Discovery and refinement of a new structural class of potent peptide deformylase inhibitors

Article abstract of DOI:10.1021/jm060910c

New classes of antibiotics are urgently needed to counter increasing levels of pathogen resistance. Peptide deformylase (PDF) was originally selected as a specific bacterial target, but a human homologue, the inhibition of which causes cell death, was rec

Full text of DOI:10.1021/jm060910c

10
J. Med. Chem. 2007, 50, 10-20
Articles
Discovery and Refinement of a New Structural Class of Potent Peptide Deformylase Inhibitors
Adrien Boularot,†,‡ Carmela Giglione,† Sylvain Petit,†,‡ Yann Duroc,† Rodolphe Alves de Sousa,‡ Vale´ry Larue,§
Thierry Cresteil,# Fre´de´ric Dardel,§ Isabelle Artaud,‡ and Thierry Meinnel*,†
UPR2355 and UPR2301, Centre National de la Recherche Scientifique, Baˆtiment 23, 1 AVenue de la Terrasse, F-91198 Gif-Sur-YVette Cedex,
France, and UMR8601 and UMR8015, CNRSsUniVersite´ Paris V, 45 Rue des Saint-Pe`res and 4 AVenue de l’ObserVatoire,
75006 Paris, France
ReceiVed July 29, 2006
New classes of antibiotics are urgently needed to counter increasing levels of pathogen resistance. Peptide
deformylase (PDF) was originally selected as a specific bacterial target, but a human homologue, the inhibition
of which causes cell death, was recently discovered. We developed a dual-screening strategy for selecting
highly effective compounds with low inhibition effect against human PDF. We selected a new scaffold in
vitro that discriminated between human and bacterial PDFs. Analyses of structure-activity relationships
identified potent antibiotics such as 2-(5-bromo-1H-indol-3-yl)-N-hydroxyacetamide (6b) with the same
mode of action in vivo as previously identified PDF inhibitors but without the apoptotic effects of these
inhibitors in human cells.
Introduction
in the same essential pathway as in bacteria, (iii) to be inhibited
in vitro by the natural compound actinonin, used to design the
compounds currently in clinical trials, and (iv) to be inhibited
in vivo by actinonin in some cell lines.6-8,11-13 Finally, when
used at a range of concentrations similar to the minimal
inhibitory concentrations (MIC) of the best compounds, acti-
nonin and its derivatives had an antiproliferative effect triggered
by mitochondrial dysfunction, including ATP depletion, mem-
brane depolarization, and apoptosis induction, leading to cell
death.6,12,14,15
Two bacterial PDF types, PDF1B and PDF2, have been
distinguished.9,16 PDF2 is found only in Gram-positive bacteria.
Bacteria may have one or several functional genes encoding
different types of PDF.9,16 For example, Gram-negative Es-
cherichia coli has one PDF1B gene (def) only, whereas Gram-
positive bacteria such as Bacillus spp. have two PDF genes:
def, encoding PDF1B, and ykrB, encoding PDF2. The human
PDF (mPDF) has been classified as a PDF1A.11
Over the past decade, there has been an alarming increase in
the prevalence of multiresistant phenotypes among pathogenic
bacteria. There is therefore an urgent need to identify new
antibiotics to combat infectious diseases. This challenge can
be met by whole-cell screening or by a combination of genomics
and mechanism-based drug design in cell-free assays.1,2 The
first step in this second approach is the search for vital genes
unique to bacteria and conserved in various bacterial genomes.
The starting criteria for suitability are essentiality and selectivity
(the so-called “druggability”) of the potential antibiotic target.
Inhibitors of the most suitable targets are then identified by high-
throughput screening in vitro, and the most promising molecules
are improved by medicinal chemistry. Whole-cell studies must
then be carried out on the most promising compounds, selected
on the basis of optimal antibiotic capacities. Constant monitoring
of the mode of action of the compound series is critical during
this process.
However, a decade after its introduction, the “cell-free”
approach may be considered disappointing, as it has brought to
the pipeline only a couple of drugs all derived from natural
products, including actinonin inhibiting peptide deformylase
(PDFa).3-5 The recent demonstration of the existence of a PDF
homologue (mitochondrial PDF or mPDF) in humans6-8 raised
major objections to the use of this otherwise very promising
target.9,10 Indeed, mPDF proved (i) to be functional, displaying
PDF activity in the human mitochondrion, (ii) to be involved
The aim of our study was to identify new compounds that
would selectively inhibit both types of bacterial PDFs (PDF1B
and PDF2) without significantly inhibiting PDF1A.
Results and Discussion
Design and Medium-Throughput Screening by NMR
Identify Indole Derivatives as Low-Affinity Ligands of PDF.
The binding potency of peptide deformylase inhibitors (PDFI)
depends essentially on the additive effects of two chemical
groups: (i) a metal-binding group and (ii) the P1′ group, which
binds the S1′ pocket of PDF.4 The increase in entropy due to
the binding of either of these low-affinity binding groups results
in the molecule being a potent PDFI, with inhibition constants
in the nanomolar range. The S1′ pocket of bacterial PDFs of
both types, PDF1B and PDF2, accepts n-butyl, n-pentyl, n-hexyl,
n-phenyl,17 and other cyclic side chains4 with low levels of
selectivity. mPDF has a modified S1′ pocket that cannot tolerate
bulky chains such as phenyl chains.8,18,19 We therefore looked
for bulky, cyclic compounds that would bind to bacterial PDF
* To whom correspondence should be addressed. Phone: +33 1 69 82
36 12. Fax: +33 1 69 82 36 07. E-mail: Thierry.Meinnel@isv.cnrs-gif.fr.
† UPR2355, Centre National de la Recherche Scientifique.
‡ UMR8601, CNRSsUniversite´ Paris V.
§ UMR8015, CNRSsUniversite´ Paris V.
# UPR2301, Centre National de la Recherche Scientifique.
a Abbreviations: HSQC, heteronuclear single quantum correlation; IC50
,
inhibition constant; ID, indole derivatives; KI, inhibition or dissociation
constant; MIC, minimum inhibitory concentration; PDF, peptide deformy-
lase; PDFI, peptide deformylase inhibitors; SAR, structure-activity relation-
ship; STD, saturation transfer difference; 3D, three-dimensional.
10.1021/jm060910c CCC: $37.00 © 2007 American Chemical Society
Published on Web 12/15/2006

Inhibitors of Bacterial Peptide Deformylase
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 1 11
Figure 1. Mode by which 6b binds to the active site pocket of PDF1B, as probed by NMR. (A) HSQC footprint of 6b (top inset) on 15N-labeled
E. coli PDF. The reference spectrum is shown in black contours, and the spectrum in the presence of a stoichiometric amount of the compound is
overlaid in red. Significantly perturbed amide groups are labeled according to NMR chemical shift assignments.34 Similar footprints were obtained
with several ID, including 5Br-indole. (B) 3D structure of E. coli PDF. The peptide chain is shown in blue. Amino acids in which the amide group
NMR shift was perturbed by the binding of 6b are shown in red (residues labeled in panel A). The catalytic metal cation is shown as a blue sphere.
(C) NMR saturation transfer difference experiment (STD) showing the transfer of magnetization from E. coli PDF to 6b. Upper trace shows the
reference spectrum of the mixture. Peaks labeled with asterisks correspond to buffer signals. The numbered peaks correspond to signals originating
from 6b, as labeled in panel A. Lower trace shows STD signals. The arrow indicates the spectral region irradiated. Normalized STD intensities are
indicated.
Scheme 1. ID Structure and Labeling: Chemical Structures of 6b, Actinonin, and 18 and Their Main PDF Inhibition Properties
but not to mPDF. A medium-throughput screening system was
used to identify such chemical groups by NMR. We used a
pipetting robot connected to a flow-injection NMR probe and
sequentially mixed peptide and non-peptide compounds with
15N-labeled E. coli PDF (PDF1B), the resulting mixture being
injected into the NMR spectrometer. Chemical shift perturba-
tions were monitored by recording {1H-15N} heteronuclear
single quantum correlation (HSQC) spectra, which were com-
pared with control data obtained with the free enzyme, as
previously described.20,21 We selected compounds that caused
resonance broadening or a shift in the HSQC protein spectrum.
The amide groups concerned were identified and located on
the three-dimensional (3D) structure of PDF1B (Figure 1A).
Because compounds such as 5-bromoindole induced modifica-
tions of the {1H-15N} 2D NMR map in a region corresponding
to residues located in the S1′ pocket of the enzyme, the indole
scaffold was selected from this screening. The effect of one of
the indole derivatives (ID; see Scheme 1 for labeling) is shown
in Figure 1B. Similar strong perturbations of the HSQC
spectrum were also observed with ID and PDF2. Reciprocal
ligand-observed saturation transfer difference (STD) experi-
ments22 were carried out to obtain information about the mode
of binding of these derivatives to both PDF1B (Figure 1C) and
PDF2 (data not shown). The strongest STD effect was observed
at R4 of the indole moiety, and the weakest effect was that for
the methylene group attached at position R3. These results
confirm that the indole group bound to both PDF1B and PDF2
and imply that position R4 is buried the most deeply in the S’1
pocket.
Structure-Activity Relationship (SAR) Analysis of ID and
3D Modeling. Indole and 5-bromoindole had binding constants
in the millimolar range, as shown by NMR titrations and/or
PDF inhibition assays (Table 1). We tried to increase the potency
of ID by introducing substituents at positions R1-6. The synthesis

12 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 1
Boularot et al.
