Design and synthesis of 3-fluoro-2-oxo-3-phenylpropionic acid derivatives as potent inhibitors of 4-hydroxyphenylpyruvate dioxygenase from pig liver

Article abstract of DOI:10.1016/S0968-0896(99)00062-0

Several rationally designed analogs of 3-fluoro-2-oxo-3-phenylpropionic acid were chemically synthesized, and the reactions of the hydrate form of these compounds with 4-hydroxyphenylpyruvate dioxygenase from pig liver as inhibitors were examined. Compounds 14a and 14b were found to be potent competitive inhibitors of the enzyme with K(i) values of 10 and 22μM, respectively. Copyright (C) 1999 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(99)00062-0

Bioorganic & Medicinal Chemistry 7 (1999) 1459±1465
Design and Synthesis of 3-Fluoro-2-oxo-3-phenylpropionic Acid
Derivatives as Potent Inhibitors of 4-Hydroxyphenylpyruvate
Dioxygenase from Pig Liver
Tung-shen Ling, Shi Shiu and Ding-yah Yang*
Department of Chemistry, Tunghai Christian University, 181, Taichung-Kong Rd. Sec.3, Taichung, Taiwan 40704,
People's Republic of China
Received 11 January 1999; accepted 3 March 1999
AbstractÐSeveral rationally designed analogs of 3-¯uoro-2-oxo-3-phenylpropionic acid were chemically synthesized, and the reac-
tions of the hydrate form of these compounds with 4-hydroxyphenylpyruvate dioxygenase from pig liver as inhibitors were exam-
ined. Compounds 14a and 14b were found to be potent competitive inhibitors of the enzyme with Ki values of 10 and 22 mM,
respectively. # 1999 Elsevier Science Ltd. All rights reserved.
Introduction
unclear. Nevertheless, a couple of possible mechanisms
have been proposed by Witkop7 and Hamilton8 as
depicted in Schemes 1 and 2.
4-Hydroxyphenylpyruvate dioxygenase (EC 1.13.11.27,
HPPD)1 is a key enzyme involved in the catabolism of
tyrosine and phenylalanine in most organisms. It cata-
lyzes the conversion of 4-hydroxyphenylpyruvate (HPP,
1) and molecular oxygen to homogentisate 2 and carbon
dioxide as shown below.
In Scheme 1, the metal ion±oxygen complex reacts with
the carbanion of HPP to form a peroxide ion 3, which
then attacks the a-carbon of the keto acid yielding cyclic
peroxide 4. After decarboxylation, the carboxymethyl
side chain of 5 migrates to the ortho-position to yield the
product 2 by a mechanism similar to the NIH shift.9
In Scheme 2, the metal ion±oxygen complex reacts with
the carbonyl group of the a-keto acid to form the cyclic
peroxide 6, followed by oxidative decarboxylation to
yield the oxoiron (IV) species 7. The proposed high-
valent iron-oxo species then attacks the ring to give
arene oxide-like intermediate 8, followed by enzyme-
catalyzed 1, 2-shift of the carboxymethyl group and
®nal aromatization to a€ord homogentisate 2.
This enzyme has been isolated homogeneously from
avian liver,2 mammalian liver,3 and a pseudomonad
strain.4 Most of these puri®ed proteins have been shown
to contain a non-heme iron as a cofactor essential for
catalytic activity. Early studies of HPPD have estab-
lished that one atom each from molecular oxygen ends
up in carboxyl and hydroxyl groups of the product
homogentisate and the produced carbon dioxide comes
from the carboxylate of HPP.5 Although kinetic studies
of the enzymes have indicated the binding of substrates
is ordered with oxygen as the second substrate,6 none of
the proposed enzyme-bound intermediates in the reac-
tion have been observed or trapped. Thus, the sequential
events catalyzed by HPPD leading to decarboxylation,
hydroxylation and rearrangement of HPP remains
According to the proposed mechanisms for HPPD in
Schemes 1 and 2, both mechanisms share a common
step of nucleophilic attack of activated oxygen onto the
carbonyl group of the a-keto acid to generate the tetra-
hedral enzyme-substrate intermediate either 4 or 6. We
speculate the catalytic process will be blocked if the a-
keto group of the natural substrate HPP change from
keto form to hydrate form, since tetrahedral structure of
a-carbon of HPP hydrate form is no longer feasible for
nucleophilic attack by activated oxygen. Thus, the
hydrate form of HPP may serve as a potential inhibitor
for HPPD. The desired a-keto hydrated HPP can be
prepared by simply introducing a ¯uorine atom at C-3
Key words: HPPD; enzyme inhibition; keto form; hydrate form.
