Inhibition of carboxypeptidase A by (S)-2-mercapto-3-phenylpropanoic acid

Article abstract of DOI:10.1039/a705741e

(S)-2-Mercapto-3-phenylpropanoic acid is an effective competitive inhibitor of carboxypeptidase A, comparable in potency to (S)-3-mercapto-2-benzylpropanoic acid.

Full text of DOI:10.1039/a705741e

View Article Online / Journal Homepage / Table of Contents for this issue
Inhibition of carboxypeptidase A by (S)-2-mercapto-3-phenylpropanoic acid
Christopher M. Lanthier, Gregory R. MacKinnon and Gary I. Dmitrienko*
Guelph-Waterloo Centre for Graduate Work in Chemistry, Waterloo Campus, Department of Chemistry, University of Waterloo,
Waterloo, Ontario, Canada N2L 3G1
(S)-2-Mercapto-3-phenylpropanoic acid is an effective
competitive inhibitor of carboxypeptidase A, comparable in
potency to (S)-3-mercapto-2-benzylpropanoic acid.
Me
HS
N
Divalent metal ion-dependent proteolytic enzymes are involved
in a wide variety of physiologically important processes and
play a role in a number of disease states. It is widely recognized
that selective inhibition of such enzymes is a viable strategy for
therapeutic intervention. Notable successes have been reported
in the development of very effective antihypertensive agents
which function through inhibition of zinc-dependent angio-
tensin converting enzyme (ACE).1 Intensive research efforts are
underway aimed at developing related specific inhibitors for
other zinc-dependent proteases such as neutral endopeptidase
24.11 and the matrix metalloproteases.2,3 The ability to carry
out rational design of inhibitors for such enzymes has been built
upon detailed observations made in the development of specific
inhibitors for the better characterized and understood zinc-
dependent proteases such as carboxypeptidase A (CPA) and
thermolysin.4 Thus CPA has served as a valuable model system
for the elucidation of guiding principles for the design of
inhibition strategies for other less well understood but therapeu-
tically important zinc-dependent proteases.
The now classical studies of the inhibition of CPA by
benzylsuccinic acid5 and by mercaptocarboxylic acid deriva-
tives6–8 played an important role in the definition of design
criteria for the extremely successful commercial antihyperten-
sive agents enaloprilat9 and captopril.10 The observation of the
relative potencies of (±)-2-mercapto-3-phenylpropanoic acid 1
(Ki = 1.2 mm)7 and its homologue (S)-2-benzyl-3-mercaptopro-
panoic acid 2 (Ki = 7.8 nm)8 as competitive inhibitors of CPA
has led to the conclusion that the insertion of a methylene group
between the metal binding mercapto group and the a-carbon of
an a-mercapto carboxylic acid derivative results in an optimal
interaction between the active site and the mercapto, carbox-
ylate and hydrophobic side chains of the inhibitor. The fact that
inclusion of a similar methylene spacer between the mercapto
group and the remainder of the structure of the ACE inhibitor
captopril results in very high inhibitory potency (Ki = 1.7 nm)10
reinforced the belief that a spacing of approximately 4.1 Å
between the sulfur atom and the carbonyl group carbon
represents the ideal spatial relationship.
HS
HS
CO2H
S
CO2H
O
CO2H
2
n
1 n = 0
2 n = 1
3
Captopril
Ph
Ph
O
O
H
N
HS
NH2
N
H
CO2H
N
H
Me
O
Me
4
O
5
acid–phenol yielded the free thioacid 1 (89% ee as measured by
1H NMR analysis of the Mosher’s thioester11 form of 1.
Progress curves† for the CPA catalyzed hydrolysis of
hippuryl-l-phenylalanine (HP)12 in the presence of 1 were
found to be biphasic with an initial low rate observed at short
reaction times and a higher reaction rate at longer reaction
times. That this phenomenon was associated with air oxidation
of the thiol to the corresponding disulfide 3 was revealed by the
fact that preincubation of solutions of 1 in air until no thiol could
be detected using DTNB13 assays prior to use in inhibition
studies abolished the biphasic nature of the progress curves. The
slopes of such progress curves were equal to those observed for
the second phase of the biphasic curves observed in experiments
employing solutions with equal concentrations of 1 which had
not been preincubated in air. The inhibition by the disulfide 3,
which was isolated from such solutions and fully characterized,
was found to be of the same order of magnitude as reported
previously for the thiol 1 [Ki = 3.6 ± 0.5 mm (competitive) for
3 vs. reported apparent Ki of 1.2 mm (ref. 7) for racemic 1].
Measurement of the true inhibitory potency of the thiol 1
required experimental conditions which inhibited the air
oxidation process. This was eventually achieved by working
with an argon atmosphere or, much more conveniently, by
adding glutathione as an antioxidant. It was found that
concentrations of glutathione as high as 0.1 mm could be
tolerated without any observable influence on the rate of CPA
In the course of studies of the hydrolysis of thioesters by CPA
we have prepared optically active 2-mercapto-3-phenyl-
propanoic acid 1 and have examined the kinetics of inhibition of
CPA by this thiol. We report herein our observation that 1 is a
much more effective inhibitor of the peptidase activity of CPA
than previously believed, rivaling the potency of 3-mercapto-
2-benzylpropanoic acid 2, and point out the possible sig-
nificance of this observation with respect to the rational design
of specific zinc-binding inhibitors for metalloproteases.
The thiol 1 was synthesized via the route shown in Scheme 1.
The known bromoacid 5, prepared with retention of configura-
tion by diazotization of (R)-phenylalanine in the presence of
bromide, was converted to the benzhydryl ester 6 which was
subjected to an SN2 displacement with 1,1-diphenylmethane-
thiol as the nucleophile. Deprotection using trifluoroacetic
CH2Ph
CO2H
CH2Ph
H
H
i
ii
(R)-Phe
Br
Br
CO2CHPh2
6
7
iii
CH2Ph
iv
1
Ph2CHS
CO2CHPh2
8
Scheme 1 Reagents and conditions: i, NaBr, NaNO2, H2SO4; ii, Ph2CN2,
ether; iii, Ph2CHSH, acetone; iv, CF3CO2H, CH2Cl2, PhOH
Chem. Commun., 1997
2309