Table 1. Structures and Associated IC50 Values with the Various Types of PDFsb
a AtPDF1A corresponds to mitochondrial PDF1A from Arabidopsis thaliana, a mitochondrial PDF whose specific activity of the purified form is higher
than that of mPDF and allows better assessment and inhibition properties of a mitochondrial PDF than mPDF.8 b ND, not determined.
of 22 different ID is depicted in Schemes 2 and 3. Because
phenyl side chains at P1′ are known to be poorly accepted by
mPDFs8,19 but strongly accepted by PDF1Bs,17,21 we prepared
the hydroxamic acid, hydroxylamine, and reverse hydroxamic
acid of N-Boc-phenylalanine (compounds 18, 20, and 21,
respectively, Scheme 4). We first assessed the ability of all
compounds to inhibit both PDF1B and PDF2. The hydroxamic
and reverse hydroxamic acid were the most efficient functional
groups introduced at R3, in terms of binding (Table 1). SAR
analysis was carried out with the hydroxamic acid derivatives
by introducing various substituents at R4-6. Because 5-Br indole
derivative 6b was the most potent inhibitor (Scheme 1), we also
assessed the introduction of a methyl group at R2 (6j) or of a
benzyloxycarbonyl group at R1 (6d). Whereas 6j was one of
the least potent inhibitors in this series, 6d was as active as the
most potent compound 6b, suggesting that the addition of bulky
substitutions at R1 has no effect on the affinity of ID. As
expected from the characteristics of the S1′ pocket, none of the
ID tested significantly inhibited mPDFs (Scheme 1 and Table
1). Finally, compounds 18 and 21 also efficiently inhibited
PDF1B with only mild effects on mPDFs. However, these
compounds, unlike ID, inhibited PDF2 only poorly (Scheme 1
and Table 1). They would therefore not be expected to display
antibiotic activity against Gram-positive bacteria in vivo.
A 3D model of 6b docked in the active site of PDF1B
indicated a perfect fit within the S1′ pocket. The 5-Br group
corresponded to the C4 methyl group of the Met side chain
(Figure S1 in Supporting Information), four-carbon chains being
optimal for acceptance in the S1′ pocket.17 In the model, position
R4 is locked deep inside the S1′ pocket, buried under the side

Inhibitors of Bacterial Peptide Deformylase
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 1 13
Scheme 2. Synthesis of Indole Derivatives: Hydroxyketones, Ketothioacetyl Derivatives, and Hydroxamic Acidsa
a (a) (1) HOBt, EDCI, NMM/DMF; (2) NH2OH‚HCl/DMF; (b) (1) (COCl)2, DMF/THF; (2) C11H28O3Si3/dioxane; (3) HCl(g), dioxane, 80 °C; (c) (1)
SOCl2/CH2Cl2, 40 °C; (2) CH2N2, diethyl ether/CH2Cl2, 0 °C; (d) HCl(g), diethyl ether, 0 °C; (e) KSAc/DMF; (f) (1) n-BuLi, ZnCl2/THF, 0 °C; (2)
BrCH2COOEt/toluene; (g) NH2OH, ethanol, 80 °C.
Scheme 3. Synthesis of Hydroxylamines and Reverse Hydroxamic Acid Indole Derivativesa
a (a) (1) POCl3, DMF; (2) NaOH; (b) NH2OH, HCl/pyridine; (c) NaBH3CN, HCl; (d) (1) NaH/THF; (2) PhSO2Cl; (e) NaBH4/ethanol; (f) (1) PPh3, NBS;
(2) BocNH2OBoc, NaH /DMF; (g) (1) CF3COOH/CH2Cl2; (h) HCOOH, Ac2O/CH2Cl2.
chain of His132, consistent with the STD data. We cocrystallized
6b with PDF1B. The difference between the electron density
maps of the free protein and of the complex confirmed that the
inhibitor bound to the active site according to the model
proposed in Figure S1. These data suggest that the antibiotic
activity of ID may target bacterial PDFs in vivo without
inhibiting animal PDFs and causing apoptosis.
Like Actinonin, 6b Shows Slow, Tight Binding to PDF.
The most potent PDFI show slow, tight binding to PDF.23,24
Detailed kinetic analysis is required to characterize this pro-
cess.25 We studied and compared the binding mechanisms of
6b and actinonin. When the reaction was triggered by adding
the enzyme (Figure S2A,B of Supporting Information), the initial
rate of reaction decreased rapidly, reaching a minimum within
3 min, for both compounds. When the reaction was initiated
by adding the substrate after incubating the enzyme with the
inhibitor (Figure S2C,D), the kinetics of the reaction were linear
over a long period of time (10 min) and the initial reaction rate
corresponded to that reached after 3 min in the second phase
of the experiment performed without prior incubation of enzyme
and inhibitor. Thus, both actinonin and 6b induced similar slow
binding to PDF via a two-step mechanism (Figure S3A). The
KI value for the inhibitor/enzyme pre-equilibrium was ∼100
nM for both compounds, similar to previously reported values.23

14 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 1
Boularot et al.
Figure 2. 6b specifically targets both PDF types in vivo. The xylose-inducible Y103 and D103 strains of B. subtilis were cultured in the presence
of various concentrations of L-xylose (Xyl) to increase intracellular PDF concentration.27 Constitutive strains (Y101 and D101) were cultured under
the same conditions for comparison. MIC values were determined for 6b (black squares, 103 series; white squares, 101 series) and compared with
those obtained for actinonin (black circles, 103 series; white circles, 101 series). (A) PDF2-deficient context. The pdf1b gene is either constitutively
expressed (strain Y101) or inducible with L-xylose (strain D103). (B) PDF1B-deficient context. The pdf2 gene is either constitutively expressed
(strain D101) or inducible with L-xylose (strain Y103).
Scheme 4. Synthesis of Pseudopeptide Derivativesa
a (a) (1) HOBt, EDCI, NMM/DMF; (2) NH2OH‚HCl/DMF; (b) (1)
NH2OH, HCl/pyridine; (2) NaBH3CN, CH3COOH/methanol; (c) (1)
PhCH2NHOH‚HCl/pyridine; (2) NaBH3CN, CH3COOH/methanol; (3)
HCOOH, acetic anhydride; (4) H2, Pd/C.
Table 2. Comparison of the Main Kinetic and Thermodynamic
Figure 3. 6b does not cause significant toxic effects by inducing
Parameters Describing PDF2 Inhibition by 6b and Actinonin
apoptosis in human cells. Apoptosis in KB cells was assessed by
measuring caspase induction. The inset shows the results of toxicity
parameter
actinonina
6ba
assays in the presence of 10 µM drug with various cell types.
KI (nM)
KI*app (nM)
KI* (nM)
185 ( 15b
91 ( 9c
10 ( 1d
15 ( 2e
2.9 ( 0.8f
64 ( 7
0.075 ( 0.008
0.0011 ( 0.0003
10
5.6 ( 1.9f
16 ( 3
0.074 ( 0.008
0.0049 ( 0.0009
2.4
KI/KI*
The KI* value (Figure S3B,C) corresponded to the PDF
concentration used in the assay, indicating tight binding. Because
we could not use lower enzyme concentrations, this KI* value
should be considered as a KI*app (Table 2). The constants for
binding to PDF2 of both actinonin and 6b were therefore
determined using Henderson’s method26 (Figure S3D). The two
compounds had similar KI*, in the nanomolar range (Table 2).
However, compound 18, the pseudopeptide analogue, displayed
neither slow nor tight binding to PDF2. This accounts for the
low IC50 of compound 18 for PDF2 (Scheme 1). This suggests
that 6b, unlike 18, is able to induce a change in the conformation
of the protein. This property might be displayed by any potent
PDFI. Because compound 6d also displayed slow and tight
binding, the ID series was considered a promising source of
potent and specific PDFI.
g
k5 (s-1
k6 (s-1
)
)
h
t
1/2 (min)h
a The PDF2 concentration used in the assay was 25 nM. The analysis in
Figure S2D indicates that the active site enzyme concentration of our
preparation was 11 ( 2 nM, i.e., 40-50% the initial enzyme concentration.
b Derived from the curve shown in Figure S2B. c Derived from the curve
shown in Figure S2C. d Derived from the curve shown in Figure S2B.
e Derived from the curve shown in Figure S2C. f Derived from the curve
shown in Figure S2D. g Derived from the curve shown in Figure S2E. h k6
) k5/(KI/KI* - 1) and t1/2 ) log 2/k5.
(D101 and Y101), each with only one type of PDF (Table 3).
