A novel type of structurally simple nonpeptide inhibitors for α- chymotrypsin. Induced-fit binding of methyl 2-allyl-3-benzene-propanoate to the S2 subsite pocket

Article abstract of DOI:10.1016/S0968-0896(97)10038-4

Unexpectedly, methyl and benzyl esters of 2-allyl-3-benzenepropanoic acid were found to be not substrates but potent competitive inhibitors for α-chymotrypsin. The inhibitory property of the structurally simple nonpeptidic compounds is ascribed to their high binding affinity to the enzyme at the S₂ rather than S₁ subsite pocket. These inhibitors exist in a flexible form in solution, but as they bind to the enzyme bulky contrained conformers present in a minute concentration play an important role, forming tighter enzyme-inhibitor complexes by binding to the large hydrophobic S₂ pocket. The contrained conformers are thought to be resulted from intramolecular CH/π interactions between a vinylic proton and the aromatic π-electron cloud in the inhibitor molecules. These compounds constitute novel examples of the induced-fit binding inhibitor of possibly simplest structure.

Full text of DOI:10.1016/S0968-0896(97)10038-4

BIOORGANIC &
MEDICINAL
CHEMISTRY
Bioorganic & Medicinal Chemistry 6 (1998) 239±249
A novel type of structurally simple nonpeptide inhibitors for
ꢀ-chymotrypsin. Induced-®t binding of methyl 2-allyl-3-benzene-
propanoate to the S2 subsite pocket
Dong H. Kim*, Zhi-Hong Li, Soo Suk Lee, Jeong-il Park, Sang J. Chung
Center for Biofunctional Molecules and Department of Chemistry, Pohang University of Science and Technology, San 31 Hyojadong,
Pohang 790-784, Korea
Received 21 August 1997; accepted 31 October 1997
Abstract
Unexpectedly, methyl and benzyl esters of 2-allyl-3-benzenepropanoic acid were found to be not substrates but potent
competitive inhibitors for ꢀ-chymotrypsin. The inhibitory property of the structurally simple nonpeptidic compounds
is ascribed to their high binding anity to the enzyme at the S2 rather than S1 subsite pocket. These inhibitors exist in a
¯exible form in solution, but as they bind to the enzyme bulky contrained conformers present in a minute concentra-
.
tion play an important role, forming tighter enzyme inhibitor complexes by binding to the large hydrophobic S2
pocket. The contrained conformers are thought to be resulted from intramolecular CH/ꢁ interactions between a vinylic
proton and the aromatic ꢁ-electron cloud in the inhibitor molecules. These compounds constitute novel examples of
the induced-®t binding inhibitor of possibly simplest structure. # 1998 Elsevier Science Ltd. All rights reserved.
Keywords: ꢀ-Chymotrypsin, methyl 2-allyl-3-benzenepropanoate, competitive inhibitors, induced-®t binding, CH/ꢁ
interaction.
1. Introduction
We found that 2a is an inhibitor rather than a substrate
for the enzyme. This communication addresses the nat-
ure and the origin of the enzyme inhibitory property
exhibited by 2a towards ꢀ-chymotrypsin.
ꢀ-Chymotrypsin (EC 3. 4. 21. 1) is a prototypic serine
protease which catalyzes the hydrolysis of protein at the
carboxyl side of amino acid residue having an aromatic
side chain [1]. Hydrolysis of esters of such amino acids is
also accelerated by the enzyme, rate of which is even
greater than that for amide hydrolysis, and thus this
enzyme is useful for kinetic resolution of esters of race-
mic acids having a hydrophobic substituent at the ꢀ-
position [2]. Indeed, optically active 2-vinyl-3-benzene-
propanoic acid which we needed in connection with
preparing an optically active carboxypeptidase A inhi-
bitor could be obtained readily by the enzymic kinetic
resolution of racemic methyl 2-vinyl-3-benzenepro-
panoate (1) using ꢀ-chymotrypsin (Scheme 1) [3]. How-
ever, when we subjected methyl ester of its homologous
acid, methyl 2-allyl-3-benzenepropanoate (2a) to the
enzymic hydrolysis, the ester resisted to be hydrolyzed.
2. Results
2.1 Synthesis of inhibitors
Racemic 2a was prepared by allylation of 3-benzene-
propanoic acid using allyl bromide in the presence of
LDA followed by esteri®cation of the product with a
large excess of methanol (Scheme 2). Similarly was pre-
pared 4 (Scheme 2). Compound 2b was obtained by
allowing 9 to react with benzyl bromide in the presence
of K2CO3 in DMF. In order to gain knowledge on the
stereospeci®city of the ꢀ-chymotrypsin inhibition, we
needed to prepare optically active 2a. The requisite com-
pounds were synthesized using an Evans chiral auxilliary
as depicted in Scheme 3. Synthesis of 3 is illustrated in
Scheme 4, in which 17 was separated from a mixture of
17 and 18 and puri®ed by column chromatography.
*Corresponding author. Tel: 82-562-279-2101; fax: 82-562-279-
5877; e-mail: dhkim@vision.potech.ac.kr
0968-0896/98/$19.00 # 1998 Elsevier Science Ltd. All rights reserved
PII: S0968-0896(97)10038-4

240
D. H. Kim et al./Bioorg. Med. Chem. 6 (1998) 239±249
Table 1
Inhibitory potency against ꢀ-chymotrypsin
Compound no.
Ki,ꢂM
(R)-2a
(S)-2a
736
731
56
2b
3
3870
Ða
4
5
Ðb
Ðb
Ðb
6
7
aCompound 4 is a substrate for ꢀ-chymotrypsin.
bKinetic parameters cannot be obtained due to their extreme
insolubility.
2.2 ꢀ-Chymotrypsin inhibition
21% in 10 days. Compounds 5±7 resisted the enzymic
hydrolysis. No kinetic experiments could be performed
with these compounds because of their extreme insolu-
bility in water.
