at 5000 rpm (4200g;
15 min, 4 °C) yielded 33 g of cells. The cell pellet was suspended in
Tris-HCl (120 mL, 50 mM, pH 8.0) containing DTT (1.0 mM), and
the cells were ruptured by passage through a French press (three times).
Cell debris was removed from the lysate by centrifugation at 10 000
rpm (16 420g; 30 min, 4 °C). The supernatant was initially dialyzed at
4 °C against Tris-HCl (2.0 L, 50 mM, pH 8.0) containing DTT (1.0
mM) and a solution of phenylmethanesulfonyl fluoride (50 mg PMSF
dissolved in 1.0 mL of CH3CN) and then exhaustively in the same
buffer lacking PMSF. It was stored at 4 °C prior to use. Note: best
yields of isochorismate in the following procedures were obtained with
freshly prepared cell extract.
Isochorismate (1). The monosodium salt of ATP (2.28 g, 4.14
mmol) was added in portions to Tris-HCl (95 mL, 50 mM, pH 8.0)
while maintaining the pH of the solution between 6.5 and 9.0. To this
solution were added FMN (140 mg, 0.28 mmol), the monopotassium
salt of phosphoenolpyruvate (1.72 g, 8.33 mmol), a solution of KCl
(1.0 M aqueous, 6.9 mL, 6.9 mmol), a solution of MgSO4 (1.0 M
aqueous, 0.69 mL, 0.69 mmol), and shikimic acid (480 mg, 2.76 mmol).
KA12/pKAD50/pKS3-02 extract (35 mL, corresponding to a lysate
fraction originating from ca. 10 g of cells with an estimated total protein
concentration of ca. 40 mg/mL) was added, and the pH of the resulting
mixture was adjusted to 8.0 using NaOH (10.0 N). After addition of
sodium dithionite (270 mg, 1.56 mmol), the reaction mixture was sealed
and gently stirred at 23 °C for 1 h. It was then brought rapidly to about
0 °C and filtered through an Amicon ultrafiltration membrane (10 K
cutoff) at 4 °C, while cooling the collected filtrate in a slurry of ice
and NaCl. When nearly dry, the membrane was washed with Tris-
HCl (20 mL, 50 mM, pH 8.0). The combined filtrate was acidified to
pH 1.5 and extracted with EtOAc (6 × 100 mL). The combined extracts
were dried over Na2SO4 and carefully concentrated in vacuo (maintain-
ing the rotary evaporator bath at ca. 15 °C). The resulting light yellow
oil was purified by preparative reverse phase HPLC (Macherey-Nagel
[2-2H]Isochorismate (1a). Compound 13 (525 mg, 2.76 mmol) was
dissolved in a mixture of THF (20.5 mL), H2O (17.6 mL), and NaOH
(1 M, 2.9 mL, 2.9 mmol). The solution was stirred at 23 °C for 3 h.
After the removal of THF by distillation in vacuo, the aqueous mixture
was lyophilized to yield 545 mg of the labeled sodium shikimate as a
white solid. Using the conditions described above for isochorismate
(1), the crude sodium [6,6-2H2]shikimate (545 mg, 2.76 mmol) was
converted chemoenzymatically into [2-2H]chorismate (46 mg, 7%) and
1
1a (49 mg, 8%). [2-2H]Chorismate. H NMR (300 MHz, CD3OD): δ
4.69 (dt, J ) 11.4, 2.7 Hz, 1H), 4.82 (d, J ) 3.0 Hz, 1H), 4.94 (d, J
) 11.4 Hz, 1H), 5.49 (d, J ) 3.0 Hz, 1H), 5.99 (ddd, J ) 10.1, 2.4,
0.9 Hz, 1H), 6.34 (dd, J ) 10.1, 2.7 Hz, 1H). [2-2H]Isochorismate. 1H
NMR (300 MHz, DMSO-d6): δ 4.53 (d, J ) 4.2 Hz, 1H), 4.82 (br s,
1H), 5.24 (br s, 1H), 6.18 (dd, J ) 9.5, 4.2 Hz, 1H), 6.30 (dd, J ) 9.5,
1
4.5 Hz, 1H), 6.88 (d, J ) 4.5 Hz, 1H). H NMR (300 MHz, D2O, pD
7.5, 60 °C): δ 4.49 (d, J ) 2.8 Hz, 1H), 4.52 (d, J ) 4.5 Hz, 1H),
5.01 (d, J ) 2.8 Hz, 1H), 5.98 (ddd, J ) 9.6, 4.4, 1.2 Hz, 1H), 6.13
2
(ddd, J ) 9.6, 5.6, 1.2 Hz, 1H), 6.59 (dd, J ) 5.6, 1.2 Hz, 1H). H
NMR (61.4 MHz, H2O, pH 6.0, 60 °C): δ 4.64 (br s, 12H).
