Synthesis and evaluation of 3-(dihydroxyboryl)benzoic acids as D,D-carboxypeptidase R39 inhibitors

Article abstract of DOI:10.1021/jm9009718

Penicillin binding proteins (PBPs) catalyze steps in the biosynthesis of bacterial cell walls and are the targets for the β-lactam antibiotics. Non-β-lactam based antibiotics that target PBPs are of interest because bacteria have evolved resistance to the

Full text of DOI:10.1021/jm9009718

J. Med. Chem. 2009, 52, 6097–6106 6097
DOI: 10.1021/jm9009718
Synthesis and Evaluation of 3-(Dihydroxyboryl)benzoic Acids as D,D-Carboxypeptidase R39 Inhibitors
Steven R. Inglis,† Astrid Zervosen,‡ Esther C. Y. Woon,† Thomas Gerards,‡ Nathalie Teller,‡ Delphine S. Fischer,†
Andre Luxen,*,‡ and Christopher J. Schofield*,†
ꢀ
†Department of Chemistry, Chemistry Research Laboratory, University of Oxford, 12 Mansfield Road, Oxford, OX1 3TA, U.K., and ‡Centre de
Recherches du Cyclotron, B30, Universite de Lieꢁge, Sart-Tilman, B-4000, Lieꢁge, Belgium
Received July 1, 2009
Penicillin binding proteins (PBPs) catalyze steps in the biosynthesis of bacterial cell walls and are the
targets for the β-lactam antibiotics. Non-β-lactam based antibiotics that target PBPs are of interest
because bacteria have evolved resistance to the β-lactam antibiotics. Boronic acids have been developed
as inhibitors of the mechanistically related serine β-lactamases and serine proteases; however, they have
not been explored extensively as PBP inhibitors. Here we report aromatic boronic acid inhibitors of the
D,D-carboxypeptidase R39 from Actinomadura sp. strain. Analogues of an initially identified inhibitor
[3-(dihydroxyboryl)benzoic acid 1, IC50 400 μM] were prepared via routes involving pinacol boronate
esters, which were deprotected via a two-stage procedure involving intermediate trifluorborate salts that
were hydrolyzed to provide the free boronic acids. 3-(Dihydroxyboryl)benzoic acid analogues contain-
ing an amide substituent in the meta, but not ortho position were up to 17-fold more potent inhibitors of
the R39 PBP and displayed some activity against other PBPs. These compounds may be useful for the
development of even more potent boronic acid based PBP inhibitors with a broad spectrum of
antibacterial activity.
Introduction
irreversible inhibition (Figure 1B). Bacteria have developed
resistance to β-lactams by mechanisms including the produc-
tion of β-lactamases that catalyze β-lactam hydrolysis
(Figure 1B), and production of PBPs that no longer recognize,
or have reduced affinity for, β-lactams. While attempts have
been made to prepare alternative lactam structures that also
leadtoastable acyl-enzyme complex (including bothsynthetic
compounds4,5 and derivatives of the natural product γ-lac-
tam, lactivicin6-9), these compounds are also likely to be
prone to hydrolysis by β-lactamases. Thus, there is a need to
develop transpeptidase inhibitors that are not acylating
agents.
Boronic acids have received attention as inhibitors of
enzymes employing nucleophilic catalysis, in particular as
inhibitors of serine and other proteases (reviewed by Yang
et al.10). Inhibition of nucleophilic enzymes by boronic acids
employs the electron deficient nature of the boron atom in
its trivalent form which enables it to form covalent, “tetra-
hedral” adducts with nucleophilic residues (Figure 1C). Boro-
nic acid inhibitors of both the AmpC11-14 and TEM β-
lactamases15-17 have been reported. However, boronic acids
have received relatively little attention as PBP inhibitors. The
kinetics of peptide based boronic acid inhibitors of PBP3,
PBP4, and PBP5 from Neisseria gonorrhoeae have been
studied18 and a crystal structure of a peptide boronic acid in
complex with PBP5 has been reported.19 Phenyl- and methyl-
boronic acids have been tested against these same enzymes,
and modest activity was observed for phenylboronic acid
against PBP3.20
The bacterial cell wall consists of a complex network of
sugars connected to peptides that are cross-linked by the
transpeptidase activity of penicillin binding proteins (PBPsa)
(reviewed in Macheboeuf et al.1 and Sauvage et al.2). PBP
catalysis proceeds via acyl-enzyme intermediates formed by
nucleophilic reaction of a serine residue onto the amide
carbonyl group of a C-terminal D-ala-D-ala dipeptide frag-
ment of a stem peptide (Figure 1A) with the concomitant loss
of the terminal D-ala fragment. The acyl-enzyme intermediate
then undergoes reaction with a nucleophilic side chain, often
lysine (in Gram-positive bacteria or diaminopimelic acid
(DAP) Gram-negative bacteria), from a neighboring stem
peptide toform a cross-linkedpeptide withregeneration of the
free PBP.
The PBPs can be classified into high molecular weight and
low molecular weight enzymes. Within the high molecular
weight class, there are two subclasses: the bifunctional class A
and monofunctional class B (which may also be bifunctional).
The low molecular weight PBPs are sometimes referred to as
class C PBPs.1 The high molecular weight proteins are gen-
erally multifunctional, and their inhibition leads to cell death.
Inhibition of the low molecular weight proteins is generally
nonlethal but leads to a change in the cell wall cross-linking
pattern.3 β-Lactam antibiotics (e.g., penicillins) react irrever-
sibly with the nucleophilic serinyl residue of the PBP, causing
*To whom correspondence should be addressed. For A.L.: phone,
þ32 4 366 2360; fax, þ32 4 366 2946; e-mail, aluxen@ulg.ac.be. For
C.J.S.: phone, þ44 1865 275 625; fax, þ44 1865 275 674; e-mail,
christopher.schofield@chem.ox.ac.uk.
a Abbreviations: PBP, penicillin binding protein; RA, residual
activity.
Here we describe the identification and development
3-(dihydroxyboryl)benzoic acid derivatives as inhibitors of
the D,D-carboxypeptidase R39 from Actinomadura sp. strain
(R39). R39 is a low molecular weight PBP that is related to
r
2009 American Chemical Society
Published on Web 09/04/2009
pubs.acs.org/jmc

6098 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 19
Inglis et al.
Figure 1. Reactions of nucleophilic serine enzymes: (A) transpeptidase activity, (B) differential reactivities of PBPs and serine β-lactamases
(βL) with a β-lactam antibiotic, and (C) reaction with a boronic acid inhibitor.
PBP4 from Escherichia coli,21 PBP4a from Bacillus subtilis,22
and PBP3 from Neisseria gonorrheae.23 With a view to devel-
opment of potent inhibitors of more clinically important high
molecular weight PBPs (e.g., PBP1b, PBP2xR6, and the
penicillin resistant strain PBP2x5204), we used R39 as a model
system with established kinetic assays for the initial develop-
ment stages. We identified lead compounds from a set of
commercially available boronic acids tested against three
different PBPs. We found that 3-(dihydroxyboryl)benzoic
acid 1 was a modest inhibitor of R39. The design and synthesis
of analogues of 1 are described, from which we developed
inhibitors with improved potency for R39.
inhibitor identified in this initial screen was 1 [residual activity
(RA) of 20% at 1 mM], in both the absence and presence of
Triton-X detergent, which is used to test for the possibility of
nonspecific interactions/aggregation.25,26 The IC50 of 1 was
determined as 400 ( 19 μM; the corresponding pinacol ester
19 showed similar activity(RA = 38% at500 μM), suggesting
that pinacol esters are hydrolyzed under the assay conditions.
