Macrocyclic inhibitors for peptide deformylase: A structure-activity relationship study of the ring size

Article abstract of DOI:10.1021/jm049592c

Peptide deformylase (PDF) catalyzes the removal of the N-terminal formyl group from newly synthesized polypeptides in eubacteria. Its essential role in bacterial cells but not in mammalian cells makes it an attractive target for antibacterial drug design. We have previously reported an N-formylhydroxylamine- based, metal-chelating macrocyclic PDF inhibitor, in which the P ₁′ and P₃′ side chains are covalently joined. In this work, we have carried out a structure-activity relationship study on the size of the macrocycle and found that 15-17-membered macrocycles are optimal for binding to the PDF active site. Unlike the acyclic compounds, which are simple competitive inhibitors, the cyclic compounds all act as slow-binding inhibitors. As compared to their acyclic counterparts, the cyclic inhibitors displayed 20-50-fold higher potency against the PDF active site (K _I* as low as 70 pM), improved selectivity toward PDF, and improved the metabolic stability in rat plasma. Some of the macrocyclic inhibitors had potent, broad spectrum antibacterial activity against clinically significant Gram-positive and Gram-negative pathogens. These results suggest that the macrocyclic scaffold provides an excellent lead for the development of a new class of antibiotics.

Full text of DOI:10.1021/jm049592c

J . Med. Chem. 2004, 47, 4941-4949
4941
Ma cr ocyclic In h ibitor s for P ep tid e Defor m yla se: A Str u ctu r e-Activity
Rela tion sh ip Stu d y of th e Rin g Size
Xubo Hu,†,‡ Kiet T. Nguyen,†,‡ Vernon C. J iang,§ Denene Lofland,§ Heinz E. Moser,§ and Dehua Pei*,†
Department of Chemistry and Ohio State Biochemistry Program, The Ohio State University, 100 West 18th Avenue,
Columbus, Ohio 43210, and GeneSoft Pharmaceuticals Inc., 7300 Shoreline Court, South San Francisco, California 94080
Received May 28, 2004
Peptide deformylase (PDF) catalyzes the removal of the N-terminal formyl group from newly
synthesized polypeptides in eubacteria. Its essential role in bacterial cells but not in mammalian
cells makes it an attractive target for antibacterial drug design. We have previously reported
an N-formylhydroxylamine-based, metal-chelating macrocyclic PDF inhibitor, in which the P1′
and P3′ side chains are covalently joined. In this work, we have carried out a structure-activity
relationship study on the size of the macrocycle and found that 15-17-membered macrocycles
are optimal for binding to the PDF active site. Unlike the acyclic compounds, which are simple
competitive inhibitors, the cyclic compounds all act as slow-binding inhibitors. As compared to
their acyclic counterparts, the cyclic inhibitors displayed 20-50-fold higher potency against
the PDF active site (KI* as low as 70 pM), improved selectivity toward PDF, and improved the
metabolic stability in rat plasma. Some of the macrocyclic inhibitors had potent, broad spectrum
antibacterial activity against clinically significant Gram-positive and Gram-negative pathogens.
These results suggest that the macrocyclic scaffold provides an excellent lead for the
development of a new class of antibiotics.
In tr od u ction
advanced into clinical trials for the treatment of hos-
pitalized community-acquired pneumonia. However,
because most of these inhibitors are metal chelators
appended to peptidomimetics, there remains some
concern about their selectivity (e.g., inhibition of other
metalloproteases) and in vivo stability (e.g., proteolysis
of the peptide bonds). A popular approach to improving
both stability against proteolysis and selectivity of
peptidomimetic inhibitors is to form cyclic peptides or
depsipeptides.23,24 Structural studies of several PDF-
inhibitor complexes21,25,26 have revealed that the inhibi-
tors are bound in an extended conformation and the P1′
and P3′ side chains are similarly oriented. While the P2′
side chain is extended toward solvent, the P1′ and P3′
side chains are engaged in intimate interactions with
the enzyme. The P1′ side chain (usually an n-butyl
group) fits into a deep hydrophobic pocket in the PDF
active site. The P3′ side chain makes hydrophobic
contacts with a shallow pocket near the active site as
well as one face of the P1′ side chain. We have recently
reported a macrocyclic PDF inhibitor, in which the P1′
and P3′ side chains are covalently cross-linked (Figure
1, compound 1b).27 Compound 1b was an exceedingly
potent PDF inhibitor (apparent KI ) 0.67 nM) and was
approximately 10-fold more potent than the acyclic
counterpart 2 (BB-3497). We attributed the improved
potency against the enzyme to the rigidity introduced
by cyclization, which may lock the inhibitor into a PDF-
binding conformation. We anticipated that the increased
rigidity should also improve both selectivity and stabil-
ity by preventing their binding to other enzymes (e.g.,
other peptidases). In this work, we report the synthesis
and evaluation of a series of macrocyclic inhibitors of
different ring sizes. These cyclic inhibitors show good
antibacterial activity and markedly improved metabolic
stability and selectivity.
Peptide deformylase (PDF) has emerged as an excit-
ing new target for designing novel antibiotics to treat
antibiotic resistant pathogens.1-4 PDF is an essential
enzyme involved in bacterial protein biosynthesis and
maturation. In bacteria and the organelles of eukary-
otes, protein synthesis starts with an N-formylmethion-
ine, and as a result, all newly synthesized polypeptides
carry transiently a formylated N terminus.5 PDF cata-
lyzes the subsequent removal of the formyl group from
the vast majority of those polypeptides, many of which
undergo further N-terminal processing by methionine
aminopeptidase to produce mature proteins. PDF is
present in all eubacteria, and its essentiality for bacte-
rial survival has been demonstrated by both genetic
studies6-8 and arrest of bacterial growth by PDF
inhibitors.1-4 On the other hand, protein synthesis in
the eukaryotic cytoplasm does not involve N-terminal
formylation, and PDF apparently has no catalytic func-
tion in the mammalian mitochondrion.9 Furthermore,
PDF is a metallopeptidase that contains a tetrahedrally
coordinated, catalytically essential ferrous ion (Fe2+) in
the active site.10,11 These properties make PDF an
attractive antibacterial drug target.
Numerous PDF inhibitors have been reported in
recent years; essentially, all of them are metal chelators
such as thiols,12-14 hydroxamates,15-20 and N-formyl-
hydroxylamines or reverse hydroxamates.21,22 Many of
these inhibitors exhibit excellent antibacterial activities
in vitro and in animal studies. One of the reverse
hydroxamates from British Biotech, BB86398, has been
* To whom correspondence should be addressed. Tel: 614-688-4068.
Fax: 614-292-1532. E-mail: pei.3@osu.edu.
† The Ohio State University.
‡ These two authors contributed equally to this work.
§ Genesoft Pharmaceuticals Inc.
10.1021/jm049592c CCC: $27.50 © 2004 American Chemical Society
Published on Web 08/24/2004

4942 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 20
Hu et al.
Sch em e 1
Sch em e 2a
F igu r e 1. Structures of PDF inhibitors.
Ch em istr y
In the 15-membered cyclic inhibitor (Figure 1; com-
pound 1b) that we reported previously, the P1′ and P3′
side chains were replaced with a nonyl group.27 Molec-
ular modeling showed that the peptide backbone in 1b
should exist in an extended conformation. We antici-
pated that the ring size could greatly affect the confor-
mation and rigidity of the peptide backbone and,
therefore, the binding affinity. To determine the optimal
ring size(s), we designed and synthesized compounds
1a -e, which contain 13-20-membered macrocycles
(Figure 1). Our attempt to form a 12-membered mac-
rocycle (n ) 0) failed at the olefin metathesis step,
presumably due to geometric constraints in the corre-
sponding diene. We chose tert-leucine as the P2′ residue
in 1a -e because the tert-butyl group is well-documented
for its ability to improve both the metabolic stability and
the oral bioavailability of peptidyl drugs.21,22 However,
one of the main goals for designing cyclic PDF inhibitors
was to improve their in vivo stability through cyclization
so that residues other than tert-leucine can be tolerated
at the P2′ position. Because PDF has very broad
specificity for the P2′ side chain,21,25,26,28,29 this would
provide an opportunity to optimize the pharmacokinetic
parameters of the inhibitor by varying the P2′ structure.
To test this notion, a lysine was chosen as the P2′
residue in 1f-h . The lysine-containing compounds
should be susceptible to proteolysis by trypsin-like
proteases, thus allowing us to evaluate the effect of
cyclization on the inhibitor selectivity and stability. For
this reason, the open chain compounds 3 and 4 were
prepared for comparison.
a
Conditions: (a) Boc-tert-Leu-OH, EDC, CH2Cl2. (b) TFA. (c)
Compound 6, EDC, CH2Cl2. (d) (Pcy3)2Cl2RudCHPh, CH2Cl2,
reflux. (e) Pd/C, H2, MeOH-EtOAc.
trifluoroacetic acid (TFA) gives the amines 7a -e. The
coupling of amines 7a -e and acid 6 gave dienes 5a -e.