NMR Kinetics. The reaction of isochorismate and [2-2H]isochoris-
1
mate (20 mM) was monitored at 60 °C by H NMR spectroscopy in
deuterated phosphate buffer (200 mM, pD 7.5) containing dioxane (1.0
mM) as an internal standard. The concentrations of isochorismate,
isoprephenate, 3-carboxyphenylpyruvate, salicylate, and pyruvate were
determined periodically by integration of representative, well-resolved
peaks and comparison to the internal standard. The kinetic data were
analyzed with the program DynaFit.11 The decomposition of [2-2H]-
isochorismate in phosphate buffer (200 mM, pH 6.0, 60 °C) was
similarly monitored by 2H NMR spectroscopy using dioxane-d8 as the
internal standard. All experiments were performed in duplicate.
Isoprephenate (2). 1H NMR (300 MHz, D2O, pD 7.5, 60 °C): δ 2.57-
2.75 (m, 2H), 3.13-3.21 (m, 1H), 4.70-4.75 (m, 1H), 5.72-5.76 (m,
1
2H), 6.42 (d, J ) 2.7 Hz, 1H). [4-2H]Isoprephenate. H NMR (300
MHz, D2O, pD 7.5, 60 °C): δ 2.57-2.75 (m, 2H), 3.13-3.22 (m, 1H),
2
5.73-5.76 (m, 2H), 6.42 (d, J ) 3.3 Hz, 1H). H NMR (61.4 MHz,
H2O, pH 6.0, 60 °C): δ 4.75 (br s, 12H). 3-Carboxyphenylpyruvate
(3). 1H NMR (300 MHz, D2O, pD 7.5, 60 °C): δ 3.17 (s, 2H), 7.14-
7.26 (m, 2H), 7.47-7.63 (m, 2H). 3-Carboxy[4-2H]phenylpyruvate. 1H
NMR (300 MHz, D2O, pD 7.5, 60 °C): δ 3.18 (s, 2H), 7.16 (dd, J )
7.7, 2.0 Hz, 1H), 7.24 (d, J ) 7.8 Hz, 1H), 7.48 (d, J ) 2.0 Hz, 1H).
2H NMR (61.4 MHz, H2O, pH 6.0, 60 °C): δ 7.65 (br s, 12H). Salicylate
(4). 1H NMR (300 MHz, D2O, pD 7.5, 60 °C): δ 6.72-6.79 (m, 2H),
1
7.21-7.27 (m, 1H), 7.62 (dd, J ) 7.7, 1.7 Hz, 1H). Pyruvate (5). H
NMR (300 MHz, D2O, pD 7.5, 60 °C): δ 2.14 (s, 3H). [3-2H]Pyruvate.
1H NMR (300 MHz, D2O, pD 7.5, 60 °C): δ 2.13 (t, J ) 2.4 Hz, 2H).
2H NMR (61.4 MHz, H2O, pH 6.0, 60 °C): δ 2.19 (br s, 12H).
Computational Studies. All calculations were performed using the
GAMESS12 and GAUSSIAN13 series of programs. Both hybrid density
(8) Miller, J. H. A Short Course in Bacterial Genetics. A Laboratory Manual
and Handbook for Escherichia coli and Related Bacteria; Cold Spring
Harbor Laboratory: Cold Spring Harbor, NY, 1992.
(9) Copley, S. D.; Knowles, J. R. J. Am. Chem. Soc. 1987, 109, 5008-5013.
(10) Gould, S. J.; Eisenberg, R. L. Tetrahedron 1991, 47, 5979-5990.
(11) Kuzmic, P. Anal. Biochem. 1996, 237, 260-273.
9
J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006 2045

A R T I C L E S
DeClue et al.
Figure 1. 1H NMR spectroscopic analysis of the nonenzymatic decomposition of isochorismate (20 mM) at 60 °C in 200 mM sodium phosphate buffer in
D2O (pD 7.5) with 1.0 mM dioxane as an internal standard. The signals associated with isochorismate (1), isoprephenate (2), 3-carboxyphenylpyruvate (3),
salicylate (4), and pyruvate (5) are indicated.
functional (B3LYP)14 and Møller-Plesset 2 (MP2)15 theories were
of isoprephenate, salicylate (4) and pyruvate (5). A typical time
considered, using the DZ+(2d,p) basis set.16 Only the former are
presented here since no substantial differences were found. Full
geometry optimizations were performed and uniquely characterized by
course for the disappearance of substrate and for the appearance
of the individual products is shown in Figure 2A. The calculated
rate constants for each step of this process (Scheme 1) are k1 )
calculating and diagonalizing the matrix of energy second derivatives
(3.63 ( 0.09) × 10-4 s-1, k2 ) (5.44 ( 0.01) × 10-5 s-1, and
(Hessian) to determine the number of imaginary frequencies (0 )
k3 ) (4.65 ( 0.22) × 10-5 s-1. Under the experimental
minima; 1 ) transition state). All results include zero-point corrections.