* Corresponding author. Tel.: +886-4-3597613; fax: +886-4-
3590426; e-mail: yang@mail.thu.edu.tw
0968-0896/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(99)00062-0

1460
T. Ling et al. / Bioorg. Med. Chem. 7 (1999) 1459±1465
e€ort to elucidate the pathway for this interesting enzy-
matic transformation and to develop strong inhibitors
for this enzyme.
Results and Discussion
Since several well-known potent inhibitors of HPPD,
such as 2-(2-nitro-4-tri¯uoromethylbenzoyl)-cyclohex-
ane-1,3-dione11a (NTBC, IC50=40 nM) and 2-(2-chlo-
ro-4-methanesulfonylbenzoyl)-cyclohexane-1,3-dione11b
(CMBC, Ki=10 nM), contain relatively large sub-
stituents at ortho and para-positions of the phenyl
ring,11c±d it is conceivable that these bulky groups on
the aromatic ring may play an important role in HPPD
inactivation. Thus, the ecacy of single ¯uorinated
substrate analogues as potential HPPD inhibitors may be
enhanced by selectively introducing the bulky groups at
ortho and/or para-position of aromatic ring to increase
the number of favorable interactions between inhibitor
and enzyme active site. In order to test this hypothesis,
several ortho and/or para-substituted 3-¯uoro-2-oxo-3-
phenylpropionic acids have also been synthesized to
investigate their e€ects on enzyme inactivation.
Scheme 1. Proposed HPPD mechanism by Witkop.
position of the natural substrate HPP. The substitution
of a hydrogen atom by ¯uorine in organic molecules
introduces few steric demands, but the high electro-
negativity of ¯uorine may induce large changes in reac-
tivity. For example, the a-keto group of 3-¯uoro-3-
phenylpyruvic acid, ¯anked by an electron-withdrawing
¯uorine atom and a carboxylic acid group, is essentially
hydrated in aqueous solution because of the adjacent
¯uorine atom.10
The procedure for the synthesis of 3-¯uoro-2-oxo-3-
phenylpropionic acid derivatives12 is outlined in Scheme
3. According to this strategy, epoxy esters 11a±d were
obtained by Darzens13 glycidic ester condensation of
commercially available ortho and/or para-substituted
benzaldehydes 10a±d with methyl chloroacetate in the
presence of sodium methoxide. Ring opening of 11a±d
with a hydrogen ¯uoride-pyridine solution produced the
b-¯uoro-a-hydroxy esters 12a±d, which were oxidized to
the corresponding hydrate form of b-¯uoro-a-keto
esters 13a±d by Jones reagent and then hydrolyzed,
under basic conditions, to give the target hydrate form
of 3-¯uoro-a-keto acids 14a±d.
In this paper, we report studies of the reactions of a
series of chemically synthesized ¯uorinated a-keto
hydrated 3-¯uoro-3-phenylpyruvic acids with 4-hydroxy-
phenylpyruvate dioxygenase from pig liver as our initial
The extent of hydration of 3-¯uoro-2-oxo-3-phenyl-
propionic acid derivatives in aqueous solution were
determined by 1H NMR analysis. 3-Fluoro-p-nitro-
phenylpyruvate 14b, for example, was found to be in
both keto and hydrate forms when dissolved in acetone-
d6 with hydrate/keto ratio of 2.6/1, based on the inte-
gration of b-H in 1H NMR spectrum shown in Figure 1.
When dissolved in deuterium oxide, however, 14b was
found exclusively in hydrate form with the complete
disappearance of the keto form b-H signal at 6.83 ppm.
With these potential HPPD inhibitors at hand, incuba-
tion experiments were then conducted to determine
kinetic and inhibition parameters as well as the type of
inhibition of 14a±d by the spectrophotometric enol
borate assay method.14 The time-course of the reaction
of pig liver HPPD in the absence and presence of
increasing amount of the hydrate form of 3-¯uoro-3-
Scheme 2. Proposed HPPD mechanism by Hamilton.