All of the compounds synthesized were tested for
substrate and as inhibitors for ꢀ-chymotrypsin. The
inhibitory activity assays were carried out using succi-
nyl-Ala-Ala-Pro-Phe-p-nitroanilide (Suc-AAPFpNA) as
substrate at pH 7.5. Semireciprocal plots of the initial
velocity of the enzymic action for the substrate in the
presence of each inhibitor at di€erent concentrations
against concentrations of the inhibitors were con-
structed according to the method of Dixon [4], from
which inhibitory constants (Ki) were estimated. The
Dixon plots represented by Fig. 1 suggest strongly that
the enzyme is inhibited in a reversible competitive fash-
ion. The inhibitory constants thus obtained are listed in
Table 1. Compound 4 was hydrolyzed by the enzyme,
yielding the corresponding acid having the (R)-con®g-
uration with (S)-4 being intact. Compound 3 was
hydrolyzed very slowly and only in partial. On the basis
of the amount of 0.02 M NaOH solution consumed the
hydrolysis of 3 was estimated to proceed only up to
3. Discussion
We were surprised to observe that 2a which is a
homolog of ꢀ-chymotrypsin substrate 1 resists to be
hydrolyzed by ꢀ-chymotrypsin. When the compound
was then tested for any inhibitory activity against the
enzyme, we found that it inhibits ꢀ-chymotrypsin com-
petitively with the Ki value of 730 ꢂM. It is remarkable
that such an abrupt change from a substrate to an inhib-
itor results upon the increment of one methylene unit.
The observation suggests that 2a binds the active site of
the enzyme but not in a productive mode, i.e., the phenyl
ring and the ester group of 2a are not accommodated in
the S1 subsite pocket and the catalytic site, respectively
[5]. When both enantiomers of 2a were evaluated as
Scheme 1.

D. H. Kim et al./Bioorg. Med. Chem. 6 (1998) 239±249
241
Scheme 2. Reagents, conditions, and (yields): (a) LDA (2.05 eq.), allyl bromide (1.05 eq.), 0ꢀC, THF (82%); (b) H2, Pd/C, MeOH, 3 h
(96%); (c) LDA (2.05 eq.), RX (1.05 eq.), 0ꢀC, THF; (d) MeOH, H+; (e) MeOH, H+ (for 2a, 95%) or K2CO3 (1.2 eq.), benzyl bro-
mide (1.0 eq.), DMF (for 2b, 87%).
Scheme 3. Reagents, conditions, and (yields): (a) LDA (1.0 eq.), allyl bromide (1.0 eq.), 78ꢀC, THF (79%); (b) MeONa (1.05 eq.),
0ꢀC, THF (96%).

242
D. H. Kim et al./Bioorg. Med. Chem. 6 (1998) 239±249
inhibitors for ꢀ-chymotrypsin, they were found to be
equally potent (Fig. 1), demonstrating that the inhibi-
tion is non-stereospeci®c. These results taken together
suggest strongly that the aromatic ring of 2a is not
accommodated in the primary substrate recognition
pocket (S1 subsite pocket) of the enzyme [6]. Since the S1
subsite pocket exhibits stereospeci®city for the `L' iso-
mer in its interaction with ligands, if 2a anchors in the
S1 pocket, the stereospeci®city should be manifested in
the inhibition. The lack of stereospeci®city shown by
optically active 2a tends to exclude also a possible
explanation that the inhibition may be arisen as a result
of reverse binding of 2a to the enzyme, i.e., the binding
in which the allyl moiety anchors in the hydrolytic sub-
site, because if it would, there is expected some di€er-
ences in the inhibitory potency between the
enantiomeric pair. Anchoring of the inhibitor in an
auxiliary binding subsite most likely in the S2 subsite
Scheme 4. Reagents, conditions, and (yields): (a) LiAlH4 (0.6 eq.), 0ꢀC, THF (95%); (b) PivCl, pyridine, CH2Cl2 (88%); (c) N-bro-
mosuccinimide (1.2 eq.), pyridinium poly(hydrogen ¯uoride), ether (65%); (d) tert-BuOK (1.2 eq.), 78ꢀC, THF (40%); (e) DIBAL (2
eq.), 78ꢀC, CH2Cl2 (94%) and Jones' reagent, acetone:water=1:1 (78%); (f) MeOH, H+ (84%).
Fig. 1. Dixon plots for inhibition of ꢀ-chymotrypsin-catalyzed hydrolysis of Suc-AAPFpNA by (S)-2a (A) and (R)-2a (B): 0.036 M
Tris bu€er, pH 7.8; 0.045 M in CaCl2; [E]=1.0Â10 M; Temperature 25ꢀC.
8

D. H. Kim et al./Bioorg. Med. Chem. 6 (1998) 239±249
243
pocket is conjectured. Figure 2 depicts the possible
binding mode of 2a. The S2 subsite pocket is known to
be a large hydrophobic pocket consisting of His-57, Ile-
99, and Trp-219, and prefers a bulky residue [7]. The
proposed binding mode of 2a is supported by the
observation that the benzyl ester 2b is an inhibitor hav-
ing binding anity (Ki=56.2 ꢂM) 13 times that of 2a. It
may be envisioned that in the binding of 2b to ꢀ-chy-
motrypsin, the S1 subsite pocket, dimensions of which is
known to be 10±12A in depth and (3.5±4.0A)Â(5.5±
6.5A) in cross section [8] is also occupied by the phenyl
ring in the ester portion of 2b to strengthen the binding
(Fig. 3).
The observation that 4 in which the double bond in
2a is replaced with an acetylenic bond is not an inhibitor
but a substrate for ꢀ-chymotrypsin suggests that the
critical structural feature that is responsible for the
unusual mode of binding in the case of 2a is the terminal
double bond. We propose that there may operate non-
bonded interactions, mostly of delocalization (charge
transfer from ꢁ to ꢃ*) and dispersion forces, between a
vinylic proton of the double bond and the ꢁ-electron
cloud of the phenyl ring in the case of 2a, forming a
hydrophobic core that binds the S2 pocket. Noncovalent
attractive interactions between CH and ꢁ electron cloud
have recently been proposed by Nishio et al. as an
Fig. 2. Methyl 2-allyl-3-benzenepropanoate (2a) binds ꢀ-chymotrypsin at the S2 subsite pocket: Since the S2 subsite pocket is larger
than the primary substrate recognition pocket (S1 pocket), the binding to the S2 pocket is expectedly much tighter than the binding to
the S1 pocket due to increased hydrophobic interactions. Consequently, the equilibrium shifts in favor of 2a binding to the S2 subsite
pocket.