The thiophene boronic acid 13, analogous with 1, displayed
less inhibition (57% at 1 mM), while the remaining thiophene
analogues 10, 11, and 14-16 and the known β-lactamase
inhibitor 1213 were even less active. The derivatives with bulky
substituents (5 and 6) were poorly soluble under the assay
conditions and could only be tested at a lower concentration
of 100 μM, at which they showed negligible inhibition. The
trifluoromethyl and formyl analogues (20 and 21) of 1 were
also tested because reactive carbonyl compounds are used as
inhibitors of nucleophilic enzymes.27-29 These compounds
displayed some inhibition; however, they were less active than
the boronicacid1. This observationis consistent withwork on
proteases where boronic acid inhibition was more potent than
with an analogous aldehyde30 or fluorinated ketone.31 None of
the set showed any significant inhibition against the PBP2xR6
and PBP2x5204 penicillin resistant strains (Supporting Infor-
mation).
Identification of Arylboronic Acids as R39 Inhibitors
Initially, a set of commercially available boronic acids
including phenylboronic acids (1-9), thiophenyl boronic
acids (10-16), an alkylboronic acid 17, a bicyclic benzoxa-
borole 18, and a boronic acid pinacol ester 19 were tested for
inhibition of the R39 D,D-peptidase from Actinomadura sp.
strain and of PBP2xR6 and PBP2x5204 (penicillin resistant)
from Streptococcus pneumoniae. The assays involved hydro-
lysis of a reporter substrate (R)-[2-(benzoylamino)prop-
ionylsulfanyl]acetic acid24 in the presence of 5,50-dithiobis(2-
nitrobenzoic acid) (monitoring at 412 nm) in a procedure
similar to that reported.7 Four compounds (1, 8, 18, and 19)
displayed at least 50% inhibition of the R39 enzyme at the
initially tested concentrations (1 mM for 1 and 18; 0.5 mM for
compounds 8 and 19, which were not fully soluble under the
assay conditions at 1 mM) (Table 1). The most potent
These initial tests suggested that phenylboronic acids with a
carboxylic acid at the meta- position (as in 1) can inhibit R39.
In the case where the carboxylate group was at the para-
position (as in 4), no inhibition was observed. Similarly, when
the carboxylate group was absent as in 8, reduced inhibition

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 19 6099
Table 1. Inhibition Data for the Initial Set of Compounds Tested as R39 Inhibitors
compd
concn (μM)
residual activity (%)a
compd
concn (μM)
residual activity (%)a
1
1000
1000
1000
1000
100
20 ( 5 (IC50 = 400 ( 19)
82 ( 3
85 ( 3
84 ( 4
104 ( 5
97 ( 5
98 ( 2
50 ( 3
60 ( 8
91 ( 5
12
13
14
15
16
17
18
19
20
21
1000
1000
1000
1000
1000
1000
1000
500
86 ( 5
57 ( 6
99 ( 2
88 ( 3
95 ( 4
91 ( 5
45 ( 5
38 ( 4
60 ( 2
77 ( 3
2
3
4
5
6
100
7
1000
500
8
9
500
1000
1000
10
11
1000
1000
88 ( 6
a Quoted values are mean ( standard deviation over three replicate experiments.
was observed. In cases where alternative substituents were
present (as in 2 and 3), only modest inhibition was observed.
Compound 9, a known inhibitor of the serine protease
elastase,13 also displayed activity against R39, suggesting that
compounds with a m-(sulfon)amido side chain may be accom-
modated at the active site. These results suggested that a
combination of meta-arrangements of both the boronic acid
and the carboxylate (as in 1) and of the boronic acid and an
amido group (as in 9) may give rise to more potent inhibitors.
Docking studies (Figure 2) based on a crystal structure for
the acyl-enzyme complex formed between the β-lactam nitro-
cefin and R3932 were then performed in an attempt to identify
improved boronic acid inhibitors. These studies assumed that
the boronic acid reacted with the nucleophilic serine (ser 49) to
form a covalent tetrahedral complex. The results suggested
that the carboxylate group of 1 binds in a similar way to the
carboxylate of nitrocefin (Figure 2A) and that 1 analogues
withano-or m-acetamido group(22a and23) wouldbindwith
the acetamido side chain interacting in a way similar to that of
nitrocefin (Figure 2B).
then explored with the phenylacetamido derivative being
chosen for method development (Scheme 1). The boronic
acid was introduced as its pinacol ester protected form, by
reaction of 24 with bis(pinacolato)diboron and catalysis using
[1,10-bis(diphenylphosphino)ferrocene]palladium(II) chloride
(PdCl2 dppf) in the presence of potassium acetate using a
3
modified version of a reported procedure35-37 to give 25. As
observed with analogous palladium catalyzed couplings,35 a
low yield of the reduced coproduct 26 was consistently
isolated.
Prior to establishing a reliable method for the deprotection
of pinacol boronate esters, hydrolysis of the methyl ester to
yield the carboxylic acid 27 was attempted, in part because we
had previously shown that the pinacol ester 19 of the lead 1
had similar inhibitory activity to the free boronic acid.
Furthermore, pinacol boronate esters have previously been
assumed to hydrolyze under the aqueous conditions used for
biological studies.11,30 The carboxylic acid 27 was prepared by
treatment of 25 with lithium hydroxide in a mixture of THF
and water. After workup, the NMR of the crude material in
acetone-d6 indicated that 27 was the major species present,
with no evidence for hydrolysis of the pinacol ester.
Synthesis of Derivatives of 1
The synthesis of derivatives of 1 with acetamido side
chains at the ortho-position relative to the boronic acid was
In order to reliably obtain boronic acids free of pinacol for
the inhibition assay, there was a need to synthesize the free

6100 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 19
Inglis et al.
Figure 2. Rationale for derivatives of 1. (A) Docking of 1 using GOLD33,34 and comparison with a nitrocefin/R39 complex crystal structure
(PDB code 1W8Y).32 Note the C-3 side chain of nitrocefin was not observed in this structure. (B) Docking of proposed derivatives of 1, 22a,
and 23.
Scheme 1. Synthesis of o-Phenylacetamido Derivative of 1a
a Reagents and conditions: (a) PhCH2COCl/N,N-dimethylacetamide, room temp overnight; (b) bis(pinacolato)diboron, PdCl2 dppf, AcOK,
3
toluene, 90 °C overnight; (c) LiOH H2O/THF/H2O, room temp overnight; (d) KHF2(aq)/MeOH, room temp 1 h; (e) TMSCl/CH3CN/H2O, room
3
temp 1 h; (f) LiOH H2O/dioxane/water, 60 °C 1 h, then room temp overnight. In solution, 29b and 22b may be in equilibrium with ring opened forms 29a
3
and 22a, respectively.
boronic acid 22a. A commonly used method for deprotection
of pinacol boronate esters is treatment with sodium periodate
under acidic conditions.38,39 When applied to 25, a low re-
covery of a mixture of products was isolated. We then explored
the method of Yuen and Hutton40 for the deprotection of
pinacol esters via the formation of intermediate trifluoro-
borate salts (Scheme 1). Thus, intermediate 28 was obtained
after treatment of 25 with aqueous potassium hydrogen
difluoride in methanol. The salt 28 was then treated with
trimethylsilylchlorideinaqueous acetonitriletogiveester29b.