conditions, the pyruvate that is produced in the elimination step
Effects of solvent were considered using the self-consistent polarized
also exchanges with solvent with an apparent rate constant of
k4 ) (3.93 ( 0.64) × 10-4 s-1. As seen from the k1/k3 ratio,
rearrangement is more favorable than elimination by a factor
of ca. 8. For comparison, k1/k3 was previously reported to be
about 4 at 100 °C, based on salicylate concentrations estimated
by fluorescence measurements.2
continuum model COSMO.17 Isotope effects were calculated by
harmonic frequency analysis using QUIVER18 with a frequency scaling
factor of 0.961, a temperature of 333 K, and infinite parabola (KIE/ip
tunneling corrections.19 The diequatorial substrate conformer of iso-
chorismate served as the reference state (see the Supporting Informa-
tion).
[2-2H]Isochorismate Synthesis and Reactivity. To gain
further insight into the elimination reaction, [2-2H]isochorismate
(1a) was prepared chemoenzymatically. As shown in Scheme
2, [6,6-2H2]shikimic acid was synthesized chemically and
converted enzymatically to the selectively labeled isochorismate
derivative.
The required shikimate derivative (13) was prepared from
aldehyde 6, which was obtained in four steps from D-mannose.20
Mild oxidation with Ag2O gave carboxylic acid 7, which was
converted to methyl ester 8. The deuterium label was then
introduced by reducing 8 with LiAlD4. Activation of the primary
alcohol 9 as a triflate, followed by substitution with methyl
diethylphosphonoacetate, provided 10 in nearly quantitative
yield over two steps. After removing the benzyl ether, compound
11 was converted to shikimate derivative 12 via an intramo-
lecular Horner-Wadsworth-Emmons reaction. Subsequent
Results
Kinetics. The thermal decomposition of isochorismate is
readily monitored by 1H NMR spectroscopy. As shown in Figure
1, the reaction proceeds cleanly at 60 °C in 200 mM sodium
phosphate buffer in D2O (pD 7.5) to produce isoprephenate (2),
3-carboxyphenylpyruvate (3), which arises from further reaction
(12) Schmidt, M. W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S. T.; Gordon, M.
S.; Jensen, J. H.; Koseki, S.; Matsunaga, N.; Nguyen, K. A.; Su, S. J.;
Windus, T. L.; Dupuis, M.; Montgomery, Jr., J. A. J. Comput. Chem. 1993,
14, 1347-1363.
(13) Frisch, M. J.; et al. Gaussian03, revision C.01; Gaussian, Inc.: Wallingford,
CT, 2004.
(14) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652.
(15) Møller, C.; Plesset, M. S. Phys. ReV. 1934, 46, 618-622.
(16) Dunning, Jr., T. H.; Hay, P. J. In Modern Theoretical Chemistry; Schaefer,
III, H. F., Ed.; Plenum: New York, 1976; Vol. 3, pp 1-27.
(17) (a) Klamt, A.; Schu¨u¨rmann, G. J. Chem. Soc., Perkin Trans. 2 1993, 799-
805. (b) Baldridge, K. K.; Klamt, A. J. Chem. Phys. 1997, 106, 6622-
6633.
(18) Saunders, M.; Laidig, K. E.; Wolfsberg, M. J. Am. Chem. Soc. 1989, 111,
8989-8994.
(19) Bell, R. P. The Tunnel Effect in Chemistry; Chapman and Hall: New York,
1980; pp 60-63.
(20) (a) Fleet, G. W. J.; Shing, T. K. M. Chem. Commun. 1983, 849-850. (b)
Fleet, G. W. J.; Shing, T. K. M.; Warr, S. M. J. Chem. Soc., Perkin Trans.
1 1984, 905-908.
9
2046 J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006

Thermal Decomposition of Isochorismate
A R T I C L E S
Figure 2. Kinetic analysis of the nonenzymatic decomposition of isochorismate (A) and [2-2H]isochorismate (B) in 200 mM phosphate buffer in D2O at
60 °C. The reactions were monitored by 1H NMR spectroscopy. Relative concentrations of isochorismate (red circles), isoprephenate (blue inverted triangles),
3-carboxyphenylpyruvate (green squares), salicylate (magenta triangles), and pyruvate (black diamonds) are plotted as a function of time. The data were
analyzed with the program DynaFit (ref 11).
