Looking glass mechanism-based inhibition of peptidylglycine α-amidating monooxygenase

Article abstract of DOI:10.1016/j.tetasy.2011.01.006

Carboxyl-terminal amidation, a required post-translational modification for the bioactivation of many peptide hormones, entails sequential enzymatic action by peptidylglycine α-monooxygenase (PAM, EC 1.14.17.3) and peptidylamidoglycolate lyase (PGL, EC 4.3.2.5). We have previously demonstrated that PAM and PGL exhibit strict tandem reaction stereospecificities, with PAM producing exclusively α-hydroxyglycine moieties of absolute configuration (S), and PGL being reactive only toward (S)-α-hydroxyglycines, and we have also shown that PAM exhibits strict P₂-subsite stereospecificity toward both peptide substrates and peptidyl competitive inhibitors. Herein, it is reported that the inhibitory stereochemistry of olefinic mechanism-based amidation inhibitors differs from the strict subsite stereospecificity exhibited by PAM toward substrates and reversible competitive inhibitors. Kinetic analyses of mechanism-based irreversible inhibition of PAM by the (S)- and (R)-enantiomers of 5-acetamido-4-oxo-6-phenyl-2-hexenoic acid were carried out using the rigorous progress curve method. The two enantiomers were found to exhibit very similar values of K_I and k_inact and in both cases kinetic analysis confirmed that irreversible inhibition occurs strictly at the substrate binding site with no ESI complex being formed during the catalytic processing of these irreversible inhibitors. Molecular docking studies were carried out to help rationalize the sharp contrast in the stereospecificity of PAM toward irreversible inhibitors versus substrates and competitive inhibitors. The results revealed that, in contrast to substrates, both docked enantiomers of the olefinic irreversible inhibitors are well-positioned to undergo catalytic processing at the Cu center that gives rise to irreversible inhibition. Taken together, these results provide one of the first clear examples where the stereospecificity of a particular enzyme toward mechanism-based irreversible inhibitors differs from that for substrates and competitive inhibitors.

Full text of DOI:10.1016/j.tetasy.2011.01.006

Tetrahedron: Asymmetry 22 (2011) 283–293
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Looking glass mechanism-based inhibition of peptidylglycine
a-amidating monooxygenase
⇑
Michael S. Foster, Charlie D. Oldham, Sheldon W. May
School of Chemistry and Biochemistry and the Petit Institute for Bioengineering and Biosciences, Georgia Institute of Technology, Atlanta, GA 30332, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Carboxyl-terminal amidation, a required post-translational modification for the bioactivation of many
Received 14 December 2010
Accepted 12 January 2011
Available online 4 March 2011
peptide hormones, entails sequential enzymatic action by peptidylglycine
a-monooxygenase (PAM, EC
1.14.17.3) and peptidylamidoglycolate lyase (PGL, EC 4.3.2.5). We have previously demonstrated that
PAM and PGL exhibit strict tandem reaction stereospecificities, with PAM producing exclusively
a-hydroxyglycine moieties of absolute configuration (S), and PGL being reactive only toward (S)-a-
hydroxyglycines, and we have also shown that PAM exhibits strict P2-subsite stereospecificity toward
both peptide substrates and peptidyl competitive inhibitors. Herein, it is reported that the inhibitory
stereochemistry of olefinic mechanism-based amidation inhibitors differs from the strict subsite stereo-
specificity exhibited by PAM toward substrates and reversible competitive inhibitors. Kinetic analyses of
mechanism-based irreversible inhibition of PAM by the (S)- and (R)-enantiomers of 5-acetamido-
4-oxo-6-phenyl-2-hexenoic acid were carried out using the rigorous progress curve method. The two
enantiomers were found to exhibit very similar values of KI and kinact and in both cases kinetic analysis
confirmed that irreversible inhibition occurs strictly at the substrate binding site with no ESI complex
being formed during the catalytic processing of these irreversible inhibitors. Molecular docking studies
were carried out to help rationalize the sharp contrast in the stereospecificity of PAM toward irreversible
inhibitors versus substrates and competitive inhibitors. The results revealed that, in contrast to sub-
strates, both docked enantiomers of the olefinic irreversible inhibitors are well-positioned to undergo
catalytic processing at the Cu center that gives rise to irreversible inhibition. Taken together, these results
provide one of the first clear examples where the stereospecificity of a particular enzyme toward mech-
anism-based irreversible inhibitors differs from that for substrates and competitive inhibitors.
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
we have shown that PGL, which converts the intermediate to the
des-glycyl amide plus glyoxylate, is reactive only toward
Carboxyl-terminal amidation is a common post-translational
event responsible for the bioactivation of approximately half of
all peptide hormones, including the potent pro-inflammatory
a
-hydroxyglycines with an (S)-absolute configuration.9,16 We have
also shown that the glycolate ester analogs of glycine-extended
peptides are potent competitive inhibitors of PAM, but only when
mediators Substance
P
and Calcitonin Gene-Related Peptide
an L
-amino acid residue is present at the P2-subsite position.16
(CGRP)1–3 often increasing the affinity for their respective recep-
tors by as much as 1000-fold compared to their glycine-extended
precursors.4 The amidation reaction is catalyzed by the sequential
Thus, PAM exhibits subsite stereospecificity toward both peptide
substrates and peptidyl competitive inhibitors.
Recognizing the potential pharmacological benefit of inhibiting
the synthesis of pro-inflammatory peptides, we have developed
new classes of mechanism-based irreversible inhibitors and transi-
tion-state analogs targeted at the post-translational amidation
process. Among these, compounds 5-acetamido-4-oxo-6-phenyl-
2-hexenoic acid (AOPHA) and 5-acetamido-4-oxo-6-thienyl-2-
hexenoic acid, which possess a C-terminal acrylate functionality
linked to a PAM-binding peptide moiety, are the most potent irre-
versible amidation inhibitors known to date.17,6,18,19 Indeed, we
have recently shown that irreversible PAM inhibitors exhibit po-
tent anti-inflammatory activity against both acute (carrageenan)
and chronic (adjuvant-induced polyarthritis) inflammation in
activities of peptidylglycine a-monooxygenase [PAM; EC 1.14.17.3]
and peptidoamidoglycolate lyase [EC 4.3.2.5] upon precursor C-
terminal glycine-extended peptide substrates.5–16 In a process that
is dependent on ascorbate and molecular oxygen, PAM stereospe-
cifically catalyzes the formation of an (S)-
intermediate from glycine-extended peptides with amino acid
a-hydroxyglycine
residues of L-configuration at the P2-position. Correspondingly,
⇑
Corresponding author. Tel.: +1 404 894 4052; fax: +1 404 894 2295.
E-mail address: sheldon.may@chemistry.gatech.edu (S.W. May).
0957-4166/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2011.01.006

284
M. S. Foster et al. / Tetrahedron: Asymmetry 22 (2011) 283–293
rats.20,21 We have also shown that the methyl ester of 4-
phenyl-3-butenoic acid (PBA-OMe), an irreversible PAM inhibitor,
is able to restore gap-junctional communication in WB-Ras-trans-
formed rat liver epithelial cells, and is selectively cytotoxic toward
transformed cells versus untransformed WB-Neo cells.22
together as a mixture. The enantiomeric purity was further inves-
tigated using polarimetry and circular dichroism spectroscopy. The
Cotton effect was clearly evident at 205 nm for the enantiomeric
pair. Polarimetry indicated that the (R)- and (S)-enantiomers rotate
polarized light by the same magnitude but in opposite directions,
Stereospecificity is obviously a very important consideration in
the design of enzyme-targeted pseudosubstrates, reversible inhib-
itors, and irreversible inhibitors. As pointed out by Kim, while
binding stereospecificities for competitive inhibitors and transi-
tion-state analogs would generally be expected to correspond to
those for substrates, evidence has begun to emerge that the inhib-
itory stereochemistry for mechanism-based irreversible inhibitors
may be quite different from that predicted on the basis of substrate
reactivity.23–25 Herein, we report that the inhibitory stereochemis-
try of olefinic mechanism-based irreversible amidation inhibitors
indeed differs from the strict subsite stereospecificity exhibited
by PAM toward both substrates and competitive inhibitors.
with specific [
a
]
values of +12 and ꢀ12 for (R)-AOPHA and (S)-
D
AOPHA, respectively. ESI mass spectrometry and NMR again
showed that both compounds had identical profiles. These results
clearly indicate that enantiomeric purity was maintained through-
out these syntheses.