The terminal alkenes were cross-linked using Grubbs’s
ruthenium catalyst30,31 to produce the macrocycles 9a -
e. Subsequent hydrogenation on Pd/C removes the
benzyl group from the hydroxylamine moiety and re-
duces the ring carbon-carbon double bond to afford 1a -
e. Compounds 1f,g were similarly prepared, except that
NR-Fmoc-Nꢀ-Boc-L-lysine was used instead of N-Boc-tert-
leucine (Scheme 3). Following ring closure and hydro-
genation, the Boc group was removed by treatment with
TFA. Compound 1h was obtained by reacting 1f with
acetic anhydride. Compound 3 was prepared by catalytic
hydrogenation of diene 11a and treatment with TFA
(Scheme 3). Compounds 2 and 4 were synthesized as
previously described.9
A convergent synthesis for the inhibitors was devel-
oped (Scheme 1). The macrocycles may be readily
obtained from their diene precursors (5) via olefin
metathesis. The dienes can be prepared by coupling the
common acid 6 to the various alkenylamine 7. The
synthesis of acid 6 has been previously reported.27 The
synthesis of compounds 1a -e, which contain tert-leucine
as the P2′ residue, is illustrated in Scheme 2. The
alkenykamines 8a -e were prepared from the corre-
sponding alcohols or halides. Condensation of the amines
with N-Boc-tert-leucine followed by treatment with
Resu lts a n d Discu ssion
Slow -Bin d in g In h ibition of P DF . The macrocycles
1a -h were initially tested against PDF by coupling the
deformylase reaction with those of formate dehydroge-
nase32,33 or Aeromonas aminopeptidase (AAP).34 Sur-
prisingly, these inhibitors exhibited a slow-binding

Macrocyclic Inhibitors for Peptide Deformylase
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 20 4943
Sch em e 3a
a
Conditions: (a) Fmoc-Lys(Boc)-OH, EDC, CH2Cl2. (b) Piperi-
dine, CH2Cl2. (c) Compound 6, EDC, CH2Cl2. (d) (Pcy3)2Cl2-
RudCHPh, CH2Cl2, reflux. (e) Pd/C, H2, MeOH-EtOAc. (f) TFA,
CH2Cl2. (g) Ac2O.
F igu r e 2. Slow-binding inhibition of PDF by compound 1c.
(A) Reaction progress curves when PDF was added as the last
component. Each assay reaction contained buffer A, DPPI (0.1
U), f-Met-Lys-AMC (130 µM), the indicated concentration of
1c (0, 50, 100, or 150 nM), and PDF (4.0 nM). (B) Reaction
progress curves when substrate was added as the last com-
ponent. PDF (100 nM) and varying concentrations of 1c were
preincubated for 2 h and rapidly diluted (10-fold) into buffer
A containing f-Met-Lys-AMC (270 µM) and DPPI (0.1 U). a,
control reaction with 1.0 nM PDF (no inhibitor); b, 120 nM
inhibitor; and c, 140 nM inhibitor.
behavior35 against PDF. Coupled with their exceptional
potency, a detailed kinetic characterization of the
inhibitors using the above assay methods proved prob-
lematic. To this end, we recently developed a new PDF
assay by coupling with dipeptidyl peptidase I (DPPI)
and employing peptide f-Met-Lys-AMC as the sub-
strate.36 Because none of the PDF inhibitors that we
have tested showed any inhibition against DPPI (DPPI
is a cysteine protease), the new assay can be carried
out in a continuous fashion and is ideally suited for
kinetic characterization of the slow-binding inhibitors
1a -h . Figure 2A shows the reaction progress curves in
the absence and in the presence of inhibitor 1c. In the
absence of inhibitors, the reaction progress curve was
essentially a linear line. Straight lines were also ob-
served in the presence of competitive, acyclic inhibitors
2-4 (not shown). However, the cyclic inhibitors 1a -h
each resulted in biphasic curves. The inhibition kinetics
can be described by the equation
concentrations of the inhibitor for 2 h prior to the
addition of substrate. To distinguish slow-binding in-
hibition from time-dependent irreversible inactivation
and to determine the rate constant k6, the inhibitors
were preincubated with PDF to form the E•I* complex,
which was then rapidly diluted into an assay solution
containing the PDF substrate, f-Met-Lys-AMC. The
slow, time-dependent recovery of PDF activity demon-
strates that the inhibition is reversible and thus slow
binding in nature (Figure 2B). Curve fitting gave the
reactivation rate constant (k6), in this case 0.0037 min-1
for inhibitor 1c.
,
k5
KI
E + I h
Table 1 lists the inhibition constants for the tested
inhibitors. All of the cyclic inhibitors exhibited a slow-
binding behavior, whereas the acyclic compounds
(2-4) all acted as simple competitive inhibitors under
the same conditions, with KI values of 11,9 74, and 23
nM, respectively. A comparison of compounds 1a -e,
which only differ in their ring sizes, reveals that a 15-
17-membered ring (n ) 6-8) is optimal for binding to
the PDF active site (KI* ) 0.22-0.33 nM against
Escherichia coli PDF). For inhibitors of smaller (e.g.,
1a ) or larger (e.g., 1e) rings, cyclization actually slightly
decreased their potency (Table 1, compare the KI* values
of 1a ,e and the KI value of 2). Note that with the
exception of the largest ring 1e, the other four cyclic
inhibitors all have similar initial KI values (64-109 nM)
to that of the acyclic control 3 (74 nM). Presumably, the
h E•I*
•
E I
k6
where KI denotes the equilibrium inhibition constant
of the initial enzyme-inhibitor complex (E•I) and k5 and
k6 are the forward and reverse rate constants for the
slow conversion of the initial E•I complex into a tight-
binding complex E•I*, respectively. The overall potency
of the inhibitor is described by the equilibrium constant,
KI* ) KIk6/(k5 + k6). The initial inhibition constant KI
was determined by initiating the assay reactions (which
contained varying concentrations of the inhibitor) by the
addition of PDF and fitting the initial rates against the
Michaelis-Menten equation. The KI* values were de-
termined by preincubating the enzyme and various

4944 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 20
Hu et al.
Ta ble 1. Inhibition Constants against E. coli PDF
In Vitr o An tiba cter ia l Activity. The cyclic inhibi-
tors were first tested against E. coli and Bacillus
subtilis, the representative Gram-negative and Gram-
positive bacteria, respectively. For B. subtilis, there is
a general correlation between the antibacterial activity
and the KI* values against PDF, consistent with PDF
as the molecular target responsible for the observed
antibacterial activity. Compounds 1b-d , which are the
most active inhibitors against PDF (Table 1), also
exhibited the most potent antibacterial effect, with
minimal inhibitory concentrations (MIC90 values) of
0.7-1.4, 0.2-0.4, and 0.08-0.16 µg/mL, respectively
(Table 2). On the other hand, inhibitors 1a ,e-g had
both poorer KI* (3.0-36 nM) and MIC90 values (1.4 to
>32 µg/mL). Most of the compounds were less effective
against E. coli, and there was no correlation between
the KI* and the MIC90 values (Tables 1 and 2). This
suggests that other factors such as efflux pumps,
membrane permeability, intracellular distribution, and/
or intracellular stability are at play. A notable exception
is inhibitor 1f, which is equally active against E. coli
and B. subtilis (MIC90 ) 4-8 µg/mL). Compound 1f
contains a lysine as the P2′ residue (Figure 1). It was
previously reported that substitution of a P2′ lysine for
a leucine substantially improved the antibacterial activ-
ity of another PDF inhibitor.13 It is not yet clear how
the inclusion of an amine group improves the cellular
activity of the inhibitors.
Next, compounds 1b,e,g were selected for testing
against a panel of clinically significant pathogens
including Enterococcus faecalis, Haemophilus influen-
zae, Moraxella catarrhalis, Staphylococcus aureus, and
three Streptococcus pneumoniae strains. Compound 1b
had a good antibacterial activity against most of the
pathogens, with MIC values ranging from 0.5 to 16 µg/
mL (Table 2). The high activity of compound 1b against
H. influenzae strain ATCC 31517 is worth noting. This
pathogen is generally difficult to treat with other
antibiotics especially with the other reported PDF
inhibitors.22,37,38 To gain some insight into the molecular
basis of the observed high efficacy, we tested compound
1b against the purified H. influenzae PDF and found
1b to be an exceptionally potent inhibitor against the
enzyme (KI* ) 0.074 ( 0.016 nM). A KI* value of 0.071
( 0.007 nM was also determined for compound 1d
(against H. influenzae PDF). Compound 1e had a good
activity against M. catarrhalis and the S. pneumoniae
strains (MIC ) 0.031-8 µg/mL) but was poorly active
against E. faecalis, H. influenzae, and S. aureus (Table
2). Compound 1g was active only against M. catarrhalis,
a result not unexpected from its relatively poor KI*
value.
c
inhibitor
KI (nM)a
63 ( 12 13.7 ( 3.0
109 ( 5
KI* (nM)b k5 (min-1
)
k6 (min-1
)
1a
1b
1c
1d
1e
1f
1g
1h
2
0.022
1.2
1.6
0.50
1.0
0.0061 ( 0.0015
0.0038 ( 0.0010
0.0037 ( 0.0010
0.0015 ( 0.0008
0.012 ( 0.0058
0.0042 ( 0.0011
0.018 ( 0.0045
0.0048 ( 0.0008
N/A
0.33 ( 0.15
96 ( 17 0.23 ( 0.06
77 ( 20 0.22 ( 0.04
2100 ( 500
92 ( 20
258 ( 28
710 ( 170
11 ( 1
74 ( 17
23 ( 6
25 ( 6
3.0 ( 1.3
36 ( 9
12 ( 4
N/Ad
0.12
0.11
0.28
N/A
N/A
N/A
3
4
N/A
N/A
N/A
N/A
a
KI values were determined by the DPPI assay in a continuous
b
fashion. KI* values were determined by the AAP assay in the
end-point format. c k6 values were determined by diluting a
preformed E•I* complex into a large volume of assay buffer
d
containing f-Met-Lys-AMC and DPPI. N/A, not applicable (no
slow-binding behavior). All values reported represent the mean
( SD from three independent sets of experiments.
cyclic inhibitors initially bind to PDF in a manner
similar to the acyclic counterpart and then undergo a
slower conformational change in the inhibitor and/or
protein structure. The nature of the conformational
change(s) is yet unclear and will be the subject of future
work. Consistent with this model, compound 1a , which
has the smallest ring size and highest rigidity, is likely
more restricted from the conformational change(s) as
reflected by its slow k5 rate (0.022 min-1) and similar
KI (63 nM) and KI* values (14 nM). The greater KI value
for 1e suggests that its larger P1′-P3′ side chains may
impede its access to the enzyme active site. Alterna-
tively, the larger rings have more conformational flex-
ibility; thus, a smaller fraction of the molecules has the
proper preexisting conformation in the solution that can
fit the contour of the PDF active site. As a result, 1e
has an overall potency similar to that of the acyclic
control 3. Cyclic inhibitors that contain an L-lysine or
acetylated lysine as the P2′ residue (1f-h ) are generally
less potent than those containing tert-leucine at this
position (e.g., compare 1b vs 1f). This is mainly due to
slower conversion from E•I to E•I* (slower k5) for the
former. The same trend is also observed for the acyclic
inhibitors (2 vs 3 and 4), although the difference is
smaller. Again, the 15-membered ring 1f is an order of
magnitude more potent than the 20-membered ring 1g.