We found that the MIC values of actinonin, 6b, and 6d followed
similar trends, being lower in the ykrB (PDF2) gene-inactivated
context. These results confirmed that def was less strongly
expressed than ykrB27 and suggested that 6b and 6d targeted
both PDF1B and PDF2. The MIC of 6b was half that of
actinonin in E. coli (Table 3). In E. coli, actinonin is detoxified
by the AcrAB-tolC efflux pump. We used a tolC mutant to
ID Have Antibiotic Activity. We determined the antibiotic
properties of certain ID. In Bacillus subtilis, the MIC of 6b
was similar to that of actinonin, in the µg/mL range, whereas
that of 6d was slightly higher (Table 3). We assessed the effects
of 6b, 6d, and actinonin on two B. subtilis strain derivatives

Inhibitors of Bacterial Peptide Deformylase
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 1 15
Conclusions
Table 3. In Vivo Antibiotic Activity of the ID Seriesc
MIC (µg/mL)
We report here the first identification of a PDFI based on
the medium-throughput NMR screening of different scaffolds,
followed by SAR analysis, kinetic characterization, and mode-
of-action determination and demonstration of antibiotic activity
for the best compounds. The rationale for this screening was
based on the structural analysis of mPDFs.18 This made it
possible to select PDFIs that inhibited both types of bacterial
PDF, without significantly inhibiting mPDF. The mLogP value
associated with 6b (1.84) makes this molecule an excellent
candidate drug. Further modification of ID, which currently have
few substituents, could be used to improve their permeability
and pharmacological properties, creating a new generation of
antibiotics. In addition, some toxicity of 6b was observed in a
number of cell lines, although to a lesser extent than actinonin
(Figure 3, inset). Because this effect is necessarily unrelated to
mPDF inhibition, identification of the target(s) involved would
be of strong interest to lower toxicity of future ID. Next
developments will therefore include the determination of the
inhibition capacities by ID of various human metallopeptidases
such as matrix metalloproteases (MMPs) or angiotensin I
converting enzyme (ACE), which are known to be slightly
inhibited by PDFI such as BB-3497 and various hydroxamate
derivatives of the actinonin series, including actinonin itself.30,31
bacterial strain
actinonin
6b
6d 6a
2
18
21
E. coli
JM101Tr 32-64
K37
4-8
120 114 >300 >300 >300
32-64
16-32 ND ND ND
ND
ND
120
CAG1284 0.125-0.25 8-16 25 95 >300 32
B. subtilis 168a
D101b
Y101b
1-2
1-2
2-4
2-4
16 160 >300 >300 >300
12 150 >300 >300 >300
0.125-0.25 1-2
6
145 >300
8
9
a Identical results were obtained with MO3482. b Both strains were
derived from 168.27 D101 is deficient in PDF1B (def gene), the minor PDF
in B. subtilis,27 and Y101 is deficient in PDF2 (ykrB gene). c ND, not
determined.
determine whether such efflux was involved in 6b detoxification.
The MIC of 6b was only slightly lower in this mutant, whereas
that of actinonin was lower than that in the wild type by a factor
of 300. Thus, this efflux system detoxified ID less efficiently
than actinonin. Compounds 2, 6a, 18, and 21 had no antibiotic
activity.
ID Specifically Target PDF in Vivo. We next assessed the
specificity of PDF targeting by 6b in vivo, using two B. subilis
mutants in which (i) both the ykrB and def genes had been
inactivated and (ii) a supplementary copy of one of the two
PDF genes was placed under the control of the XylR promoter.
Under these conditions, the synthesis of PDF1B or PDF2 from
the gene concerned depends on L-xylose.27 The MIC values of
both actinonin and 6b depended strongly on L-xylose concentra-
tion in both mutant strains (Figure 2). The range of induction
was similar in both genetic contexts, with a plateau reached at
the same concentration. Moreover, at concentrations 4 times
the MIC (as indicated in Table 3), both actinonin and 6b induced
bacteriostatic activity in the first 4 h after drug addition. Finally,
we determined the frequency with which resistance to 6b and
actinonin appeared in B. subtilis. Similar values (5 × 10-7) were
obtained for these two compounds, in keeping with previous
reports for actinonin.28 The resistant strains isolated on 6b at 8
times the MIC showed (i) resistance to actinonin at 8 times the
MIC and (ii) a decrease in doubling time by a factor of 3-8.
Actinonin-resistant strains also proved to be resistant to 6b. The
resistant strains were sensitive to other antibiotics, consistent
with the resistance being specific to the inhibited target, PDF.
Resistance to 6b and actinonin was consistent with the bypassing
of PDF activity induced by Met-tRNAfMet transformylase gene
(fmt) inactivation.28 The sequencing of a number of 6b-resistant
clones led to the identification of various mutations in fmt, all
of which were predicted to lead to gene-function inactivation.
Because the major fitness cost associated with this fmt inactiva-
tion conferring PDFI resistance can be compensated only partly
by further mutations, at extremely low (10-14) rates,29 PDF may
be considered a target of choice for new antibiotics. These data
provide further evidence that the bacterial target of 6b, like that
of actinonin, is PDF.
Experimental Section
Protein Analysis. Escherichia coli PDF (PDF1B) and Bacillus
stearothermophilus PDF2 (PDF2) were used as the representatives
of the two bacterial PDF classes. They were purified in the presence
of nickel to yield highly active Ni-PDF.32 PDF activity was
measured as previously described.33 We monitored, at 37 °C, the
absorbance at 340 nm of NADH (ꢀM ) 6300 M-1 cm-1), essentially
as previously described.33 Fo-Met-Ala-Ser (Bachem AG, Well-am-
Rhein, Switzerland) was used as the substrate, and actinonin (Sigma,
L’Isle d’Abeau Chesnes, France) was used as a PDF inhibitor for
the purposes of comparison. The reaction was started by adding
5-15 µL of purified enzyme. All inhibitors were diluted in
dimethylsulfoxide, and the final assay buffer contained 10% of this
solvent. For the determination of all IC50 values (IC50 is the
concentration giving 50% inhibition of enzyme activity), to prevent
the effects associated with slow binding,24,25 each inhibitor was
incubated with the enzyme for 10 min at 25 °C before kinetic
analysis, which was initiated by adding the substrate. 15N-Labeled
PDFs were produced as previously described.34 NMR experiments
were carried out with a Bruker Avance 600 MHz NMR spectrom-
eter equipped with a 3 mm triple-resonance flow-injection probe.
The probe was connected to a Gilson liquid handler controlled by
the NMR console (Bruker BEST system). The injection protocol
was as previously described.35 For chemical shift perturbation
experiments, 90 µL of 1 mM 15N-labeled PDF1B was mixed with
an equal volume of the tested ligand (3 mM), dissolved in the same
buffer, in a 96-well plate. The mixture was refrigerated at 4 °C on
a Gilson 242 Peltier rack before injection and HSQC experiments.
The conditions for saturation transfer difference22 experiments were
similar except that the final PDF (unlabeled) and ligand concentra-
tions were 20 µM and 1 mM, respectively. STD irradiation was
carried out for 1 or 2 s, with the carrier set on the solid methyl
protein (0.5 ppm), using a field strength γB1/(2π) of 100 Hz.
Chemistry. All solvents and chemicals were purchased from
SDS and Aldrich, respectively. DMF, THF, and CH2Cl2 were dried
according to standard procedures and stored over 4 Å molecular
ID Do Not Trigger Apoptosis. Actinonin has been reported
to induce apoptosis in some cell lines, with cytotoxicity, and
with IC50 values in the range of 2-10 µM.6,12,14,15 Lower levels
of toxicity were observed with 6b than with actinonin in various
cell lines (Figure 3). Apoptosis was strongly induced by
actinonin, as shown by the measurement of caspase activation
in KB cells (Figure 3). At the IC50 of the two compounds,
actinonin induced apoptosis at least 10 times more strongly than
6b in KB cells. At the IC50 of actinonin in KB cells, 6b gave
negligible levels of apoptosis. Thus, ID fulfill in vivo the
conditions imposed at the beginning of the study; they do not
block mPDF because they do not trigger apoptosis, but they do
have potent antibiotic activity.
1
sieves under argon. H NMR spectra were recorded on a Bruker
ARX-250 spectrometer, and chemical shifts were expressed in ppm
downfield from TMS. IR spectra were obtained with a Perkin-Elmer
Spectrum One FT-IR spectrometer equipped with a MIRacle single-
reflection horizontal ATR unit (zirconium-selenium crystal). FAB
and CI mass spectra were recorded at the ENS in Paris. Elemental
analyses were carried out by the microanalysis service at Paris VI

16 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 1
Boularot et al.
University (France) or at Gif-Sur-Yvette (CNRS, France). Experi-
ments performed under argon were run on a vacuum line. The
various synthetic pathways for the products are depicted in Schemes
2-4. General procedures for the chemical syntheses of all
compounds described are reported below, and elemental analysis
results for target compounds are shown in Table S1 in Supporting
Information.
MHz, CDCl3): δ 3.76 (s, 2H), 5.28 (s, 1H), 7.14-7.28 (m, 3H),
7.39 (d, J ) 7.8, 1H), 7.48 (d, J ) 7.7, 1H), 8.12 (s, 1H) ppm.
CI-MS: m/z ) 217 [M + NH4+], 50%; 200 ([M + H]+), 30%. An
etheral solution of HCl(g) (630 µL, C ) 6 N, 3.78 mmoL, 2.2
equiv) was added to a solution (10 mL) of 3a in diethyl ether. We
added 5 mL of DMF, and diethyl ether was carefully evaporated.
A solution (5 mL) of KSAc (432 mg, M ) 114.20, 3.78 mmoL,
2.2 equiv) in DMF was added, via a syringe, under argon. The
solution was stirred overnight at room temperature. Evaporation
of the solvent under vacuum yielded 5a, which was further purified
by column chromatography, with elution with a 8/2 v:v mixture of
ethyl acetate/cyclohexane (220 mg, yield of 52%). Rf (silica gel,
ethyl acetate/cyclohexane, 1/1 v:v mixture) ) 0.6. IR (cm-1): 1788
3-Carboxymethylindol-1-carboxylic Acid Benzyl Ester (1c).
A solution of 2-(indol-3-yl)acetic acid (500 mg, M ) 175.19, 2.85
mmoL) and LHMDS (1 M in THF, 6.27 mmoL, 2.2 equiv, 6.27
mL) in freshly distilled THF (10 mL) was stirred at -78 °C under
argon for 1 h. Benzyl chloroformate (505 µL, M ) 170.60, d )
1.195, 3.42.mmoL, 1.2 equiv) was then added. The reaction mixture
was stirred at -78 °C for 2 h, and solvents were then evaporated
under vacuum. The residue was dissolved in water and extracted
with diethyl ether. The aqueous layer was then acidified to pH 3
with aqueous 0.1 N HCl, resulting in the formation of 3-carboxy-
methylindol-1-carboxylic acid benzyl ester 1c as a white precipitate,
which was filtered off and washed with pentane (795 mg, yield of
90%). Rf (silica gel MerckF 254, CH2Cl2/CH3OH, 9/1 v:v mixture)
1
(νCO). H NMR (250 MHz, CDCl3): δ 2.34 (s, 3H), 3.78 (s, 2H),
3.95 (s, 2H), 7.09-7.57 (m, 5H), 8.19 (s, 1H) ppm. CI-HRMS:
expected for C13H14NO2S+ ([M + H]+), 248.0445. Observed,
248.0448.