Scheme 2. Synthesis of [2-2H]isochorismate (1a)a
a Conditions: (a) AgNO3, KOH, EtOH, H2O, 90%; (b) (CH3)3SiCl, MeOH, 71%; (c) LiAlD4, THF, 0 °C, 93%; (d) (CF3SO2)2O, py, CH2Cl2, -30 °C; (e)
(EtO)2P(O)CH2CO2Me, NaH, DMF, 99% over two steps; (f) Pd/C, MeOH, HCO2NH4, 50 °C, 99%; (g) MeOH, Na0, 0 °C, 31%; (h) Dowex 50WX8-200,
MeOH, H2O, 92%; (i) NaOH, THF, H2O; (j) KA12/pKAD50/pKS3-02, Tris-HCl (pH 8.0), PEP, ATP, FMN, KCl, MgSO4, NaS2O4, 8% over two steps.
acid-catalyzed hydrolysis of the isopropylidene ketal with
Dowex 50W-X8 resin gave methyl [6,6-2H2]shikimate 13, which
afforded selectively labeled sodium shikimate upon saponifica-
tion.
ratio. Extraction of the crude reaction mixture and purification
by preparative reversed-phase HPLC4 affords milligram amounts
of both compounds. In the same way, roughly 50 mg of [2-2H]-
isochorismate was obtained from 545 mg of methyl [6,6-2H2]-
shikimate, corresponding to an overall yield of 8%.
For the enzymatic transformation of shikimate, we modified
a previously engineered E. coli strain, KA12/pKAD50, which
overproduces the biosynthetic enzymes shikimate kinase, EPSP
synthase, and chorismate synthase and efficiently converts
shikimate to chorismate.21 Transformation of KA12/pKAD50
with plasmid pKS3-02,6 which carries the entC gene encoding
isochorismate synthase, enables further metabolism of choris-
mate to isochorismate. Incubating crude sodium shikimate and
phosphoenolpyruvate (PEP) with cell extracts from KA12/
pKAD50/pKS3-02 for 1 h at 23 °C produces an equilibrium
mixture of unlabeled chorismate and isochorismate in a ca. 3:2
The KA12/pKAD50/pKS3-02 production strain represents an
attractive alternative to a recombinant strain of Klebsiella
pneumoniae 62-1, a class 2 pathogenic organism, which has
been used previously to produce isochorismate.6 The combined
yield of chorismate and isochorismate (about 15%) is substan-
tially lower than the ca. 50% yield of chorismate previously
observed for the KA12/pKAD50 system,21 presumably due to
the relative instability of isochorismate and its continuous
metabolic depletion in the cell extract. Nevertheless, prepara-
tively useful quantities of selectively labeled isochorismate
derivatives can be easily produced.
(21) Gustin, D. J.; Hilvert, D. J. Org. Chem. 1999, 64, 4935-4938.
9
J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006 2047

A R T I C L E S
DeClue et al.
2
Figure 3. Decomposition of [2-2H]isochorismate (1a) in 200 mM phosphate buffer (pH 6.0 and 60 °C) monitored by H NMR spectroscopy. Dioxane-d8
(1.00 mM) was used as an internal standard.
The effect of deuterium substitution at C2 on the reaction
manifold was determined by monitoring the decomposition of
[2-2H]isochorismate at 60 °C by 1H NMR as described for the
unlabeled compound. A representative time course is shown in
Figure 2B. The rate constants derived for the rearrangement,
aromatization, and elimination steps are k1 ) (3.43 ( 0.01) ×
10-4 s-1, k2 ) (5.37 ( 0.14) × 10-5 s-1, and k3 ) (2.03 (
0.02) × 10-5 s-1, respectively. Comparison with the corre-
Figure 4. Pseudodiequatorial (left) and pseudodiaxial (right) conformation
of isochorismate (1) (B3LYP/DZ+(2d,p)). The O-C2-C3-O dihedral
angle for these structures is 77.2° and 141.0°, respectively.
sponding kinetic constants obtained with unlabeled isochoris-
mate gives deuterium isotope effects of 1.06 ( 0.03, 1.01 (
0.03, and 2.29 ( 0.11 on the k1, k2, and k3 steps.
and can easily adopt multiple orientations with minimal
energetic penalty. In contrast, pseudodiaxial conformers, in
which the hydrogen bond between the C1 carboxylate and the
C2 alcohol is broken (Figure 4, right), are substantially higher
in energy. In the gas phase, the latter are at least 10.1 kcal/mol
less stable than the lowest energy pseudodiequatorial conformer.
The calculated difference drops to 8.0 kcal/mol when the two
carboxylate groups are shielded in a dielectric continuum that
mimics water,17 but this large value makes it unlikely that
pseudodiaxial conformers are significantly populated in solution.