T. Ling et al. / Bioorg. Med. Chem. 7 (1999) 1459±1465
1461
Scheme 3. Synthesis of compounds 14a±d. (a) ClCH2CO2Me, NaOMe, MeOH; (b) HF/pyridine, CH2Cl2; (c) Jones reagent, acetone; (d) H2O,
i-PrOH, NaHCO3.
phenylpyurvate is shown in Figure 2. The activity of
HPPD was obviously inhibited by 14a. Almost complete
inhibition was observed at a concentration of 100 mM.
The inhibition constants for the reactions of 14a±d with
HPPD are listed in Table 1. The Lineweaver±Burk plots
indicated that 14a and 14b were competitive inhibitors
of HPPD with respect to HPP, whereas 14c and 14d
were noncompetitive inhibitors. Compounds 14a and
14b were found to be potent inhibitors of the enzyme
with Ki values of 10 and 22 mM, respectively. Appar-
ently, the substituent on para-position of the phenyl
ring is not essential for binding, since the introduction
of p-nitro group resulted in little change in potency. The
failures of 14c and 14d as competitive inhibitors of
HPPD are consistent with the results reported by Pas-
cal,15 who has shown that o-chloro and o-methyl-HPP
were noncompetitive inhibitors of HPPD. Perhaps the
relatively bulky substituent at ortho-position of phenyl
moiety distorts the molecular geometry which is
required for binding. Furthermore, the fact that higher
steric demands on the ortho-position of HPP ring sub-
stituents as the substrate for HPPD suggest that 14a and
14b may share di€erent inhibition mechanism of action
with NTBC and CMBC, since latter compounds
required a bulky ortho subsutituent like nitro or chloro
group for potent inhibition activity.16
Conclusion
A series of analogues of the hydrate form of 3-¯uoro-2-
oxo-3-phenylpropionic acid were rationally designed
and synthesized to mimic the tetrahedral structure of
enzyme±substrate complex, and the reactions of these
compounds with 4-hydroxyphenylpyruvate dioxygenase
puri®ed from pig liver as inhibitors were examined.
Compounds 14a and 14b were found to be potent
competitive inhibitors of the enzyme HPPD, whereas
compound 14c and 14d were noncompetitive inhibitors.
The substrate analogues for potent HPPD inhibition
appear to have higher steric demands on ortho-position
Figure 1. 1H NMR spectra (300 MHz) of 14b in (a) acetone-d6 (b)
than para-position of the phenyl moiety. The results of
D2O. The more up®eld doublet at 5.88 ppm in (a) was assigned to b-H
our studies will provide useful information for designing
inhibitors for treatment of the fatal disease tyrosinemia
of 14b with JH F=44.7 Hz and the small doublet at 6.83 ppm was
assigned to b-H of keto form of 14b with JH F=47.1 Hz.

1462
T. Ling et al. / Bioorg. Med. Chem. 7 (1999) 1459±1465
Figure 2. Time-course of the reaction of HPPD and 4-hydroxyphenylpyruvate (35 mM) with varying concentrations of 14a (100, 30, 15, 7.5, and
0 mM). The protein concentration in the assay was 1 mg/mL.
type I11a and developing new class of bleaching herbi-
cides17 for selected plants as well.
Enzyme puri®cation and assay
Phenylpyruvate tautomerase was puri®ed from pig liver
by the method of Knox.18 HPPD was puri®ed from pig
liver by the method of Hamilton19 with the speci®c
activity of 0.1 mmol min 1 mg 1. The protein concentra-
tion was determined by BCA method.20 Routine assay
of the enzyme utilized the spectrophotometric enol
borate method of Lindstedt and Rundgren.14 The typi-
cal assay mixtures contained 0.85 mL potassium phos-
phate/borate bu€er (prepared by adjusting the pH of
0.42 M H3BO3 to 6.2 with a 0.17 M Na3PO4 solution),
0.06 mL HPP (1.8 mM, in 0.2 M Na3PO4 bu€er), 30 mL
dichlorophenolindophenol (reduced form, prepared by
mixing 1 mL of 3.3 mM sodium dichloro-phenolindo-
phenol in H2O and 0.16 M glutathione in 0.2 M sodium
phosphate bu€er), 0.01 mL phenylpyruvate tautomerase
(10 U/mL). The above solution was equilibrated for
about 15 min (monitored at 308 nm), then HPPD to be
tested was added (0.05 mL solution). For calculation of
dioxygenase activity the change in absorbance between
the 8th and 10th min was used. The interaction of 3-
¯uoro-2-oxo-3-phenylpropionic acid derivatives and the
enzyme HPPD was evaluated by measuring the decrease
in absorbance at 308 nm over a 15 min period following
coadministration of varying concentration of the sub-
strate and inhibitors.