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 19 6101
The structure of 29b was assigned, at least in the solid state, as
the bicyclic benzoxazaborinine on the basis of elemental
analysis and mass spectrometric data. The formation of such
a bicyclic boron containing heterocycle, instead of the free
boronic acid form 29a, has been observed previously in
attempts to prepare o-amidophenylboronic acids.41,42 It has
been suggested that such compounds exist in equilibrium with
the monocyclic boronic acid form in solution.43 An analogous
ortho-substituted thiophenylboronic acid 30 (Table 2 below),
derived from 13, was alsoprepared (Supporting Information).
Benzoxazaborinine 29b was then hydrolyzed with lithium
hydroxide to give carboxylic acid 22b. 29b and 22b have very
similar NMR spectra (aside from the anticipated loss of the
ester CH3 peak and the appearance of a CO2H peak),
suggesting that the form of the product obtained in the
NMR solution was unchanged following exposure to the
basicand then acidicdeprotectionconditions (i.e., the product
is predominantly in the bicyclic form 22b). NMR analysis
under aqueous conditions indicated that pinacol ester 27
undergoes rapid hydrolysis to give 22b (Supporting Infor-
mation). However, wefound thatthe pinacol ester 27 re-forms
during acidic workup if the pinacol is not removed.
Meta-substituted boronic acids (23 and 31-37) were then
prepared by a route analogous to that used for the ortho-
derivative 22b (Scheme 2). In contrast to the comparable
reaction with the ortho-derivative, no reduced byproduct
analogous with 26 (Scheme 1) was obtained during the
palladium catalyzed preparation of the m-pinacol esters lead-
ing to higher yields in most cases. There was no evidence for
intramolecular interactions between the m-amido group and
the boronic acid for this series.
Table 2. Inhibitory Activity of Ayrlboronic Acids against R39a
Activity of Boronic Acid Analogues against PBPs
The two ortho-substituted 1 derivatives 27 and 22b and the
analogous thiophene methyl ester 30, derived from 13, were
poor inhibitors of R39 (RA = 80-100% at 1 mM, Table 2),
displaying a reduction in potency compared with their respec-
tive leads 1 and 13. This lack of activity may be a consequence
of the tendency of these compounds to exist in the bicyclic
benzoxazaborinine form of structure 22b (Scheme 1). How-
ever, it is notable that the bicyclic benzoxaborole 18 (Table 1)
shows a better level of inhibition than 22b, suggesting that the
position of the substituents on the structure 22b may be
responsible for its poor activity.
In contrast to the ortho-compounds, nearly all of the meta-
substituted 1 derivatives (23, 31, 32 and 34-36) displayed
improved inhibition against R39 relative to 1, with residual
activities in the range 2-18% at 1 mM (Table 2). Inhibition
for selected compounds was also observed in the presence
of 0.01% Triton-X (data not shown). The most active
compounds had IC50 values in the range 20-90 μM. The
2-methoxybenzoyl derivative 32 was the most potent inhibitor
(IC50 = 23 μM), a significant improvement from the parent
compound 1 (IC50 = 400 μM).
The results indicate that the number of atoms between the
aromatic side chain and carbonyl group is important; the two
arylacetamido derivatives 23 and 35 were less potent (IC50
≈
80-90 μM) than the derivatives, which have the aromatic
group attached directly to the carbonyl group (as in 31, 32,
and 34, IC50 ≈ 20-30 μM) or the phenoxyacetamido deriva-
tive 36 (IC50 = 28 μM). The observation that the derivative
containing the 2,6-dimethoxybenzoyl side chain 33 was a poor
inhibitor indicates that this side chain is too bulky to be
accommodated into the adjacent pocket in the active site.
This result is of interest given that the analogue with a single
methoxy substituent (32) was the most potent inhibitor,
demonstrating that one, but not two, methoxy group can be
accommodated. In the case where the m-amido group is
less bulky and lacked an aromatic group as in acetamide
37, a reduction in inhibition was observed (RA=50% at
1 mM) relative to 1, suggesting that the aromatic group is
important and that a lipophilic contact is likely to provide the
enhanced inhibition. The importance of the carboxylate
group on the 1 derivatives was shown by testing the methyl
ester of 23, as in 38, which displayednoinhibition. Someof the
a Symbols in the table have the following meaning: (/) mean values (
standard deviation over three replicate experiments; (†) values calcu-
lated based on three replicate experiments and adjusted to account for
standard error of regression of about 50% inhibition; (§) not determined.

6102 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 19
Scheme 2. Synthesis of m-Amido Derivatives of 1a
Inglis et al.
a Reagents and conditions: (a) SOCl2/MeOH, Δ, 2 h; (b) RCOCl/N,N-dimethylacetamide, room temp overnight; R = Me, Ac2O/THF, room temp
overnight; (c) bis(pinacolato)diboron, PdCl2 dppf, AcOK, toluene, 90 °C overnight; (d) KHF2(aq)/MeOH/THF, room temp 1 h; (e) TMSCl/CH3CN/
3
H2O, room temp 1 h; (f) LiOH H2O/THF/H2O, room temp overnight.
3
meta-substituted 1 derivatives also displayed activity against
PBP2xR6, PBP2x5204 (penicillin resistant), and PBP1b, from
Streptococcus pneumoniae (Supporting Information).
Lakaye et al.46 (R)-[2-(Benzoylamino)propionylsulfanyl]acetic
acid S2d was prepared as described by Adam et al.24 and Schwyzer
and Hurlimann.47
Testing of Compounds for Inhibition of R39, PBP2xR6, and
PBP2x5204. Inhibition experiments with R39, PBP2xR6, and
PBP2x5204 were performed by monitoring the degree of hydro-
lysis of the substrate S2d in microtiter 96-well plates using a
Power Wave microtiter plate reader, using procedures pre-
viously described.7 Enzyme residual activity (RA) was deter-
mined after preincubation of the PBPs in the presence of
potential inhibitors. The initial rate of hydrolysis of S2d
(1 mM) was determined in the presence of 5,50-dithiobis(2-
nitrobenzoic acid), ε[Δε] = 13 600 M-1 s-1 monitoring at 412 nm.
The rate of spontaneous hydrolysis of S2d in the presence of
the inhibitors was also determined in absence of the enzymes. All
experiments were performed in triplicate. Activity of PBPs in
the absence of inhibitors (RA=100%) was measured with six
replicates on each plate. Compounds were initially prepared
as 100 mM solutions in DMF and diluted to 10 mM in sodium
phosphate buffer (pH 7.0). In cases where compounds were
insoluble under these conditions, solutions of compound were
prepared at 1 mM in 10% DMF. For initial tests, potential
inhibitors were incubated with the enzymes at a concentration of
either 1.18 mM or 118 μM (depending on the solubility) in the
presence of 3.5 nM R39, 10 mM sodium phosphate buffer
(pH 7.2) with 100 mM NaCl, 100 mM D-alanine, and 0.01 mg/
mL BSA for 60 min at 25 °C (total volume = 127 μL). After
incubation, the substrate mixture [S2d and 5,50-dithiobis(2-
nitrobenzoic acid)] was added (23 μL) to give the final solution
(150 μL) [final concentration is 1 mM for S2d, 0.5 mM for 5,50-
dithiobis(2-nitrobenzoic acid), 1 mM for potential inhibitor (or
100 μM for low solubility compounds), and 1% v/v DMF], and
the RA was then measured at 412 nm. In the case of PBP2xR6,
0.09 μM enzyme was incubated in the presence of potential
inhibitors (1.18 mM or 118 μM) in 10 mM sodium phosphate
Conclusions
We have used the low molecular weight PBP, R39, from
Actinomadura sp. strain as a model system for the develop-
ment of inhibitors against higher molecular weight, more
clinically important and penicillin resistant PBPs. Following
the identification of 3-(dihydroxyboryl)benzoic acid 1 as an
R39 inhibitor (IC50 = 400 μM), derivatives of 1 with sub-
stitution in both the ortho- and meta-positions were prepared.