The molecular basis of the slower k5 for 1f-h is not yet
clear. Another general trend is that formation of the
optimal-sized macrocycle (15-17-membered) improves
the overall potency of an inhibitor by 20-50-fold over
their acyclic counterparts (compare 1b-d vs 2; 1f vs
3). We believe that the improved potency is due to the
cross-linked side chain locking the inhibitor into the
PDF-binding conformation.
Ta ble 2. In Vitro Antibacterial Activity of PDF Inhibitors
MIC90 (µg/mL)
bacteria
1a
>24
1.4-2.8
NDa
ND
ND
ND
ND
ND
1b
1c
>12
1d
1e
1f
1g
E. coli
B. subtilis
∼12
>24
0.08-0.16
ND
>24
∼16
>32
>32
<0.031
>32
8
4-8
∼8
>32
>32
32
32
1
0.7-1.4
0.2-0.4
ND
E. faecalis (ATCC 29212)
H. influenzae (ATCC 31517)
M. catarrhalis (ATCC 37054)
S. aureus (ATCC 29213)
S. pneumoniae (ATCC 49619)
S. pneumoniae (ATCC 6301)
S. pneumoniae (ATCC 6303)
>32
0.5
0.62
16
4
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
32
>32
>32
>32
4
2
8
4
ND
a
ND, not determined.

Macrocyclic Inhibitors for Peptide Deformylase
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 20 4945
Ta ble 3. Inhibition of MMPsa
MMP-1 (%) MMP2 (%) MMP-3 (%) MMP-9 (%)
1.0
10
1.0
10
1.0
10
1.0
10
inhibitor µM
µM
µM
µM
µM
µM
µM
µM
1a
1b
1e
1g
1h
3
0
7
7
1
3
1
14
8
10
4
0
15
8
0
15
20
98
55
3
3
3
0
0
20
9
4
6
8
8
5
4
4
8
11
9
6
5
8
7
6
13
100
97
17
73
34
50
50
81
68
99
73
61
54
94
89
4
a
The values reported are the percentages of MMP activity
inhibited in the presence of indicated concentrations of inhibitors.
Each inhibitor was tested at two different concentrations (1.0 and
10 µM).
F igu r e 3. Comparison of the in vitro stability of PDF
inhibitors 1f (cyclic) and 3 (acyclic) in rat plasma.
Cycliza tion Im p r oves In h ibitor Selectivity. A
major concern of any metal-chelating PDF inhibitor is
its potential inhibition of the numerous metalloprotein-
ases in the host, such as the distantly related matrix
metalloproteases (MMPs). We tested several of the cyclic
inhibitors (1a ,b,e,g,h ) and the acyclic controls 3 and 4
against a panel of human MMPs (MMP-1, MMP-2,
MMP-3, and MMP-9). At 10 µM, the acyclic compound
3 resulted in essentially complete inhibition of all four
MMPs; even at 1 µM, it caused 50-80% inhibition
(Table 3). Acyclic compound 4 had a similar inhibition
profile. In contrast, none of the cyclic inhibitors showed
greater than 20% inhibition of MMP activity at 10 µM
inhibitor concentration (Table 3). It has previously been
reported that acyclic compound 2 is highly selective for
PDF, showing minimal inhibition of MMPs, enkepha-
linase, or ACE.21 The difference between inhibitors 2
and 3 or 4 is likely due to the presence of a bulky tert-
butyl group at the P2′ position of 2, which prevents
binding to the active sites of these other enzymes.
Remarkably, even the cyclic inhibitors that do not
contain the tert-butyl group at the P2′ position (e.g.,
compounds 1g,h ) showed little inhibition of the MMPs.
Thus, cyclization per se was sufficient to largely prevent
the inhibition of MMPs. Presumably, cyclization locked
the inhibitors into conformations that do not fit the
active sites of these MMPs.
Cycliza tion Im p r oves th e Sta bility of In h ibitor s.
Inhibitors 1f and 3 were chosen to test for the effect of
cyclization on inhibitor stability. They both have an
L-lysine as the P2′ residue and are expected to be
sensitive to proteolytic degradation by trypsin-like
enzymes. Both inhibitors were incubated in rat plasma
at 37 °C, and aliquots were withdrawn at various times
and analyzed by liquid chromatography-mass spec-
trometry (LC-MS). The cyclic inhibitor (1f) was very
stable under the experimental conditions, showing no
detectable degradation after 5 h (Figure 3). In contrast,
the acyclic compound (3) showed time-dependent deg-
radation, with approximately 25% loss after 5 h. There-
fore, cyclization significantly improves the stability of
the inhibitors against metabolic degradation. Our at-
tempts to identify the degradation products by MS
failed.
the two side chains. The ring size greatly affects the
inhibitor properties, with 15-17-membered rings as the
optimal ring size. As compared to their acyclic counter-
parts, the cyclic inhibitors of the optimal ring sizes show
a much higher potency against PDF (>20-fold), im-
proved stability against proteolytic degradation, and
higher selectivity for PDF over other metalloproteases.
To the best of our knowledge, the cyclic inhibitors rank
among the most potent inhibitors reported to date
against the PDF enzyme. Some of the inhibitors have
good antibacterial activity against a wide spectrum of
pathogens. Further development of these cyclic scaffolds
may provide novel PDF inhibitors with improved meta-
bolic stability and higher selectivity against human
MMPs.
Exp er im en ta l Section
Gen er a l. E. coli PDF was overexpressed and purified to
apparent homogeneity with cobalt(II) as the divalent metal
cofactor.39 H. influenzae PDF was overproduced in E. coli with
cobalt(II) as the metal cofactor and a C-terminal histidine tag
and was purified on a cobalt(II) affinity column in a manner
similar to that described for E. coli and B. subtilis PDF.13,39
Mca-P-K-G-Dpa-A-R-NH2 fluorogenic peptide substrate I and
Mca-R-P-K-P-V-E-Nval-W-R-K(Dnp)-NH2 fluorogenic peptide
substrate II were purchased from R&D Systems (Minneapolis,
MN). Cathepsin C (DPPI) and human MMPs (MMP-1, MMP-
2, MMP-3, and MMP-9) were purchased from Sigma Chemical
Co. (St. Louis, MO). Other chemicals were obtained from either
Sigma-Aldrich (St. Louis, MO) or Bio-Rad Laboratories (Her-
cules, CA). Spectroscopic measurements were performed on a
Perkin-Elmer Lambda 20 UV/vis spectrophotometer. Fluoro-
genic assays were performed with an Aminco-Bowman Series
2 Luminescence spectrometer. Molecular modeling was per-
formed with inhibitor 1b and was based on the crystal
structure of E. coli PDF complexed with inhibitor 221 (PDB
access code 1G27) with the modeling program FLO/QXP.40 The
binding site model consisted of all protein residues within 10.0
Å of inhibitor 2. In the binding site model, the P1′ n-butyl and
P3′ side chains of 2 were cross-linked with polymethylene
linkers of various lengths, followed by energy minimization
with the peptide backbone fixed. Only the molecule with a
nonyl group between the P1′ CR carbon and the P3′ amino
group was able to fit in the binding site while maintaining
staggered conformations in the linker. Subsequently, the entire
molecule was subjected to torsional Monte Carlo docking
searches, but no lower energy binding modes were found other
than the designed one.
Con clu sion
Bu ffer s. Buffer A: 50 mM Hepes, 10 mM NaCl, 5 mM
dithiothreitol, pH 7.0. Buffer B: 100 mM Tris, 100 mM NaCl,
10 mM CaCl2, 0.04% Brij-35. Buffer C: 50 mM Mops, pH 7.0.
Gen er a l P r oced u r e for th e Syn th esis of Am in oa lk en es
(8a -e). To a solution of terminal alkenol (5.0 mmol) and
On the basis of earlier observations that the P1′ and
P3′ side chains of PDF inhibitors are closely packed in
the PDF-inhibitor complex, a new class of macrocyclic
PDF inhibitors has been developed by covalently linking

4946 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 20
Hu et al.
triethylamine (5.5 mmol) in dichloromethane (20 mL) was
added methanesulfonyl chloride (5.5 mmol) dropwise at -5 °C.