Experimental
General
Melting points were determined on a Mel-Temp melting
point apparatus in open capillaries and are uncorrected.
High resolution FAB-MS was measured with a JEOL
JMS-SX/SX 102A spectrometer. Ultraviolet-visible
spectroscopy was performed on Hewlett±Packard 8453
spectrophotometer. 1H and 13C NMR spectra were
recorded at 300 MHz on a Varian VXR300 spectro-
meter. Chemical shifts were reported in ppm on the d
scale relative to internal standard (tetramethylsilane, or
appropriate solvent peaks) with coupling constants
1
given in hertz. H and 13C NMR multiplicity data were
denoted by s (singlet), d (doublet), t (triplet), q (quartet),
m (multiplet), and b (broad). Flash chromatography
was performed in columns of various diameters with
Merck silica gel (230±400 mesh ASTM 9385 kieselgel
60H) by elution with the solvents reported. Analytical
thin-layer chromatography (TLC) was carried out on
Merck silica gel 60G-254 plates (25 mm) and devel-
oped with the solvents mentioned. Solvents, unless
otherwise speci®ed, were reagent grade and distilled
once prior to use.
General procedure for preparation of epoxy esters 11a±
d. A solution of benzaldehyde (2.12 g, 10.0 mmol) and
methyl chloroacetate (2.0 g, 20.0 mmol) was added a
solution of sodium methoxide in anhydrous MeOH (2 g
of sodium dissolved in 40 mL of MeOH), with the tem-
perature kept at 10 ꢀC. After the addition was com-
pleted, the solution was stirred at 0 ꢀC for 2 h and then
3 h at room temperature. The mixture was then poured
into 0.1 N HCl cold water and the crude product was
extracted with ethyl acetate, dried with MgSO4, con-
centrated in vacuo to give the yellow solid or liquid
which can be further recrystallized in ethanol or puri®ed
by ¯ash chromatography to give the desired product.
Table 1. Kinetic constants for reactions of 14a±d with HPPD from
pig liver
Compound
R1 R2
typea
Ki (mM)b
10‹2
14a
14b
14c
14d
H
H
Cl
H
NO2
H
C
C
NC
NC
22‹2
250‹20
380‹25
Cl Cl
a C=competitive inhibitor; NC=noncompetitive inhibitor.
b Assay detail is described in Experimental.

T. Ling et al. / Bioorg. Med. Chem. 7 (1999) 1459±1465
1463
Methyl 3-phenylglycidate (11a). This compound was
obtainedꢀ as a white solid in 52% yield, mp 84±85 ꢀC
(lit.12 84 C).
(CDCl3) d 171.5, 171.3 (COOMe), 133.5±126.7 (Ar C's),
91.4 (d, J=180.9; Cb), 90.4 (d, J=179.8; Cb), 72.4 (d,
J=25.3; Ca), 70.9 (d, J=22.6; Ca), 53.1 (OMe), 52.4
(OMe). HRMS: calcd for C10H10ClFO3 232.0303, found
232.0312.
Methyl trans-3-(4-nitrophenyl)glycidate (11b). This com-
pound was obtained as a white solid in a 65% yield, mp
139±140 ꢀC (lit.21 138±140 ꢀC).
Methyl 3-¯uoro-2-hydroxy-(2,4-dichlorophenyl)propionate
(12d). This compound was obtained as a yellow liquid
in a 68% yield. Rf=0.53 (25% EtOAc/hexanes). 1H
NMR (erythro+threo) (CDCl3) d 7.52±7.29 (6H; m; Ar
H's), 6.16 (1H; dd, J=45.5, 1.5; Hb), 6.05 (1H; dd,
J=44.7, 3.0; Hb), 4.68 (1H; dd, J=18.2, 8.1; Ha), 4.56
(1H; dd, J=30.0, 6.9; Ha), 3.92 (3H; s; OMe), 3.71(3H;
s; OMe), 3.31 (1H; d, J=8.1; OH), 3.23 (1H; d, J=7.2;
OH). 13C NMR (erythro+threo) (CDCl3) d 171.7, 171.5
(COOMe), 135.3±126.9 (Ar C's), 91.0 (d, J=181.4; Cb),
90.0 (d, J=180.4; Cb), 72.4 (d, J=22.0; Ca), 71.0 (d,
J=22.6; Ca), 53.1 (OMe), 52.5 (OMe). HRMS: calcd
for C10H9Cl2FO3 265.9914, found 265.9925.