Kinetic analyses of the irreversible inhibition of PAM by the (S)-
and (R)-enantiomers of AOPHA were carried out using both the
conventional ‘dilution assay’ method and the much more rigorous
‘progress curve’ method.17,26,27 Whereas the dilution assay is quite
straightforward, it reliably provides only the value of kinact/KI and
not the individual constants. On the other hand, the progress curve
method, owing to the simultaneous incubation of both substrate
and inhibitor, allows for the elucidation of all possible kinetic con-
stants and is therefore able to clearly indicate the type of inhibition
which occurs. As detailed by Tsou,28 progress curves of inhibition
are obtained at a series of substrate and inhibitor concentrations;
then, through a series of replots, numerical values are obtained
for each of the kinetic parameters shown in the scheme in Table 1.
This scheme illustrates the nature of all possible pathways when
the enzyme is incubated simultaneously in the presence of both
substrate and inhibitor.
2. Results
2.1. Kinetic analyses of PAM inhibition by (S)- and (R)-5-
acetamido-4-oxo-6-phenyl-2-hexenoic acid
In order to ensure the enantiomeric purity of the tested com-
pounds, each synthesis began with the appropriate enantiomeri-
cally pure amino acid methyl ester. Prior to the final hydrolysis
step, each enantiomer of AOPHA-Me was eluted from a chiral
Representative progress curves and double reciprocal plots for
both enantiomers of AOPHA are shown in Figure 1, and the kinetic
parameters obtained from these experiments are listed in Table 1.
It is evident that the KI values for each enantiomer are remarkably
D-penicillamine column as a single peak (21 and 28 min for the D
and L-enantiomers, respectively) using the procedure outlined in
Section 4. Both enantiomers had identical profiles by ESI mass
spectrometry and NMR. Each enantiomeric ester was then enzy-
matically hydrolyzed to the free acid using porcine liver esterase
(E.C. 3.1.1.1). This enzyme was employed to allow for more mild
alkaline conditions than would be utilized in a base-catalyzed
hydrolysis. Each enantiomer of AOPHA showed only one peak upon
similar, with values of 53.7 and 60.0 lM obtained from the pro-
gress curve assay for the (S)- and (R)-enantiomers, respectively.
The rate constants for irreversible inhibition are also very similar:
0.38 and 0.29 minꢀ1 for the (S)- and (R)-enantiomers, respectively.
Double-reciprocal plots for both enantiomers (Fig. 1c and d) show
a convergence of all lines (each representative of a different
eluting from the D-penicillamine column, and two peaks when run
Table 1
Inhibition scheme and kinetic parameters for PAM inhibition by olefins
(S)-5-Acetamido-4-oxo-6-phenyl-2-
(R)-5-Acetamido-4-oxo-6-phenyl-2-
(S)-5-Acetamido-7-methylthio-4-oxo-2-
hexenoic acid
hexenoic acid
heptenoic acid
KI (
l
M)
54
1
1
60
1
2
57
1
1
K0I (MM)
kinact (minꢀ1
k0inact (minꢀ1
k02 (minꢀ1
)
)
0.38 0.04
0
0.29 0.02
0
0.14 0.01
0
0
0
0
)
kinact/KI (Mꢀ1 minꢀ1 ꢁ 10ꢀ3
kinact/KI (Mꢀ1 minꢀ1 ꢁ 10ꢀ3
(dilution assay)
)
)
7.0 0.2
7.3 0.1
4.8 0.1
4.1 0.1
2.4 0.1
ND
KM (methyl ester, PLE)
Vmax (methyl ester, PLE)
8.6 mM
6.1 mM
ND
ND
0.058 mM sꢀ1
0.048 mM sꢀ1

M. S. Foster et al. / Tetrahedron: Asymmetry 22 (2011) 283–293
285
Figure 1. Progress curves and double-reciprocal plots for PAM inhibition by AOPHA enantiomers. Product formation in the presence of various concentrations of (S)-5-
acetamido-4-oxo-6-phenyl-2-hexenoic acid and (R)-5-acetamido-4-oxo-6-phenyl-2-hexenoic acid. (A) Inhibition by (S)-5-acetamido-4-oxo-6-phenyl-2-hexenoic acid
between 19.2 and 64.1
lM at a substrate (TNP-YVG) concentration of 5.86
lM. (B) Inhibition by (R)-5-acetamido-4-oxo-6-phenyl-2-hexenoic acid between 46.7 and 104 lM
at a substrate concentration of 18.4
lM. Asymptotes represent the product concentration at time = infinity. Subsequent replots yield the kinetic parameters listed in Table 1.
(C) Double-reciprocal plot of 1/kobs versus 1/[I] for (S)-5-acetamido-4-oxo-6-phenyl-2-hexenoic acid. (D) Double-reciprocal plot of 1/kobs versus 1/[I] for (R)-5-acetamido-4-
oxo-6-phenyl-2-hexenoic acid.
substrate concentration) at the same point on the y-axis, as ex-
pected for inhibition that occurs only at the substrate binding
site.28,17 In addition, ki0nact was found to be zero for both enantio-
mers, indicating that no inactivated ESI complex is formed from
ESI; the fact that k02 is zero for each compound also shows that
no product is formed and released from the ESI complex. Moreover,
the inhibition constant K0I was found to approach infinity for both
enantiomers, indicating that a ternary complex never forms in any
case, and is consistent with the zero values for k02 and ki0nact. Taken
together, these results clearly demonstrate that inhibition of PAM
by both (R)- and (S)-AOPHA occurs strictly at the PAM substrate
binding site, with no ‘ESI’ complex being formed during catalytic
processing of these irreversible inhibitors.
In view of this near equipotency of the two AOPHA enantiomers
as irreversible inhibitors of PAM, we also investigated the relative
reactivities of the two enantiomers of AOPHA-Me as substrates of
Porcine Liver Esterase (PLE). Michaelis–Menten plots and Hanes–
Woolf replots for (S)-AOPHA-Me and (R)-AOPHA-Me are shown
in Figure 2. The KM values for the (S)- and (R)-enantiomers are
8.6 and 6.1 mM, respectively, and the respective Vmax values at
identical concentrations of enzyme are 0.06 and 0.05 mM sꢀ1. PLE
has been shown, in many different cases, to be stereoselective,
but generally research has been limited to molecules which
possess stereocenters either immediately adjacent to, or within
two atoms of, the oxygen atom of the ester moiety.29
2.2. Molecular docking of substrate enantiomers
We carried out a series of protein–ligand molecular docking
studies in an attempt to rationalize the sharp contrast in the ste-
reospecificity of PAM toward the mechanism-based irreversible
AOPHA inhibitors on the one hand, versus peptide substrates and
glycolate-ester competitive inhibitors on the other. Obviously, a
key structural difference between these three classes of com-
pounds is the nature of the atom present at the position analogous
to that of the amide nitrogen (hydrogen bond donor) of the C-
terminal-Gly in PAM substrates, such as N-Ac-Phe-Gly; the com-
petitive inhibitors possess an ester oxygen atom in this position
(hydrogen bond acceptor), and the olefinic irreversible inhibitors
possess an sp2-hybridized carbon atom (neither a hydrogen bond
donor nor acceptor).