Fig. 3. Schematic representation that shows the binding mode of benzyl 2-allyl-3-benzenepropanoate (2b). Compound 2b binds ꢀ-
chymotrypsin with much higher binding anity than 2a does as a result of the additional hydrophobic interactions that arises from
anchoring the phenyl ring of the ester portion in the S1 subsite pocket.

244
D. H. Kim et al./Bioorg. Med. Chem. 6 (1998) 239±249
important binding force in the biological system [9].
Such constrained conformer in the case of 2a was sug-
gested from the PM3 calculations [10]: The distance
between the center of the phenyl ring and the internal
vinylic proton in one of energy minimized conformers of
2a is 2.99A which falls within the range of noncovalent
interactions (Fig. 5(A)) [9,11]. The ®nding that 4 is not
an inhibitor but a substrate for the enzyme can now be
envisaged on the following ground that because of the
linear nature of the triple bond, the intramolecular CH/
ꢁ interaction observed in 2a is not feasible in 4, and no
bulky constrained conformer postulated for 2a can
be formed. Accordingly, 4 binds the enzyme in a pro-
ductive mode for the catalytic hydrolysis with its aro-
matic ring being ®tted into the S1 pocket as shown in
Fig. 4.
It was thought that the binding mode proposed for 2a
would become impossible if the internal vinylic proton is
replaced with a ¯uorine and such a compound should be
hydrolyzed by the enzyme. Indeed, 3 was a substrate for
the enzyme although the rate of the hydrolysis was
extremely slow. However, unexpectedly, 3 also exhibited
an inhibitory activity for the enzyme albeit very weak
with Ki value of 3.87 mM. We thought initially that the
external vinylic proton may also interact with the ꢁ
electron cloud of the phenyl ring, leading to form
another constrained conformer such as that depicted in
Fig. 5(B), but the PM3 calculations revealed that the
distance between the external vinylic proton and the
center of the aromatic ring is 4.24A which is too far for the
CH/ꢁ interactions. The molecular origin of the ꢀ-chy-
motrypsin inhibitory action of 3 is not apparent to us.
Recently, Shimohigashi et al. reported that H-d-Leu-
Phe-O-CH2Ph is a potent inhibitor for ꢀ-chymotrypsin
having the Ki value of 3.6 ꢂM [12]. Its inhibitory prop-
erty was attributed to the bulky constrained conforma-
tion (hydrophobic core) resulted from intramolecular
CH/ꢁ interactions between the isobutyl group of d-Leu
and the phenyl ring of Phe. The bulky hydrophobic core
thus formed was proposed to anchor in the S2 subsite
with the phenyl ring in the ester portion being ®tted in
the S1 subsite pocket. The formation of the bulky
hydrophobic core was substantiated by the 1H NMR
spectrum in which the protons of the isobutyl side chain
of d-Leu experience a substantial up®eld chemical
shifts. In comparison, no such up®eld chemical shift was
Fig. 4. Schematic representation that shows methyl 3-benzene-2-propagylpropanoate (4) binds ꢀ-chymotrypsin in a productive mode
resulting in the hydrolysis of the ester.
Fig. 5. Stick and ball drawing of constrained conformers of 2a generated by MOPAC using PM3 program.

D. H. Kim et al./Bioorg. Med. Chem. 6 (1998) 239±249
245
noted in the 1H NMR spectrum of 2a, demonstrating
that in the case of 2a the hydrophobic core such as that
observed with H-d-Leu-Phe-O-CH2Ph is not formed
to any meaningful concentration, which suggests that
the non-bonded interactions of the vinylic proton with
the aromatic ꢁ-electron cloud in 2a become to play
an important role only when 2a binds the enzyme. Thus,
in the case of 2a which exists in a ¯exible form in solu-
tion, the hydrophobic constrained conformer may
develop as it binds to the S2 subsite pocket. Presumably,
there may be in e€ect hydrophobic and van der Waals
interactions between the S2 subsite pocket and the con-
strained conformer upon binding. The hydrophobic
interactions are known to increase in proportion to the
contact area of the interacting moieties both at mol-
ecular and macromolecular levels [13]. Thus, it may be
plausible that since the S2 pocket is larger than the S1
pocket, the binding of 2a to the S2 pocket is expectedly
much tighter compared with that resulting from the
anchoring of just the phenyl ring alone in the S1 pocket
as illustrated schematically in Fig. 2. Therefore, the Ki
value is much smaller than the Km value in the case of
binding 2a to ꢀ-chymotrypsin, and as a consequence 2a
behaves as a competitive inhibitor (Fig. 2). Thus, the
inhibition of ꢀ-chymotrypsin by 2a may be referred to
the binding of 2a to the S2 subsite pocket in an induced-
®t manner, but double-induced ®t binding, in which the
enzyme as well as the inhibitor change their conforma-
tions upon forming complex may not be excluded.
The induced-®t theory proposed by Koshland has
provided a plausible explanation for the catalytic action
of enzymes since its inception in 1958 [14]. According to
the theory, binding of a substrate to an enzyme induces
conformational changes of the enzyme and these con-
formational changes would result in the catalytic func-
tional groups to orient themselves for the ensuing
catalytic process. Thus, the theory focuses the con-
formational changes of the enzyme in explaining the
enzymic action, but recent works have drawn attention
on the possibility of conformational changes in ligands
[15]. The inhibition of the enzymic activity of ꢀ-chymo-
trypsin by 2a may constitute a novel example of such
induced-®t binding of a ligand to enzyme and 2a
may possibly be a minimum structural unit for such
binding.
the large hydrophobic S2 subsite pocket. Such a mode of
binding in the case of 2a and analogs is possibly ascri-
bed to the unique property of these molecules to form
bulky constrained conformers resulting from the intra-
molecular CH/ꢁ interactions between the internal vinylic
proton and the phenyl aromatic ring. Although such con-
strained conformers exist in a minute quantity, they are
thought to bind preferentially to the large hydrophobic
pocket of the S2 subsite in the active site of ꢀ-chymo-
trypsin because the hydrophobic and van der Waals
forces involved in the binding are stronger than those
resulting from the normal binding of 2a and its analogs
to the S1 subsite. Inhibitors 2a and its analogs exemplify
the induced-®t binding of a ¯exible ligand to an enzyme.