The mixture was continuously stirred until the reaction was
complete (∼2 h). Dichloromethane (50 mL) was added to the
mixture followed by washing with 10% sodium bicarbonate
solution (2 × 15 mL) and brine (15 mL). The organic layer
was dried over sodium sulfate and concentrated to give the
alkenyl methanesulfonyl ester in quantitative yield, which was
used next without further purification. The methanesulfonyl
ester (4.0 mmol) and sodium azide (10 mmol) were dissolved
in 8:1 dimethyl formamide-H2O (16 mL) and stirred overnight
at 50 °C. The mixture was allowed to cool to room temperature,
diluted in 50 mL of water, and extracted with diethyl ether (3
× 30 mL). The ethereal layers were combined, washed with
brine (2 × 15 mL), dried over sodium sulfate, and concentrated
in vacuo to give the corresponding azidoalkene. The azide (5.0
mmol) was dissolved in diethyl ether (50 mL), and lithium
aluminum hydride (5.0 mmol) was added in three portions
under argon in a water bath. The grayish suspension was
vigorously stirred for ∼40 min and then quenched by the
addition of water (5 mL). The sodium hydroxide solution was
added to dissolve the solid, and the resulting solution was
extracted with diethyl ether (5 × 30 mL). The organic phase
was dried over sodium sulfate and concentrated to give the
corresponding amines 8d ,e. For amines 8a ,b, after the LiAlH4
reaction was quenched with water, Boc2O (5.5 mmol) was
added and stirred overnight. The ether layer was separated,
and the residue was rinsed with ether (2 × 20 mL). The ether
solutions were combined, dried over sodium sulfate, and
concentrated to give the corresponding Boc-aminoalkenes.
Removal of the Boc group with TFA in dichloromethane
followed by exhaustive pumping gave amines 8a ,b as trifluo-
roacetate salts.
(m, 2H), 1.53-1.27 (m, 6H), 1.43 (s, 9H), 1.00 (s, 9H). 13C NMR
(63 MHz, CDCl3): δ 171.2, 156.3, 139.0, 114.8, 79.9, 62.7, 39.7,
34.8, 33.9, 29.8, 28.9, 28.7, 27.0, 26.8. ESI-HRMS: Calcd for
C
18H34N2O3Na+ ([M + Na]+), 349.2462; found, 349.2456. This
amide (0.75 g, 2.30 mmol) was dissolved in dichloromethane
(4 mL), and TFA (3 mL) was added and stirred for 2 h until
all of the amide was consumed. The volatile substances were
removed under vacuum. The residue was dissolved in ethyl
acetate (50 mL) and was washed with 5% NaHCO3 (2 × 40
mL) and brine, dried (sodium sulfate), filtered, and concen-
trated to give the amine (0.49 g, 94%). 1H NMR (250 MHz,
CDCl3): δ 6.84 (m, 1H), 5.80 (m, 1H), 4.96 (m, 2H), 3.26 (m,
2H), 3.13 (brs, 1H), 2.19 (brs, 2H), 2.04 (m, 2H), 1.57-1.28
(m, 6H), 1.00 (s, 9H). 13C NMR (63 MHz, CDCl3): δ 173.4,
139.1, 114.8, 64.6, 39.4, 34.0, 29.9, 28.9, 27.1, 26.8. ESI-
HRMS: Calcd for C13H26N2ONa+ ([M + Na]+), 249.1937; found,
249.1950.
Nr-{[(2R-N-Ben zyloxy-N-for m yla m in o)m eth yl]-6-h ep -
t en oyl}-N-(6-h ep t en -1-yl)-L-ter t-leu cin a m id e (5c). This
compound was prepared by coupling amine 7c and acid 6 in a
manner similar to that described for 7c (88% yield). Compound
5c exists as a mixture of cis and trans isomers at the
formamide moiety. 1H NMR (250 MHz, CDCl3): δ 8.11 (brs,
0.6H), 7.86 (brs, 0.4H), 7.36 (m, 5H), 6.70 (brs, 1H), 6.61 (d, J
) 9.5 Hz, 1H), 5.76 (m, 2H), 5.09-4.79 (m, 6H), 4.31 (d, J )
9.5 Hz, 1H), 3.68 (m, 1.7H), 3.36 (m, 1H), 3.08 (m, 1.3H), 2.66
(m, 0.8H), 2.43 (m, 0.2H), 2.01 (m, 4H), 1.54-1.23 (m, 10H),
0.96 (s, 9H). 13C NMR (63 MHz, CDCl3): δ 173.6, 170.7, 163.3,
139.0, 138.3, 134.6, 129.9, 129.5, 129.1, 115.4, 114.9, 60.9, 46.5,
45.6, 39.8, 34.8, 33.9, 30.6, 29.9, 28.9, 27.0, 26.8, 26.5. ESI-
HRMS: Calcd for C29H45N3O4Na+ ([M + Na]+), 522.3302;
found, 522.3324.
N-(3-ter t-Bu t yl-2,5-d ioxo-1,4-d ia za -cycloh exa d ec-10-
en -6-yl)m eth yl-N-ben zyloxy-for m a m id e (9c). To a solution
of alkene 5c in dichloromethane (1.0 mM) was added 7% molar
equiv of Grubb’s catalyst.30,31 The solution was stirred at 40
°C for ∼20 h, allowed to cool to room temperature, and
concentrated to dryness. The residue was purified by flash
chromatography to give the cyclic monomer 9c as a mixture
of cis and trans isomers at the formamide moiety (93% yield).
1H NMR (250 MHz, CDCl3): δ 8.10 (brs, 0.59H), 7.88 (brs,
0.41H), 7.37 (m, 5H), 6.49-6.28 (m, 2H), 5.45 (m, 0.16H), 5.34
(m, 0.84H), 5.20 (m, 1H), 4.98 (brs, 0.84H), 4.78 (brs, 1.16H),
4.28 (m, 1H), 3.79 (m, 2.66H), 3.08 (m, 0.34H), 2.83 (m, 1H),
2.57 (m, 1H), 2.11-1.78 (m, 4H), 1.55-1.03 (m, 10H), 0.94 (s,
9H). 13C NMR (63 MHz, CDCl3): δ 173.7, 170.9, 163.2, 134.6,
131.2, 131.1, 129.9, 129.4, 129.1, 61.1, 46.2, 39.2, 35.3, 32.8,
32.3, 30.2, 28.7, 26.9, 25.5. ESI-HRMS: Calcd for C27H41N3O4-
Na+ ([M + Na]+), 494.2989; found, 494.2972.
3-Bu ten yla m m on iu m Tr iflu or oa ceta te (8a ).41 1H NMR
(250 MHz, CDCl3): δ 7.53 (brs, 3H), 5.74 (m, 1H), 5.25 (m,
2H), 3.13 (m, 2H), 2.45 (m, 2H).
5-Hexen yla m m on iu m Tr iflu or oa ceta te (8b).41 1H NMR
(250 MHz, D2O): δ 6.11-5.95 (m, 1H), 5.26-5.14 (m, 2H), 3.14
(t, J ) 7.3 Hz, 2H), 2.25 (m, 2H), 1.87-1.75 (m, 2H), 1.67-
1.58 (m, 2H).
1
7-Octen yla m in e (8d ). H NMR (250 MHz, CDCl3): δ 5.82
(m, 1H), 4.95 (m, 2H), 2.68 (t, J ) 6.8 Hz, 2H), 2.07 (m, 2H),
1.49-1.11 (m, 10H).
10-Un d ecen yla m in e (8e).42 1H NMR (250 MHz, CDCl3):
δ 5.79 (m, 1H), 4.96 (m, 2H), 2.68 (t, J ) 6.8 Hz, 2H), 2.03 (m,
2H), 1.53-1.04 (m, 16H).
6-Hep ten yla m in e (8c).43 LiAlH4 (0.42 g, 11.0 mmol) and
AlCl3 (1.47 g, 11. 2 mmol) were added to ether (20 mL) in an
ice bath. After this mixture was stirred for 5 min, a solution
of 6-heptenitrile (1.00 g, 9.16 mmol) in ether (6 mL) was added
dropwise and stirred until all of the nitrile was consumed.
Then, 1.0 M NaOH was added to adjust the pH to ∼10 and
Boc2O (2.18 g, 10.0 mmol) was added. About 3 h later, the
organic layer was separated and the aqueous layer was
extracted with ether (20 mL). The ether extracts were com-
bined and washed with water (2 × 20 mL), dried, and purified
to give 1-N-Boc-6-heptenylamine (1.85 g, 95%). Removal of the
Boc group with TFA gave the titled compound as its trifluo-
roacetate form, in quantitative yield. 1H NMR (250 MHz,
D2O): δ 5.78 (m, 1H), 4.89 (m, 2H), 2.86 (t, d ) 7.4 Hz, 2H),
1.97 (m, 2H), 1.54 (m, 2H), 1.37-1.18 (m, 4H).
N-(3-ter t-Bu tyl-2,5-d ioxo-1,4-d ia za -cycloh exa d ec-6-yl)-
m eth yl-N-h yd r oxy-for m a m id e (1c). The cyclized compound
9c (50.0 mg) was dissolved in ethyl acetate and methanol
(1:1, 10 mL), and 20 mg of 10% Pd/C was added. The mixture
was exposed to 1 atm of hydrogen until all of the starting
material was consumed. The charcoal was removed by filtra-
tion, and the filtrate was concentrated to give the desired
compound in 97% yield (as a mixture of cis and trans isomers).
Reversed-phase HPLC: 99% purity. 1H NMR (250 MHz,
CDCl3): δ 8.35 (brs, 0.23H), 7.82 (brs, 0.77H), 7.50 (d, J ) 9.3
Hz, 0.23H), 7.01 (d, J ) 9.1 Hz, 1H), 6.58 (m, 0.77H), 4.41 (m,
1H), 3.81 (m, 2H), 3.42 (m, 1H), 2.92-2.84 (m, 2H), 1.58-1.21
(m, 18H), 0.94 (s, 9H). 13C NMR (63 MHz, CDCl3): δ 173.6,
171.2, 157.5, 61.2, 53.2, 45.5, 39.5, 35.1, 30.6, 29.5, 28.0, 27.4,
27.0, 26.8, 26.4, 25.0, 24.8, 21.4. ESI-HRMS: Calcd for
N-(6-Hep t en -1-yl)-L-t er t -leu cin a m id e (7c). L-Boc-tert-
leucine (0.65 g, 2.81 mmol) and amine 8c (as its trifluoro-
acetate salt, 0.60 g, 2.64 mmol) were dissolved in a solution of
dichloromethane (15 mL) and triethylamine (0.42 mL, 3.0
mmol), and then, 1-(3-dimethylaminopropyl)-3-ethylcarbodi-
imide hydrochloride (0.54 g, 2.82 mmol) was added. After it
was stirred for about 2 h, the mixture was diluted with ethyl
acetate (60 mL), washed with 5% NaHCO3 (2 × 40 mL) and
brine, dried over sodium sulfate, filtered, and concentrated to
dryness. The residue was purified by flash chromatography
C
20H37N3O4Na+ ([M + Na]+), 406.2676; found, 406.2673.