Thioacetic Acid S-[3-(5-Bromo-1H-indol-3-yl)-2-oxopropyl]
Ester (5b). The same procedure applied to (5-bromoindol-3-yl)-
acetic acid (300 mg, M ) 254.09, 1.18 mmoL) gave 227 mg of
thioacetic acid 5-[3-(5-bromo-1H-indol-3-yl)-2-oxopropyl] ester 5b,
giving a yield of 59%. Rf (silica gel, ethyl acetate/cyclohexane, 1/1
v:v mixture) ) 0.5. 1H NMR (250 MHz, CDCl3): δ 2.36 (s, 3H),
3.78 (s, 2H), 3.92 (s, 2H), 7.13-7.29 (m, 3H), 7.72 (s, 1H), 7.48
(d, J ) 7.7, 1H), 8.18 (s, 1H) ppm. CI-HRMS: observed for C13H13-
BrNO2S+ ([M + H]+), 79Br 325.9850, 81Br 327.9830; observed,
79Br 325.9843, 81Br 327.9837.
1
) 0.4. Tf ) 152 °C. IR (cm-1): 1728 and 1694 (νCO). H NMR
(250 MHz, DMSO-d6): δ 3.77 (s, 2H), 5.53 (s, 2H), 7.29-7.66
(m, 8H), 7.74 (s, 1H), 8,14 (d, J ) 8.1, 1H) ppm. CI-MS: m/z )
327 [M + NH4+], 100%. Expected values for C18H15NO4: C, 69.89;
H, 4.89; N, 4.53. Observed: C, 69.78; H, 5.00; N, 4.55.
3-Carboxymethyl-5-bromoindol-1-carboxylic Acid Benzyl
Ester (1d). The same procedure applied to 2-(5-bromoindol-3-yl)-
acetic acid (200 mg, M )254.09, 0.787 mmoL) yielded 1d (280
mg, yield of 92%). Rf (silica gel CH2Cl2/CH3OH, 9/1 v:v mixture)
Diazo Compound 3b. 3b was isolated in a 95% yield. Rf (silica
gel, cyclohexane/ethyl acetate, 3/7 v:v mixture) ) 0.5. IR (cm-1):
1
1
) 0.5, Tf ) 88 °C. IR (cm-1): 1729 and 1692 (νCO). H NMR
2101 (νCN), 1715 (νCO). H NMR (250 MHz, CDCl3): δ 3.75 (s,
3
4
2H), 5.68 (s, 1H), 7.22 (dd, J ) 8.5, J ) 1.5, 1H), 7.34 (s, 1H),
7.37 (d, 3J ) 8.5, 1H), 7.75 (d, 4J ) 1.5, 1H) ppm. CI-MS: m/z )
278, 280 [M+], 60%; 250, 252 ([M - N2]+), 100%.
(250 MHz, acetone-d6): δ 3.80 (s, 2H), 5.51 (s,2H), 7.15 (s,1H),
7.24-7.82 (m, 7H), 8.12 (d, J ) 8,1, 1H), 11.16 (s, 1H) ppm. CI-
MS: m/z ) 405, 407 [M + NH4+], 40%; 388, 390 [M+], 30%.
Expected for C18H14BrNO4‚0.5H2O: C, 54.53; H, 3.87; N, 3.53.
Observed: C, 54.67; H, 3.72; N, 3.44.
3-(3-Acetylsulfanyl-2-oxopropyl)indole-1-carboxylic Acid Ben-
zyl Ester (5c). The same procedure applied to 3-carboxymethylin-
dol-1-carboxylic acid benzyl ester (200 mg, M ) 309.32, 0.647
mmoL) gave 178 mg of 5c, corresponding to a yield of 72%. Rf
(ethyl acetate/cyclohexane, 1/1 v:v mixture) ) 0.6. 1H NMR (250
MHz, acetone-d6): δ 2.34 (s, 3H), 3.98 (s, 2H), 4.07 (s, 2H), 5.50
(s, 2H), 7.15-7.80 (m, 9H), 8.15 (d, 4J ) 1.5, 1H) ppm.
CI-HRMS: expected for C21H23NO4S+ [M + NH4+], 399.1379;
observed, 399.1375.
1-(5-Bromo-1H-indol-3-yl)-3-hydroxypropan-2-one (2). Oxalyl
chloride (104 µL, M ) 126.93, d ) 1.455, 1.18 mmoL, 1.5 equiv)
was added at 0 °C, under argon, to a solution (10 mL) of 2-(5-
bromo-1H-indol-3-yl)acetic acid (200 mg, M ) 254.09, 0.788
mmoL) in THF containing a few drops of DMF. The mixture was
stirred at room temperature for 2 h, and the solvent was then
evaporated under vacuum. Tris(trimethylsilyloxy)ethylene (675 µL,
M ) 292.59, d ) 0.885, 1.94 mmoL, 2.3 equiv) was added, under
argon, to the residue dissolved in 10 mL of dioxane. After the
mixture was stirred at room temperature for 10 h, 10 mL of aqueous
HCl (0.1 N) was added and the solution was heated at 80 °C for
30 min. NaCl was added, and the solution was extracted with diethyl
ether. The organic layer was washed with brine, dried over MgSO4,
and evaporated to dryness under vacuum. After purification by
column chromatography (silica gel, elution with CH2Cl2/CH3OH,
95/5 v:v mixture), 1-(5-bromo-1H-indol-3-yl)-3-hydroxypropan-2-
one 2 was obtained (45 mg, yield of 21%). Rf (silica gel, CH2Cl2/
CH3OH, 9/1 v:v mixture) ) 0.5. Tf ) 88 °C. IR (cm-1): 1702
Diazo Compound 3c. 3c was isolated with a yield of 91%. Rf
(silica gel, ethyl acetate/cyclohexane, 1/1 v:v mixture) ) 0.5. IR
1
(cm-1): 2102 (νCN), 1730, and 1636 (νCO). H NMR (250 MHz,
acetone-d6): δ 3.78 (s, 2H), 5.50 (s, 2H), 5.82 (s, 1H), 7.26-7.69
(m, 8H), 7.70 (s, 1H), 8.17 (d, J ) 8, 1H) ppm. CI-MS: m/z )
351 [M + NH4+], 100%.
2-(Indol-3-yl)-N-hydroxyacetamide (6a). A DMF solution (10
mL) of 2-(indol-3-yl)acetic acid (500 mg, M ) 175.19, 2.85 mmol),
1-hydroxybenzotriazole (HOBT) (425 mg, M ) 135.13, 3.14 mmol,
1.1 equiv), and 1-(3-dimethylaminopropyl)-3-ethylcarboxylimine
hydrochloride, EDCI (602 mg, M ) 191.71, 3.14 mmol, 1.1 equiv),
was stirred under argon for 5 min. We then added N-methylmor-
pholine (NMM) (345 µL, M ) 101.15, d ) 0.92, 3.14 mmol, 1.1
equiv). The reaction mixture was stirred at room temperature for 2
h, and hydroxylamine hydrochloride (218 mg, M ) 69.5, 3.14
mmol, 1.1 equiv) was added. The solution was stirred overnight,
and the DMF was then evaporated under vacuum to give a yellow
oil that was dissolved in ethyl acetate. This solution was succes-
sively washed with water and aqueous NaHCO3. After the usual
workup, solvent evaporation gave 6a as a white solid (260 mg,
yield of 60%). IR (cm-1): 1638 (νCO). 1H NMR (250 MHz, DMSO-
d6): δ 3.44 (s, 2H), 7.02 (t, J ) 7.1, 1H), 7.12 (t, J ) 7.1, 1H),
7.19 (s, 1H), 7.39 (d, J ) 7.8, 1H), 7.62 (d, J ) 7.6, 1H), 8.74 (s,
1H), 10.65 (s,1H), 11.85 (s, 1H) ppm. Expected for C10H10N2O2:
C, 63.15; H, 5.30; N, 14.73. Observed: C, 62.96, H, 5.31, N, 14.63.
2-(5-Bromo-1H-indol-3-yl)-N-hydroxyacetamide (6b). Starting
from (5-bromo-1H-indol-3-yl)acetic acid (200 mg, M ) 254.09,
0.787 mmol), 116 mg of 6b was obtained, with a yield of 55%. Rf
(C18 silica gel, ethyl acetate/methanol, 1/1 v:v mixture) ) 0.8. Tf
1
(νCO). H NMR (250 MHz, CDCl3): δ 3.00 (s, 1H), 3.81 (s, 2H),
4.31 (s,2H), 7.15 (s,1H), 7.19-7.32 (m, 2H), 7.64 (s, 1H), 8.21 (s,
1H) ppm. CI-MS: m/z ) 285, 287 [M + NH4+], 80%; 268, 270
[M+], 100%. Expected for C11H10BrNO2: C, 49.28; H, 3.76; N,
5.22. Observed: C, 49.10; H, 3.75; N, 5.17.
Thioacetic Acid S-[3-(Indol-3-yl)-2-oxopropyl] Ester (5a).
Thionyl chloride (150 µL, M ) 118.97, d ) 1.63, 2.06 mmoL, 1.2
equiv) was added, under argon, to a solution (10 mL) of 2-(indol-
3-yl)acetic acid (300 mg, M ) 175.19, 1.72 mmoL) in CH2Cl2.