The structures of AOPHA and many of our competitive inhibi-
tors were designed on the basis of the active PAM substrate,
N-Ac-
structure determined by Amzel et al. was for the PAM catalytic
core complexed with the ligand N-Ac-3,5-diiodo-
-Tyr-Gly.30 In
L-Phe-Gly (KM = 7.9 l
M).16 On the other hand, the crystal
L

286
M. S. Foster et al. / Tetrahedron: Asymmetry 22 (2011) 283–293
Figure 2. (A) Plot of rate of hydrolysis of methyl-(S)-5-acetamido-4-oxo-6-phenyl-2-hexenoic acid versus concentration, with Hanes–Woolf plot inset. KM and Vmax are
8.6 mM and 0.058 mM/min, respectively. (B) Plot of rate of hydrolysis of methyl-(R)-5-acetamido-4-oxo-6-phenyl-2-hexenoic acid versus concentration, with Hanes–Woolf
plot inset. KM and Vmax are 6.1 mM and 0.048 mM/min, respectively.
the crystal structure, Amzel et al. use the notation CuM to refer to
the copper ion at the substrate binding site which is ligated by
one methionine residue and two histidines. We therefore first pro-
these interactions are maintained in the case of N-Ac-
and we calculate an rms deviation of 1.22 between our docked
N-Ac- -Phe-Gly and the crystal structure ligand. Moreover, the
glycyl methylene of N-Ac- -Phe-Gly is oriented such that the
L-Phe-Gly,
L
ceeded to demonstrate that docked N-Ac-
L
-Phe-Gly recapitulates
L
the major interactions between substrate and enzyme that were
observed in the crystal structure. As shown in Figure 3A, which
pro-(S) hydrogen is directed toward CuM, whereas the pro-(R)
hydrogen is oriented away from CuM toward the interior of the
hydrophobic pocket of the enzyme. This is consistent with the
reaction stereospecificity of PAM, which abstracts only the pro-
(S) hydrogen of peptide substrates and forms (S)-hydroxyglycines
exclusively.16
illustrates an overlay of docked N-Ac-
coordinates of the complex with N-Ac-3,5-diiodo-
L
-Phe-Gly onto the X-ray
-Tyr-Gly, this
L
is indeed the case. As pointed out by Amzel et al., four major
interactions are evident: (1) a salt bridge between the glycyl car-
boxylate of the ligand and the positively-charged guanidinium
functionality of Arg240; (2) a variety of van der Waals’ contacts,
both between the main peptide chain of the substrate and the
CuM ligands (Met314 and His242), and between the P2 benzyl side
chain and various residues of the large hydrophobic pocket of the
active site (especially Phe112); (3) a hydrogen bond between the
amide nitrogen of the substrate and the side-chain carbonyl group
of Asn316; (4) a hydrogen bond between the substrate carboxylate
and the phenol of Tyr318. It is evident from Figure 3A that all of
In sharp contrast, as illustrated in Figure 3B, the substrate enan-
tiomer, N-Ac-D-Phe-Gly does not closely associate with any PAM
active site residues. Instead, as the simulation necessarily confines
the ligand within the space outlined by the parameters of the dock-
ing box, this ‘ligand’ occupies the extensive available free space be-
tween Asn316, Lys134, and Leu206. This is consistent with our
previous report16 that N-Ac-
D-Phe-Gly, is not a PAM substrate
and exhibits only extremely weak binding (KI = 1.3 mM) when
evaluated as a potential inhibitor for the enzyme.
Figure 3. (A) N-Ac-
Ac- -Phe-Gly forms a bidentate salt bridge with Arg240, hydrogen bonds with Asn 316 and Tyr318, and the phenyl ring associates closely with Phe112. The pro-(S) hydrogen
would be directed toward CuM. (B) N-Ac- -Phe-Gly docked to PAM active site fails to bind or associate with Arg240, Phe112, or Tyr318. The carboxylate of the ligand is in
proximity with the positively-charged amino group of Lys134.
L-Phe-Gly docked to PAM active site and overlaid with the crystal structure ligand N-Ac-L
-20,50-diiodo-Tyr-Gly24 to illustrate similarity in binding mode. N-
L
D

M. S. Foster et al. / Tetrahedron: Asymmetry 22 (2011) 283–293
287
2.3. Mechanism-based inhibitor enantiomers: (S)-AOPHA versus
(R)-AOPHA
inhibitors of PAM (KI = 59.8 and 45.2
N-Ac-
l
M, respectively).16 Docked
L
-Leu-OCH2COOH (Fig. 6A) displays the aforementioned
Arg240, Tyr318 and Phe112 interactions. The corresponding D-
enantiomers of these esters, on the other hand, are very poor com-
petitive inhibitors of PAM, with KI values in excess of 2 mM. As
As shown in Figure 4A, the orientation of docked (S)-AOPHA is
very similar overall to that of N-Ac- -Phe-Gly. Again, the ligand car-
L
boxylate forms both a salt bridge with the guanidinium of Arg240
and a hydrogen bond with Tyr318, the phenyl ring of the ligand
associates closely with Phe112, and the backbone is in VDW
contact with Met314/His242. As shown in Figure 4B, these three
binding interactions are still present for the olefinic enantiomer,
(R)-AOPHA. The salt bridge to Arg240 and the hydrogen bond inter-
action with Tyr318 are both quite apparent. While the location of
the phenyl ring differs from that in Figure 4A, a compensating
hydrophobic interaction with Met208 is evident; the similar KI
values for the two enantiomers suggest that this effectively offsets
the binding energy of the lost Phe112 association. Obviously, since
the olefinic irreversible inhibitors possess an sp2 carbon at the
position of the amide nitrogen of PAM substrates, a substrate-type
hydrogen bond interaction with Asn316 is not possible.
illustrated in Figure 6B, N-Ac-D-Leu-OCH2COOH, like N-Ac-D-Phe-
Gly, does not dock productively within the confines of the PAM ac-
tive site, instead occupying the largely empty space between
Met208, Leu206, and Lys134. Thus, these docking results are fully
consistent with the kinetic data for our glycolate-ester competitive
inhibitors, which exhibit the same subsite stereospecificity as the
substrates.
2.5. ‘Un-natural’ substrates: O-Ac-(S)-Mandelyl-Gly and O-Ac-
(R)-Mandelyl-Gly
These two enantiomers are O-acetylated phenylglycine analogs
of N-Ac-Phe-Gly. We have previously shown that O-Ac-(S)-Mand-
elyl-Gly is a PAM substrate, whereas its (R)-enantiomer is not.16
Docked O-Ac-(S)-Mandelyl-Gly (Fig. 7A) displays the aforemen-
tioned Arg240, Tyr318 and Phe112 interactions. Due to the lack
of the benzylic methylene relative to N-Ac-Phe-Gly, the O-Ac-(S)-
Mandelyl-Gly is drawn upwards into the active site resulting in
the formation of a hydrogen bond between the ligand’s amide
hydrogen and Tyr318, rather than Asn316. As a result, the ligand’s
main chain is no longer in van der Waals’ contact with Met314 and
His242; the loss of these interactions likely contributes to the
greatly increased KM value of O-Ac-(S)-Mandelyl-Gly (more than
It is important to note that the olefinic moieties of both (S)- and
(R)-AOPHA are well-positioned to undergo catalytic processing at
the CuM center that gives rise to irreversible inhibition, as these
atoms are nearly superimposable upon the homologous atoms of
both our docked N-Ac-L-Phe-Gly and Amzel’s crystal-structure
ligand. Indeed, this is also the case for docked (S)-5-acetamido-7-
methylthio-4-oxo-2-heptenoic acid (the methionine analog of AO-
PHA) and its (R)-enantiomer, shown in Figure 5A and B. Here again,
the docked structures recapitulate the salt bridge with Arg240, the
hydrogen bond with Tyr318, the protrusion of the P2 side chain
into the hydrophobic pocket, and the positioning of the olefinic
functionality into close proximity with the CuM reaction center.
an order of magnitude greater than that of N-Ac-L-Phe-Gly; see
Scheme 1). As with N-Ac- -Phe-Gly, the stereotopically equivalent
D
(R)-O-Ac-Mandelyl-Gly does not bind productively to the active
site of the enzyme (Fig. 7B), with no hydrogen bonds forming be-
tween ligand and receptor, and no salt bridge formation between
the substrate carboxylate and Arg240. Thus it is evident from both
the kinetic data and the molecular docking studies that the change
from an N-acetyl group to an O-acetyl group does not alter the
stereoselectivity of PAM toward substrate enantiomers.
2.4. Reversible competitive inhibitors: N-Ac-L-Leu-OCH2COOH
versus N-Ac- -Leu-OCH2COOH
D
We have previously shown that the glycolate esters, N-Ac-
L-
Leu-OCH2COOH and N-Ac- -Phe-OCH2COOH, are competitive
L
Figure 4. (A) (S)-5-Acetamido-4-oxo-6-phenyl-2-hexenoic acid docked to PAM active site adopts a conformation very similar to that of N-Ac-L-Phe-Gly, forming a salt bridge
with Arg240, a hydrogen bond with Tyr318, and the phenyl ring of ligand in close proximity to Phe112. (B) (R)-5-Acetamido-4-oxo-6-phenyl-2-hexenoic acid docked to PAM
active site has a salt-bridge formed with the guanidinium moiety of Arg240 and a hydrogen bond formed with Tyr318. The phenyl ring, unlike the L-enantiomer, although still
directed into the hydrophobic pocket of the enzyme, associates closely with Met208 rather than Phe112.