N-(3-ter t-Bu t yl-2,5-d ioxo-1,4-d ia za -cyclot r id ec-6-yl)-
m eth yl-N-h yd r oxy-for m a m id e (1a ). This compound (a mix-
ture of cis and trans isomers) was synthesized in a manner
similar to 1c. Reversed-phase HPLC: 99% purity. 1H NMR
(400 MHz, CDCl3): δ 8.36 (brs, 0.25H), 7.85 (brs, 0.75H), 7.22
(brs, 0.42H), 6.88 (brs, 1.58H), 4.34 (d, J ) 6.0 Hz, 0.25H),
4.30 (d, J ) 5.3 Hz, 0.75H), 4.00 (brs, 0.25H), 3.84 (brs, 0.75H),
3.73 (m, 1H), 3.42 (m, 1H), 2.97 (m, 1H), 2.71 (m, 1H), 1.62-
1.24 (m, 13H), 1.00 (s, 9H). 13C NMR (100 MHz, CDCl3): δ
1
to give the amide (0.75 g, 87%). H NMR (250 MHz, CDCl3):
δ 6.32 (m, 1H), 5.78 (m, 1H), 5.38 (d, J ) 9.5 Hz, 1H), 4.95 (m,
2H), 3.86 (d, J ) 9.5 Hz, 1H), 3.28 (m, 1H), 3.16 (m, 1H), 2.03

Macrocyclic Inhibitors for Peptide Deformylase
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 20 4947
174.5, 171.9, 157.6, 61.9, 52.2, 44.7, 39.4, 34.0, 30.0, 27.2, 26.7,
26.3, 24.4, 23.5. ESI-HRMS: Calcd for C17H31N3O4Na+ ([M +
Na]+), 364.2207; found, 364.2207.
N-(3-ter t-Bu tyl-2,5-d ioxo-1,4-dia za -cyclop en ta d ec-6-yl)-
m eth yl-N-h yd r oxy-for m a m id e (1b). The synthesis of this
compound has previously been described.27 Reversed-phase
HPLC: 96% purity.
(m, 3.5H), 2.53 (brs, 1H), 1.99 (m, 4H), 1.79-1.26 (m, 24H),
1.43 (s, 9H). 13C NMR (63 MHz, CDCl3): δ 171.6, 156.6, 131.1,
130.1, 129.6, 129.1, 53.3, 40.5, 39.6, 33.1, 31.6, 29.9, 29.4, 28.9,
28.3, 27.9, 27.6, 27.1, 26.5, 25.7, 23.0. ESI-HRMS: Calcd for
C
36H58N4O6Na+ ([M + Na]+), 665.4249; found, 665.4227.
N-[3-(4-Am in obu tyl)-2,5-d ioxo-1,4-d ia za -cycloicos-6-yl]-
m eth yl-N-h yd r oxy-for m a m id e (1g). Compound 12b was
hydrogenated to give the Boc-protected 1g in quantitative
yield. 1H NMR (250 MHz, CDCl3): δ 8.37 (brs, 0.1H), 7.76 (brs,
0.9H), 7.22 (m, 1H), 6.58 (brs, 0.1H), 6.37 (brs, 0.8H), 5.99 (brs,
0.1H), 5.08 (brs, 0.7H), 4.90 (brs, 0.1H), 4.39 (m, 1H), 3.99 (brs,
0.2H), 3.83-3.65 (m, 1H), 3.57-3.35 (m, 2H), 3.10-2.88 (m,
4.8H), 2.63 (m, 0.2H), 1.79-1.25 (m, 32H), 1.43 (m, 9H). ESI-
HRMS: Calcd for C29H54N4O6Na+ ([M + Na]+), 577.3936;
found, 577.3917. This intermediate (20 mg) was dissolved in
dichloromethane (1 mL) and treated with TFA (0.5 mL). After
it was stirred for 2 h at room temperature, the mixture was
concentrated to dryness to give 1g. Reversed-phase HPLC:
95% purity. 1H NMR (250 MHz, CD3OD): δ 8.32-8.26 (m,
0.4H), 7.87 (brs, 0.6H), 4.44 (m, 1H), 3.89-3.71 (m, 1H), 3.58-
3.45 (m, 2H), 3.00 (m, 4H), 1.82-1.37 (m, 32H). 13C NMR (63
MHz, CD3OD): δ 176.0, 174.1, 160.4, 54.4, 46.2, 41.2, 40.9,
33.2, 31.8, 30.8, 30.3, 29.9, 29.3, 28.9, 28.6, 28.3, 28.2, 27.6,
27.5, 24.2. ESI-HRMS: Calcd for C24H46N4O4Na+ ([M + Na]+),
477.3411; found, 477.3410.
N -(5-H e x e n -1-y l)-N E-t er t -b u t o x y c a r b o n y l-L -ly s in -
a m id e (10a ). This compound was synthesized as described
for 10b. Amine 8b was coupled with NR-Fmoc-Nꢀ-Boc-L-lysine
to give NR-Fmoc-Nꢀ-Boc-lysyl (5-hexenyl)amide in 44% yield.
1H NMR (250 MHz, CDCl3): δ 7.71 (d, J ) 7.4 Hz, 2H), 7.53
(d, J ) 7.4 Hz, 2H), 7.38-7.25 (m, 4H), 6.62 (brs, 1H), 5.98 (d,
J ) 7.8 Hz, 1H), 5.72 (m, 1H), 4.92 (m, 2H), 4.71 (brs, 1H),
4.32 (m, 2H), 4.14 (m, 2H), 3.19 (m, 2H), 3.07 (m, 2H), 1.98
(m, 2H), 1.80-1.28 (m, 10H), 1.38 (s, 9H). ESI-HRMS: Calcd
for C32H43N3O5Na+ ([M + Na]+), 572.3095; found, 572.3093.
This amide was treated with piperidine to afford 10a in 81%
yield. 1H NMR (250 MHz, CDCl3): δ 7.31 (m, 1H), 5.77 (m,
1H), 4.96 (m, 2H), 4.69 (brs, 1H), 3.10 (m, 1H), 3.22 (m, 2H),
3.09 (m, 2H), 2.06 (m, 2H), 1.57 (m, 1H), 1.51-1.34 (m, 9H),
1.50 (s, 2H), 1.42 (s, 9H). 13C NMR (63 MHz, CDCl3): δ 175.3,
156.5, 138.6, 115.0, 79.0, 55.3, 40.4, 39.1, 34.9, 33.5, 30.1, 29.3,
28.7, 26.4, 23.2.
N-(3-ter t-Bu tyl-2,5-d ioxo-1,4-dia za -cycloh ep ta d ec-6-yl)-
m eth yl-N-h yd r oxy-for m a m id e (1d ). This compound (a mix-
ture of cis and trans isomers) was prepared in a manner
similar to 1c. Reversed-phase HPLC: 99% purity. 1H NMR
(400 MHz, CDCl3): δ 9.33 (brs, 1H), 8.37 (s, 0.27H), 7.85 (s,
0.73H), 7.37 (brd, J ) 8.1 Hz, 0.27H), 6.88 (brd, J ) 8.1 Hz,
0.73H), 6.55 (brd, J ) 6.5 Hz, 0.27H), 6.25 (brd, J ) 6.5 Hz,
0.73H), 4.33 (d, J ) 8.2 Hz, 1H), 3.86-3.74 (m, 2H), 3.42 (m,
1H), 2.85 (m, 2H), 1.59 (m, 2H), 1.47-1.24 (m, 18H), 0.95 (s,
9H). 13C NMR (100 MHz, CDCl3): δ 173.5, 171.0, 157.5, 61.2,
54.3, 53.1, 45.3, 39.0, 35.4, 32.1, 30.9, 29.7, 28.7, 28.6, 28.1,
27.4, 27.0, 25.6. ESI-HRMS: Calcd for C21H39N3O4Na+ ([M +
Na]+), 420.2833; found, 420.2783.
N-(3-ter t-Bu tyl-2,5-dioxo-1,4-diaza-cycloicos-6-yl)m eth yl-
N-h yd r oxy-for m a m id e (1e). This compound was prepared
as a mixture of cis and trans isomers (formamide) in a manner
similar to 1c. Reversed-phase HPLC: 72% purity. 1H NMR
(250 MHz, CDCl3): δ 8.38 (brs, 0.24H), 7.83 (s, 0.7H), 8.16
(brs, 0.66H), 7.37 (s, 0.10H), 7.09 (d, J ) 8.7 H, 0.23H), 6.73
(brd, J ) 9.3 Hz, 0.67H), 6.14 (m, 0.23H), 6.04 (m, 0.67H), 5.93
(m, 0.1H), 4.25 (m, 1H), 4.11 (m, 0.3H), 3.81 (m, 0.7H), 3.64-
3.40 (m, 2H), 2.86 (m, 1.7H), 2.72 (m, 0.3H), 1.77-1.23 (m,
26H), 1.02 (s, 2.7H), 0.97 (s, 6.3H). 13C NMR (63 MHz,
CDCl3): δ 173.1, 170.3, 156.6, 61.1, 60.9, 51.9, 45.2, 39.6, 34.6,
30.5, 29.5, 28.9, 28.5, 27.9, 27.6, 27.5, 27.0, 26.6, 26.5, 26.0.