View Article Online
Fig. 1 Stereoview of the predicted binding mode of 1 to the active site of CPA
catalyzed hydrolysis of HP. Inhibition of the CPA catalyzed
hydrolysis of HP by 1 in the presence of glutathione was found
to fit a competitive inhibition model as indicated in the
Lineweaver–Burk plots.
were found to be necessary in kinetic investigations with these
compounds. Studies aimed at developing a mechanistic ration-
ale for the susceptibility of 2-mercaptocarboxylates towards air
oxidation are in progress.
Estimation of the potency of inhibition by 1 was complicated
by the fact that the affinity of 1 for CPA was sufficiently high
that inhibitor and enzyme depletion phenomena are significant
and render invalid the normal assumptions associated with
kinetic models of reversible inhibition. As a result, the graphical
method of Dixon14 which avoids such assumptions was
employed to determine Ki for inhibition of CPA catalyzed
hydrolysis of HP to be 12.6 ± 1.0 nm. It thus appears likely that,
in the earlier studies which concluded that the apparent Ki for
inhibition of peptidase activity by (±)-1 was 1.2 mm (ref. 7), the
inhibition observed resulted largely from the presence of the
disulfide 3 rather than the thiol 1.
Footnotes and References
* E-mail: dmitrien@muskie.uwaterloo.ca
† CPA activity assays were conducted in 25 mm tris-HCl, 0.5 m NaCl, pH
7.5 containing 3.5% ethanol (v/v) at 25 °C using hippuryl-l-phenylalanine
as substrate. Enzyme concentrations ranged from 113 to 126 nm, as
determined using absorption at 280 nm (e280 = 6.42 3 104 m21 cm21).7
‡ Molecular modeling studies were conducted using the software package
SYBYL 6.2 (Tripos Associates Inc.) with crystal coordinates for carboxy-
peptidase
A
bound
to
[(2)-2-carboxy-3-phenylpropyl]methyl-
sulfodiimine17 (pdb1cps) obtained from the Brookhaven Protein Data
Bank.
The recognition that 1 (Ki = 12.6 nm) is comparable in
potency to 2 (Ki = 7.8 nm)8 raises interesting questions
concerning the dimensions of the pharmacophore for selective
zinc-binding inhibitors for Zn2+-dependent proteases. It has
been assumed previously that the apparent substantially higher
inhibitory potency of 2 arose from the ability of the benzyl side
chain of 2 to bind in the hydrophobic S1A pocket in the active site
with the thiol sulfur atom interacting favorably with the Zn2+
ion and the carboxylate group interacting with Arg-145.
Docking of a model of 2 (not shown) in the active site of CPA‡
as defined by X-ray crystallographic coordinates suggests that
such a hypothesis is reasonable and consistent with the active
site interactions observed between the related thiol-type
inhibitor 4 and the zinc-dependent endopeptidase thermoly-
sin.15 The apparent lower potency of the thiol inhibitor 1 has
been assumed to result from an inability of the hydrophobic side
chain, the thiol and the carboxylate group to simultaneously
interact with the potential binding sites which are accessible to
the same groups in the homologue 2. A molecular modeling
analysis of 1 docked in the active site of CPA confirms that the
benzyl group in 1 has a diminished interaction with the
hydrophobic pocket in the S1A subsite when the sulfhydryl sulfur
atom is liganded to zinc and also reveals an interesting favorable