Methyl trans-3-(2-chlorophenyl)glycidate (11c). 22 This
compound was obtained as a yellow liquid in a 68%
yield.
Methyl trans-3-(2,4-dichlorophenyl)glycidate (11d). This
compound was obtained as a colorless liquid with a
1
62% yield. Rf=0.66 (25% EtOAc/hexanes). H NMR
(CDCl3) d 7.38±7.16 (3H; m; Ar H's), 4.38 (1H; d,
J=2.1; Hb), 3.85 (3H; s; OMe), 3.37 (1H; d, J=2.1; Ha).
13C NMR (CDCl3) d 167.8 (COOMe), 134.9, 133.9,
131.7, 129.1, 127.4, 126.8 (Ar C's), 55.7, 54.8 (Ca, Cb),
52.6 (OMe). HRMS: calcd for C10H8Cl2O3 245.9851,
found 245.9847.
General procedure for preparation of a-keto esters 13a±d.
To a solution of a-hydroxy ester 12 (500 mg, 2.53 mmol)
in acetone (15 mL) was added Jones reagent dropwise
under room temperature until the reaction was com-
pleted. The excess of Jones reagent was destroyed by
adding isopropanol into the reaction mixture. After
concentration by rotatory evaporation, the residue was
added to water and followed by routine extraction and
puri®cation procedure to give the pure compound.
General procedure for preparation of a-hydroxy esters
12a±d. To a solution of epoxy ester 11 (392.0 mg,
2.2 mmol) in dry methylene chloride (10 mL) in a plastic
beaker was added dropwise a solution of HF/pyridine
(70%, 0.2 mL) under 20 ꢀC. After the reaction is com-
pleted within 5 min, the mixture was added to ice-cold
water, extracted with methylene chloride, dried with
MgSO4, concentrated in vacuo, and puri®ed by column
chromatography to give the pure product.
Methyl 3-¯uoro-3-phenylpyruvate (13a).12 This com-
pound was obtained as a colorless liquid in a 55% yield.
Methyl 3-¯uoro-2-hydroxy-3-phenylpropionate (12a). 12
This compound was obtained as a light yellow liquid in
a 90% yield.
Methyl 3-¯uoro-p-nitrophenylpyruvate (13b). This com-
pound was obtained as a yellow solid in a 40% yield.
mp 106±108 ꢀC. Rf=0.48 (50% EtOAc/hexanes). 1H
NMR (keto form) (CDCl3) d 8.29 (2H; d, J=8.4; Ar
H's), 7.70 (2H; d, J=8.7; ArH's), 6.54 (1H; d,
Methyl 3-¯uoro-2-hydroxy-(4-nitrophenyl)propionate
(12b). This compound was obtained as a yellow solid in
a 60% yield. mp 86±89 ꢀC. Rf=0.39 (25% EtOAc/hex-
anes). 1H NMR (erythro+threo) (CDCl3) d 8.27 (2H; d,
J=8.4; Ar H's), 8.24 (2H; d, J=9.0; Ar H's), 7.58 (2H;
d, J=9.0; Ar H's), 7.51 (2H; d, J=9.0; Ar H's), 5.91
(1H; dd, J=45.0, 1.8; Hb), 5.84 (1H; dd, J=45.3, 3.6;
Hb), 4.69 (1H; d, JF H=16.8; Ha), 4.48 (1H; dd,
J=28.2, 4.2, Ha), 3.91 (3H, s; OMe), 3.79 (3H; s; OMe),
3.21 (1H; d, J=6.3; OH), 3.13 (1H; d, J=6.6; OH). 13C
1
JF H=46.8; Hb), 3.88 (3H; s; OMe). H NMR (hydrate
form) (CDCl3) d 8.27 (2H; d, J=8.4; Ar H's), 7.68 (2H;
d, J=8.7; Ar H's), 5.78 (1H; d, JF H=44.7; Hb), 4.51
(1H; bs; OH), 3.95 (3H; s; OMe), 3.19 (1H; bs; OH). 13C
NMR (hydrate form) (Acetone-d6) d 171.5 (COOMe),
149.2 (Ar C-4's), 142.9 (d, J=20.2; Ar C-1's), 130.1 (d,
J=7.7; Ar C-2, Ar C-6's), 123.4 (Ar C-3, Ar C-5's), 94.3
(d, J=26.3; Ca), 94.0 (d, J=179.3; Cb), 53.2 (OMe).