Ortho-substituted derivatives were poor inhibitors; however,
compounds with up to 17-fold improved potency were identi-
fied from the meta-substituted series [e.g., 3-(dihydroxy-
boryl)-5-(2-methoxybenzamido)benzoic acid 32, IC50 = 23
μM]. Some of the more potent derivatives displayed activity
against higher molecular weight PBPs. These compounds may
be useful for the development of even more potent boronic
acid based PBP-inhibitors with a broad spectrum of antibac-
terial activity. Given the potential of boronic acids to inhibit
serine β-lactamases,11-17 it is also possible that they could be
used in conjunction with a β-lactam both for protection from
β-lactamases and to extend the spectrum of antibacterial
activity.
Experimental Section
Biological Reagents. R39 D,D-peptidase from Actinomadura
was prepared and purified as described by Granier et al.44
PBP2x-5204, PBP2x-R6, and PBP1b from Streptococcus
pneumoniae were prepared as described by Carapito et al.45
Fluorescein-labeled ampicillin was prepared as described by

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 19 6103
buffer (pH 7.0) and 0.01 mg/mL BSA for 60 min at 25 °C, and for
PBP2x5204, 0.6 μM enzyme was incubated in the presence of
potential inhibitors (1.18 mM or 118 μM) in 10 mM sodium
phosphate buffer (pH 7.0), 70 mM D-alanine, and 0.01 mg/mL
BSA for 4 h at 25 °C, prior to the addition of the substrate
mixtures and measurement of the RA. When RA was <80%,
the RA measurement was repeated in the presence of 0.01%
Triton-X-100 v/v to identify possible false positive results. If the
RA for compounds was <50% at 1 mM (or 100 μM), the RA
was measured over a range of concentrations from which IC50
values were determined by performing a nonlinear regression
analysis using Sigma Plot (Systat software) and fitting the
data to the equation7 y = y0 þ (ab)/(b þ x) where y0 is the
activity with high concentration of inhibitor (y0 ≈ 0), y0 þ a is
the activity in the absence of the inhibitor (x = 0), and b =
IC50(a - y0)/(a þ y0); so for y0 ≈ 0, b ≈ IC50. The IC50 values are
adjusted to account for standard error of regression of about
50% inhibition.
Testing of Compounds for PBP1b Inhibition. PBP1b (0.7 μM)
was incubated with inhibitors (1 mM) in 10 mM sodium
phosphate (pH 7.0) for 60 min at 30 °C. Active PBP was then
counterlabeled with fluorescein-labeled ampicillin (10 μM) for
20 min. The reaction was stopped and analyzed by SDS-PAGE
followed by fluorescence visualization using a Molecular Imager
FX (Bio-Rad) and the program Quantity One (Bio-Rad). Back-
ground fluorescence was subtracted.
phosphino)ferrocene]palladium(II) chloride with dichloro-
methane (1:1) (PdCl2 dppf) were from Alfa Aesar. All solvents
3
used were HPLC grade. Where possible, reactions were moni-
tored by TLC, which was performed on precoated aluminum-
backed plates (Merck, silica 60 F254). Spots were visualized
using UV light (λ = 254 or 221 nm) and/or by staining with a
potassium permanganate/ethanol solution. Chromatography
was performed using a Biotage SP4 chromatography system,
using prepacked Biotage columns, or by filtration through silica
purchased from VWR (40-63 μm particle size). NMR spectra
were recorded on a Bruker DPX250 NMR spectrometer fitted
with a 5 mm broadband heteronuclear probe (1H, 250.13 MHz),
a Bruker AV400 spectrometer fitted with a 5 mm z-gradient
1H/13C/19F/31P quad-nucleus probe (1H, 400.133; 13C, 100.630
MHz; 19F, 376.507 MHz) or a Bruker AVII 500 spectrometer
fitted with a 5 mm 13C(1H) dual cryoprobe optimized for 13C
detection (1H, 500.303 MHz; 13C, 125.813 MHz). Chemical
shifts (δ) are given in ppm, and the multiplicities are given as
singlet (s), doublet (d), triplet (t), quartet (q), multiplet (m),
broad (br). Coupling constants J are given in Hz to the nearest
0.1 Hz. Spectra were recorded as solutions in CDCl3 (δH at
7.26 ppm, δC at 77.0 ppm), acetone-d6 (δH at 2.05 ppm, δC at
29.9 ppm), or DMSO-d6 (δH at 2.50 ppm, δC at 39.5 ppm), used
as internal references. Low resolution mass spectra were re-
corded on a LCT Premier, operating in electrospray ionization
(ESI) mode with medium-high (7000) resolution running Open-
Lynx. Samples were introduced by a PAL autosampler and were
prepared as methanol solutions, unless otherwise stated. Field
ionization (FI) mass spectra were recorded on a Micromass
GCT reflectron time-of-flight spectrometer in FI mode with
a temperature programmed solids probe inlet system (resolu-
tion = 5000 fwhm). High resolution mass spectra (HRMS) were
recorded under ESI conditions on a Bruker MicroTOF
(resolution = 10 000 fwhm). Unless otherwise stated, infrared
spectra were recorded from Nujol mulls between sodium chlor-
ide plates, using a Bruker Tensor 27 FT-IR spectrometer.
Elemental analyses were performed at London Metropolitan
University, London, U.K. The results of elemental analysis are
given in Table S-1. Melting points were determined using a Leica
Galen III hot-stage melting point apparatus and microscope.
The 13C NMR spectra of some of the compounds contained
some resonances that were observed as two peaks, likely due to
conformational isomers due to hindered rotation about amide
groups. This feature was observed for methyl esters 38 and
41-43, where the carbons adjacent to the amide functionality
were observed as two signals separated by ∼0.1 ppm in ratios
of ∼ 3:1; in these cases only the major peaks have been reported.
For 43, a 1:1 ratio was observed and both sets of peaks are
reported. Similarly, small additional peaks were observed for
the same carbon atoms of carboxylic acids 23 and 31-37 for
which only the major species is reported. The signals for the
quaternary carbons directly attached to the boron atom of all
the boronic acid derivatives were usually observed as small
broad humps in the baseline. The signal could be enhanced by
increasing the line broadening to ∼20 Hz in the exponential
multiplication window function, and this has been used to
identify and assign this signal where possible.
Docking Studies. Docking of boronic acids was performed
using the R39/nitrocefin structure (PDB code 1W8Y). Mono-
mer A was selected from an ensemble of four structures, and
the nitrocefin molecule was removed from the active site.