The mixture was heated at 40 °C for 3 h, and the solvent was
evaporated at low pressure. The residue was dissolved in CH2Cl2
(10 mL) and added to a freshly prepared solution of diazomethane
in diethyl ether under argon via a syringe (10.4 mL, C ) 0.66 M,
6.88 mmoL, 4 equiv). After being stirred at 0 °C for 4 h, the solution
was evaporated under reduced pressure to give in a 95% yield the
diazo compound 3a as a yellow oil that was used without further
purification. 3a: Rf (silica gel, ethyl acetate/cyclohexane, 1/1 v:v
mixture) ) 0.65. IR (cm-1): 2100 (νCN), 1726 (νCO). 1H NMR (250

Inhibitors of Bacterial Peptide Deformylase
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 1 17
1
) 145 °C. IR (cm-1): 3417 (νOH), 3213 (νNH), 1633 (νCO). 1H NMR
(250 MHz, DMSO-d6): δ 3.49 (s, 1H), 7.20-7.44 (m, 3H), 7.87
(s, 1H), 8.86 (s, 1H), 10.59 (s, 1H), 11.14 (s, 1H) ppm. CI-MS:
m/z ) 270, 272 [M + H]+, 100%. Expected for C10H9BrN2O2: C,
44.63; H, 3.37; N, 10.41. Observed: C, 44.52, H, 3.52; N, 10.45.
5-Bromo-3-hydroxycarbamoylmethylindole-1-carboxylic Acid
Benzyl Ester (6d). Starting from 5-bromo-3-carboxymethylindol-
1-carboxylic acid benzyl ester (120 mg, M ) 388.21, 0.309 mmol),
50 mg of 6c was obtained with a yield of 40%. IR (cm-1): 1650
IR (cm-1): 3348 (νNH), 3190 (νOH), 1625 (νCO). H NMR (250
MHz, acetone-d6): δ 3.59 (s, 2H), 7.10 (d, J ) 8.5, 1H), 7.35 (s,
1H), 7.41 (d, J ) 8.5, 1H), 7.67 (s, 1H), 10.00 (s, 1H), 10.29 (s,
1H) ppm. CI-HRMS: expected for C10H10N2O2Cl ([M + H+]),
225.0431 (100%), 227.0461 (33%); observed 225.0432 (100%),
227.0413 (32.4%).
(5-Methoxy-1H-indol-3-yl)acetic Acid Ethyl Ester (8h). We
obtained 219 mg of the ester 8h (yield of 46%) from 300 mg of
5-methoxy-1H-indole 7h (M ) 147.18, 2.04 mmol). 1H NMR (250
MHz, acetone-d6): δ 1.23 (t, J ) 7, 3H), 3.73 (s, 2H), 3.82 (s,
3H), 4.15 (q, J ) 7, 2H), 6.79 (d, J ) 8.5, 1H), 7.11 (s, 1H), 7.26-
7.32 (m, 2H), 9.97 (s, 1H) ppm.
1
and 1690 (νCO). H NMR (250 MHz, DMSO-d6): δ 3.40 (s, 2H),
5.56 (s, 2H), 7.45-8.10 (m, 9H), 8.95 (s, 1H), 10.75 (s, 1H) ppm.
Expected for C18H15BrN2O4: C, 53.62; H, 3.75; N, 6.95. Ob-
served: C, 53.73; H, 3.54; N, 7.11.
2-(5-Methoxy-1H-indol-3-yl)-N-hydroxyacetamide (6h). Start-
ing from 219 mg of the ester 8g (M ) 233.26, 0.94 mmol), we
obtained 90 mg (yield of 43%) of the hydroxamic acid 6h (M )
220.22). IR (cm-1): 3307 (νNH), 3160 (νOH), 1620 (νCO). 1H NMR
(4-Fluoro-1H-indol-3-yl)acetic Acid Ethyl Ester (8e). To a
cooled solution of 4-fluoro-1H-indole 7e (M ) 135.05, 300 mg,
2.22 mmol) in 3.3 mL of THF, we added 1.39 mL (2.22 mmol) of
1.6 M n-BuLi in hexane, keeping the solution below 0 °C with an
ice bath. After 15 min, 2.22 mL of 1 N ZnCl2 in diethyl ether was
added. The cooling bath was removed, and the mixture was stirred
for 24 h. The solvent was then evaporated under vacuum to give a
wax, which was further dissolved in anhydrous toluene (3.3 mL).
Ethyl 2-bromoacetate (246 µL, 2.22 mmol) was added, and the
solution was stirred for 24 h. The mixture was then acidified with
1 N HCl and poured into ethyl acetate. The organic layer was
washed with brine and dried over MgSO4. The ester was subjected
to chromatography on silica gel, eluting with 10% ethyl acetate/
3
(250 MHz, acetone-d6): δ 3.57 (s, 2H), 3.81 (s, 3H), 6.77 (dd, J
) 8.8, 4J ) 2.3, 1H), 7.15 (d, J ) 2.3, 1H), 7.22 (s, 1H), 7.28 (d,
J ) 8.8, 1H), 7.93 (s, 1H) ppm, 10.04 (s, 1H). HRMS: expected
for C11H13N2O3 ([M+H+]), 221.0926; observed 221.0928.
(6-Bromo-1H-indol-3-yl)acetic Acid Ethyl Ester (8i). We
obtained 192 mg of the ester 8i (yield of 45%) from 300 mg of
6-bromo-1H-indole (7i) (M ) 196.05, 1.53 mmol). 1H NMR (250
MHz, DMSO-d6): δ 1.23 (t, J ) 7.2, 3H), 3.76 (s, 2H), 4.13 (q,
J ) 7.2, 2H), 7.19 (dd, 3J ) 8.5, 4J ) 1.5, 1H), 7.34 (s, 1H), 7.55
(d, J ) 8.5, 1H), 7.62 (d, J ) 1.5, 1H), 10.28 (s, 1H) ppm.
2-(6-Bromo-1H-indol-3-yl)-N-hydroxyacetamide (6i). Starting
from 192 mg of the ester 8i (M ) 282.16, 0.68 mmol), we obtained
136 mg (yield of 74%) of the hydroxamic acid 6i (M ) 269.09).
1
cyclohexane to give 158 mg (32%) of 8e. H NMR (250 MHz,
acetone-d6): δ 1.24 (t, J ) 7, 3H), 3.88 (s, 2H), 4.16 (q, J ) 7,
3
3
3
4
2H), 6.71 (dd, JHF ) 11, JHH ) 8, 1H), 7.06 (td, JHH ) 8, JHF
) 5.4, 1H), 7.23 (d, J ) 8, 1H), 7.28 (d, 5JHF ) 1.9, 1H), 10.34 (s,
1H) ppm.
1
IR (cm-1): 3335 (νNH), 3230 (νOH), 1615 (νCO). H NMR (250
3
4
MHz, acetone-d6): δ 3.60 (s, 2H), 7.17 (dd, J ) 8.5, J ) 1.5,
1H), 7.30 (s, 1H), 7.6 (m, 2H), 8.0 (s, 1H), 10.0 (s, 1H), 10.27 (s,
1H) ppm. CI-HRMS: expected for C10H10N2O279Br ([M + H+]),
290.9745; observed 290.9743.
2-(4-Fluoro-1H-indol-3-yl)-N-hydroxyacetamide (6e). NH2OH‚
HCl (497 mg, 7.15 mmol) was added as a powder to 6.4 mL of a
1 M solution of sodium ethoxide in ethanol. This solution was added
to an ethanol solution (10 mL) of the ester 8e (158 mg, 0.715 mmol,
M ) 221.03). The mixture was stirred under argon at 80 °C for 24
h. After the mixture was cooled and the solvent was evaporated
under vacuum, the residue was dissolved in ethyl acetate. This
organic layer was washed with brine, aqueous NaHCO3, 0.1 N HCl,
and finally with brine and dried over MgSO4. After evaporation
under vacuum, the hydroxamic acid 6e was dissolved in a 1:1
mixture of acetone/cyclohexane. Slow evaporation of acetone at
low pressure gave 75 mg of 6e as a white solid (51%). IR (cm-1):
3354 (νNH), 3280 (νOH), 1631 (νCO). 1H NMR (250 MHz, acetone-
2-(5-Bromo-2-methyl-1H-indolyl)acetic Acid Ethyl Ester (8j).
We obtained 201 mg of the ester 8j (yield of 48%) from 300 mg
of 5-bromo-2-methyl-1H-indole 7j (M ) 210.06, 1.43 mmol). H
NMR (250 MHz, acetone-d6): δ 1.18 (t, J ) 7, 3H), 2.33 (S, 3H),
3.66 (s, 2H), 4.06 (q, J ) 7, 2H), 7.10 (d, J ) 8, 1H), 7.22 (d,
J ) 8, 1H), 7.56 (s, 1H), 10.11 (s, 1H) ppm.
1
2-(5-Bromo-2-methyl-1H-indolyl)-N-hydroxyacetamide (6j).
Starting from 201 mg of the ester 8j (M ) 296.16, 0.68 mmol),
we obtained 160 mg (yield of 83%) of the hydroxamic acid 6j (M
) 283.12). IR (cm-1): 1690 (νCO). H NMR (250 MHz, acetone-
1
3
3
d6): δ 3.72 (s, 2H), 6.70 (dd, JHF ) 11, JHH ) 8, 1H), 7.05 (td,
3JHH ) 8, 4JHF ) 5.4, 1H), 7.22 (d, J ) 8, 1H), 7.28 (s, 1H), 10.00
(s, 1H), 10.43 (s, 1H). CI-HRMS: expected for C10H13N3O2F ([M
+ NH4+]), 226.0992; observed 226.0991.
d6): δ 2.44 (s, 3H), 3.69 (s, 2H), 7.14 (d, J ) 8.4, 1H), 7.26 (d,
J ) 8.4, 1H), 7.67 (s, 1H), 10.20 (s, 1H) ppm. CI-HRMS: expected
for C11H14N2O279Br ([M + 3H]+), 285.0239; observed 285.0234.