HRMS: calcd for C10H10FNO6 259.0492, found
259.0485.
NMR (erythro+threo) (CDCl3)
d
171.2, 171.1
(COOMe), 148.2, 148.1 (Ar C-4's), 142.6 (d, J=20.7; Ar
C-1's), 127.2±126.8 (Ar C-2, Ar C-6's), 123.8±123.4 (Ar
C-3, Ar C-5's), 92.8 (d, J=183.2; Cb), 92.1 (d, J=183.1;
Cb), 73.3 (d, J=22.6; Ca), 73.0 (d, J=20.9; Ca), 53.3
(OMe), 52.9 (OMe). HRMS: calcd for C10H10FNO5
243.0543, found 243.0531.
Methyl 3-¯uoro-o-chlorophenylpyruvate (13c). This com-
pound was obtained as a yellow liquid in a 56% yield.
1
Rf=0.58 (50% EtOAc/hexanes). H NMR (keto form)
(CDCl3) d 7.51±7.36 (4H; m; Ar H's), 6.82 (1H; d,
J=46.2; Hb), 3.83 (3H; s; OMe). 1H NMR (hydrate
form) d 7.51±7.36 (4H; m; Ar H's), 6.22 (1H; d, J=44.3;
Hb), 3.92 (3H; s; OMe). 13C NMR (keto+hydrate form)
(CDCl3) d 186.9 (d, J=24.8; keto form Ca), 171.2
(hydrate form COOMe), 160.0 (d, J=2.2, keto form
COOMe), 134.3±127.5 (m; keto+hydrate form Ar C's),
93.5 (d, J=29.1; hydrate form Ca), 89.9 (d, J=187.0;
keto form Cb), 88.9 (d, J=186.4; hydrate form Cb), 53.9
(hydrate form OMe), 53.2 (keto form OMe). HRMS:
calcd for C10H10ClFO4 248.0252, found 248.0257.
Methyl 3-¯uoro-2-hydroxy-(2-chlorophenyl)propionate
(12c). This compound was obtained as a yellow liquid
in a 78% yield. Rf=0.45 (25% EtOAc/hexanes) 1H
NMR (erythro+threo) (CDCl3) d 7.62±7.31 (8H; m; Ar
H's), 6.18 (1H; dd, J=44.4, 1.2; Hb), 6.08 (1H; ddd,
J=45.3, 3.0, 0.6; Hb), 4.71 (1H; ddd, J=18.0, 8.1, 3.0;
Ha), 4.59 (1H; ddd, J=30.1, 7.2, 1.2; Ha), 3.92 (3H; s;
OMe), 3.68 (3H, s; OMe), 3.31 (1H; d, J=8.1; OH),
3.24 (1H; d, J=7.2; OH). 13C NMR (erythro+threo)

1464
T. Ling et al. / Bioorg. Med. Chem. 7 (1999) 1459±1465
Methyl 3-¯uoro-3-(2,4-dichlorophenyl)pyruvate (13d).
This compound was obtained as a colorless liquid in a
64% yield. Rf=0.32 (25% EtOAc/hexanes). H NMR
(d, J=184.7; hydrate form Cb). HRMS: calcd for
C9H8ClFO4 234.0095, found 234.0089.
1
(keto form) (CDCl3) d 7.65±7.26 (3H; m; Ar H's), 6.76
(1H; d, J=45.9; Hb), 3.87 (3H; s; OMe). 1H NMR
(hydrate form) d 7.51±7.36 (4H; m; Ar H's), 6.18 (1H; d,
J=43.8; Hb), 4.49 (1H; s; OH), 3.96 (3H; s; OMe), 3.12
(1H; s; OH). 13C NMR (keto+hydrate form) (CDCl3) d
188.1 (d, J=22.1; keto form Ca), 173.2 (hydrate form
COOMe), 160.3 (keto form COOMe), 134.1±127.2 (m;
keto+hydrate form Ar C's), 93.5 (d, J=29.2; hydrate
form Ca), 89.2 (d, J=185.1; keto form Cb), 88.6 (d,
J=183.4; hydrate form Cb), 53.6 (hydrate form OMe),
53.4 (keto form OMe). HRMS: calcd for C10H9Cl2FO4
281.9863, found 281.9869.