5. Experimental
Infrared (IR) spectra were obtained with a Bruker
Equinox 55 FT-IR spectrometer. 1H and 19F NMR
spectra were recorded on a Bruker AM 300 NMR
spectrometer in CDCl3 using tetramethylsilane (TMS)
as the internal reference. Chemical shifts of 19F NMR
spectra are reported in ꢄ values up®eld from the 19F
signal of CF3CO2H. Low and high resolution mass
spectral analyses were performed at the Inter-University
Center for Natural Science Research Facilities, Seoul
National University. Optical rotations were measured
on Rudolph Research Autopol III digital polarimeter.
Flash chromatography was performed using silica gel 60
(230ꢁ400 mesh). Thin layer chromatography (TLC) was
carried out on silica coated glass sheets (Merck silica gel
60 F-254). The pH-stat used in enzymic kinetic resolu-
tion is a Cole-Parmer pH/mV controller type 5997-20
which is coupled with a Orion gel-®lled combination pH
electrode and Pharmacia Peristaltic Pump P-1. Enzyme
assays were carried out using Hewlett-Packard 8452a
UV spectrometer ®tted with a cell temperature con-
troller. ꢀ-Chymotrypsin and its substrate, succinyl-Ala-
Ala-Pro-Phe-p-nitroanilide (sucAAPFpNA) were pur-
chased from Sigma Chemical Co.
5.1 2-Allyl-3-bezenepropanoic acid (9)
To an ice-chilled solution of diisopropylamine (5.9 ml,
42.0 mmole) in dried THF (30 ml) was added slowly n-
butyllithium (16.8 ml of 2.5 M solution in n-hexane,
42.0 mmole) under nitrogen atmosphere. After the mix-
ture was stirred for 30 min, hydrocinnamic acid (8, 3.0 g,
20.0 mmole) in dried THF (20 ml) was added dropwise
over a period of 20 min. The resulting yellowish solution
was stirred for 40 min, and then allyl bromide (1.8 ml,
21.0 mmole) was added dropwise. The reaction mixture
was stirred for 3 h to give a colorless solution. The pH of
the solution was adjusted to 2.5 with 3 N HCl solution.
The organic layer was separated and extracted with
4. Conclusion
Methyl 2-allyl-3-benzenepropanoate 2a which is a
homolog of methyl 2-vinyl-3-benzenepropanoate,
a
substrate of ꢀ-chymotrypsin resists the hydrolysis by ꢀ-
chymotrypsin. Rather, it inhibited the catalytic activity
of the enzyme. This unexpected inhibitory property
shown by 2a and its analogs towards ꢀ-chymotrypsin is
attributed to their high binding anity to the enzyme at

246
D. H. Kim et al./Bioorg. Med. Chem. 6 (1998) 239±249
saturated sodium bicarbonate solution. The aqueous
layer was acidi®ed with 3 N HCl solution to pH 2,
saturated with sodium chloride, and extracted with ethyl
acetate (50 mlÂ3). The combined extracts were dried
over anhydrous MgSO4, ®ltered, and evaporated under
reduced pressure. The residue was puri®ed by column
chromatography, eluting with a solution of ethyl acetate
n-butyllithium (10 ml of 2.5 M solution in n-hexane,
25 mmole) under nitrogen atmosphere. After the mix-
ture was stirred for 30 min and cooled to 78ꢀC, (4S)-3-
(3-phenylpropanoyl)-4-benzyl-1,3-oxazolidin-2-one [16]
((S)-11, 7.73 g, 25 mmole) in dried THF (20 ml) was
added dropwise over a period of 20 min. The resulting
yellowish solution was stirred for 1.5 h, and then allyl
bromide (2.2 ml, 25 mmole) was added dropwise. The
reaction mixture was allowed to warm to room tem-
perature over 7 h, then quenched with saturated ammo-
nium chloride solution and partitioned between water
and ether. The combined ether extracts were washed
with saturated sodium bicarbonate solution and brine,
then evaporated under reduced pressure. The oily resi-
due was subjected to ¯ash chromatography, eluting with
a solution of methylene chloride and petroleum ether
(50:50) to give (S)-12 [17] (6.9 g, 79%): 1H NMR
(CDCl3) ꢄ 7.18±7.36 (m, 10H), 5.82±5.93 (m, 1H), 5.08±
5.18 (m, 2H), 4.32±4.51 (m, 2H), 4.01±4.05 (dd, 1H),
3.84 (t, 1H), 3.25 (dd, 1H), 2.85±3.02 (m, 2H), 2.67 (dd,
1H), 2.30±2.62 (m, 2H).
and n-hexane (25:75) to give 9 (3.12 g, 82% yield) as a
1
colorless oil: IR (neat) 1712 (C=O) cm
;
1H NMR
(CDCl3) ꢄ 7.13±7.35 (m, 5H), 5.65±5.74 (m, 1H), 4.98±
5.06 (m, 2H), 2.91±2.97 (m, 1H), 2.70±2.81 (m, 2H),
2.26±2.39 (m, 2H).
5.2 Methyl 2-allyl-3-bezenepropanoate (2a)
To an ice-chilled solution of 9 (2.83 g, 15 mmole) in
anhydrous methanol (50 ml) was added 2 ml of acetyl
chloride. The resulting mixture was stirred for 12 h at
room temperature, then evaporated under reduced
pressure. The residue was dissolved in ether, and the
ether solution was washed with water (30 ml), 5%
sodium bicarbonate solution (30 mlÂ3), and brine
(30 ml). The organic solution was dried over anhydrous
MgSO4, ®ltered, and evaporated under reduced pres-
sure. The residue was puri®ed by column chromato-
graphy, eluting with a solution of ethyl acetate and n-
hexane (10:90) to give 2a (3.20 g, 95%) as a colorless oil:
A
methanol solution of sodium methoxide
(7.3 mmole) was added dropwise to the stirred solution
of (S)-12 (2.44 g, 7 mmole) in 20 ml of anhydrous
methanol at 0ꢀC under nitrogen atmosphere, then stir-
red for 2 h. The reaction mixture was quenched with
saturated ammonium chloride solution and the remain-
ing methanol was evaporated under reduced pressure.