288
M. S. Foster et al. / Tetrahedron: Asymmetry 22 (2011) 283–293
Figure 5. (A) (S)-5-Acetamido-7-methylthio-4-oxo-2-heptenoic acid docked to PAM active site, adopts a conformation similar to that of (S)-5-acetamido-4-oxo-6-phenyl-2-
hexenoic acid. (B) (R)-5-Acetamido-7-methylthio-4-oxo-2-heptenoic acid docked to PAM active site, adopts a slightly different conformation in the methionyl side chain than
that seen for (R)-5-acetamido-4-oxo-6-phenyl-2-hexenoic acid.
Figure 6. (A) N-Ac-
hydrophobic pocket corresponding to that of the aromatic ring of N-Ac-
productively to any active site residues, and fails to adopt an orientation characteristic of a substrate or competitive inhibitor.
L
-Leu-OCH2COOH docked to PAM active site. The carboxylate forms a salt bridge with Arg240 and the sec-butyl side chain occupies space in the
L
-Phe-Gly. (B) N-Ac- -Leu-OCH2COOH docked to the PAM active site. D-enantiomer is unable to bind
D
of carboxypeptidase A (CPA) a zinc-containing metalloprotease
which hydrolyzes C-terminal hydrophobic residues of -configura-
tion adjacent to P2 residues also with an -configuration (i.e., only
one of four possible diastereomers). The molecule 2-benzyl-3,4-
epoxybutanoic acid was designed as a mechanism-based inhibitor
that alkylates the catalytic residue Glu270; unexpectedly, the
(2S,3R)-isomer, with a reverse stereochemistry at the 2-position
to that of the substrates, was found to be twice as potent as its
(2R,3S)-enantiomer. Similarly, it was found that all four diastereo-
3. Discussion
L
L
The results reported herein demonstrate that the inhibitory ste-
reochemistry of olefinic mechanism-based irreversible amidation
inhibitors differs from the strict subsite stereospecificity exhibited
by PAM toward both substrates and reversible competitive inhibi-
tors. To the best of our knowledge, there is only one other clear
example in the literature where the stereospecificity of a particular
enzyme toward mechanism-based irreversible inhibitors differs
from that for substrates and competitive inhibitors. This example
is the work of Kim et al.23,24,31,25 on mechanism-based inhibitors
mers of the related molecule
a-benzyl-2-oxo-1,3-oxazolidine-4-
acetic acid irreversibly inhibit CPA with comparable potency, in

M. S. Foster et al. / Tetrahedron: Asymmetry 22 (2011) 283–293
289
Figure 7. (A) O-Ac-(S)-Mandelyl-Gly docked to the PAM active site. Binds in a similar manner to PAM substrate N-Ac-
L-Phe-Gly. The shorter P2 side chain results in the
molecule being pulled upwards toward Phe112, and two hydrogen bonds are formed with Tyr318. (B) O-Ac-(R)-Mandelyl-Gly docked to PAM. The molecule was unsuccessful
in binding in the manner of substrate, competitive inhibitor or mechanism-based inhibitor.
O
O
O
O
O
O
H
N
H
N
O
N
H
OH
O
OH
N
H
OH
O
O
O
N-Ac-Phe-Gly
N-Ac-Leu-Glycolate
O-Ac-Mandelyl-Gly
KM = 7.9 µM (L)
KI = 2.0 µM (L)
KI = 1300 µM (D)
K
K
I = 60 µM (L)
I = 2100 µM (D)
KM= 170 µM (L)
KI = 8100 µM (D)
S
O
O
O
O
N
H
OH
N
H
OH
O
O
5-acetamido-4-
oxo-6-phenyl-2-
hexenoic acid
5-acetamido-7-
methylthio-4-oxo
2-heptenoic acid
KI = 54 µM (L)
KI = 57 µM (L)
K
I = 60 µM (D)
Scheme 1. Kinetic parameters of compounds used in molecular modeling studies.
contrast to the strict stereospecificity of CPA catalysis. Mechanisti-
cally, Kim et al. propose that for the oxazolidinone inhibitors, the
Glu270 carboxylate attacks at the carbamate carbonyl rather than
at the 5-position of the oxazolidinone ring. It should be noted that
Moreira et al.32 have reported an apparent lack of stereospecificity
in the mechanism-based inhibition of human leukocyte elastase by
oxazolidin-2,4-dione derivatives. However, since only the diaste-
reomeric pairs were purified, it is unclear whether or not all four
stereoisomers are active, although the authors’ docking results
suggested that only the (4R,50S)- and (4S,50S)-diastereomers, with
chirality corresponding to that at the P1-position of normal sub-
strates, are active.
Turning to reversible inhibitors, Fleet et al. have shown that the
‘un-natural’ enantiomers of a number of iminosugar glycomimetics
are potent glucosidase inhibitors,33–35 and these investigators have
termed them ‘looking-glass inhibitors’. For example,
(2,5-dideoxy-2,5-imino- -mannitol) was found to be 2–4 orders
of magnitude more potent in inhibiting plant and mammalian
-glucosidases than its D-enantiomeric natural product. Similar
L-DMDP
L
a
results have been noted for a variety of natural product sugar
analogs and their enantiomeric counterparts, such as L-DIM
(1,4-dideoxy-1,4-imino-
imino- -arabinitol). However, kinetic analyses of the
ars proved that the unnatural ‘looking-glass inhibitors’ of DMPD
L
-mannitol) and L-AB1 (1,4-dideoxy-1,4-
L
L
- and -sug-
D

290
M. S. Foster et al. / Tetrahedron: Asymmetry 22 (2011) 283–293
and AB1 actually exhibit non-competitive inhibition, as opposed to
the competitive inhibition exhibited by the natural -sugar ana-
4. Experimental
D
logs. The authors interpreted this as evidence that the ‘unnatural’
analogs actually bind to a site on the enzyme other than the active
site.36 More recently, these investigators have found that both
enantiomers of N-benzyl-1,4-dideoxy-1,4-imino-lyxitol are mod-
4.1. Synthesis of (R)-5-acetamido-4-oxo-6-phenyl-2-hexenoic
acid
4.1.1. N-Ac-D-Phe-OMe
In a 100 mL round-bottomed flask 9.4 mL (120 mmol) pyridine
and 11.0 mL (120 mmol) acetic anhydride were combined at 0 °C.
erate competitive inhibitors of
Schramm et al. have found that for
tion-state analog inhibitors of purine nucleoside phosphorylase,
both
-50-deaza-l’-aza-20-deoxy-l’-(9-methylene)-immucillin-H
a-
D
-galactosidase.37 Similarly,
- and -immucillin, transi-
D
L
Next, 5.0 g (23 mmol) D-Phe-OMe HCl was dissolved in the above
L
solution and the ice bath was removed. The reaction was allowed
to proceed overnight at room temperature, turning a rich purple
color. The reaction was quenched by the addition of 100 mL
water/ice slurry. The resulting solution was again allowed to return
to room temperature and then extracted four times with CH2Cl2.
The extracts were pooled and then rinsed three times each with
saturated NaHCO3 solution, 0.1 M HCl solution, and water. The
methylene chloride solution was dried over MgSO4 and evaporated
to dryness under reduced pressure, yielding 4.5 g (20 mmol, 87%)
and its ‘natural’ enantiomer are competitive inhibitors with KI
values differing by a factor of three.38
Several cellular kinases that are involved in the bioactivation of
nucleoside pro-drugs, such as 20,30-dideoxy-30-thiacytidine (lami-
vudine/3TC) and L-thymidine, into their active triphosphate forms
have been shown to exhibit ‘relaxed enantioselectivity’. In many of
these cases, although the relevant substrates are chiral, stereose-
lectivity is very low and this feature also extends to very low
enantioselectivity in the binding efficacy of competitive inhibitor
nucleoside analogs. Thus, for example, cytosolic deoxycytidine ki-
nase (dCK) and mitochondrial thymidine kinase (TK2) catalyze
of N-Ac-D
-Phe-OMe as a white solid. 1H NMR ([2H]chloroform, tet-
ramethylsilane (TMS) = 0.0 ppm): d 1.98 (s, 3H), d 3.12 (m, 2H), d
3.73 (s, 3H), d 4.89 (m, 1H), d 5.90 (broad, 1H), d 7.09 (m, 2H), d
7.27 (m, 3H).