3-Fluoro-3-(2,4-dichlorophenyl)pyruvic acid (14d). This
compound was obtained as a white solid in a 42% yield.
mp 79±81 ꢀC. Rf=0.46 (20% MeOH/EtOAc). 1H NMR
(keto form) (Acetone-d6) d 7.82±7.41 (3H; m; Ar H's),
6.93 (1H; d, J=45.9; Hb). 1H NMR (hydrate form)
(Acetone-d6) d 7.82±7.41 (3H; m; Ar H's), 6.24 (1H; d,
J=44.1; Hb). 13C NMR (keto+hydrate form) (Acetone-
d6) d 188.5 (d, J=22.7; keto form Ca), 171.8 (hydrate
form COOH), 160.6 (keto form COOH), 133.1±127.6
(m; keto+hydrate form Ar C's), 93.8 (d, J=27.6;
hydrate form Ca), 90.9 (d, J=184.8; keto form Cb), 99.8
(d, J=181.9; hydrate form Cb). HRMS: calcd for
C9H7Cl2FO4 267.9706, found 267.9715.
General procedure for preparation of a-keto acids 14a±d.
To a solution of a-keto ester 13 (300 mg, 1.5 mmol) in
10 mL of H2O/i-PrOH (1/1) was added solid sodium
bicarbonate until it was saturated under room tempera-
ture. After the hydrolysis was completed within 1 h,
aqueous HCl solution (0.1 N) was added to adjust pH
to about 3. The product was then extracted with ethyl
acetate, dried with MgSO4, concentrated in vacuo, and
puri®ed by ¯ash chromatography to give the ®nal target
a-keto acid 14 with low yield.
Acknowledgements
The ®nancial assistance provided by National Science
Council of Republic of China is thankfully acknowl-
edged. The authors are grateful to Dr. Andrew Yeh for
reading the manuscript.
References
1. (a). Scheparts, B.; Gurin, S. J. Biol. Chem. 1949, 180, 663.
(b). Vitol, M. J.; Vilks, S. R.; Zabarovska, I. M.; Maurinia, K.
A. Kokl. Akad. Nauk SSSR 1970, 192, 908.
2. Sada, G. H.; Fellman, J. H.; Fujita, T. S.; Roth, E. S. J.
Biol. Chem. 1975, 250, 6720.
3-Fluoro-3-phenylpyruvic acid (14a). This compound
was obtained as a white solid in a 45% yield, mp 70±
71 ꢀC. Rf=0.35 (20% MeOH/EtOAc). H NMR (keto
1
form) (Acetone-d6) d 7.50±7.33 (5H; m; Ar H's), 6.64
(1H; d, JF H=47.1; Hb). 1H NMR (hydrate form)
(Acetone-d6) dd 7.50±7.33 (5H; m; Ar H's), 5.69 (1H; d,
JF H=44.7; Hb). 13C NMR (keto+hydrate form)
(Acetone-d6) d 188.3 (d, J=24.5; keto form Ca), 171.8
(hydrate form COOH), 160.5 (keto form COOH),
135.1±127.3 (m; keto+hydrate form Ar C's), 96.3 (d,
J=27.2; hydrate form Ca), 94.0 (d, J=177.0; keto form
Cb), 93.2 (d, J=184.2; hydrate form Cbb). HRMS: calcd
for C9H9FO4 200.0485, found 200.0512.
3. (a). Nakai, C.; Nozaki, M.; Hayaishi, O. Biochem. Biophys.
Res. Commun. 1975, 67, 590. (b). Lindblad, B.; Lindstedt, G.;
Lindtedt, S.; Rundgren, M. J. Biol. Chem. 1977, 252, 5073. (c).
Leinberger, R.; Hull, W. E.; Simon, H.; Retey, J. Eur. J. Bio-
chem. 1981, 117, 311. (d). Roche, P. A.; Moorehead, T. J.;
Hamilton, G. A. Arch. Biochem. Biophys. 1982, 216, 62.
4. Lindstedt, S.; Odelhog, B.; Rundgren, M. Biochemistry
1977, 16, 3369.
5. Lindblad, B.; Lindstedt, G.; Lindstedt, S. J. Am. Chem.
Soc. 1970, 92, 7446.
6. (a). Rundgren, M. J. Biol. Chem. 1977, 252. 5094. (b).
Rundgren, M. Eur. J. Biochem. 1983, 133, 657.