The residue was diluted with ethyl acetate, and washed
with water and brine, then dried over anhydrous
MgSO4, ®ltered, and evaporated under reduced pres-
sure. The residue was subjected to ¯ash column chro-
matography, eluting with a solution of ethyl acetate and
hexane (20:80) to give (S)-2a (1.37 g, 96%): IR (neat)
1
IR (neat) 1722 (C=O) cm 1; H NMR (CDCl3) ꢄ 7.13±
7.29 (m, 5H), 5.70±5.80 (m, 1H), 5.01±5.09 (m, 2H), 3.59
(s, 3H), 2.91±2.99 (m, 1H), 2.73±2.81 (m, 2H), 2.26±2.40
(m, 2H); MS (m/z) 204, 172, 163, 144, 131, 103, 91.
5.3 Benzyl 2-allyl-3-benzenepropanoate (2b)
1
To a stirred mixture of 9 (380 mg, 2 mmole) and
K2CO3 (332 mg, 2.4 mmole) in 7 ml of DMF was added
slowly benzyl bromide (360 mg, 2 mmole) at 0ꢀC. The
stirring was continued for about 2 h, and then the reac-
tion mixture was diluted with ethyl acetate. The solution
thus obtained was washed with sodium thiosulfate
(20 ml), 5% sodium bicarbonate solution (20 ml), 10%
citric acid (20 ml), and brine (20 ml), successively, then
dried over anhydrous MgSO4, and evaporated under
reduced pressure. The oily residue was puri®ed by col-
umn chromatography, eluting with a solution of ethyl
3018, 2961, 1722 (C=O), 1212 cm 1; H NMR (CDCl3)
ꢄ 7.13±7.29 (m, 5H), 5.70±5.80 (m, 1H), 5.10±5.09 (m,
2H), 3.59 (s, 3H), 2.91±2.99 (m, 1H), 2.73±2.81 (m, 2H),
2.26±2.40 (m, 2H); MS (m/z) 204, 172, 163, 144, 131,
113, 103, 91; [ꢀ]D=+30.0ꢀ(c=1.0, EtOH).
5.5 Methyl (R)-2-allyl-3-benzenepropanoate ((R)-2a)
This was prepared in a fashion similar to that used for
the preparation of (S)-12 using (R)-11: 1H NMR
(CDCl3) ꢄ 7.18±7.36 (m, 10H), 5.82±5.93 (m, 1H), 5.08±
5.18 (m, 2H), 4.32±4.51 (m, 2H), 4.01±4.05 (dd, 1H),
3.84 (t, 1H), 3.25 (dd, 1H), 2.85±3.02 (m, 2H), 2.67 (dd,
1H), 2.30±2.62 (m, 2H). (R)-12 was then treated with a
methanol solution of sodium methoxide in a similar
acetate and hexane (5:95) to give 2b (485 mg, 87%): IR
1
(neat) 3067, 3031, 1733 (C=O), 1496, 1449, 1181 cm
;
1H NMR (CDCl3) ꢄ 7.15±7.35 (m, 5H), 5.77 (m, 1H),
5.02±5.11 (m, 2H), 5.05 (s, 2H), 2.98 (m, 1H), 2.79±2.87
(m, 2H), 2.44 (m, 1H), 2.30 (m, 1H); MS (m/z) 281
(M++1) 280, 239, 228, 190, 181, 144, 127, 111.
fashion to that used for preparation of (S)-2a to give
1
(R)-2a: IR (neat) 3015, 2958, 1721 (C=O), 1220 cm
;
1H NMR (CDCl3) ꢄ 7.11±7.31 (m, 5H), 5.68±5.76 (m,
1H), 5.07±5.10 (m, 2H), 3.55 (s, 3H), 2.89±2.96 (m, 1H),
2.72±2.79 (m, 2H), 2.24±2.37 (m, 2H); MS (m/z) 204,
172, 163, 144, 131, 113, 103, 91; [ꢀ]D= 28.5ꢀ(c=1.0,
EtOH).
5.4 Methyl (S)-2-allyl-3-benzenepropanoate ((S)-2a)
To an ice-chilled solution of diisopropylamine (3.5 ml,
25 mmole) in dried THF (100 ml) was added slowly

D. H. Kim et al./Bioorg. Med. Chem. 6 (1998) 239±249
247
5.6 Methyl 2-propagyl-3-benzenepropanoate (4)
3.60 (m, 2H), 2.63±2.66 (m, 2H), 2.12±2.17 (m, 2H),
1.90±1.94 (m, 1H), 1.26 (br, 1H).
3-Benzene-2-propagylpropanoic acid (10) was prepared
in a similar fashion to that used for the preparation of 9
using propagyl bromide instead of allyl bromide: IR
5.11 1-Benzyl-3-butenyl pivaloate (14)
1
(neat) 1720 (C=O) cm 1; H NMR (CDCl3) ꢄ 7.13±7.26
To a solution of 13 (4 g, 22.7 mmole) in 20 ml of dry
CH2Cl2 was added 10 ml of dry pyridine. The resulting
mixture was chilled in an ice-bath. To the ice-chilled
solution was added dropwise pivaloyl chloride (3.08 ml,
25 mmole) over a period of 30 min under argon atmo-
sphere. The reaction mixture was allowed to warm to
room temperature, stirred for 14 h, then evaporated
under reduced pressure, and the residue was puri®ed by
column chromatography, eluting with a solution of
ether and petroleum ether (50:50) to give 14 (5.2 g, 88%)
(m, 5H), 3.02 (dd, 1H), 2.82±2.95 (m, 2H), 2.37±2.42 (m,
2H), 2.00 (t, 1H). Compound 10 was then converted
into methyl ester in a fashion similar to that used for the
preparation of 2a to give 4: IR (neat) 2349, 1737 (C=O),
1
1442, 1370 cm
;
1H NMR (CDCl3) ꢄ 7.18±7.32 (m,
5H), 3.67 (s, 3H), 3.06 (dd, 1H), 2.85±2.98 (m, 2H),
2.42±2.46 (m, 2H), 2.06 (t, 1H) ; MS (m/z) 203 (M++1),
171, 163, 141, 131, 115, 103, 91, 79.