The PDB file was prepared for the docking experiment using
WHAT IF48 online resource (http://swift.cmbi.ru.nl/servers/html/
index.html) using the “prepare PDB file for docking programs”
feature. Ligand files were prepared using Marvin Sketch/View,
version 4.1.7.49 Boronic acids were prepared in tetrahedral form
by incorporation of an additional oxygen atom bonded to the
boron to facilitate covalent linkage. Explicit hydrogen atoms
were added before optimization of structures and writing output
files as MDL mol files. Docking experiments were performed
using GOLD,33,34 version 3.2. The ligand binding site was
centered about the nucleophilic serine residue O-γ (serine 49)
˚
atom with a radius of 15 A. Atom-to-atom covalent linkage was
used with connection of the serine 49 O-γ atom with the added
oxygen atom from the ligands (see above). Flip ring corners, flip
amide bonds, flip all planar R-NR1R2 (including ring NH-R,
including ring NR1NR2) were used under the fitness and
searching options. Otherwise all settings were as per default.
Orientations of docked ligands were inspected using Pymol,
version 0.99.50
Sources and Purities of Compounds for Inhibition Studies.
Boronic acids 2, 5, 6, 8, 9, and 19 were from Aldrich, and their
stated purities were all >95%. 10, 12, 14-16, and 18 were from
Alfa Aesar with stated purities of >95%. Boronic acid 13 was
from Maybridge with stated purity of 95%. Boronic acids 3, 4, 7,
11, and 17 were from Aldrich and were specified to contain
varying amounts of anhydrides and were tested for inhibition
without purification. Aldehyde 21 was from Aldrich, and its
purity was 97%. Compound 1 was from Aldrich and was
analytically pure as determined by C and H elemental analysis.
Trifluoromethylketone 20 and thiophenylboronic acid ana-
logue 30 were synthesized as described in the Supporting Infor-
mation and were >95% pure (HPLC). All other boronic acids
used for inhibition assays were prepared as described in the
example procedures below or in the Supporting Information
and were analytically pure as determined by C, H, and N
elemental analysis (Table S-1), with the exception of 22b which
was 90% pure (HPLC).
General Procedure for the Synthesis of Anilides. The aniline 39
or 44 (4.37 mmol) was stirred in N,N-dimethylacetamide
(10 mL) under an atmosphere of nitrogen and ice cooling
before addition of the appropriate acid chloride (8.73 mmol)
dropwise over 20 min. The mixture was stirred on ice for 1 h and
then at room temperature overnight. The solution was diluted
with ethyl acetate (70 mL), washed with 1 M HCl (3ꢀ40 mL),
0.2 M NaOH (3 ꢀ 40 mL), brine (40 mL), and dried (Na2SO4),
and the solvent was removed to afford the products which were
used as indicated.
General Synthetic Considerations. Compounds 39 and 40
were from Aldrich and used without further purification.
Bis(pinacolato)diboron and the complex of [1,10-bis(diphenyl-
Methyl 3-Bromo-5-(2-phenylacetamido)benzoate 45. Methyl
3-bromo-5-aminobenzoate 44 (1.00 g, 4.37 mmol) was treated
with phenylacetyl chloride (1.15 mL, 1.35 g, 8.73 mmol) in

6104 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 19
Inglis et al.
-
N,N-dimethylacetamide (10 mL) as described in the general
procedure. After workup, the residue was purified by filtration
through silica gel using 10% ethyl acetate/dichloromethane as
eluant to afford the title compound (1.35 g, 89%) as a white
waxy solid, mp 86-90 °C. Anal. (C16H14BrNO3) C, H, N. IR
(nujol) ν/cm-1: 3284 (NH), 3246 (NH), 3184 (NH), 3092 (NH),
1726 (CO2), 1655 (CON). 1H NMR (500 MHz, CDCl3) δ: 3.76
(2H, s, CH2), 3.89 (3H, s, CH3), 7.30 (1H, br s, NH), 7.32-7.43
(5H, m, H20-H60), 7.77 (1H, br t, J = 1.5 Hz, H6), 7.88 (1H, br t,
J = 1.5 Hz, H2), 8.11 (1H, m, H4). 13C NMR (125 MHz,
CDCl3) δ: 44.7 (CH2), 52.2 (CH3), 119.1 (C6), 122.8 (C3), 126.8
(C4), 128.0 (C40), 128.2 (C2), 129.4 ([C20 and C60] or [C30 and
C50]), 129.5 ([C30 and C50] or [C20 and C60]), 132.1 (C1), 133.7
(C10), 138.9 (C5), 165.4 (CO2), 169.3 (CON). m/z (ESI, negative
ion) 697 ({2M} - H, 85%,), 695 ({2M} - H, 100), 693 ({2M} -
H, 80), 349 (30), 348 (M - H[81Br], 85), 347 (30), 346 (M -
H[79Br], 85).
found, 50.38. HRMS (ESI, negative ion) C16H1411BF3NO3
(M - Kþ) requires 336.1024. Found 336.1016. IR (Nujol) ν/
cm-1: 3356 (NH), 1703 (CO2), 1698 (CON). 1H NMR (500
MHz, DMSO-d6) δ: 3.62 (2H, s, CH2), 3.80 (3H, s, CH3),
7.22-7.26 (1H, m, H40), 7.31-7.36 (4H, m, H20, H30, H50,
H60), 7.60 (1H, br d, J = 2.0 Hz, H6), 7.69 (1H, br s, H2),
8.19 (1H, t, J = 2.0 Hz, H4), 10.06 (1H, br s, NH). 13C NMR
(125 MHz, DMSO-d6) δ: 43.4 (CH2), 51.6 (CH3), 117.0 (C4),
126.4 (C40), 127.3 (C6), 127.7 (C2), 127.8 (C3), 128.3 ([C20 and
C60] or [C30 and C50]), 129.0 ([C30 and C50] or [C20 and C60]),
136.2 (C10), 137.7 (C5), 167.3 (CO2), 168.8 (CON), C1 not
observed. 19F NMR (DMSO-d6, 366 MHz) δ: -139.8 (3F, br
s, BF3). m/z (ESI, negative ion) 337 (60%), 336 (M - K[11B]),
100), 335 (M - K[10B], 80).
General Procedure for the Preparation of Boronic Acids from
Trifluoroborate Salts. A stirred solution of the trifluoroborate
salt (1.90 mmol) in acetonitrile and water as indicated (total
volume of ∼35 mL) was treated with TMSCl (7.25 mmol) for
1 h. The resulting mixture was treated with saturated NaHCO3
(4 mL) before drying (Na2SO4) and concentrating to afford the
product.
General Procedure for the Preparation of Pinacol Boronate
Esters from Aryl Bromides. The aryl bromide (3.24 mmol) was
treated with bis(pinacolato)diboron (3.57 mmol), [1,10-bis-
(diphenylphosphino)ferrocene]palladium(II) chloride, complex
with dichloromethane (1:1), (hereafter referred to as PdCl2 dppf)
Methyl 3-(Dihydroxyboryl)-5-(2-phenylacetamido)benzoate
38. A solution of potassium trifluoro{3-(methoxycarbonyl)-
5-[2-phenylacetamido]phenyl}borate 61 (0.440 g, 1.17 mmol)
in acetonitrile (18 mL) and water (5 mL) was treated with
TMSCl (0.560 mL, 0.480 g, 4.44 mmol) for 1 h, followed by
addition of NaHCO3 solution (3.5 mL), as described in the
general procedure. After workup the title compound (0.269 g,
73%) was obtained as a white solid, mp 273-276 °C. Anal.