5-Bromo-2-methyl-1H-indole-3-carbaldehyde (9). POCl3 (267
µL, 2.86 mmoL, d ) 1.64) was added under argon at 0 °C to a
solution of 5-bromo-2-methyl-1H-indole 7j (500 mg, 2.38 mmol)
in 5 mL of DMF. This mixture was stirred at room temperature
overnight, and 2 mL of 2 N aqueous NaOH was then added. The
solution was stirred for 2 h. It was then poured into ethyl acetate.
After the mixture was washed with water, dried, and evaporated to
dryness, 526 mg of 9 was isolated as a white solid (yield of 93%).
IR (cm-1): 1628 (νCdO). 1H NMR (250 MHz, acetone-d6): δ 2.78
(5-Fluoro-1H-indol-3-yl)acetic Acid Ethyl Ester (8f). We
obtained 91 mg of the ester 8f (yield of 44%) from 125 mg of
5-fluoro-1H-indole (7f) (M ) 135.05, 0.926 mmol). 1H NMR (250
MHz, acetone-d6): δ 1.24 (t, J ) 7.2, 3H), 3.74 (s, 2H), 4.13 (q,
J ) 7.2, 2H), 6.92 (t, J ) 9, 1H), 7.30 (d, J ) 9, 1H), 7.38 (m,
2H), 10.21 (s, 1H) ppm.
2-(5-Fluoro-1H-indol-3-yl)-N-hydroxyacetamide (6f). Starting
from 91 mg of the ester 8f (M ) 221.23, 0.41 mmol), we obtained
21 mg (yield of 25%) of the hydroxamic acid 6f (M ) 208.19). IR
(cm-1): 3360 (νNH), 3175 (νOH), 1625 (νCO). 1H NMR (250 MHz,
3
3
(s, 3H), 7.33 (d, J ) 8.45, 1H), 7.36 (d, J ) 8.45, 1H), 8.33 (s,
1H), 10.15 (s, 1H), 11.04 (s, 1H) ppm. CI-MS: m/z ) 255, 257
[M + NH4+], 85%; 238-240, [M+], 100%. Expected for C10H8-
BrNO: C, 50.45; H, 3.39; N, 5.88. Observed: C, 50.42, H, 3.49;
N, 5.87.
3
3
4
acetone-d6): δ 3.57 (s, 2H), 6.91 (td, JHF ) JHH ) 9.3, JHH
)
2.3, 1H), 7.33-7.41 (m, 3H), 8.00 (s, 1H), 10.08 (s, 1H), 10.22 (s,
1H) ppm. CI-HRMS: expected for C10H10N2O2F ([M + H+]),
209.0726; observed 209.0720.
5-Bromo-2-methyl-1H-indole-3-carbaldehydeoxime (10). NH2-
OH‚HCl (70 mg, 1.01 mmol) was added to a solution of 9 (200
mg, 0.84 mmoL) in 5 mL of pyridine. The solution was stirred at
room temperature for 5 h and then evaporated to dryness. The
residue was dissolved in ethyl acetate, and the resulting solution
was successively washed with 1 N HCl and brine and dried over
MgSO4. The solvent was evaporated, and 210 mg of 10 was isolated
(5-Chloro-1H-indol-3-yl)acetic Acid Ethyl Ester (8g). We
obtained 198 mg of the ester 8g (yield of 43%) from 300 mg of
5-chloro-1H-indole (7g) (M ) 151.60, 1.98 mmol). 1H NMR (250
MHz, acetone-d6): δ 1.24 (t, J ) 7, 3H), 3.77 (s, 2H), 4.14 (q,
J ) 7, 2H), 7.24 (d, J ) 8.5, 1H), 7.35-7.40 (m, 2H), 7.79 (s,
1H), 10.33 (s, 1H) ppm.
2-(5-Chloro-1H-indol-3-yl)-N-hydroxyacetamide (6g). Starting
from 198 mg of the ester 8g (M ) 233.26, 0.85 mmol), we obtained
194 mg (yield of 95%) of the hydroxamic acid 6g (M ) 224.64).
as a yellow oil in quantitative yield. IR (cm-1): 1622 (νCdN). H
1
NMR (250 MHz, acetone-d6): δ 2.53 (s, 3H), 7.2-7.35 (m, 2H),
7.80 (s, 1H), 8.32 (s, 1H), 9.69 (s, 1H), 10.49 (s, 1H) ppm. Expected

18 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 1
Boularot et al.
for C10H9BrN2O‚0.25H2O: C, 46.63; H, 3.72; N, 10.87. Ob-
served: C, 46.69; H, 3.66; N, 10.58.
CH2Cl2. The solution was stirred for 2 h at 0 °C under argon, then
washed with water and aqueous NaHCO3 and dried over MgSO4.
Filtration and evaporation at low pressure gave 80 mg of 15, which
was tested without further purification because of the instability of
this molecule during silica gel chromatography. 1H NMR (250 MHz,
CD2Cl2): δ 8.13 (d, J ) 9, 1H), 7.83 (d, J ) 7.4, 2H), 7.73 (d,
J ) 1.7, 1H), 7.6 (m, 4H), 4.47 (s, 2H), 3.4 (s, 1H), 2.69 (s, 3H).
CI-HRMS: expected for C16H1403N2BrS ([M - H]+), 392.9909
(96%), 394.9889 (100%); observed 392.9911 (24.7%), 394.9895
(24.2%).
N-[2-(5-Bromo-2-methyl-1H-indole-3-yl)methyl]hydroxyl-
amine (11). To a 10 mL solution of (10) (110 mg, 0.43 mmol) in
methanol, we added a few crystals of methyl orange followed by
a few drops of an ethanolic solution of HCl(g) (2 N). The solution
turned reddish purple. The red color of the solution was maintained
by adding HCl, and an amount of 2 equiv of NaBH3CN (55 mg,
0.86 mmol) in 5 mL of THF was added. The solution was stirred.
A few hours later, the red color was stable and the solution was
left with stirring overnight. The solvent was evaporated, the residue
was dissolved in 5 mL of methanol, and 5 mL of water was added.
The pH was adjusted to 9 with 1 N aqueous NaOH. This solution
was extracted with CH2Cl2. After the usual workup, 11 (110 mg)
was isolated as a yellow oil in a quantitative yield and was
characterized without further purification. 1H NMR (250 MHz,
acetone-d6): δ 2.51 (s, 3H), 5.16 (s, 2H), 7.1-7.3 (m, 2H), 8.10
(s, 1H), 10.34 (s, 1H) ppm. CI-MS: m/z ) 255, 257 [M+], 30%;
224, 226 ([M - NHOH]+), 100%. Expected for C10H11BrN2O: C,
47.08; H, 4.35; N, 10.98. Observed: C, 46.97; H, 4.51; N, 11.13.
5-Bromo-2-methyl-1-(phenylsulfonyl)-1H-indole-3-carbalde-
hyde (12). A solution of 5-bromo-2-methyl-1H-indole-3-carbalde-
hyde 9 (500 mg, 2.10 mmol) was added at 0 °C under argon to a
suspension of NaH (110 mg, 4.62 mmol, 2.2 equiv) in 10 mL of
THF. The solution was stirred for 1 h, and a solution of
benzenesulfonyl chloride (320 µL, 2.52 mmol, 1.2 equiv) was
added. The mixture was stirred overnight at room temperature. We
then added 50 mL of water and extracted the solution with ethyl
acetate. The organic layer was washed successively with 0.1 N HCl,
aqueous NaHCO3, and brine and dried over MgSO4. Filtration over
Celite and evaporation gave 700 mg of 12 (95% yield) after
N-(1-Benzenesulfonyl-5-bromo-2-methyl-1H-indol-3-ylmethyl)-
N-hydroxyformamide (16). Formic acid (3.77 mL, 0.1 mol) was
added dropwise at 0 °C to acetic anhydride (9.45 mL, 0.1 mol).
The mixture was then heated at 50 °C for 2 h and left to cool to
room temperature. We added 15 (80 mg, 0.2 mmol) dissolved in
10 mL of CH2Cl2 dropwise, at room temperature, to this solution.
The mixture was stirred overnight, and evaporation of the solvent
under vacuum gave an oil, which was subjected to silica gel
chromatography, with elution with CH2Cl2/C6H12, 99.5/0.5 (v:v
mixture) in the presence of a few drops of acetic acid. Following
evaporation, we obtained 50 mg of 16 (yield of 60%) upon
1
precipitation from an ether solution into pentane. H NMR (250
3
MHz, acetone-d6): δ 2.72 (s, 3Η), 4.81 (s, 2H), 7.47 (dd, J )
8.8, 4J ) 1.9, 1H), 7.63 (t, J ) 7.5, 2H), 7.73 (d, J ) 7.5, 1H), 7.8
(d, J ) 1.9, 1H), 7.94 (d, J ) 7.5, 2H), 8.13 (d, J ) 8.8, 1H), 8.35
(s, 1H), 9.10 (s, 1H) ppm. FAB+-MS: 423, 425 ([M + H]+), 10%;
461, 463 ([M + NH4+]), 10%. CI-HRMS: expected for C17H19O4N3-
BrS ([M + NH4+]), 440.0280 (95.6%), 442.0260 (100%); observed
440.0292 (86.4%), 442.0247 (92.9%).