The conversion of isochorismate to isoprephenate is formally
a [3,3]-sigmatropic rearrangement, and two different aromatic
transition structures, corresponding to chairlike and boatlike
geometries, respectively, were located at the B3LYP/DZ+(2d,p)
level of theory (Figure 5). Both are highly asymmetric, with
C-O bond cleavage significantly preceding C-C bond forma-
tion. The O-C2-C3-O dihedral angles in these structures
(138° for the chair and 126° for the boat) approach that of the
pseudodiaxial ground state conformer (141°), suggesting that
substrate must undergo an unfavorable conformational change
prior to rearrangement that contributes significantly to the overall
reaction barrier. The calculated activation energy for rearrange-
ment via the boat transition state is 26.5 kcal/mol, in reasonable
agreement with the experimental value of 24.8 kcal/mol (Table
1). However, agreement between theory and experiment is poor
The fate of the deuterium atom at C2 during decomposition
2
of [2-2H]isochorismate was established by H NMR spectros-
copy at 60 °C (Figure 3). A pH of 6.0 was chosen for these
experiments to minimize exchange with solvent. In agreement
with the kinetic data (Figure 2B), both [4-2H]isoprephenate and
3-carboxy[4-2H]phenylpyruvate, which derive from the energeti-
cally more favorable rearrangement pathway, are readily
detected by the characteristic signal of their single deuterium
atom. In addition, sizable amounts of [3-2H]pyruvate, which
arises from the minor elimination pathway, are also observed.
Roughly 70% of the pyruvate formed within 3 h contains
deuterium originating from C2 of the labeled isochorismate as
judged from direct integration of the respective signals for
1
deuterated and deuterium-free pyruvate observed by H NMR
spectroscopy after lyophilization of the NMR sample. Given
the observed rate of exchange under the experimental conditions,
formation of unlabeled pyruvate can be entirely explained by
direct exchange of [3-2H]pyruvate with solvent.
Computation. Isochorismate is a flexible molecule. HDFT
calculations at the B3LYP/DZ+(2d,p) level of theory show that
pseudodiequatorial conformations, which are stabilized by an
intramolecular hydrogen bond between the C1 carboxylate group
and the C2 alcohol, are lowest in energy (Figure 4, left); the
enolpyruvyl side chain is not particularly constrained, however,
9
2048 J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006

Thermal Decomposition of Isochorismate
A R T I C L E S
Figure 5. Chairlike (left) and boatlike (right) transition structures for the
[3,3]-sigmatropic rearrangement of isochorismate to isoprephenate (B3LYP/
DZ+(2d,p)). These structures have an O-C2-C3-O dihedral angle of
138.0° and 126.3°, respectively.
Figure 6. Chairlike (left) and twist boat (right) transition structures for
the [1,5]-sigmatropic rearrangement of isochorismate to give salicylate and
pyruvate (B3LYP/DZ+(2d,p)). The chair transition state has an O-C2-
C3-O dihedral angle of 87.5°, whereas the value for the twist boat is 114.7°.
Table 1. Calculated Activation Energies and Isotope Effects for
the Thermal Decomposition of Isochorismate at 333.15 K
Ea
Ea-Solution
(kcal/mol)a
reaction
(kcal/mol)a
2H KIEb
Claisen TS1 (chair)
Claisen TS2 (boat)
elimination TS1 (chair)
elimination TS2 (twist boat)
35.8
26.5
25.0
26.4
24.9
26.2
22.0
24.8
1.027
0.989
2.071
1.830
Figure 7. Transition structure for ionization of isochorismate to an ion
pair intermediate (B3LYP/DZ+(2d,p)).
a Delta energy difference between the transition state indicated and the
pseudoequatorial ground state conformation. For comparison, the experi-
mental activation energies for the Claisen rearrangement and the elimination
reaction are 24.8 and 26.2 kcal/mol, respectively, at this temperature.
b Kinetic isotope effects (KIE) calculated for [2-2H]isochorismate. For
comparison, the experimental values for the Claisen rearrangement and the
elimination reaction are 1.06 ( 0.03 and 2.29 ( 0.11, respectively, at T )
333.15 K, as described in the text.
25.0 kcal/mol, in good agreement with experiment (26.2 kcal/
mol); the predicted 2H isotope effect for the transferred hydrogen
atom of 2.071 also agrees well with the experimental value of
2.29 ( 0.11 (Table 1).23 For comparison, the higher energy twist
boat transition state affords a substantially lower isotope effect
(1.830). In this case, application of the dielectric continuum
model does not alter the intrinsic preference for the chair
geometry.
for the 2H isotope effect at C2 (predicted, 0.989; observed, 1.06
( 0.03). In contrast, the chairlike transition state predicts a
normal isotope effect of 1.027 that is much closer to the
observed value, but the barrier in this case is unrealistically high
(35.8 kcal/mol) (Table 1). Chairlike transition states are usually
favored for Claisen rearrangements,22 but in isochorismate
juxtaposition of the negatively charged carboxylates greatly
destabilizes the chair relative to the boat in the gas phase.
Consistent with this explanation, when the charges are shielded
in a dielectric continuum to model an aqueous environment,17
rearrangement via the chair is predicted to be favored over the
boat by 1.3 kcal/mol (Table 1).