(C16H16BNO5) C, H, N. IR (Nujol) ν/cm-1: 3404 (NH or OH),
3340 (NH or OH), 3262 (NH or OH), 1712 (CO2), 1652 (CON).
1H NMR (500 MHz, acetone-d6) δ: 3.76 (2H, s, CH2), 3.88 (3H,
s, CH3), 7.25-7.28 (1H, m, H40), 7.33-7.36 (2H, m, [H20 and
H60] or [H30 and H50]), 7.42-7.43 (2H, m, [H30 and H50] or [H20
and H60]), 7.45 (2H, s, 2 ꢀ BOH), 8.23 (1H, br s, H2), 8.25 (1H,
br s, H4), 8.47 (1H, br s, H6), 9.44 (1H, br s, NH). 13C NMR (125
MHz, acetone-d6) δ: 44.8 (CH2), 52.3 (CH3), 122.8 (C6), 127.5
(C40), 129.2 ([C20 and C60] or [C30 and C50]), 130.2 ([C30 and C50]
or [C20 and C60]), 130.3 (C4), 130.9 (C2), 131.0 (C1), 135.6 (v br,
C3), 136.8 (C10), 140.0 (C5), 167.4 (CO2), 170.0 (CON). m/z
(ESI, negative ion) 326 (90%), 313 (20), 312 (M - H[11B], 100),
311 (M - H[10B], 50).
3
(0.1 mmol) and potassium acetate (9.70 mmol) in dry degassed
toluene (32 mL) with stirring at 95 °C overnight under an atmo-
sphere of nitrogen. After cooling, the mixture was concentrated,
and the residue was resuspended in ethyl acetate (70 mL), washed
with water (1ꢀ 40 mL) and brine (1ꢀ 40 mL), anddried(Na2SO4),
and the solvent was removed. The residue (a dark-brown solid) was
purified by chromatography on silica gel as indicated to afford the
pure product.
Methyl 3-(4,4,5,5-Tetramethyl-1,3,2-dioxaborolan-2-yl)-5-(2-
phenylacetamido)benzoate 53. Methyl 3-bromo-5-(2-phenylace-
tamido)benzoate 45 (0.430 g, 1.24 mmol) was treated with
bis(pinacolato)diboron (0.347 g, 1.36 mmol), PdCl2 dppf
3
(0.031 g, 0.037 mmol), and potassium acetate (0.364 g, 3.71
mmol) in toluene (13 mL) as described in the general procedure.
After workup the residue was purified by filtration through
silica gel using 20% ethyl acetate/dichloromethane as eluant to
afford the title compound (0.330 g, 67%) as an off-white solid,
mp 132-137 °C. Anal. (C22H26BNO5) C, H, N. IR (nujol) ν/
cm-1: 3300 (NH), 1724 (CO2), 1662 (CON). 1H NMR (500
MHz, CDCl3) δ: 1.33 (12H, s, 4 ꢀ CH3[pinacolyl]), 3.76 (2H, s,
CH2), 3.89 (3H, s, CH3O), 7.18 (1H, br s, NH), 7.33-7.43 (5H,
m, H20-H60), 7.87 (1H, br d, J = 1.5 Hz, H4), 8.20 (1H, bs s, H2),
8.31 (1H, br s, H6). 13C NMR (125 MHz, CDCl3) δ: 24.8 (4 ꢀ
CH3[pinacolyl]), 44.8 (CH2), 52.1 (CH3), 84.2 (C400 and C500),
123.7 (C6), 127.8 (C40), 129.4 ([C20 and C60] or [C30 and C50]),
129.6 ([C30 and C50] or [C20 and C60]), 130.1 (C4), 130.4 (v br,
C3), 130.5 (C10), 131.9 (C2), 134.1 (C10), 137.3 (C5), 166.7
(CO2), 169.1 (CON). m/z (ESI, negative ion) 822 ({2M} þ
MeO, 75%), 821 ({2M} þ MeO, 100), 820 ({2M} þ MeO, 70),
427 (30), 426 (M þ MeO[11B], 90), 425 (M þ MeO[10B], 40), 395
(15), 394 (M - H[11B], 50), 393 (M - H[10B], 12).
General Procedure for Hydrolysis of Methyl Esters to Car-
boxylic Acids. A mixture of the methyl ester (0.82 mmol) and
LiOH H2O (2.86 mmol) was stirred overnight in THF (10 mL)
3
and water (7.5 mL). The mixture was concentrated to ∼2 mL
and the resulting solution diluted with additional water (1 mL).
This mixture was extracted with ethyl acetate (2 ꢀ 2 mL), and
the aqueous layer was acidified with 1 M HCl until substantial
precipitation was observed, with assistance by scratching if
necessary. The precipitate was collected and dried to afford
the carboxylic acids. The crude products were accompanied by
small amounts of apparent impurities thought to be correspond-
ing anhydrides or semianhydrides. In all cases, the analytically
pure free boronic acids were obtained after recrystallization
from water/methanol mixtures as indicated.
General Procedure for Conversion of Pinacol Esters to Tri-
fluoroborate Salts. The pinacol ester (2.32 mmol) is dissolved
in a mixture of methanol (15 mL) and THF (10 mL) with stirr-
ing before addition of 4.5 M KHF2(aq) (20.7 mmol). The
resulting mixture was stirred for 1 h before concentrating to
dryness. The residue, a white solid, was collected by filtration,
washed several times with hot diethyl ether (total of ∼100 mL)
followed by water (∼50 mL), and dried to afford the trifluoro-
borate salt.
Potassium Trifluoro{3-(methoxycarbonyl)-5-[2-phenylaceta-
mido]phenyl}borate 61. A solution of methyl 3-(4,4,5,5-tetra-
methyl-1,3,2-dioxaborolan-2-yl)-5-(2-phenylacetamido)benzo-
ate 53 (0.400 g, 1.01 mmol) in methanol (6 mL) and THF (2 mL)
was treated with 4.5 M KHF2(aq) (2.0 mL, 9.00 mmol) as
described in the general procedure. After workup, the title
compound (0.292 g, 77%) was isolated as a white solid, mp
278-284 °C. Anal. (C16H14BF3KNO3) H, N. C: calcd 51.22;
3-(Dihydroxyboryl)-5-(2-phenylacetamido)benzoic Acid 23.
Methyl 3-(dihydroxyboryl)-5-(2-phenylacetamido)benzoate 38
(0.210 g, 0.67 mmol) was treated with LiOH H2O (0.099 g, 2.35
3
mmol) in THF (9 mL) and water (7 mL) as described in the
general procedure. After workup the title compound (0.147 g,
74%) was obtained as a white solid. A portion was recrystallized
from 30% methanol/water solution (3ꢀ) to obtain an analytical
sample as a white fluffy solid, mp 200-210 °C. Anal.