5.7 Methyl 2-benzylpentanoate (5)
as a colorless oil: IR (neat) 2976 (s, C-H), 1738 (s,
1
C=O), 1653, 1640 (w, C=C), 1161 (s, C-O) cm
;
1H
To a solution of 2a (200 mg, 0.98 mmole) in anhy-
drous methanol (10 ml) was added a catalytic amount of
10% Pd/C and the resulting solution was stirred under
hydrogen gas (1 atm) for 3 h. The catalyst was removed
by ®ltration and the ®ltrate was evaporated under
NMR (CDCl3) ꢄ 7.14±7.30 (m, 5H), 5.72±5.86 (m, 1H),
5.02±5.08 (m, 2H), 3.94±3.98 (d, 2H), 2.64±2.66 (d, 2H),
2.06±2.15 (m, 3H), 1.21±1.26 ( m, 9H).
5.12 1-Benzyl-3-¯uoro-4-bromobutyl pivaloate (15) and
1-benzyl-3-bromo-4-¯uorobutyl pivaloate (16)
reduced pressure to give 5 (195 mg, 96%): IR (neat)
1
1730 (C=O) cm
;
1H NMR (CDCl3) ꢄ 7.13±7.29 (m,
5H), 3.60 (s, 3H), 2.68±2.96 (dq, 2H), 2.66±2.70 (m, 1H),
1.45±1.65 (m, 2H), 1.23±1.35 (m, 2H), 0.88 (t, 3H).
To a stirred solution of N-bromosuccinimide (3.79 g,
21.3 mmole) obtained by dissolving it in ether (20 ml) in
a plastic bottle was added pyridinium poly(hydrogen
¯uoride) solution (20 ml) under nitrogen atmosphere.
The mixture was stirred in an ice bath for 15 min, at
which time 14 (5.03 g, 19.3 mmole) dissolved in ether
(10 ml) was added slowly. The reaction mixture was
allowed to warm to room temperature over 30 min.
After being stirred at room temperature for 2 h, the
reaction mixture was poured into saturated sodium
bicarbonate solution (50 ml), and extracted each with
ether (100 mlÂ4), dried over anhydrous Na2SO4, and
evaporated under reduced pressure. The resulting yel-
lowish oil was chromatographed eluting with a solution
of ethyl acetate and hexane (10:90) to give a mixture
(4.94 g, 65% yield) of 15 and 16 (1:1). A small amount
of 15 could be isolated and puri®ed by chromatography
5.8 Methyl 2-benzylhexanoate (6)
This was prepared in a fashion similar to that used for
the preparation of 4 using 1-iodobutane and subsequent
1
esteri®cation with methanol: IR (neat) 1731 (C=O)cm
;
1H NMR (CDCl3) ꢄ 7.13±7.30 (m, 5H), 3.60 (s, 3H),
2.70±2.96 (dq, 2H), 2.62±2.68 (m, 1H), 1.49±1.63
(m, 2H), 1.25±1.31 (m, 4H), 0.87 (t, 3H).
5.9 Methyl 2-benzylheptanoate (7)
This was prepared in a fashion similar to that used for
the preparation of 4 using 1-bromopentane and sub-
sequent esteri®cation with methanol: IR (neat) 1728
1
(C=O) cm
;
1H NMR (CDCl3) ꢄ 7.14±7.30 (m, 5H),
for identi®cation: IR (neat) 2972 (s, C-H), 1733 (s,
1
3.58 (s, 3H), 2.71±2.97 (dq, 2H), 2.62±2.68 (m, 1H),
1.49±1.65 (m, 2H), 1.26±1.32 (m, 6H), 0.86 (t, 3H).
C=O), 1161 (s, C-O) cm
;
1H NMR (CDCl3) ꢄ 7.14±
7.32 (m, 5H), 4.65±4.86 (m, 1H), 3.90±4.10 (m, 2H),
3.38±3.48 (m, 2H), 2.67±2.74 (m, 2H), 2.17±32 (m, 1H),
1.68±1.88 (m, 2H), 1.21 (s, 9H); 19F NMR (CDCl3) ꢄ
103.60, 103.66. The mixture of 15 and 16 was used
for the following reaction without separation.