the phosphorylation of either ‘un-natural’
L-thymidine or
D-thymi-
dine to their respective monophosphates (
L
-TMP/ -TMP) with
D
nearly equal facility,39,40 and UMP/CMP kinase and TMP kinase
generate diphosphate nucleotides with a similar lack of enantiose-
lectivity.41–43 Similarly, human 3-phosphoglycerate kinase (hPGK),
perhaps the most well-studied of these kinases, is also non-specific
toward its nucleotide substrates.44,45 Finally, a variety of viral thy-
midine kinases, including those from Herpes Simplex I, Herpes
Simplex 2, Pseudorabies, and Varicella zoster, coined as ‘ambidex-
trous enzymes’, have also been found to be promiscuous with re-
4.1.2. N-Ac-D-Phe-a-ketophosphonate
In a 3-necked 250 mL round-bottomed flask at ꢀ78 °C under ar-
gon, 4.4 mL (41 mmol) dimethylmethylphosphonate was added to
45 mL dry THF (dried over sodium metal). Next, 16.5 mL of n-BuLi
(2.5 M in hexanes, 42 mmol) was added dropwise over 30 min. A
small quantity of white precipitate formed, which disappeared
over 15 min. Next, 4.5 g (20 mmol) N-Ac-D-Phe-OMe in 50 mL dry
THF were added all at once. The reaction was allowed to proceed
overnight and return to room temperature, turning yellow-orange.
The reaction mixture was quenched by the addition of 100 mL
water and washed twice with 50 mL diethyl ether. The aqueous
layer was acidified by 0.1 M HCl (pH 1.0), and then extracted four
times with 50 mL CH2Cl2. The methylene chloride extracts were
pooled, dried over MgSO4, and evaporated to dryness under re-
spect to ‘un-natural’ L
-thymidine.46–49
In the case of our olefinic irreversible inhibitors, the AOPHA
enantiomers have three major points of contact with the active site
of the enzyme: the salt bridge with Arg240, the hydrogen bond
with Tyr318, and the close hydrophobic interactions with
Phe112/Tyr318. Unlike the corresponding N-Ac-L-Phe-Gly sub-
duced pressure, yielding a crude yellow oil. The crude N-Ac-
D-
strate, the inclusion of the olefinic moiety results in the elimination
of the hydrogen bond that would form between a peptide substrate
and Asn316. We suggest that the lack of this hydrogen bond inter-
action with Asn316 gives rise to the loss of the subsite stereospec-
ificity of the enzyme. Indeed, flexible alignment overlays confirm
that both AOPHA enantiomers are able to closely overlay with N-
Phe- -ketophosphonate was purified by silica gel chromatography
a
from chloroform/methanol (20:1 v/v) yielding 6.1 g yellow oil
(19 mmol, 95%). 1H NMR ([2H]chloroform, TMS = 0.0 ppm): d 1.98
(s, 3H), d 2.98–3.30 (m, 4H), d 3.72–3.78 (m, 6H), d 4.82–4.94 (q,
1H), d 6.52 (d, 1H), d 7.10–7.32 (m, 5H).
Ac-L-Phe-Gly, whereas the substrate enantiomers themselves are
4.1.3. Methyl glyoxylate
poorly aligned at the glycine residue, and is evident from Figure 4
that the olefinic moieties of both (S)- and (R)-AOPHA are well-posi-
tioned to undergo the catalytic processing at the CuM center that
gives rise to irreversible inhibition. Thus, our modeling results
are remarkably consistent with the kinetic data for these mecha-
nism-based irreversible inhibitors.
First, 3.6 g (39 mmol) glyoxylic acid monohydrate and 100 mg
p-toluenesulfonic acid were dissolved in 4.6 mL (29 mmol) methyl
dimethoxyacetate in a 100 mL round-bottomed flask. The reaction
mixture was heated at reflux overnight. The next morning, the
reaction mixture was cooled to room temperature and 4 g phos-
phorus pentoxide were slowly added. The reaction mixture was
again heated at reflux (80 °C) for 4 h, allowed to cool to room tem-
perature and then distilled under reduced pressure. The product
which distilled at 70 °C was collected (3.0 mL) as a yellow oil. 1H
NMR ([2H]chloroform, TMS = 0.0 ppm) d 3.76 (singlet).
Amidation represents a potentially attractive target point for
modulating the production of bioactive peptides and thereby
affecting certain disease states. Amidated peptides such as sub-
stance P and calcitonin gene-related peptide are well established
mediators of inflammation, and we have demonstrated that our
irreversible amidation inhibitors exhibit potent anti-inflammatory
activity against both acute and chronic inflammation in rats.20,21
The proliferation of some tumor cells is dependent on autocrine
growth loops that require amidated autocrine growth factors,
and we have shown that an irreversible PAM inhibitor is selectively
cytotoxic toward WB-Ras-transformed rat liver epithelial cells ver-
sus untransformed cells, and also restores gap-junctional commu-
nication in these transformed cells.22 Since stereospecificity is a
key aspect in the design of enzyme-targeted inhibitors, the results
reported here should facilitate the future development of new clas-
ses of amidation inhibitors with therapeutic potential.
4.1.4. Methyl-(R)-5-acetamido-4-oxo-6-phenyl-2-hexenoic acid
At first, 6.1 g (19 mmol) N-Ac-D-Phe-a-ketophophonate were
placed in a 100 mL round-bottomed flask over ice. Next, 10 mL water
and 2.5 mL methyl glyoxylate were added, and after 5 min 20 mL of
potassium carbonate solution (0.25 g/mL) were added. A white pre-
cipitate formed immediately upon the addition of the carbonate
solution. The reaction was allowed to proceed for 30 min. The white
solid was collected by vacuum filtration, rinsed with cold water, and
recrystallized from ethanol/water. The white solid was again
collected by vacuum filtration and dried under vacuum over

M. S. Foster et al. / Tetrahedron: Asymmetry 22 (2011) 283–293
291
phosphorus pentoxide (1.1 g, 19%). 1H NMR ([2H]chloroform,
TMS = 0.0 ppm) d 1.98 (s, 3H), d 3.00–3.22 (m, 2H), d 3.80 (s, 3H), d
5.05–5.12 (q, 1H), d 6.04 (br s, 1H), d 6.76 (d, 1H, J = 15.9 Hz), d
7.05 (m, 2H), d 7.12 (d, 1H, J = 15.9 Hz), d 7.25 (m, 3H).
described above) and allowed to stir on ice for 5–10 min. Then,
6 mL potassium carbonate (20% w/v) was added to initiate the
reaction. A yellow precipitate formed immediately. The precipi-
tate was collected by vacuum filtration, and washed with water
and ethyl ether, yielding 0.70 g (2.7 mmol, 54%) off-white crystals.
Mp = 104–106 °C. 1H NMR ([2H]-chloroform, TMS = 0.0 ppm): d
1.80–2.29 (m, 2H), d 2.05–2.07 (s, 3H), d 2.08–2.12 (s, 3H), d
2.45–2.58 (m, 2H), d 3.82 (s, 3H), d 4.95–5.10 (m, 1H), d 6.30–
6.40 (bd, 1H), d 6.80–8.90 (d, 1H), d 7.22–7.28 (d, 1H). MS
(ESI+): 260 (M+1).
4.1.5. (R)-5-Acetamido-4-oxo-6-phenyl-2-hexenoic acid
Typically, 100 mg (0.4 mmol) methyl-(R)-AOPHA was dissolved
in 8 mL EtOH, and slowly added to 40 mL 100 mM TRIS buffer (pH
7.1) at 37 °C. To initiate the reaction, 100 lL of 6 mg/mL reconsti-
tuted pig liver esterase (E.C. 3.1.1.1, 15 U/mg) was added. The reac-
tion was monitored on a C8 Alltech Allsphere column at 260 nm
(25.0% CH3CN/74.9% water/0.1% TFA) at a flow rate of 1.5 mL/min.