(C15H14BNO5) C, H, N. IR (Nujol) ν/cm-1: 3273 (NH and
OH), 1698 (CO2), 1631 (CON). 1H NMR (500 MHz, acetone-d6)
δ: 3.75 (2H, s, CH2), 7.25-7.28 (1H, m, H40), 7.33-7.36 (2H, m,
[H20 and H60] or [H30 and H50]), 7.40-7.44 (4H, m, [H30 and H50]
or [H20 and H60] and 2 ꢀ BOH), 8.26 (2H, br s, H2 and H4), 8.48
(1H, t, J = 2.0 Hz, H6), 9.42 (1H, br s, NH), 11.20 (1H, br s,

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 19 6105
CO2H). 13C NMR (125 MHz, acetone-d6) δ: 44.8 (CH2), 123.2
(C6), 127.5 (C40), 129.2 ([C20 and C60] or [C30 and C50]), 130.1
([C30 and C50] or [C20 and C60]), 130.3 (C4), 131.2 (C2), 131.3
(C1), 135.7 (v br, C3), 136.9 (C10), 139.9 (C5), 167.8 (CO2), 170.0
(CON). m/z (ESI, negative ion) 313 (20%), 312 (M þ MeO -
OH - H[11B], 60), 311 (M þ MeO - OH - H[10B], 25), 299 (40),
298 (M - H[11B], 100), 297 (M - H[10B], 50).
(17) Strynadka, N. C. J.; Martin, R.; Jensen, S. E.; Gold, M.; Jones, J. B.
Structure-based design of a potent transition state analogue for
TEM-1 β-lactamase. Nat. Struct. Biol. 1996, 3, 688–695.
(18) Pechenov, A.; Stefanova, M. E.; Nicholas, R. A.; Peddi, S.;
Gutheil, W. G. Potential transition state analogue inhibitors for
the penicillin-binding proteins. Biochemistry 2003, 42, 579–588.
(19) Nicola, G.; Peddi, S.; Stefanova, M.; Nicholas, R. A.; Gutheil, W.
G.; Davies, C. Crystal structure of Escherichia coli penicillin-
binding protein 5 bound to a tripeptide boronic acid inhibitor: a
role for Ser-110 in deacylation. Biochemistry 2005, 44, 8207–8217.
(20) Stefanova, M. E.; Tomberg, J.; Davies, C.; Nicholas, R. A.;
Gutheil, W. G. Overexpression and enzymatic characterization of
Neisseria gonorrhoeae penicillin-binding protein 4. Eur. J. Biochem.
2004, 271, 23–32.
(21) Granier, B.; Duez, C.; Lepage, S.; Englebert, S.; Dusart, J.;
Dideberg, O.; Van Beeumen, J.; Frere, J. M.; Ghuysen, J. M.
Primary and predicted secondary structures of the Actinomadura
R39 extracellular DD-peptidase, a penicillin-binding protein (PBP)
related to the Escherichia coli PBP4. Biochem. J. 1992, 282, 781–
788.
Acknowledgment. We thank the European Union (European
Community Sixth Framework Programme) via the EUR-
INTAFAR project for the financial support for this research.
N.T. was supported by Grant ARC 03/08-297.
Supporting Information Available: Additional synthetic pro-
cedures, elemental analysis results for compounds, NMR spec-
tra of the solution studies of aromatic boronic acids, and
inhibition data from testing compounds against other PBPs.
This material is available free of charge via the Internet at http://
pubs.acs.org.
(22) Pedersen, L. B.; Murray, T.; Popham, D. L.; Setlow, P. Character-
ization of dacC, which encodes a new low-molecular-weight peni-
cillin-binding protein in Bacillus subtilis. J. Bacteriol. 1998, 180,
4967–4973.
(23) Stefanova, M. E.; Tomberg, J.; Olesky, M.; Hoeltje, J.-V.; Gutheil,
W. G.; Nicholas, R. A. Neisseria gonorrhoeae penicillin-binding
protein 3 exhibits exceptionally high carboxypeptidase and β-
lactam binding activities. Biochemistry 2003, 42, 14614–14625.
(24) Adam, M.; Damblon, C.; Plaitin, B.; Christiaens, L.; Frere, J. M.
Chromogenic depsipeptide substrates for β-lactamases and peni-
cillin-sensitive DD-peptidases. Biochem. J. 1990, 270, 525–529.
(25) Feng, B. Y.; Shoichet, B. K. A detergent-based assay for the
detection of promiscuous inhibitors. Nat. Protoc. 2006, 1, 550–553.
(26) Shoichet, B. K. Screening in a spirit haunted world. Drug Discovery
Today 2006, 11, 607–615.
(27) Leung, D.; Abbenante, G.; Fairlie, D. P. Protease inhibitors:
current status and future prospects. J. Med. Chem. 2000, 43, 305–
341.
(28) Peet, N. P.; Burkhart, J. P.; Angelastro, M. R.; Giroux, E. L.;
Mehdi, S.; Bey, P.; Kolb, M.; Neises, B.; Schirlin, D. Synthesis of
peptidyl fluoromethyl ketones and peptidyl r-keto esters as in-
hibitors of porcine pancreatic elastase, human neutrophil elastase,
and rat and human neutrophil cathepsin-G. J. Med. Chem. 1990,
33, 394–407.
References
(1) Macheboeuf, P.; Contreras-Martel, C.; Job, V.; Dideberg, O.;
Dessen, A. Penicillin binding proteins: key players in bacterial cell
cycle and drug resistance processes. FEMS Microbiol. Rev. 2006,
30, 673–691.
(2) Sauvage, E.; Kerff, F.; Terrak, M.; Ayala, J. A.; Charlier, P. The
penicillin-binding proteins: structure and role in peptidoglycan
biosynthesis. FEMS Microbiol. Rev. 2008, 32, 234–258.
(3) Gordon, E.; Mouz, N.; Duee, E.; Dideberg, O. The crystal struc-
ture of the penicillin-binding protein 2x from Streptococcus pneu-
moniae and its acyl-enzyme form: implication in drug resistance.
J. Mol. Biol. 2000, 299, 477–485.
(4) Baldwin, J. E.; Lowe, C.; Schofield, C. J.; Lee, E. A γ-lactam analog
of penems possessing antibacterial activity. Tetrahedron Lett. 1986,
27, 3461–3464.
(5) Boyd, D. B.; Elzey, T. K.; Hatfield, L. D.; Kinnick, M. D.; Morin,
J. M. γ-Lactam analogs of the penems. Tetrahedron Lett. 1986, 27,
3453–3456.
(6) Harada, S.; Tsubotani, S.; Hida, T.; Ono, H.; Okazaki, H. Struc-
ture of lactivicin, an antibiotic having a new nucleus and similar
biological activities to β-lactam antibiotics. Tetrahedron Lett. 1986,
27, 6229–6232.
(7) Macheboeuf, P.; Fischer, D. S.; Brown, T., Jr.; Zervosen, A.;
Luxen, A.; Joris, B.; Dessen, A.; Schofield, C. J. Structural and
mechanistic basis of penicillin-binding protein inhibition by lacti-
vicins. Nat. Chem. Biol. 2007, 3, 565–569.
(8) Nozaki, Y.; Katayama, N.; Harada, S.; Ono, H.; Okazaki, H.
Lactivicin, a naturally occurring non-β-lactam antibiotic having
β-lactam-like action: biological activities and mode of action.
J. Antibiot. (Tokyo) 1989, 42, 84–93.
(9) Nozaki, Y.; Katayama, N.; Ono, H.; Tsubotani, S.; Harada, S.;
Okazaki, H.; Nakao, Y. Binding of a non-β-lactam antibiotic to
penicillin-binding proteins. Nature 1987, 325, 179–180.