The isochorismate elimination reaction can also be described
as a pericyclic process, i.e., a [1,5]-sigmatropic rearrangement
in which the C-O bond is cleaved simultaneously with
hydrogen atom transfer from C2 to C9. Again, two conforma-
tionally distinct transition state structures were located, which
differ in energy by 1.4 kcal/mol in the gas phase (Figure 6;
Table 1). The more favorable has a chairlike geometry, with
the carboxylate on the side chain directed over the ring, whereas
the higher energy structure is better described as a twist boat in
which the side chain carboxylate is directed away from the ring.
As in the case of the Claisen rearrangement, both elimination
transition structures correspond to highly asynchronous pro-
cesses, with C-O cleavage far advanced compared to hydrogen
atom transfer. To access the chairlike transition state, which
has an O-C2-C3-O dihedral angle of 87.5°, only relatively
minor adjustments in the more stable pseudodiequatorial ground
state conformation, which has a dihedral angle of 77°, are
necessary. The calculated activation energy for this process is
While our calculations support the feasibility of isochorismate
decomposing via competitive sigmatropic processes, a fully
dissociative pathway, in which rate-determining C-O bond
scission affords an ion pair intermediate that subsequently
partitions to the observed rearrangement and/or elimination
products, represents a conceivable alternative mechanism. The
different experimental 2H isotope effects on rearrangement and
elimination effectively rule out mechanistic linkage of the two
pathways via a common intermediate in this way, but a two-
step pathway might apply to either one of the reaction channels
alone. A dissociative transition state is expected to be highly
unstable in the gas phase due to extensive charge separation.
However, we found a region of the potential energy surface
where such a species exists, allowing us to estimate the
energetics of the ionic pathway. The dissociative structure shown
in Figure 7, which is 56.1 kcal/mol higher in energy than the
pseudodiequatorial ground state conformer, has a single negative
eigenvalue associated with C-O bond scission, but the gradient
of this structure is not identically zero because of the flatness
of the surrounding energy surface. Application of the dielectric
continuum model does not change this estimate much so that
the ionic pathway remains more than 20 kcal/mol less favorable
than either sigmatropic mechanism. Furthermore, the dissociative
(23) While conventional transition state theory predicts heavy atom isotope
effects well (see, for example: (a) Kohen, A. Prog. React. Kinet. Mech.
2003, 28, 119-156. (b) Scheiner, S. Biochim. Biophys. Acta 2000, 1458,
28-42. (c) Meyer, M. P.; DelMonte, A. J.; Singleton, D. A. J. Am. Chem.
Soc. 1999, 121, 10865-10874), calculated primary deuterium isotope effects
appear to be less accurate. Since higher level variational treatments are
currently very difficult, the theoretical isotope effects reported here for the
elimination reaction should be viewed with caution.
(22) Ziegler, F. E. Chem. ReV. 1988, 88, 1423-1452.
9
J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006 2049

A R T I C L E S
DeClue et al.
Scheme 3. Comparison of the Uncatalyzed Reactions of
transition state predicts a deuterium isotope effect of 1.274 at
C2, which is substantially higher than the experimental value
of 1.06 ( 0.03 observed for the rearrangement and substantially
smaller than the experimental value of 2.29 ( 0.11 observed
for the elimination. On the basis of these observations, a fully
dissociative mechanism appears unlikely for either process.
Isochorismate and Chorismatea
Discussion
The decomposition modes of isochorismate parallel the
uncatalyzed reactions of chorismate, which is another branch-
point intermediate in the shikimate biosynthetic pathway as well
as the direct precursor of isochorismate (Scheme 3).1 Like
isochorismate, chorismate reacts thermally via competitive
rearrangement and elimination pathways, leading to prephenate,
on the one hand, and to 4-hydroxybenzoate and pyruvate, on
the other.24
a Isochorismate and chorismate are equilibrated by the enzyme isocho-
rismate synthase (ICS).
Table 2. Kinetic Data for Decomposition of Isochorismate and
Chorismatea
1
1
substrate
104k1 (s-
)
Hk1
/
Dk1
104k 3 (s-
)
Hk 3 Dk 3
/
isochorismate 3.63 ( 0.09 1.06 ( 0.03 0.465 ( 0.022 2.29 ( 0.11
chorismateb
2.98 ( 0.01 1.07 ( 0.04 0.326 ( 0.009 1.85 ( 0.06
Although the mechanism of the chorismate transformations
has been a matter of some debate, it is clear that the corre-
sponding transition states are relatively polar compared to
chorismate based on the effects of substituents25 and solvent9
on the reaction rates. However, fully dissociative mechanisms
involving a shared ion pair intermediate have been viewed as
unlikely.26 Instead, kinetic isotope effects and computation
support concerted but highly asynchronous one-step mecha-
nisms, with considerable C-O bond cleavage at the transition
state and comparatively little C-C bond formation (Claisen
rearrangement) or hydrogen atom transfer to the enolpyruvyl
side chain (elimination).27
Our results support similar conclusions for isochorismate
(Table 2). The rearrangement and elimination reactions occur
with similar rates as their chorismate counterparts, with an 8-9-
fold preference for rearrangement over elimination in both cases.