The reaction went to >99% completion after 24 h. The reaction mix-
ture was ultrafiltered through an Amicon YM-10 membrane, the fil-
trate was acidified (pH 2.0) by addition of 0.1 M HCl, and extracted
four times with 10 mL EtOAc, dried over MgSO4, and evaporated un-
der reduced pressure to yield 80 mg (0.3 mmol, 75%) white solid. RP-
HPLC analysis indicates purity of greater than 99%. 1H NMR (DMSO-
d6, DMSO = 2.49 ppm): d 1.78 (s, 3H), d 2.71–3.08 (m, 2H), d 4.72 (m,
1H), d 6.54 (d, 1H, J = 15.9 Hz), d 7.06 (d, 1H, J = 15.9 Hz), d 7.21 (m,
4.2.4. (S)-5-Acetamido-7-methylthio-4-oxo-2-heptenoic acid
Crude ester, 1.8 g (6.9 mmol), was dissolved in a minimal
amount, (ꢄ10 mL) of CH3CH2OH. The dissolved sample was then
slowly added to 250 mL 100 mM Tris–Cl buffer, pH 7.3. The reac-
tion was initiated by the addition of 0.0065 g reconstituted porcine
liver esterase (E.C. 3.1.1.1, lyophile, >15 U/mL) and incubated at
37 °C. After several hours a finely dispersed precipitate was
formed. The reaction was monitored by RP-HPLC at a wavelength
of 214 nm on an Alltech Allsphere C-8 column, using a mobile
phase of 15.0% ACN:84.9% H2O:0.1% TFA at a flow rate of 1.5 mL/
min. The reaction was allowed to progress until the substrate peak
had disappeared completely, with the concomitant formation of a
new peak representing the free acid. Retention times for the ester
and the product acid were 21 and 6 min, respectively. The reaction
mixture was ultrafiltered using an Amicon YM-10 membrane. The
filtrate was rinsed three times with 50 mL Et2O, acidified to a pH of
1.0 with HCl, extracted four times with 50 mL CH2Cl2, dried over
MgSO4, filtered, and evaporated under reduced pressure to yield
a white solid. Recrystallized from methylene chloride and isolated
by vacuum filtration to yield 0.7 g white solid (2.9 mmol, 42%). The
product was analyzed by NMR and ESI Mass Spectrometry. Purity
was greater than 99%, as determined by RP-HPLC. 1H NMR ([2H]-
DMSO-d6, DMSO = 2.49 ppm): d 1.64–2.0 (m, 2H), d 2.82 (s, 3H), d
2.00 (s, 3H), d 2.40–2.55 (m, 2H), d 4.52–4.60 (m, 1H), d 6.57–
6.62 (d, 1H), d 7.00–7.07 (d, 1H), d 8.35–8.40 (bd, 1H). MS (ESI+):
246 (M+1).
5H), d 8.41 (d, 1H). Polarimetry: ½a D25
¼ 12:5 ꢃ 1:4 (c 0.10, MeOH).
ꢂ
4.2. Synthesis of (S)-5-acetamido-7-methylthio-4-oxo-2-
heptenoic acid
4.2.1. N-Ac-L-Met-OMe
In a typical reaction, 73 mL of acetic anhydride (780 mmol) and
63 mL pyridine (780 mmol) were combined in a round-bottomed
flask and chilled on ice. After 5–10 min, 8.6 g (100 mmol) L-methi-
onine methyl ester HCl were added and the reaction mixture was
allowed to slowly return to room temperature overnight. The next
morning, the reaction was quenched with cold water and extracted
four times with 75 mL of methylene chloride. The extracts were
then rinsed three times each with 1 M HCl, saturated sodium bicar-
bonate solution, and water. The extracts were then dried over
MgSO4, filtered, and evaporated under reduced pressure, yielding
a yellow oil which crystallized upon standing. The crude product
was recrystallized in ethyl ether at ꢀ20 °C. Crystals (19.3 g,
94 mmol, 94%) were isolated by vacuum filtration. Mp = 41.7–
42.4 °C. 1H NMR ([2H]-chloroform, TMS = 0.0 ppm): d 1.86–2.20
(m, 2H), d 2.00 (s, 3H), d 2.05 (s, 3H), d 2.45–2.60 (m, 2H), d 3.75
(s, 3H), d 4.62–4.70 (m, 1H), d 6.13–6.19 (bd, 1H).
4.3. Synthesis of TNP-D-Tyr-L-Val-Gly (TNP-YVG)
The synthesis of TNP-YVG was carried out according to previ-
ously described procedures.50 Briefly, 12 mg 2–4-6-trinitroben-
zenesulfonic acid hydrate and 11 mg
D-Tyr-Val-Gly were
4.2.2. N-Ac-L-Met-a-ketophosphonate
dissolved in 20 mL of MeOH/water (1:4) in a foil-wrapped round-
bottomed flask. The reaction was initiated via the addition of three
drops of Et3N, upon which the reaction mixture turned a deep or-
ange. After 30 min, 100 mL water was added and the reaction mix-
ture was acidified to pH 2.0 by the addition of 0.1 M HCl, turning
yellow. The reaction mixture was extracted three times with
25 mL EtOAc, and the extracts were pooled, dried over MgSO4,
and evaporated to dryness under reduced pressure. The residue
was dissolved in a minimum volume of 1:1 Et2O/EtOAc, and hex-
anes were added until the solution turned cloudy. After storing
overnight at ꢀ20 °C, the yellow precipitate was collected by vac-
uum filtration, dissolved in approximately 5 mL of MeOH, and
stored at ꢀ70 °C. Concentration was determined by UV–vis spec-
troscopy. The extinction coefficient of TNP-YVG is 12.2 ꢁ 103 at
350 nm. A yield of 50% is typical for this synthesis. RP-HPLC anal-
ysis revealed no impurities.
At first, N-Ac-
L
-Met-OMe was dried in a vacuum desiccator over
P2O5 flushed with argon gas. In a three-neck round-bottomed flask
under argon pressure, 11.0 mL n-butyllithium (2.5 M in hexanes,
28.0 mmol) was added dropwise to 3.4 mL dimethylmethylphosph-
onate (32.0 mmol) in 100 mL dried, distilled tetrahydrofuran at
ꢀ78 °C, and allowed to stand for 15 min, upon which a white pre-
cipitate formed. Then, 2.8 g (14.0 mmol) N-Ac-L-Met-OMe dis-
solved in 50 mL THF were added all at once. The reaction was
allowed to return to room temperature (15 h) and quenched with
100 mL water. Reaction mixture was rinsed twice with 100 mL
ethyl ether, acidified with dilute HCl to pH 1.0, and extracted three
times with 50 mL methylene chloride. Extracts were combined,
dried over magnesium sulfate, and evaporated under reduced pres-
sure to yield 4.0 g of a yellow oil (13.5 mmol, 96%). 1H NMR ([2H]-
chloroform, TMS = 0.0 ppm): d 1.75–2.20 (m, 2H), d 2.00 (s, 3H), d
2.02–2.10 (s, 3H), d 3.05–3.38 (m, 4H), d 3.62–3.78 (m, 6H), d
4.61–4.72 (m, 1H), d 7.09–7.20 (bd, 1H). MS (ESI+): 298 (M+1).
4.4. Synthesis of (S)-5-acetamido-4-oxo-6-phenyl-2-hexenoic
acid
4.2.3. Methyl-(S)-5-acetamido-7-methylthio-4-oxo-2-heptenoic
acid
This compound was synthesized using the previously described
procedure.17
First, 1.5 g (5.0 mmol) N-Ac-
combined with 0.5 g (5.7 mmol) methyl glyoxylate (synthesis
L-Met-a-ketophosphonate were

292
M. S. Foster et al. / Tetrahedron: Asymmetry 22 (2011) 283–293
4.5. Chiral HPLC chromatography
concentrations were determined by RP-HPLC analysis on an Alltech
Allsphere C8 column (250 ꢁ 4.2 mm, 5 m particle size) with an
authentic product standard curve. Mobile phase was 44.0:55.9:
0.1 acetonitrile–water–trifluoroacetic acid at flow rate of
l
Chiral separations of the enantiomers of 5-acetamido-4-oxo-6-
phenyl-2-hexenoic acid and methyl 5-acetamido-4-oxo-6-phenyl-
a
2-hexenoate were carried out on a Phenomenex (
D
)-Penicillamine
1.5 mL/min. Absorbance was measured at 344 nm.
column. All compounds were dissolved in HPLC-grade methanol
prior to injection. Absorbance was monitored at 260 nm on a
Waters LC-Module 1 Plus. The mobile phase was 74.5:25:0.5 3 M
CuSO4/methanol/trifluoroacetic acid. Flow rate was 1.0 mL/min.