(10) Yang, W.; Gao, X.; Wang, B. Boronic acid compounds as potential
pharmaceutical agents. Med. Res. Rev. 2003, 23, 346–368.
(11) Venturelli, A.; Tondi, D.; Cancian, L.; Morandi, F.; Cannazza, G.;
Segatore, B.; Prati, F.; Amicosante, G.; Shoichet, B. K.; Costi, M.
P. Optimizing cell permeation of an antibiotic resistance inhibitor
for improved efficacy. J. Med. Chem. 2007, 50, 5644–5654.
(12) Morandi, F.; Caselli, E.; Morandi, S.; Focia, P. J.; Blazquez, J.;
Shoichet, B. K.; Prati, F. Nanomolar inhibitors of AmpC β-
lactamase. J. Am. Chem. Soc. 2003, 125, 685–695.
(29) Walker, B.; Lynas, J. F. Strategies for the inhibition of serine
proteases. Cell. Mol. Life Sci. 2001, 58, 596–624.
(30) Kettner, C. A.; Shenvi, A. B. Inhibition of the serine proteases
leukocyte elastase, pancreatic elastase, cathepsin G, and chymo-
trypsin by peptide boronic acids. J. Biol. Chem. 1984, 259, 15106–
15114.
(31) Takahashi, L. H.; Radhakrishnan, R.; Rosenfield, R. E.; Meyer, E.
F. Crystallographic analysis of the inhibition of porcine pancreatic
elastase by a peptidyl boronic acid. Structure of a reaction inter-
mediate. Biochemistry 1989, 28, 7610–7617.
(32) Sauvage, E.; Herman, R.; Petrella, S.; Duez, C.; Bouillenne, F.;
Frere, J.-M.; Charlier, P. Crystal structure of the Actinomadura
R39 DD-peptidase reveals new domains in penicillin-binding pro-
teins. J. Biol. Chem. 2005, 280, 31249–31256.
(33) Jones, G.; Willett, P.; Glen, R. C. Molecular recognition of
receptor sites using a genetic algorithm with a description of
desolvation. J. Mol. Biol. 1995, 245, 43–53.
(34) Verdonk, M. L.; Cole, J. C.; Hartshorn, M. J.; Murray, C. W.;
Taylor, R. D. Improved protein-ligand docking using GOLD.
Proteins 2003, 52, 609–623.
(35) Han, J. W.; Castro, J. C.; Burgess, K. Microwave-assisted functio-
nalization of bromo-fluorescein and bromorhodamine derivatives.
Tetrahedron Lett. 2003, 44, 9359–9362.
(36) Ni, A.; France, J. E.; Davies, H. M. Diversity synthesis using the
complimentary reactivity of rhodium(II)- and palladium(II)-cata-
lyzed reactions. J. Org. Chem. 2006, 71, 5594–5598.
(13) Weston, G. S.; Blazquez, J.; Baquero, F.; Shoichet, B. K. Structure-
based enhancement of boronic acid-based inhibitors of AmpC β-
lactamase. J. Med. Chem. 1998, 41, 4577–4586.
(37) Yuan, S. C.; Chen, H. B.; Zhang, Y.; Pei, J. Rigid linear and star-
shaped pi-conjugated 2,20:60,200-terpyridine ligands with blue emis-
sion. Org. Lett. 2006, 8, 5701–5704.
(38) Ahmed, V.; Liu, Y.; Silvestro, C.; Taylor, S. D. Boronic acids as
inhibitors of steroid sulfatase. Bioorg. Med. Chem. 2006, 14, 8564–
8573.
(39) Tzschucke, C. C.; Murphy, J. M.; Hartwig, J. F. Arenes to anilines
and aryl ethers by sequential iridium-catalyzed borylation and
copper-catalyzed coupling. Org. Lett. 2007, 9, 761–764.
(40) Yuen, A. K. L.; Hutton, C. A. Deprotection of pinacolyl boronate
esters via hydrolysis of intermediate potassium trifluoroborates.
Tetrahedron Lett. 2005, 46, 7899–7903.
(14) Morandi, S.; Morandi, F.; Caselli, E.; Shoichet, B. K.; Prati, F.
Structure-based optimization of cephalothin-analogue boronic
acids as β-lactamase inhibitors. Bioorg. Med. Chem. 2008, 16,
1195–1205.
(15) Ness, S.; Martin, R.; Kindler, A. M.; Paetzel, M.; Gold, M.; Jensen,
S. E.; Jones, J. B.; Strynadka, N. C. J. Structure-based design
guides the improved efficacy of deacylation transition state analo-
gue inhibitors of TEM-1 β-lactamase. Biochemistry 2000, 39, 5312–
5321.
(16) Wang, X.; Minasov, G.; Blazquez, J.; Caselli, E.; Prati, F.; Shoi-
chet, B. K. Recognition and resistance in TEM β-lactamase.
Biochemistry 2003, 42, 8434–8444.

6106 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 19
Inglis et al.
(41) Soloway, A. H. Synthesis of aromatic diboronic acids. J. Am.
Chem. Soc. 1960, 82, 2442–2444.
(42) Groziak, M. P.; Ganguly, A. D.; Robinson, P. D. Boron hetero-
cycles bearing a peripheral resemblance to naturally-occurring
purines: design, syntheses, structures, and properties. J. Am. Chem.
Soc. 1994, 116, 7597–7605.
(43) Yang, X.; Lee, M. C.; Sartain, F.; Pan, X.; Lowe, C. R. Designed
boronate ligands for glucose-selective holographic sensors.
Chem.;Eur. J. 2006, 12, 8491–8497.
(45) Carapito, R.; Chesnel, L.; Vernet, T.; Zapun, A. Pneumococcal β-
lactam resistance due to a conformational change in penicillin-
binding protein 2x. J. Biol. Chem. 2006, 281, 1771–1777.
(46) Lakaye, B.; Damblon, C.; Jamin, M.; Galleni, M.; Lepage, S.; Joris,
B.; Marchand-Brynaert, J.; Frydrych, C.; Frere, J.-M. Synthesis,
purification and kinetic properties of fluorescein-labeled penicil-
lins. Biochem. J. 1994, 300, 141–145.
(47) Schwyzer, R. Preparation of thio esters of amino acids and of
peptides. Helv. Chim. Acta 1954, 37, 647–649.
(44) Granier, B.; Jamin, M.; Adam, M.; Galleni, M.; Lakaye, B.; Zorzi,
W.; Grandchamps, J.; Wilkin, J.-M.; Fraipont, C.; Joris, B.; Duez,
C.; Nguyendisteche, M.; Coyette, J.; Leyhbouille, M.; Dusart, J.;
Christiaens, L.; Frere, J. M.; Ghuysen, J. M. Serine-type D-Ala-D-
Ala peptidases and penicillin-binding proteins. Methods Enzymol.
1994, 244, 249–266.
(48) Vriend, G. WHAT IF: a molecular modeling and drug design
program. J. Mol. Graphics 1990, 8, 52–56.
(49) Marvin Sketch/Marvin View, version 4.1.7; ChemAxon: Budapest,
Hungary, 2007; http://www.chemaxon.com.
(50) DeLano, W. L. The PyMOL Molecular Graphics System; DeLano
Scientific LLC: Palo Alto, CA, 2002, http://www.pymol.org.