The effects of deuterium substitution at the respective alcohol
center on both reactions are also comparable. These findings
are satisfactorily explained by competing [3,3]- and [1,5]-
sigmatropic rearrangements that proceed via different orienta-
tions of the enolpyruvyl side chain. Because the reactions
involve dianions in solution, they are difficult to characterize
theoretically. Nevertheless, the aromatic but asynchronous
transition structures with chairlike geometries which are pre-
dicted by theory for both rearrangement and elimination afford
isotope effects in good agreement with experiment for both
chorismate27 and isochorismate.23 Although the HDFT calcula-
tions incorrectly predict that the elimination reaction of iso-
chorismate should be favored over rearrangement, the small
difference in the barriers is within the error of the computational
methods.
a
Reactions were carried out in 200 mM sodium phosphate buffer in
D2O (pD 7.5) at 60 °C. b Data from ref 27.
been isolated, evidence for the enzymatic conversion of iso-
chorismate to isoprephenate has been obtained in crude extracts
derived from the plant Nicotiana silVestria.29 By analogy with
the reaction catalyzed by chorismate mutase, which accelerates
the [3,3]-sigmatropic rearrangement of chorismate to prephenate
by more than a million-fold,24 the name isochorismate mutase
was suggested for the putative enzyme.29 Like its well-studied
congener,27,30 it may exploit a combination of conformational
control and electrostatic transition state stabilization to catalyze
the Claisen rearrangement of isochorismate.
Salicylate-producing enzymes have been better characterized
due to the importance of salicylate in plant defense mechanisms
known as local and systemic acquired resistance31 and because
of their role in the biosynthesis of salicylate-based siderophores
in several pathogenic bacteria.32 The conversion of isochorismate
to salicylate is catalyzed by the enzyme isochorismate pyruvate
lyase (IPL).33 The enzyme from the opportunistic human
pathogen Pseudomonas aeruginosa, which is required for the
biosynthesis of the siderophore pyochelin, has been studied in
some detail.34,35 It accelerates the elimination reaction by an
estimated factor of roughly 4 × 105-fold over the spontaneous
thermal process.36 When [2-2H]isochorismate was employed as
a substrate, quantitative transfer of deuterium from C2 to C9
was observed.37 In addition, a deuterium kinetic isotope effect
of 2.34 ( 0.08 on kcat was determined, which is within
(29) Zamir, L. O.; Nikolakakis, A.; Bonner, C. A.; Jensen, R. A. Bioorg. Med.
Chem. Lett. 1993, 3, 1441-1446.
(30) (a) Chook, Y. M.; Ke, H.; Lipscomb, W. N. Proc. Natl. Acad. Sci. U.S.A.
1993, 90, 8600-8603. (b) Chook, Y. M.; Gray, J. V.; Ke, H.; Lipscomb,
W. N. J. Mol. Biol. 1994, 240, 476-500. (c) Lee, A. Y.; Karplus, P. A.;
Ganem, B.; Clardy, J. J. Am. Chem. Soc. 1995, 117, 3627-3628. (d) Stra¨ter,
N.; Schnappauf, G.; Braus, G.; Lipscomb, W. N. Structure 1997, 5, 1437-
1452.
In nature, the rearrangement reaction of isochorismate has
been implicated in the biosynthesis of m-carboxy-substituted
aromatic amino acids,2,28 which are produced as secondary
metabolites in plants. Although the responsible catalyst has never
(31) (a) Wildermuth, M. C.; Dewdney, J.; Wu, G.; Ausubel, F. M. Nature 2001,
414, 562-565. (b) Durrant, W. E.; Dong, X. Annu. ReV. Phytopathol. 2004,
42, 185-209.
(32) (a) Marshall, B. J.; Ratledge, C. Biochim. Biophys. Acta 1972, 264, 106-
116. (b) Okujo, N.; Saito, M.; Yamamoto, S.; Yoshida, T.; Miyoshi, S.;
Shinoda, S. Biometals 1994, 7, 109-116. (c) Serino, L.; Reimmann, C.;
Baur, H.; Beyeler, M.; Visca, P.; Haas, D. Mol. Gen. Genet. 1995, 249,
217-228. (d) Pelludat, C.; Brem, D.; Heesemann, J. J. Bacteriol. 2003,
185, 5648-5653.
(33) This enzyme is encoded by pchB, and the protein is thus also referred to
as PchB. In addition to IPL activity, PchB also exhibits weak chorismate
mutase activity (ref 34).
(34) Gaille, C.; Kast, P.; Haas, D. J. Biol. Chem. 2002, 277, 21768-21775.
(35) Ku¨nzler, D. E.; Sasso, S.; Gamper, M.; Hilvert, D.; Kast, P. J. Biol. Chem.