4.9. Dilution assay
Purified PAM was initially incubated in the above assay solution
(37 °C), in the absence of TNP-YVG substrate, but in the presence of
various concentrations of Triton X-100 and 80 mM KI. Inhibition
reactions were initiated by the addition of purified PAM. Aliquots
4.6. Expression of X. Laevis skin AE-1
Suspension cultures of Sf9 cells were treated with a Baculovirus
Expression Vector (BEV) containing the gene for AE-1 created by
Nishikawa et al.51 Cells were counted via a hemocytometer and in-
fected at 10 plaque-forming units per cell. Briefly, the cell suspen-
sion was centrifuged at low speeds to pellet the Sf9 cells, which
were then resuspended with the appropriate volume of virus solu-
tion. Cells were incubated for 1 h with end-over-end rocking and
Ex-Cell medium was added such that the final cell concentration
was approximately 1 ꢁ 106 cells/mL. Cells were incubated in sus-
pension culture for five days and then harvested by Fast Protein Li-
quid Chromatography (FPLC).
of 10
reincubated in the standard PAM assay solution described above,
containing saturating concentrations of TNP-YVG (50 M), 0.1%
v/v Triton X-100, and 80 mM KI. The reactions were quenched after
30 min and analyzed as described above. Residual PAM activity
was measured as % initial activity of untreated control enzyme.
lL were withdrawn over an appropriate time course and
l
4.10. Progress curve inhibition kinetics
Purified PAM was incubated in the presence of varying concen-
trations of TNP-YVG (2.0–15
species (20–100 M) in 5 mL of the standard PAM assay solution
containing 0.1% v/v Triton X-100 and 80 mM KI. The reaction was
initiated by the addition of enzyme and 250 L aliquots were
lM) and the appropriate inhibitor
l
4.7. Enzyme isolation
l
First, AE-I was isolated from Sf9 suspension-cultured medium
using a procedure developed by Suzuki.51 and modified by Feng.17
Typical starting volumes were 500 mL to 1 L. Cells were pelleted by
low-speed centrifugation. The supernatant was loaded onto a
Substance P-Sepharose Fast Flow cation-exchange column
(2.6 ꢁ 40 cm, Pharmacia) equilibrated with 50 mM MES-sodium,
pH 6.2. After the medium was loaded, the column was eluted for
120 min at a flow-rate of 1.5 mL/min to remove non-adsorbed
material. The column was then eluted with a step-wise NaCl gradi-
ent: 0–150 mM NaCl over 30 min, 150–250 mM NaCl over
200 min, 250–500 mM NaCl over 50 min, and 500 mM NaCl for
300 min. Fractions were assayed for PAM activity by the procedure
given below. Fractions with activity were pooled and concentrated
over a 10,000 molecular-weight Amicon YM-10 membrane, and
the buffer was exchanged to 50 mM Tris–Cl, pH 8.5 buffer.
removed every minute, quenched with 10% v/v 3 M HClO4, centri-
fuged, and the supernatant was analyzed for product concentration
according to the chromatographic conditions described above.
4.11. PAM/ligand docking
All simulations were run using Molecular Operating Environ-
ment (MOE) software, Chemical Computing Group, Inc. (Montreal,
Canada). Additional code ‘more_dock.svl’ was obtained from the
SVL Exchange website (http://svl.chemcomp.com). All ligands
were constructed using the MOE Builder module and built from
the crystal-structure coordinates of the side-chain phenyl ring of
the 1OPM ligand IYG,24 thereby placing the side chains of all li-
gands in the hydrophobic pocket of the enzyme at the start of each
docking run. Polar hydrogens were added to the ligands (including
sp2 hydrogens) and each ligand was minimized using the Engh–
Huber force field with default parameters and ‘solvation’ enabled.
Non-bonded and bonded cutoffs were changed to 5.5 and 4.5 Å,
respectively.
The receptor used for molecular docking was the oxidized PAM
crystal structure, 1OPM.30 All water molecules, carbohydrates, and
metal ions except for CuH and CuM were deleted from the receptor
prior to molecular docking. Polar hydrogen atoms were added to
the PAM crystal structure receptor, heavy atoms were locked,
and the structure was energy-minimized using the Engh–Huber
force field. Default energy parameters were used for the minimiza-
tion and ‘solvation’ was enabled. Non-bonded and bonded cutoffs
were changed, as for the ligand minimization, to 5.5 and 4.5 Å,
respectively.
The concentrated AE-1 solution was then applied to a MonoQ
HR 10/10 column (Pharmacia) equilibrated with 50 mM Tris–Cl,
pH 8.5 buffer. The column was washed for 30 min at 1.0 mL/min
with starting buffer, and then eluted with a step-wise NaCl gradi-
ent: 0–50 mM NaCl for 15 min, 50–200 mM NaCl over 90 min,
200–500 mM NaCl over 30 min, and 500 mM NaCl for 30 min. Frac-
tions with PAM activity were pooled and concentrated over an
Amicon YM-10 membrane and applied to a Superose-12 HR 10/
30 gel-filtration column (Pharmacia). Activity was eluted with
HEPES pH 7.0 at a flow-rate of 0.5 mL/min for 90 min. Purified
AE-1 eluted as a single peak over several fractions. Activity was
pooled and concentrated, and stored as a 1:1 buffer/ethylene glycol
solution at ꢀ20 °C. Activity is expressed in mU/mL, where 1 U is
the amount of enzyme required to produce 1 lM TNP-YV(OH)G
per minute at 37 °C.
Docking simulations were performed using the Simulated
Annealing algorithm (Engh–Huber force field), with all default
parameters except that ‘solvation’ was enabled and the non-
bonded and bonded cutoffs were changed to 5.5 and 4.5 Å,
respectively. The docking box was centered at 40.4032 ꢁ
25.3060 ꢁ 40.4527 (x,y,z) of the crystal structure receptor and
had arbitrary dimensions of 60 ꢁ 60 ꢁ 60 (x,y,z). All residues that
were at least partially within the confines of the docking box were
selected, and then all residues within 5.5 Å (the non-bonded cut-
off) of those residues were selected. All other atoms were deleted.
Minimized ligands were then subjected to a non-random start
4.8. PAM activity assay
Briefly, the assay solution contains 1 mg/mL catalase, 4 mM
ascorbic acid, copper sulfate, and 35 TNP-YVG in
250 L 50 mM MES-Na, pH 6.5 at 37 °C. The conversion of TNP-
4
lM
lM
l
YVG to TNP-YV(OH)G was kept to under 30% in order to obtain
consistent initial rates. The reaction was quenched by the addition
of 10% v/v 3 M HClO4. Samples were centrifuged at 10,000 g for
30 min and the supernatant was assayed by RP-HPLC. Product

M. S. Foster et al. / Tetrahedron: Asymmetry 22 (2011) 283–293
293
22. Sunman, J. A.; Foster, M. S.; Folse, S. L.; May, S. W.; Matesic, D. F. Mol. Carcinog.
2004, 41, 231–246.
23. Chung, S. J.; Chung, S.; Lee, H. S.; Kim, E. J.; Oh, K. S.; Choi, H. S.; Kim, K. S.; Kim,
Y. J.; Hahn, J. H.; Kim, D. H. J. Org. Chem. 2001, 66, 6462–6471.
24. Kim, D. H. Mini-Rev. Med. Chem. 2001, 1, 155–161.
25. Lee, S. S.; Li, Z.-H.; Lee, D. H.; Kim, D. H. J. Chem. Soc., Perkin Trans. 1 1995, 2877–
2882.
26. Wang, Z. X.; Tsou, C. L. J. Theor. Biol. 1990, 142, 531–549.
27. Wang, Z. X. J. Theor. Biol. 1990, 147, 497–508.
simulation, and all bonds that are able to undergo rotation were al-
lowed to deviate continuously 30° from the minimized configura-
tion. Initial temperatures were 1000 K. Typically, twenty-five runs
were performed for each ligand, resulting in 25 docked conforma-
tions per ligand. Each run included six cycles of decreasing initial
temperature, with 8000 iterations per cycle. Top scorers were
determined from the electrostatic and van der Waals’ energy
contributions.
28. Tsou, C. L. Adv. Enzymol. Relat. Areas Mol. Biol. 1988, 61, 381–436.
29. Wilson, W. B. S.; Barber, Y.; Scallen, T.; Morrow, C. J. Org. Chem. 1983, 48, 3960–
3966.