(1-Hydroxycarbamoylmethyl-2-phenylethyl)carbamic Acid
tert-Butyl Ester (18). We prepared 50 mg of 18 with a 55% yield,
following the usual procedure for hydroxamic acid synthesis starting
from 3-tert-butoxycarbonylamino-4-phenylbutyric acid 17 (0.308
1
recrystallization from a pentane/acetone mixture. H NMR (250
3
4
MHz, acetone-d6): δ 3.05 (s, 3H), 7.57 (dd, J ) 8.9, J ) 2.1,
1H), 7.6-8.0 (m, 5H), 8.19 (d, J ) 8.9, 1H), 8.41 (d, J ) 2.1,
1H), 10.32 (s, 1H) ppm. CI-MS: m/z ) 378, 380 [M+], 100%.
2-(5-Bromo-2-methyl-1-(phenylsulfonyl)-1H-indol-3-yl)etha-
nol (13). To a solution of 12 (700 mg, 1.85 mmol) in 10 mL of
methanol, we added a solution of NaBH4 (84 mg, 2.22 mmol) in
10 mL of methanol. After the mixture was stirred for 6 h at room
temperature, the solution was poured into ethyl acetate and washed
with water and brine. The organic layer was dried over MgSO4,
and the solvent was evaporated under vacuum. The alcohol was
subjected to silica gel chromatography, with elution with cyclo-
hexane/ethyl acetate, 70/30, yielding 700 mg of 13 in quantitative
1
mmoL, M ) 279.33, m ) 86 mg). H NMR (250 MHz, DMSO-
d6): δ 1.36 (s, 9H), 2.16 (m, 2H), 2.72 (m, 2H), 3.99 (m, 1H),
6.71 (d, J ) 8.6, 1H), 7.19-7.35 (m, 5H), 8.83 (s, 1H), 10. 40 (s,
1H) ppm. Expected for C14H20N2O4‚0.25H2O: C, 60.29; H, 7.59;
N, 9.37. Observed: C, 60.19; H, 7.38; N, 9.58.
(1-Hydroxyaminomethyl-2-phenylethyl)carbamic Acid tert-
Butyl Ester (20). Compound 20 was prepared from (1-formyl-2-
phenylethyl)carbamic acid tert-butyl ester 19 following the same
procedure as for 11. The product was purified by flash chroma-
tography over silica gel (95:5 CH2Cl2/AcOEt) to give a white solid
1
(45%). H NMR (250 MHz, CDCl3): δ 1.4 (s, 9H), 2.6-2.7 (m,
1
yield as a pale-yellow powder. H NMR (250 MHz, acetone-d6):
1H), 2.8 (d, J ) 6.7, 2H), 3.1-3.3 (m, 1H), 4.1-4.2 (m, 1H), 4.5-
4.6 (m, 1H), 7.2-7.5 (m, 7H). Expected for C14H22N2O3: C, 63.13;
H, 8.32; N, 10.51. Observed: C, 63.17; H, 8.14; N, 10.43.
(1-Hydroxycarbamoylmethyl-2-phenylethyl)carbamic Acid
tert-Butyl Ester (21). Compound 21 was synthesized as described36
from (1-formyl-2-phenylethyl)carbamic acid tert-butyl ester 19. IR
(cm-1): 1682 (νCO). 1H NMR (250 MHz, CDCl3): δ 1.38 (s, 1H),
2.83 (d, J ) 6.7, 2H), 3.08-4.01 (m, 2H), 4.19 (m, 1H), 4.61 (d,
J ) 3.9, 1H), 7.15-7.80 (m, 5H), 8.32 (s, 1H), 8.85 (s, 1H).
Expected for C15H22N2O4‚0.2H2O: C, 60.47; H, 7.58; N, 9.40.
Observed: C, 60.67; H, 7.57; N, 9.31.
3
4
δ 2.64 (s, 3H), 4.02 (s, 1H), 4.72 (s 2H), 7.45 (dd, J ) 8.9, J )
2, 1H), 7.83 (d, J ) 2, 1H), 7.6-8.0 (m, 5H), 8.13 (d, J ) 8.9,
1H) ppm. CI-MS: m/z ) 380, 382 [M+], 100%.
tert-Butyl 2-(5-Bromo-2-methyl-1-(phenylsulfonyl)-1H-indol-
3yl)methyl(tert-butoxycarbonyloxy)carbamate (14). To a solution
of 13 (220 mg, 0.58 mmol) in 10 mL of THF, we successively
added, at room temperature under argon, PPh3 (1.1 equiv, 167 mg,
0.64 mmol). We stirred the mixture for 15 min and then added
N-bromosuccinimide (1.1 equiv, 113 mg, 0.64 mmol). The mixture
was stirred at room temperature overnight. The bromo derivative
was not isolated and was subsequently tested in this solution. A
solution of N,O-bis(tert-butoxycarbonyl)hydroxylamine (135 mg,
0.58 mmol) in 5 mL of DMF previously stirred for 15 min with
NaH (1.1 equiv, 15 mg) was added to the bromo derivative. The
resulting solution was stirred for 2 h. After the addition of CH2-
CL2, the solution was washed with water, aqueous NH4Cl, and brine
and dried over MgSO4. The solvent was evaporated, and the residue
was subjected to silica gel chromatography, with elution in CH2-
Cl2/C6H12, 70/30 and 80/20, giving 150 mg of 14 (yield of 44%).
1H NMR (250 MHz, acetone-d6): δ 1.35 (s, 9H), 1.49 (s, 9H),
2.67 (s, 3H), 4.87 (s, 2H,), 7.47 (dd, 3J ) 8.8, 4J ) 1.8, 1H), 7.63
(m, 3H), 7.78 (d, J ) 1.8, 1H), 7.9 (d, J ) 7.5, 2H), 8.13 (d, J )
8.8, 1H) ppm. CI-MS: m/z ) 612, 614 [M+], 100%.
Microbiology. Two wild-type Bacillus subtilis strains were
used: MO3482, a prototrophic skin prophage-cured derivative of
JH642, and 168 (trpC2).37 Strains Y101 (∆ykrB::nm, trpC2) and
D101 (∆def::nm, trpC2) were derived from 168.27 Two 168
derivatives (one in which both wild-type PDF loci (def and ykrB)
were deleted and another in which one PDF locus was conditionally
expressed under the control of L-xylose) were used:27 Y103 (∆def::
nm, ∆ykrB::erm, ∆thrC::xylR-PxylA-ykrB-spc, trpC2) and D103
(∆def::erm, ∆ykrB::nm, ∆thrC::xylR-PxylA-def-spc, trpC2). E. coli
strains K37 (galK), JM101Tr (galK, rpsL, recA56, srl-300::Tn10),
and CAG1284 (λ-, tolC210::Tn10, rph-) have been described
elsewhere.38-40 Drug susceptibility was determined by culturing an
inoculum containing (1-4) × 104 CFU. The inoculum was cultured
for 18 h at 37 °C in 3 mL of Mueller-Hinton broth medium
(Sigma) in the presence of various dilutions of the drug. The MIC
value obtained corresponds to the lowest concentration of the drug
N-((5-Bromo-2-methyl-1(phenylsulfonyl)-1H-indol-3-yl)meth-
yl)hydroxylamine (15). We added 775 µL of trifluoroacetic acid
(40 equiv) to a solution of 14 (150 mg, 0.25 mmol) in 10 mL of

Inhibitors of Bacterial Peptide Deformylase
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 1 19
(10) Yuan, Z.; Trias, J.; White, R. J. Deformylase as a novel antibacterial
target. Drug DiscoVery Today 2001, 6, 954-961.
(11) Giglione, C.; Boularot, A.; Meinnel, T. Protein N-terminal methionine
excision. Cell. Mol. Life Sci. 2004, 61, 1455-1474.
at which less than 10% control levels of growth occurred, as
assessed by the measurement of optical density at 600 nm (1 OD600
≈ 3 × 109 CFU/mL).
Cell-Based Assays. Apoptosis in KB cells was assessed by
measuring luminescence due to caspase-3/7 induction, with the
Caspase-Glo 3/7 kit (Promega). In this assay, a Z-DEVD prolu-
minescent substrate produces light in the presence of caspase-3/7,
ATP, O2, and Glo luciferase. A 10-8 M docetaxel concentration
used as internal standard induced caspase activity by a factor of 2.
Toxicity assays were performed with the CellTiter 96 AQeous One
solution proliferation assay kit (Promega), using colorimetric
measurement at 490 nm. All assays were carried out in triplicate
with a Biomek workstation (Beckman) in 96-well plates.
3D Modeling. All PDF sequences were numbered as for the
EcPDF sequence, as previously described.16 A substitution of
residue X in any PDF indicates that the substitution concerns the
residue corresponding to amino acid X in EcPDF. Three-
dimensional (3D) modeling was carried out with Insight II software
(Accelrys). The 3D structures of Pseudomonas aeruginosa PDF
(PaPDF1) bound to a benzathiozinone-hydroxamic (BTH) deriva-
tive (PDB entry 1S17)41 and PaPDF1 bound to actinonin (PDB
entry 1LRY)16 were aligned. Compound 6b was constructed with
the Sketcher module, aligned on the structure of both actinonin
and BTH, using the hydroxamate group as a fixed anchor and was
used to replace either compound in the 3D structure. The 6b
structure docked to PaPDF1 was subjected to further energy
minimization with the CharmM force field, and the lowest energy
structure was selected.
(12) Lee, M. D.; She, Y.; Soskis, M. J.; Borella, C. P.; Gardner, J. R.;
Hayes, P. A.; Dy, B. M.; Heaney, M. L.; Philips, M. R.; Bornmann,
W. G.; Sirotnak, F. M.; Scheinberg, D. A. Human mitochondrial
peptide deformylase, a new anticancer target of actinonin-based
antibiotics. J. Clin. InVest. 2004, 114, 1107-1116.