2005, 280, 32827-32834.
(24) Andrews, P. R.; Smith, G. D.; Young, I. G. Biochemistry 1973, 12, 3492-
3498.
(25) (a) Coates, R. M.; Rogers, B. D.; Hobbs, S. J.; Peck, D. R.; Curran, D. P.
J. Am. Chem. Soc. 1987, 109, 1160-1170. (b) Gajewski, J. J.; Jurayj, J.;
Kimbrough, D. R.; Gande, M. E.; Ganem, B.; Carpenter, B. K. J. Am.
Chem. Soc. 1987, 109, 1170-1186.
(26) Gajewski, J. J.; Brichford, N. L. J. Am. Chem. Soc. 1994, 116, 3165-
3166.
(27) Wright, S. K.; DeClue, M. S.; Mandal, A.; Lee, L.; Wiest, O.; Cleland, W.
W.; Hilvert, D. J. Am. Chem. Soc. 2005, 127, 12957-12964.
(28) Larsen, P. O.; Onderka, D. K.; Floss, H. G. Chem. Commun. 1972, 842-
843; Biochim. Biophys. Acta 1975, 381, 397-408.
9
2050 J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006

Thermal Decomposition of Isochorismate
A R T I C L E S
experimental error of the value of 2.222 predicted for the
chairlike elimination transition state at 30 °C, the temperature
of the enzymatic experiment.37 On the basis of this evidence,
the IPL-catalyzed reaction, like its thermal counterpart, appears
to be pericyclic in nature. How does catalysis occur? Sequence
comparisons show that IPL is structurally related to chorismate
mutases of the AroQ class; it has even been shown to have low
chorismate mutase activity.34 It is therefore likely that the two
enzymes utilize similar mechanistic strategies for accelerating
their respective sigmatropic rearrangements,35 although the
relative importance of conformation control versus electrostatic
transition state complementarity in the different systems remains
an open issue.38
Detailed investigations of the structure and mechanism of IPL
(and isochorismate mutase) promise to extend our understanding
of how enzymes stabilize aromatic transition states. Insights into
the isochorismate reaction pathway are also relevant to the
chemistry of related shikimate derivatives, including chorismate,
4-amino-4-deoxychorismate, and 2-amino-2-deoxyisochoris-
mate, which undergo similar reactions and serve as building
blocks for diverse aromatic metabolites.1,39 Improved mecha-
nistic understanding of the plant, fungal, and bacterial enzymes
responsible for converting these compounds into essential
ubiquinones, siderophores, folates, and amino acids will benefit
efforts to develop selective herbicides, fungicides, and antibacte-
rial agents.
Acknowledgment. This work was supported by the Schweiz-
erischer Nationalfonds. We are grateful to Professor E. Leistner
(University of Cologne, Cologne, Germany) for plasmid pKS3-
02, D. Ku¨nzler for the generation and characterization of the
transformant KA12/pKAD50/pKS3-02, and Professor B. Jaun
(ETH Zu¨rich) for expert assistance with the NMR measure-
ments.
Supporting Information Available: Coordinates and com-
putational details for calculation of the theoretical isotope effects,
and complete ref 13. This material is available free of charge
via the Internet at http://pubs.acs.org.
JA056714X
(36) At 30 °C the kcat value for IPL was determined to be 1.06 s-1 (ref 35).
Under the same conditions, isochorismate decomposes with an apparent
rate constant of (2.0 ( 0.5) × 10-5 s-1 (D. E. Ku¨nzler, unpublished).
Assuming that the relative rates of the rearrangement and elimination
reactions are not strongly temperature dependent, we estimate the rate
constant for the uncatalyzed elimination reaction to be approximately (2.5
( 0.6) × 10-6 s-1. The acceleration achieved by IPL (kcat/kuncat) is therefore
about 4 × 105.
(38) (a) Kast, P.; Grisostomi, C.; Chen, I. A.; Li, S.; Krengel, U.; Xue, Y.;
Hilvert, D. J. Biol. Chem. 2000, 275, 36832-36838. (b) Kienho¨fer, A.;
Kast, P.; Hilvert, D. J. Am. Chem. Soc. 2003, 125, 3206-3207. (c) Strajbl,
M.; Shurki, A.; Kato, M.; Warshel, A. J. Am. Chem. Soc. 2003, 125,
10228-10237. (d) Zhang, X. D.; Zhang, X. H.; Bruice, T. C. Biochemistry
2005, 44, 10443-10448.
(37) DeClue, M. S.; Ku¨nzler, D. E.; Baldridge, K. K.; Kast, P.; Hilvert, D. J.
Am. Chem. Soc. 2005, 127, 15002-15003.
(39) Walsh, C. T.; Liu, J.; Rusnak, F.; Sakaitani, M. Chem. ReV. 1990, 90, 1105-
1129.
9
J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006 2051