interaction between the carboxylate of 1 and Arg-127 which
might offset the loss of binding energy associated with the
diminished hydrophobic interactions (Fig. 1). This model,
although speculative, does suggest strategies for modification
of the structure of 2-mercaptocarboxylic acid inhibitors to
enhance inhibitor potency by increasing the interaction with the
hydrophobic pocket which are now being pursued in this
laboratory.
1 L. H. Opie, Angiotensin Converting Enzyme Inhibitors, Wiley-Liss,
New York, 1992.
2 R. Bohacek, S. De Lombaert, C. McMartin, J. Priestle and M. Gru¨tter,
J. Am. Chem. Soc., 1996, 118, 8231; B. P. Roques, F. Noble, V. Dauge´,
M. C. Fournie´-Zaluski and A. Beaumont, Pharmacol. Rev., 1993, 45,
87; E. Fillion and D. Gravel, Bioorg. Med. Chem. Lett., 1996, 6,
2097.
3 B. Lovejoy, A. Cleasby, A. M. Hassell, K. Longley, M. A. Luther, D.
Weigl, G. McGeehan, A. B. McElroy, D. Drewry, M. H. Lambert and S.
R. Jordan, Science, 1994, 263, 375; M. F. Browner, W. W. Smith and A.
L. Castelhano, Biochemistry, 1995, 34, 6602; R. P. Beckett, A. H.
Davidson, A. H. Drummond, P. Huxley and M. Whittaker, Drug
Discovery Today, 1996, 1, 16.
4 W. N. Lipscomb and N. Stra¨ter, Chem. Rev., 1996, 96, 2375; B. W.
Matthews, Acc. Chem. Res., 1988, 21, 333; H. Kim and W. N.
Lipscomb, Biochemistry, 1991, 30, 8171; D. W. Christianson and W. N.
Lipscomb, Acc. Chem. Res., 1989, 22, 62–69; P. A. Bartlett and C. K.
Marlowe, Biochemistry, 1983, 22, 4618.
5 L. D. Byers and R. Wolfenden, Biochemistry, 1973, 12, 2070;
L. D. Byers and R. Wolfenden, J. Biol. Chem., 1972, 247, 606.
6 J. Suh, S. H. Lee and J. Y. Uh, Bioorg. Med. Chem. Lett., 1995, 5,
585.
7 B. Holmquist and B. L. Vallee, Proc. Natl. Acad. Sci. USA, 1979, 76,
6216.
8 M. A. Ondetti, M. E. Condon, J. Reid, E. F. Sabo and H. S. Cheung,
Biochemistry, 1979, 18, 1427; D. H. Kim and Y. J. Kim, Bioorg. Med.
Chem. Lett., 1993, 3, 2681.
9 A. A. Patchett, Nature, 1980, 288, 280; R. Shiparo and J. F. Riordan,
Biochemistry, 1984, 23, 5225.
10 D. W. Cushman, H. S. Cheung, E. F. Sabo and M. A. Ondetti,
Biochemistry, 1977, 16, 5484.
11 J. A. Dale, D. L. Dull and H. S. Mosher, J. Org. Chem., 1969, 34,
2543.
Although our observations clearly indicate the need for
consideration of the stability of thiol type inhibitors in
evaluation of inhibitory potency against zinc dependent pro-
teases, it should be emphasized that some thiols are much better
behaved than 1 and do not require the use of anaerobic
conditions or radical scavengers in inhibition studies. For
example the rates of air oxidation of the 3-mercaptocarboxylic
acid 2 or the 2-mercaptocarboxamide 5, which is a potent
inhibitor of angiotensin converting enzyme,16 are substantially
lower than that observed for 1 such that no special precautions
12 J. E. Folk and E. W. Schirmer, J. Biol. Chem., 1963, 238, 3884.
13 P. W. Riddles, R. L. Blakely and B. Zerner, Anal. Biochem., 1979, 94,
75.
14 M. Dixon, Biochem. J., 1972, 129, 197.
15 A. F. Monzino and B. W. Matthews, Biochemistry, 1982, 21, 3390.
16 C. M. Lanthier, PhD Thesis, University of Waterloo, 1996.
17 A. M. Cappalonga, R. S. Alexander and D. W. Christianson, J. Biol.
Chem., 1992, 267, 19 192.
Received in Corvallis, OR, USA, 4th August 1997; 7/05741E
2310
Chem. Commun., 1997