7. Goodwin, S.; Witkop, B. J. Am. Chem. Soc. 1975, 79, 179.
(b). Groves, J. T.; Van der Puy, M. J. Am. Chem. Soc. 1976,
98, 5290.
8. Hamilton, G. A. Prog. Bioorg. Chem. 1971, 1, 83.
9. Guro€, G.; Daly, J. W.; Jerina, D. M.; Renson, J.; Witkop,
B.; Undenfriend, S. Science 1967, 157, 1524.
10. Parisi, M. F.; Abeles, R. H. Biochemistry 1992, 31, 9429.
11. (a). Ellis, M. K.; Whit®eld, A. C.; Gowans, L. A.; Auton,
T. R.; Provan, W. M.; Lock, E. A.; Smith, L. L. Toxicol. Appl.
Pharmacol. 1995, 133, 12. (b). Secor, J. Plant Physiol. 1994,
106, 1429. (c). Lindstedt, S.; Holme, E.; Lock, E.; Hjalmarson,
O.; Strandvik, B. Lancet 1992, 340, 813. (d). Schulz, A.; Ort,
O.; Beyer, P.; Kleinig, H. FEBS 1993, 318, 162.
12. Ayi, A. I.; Remli, M.; Condom, R.; Guedj, R. J. Fluorine
Chem. 1981, 17, 565.
13. Darzens, G. Compt. Rend. 1904, 139, 1214.
14. Lindsteadt, S.; Rundgren, M. Biochim. Biophys. Acta
1982, 704, 66.
15. Pascal, Jr. R. A.; Oliver, M. A.; Chen Y.-c. Biochemistry
1985, 24, 3158.
16. Lee, D. L.; Prisbylla, M. P.; Provan, W. M.; Fraser, T.;
Mutter, L. C. Weed Science 1997, 45, 601.
17. Schulz, A.; Ort, O.; Beyer, P.; Kleinig, H. FEBS 1993, 318,
162.
3-Fluoro-p-nitrophenylpyruvic acid (14b). This com-
pound was obtained as a white solid in a 25% yield. mp
88 ꢀC dec. Rf=0.38 (20% MeOH/EtOAc). 1H NMR
(hydrate form) (D2O) d 8.17 (2H; d, J=8.4; Ar H's),
7.57 (2H; d, J=8.7; Ar H's), 5.74 (1H; d, JF H=44.7;
Hb). 13C NMR (D2O) d 172.9 (COOH), 148.0 (Ar C-
4's), 141.1 (d, J=20.4; Ar C-1's), 128.3, (d, J=7.7; Ar
C-2, Ar C-6's), 123.2 (Ar C-3, Ar C-5's), 93.6 (d,
J=180.4, Cb), 93.3 (d, J=27.6; Ca). HRMS: calcd for
C9H8FNO6 245.0335, found 245.0330.
3-Fluoro-o-chlorophenylpyruvic acid (14c). This com-
pound was obtained as a white solid in a 37% yield. mp
72±73 ꢀC. Rf=0.43 (20% MeOH/EtOAc). 1H NMR
(keto form) (Acetone-d6) d 7.62±7.42 (4H; m; Ar H's),
6.94 (1H; d, J=46.5; Hb). 1H NMR (hydrate form)
(Acetone-d6) d 7.62±7.42 (4H; m; Ar H's), 6.28 (1H; d,
J=45.0; Hb). 13C NMR (keto+hydrate form) (Acetone-
d6) d 189.2 (d, J=24.2; keto form Ca), 172.2 (hydrate
form COOH), 161.8 (keto form COOH), 135.1±127.3
(m; keto+hydrate form Ar C's), 94.3 (d, J=28.5;
hydrate form Ca), 91.4 (d, J=183.6; keto form Cb), 90.2

T. Ling et al. / Bioorg. Med. Chem. 7 (1999) 1459±1465
1465
18. Knox, W. E. Methods in Enzymol. 1955, 2, 289.
19. Buckthal, D. J.; Roche, P. A.; Moorehead, T. J.; Forbes,
B. J. R.; Hamilton, G. A. Methods in Enzymol. 1987, 142, 132.
20. Smith, P. K. Anal. Biochem. 1985, 150, 76.
21. Svoboda, J.; Kocfeldova, Z.; Palecek, J. Collect. Czech.
Chem. Commun. 1988, 53, 822.
22. Legters, J.; Thijs, L.; Zwanenburg, B. Recl. Trav. Chim.
Pays-Bas 1992, 111, 1.