Acknowledgments
30. Prigge, S. T.; Kolhekar, A. S.; Eipper, B. A.; Mains, R. E.; Amzel, L. M. Science 1997,
278, 1300–1305.
31. Kim, D. H. Curr. Top. Med. Chem. 2004, 4, 1217–1226.
32. Moreira, R.; Santana, A. B.; Iley, J.; Neres, J.; Douglas, K. T.; Horton, P. N.;
Hursthouse, M. B. J. Med. Chem. 2005, 48, 4861–4870.
33. Fleet, G. W. J.; Nicholas, S. J.; Smith, P. W.; Evans, S. V.; Fellows, L. E.; Nash, R. J.
Tetrahedron Lett. 1985, 26, 3127–3130.
34. Hakansson, A. E.; van Ameijde, J.; Guglielmini, L.; Horne, G.; Nash, R. J.; Evinson,
E. L.; Kato, A.; Fleet, G. W. J. Tetrahedron: Asymmetry 2007, 18, 282–289.
35. Yu, C. Y.; Asano, N.; Ikeda, K.; Wang, M. X.; Butters, T. D.; Wormald, M. R.;
Dwek, R. A.; Winters, A. L.; Nash, R. J.; Fleet, G. W. Chem. Commun. 2004, 1936–
1937.
36. Asano, N.; Ikeda, K.; Yu, L.; Kato, A.; Takebayashi, K.; Adachi, I.; Kato, I.; Ouchi,
H.; Takahata, H.; Fleet, G. W. J. Tetrahedron: Asymmetry 2005, 16, 223–229.
37. Mercer, T.; Jenkinson, S.; Bartholemew, B.; Nash, R.; Miyauchi, S.; Kato, A.;
Fleet, G. Tetrahedron: Asymmetry 2009, 20, 2368–2373.
38. Rinaldo-Matthis, A.; Murkin, A. S.; Ramagopal, U. A.; Clinch, K.; Mee, S. P.;
Evans, G. B.; Tyler, P. C.; Furneaux, R. H.; Almo, S. C.; Schramm, V. L. J. Am. Chem.
Soc. 2008, 130, 842–844.
39. Sabini, E.; Hazra, S.; Konrad, M.; Burley, S.; Lavie, A. Nucleic Acids Res. 2007, 35,
186–192.
40. Sabini, E.; Hazra, S.; Konrad, M.; Lavie, A. J. Med. Chem. 2007, 50, 3004–3014.
41. Ostermann, N.; Segura-Pena, D.; Meier, C.; Veit, T.; Monnerjahn, C.; Konrad, M.;
Lavie, A. Biochemistry 2003, 42, 2568–2577.
42. Pasti, C.; Gallois-Montbrun, S.; Munier-Lehmann, H.; Veron, M.; Gilles, A. M.;
Deville-Bonne, D. Eur. J. Biochem. 2003, 270, 1784–1790.
43. Alexandre, J. A.; Roy, B.; Topalis, D.; Pochet, S.; Perigaud, C.; Deville-Bonne, D.
Nucleic Acids Res. 2007, 35, 4895–4904.
44. Gondeau, C.; Chaloin, L.; Lallemand, P.; Roy, B.; Perigaud, C.; Barman, T.; Varga,
A.; Vas, M.; Lionne, C.; Arold, S. T. Nucleic Acids Res. 2008, 36, 3620–3629.
45. Varga, A.; Lionne, C.; Lallemand, P.; Szabo, J.; Adamek, N.; Valentin, C.; Vas, M.;
Barman, T.; Chaloin, L. Biochemistry 2009, 48, 6998–7008.
46. Focher, F.; Maga, G.; Bendiscioli, A.; Capobianco, M.; Colonna, F.; Garbesi, A.;
Spadari, S. Nucleic Acids Res. 1995, 23, 2840–2847.
47. Magrassi, L.; Finocchiaro, G.; Milanesi, G.; Benvenuto, F.; Spadari, S.; Focher, F.
Gene Ther. 2003, 10, 2052–2058.
48. Spadari, S.; Maga, G.; Focher, F.; Ciarrocchi, G.; Manservigi, R.; Arcamone, F.;
Capobianco, M.; Carcuro, A.; Colonna, F.; Iotti, S.; Garbesi, A. J. Med. Chem. 1992,
35, 4214–4220.
49. Spadari, S.; Maga, G.; Verri, A.; Bendiscioli, A.; Tondelli, L.; Capobianco, M.;
Colonna, F.; Garbesi, A.; Focher, F. Biochimie 1995, 77, 861–867.
50. Katopodis, A. G.; May, S. W. Biochem. Biophys. Res. Commun. 1988, 151, 499–
505.
51. Suzuki, K.; Shimoi, H.; Iwasaki, Y.; Kawahara, T.; Matsuura, Y.; Nishikawa, Y.
EMBO J. 1990, 9, 4259–4265.
We gratefully acknowledge the partial support of this work by
the National Institutes of Health (GM40540). We also gratefully
acknowledge Mrs. Maggie Schwab Gunnerson for her contributions
to the synthesis of the methionyl derivative.
References
1. Bradbury, A. F.; Smyth, D. G. Eur. J. Biochem. 1987, 169, 579–584.
2. Bradbury, A. F.; Smyth, D. G. Biosci. Rep. 1987, 7, 907–916.
3. Eipper, B. A.; Mains, R. E. Annu. Rev. Physiol. 1988, 50, 333–344.
4. Eipper, B. A.; Stoffers, D. A.; Mains, R. E. Annu. Rev. Neurosci. 1992, 15, 57–85.
5. Kato, I.; Yonekura, H.; Tajima, M.; Yanagi, M.; Yamamoto, H.; Okamoto, H.
Biochem. Biophys. Res. Commun. 1990, 172, 197–203.
6. Katopodis, A. G.; May, S. W. Biochemistry 1990, 29, 4541–4548.
7. Katopodis, A. G.; Ping, D.; Smith, C. E.; May, S. W. Biochemistry 1991, 30, 6189–
6194.
8. Perkins, S. N.; Husten, E. J.; Eipper, B. A. Biochem. Biophys. Res. Commun. 1990,
171, 926–932.
9. Ping, D.; Katopodis, A. G.; May, S. W. J. Am. Chem. Soc. 1992, 114, 3998–4000.
10. Tajima, M.; Iida, T.; Yoshida, S.; Komatsu, K.; Namba, R.; Yanagi, M.; Noguchi,
M.; Okamoto, H. J. Biol. Chem. 1990, 265, 9602–9605.
11. Takahashi, K.; Okamoto, H.; Seino, H.; Noguchi, M. Biochem. Biophys. Res.
Commun. 1990, 169, 524–530.
12. Bradbury, A. F.; Finnie, M. D. A.; Smyth, D. G. Nature 1982, 298, 686–688.
13. Eipper, B. A.; Mains, R. E.; Glembotski, C. C. Proc. Natl. Acad. Sci. U.S.A. 1983, 80,
5144–5148.
14. Ramer, S. E.; Cheng, H.; Palcic, M. M.; Vederas, J. C. J. Am. Chem. Soc. 1988, 110,
8526–8532.
15. Bradbury, A. F.; Smyth, D. G. Biochem. Biophys. Res. Commun. 1983, 112, 372–
377.
16. Ping, D.; Mounier, C. E.; May, S. W. J. Biol. Chem. 1995, 270, 29250–29255.
17. Feng, J.; Shi, J.; Sirimanne, S. R.; Mounier-Lee, C. E.; May, S. W. Biochem. J. 2000,
350, 521–530.
18. Mounier, C. E.; Shi, J.; Sirimanne, S. R.; Chen, B. H.; Moore, A. B.; Gill-
Woznichak, M. M.; Ping, D.; May, S. W. J. Biol. Chem. 1997, 272, 5016–5023.
19. Moore, A. B.; May, S. W. Biochem. J. 1999, 341, 33–40.
20. Bauer, J. D.; Sunman, J. A.; Foster, M. S.; Thompson, J. R.; Ogonowski, A. A.;
Cutler, S. J.; May, S. W.; Pollock, S. H. J. Pharmacol. Exp. Ther. 2007, 320, 1171–
1177.
21. Ogonowski, A. A.; May, S. W.; Moore, A. B.; Barrett, L. T.; O’Bryant, C. L.; Pollock,
S. H. J. Pharmacol. Exp. Ther. 1997, 280, 846–853.