5.10 2-Benzyl-4-pentenol (13)
To an ice-chilled solution of 2a (4.62 g, 22.7 mmole) in
150 ml of dried THF was added LiAlH4 (0.52 g,
13.6 mmole) slowly and stirred for 3 h at 0ꢀC. The reac-
tion mixture was quenched with 3 ml of water, dried
over anhydrous MgSO4, and ®ltered. The ®ltrate was
concentrated under reduced pressure and chromato-
5.13 1-Benzyl-3-¯uoro-3-butenyl pivaloate (17)
To a stirred and chilled ( 78ꢀC) solution of the mixture
of 15 and 16 (3.94 g, 11 mmole) in 20 ml of dried THF
was added slowly 15 ml of 1.0 M potassium tert-butoxide
in THF. The reaction mixture was allowed to warm to
30ꢀC and kept between 30 and 20ꢀC for 2 h. The
graphed to give 13 (3.79 g, 95%): IR (neat) 3064, 2976,
1
1480, 1285, 1163 cm
;
1H NMR (CDCl3) ꢄ 7.18±7.32
(m, 5H), 5.80±5.89 (m, 1H), 5.04±5.13 (m, 2H), 3.50±

248
D. H. Kim et al./Bioorg. Med. Chem. 6 (1998) 239±249
temperature of the reaction mixture was lowered to
78ꢀC, and 15 ml of ether and 0.5 ml of acetic acid were
added successively. The organic solvents were evaporated
under reduced pressure to give a mixture of 17 and 1-ben-
zyl-4-¯uorobutenyl pivaloate (18). Compound 17 (1.23 g,
40%) was separated from the mixture by column chro-
matography, eluting with a solution of ethyl acetate and
room temperature, then evaporated under reduced
pressure. The residue was dissolved in ether, and the
ether solution was washed with water (30 ml), 5%
sodium bicarbonate solution (30 mlÂ3), and brine
(30 ml). The organic solution was dried over anhydrous
MgSO4, ®ltered, and evaporated under reduced pres-
sure. The residue was puri®ed by column chroma-
tography, eluting with a solution of ethyl acetate and n-
hexane (20:80): IR (neat) 2975 (s, C-H), 1732 (s, C=O),
1
1674 (m, C=C), 1284, 1160 (s, C-O) cm
;
1H NMR
hexane (10:90) to give 3 (220 mg, 84%) as a colorless oil:
1
(CDCl3) ꢄ 7.15±7.32 (m, 5H), 4.58±4.65 (dd, 1H), 4.19±
4.36 (dd, 1H), 3.92±4.07 (m, 2H), 2.63±2.72 (m, 2H), 2.22±
2.32 (m, 3H), 1.24 (s, 9H); 19F NMR (CDCl3) ꢄ 16.0.
IR (neat) 1739 (s, C=O), 1684 (m, C=C) cm
;
1H
NMR (CDCl3) ꢄ 7.14±7.31 (m, 5H), 4.55±4.61 (dd, 1H),
4.20±4.37 (dd, 1H), 3.61 (s, 3H), 2.91±2.98 (m, 2H),
2.81±2.87 (m, 1H), 2.49±2.63 (m, 1H), 2.33±2.43 (m,
1H); 19F NMR (CDCl3) ꢄ 19.4; MS (m/z) 222, 202,
191, 162, 131, 103, 91, 77, 65; HRMS (EI+) obsd.
222.1057, calc. 222.1056 (C13H15O2F).
5.14 2-Benzyl-4-¯uoro-4-pentenoic acid (19)
To a stirred and chilled ( 78ꢀC) solution of 17
(628 mg, 2.26 mmol) in 20 ml of dry CH2Cl2 was added
dropwise diisobutylaluminium hydride (5.6 ml, 1.0
molar solution in hexane) under nitrogen atmosphere.
After 30 min stirring, the reaction mixture was trans-
ferred to a separatory funnel containing 50 ml of satu-
rated solution of sodium tartarate and 100 ml of ethyl
acetate. The organic layer was separated after vigorous
shaking, dried over anhydrous MgSO4, and evaporated
under reduced pressure. The residue was column chro-
matographed, eluting with a solution of ethyl acetate
and hexane (15:85) to give 2-benzyl-4-¯uoro-4-pentenol
5.16 Enzyme-catalyzed hydrolysis of racemic methyl 2-
propagyl-3-benzenepropanoate (4)
Racemic 4 (202 mg, 1.0 mmole) was suspended in
10 ml of 0.01 M phosphate bu€er solution. To the mix-
ture was added ꢀ-chymotrypsin (40 mg), and stirred
slowly. The pH of the reaction mixture was maintained
at 7.8 by addition of sodium hydroxide solution (0.1 N)
using a pH-stat. When 3.5 ml of the alkaline solution
was consumed after 24 h, the reaction mixture was
acidi®ed with 1N HCl solution and saturated with
sodium chloride, then extracted with ethyl acetate. The
extract was dried over anhydrous MgSO4, ®ltered, and
evaporated under reduced pressure. The residue was
puri®ed by column chromatography, eluting with a
solution of ethyl acetate and n-hexane (20:80) to give
(R)-3-benzene-2-propagylpropanoic acid and methyl
(S)-3-benzene-2-ethynylpropanoate. The spectral data
of these compounds were identical to those of respective
racemic compounds, i.e., 10 and 4.
(410 mg, 94%) as a colorless oil: IR (neat) 3361 (br, O-
1
H), 2923 (m, C-H), 1674 (s, C=C), 1030 (m, C-O) cm
;
1H NMR (CDCl3) ꢄ 7.17±7.33 (m, 5H), 4.57±4.64 (dd,
1H), 4.21±4.38 (dd, 1H), 3.52±3.64 (m, 2H), 2.68±2.71
(d, 2H), 2.22±2.36 (m, 2H), 2.10±2.17 (m, 1H), 1.39 (br,
1H); 19F NMR (CDCl3) ꢄ 17.8.
To an ice-chilled solution of 2-benzyl-4-¯uoro-4-pen-
tenol (380 mg, 1.96 mmole) in 10 ml of acetone was
added slowly the Jones' reagent until the brownish color
of the solution remains over 20 min, then 2-propanol
was added until the solution became clear. The pre-
cipitate that separated was ®ltered using a celite pad and
the ®ltrate was evaporated under reduced pressure. The
residue was diluted with 5 ml of 1 N HCl then extracted
with ethyl acetate (30 mlÂ3). The extract was dried over
anhydrous MgSO4 and evaporated under reduced pres-
sure to give a yellowish syrup (320 mg, 78%) of 19
which was used in the next reaction without puri®cation:
IR (neat) 3031 (br, COOH), 1713 (s, C=O), 1683 (s,
C=CF), 1252 (m, C-O) cm 1; 1H NMR (CDCl3) ꢄ 10.84
(br, 1H), 7.15±7.34 (m, 5H), 4.59±4.66 (dd, 1H), 4.24±4.42
(dd, 1H), 2.94±3.06 (m, 2H), 2.84±2.89 (m, 1H), 2.46±2.60
(m, 1H), 2.39±2.45 (m, 1H); 19F NMR (CDCl3) ꢄ 22.1.