ESI-HRMS: Calcd for C24H45N3O4Na+ ([M + Na]+), 462.3302;
found, 462.3307.
N-(10-Un d ecen -1-yl)-NE-ter t-b u t oxyca r b on yl-L-lysin -
a m id e (10b). NR-Fmoc-Nꢀ-Boc-lysine was coupled with amine
8e as described for 7c to give N-(10-undecen-1-yl-NR-Fmoc-
1
Nꢀ-Boc-lysinamide in 82% yield. H NMR (250 MHz, CDCl3):
δ 7.77 (d, J ) 7.5 Hz, 2H), 7.59 (d, J ) 7.4 Hz, 2H), 7.43-7.26
(m, 4H), 6.02 (brs, 1H), 5.82 (m, 1H), 5.44 (brs, 1H), 4.95 (m,
2H), 4.57 (brs, 1H), 4.41 (brd, J ) 6.1 Hz, 2H), 4.20 (t, J ) 6.8
Hz, 1H), 4.07 (m, 1H), 3.23 (m, 2H), 3.11 (m, 2H), 2.05 (m,
2H), 1.84 (m, 1H), 1.69-1.18 (m, 21H), 1.43 (s, 9H). This amide
(1.86 g, 3.0 mmol) was treated with piperidine (3 mL) in
dichloromethane (20 mL) for ∼40 min with stirring. The
volatile solvents were removed under vacuum, and the residue
was purified by silica gel chromatography to afford 10b in 87%
yield. 1H NMR (250 MHz, CDCl3): δ 7.26 (m, 1H), 5.83 (m,
1H), 4.97 (m, 2H), 4.55 (m, 1H), 3.29 (m, 1H), 3.23 (m, 2H),
3.12 (m, 2H), 2.04 (m, 2H), 1.83 (m, 1H), 1.60-1.28 (m, 21H),
1.44 (s, 9H). 13C NMR (63 MHz, CDCl3): δ 175.1, 156.4, 139.6,
114.5, 55.5, 40.5, 39.4, 35.1, 34.2, 30.4, 30.0, 29.0, 29.8, 29.7,
29.5, 29.3, 28.8, 27.3, 23.3. ESI-HRMS: Calcd for C22H43N3O3-
Na+ ([M + Na]+), 420.3197; found, 420.3174.
Nr-{[(2R-N-Ben zyloxy-N-for m yla m in o)m eth yl]-6-h ep -
ten oyl}-N-(5-h exen -1-yl)-NE-ter t-bu toxyca r bon yl-L-lysin -
a m id e (11a ). 1H NMR (250 MHz, CDCl3): δ 8.16 (brs, 0.56H),
7.83 (brs, 0.44H), 7.38 (m, 5H), 6.69 (m, 1H), 6.60 (brs, 1H),
5.84-5.63 (m, 2H), 5.33 (brs, 0.44H), 5.02-4.82 (m, 5.56H),
4.36 (brs, 1H), 3.87 (brs, 0.56H), 3.60 (m, 1H), 3.27-3.02 (m,
4.44H), 2.60 (m, 1H), 2.04 (m, 4H), 1.80-1.17 (m, 14H), 1.43
(s, 9H). 13C NMR (63 MHz, CDCl3): δ 174.1, 173.9, 171.8,
156.7, 138.7, 138.3, 130.1, 129.6, 129.1, 115.5, 115.2, 53.3, 45.4,
40.5, 39.7, 39.0, 33.8, 33.6, 29.8, 29.2, 28.9, 26.6, 26.5, 22.9.
ESI-HRMS: Calcd for C33H52N4O6Na+ ([M + Na]+), 623.3779;
found, 623.3799.
N r-{[(2R -N -B e n zy lo x y -N -fo r m y la m in o )m e t h y l]-6-
h epten oyl}-N-(10-u n decen -1-yl)-NE-ter t-bu toxycar bon yl-L-
lysin a m id e (11b). This compound was synthesized from
amine 10b and acid 6 as described for 7c (85% yield). 1H NMR
(250 MHz, CDCl3): δ 8.18 (brs, 0.55H), 7.83 (brs, 0.45H), 7.38
(m, 5H), 6.66 (brs, 1H), 6.54 (brs, 1H), 5.85-5.70 (m, 2H), 5.38
(brs, 0.45H), 5.03-4.77 (m, 6.55H), 4.35 (brs, 1H), 3.72 (brs,
1.45H), 3.23 (m, 2H), 3.03 (m, 2.55H), 2.58 (brs, 1H), 2.07-
1.99 (m, 4H), 1.81-1.18 (m, 24H), 1.43 (s, 9H). 13C NMR (100
MHz, CDCl3): δ 171.7, 156.6, 139.5, 138.3, 130.1, 129.6, 129.1,
115.5, 114.5, 53.3, 40.5, 40.0, 34.2, 33.9, 29.9, 29.8, 29.6, 29.5,
29.3, 28.9, 27.3, 26.7, 22.9. ESI-HRMS: Calcd for C38H62N4O6-
Na+ ([M + Na]+), 693.4562; found, 693.4561.
N-{3-[4-(ter t-Bu t oxyca r b on yla m in o)b u t yl]-2,5-d ioxo-
1,4-d ia za -c y c lo ic o s -10-e n -6-y l}m e t h y l-N -b e n zy lo x y -
for m a m id e (12b). This compound was synthesized from
amine 11b as described for 9c (84% yield). 1H NMR (250 MHz,
CDCl3): δ 8.19 (brs, 0.53H), 7.85 (brs, 0.47H), 7.39 (m, 5H),
6.40 (d, J ) 8.0 Hz, 1H), 6.37 (brs, 1H), 5.32 (m, 2H), 5.01-
4.60 (m, 3H), 4.30 (m, 1H), 3.73-3.43 (m, 2.5H), 3.12-2.87
N-[3-(4-Am in obu tyl)-2,5-dioxo-1,4-diaza-cyclopen tadec-
6-yl]m eth yl-N-h yd r oxy-for m a m id e (1f). This compound
was prepared from 11a in a manner similar to 1g. Reversed-
phase HPLC: 96% purity. 1H NMR (400 MHz, CD3OD): δ 8.12
(brs, 0.17H), 7.69 (brs, 0.83H), 4.25 (m, 1H), 3.71-3.27 (m, 4H),
2.82 (m, 2.83H), 2.61 (m, 0.17H), 1.55-1.11 (m, 22H). 13C NMR
(100 MHz, CD3OD): δ 176.3, 174.3, 74.21, 64.7, 54.9, 46.2, 41.2,
40.1, 32.6, 30.8, 29.5, 29.1, 29.0, 28.8, 28.4, 28.3, 28.2, 26.8,
24.1. ESI-HRMS: Calcd for C19H36N4O4Na+ ([M + Na]+),
407.2629; found, 407.2642.
N-[3-(4-Aceta m id obu tyl)-2,5-d ioxo-1,4-d ia za -cyclop en -
ta d ec-6-yl]m eth yl-N-h yd r oxy-for m a m id e (1h ). Compound
12a was dissolved in TFA. After concentration in vacuo, the
resulting salt (9.0 mg, 13.7 µmol) was dissolved in dichloro-
methane (3 mL) containing triethylamine (2.1 mg, 20.6 µmol)
and acetic anhydride (1.8 mg, 17.8 µmol). The product was
purified by silica gel chromatography. Hydrogenation with
Pd/C afforded compound 1h . Reversed-phase HPLC: 77%
purity. 1H NMR (400 MHz, DMSO): δ 9.91 (s, 0.44H), 9.49 (s,
0.66H), 8.22 (s, 0.44H), 8.03 (m, 1.66H), 7.83-7.74 (m, 2H),

4948 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 20
Hu et al.
4.28 (m, 1H), 3.60-3.35 (m, 2H), 2.95 (m, 2H), 2.76-2.50 (m,
3H), 1.77 (s, 3H), 1.55-1.11 (m, 22H). ESI-HRMS: Calcd for
In h ibition of MMP s. The effect of PDF inhibitors on
human MMP-1, MMP-2, MMP-3, and MMP-9 was assessed
in reactions (400 µL total volume) containing buffer B, 15 µM
fluorogenic MMP peptide substrate, and 0, 1, or 10 µM PDF
inhibitor. For MMP-1, MMP-2, and MMP-3, peptide Mca-P-
K-G-Dpa-A-R-NH2 was employed as the substrate. For the
detection of MMP-3 activity, peptide Mca-R-P-K-P-V-E-Nval-
W-R-K(Dnp)-NH2 was added to the assay. The assay reactions
were initiated by the addition of 0.10 µg of the proper MMP
enzyme, and the remaining MMP activity was monitored
continuously on an Aminco-Bowman Series 2 luminescence
spectrometer (excitation wavelength at 320 nm and emission
at 405 nm).
C
21H38N4O5Na+ ([M + Na]+), 449.2721; found, 449.2734.