(13) Leszczyniecka, M.; Bhatia, U.; Cueto, M.; Nirmala, N. R.; Towbin,
H.; Vattay, A.; Wang, B.; Zabludoff, S.; Phillips, P. E. MAP1D, a
novel methionine aminopeptidase family member is overexpressed
in colon cancer. Oncogene, in press.
(14) Xu, Y.; Lai, L. T.; Gabrilove, J. L.; Scheinberg, D. A. Antitumor
activity of actinonin in Vitro and in ViVo. Clin. Cancer Res. 1998, 4,
171-176.
(15) Grujic, M.; Renko, M. Aminopeptidase inhibitors bestatin and
actinonin inhibit cell proliferation of myeloma cells predominantly
by intracellular interactions. Cancer Lett. 2002, 182, 113-119.
(16) 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.
(17) Ragusa, S.; Mouchet, P.; Lazennec, C.; Dive, V.; Meinnel, T.
Substrate recognition and selectivity of peptide deformylase. Similari-
ties and differences with metzincins and thermolysin. J. Mol. Biol.
1999, 289, 1445-1457.
(18) Fieulaine, S.; Juillan-Binard, C.; Serero, A.; Dardel, F.; Giglione,
C.; Meinnel, T.; Ferrer, J. L. The crystal structure of mitochondrial
(type 1A) peptide deformylase provides clear guidelines for the design
of inhibitors specific for the bacterial forms. J. Biol. Chem. 2005,
280, 42315-42324.
(19) Serero, A.; Giglione, C.; Meinnel, T. Distinctive features of the two
classes of eukaryotic peptide deformylases. J. Mol. Biol 2001, 314,
695-708.
(20) Meinnel, T.; Blanquet, S.; Dardel, F. A new subclass of the zinc
metalloproteases superfamily revealed by the solution structure of
peptide deformylase. J. Mol. Biol. 1996, 262, 375-386.
(21) Meinnel, T.; Patiny, L.; Ragusa, S.; Blanquet, S. Design and synthesis
of substrate analogue inhibitors of peptide deformylase. Biochemistry
1999, 38, 4287-4295.
(22) Mayer, M.; Meyer, B. Group epitope mapping by saturation transfer
difference NMR to identify segments of a ligand in direct contact
with a protein receptor. J. Am. Chem. Soc. 2001, 123, 6108-6117.
(23) Van Aller, G. S.; Nandigama, R.; Petit, C. M.; Dewolf, W. E., Jr.;
Quinn, C. J.; Aubart, K. M.; Zalacain, M.; Christensen, S. B.;
Copeland, R. A.; Lai, Z. Mechanism of time-dependent inhibition
of polypeptide deformylase by actinonin. Biochemistry 2005, 44,
253-260.
(24) 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.
(25) Morrison, J. F.; Walsh, C. T. The behavior and significance of slow-
binding enzyme inhibitors. AdV. Enzymol. Relat. Areas Mol. Biol.
1988, 61, 201-301.
Acknowledgment. This work was supported by Grants
MIME-2006 (Agence Nationale de la Recherche), PCV-00/027,
and MCT-2000, a “Bourse de Docteur Inge´nieur” to A.B.
(Centre National de la Recherche Scientifique, France), and
Grant 3363 (Association pour la Recherche sur le Cancer,
Villejuif, France). We warmly thank C. Freiberg (Bayer AG,
Wuppertal, Germany) and P. Stragier (IBPC, CNRS, Paris,
France) for generously providing us with strains and for helpful
advice on the B. subtilis culture. We thank G. Aubert (CNRS,
Gif/Yvette, France) for carrying out the toxicity experiments.
Supporting Information Available: Additional details for the
synthesis of the various PDFI described in Schemes 2 and 3, a
table listing the purities of target compounds, Figure S1 showing
the results of 3D modeling of the PDF/6b complex, and Figures
S2 and S3 showing the results from kinetic studies of the interaction
between PDF and 6b. This material is available free of charge via
the Internet at http://pubs.acs.org.
References
(1) Hughes, D. Exploiting genomics, genetics and chemistry to combat
antibiotic resistance. Nat. ReV. Genet. 2003, 4, 432-441.
(2) Miesel, L.; Greene, J.; Black, T. A. Genetic strategies for antibacterial
drug discovery. Nat. ReV. Genet. 2003, 4, 442-456.
(26) Henderson, P. J. A linear equation that describes the steady-state
kinetics of enzymes and subcellular particles interacting with tightly
bound inhibitors. Biochem. J. 1972, 127, 321-333.
(3) Mills, S. D. When will the genomics investment pay off for
antibacterial discovery? Biochem. Pharmacol. 2006, 71, 1096-1102.
(4) Boularot, A.; Giglione, C.; Artaud, I.; Meinnel, T. Structure-activity
relationship and therapeutic potential of peptide deformylase inhibi-
tors. Curr. Opin. InVest. Drugs 2004, 5, 809-822.
(27) Haas, M.; Beyer, D.; Gahlmann, R.; Freiberg, C. YkrB is the main
peptide deformylase in Bacillus subtilis, a eubacterium containing
two functional peptide deformylases. Microbiology 2001, 147, 1783-
1791.
(28) Giglione, C.; Meinnel, T. Resistance to anti-peptide deformylase
drugs. Expert Opin. Ther. Targets 2001, 5, 415-418.
(29) Nilsson, A. I.; Zorzet, A.; Kanth, A.; Dahlstrom, S.; Berg, O. G.;
Andersson, D. I. Reducing the fitness cost of antibiotic resistance
by amplification of initiator tRNA genes. Proc. Natl. Acad. Sci. U.S.A.
2006, 103, 6976-6981.
(30) Clements, J. M.; Beckett, R. P.; Brown, A.; Catlin, G.; Lobell, M.;
Palan, S.; Thomas, W.; Whittaker, 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.
(5) Jain, R.; Chen, D.; White, R. J.; Patel, D. V.; Yuan, Z. Bacterial
peptide deformylase inhibitors: a new class of antibacterial agents.
Curr. Med. Chem. 2005, 12, 1607-1621.
(6) Lee, M. D.; Antczak, C.; Li, Y.; Sirotnak, F. M.; Bornmann, W. G.;
Scheinberg, D. A. A new human peptide deformylase inhibitable by
actinonin. Biochem. Biophys. Res. Commun. 2003, 312, 309-315.
(7) Nguyen, K. T.; Hu, X.; Colton, C.; Chakrabarti, R.; Zhu, M. X.;
Pei, D. Characterization of a human peptide deformylase: Implica-
tions for antibacterial drug design. Biochemistry 2003, 42, 9952-
9958.
(8) Serero, A.; Giglione, C.; Sardini, A.; Martinez Sanz, J.; Meinnel, T.
An unusual peptide deformylase features in the human mitochondrial
N-terminal methionine excision pathway. J. Biol. Chem. 2003, 278,
52953-52963.
(9) Giglione, C.; Pierre, M.; Meinnel, T. Peptide deformylase as a target
for new generation, broad spectrum antimicrobial agents. Mol.
Microbiol. 2000, 36, 1197-1205.
(31) Apfel, C.; Banner, D. W.; Bur, D.; Dietz, M.; Hirata, T.; Hubschw-
erlen, C.; Locher, H.; Page, M. G.; 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.

20 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 1
Boularot et al.
(32) Ragusa, S.; Blanquet, S.; Meinnel, T. Control of peptide deformylase
activity by metal cations. J. Mol. Biol. 1998, 280, 515-523.
(33) Lazennec, C.; Meinnel, T. Formate dehydrogenase-coupled spectro-
photometric assay of peptide deformylase. Anal. Biochem. 1997, 244,
180-182.
(39) Hirel, P.-H.; Le´veˆque, F.; Mellot, P.; Dardel, F.; Panvert, M.;
Mechulam, Y.; Fayat, G. Genetic engineering of methionyl-tRNA
synthetase: in Vitro regeneration of an active synthetase by proteolytic
cleavage of a methionyl-tRNA synthetase-â-galactosidase chimeric
protein. Biochimie 1988, 70, 773-782.
(40) Singer, M.; Baker, T. A.; Schnitzler, G.; Deischel, S. M.; Goel, M.;
Dove, W.; Jaacks, K. J.; Grossman, A. D.; Erickson, J. W.; Gross,
C. A. A collection of strains containing genetically linked alternating
antibiotic resistance elements for genetic mapping of Escherichia
coli. Microbiol. ReV. 1989, 53, 1-24.
(41) Molteni, V.; He, X.; Nabakka, J.; Yang, K.; Kreusch, A.; Gordon,
P.; Bursulaya, B.; Warner, I.; Shin, T.; Biorac, T.; Ryder, N. S.;
Goldberg, R.; Doughty, J.; He, Y. Identification of novel potent
bicyclic peptide deformylase inhibitors. Bioorg. Med. Chem. Lett.
2004, 14, 1477-1481.
(34) Dardel, F.; Ragusa, S.; Lazennec, C.; Blanquet, S.; Meinnel, T.
Solution structure of nickel-peptide deformylase. J. Mol. Biol. 1998,
280, 501-513.
(35) Tisne, C.; Dardel, F. Optimisation of a peptide library for screening
specific RNA ligands by flow-injection NMR. Comb. Chem. High
Throughput Screening 2002, 5, 523-529.
(36) Xiang, J.; Christensen, S.; Liao, X. Peptide deformylase inhibitors,
Patent WO-02070653. 2002.
(37) Anagnostopoulos, C.; Crawford, I. P. Transformation studies on the
linkage of markers in the tryptophan pathway in Bacillus subtilis.
Proc. Natl. Acad. Sci. U.S.A. 1961, 47, 378-390.
(38) Miller, H. I.; Friedman, D. I. An E. coli gene product required for λ
site-specific recombination. Cell 1980, 20, 711-719.
JM060910C