5.17 Enzyme-catalyzed hydrolysis of Rracemic methyl 2-
benzyl-4-¯uoro-4-pentenoate (3)
Racemic 3 (111 mg, 0.5 mmole) was suspended in
20 ml of 0.01 M phosphate bu€er solution. To the mix-
ture was added ꢀ-chymotrypsin (10 mg), and stirred
slowly. The pH of the reaction mixture was maintained
at 7.8 by addition of sodium hydroxide solution (0.02 N)
using a pH-stat. The alkaline solution was consumed
slowly up to 5.2 ml in 2 days. No further consumption
was noticed in the following 7 days. The reaction mix-
ture was acidi®ed with 1N HCl solution and saturated
with sodium chloride, then extracted with ethyl acetate.
The extract was dried over anhydrous MgSO4, ®ltered,
and evaporated under reduced pressure. The residue
was puri®ed by column chromatography, eluting with a
solution of ethyl acetate and n-hexane (20:80) to give
(R)-2-benzyl-4-¯uoro-4-pentenoic acid ([ꢀ]D=+11.2ꢀ
5.15 Methyl 2-benzyl-4-¯uoro-4-pentenoate (3)
To an ice-chilled solution of 19 (250 mg, 1.2 mmole) in
anhydrous methanol (20 ml) was added 1 ml of acetyl
chloride. The resulting mixture was stirred for 12 h at

D. H. Kim et al./Bioorg. Med. Chem. 6 (1998) 239±249
249
(c=1.0, CHCl3)) and (S)-2-benzyl-4-¯uoro-4-pentenoic
acid methyl ester. The spectral data of these compounds
were identical to those of respective racemic com-
pounds, i.e., 19 and 3.
Use of Isolated Enzymes and Whole Cell System, New
York: Academic Press, 1989:26±32.
[3] Kim YM, Kim DH. Bull Korean Chem Soc 1996;17:967±
969.
[4] Dixon M. Biochem J 1953;55:170±171.
[5] Kim DH. Bioorg Med Chem Lett 1993;3:1313±1318.
[6] Fersht A. Enzyme Structure and Mechanism. 2nd ed. New
York: Freeman and Co., 1985:29.
5.18 Enzyme inhibition assay
[7] (a) Segal DM. Biochemistry 1972;11:349±356. (b) Blevins
RA, Tulinsky A. J Biol Chem 1985;260:8865±8872. (c)
Schellenberger V, Braune K, Hof H-J, Jabuke HD. Eur J
Biochem 1991;199:623±636.
[8] (a) Blevins RA, Tulinsky A. J Biol Chem 1985;260:4264±
4275. (b) Steitz TA, Henderson R, Blow DM. J Mol Biol
1969;46:337±348.
[9] (a) Nishio M, Hirota M. Tetrahedron 1989;45:7201±7245.
(b) Nishio M, Umezawa Y, Hirota M, Takeuchi Y. Tet-
rahedron 1995;51:8665±8701. (c) Gung BW, Zhu Z, Fouch
RA. J Am Chem Soc 1995;117:1783±1788.
[10] Stewart JJP. MOPAC 93.00 Manual. Tokyo: Fujitsu Ltd.
[11] Fujimoto T, Yanagihara R, Kobayashi K, Aoyama Y.
Bull Chem Soc Jpn 1995;68:2113±2124.
[12] (a) Shimohigashi Y, Maeda I, Nose T, Ikesue K, Saka-
moto H, Ogawa T, Ide Y, Kawahara M, Ohno M. J Chem
Soc Perkin Trans 1 1996:2479±2485. (b) Sakamoto H,
Shimohigashi Y, Maeda I, Nose T, Nakashima K, Naka-
mura I, Ogawa T, Kawano K, Ohno M. J Mol Recog
1993;6:95±100.
All solutions were prepared using deionized water.
The stock solutions of ꢀ-chymotrypsin and substrate
(sucAAPFpNA) were prepared in Tris bu€er (0.036 M
Tris-0.045 CaCl2, pH 7.8 at 25ꢀC). The stock solution of
inhibitors were prepared in DMSO. The stock solutions
were ®ltered prior to use. The enzyme reaction was
inhibited by adding an aliquot of the ꢀ-chymotrypsin
stock solution to a mixture of inhibitor and substrate in
a 1 ml cuvett at 25ꢀC. Final concentration of ꢀ-chymo-
trypsin was 1.0 ꢂg/ml. Initial rates of the hydrolysis of
the substrate, in the absence and presence of di€erent
concentrations of inhibitors, were determined from the
change in absorbance at 400 nm. The reversible inhibi-
tory constants (Ki) were calculated from plots obtained
by the method of Dixon [4].
[13] (a) Kim DH. In Murakami Y. editor Supramolecular
aspects of enzymes. Comprehensive Supramolecular
Chemistry, vol. 4. Oxford: Pergamon, (b) Cramer RD. J
Am Chem Soc 1977;99:5408±5412. (c) Goto K. Biochem
Biophys Res Commun 1995;206:497±501.
[14] (a) Koshland DE Jr. Proc Natl Acad Sci USA 1958;44:98±
104. (b) Koshland DE Jr. Angew Chem Int Ed Engl
1994;33:2375±2378.
Acknowledgements
We thank the Korea Science and Engineering Founda-
tion for the ®nancial support of this work.
[15] (a) Burgen ASV, Roberts GCK, Feeney J. Nature
1975;253:754±755. (b) Wuthrich K, von Freyberg B,
Weber C, Wider G, Traber R, Widmer H, Braun W. Sci-
ence 1991;254:953±954. (c) Nicklaus MC, Wang S, Dris-
coll JS, Milne GWA. Bioorg Med Chem 1995;3:411±428.
(d) Perona JJ, Tsu CA, Craik CS, Fletterick RJ. Bio-
chemistry 1997;36:5381±5392.
References and Notes
[1] (a) Blow DM. In: Boyer PD editor. The Enzyme, vol. 3.
3rd ed, New York: Academic Press, 1971:185±212. (b)
Blow DM. Acc Chem Res 1976;9:145±152.
[2] Davies HG, Green RH, Kelly DR, Roberts SM. Bio-
transformations in Preparative Organic Chemistry: The
[16] Ager DJ, Allen DR, Schaad DR. Synthesis 1996:1283±1285.
[17] Wissner A, Trova MP. US Patent US5,428,167(1995).