Nr-[(2R-N-For m yl-N-h ydr oxylam in o)m eth yl]h eptan oyl-
N-h exyl-L-lysin a m id e (3). Compound 11a was hydrogenated
1
on Pd/C to give Boc-protected 3 as described for 1c. H NMR
(400 MHz, CD3OD): δ 8.30 (s, 0.34H), 7.84 (s, 0.66H), 4.32
(m, 1H), 3.78 (m, 1H), 3.64 (dd, J ) 5.5, 14.1 Hz, 0.5H), 3.45
(dd, J ) 4.4, 14.1 Hz, 0.5H), 3.19 (t, J ) 7.0 Hz, 2H), 3.05 (t,
J ) 6.6 Hz, 2H), 2.93 (m, 0.66H), 2.75 (m, 0.34H), 1.74-1.33
(m, 22H), 1.46 (s, 9H), 0.92 (m, 6H). 13C NMR (100 MHz, CD3-
OD): δ 174.8, 172.9, 157.5, 78.8, 53.7, 53.6, 44.8, 44.5, 40.1,
39.4, 31.9, 31.7, 30.4, 30.2, 29.5, 29.4, 27.8, 26.9, 26.7, 23.1,
22.6, 22.5, 13.4, 13.4. This resulting compound was subjected
to TFA treatment to afford 3 as described for 1g. Reversed-
phase HPLC: 98% purity. 1H NMR (400 MHz, CD3OD): δ 8.19
(s, 0.19H), 7.73 (brs, 0.81H), 4.28 (m, 1H), 3.82-3.35 (m, 2H),
3.09 (m, 2H), 2.87 (m, 2.81H), 2.66 (m, 0.19), 1.69-1.21 (m,
22H), 0.82 (m, 6H). 13C NMR (100 MHz, CD3OD): δ 176.1,
174.2, 54.4, 46.1, 45.9, 41.2, 40.8, 33.3, 33.1, 33.0, 31.6, 30.8,
28.4, 28.2, 28.0, 24.2, 24.0, 23.9, 14.7. ESI-HRMS: Calcd for
An tim icr obia l Su scep tibility Testin g. The antimicrobial
susceptibility testing was performed using a microdilution
broth assay based on the guidelines of the National Committee
for Clinical Laboratory Standards (NCCLS) document M7-A5
(NCCLS, 2000). Cation-adjusted Mueller-Hinton broth (CAM-
HB) was used in the microtiter plates for dilution of the agent
stock solution and for diluting the bacterial inoculum. For
Streptococci, 5% lysed horse blood was added to the CAMHB.
Haemophilus Test Medium purchased from REMEL Inc. was
used for Haemophilus species. Inoculum was prepared by
taking a sample of bacteria from a 16-20 h plate culture and
adjusting it in sterile saline to an OD600 that corresponds to
∼1 × 108 colony-forming units per milliliter (CFU/mL). The
saline suspension was then diluted to ∼1 × 106 CFU/mL in
appropriate media to produce the standardized inoculum. A
sample of the inoculating culture is diluted in sterile saline
through a series of six 1:10 dilutions, and all dilutions are
plated by spotting 10 µL of each dilution, incubated overnight
at 35-37 °C, and counted the next day to verify the inoculum
size.
C
21H42N4O4Na+ ([M + Na]+), 437.3098; found, 437.3056.
P DF In h ibition Assa ys. Two different assay methods were
employed in this work. Method A couples the PDF reaction
with AAP.34 This assay was employed to determine the KI*
values. PDF inhibition assays were performed by incubating
PDF (8 nM) in 50 mM Hepes, pH 7.0, 150 mM NaCl, 100 µM
bovine serum albumin, and 0-300 nM inhibitor for 2 h on ice.
The solution was brought to room temperature in a water bath,
and the reaction was initiated by the addition of f-ML-pNA
substrate (80 µM). After 10 min at room temperature, the
reaction was quenched by heating at 95 °C for 15 min and
cooled to room temperature in a water bath. AAP (1.0 unit)
was added, and the resulting mixture was incubated at room
temperature for 15 min. The absorbance at 405 nm was
measured in a quartz cuvette on a UV/vis spectrophotometer.
The background absorbance was determined by repeating the
reaction but in the absence of PDF and subtracted from the
observed absorbance values. The inhibition constant, KI*, was
calculated from the initial rates using the Michaelis-Menten
equation: V ) (Vmax × [S])/{KM(1 + [I]/KI*) + [S]}. The
percentage of substrate to product conversion was kept at
<20%.
Method B couples the PDF reaction with a dipeptidyl
aminopeptidase, DPPI,36 and was employed to determine the
KI and k6 values. Assay reactions (total volume 200 µL) were
performed at room temperature in a quartz cuvette containing
5-150 µM f-Met-Lys-aminomethylcoumarin (f-Met-Lys-AMC)
as the substrate, buffer A, and 0.1 U of DPPI. Prior to use,
DPPI was activated by the treatment with 5 mM dithiothreitol
in 50 mM Hepes (pH 7.0) and 10 mM NaCl for 30 min. The
reactions were initiated by the addition of 1-10 µL (2-10 ng)
of PDF and monitored continuously at 360 nm in a UV/vis
spectrophotometer. The initial rates were obtained from the
early part of the reaction progression curves (<30 s). The
background signal was measured under the same conditions
but in the absence of PDF and was subtracted from the
observed PDF reaction rates. The PDF inhibition assays were
performed in a similar manner except that the reactions
contained 0-300 nM inhibitor and fixed concentrations of PDF
(4.0 nM), DPPI (0.1 U), and f-Met-Lys-AMC (130 µM). The
inhibition constant, KI, was calculated from the initial rates
using the above Michaelis-Menten equation.
Microtiter plates (Evergreen 96 well microplate number 222-
8032-01R) were prepared with the use of the Beckman
BiomekR 1000 automated laboratory workstation in combina-
tion with manual techniques. The microtiter plate was filled
with 50 µL of diluent broth using the BiomekR 1000 instru-
ment. The agent or control antibiotic stock solution was
manually added to the drug well of the microtiter plate using
a Calibra 852 Pipettor. The agent/antibiotic was serially
diluted in 2-fold dilutions using the BiomekR 1000 instrument.
Fifty microliters of the standardized bacterial inoculum was
added to each well using the BiomekR 1000 instrument,
producing a final inoculum of ∼5 × 105 CFU/mL. In addition,
a nontreated growth control and commercially available control
antibiotics were included to validate the assay. The final
concentrations ranged from 0.03 to 32 µg/mL for test agents
and 0.06 to 64 µg/mL for control antibiotics. For controls, all
stock and working concentrations were 1 mg/mL. The micro-
titer plates were incubated overnight (16-20 h) in a 35-37
°C incubator with the exception of the Haemophilus and
Streptococcus species, which were incubated at 35-37 °C and
in the presence of 5% CO2 for 24 h. The MIC90 is defined as
the lowest concentration of agent that prevented visible growth
of the organism. The end point was determined visually using
a plate reading mirror.
In Vitr o Sta bility of P DF In h ibitor s in Ra t P la sm a .
Compounds were diluted 1000-fold into rat plasma (Lot 27624,
Pel-Freez, Rogers, AK) to result in an incubation concentration
of 5 µg/mL. All samples were incubated at 37 °C for 0-5 h.
Aliquots were withdrawn at various time points, the proteins
were precipitated with acetonitrile, and internal standard was
added (10 µg/mL). After the proteins were removed by cen-
trifugation, the supernatant was analyzed for the amount of
remaining parent compound by LC-MS/MS. LC was conducted
on an Agilent 1100 Binary HPLC equipped with a C18 column.
The chromatography parameters were optimized so that the
analyte (1f and 3) coeluted with the internal standard. MS
was performed on a ThermoFinnigan LCQ-DUO instrument.
MS parameters were optimized for maximal ion signal with
the AutoTune routine of the Xcalibur software (ThermoFinni-
gan). The samples were prepared in duplicate.
To determine the rate constant, k6, PDF (40-160 nM) and
inhibitor (80-200 nM) were preincubated in 100 µL of buffer
A and 100 µM BSA for 2 h on ice. The mixture was rapidly
diluted into 900 µL of buffer A containing 270 µM f-Met-Lys-
AMC and 0.1 U DPPI, and the reaction was monitored at 360
nm on
a UV/vis spectrophotometer. After correction for
background hydrolysis by DPPI, the progress curve was fitted
t
to the equation: Abs360 ) Vs[t - (1 - e-k )/k6], where Vs is
6
the final steady state velocity. The rate constant k5 was
calculated from KI, KI*, and k6 values by using the equation:
KI* ) KIk6/(k5 + k6).

Macrocyclic Inhibitors for Peptide Deformylase
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 20 4949
(19) Roblin, P. M.; Hammerschlag, M. R. In vitro activity of a new
antibiotic, NVP-PDF386 (VRC4887), against Chlamydia pneu-
moniae. Antimicrob. Agents Chemother. 2003, 47, 1447-1448.
(20) Coats, R. A.; Lee, S.-L.; Davis, K. A.; Patel, K. M.; Rhoads, E.
K.; Howard, M. H. Stereochemical definition and chirospecific
synthesis of the peptide deformylase inhibitor Sch 382583. J .
Org. Chem. 2004, 69, 1734-1737.
(21) Clements, J . M.; Beckett, R. P.; Brown, A.; Catlin, G.; Lobell,
M.; Palan, S.; Thomas, W.; Wittaker, M.; Wood, S.; Salama, S.;
Baker, P. J .; Rodgers, H. F.; Barynin, V.; Rice, D. W.; Hunter,
M. G. Antibiotic activity and characterization of BB-3497, a novel
peptide deformylase inhibitor. Antimicrob. Agents Chemother.
2001, 45, 563-570.
Ack n ow led gm en t. We thank Professors Wim G. J .
Hol and Christophe L. M. J . Verlinde (University of
Washington, Seattle) for their initial computer modeling
studies that suggested the feasibility of designing cyclic
PDF inhibitors and Susan Hatcher (Campus Chemical
Instrument Center, Ohio State University) for perform-
ing all of the mass spectrometric analysis. This work
was supported by a grant from the National Institutes
of Health (AI40575).
(22) Gross, M.; Clements, J .; Beckett, R. P.; Thomas, W.; Taylor, S.;
Lofland, D.; Ramanathan-Girish, S.; Garcia, M.; Difuntorum, S.;
Hoch, U.; Chen, H.; J ohnson, K. W. Oral anti-pneumococcal
activity and pharmacokinetic profiling of a novel peptide de-
formylase inhibitor. J . Antimicrob. Chemother. 2004, 53, 487-
493.
Su p p or tin g In for m a tion Ava ila ble: Additional experi-
mental details and spectral data of synthetic intermediates.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(23) Xue, C.-B.; He, X.; Roderick, J .; et al. Design and synthesis of
cyclic inhibitors of matrix metalloproteases and TNF-R produc-
tion. J . Med. Chem. 1998, 41, 1745-1748.
Refer en ces
(1) Giglione, C.; Pierre, M.; Meinnel, T. Peptide deformylase as a
target for new generation, broad spectrum antimicrobial agents.
Mol. Microbiol. 2000, 36, 1197-1205.
(24) Wei, C.-Q.; Gao, Y.; Lee, K.; et al. Macrocyclization in the design
of Grb2 SH2 domain-binding ligands exhibiting high potency in
whole-cell systems. J . Med. Chem. 2003, 46, 244-254.
(25) Hao, B.; Gong, W.; Rajagopalan, P. T. R.; Zhou, Y.; Pei, D.; Chan,
M. K. Structural basis for the design of antibiotics targeting
peptide deformylase. Biochemistry 1999, 38, 4712-4719.
(26) Guilloteau, J .-P.; Mathieu, M.; Giglione, C.; Blanc, V.; Dupuy,
A.; Chevrier, M.; Gil, P.; Famechon, A.; Meinnel, T.; Mikol, V.
The crystal structures of four peptide deformylases bound to the
antibiotic actinonin reveal two distinct types: A platform for the
structure-based design of antibacterial agents. J . Mol. Biol. 2002,
320, 951-962.
(2) Pei, D. Peptide deformylase: A target for novel antibiotics?
Emerging Ther. Targets 2001, 5, 23-40.
(3) Yuan, Z.; Trias, J .; White, R. J . Deformylase as
a novel
antibacterial target. Drug Discovery Today 2001, 6, 954-961.
(4) Clements, J . M.; Ayscough, A. P.; Keavey, K.; East, S. P. Peptide
deformylase inhibitors, potential for
spectrum antibacterials. Curr. Med. Chem. Anti-Infective Agents
2002, 1, 239-249.
a new class of broad
(5) Meinnel, T.; Mechulam, Y.; Blanquet, S. Methionine as trans-
lational start signal: A review of the enzymes of the pathway
in Escherichia coli. Biochimie 1993, 75, 1061-1075.
(6) Mazel, D.; Pochet, S.; Marliere, P. Genetic characterization of
polypeptide deformylase, a distinctive enzyme of eubacterial
translation. EMBO J . 1994, 13, 914-923.
(27) Hu, X.; Nguyen, K. T.; Verlinde, C. L. M. J .; Hol, W. G. J .; Pei,
D. Structure-based design of a macrocyclic inhibitor for peptide
deformylase. J . Med. Chem. 2003, 46, 3771-3774.
(28) Hu, Y. J .; Wei, Y.; Zhou, Y.; Rajagopalan, P. T. R.; Pei, D.
Determination of substrate specificity for peptide deformylase
through the screening of a combinatorial peptide library. Bio-
chemistry 1999, 38, 643-650.
(29) Ragusa, S.; Mouchet, P.; Lazennec, C.; Dive, V.; Meinnel, M.
Substrate recognition and selectivity of peptide deformylase.
Similarities and differences with metzincins and thermolysin.
J . Mol. Biol. 1999, 289, 1445-1457.
(7) Meinnel, T.; Blanquet, S. Characterization of the Thermus
thermophilus locus encoding peptide deformylase and methionyl-
tRNA(fMet) formyltransferase. J . Bacteriol. 1994, 176, 7387-
7390.
(8) Margolis, P. S.; Hackbarth, C. J .; Young, D. C.; Wang, W.; Chen,
D.; Yuan, Z.; White, R.; Trias, J . Peptide deformylase in S.
aureus: Resistance to inhibition is mediated by mutations in
the formyltransferase gene. Antimicrob. Agents Chemother.
2001, 44, 1825-1831.
(9) Nguyen, K. T.; Hu, X.; Colton, C.; Chakratarti, R.; Zhu, M. X.;
Pei, D. Characterization of a human peptide deformylase:
Implications for antibacterial drug design. Biochemistry 2003,
42, 9952-9958.
(10) Rajagopalan, P. T. R.; Yu, X. C.; Pei, D. Peptide deformylase: A
new type of mononuclear iron protein. J . Am. Chem. Soc. 1997,
119, 12418-12419.
(11) Groche, D.; Becker, A.; Schlichting, I.; Kabasch, W.; Schultz, S.;
Wagner, A. F. V. Isolation and crystallization of functionally
competent Escherichia coli peptide deformylase forms containing
either iron or nickel in the active site. Biochem. Biophys. Res.
Commun. 1998, 246, 342-346.
(12) Meinnel, T.; Patiny, L.; Ragusa, S.; Blanquet, S. Design and
synthesis of substrate analogue inhibitors of peptide deformy-
lase. Biochemistry 1999, 38, 4287-4295.
(13) Huntington, K. M.; Yi, T.; Wei, Y.; Pei, D. Synthesis and
antibacterial activity of peptide deformylase inhibitors. Bio-
chemistry 2000, 39, 4543-4551.
(14) Wei, Y.; Yi, T.; Huntington, K. M.; Chaudhury, C.; Pei, D.
Identification of a potent peptide deformylase inhibitor from a
rationally designed combinatorial library. J . Comb. Chem. 2000,
2, 650-657.
(15) Chen, D. Z.; Patel, D. V.; Hackbarth, C. J .; Wang, W.; Dreyer,
G.; Young, D.; Margolis, P. S.; Wu, C.; Ni, Z.-J .; Trias, J .; White,
R.; Yuan, Z. Actinonin, a naturally occurring antibacterial agent,
is a potent deformylase inhibitor. Biochemistry 2000, 39, 1256-
1262.
(16) Apfel, C.; Banner, D. W.; Bur, D.; Dietz, M.; Hirata, T.;
Hubschwerlen, C.; Locher, H.; Page, M. G. P.; Pirson, W.; Rosse,
G.; Specklin, J .-L. Hydroxamic acid derivatives as potent peptide
deformylase inhibitors and antibacterial agents. J . Med. Chem.
2000, 43, 2324-2331.
(30) Miller, S. J .; Blackwell, H. E.; Grubbs, R. H. Application of
ringclosing metathesis to the synthesis of rigidified amino acids
and peptides. J . Am. Chem. Soc. 1996, 118, 9606-9614.
(31) Furstner, A. Olefin metathesis and beyond. Angew. Chem., Int.
Ed. 2000, 39, 3012-3043.
(32) Rajagopalan, P. T. R.; Datta, A.; Pei, D. Purification, character-
ization, and inhibition of peptide deformylase from Escherichia
coli. Biochemistry 1997, 36, 13910-13918.
(33) Lazennec, C.; Meinnel, T. Formate dehydrogenase-coupled spec-
trophotometric assay of peptide deformylase. Anal. Biochem.
1997, 244, 180-182.
(34) Wei, Y.; Pei, D. Continuous spectrophotometric assay of peptide
deformylase. Anal. Biochem. 1997, 250, 29-34.
(35) Morrison, J . F.; Walsh, C. T. The behavior and significance of
slow-binding enzyme inhibitors. Adv. Enzymol. 1988, 61, 201-
301.
(36) Nguyen, K. T.; Hu, X.; Pei, D. Slow-binding inhibition of peptide
deformylase by cyclic peptidomimetics as revealed by a new
spectrophotometric assay. Bioorg. Chem. 2004, 32, 178-191.
(37) Wise, R.; Andrews, J . M.; Ashby, J . In vitro activities of peptide
deformylase inhibitors against Gram-positive pathogens. Anti-
microb. Agents Chemother. 2002, 46, 1117-1118.
(38) J ones, R. N.; Rhomberg, P. R. Comparative spectrum and activity
of NVP-PDF386 (VRC4887), a new peptide deformylase inhibi-
tor. J . Antimicrob. Chemother. 2003, 51, 157-161.
(39) Rajagopalan, P. T. R.; Grimme, S.; Pei, D. Characterization of
cobalt(II)-substituted peptide deformylase: Function of the metal
ion and the catalytic residue Glu-133. Biochemistry 2000, 39,
779-790.
(40) McMartin, C.; Bohacek, R. J . QXP: Powerful, rapid computer
algorithms for structure-based drug design. J . Comput.-Aided
Mol. Des. 1997, 11, 333-344.
(41) Osowska-Pacewicka, K.; Zwierzak, A. Two-carbon homologation
of Grignard reagents to primary amines. Synthesis 1996, 3, 333-
335.
(17) J ayasekera, M. M. K.; Kendall, A.; Shammas, R.; Dermyer, M.;
Tomala, M.; Shapiro, M. A.; Holler, T. P. Novel nonpeptidic
inhibitors of peptide deformylase. Arch. Biochem. Biophys. 2000,
381, 313-316.
(18) Thorarensen, A.; Douglas, M. R., J r.; Rohrer, D. C.; et al.
Identification of novel potent hydroxamic acid inhibitors of
peptidyl deformylase and the importance of the hydroxamic acid
functionality on inhibition. Bioorg. Med. Chem. Lett. 2001, 11,
1355-1358.
(42) Yebassa, D.; Balakrishnan, S.; Feresenbet, E.; Raghavan, D.;
Start, P. R.; Hudson, S. D. Chemically functionalized clay vinyl
ester nanocomposites: effect of processing parameters. J . Polym.
Sci., Part A: Polym. Chem. 2004, 42, 1310-1321.
(43) Fields, L. B.; J acobsen, E. N. Synthesis and characterization of
chiral bimetallic complexes bearing hard and soft Lewis acidic
sites. Tetrahedron: Asymmetry 1993, 4, 2229-2240.
J M049592C