Synthesis and evaluation of boronic acids as inhibitors of Penicillin Binding Proteins of classes A, B and C

Article abstract of DOI:10.1016/j.bmc.2012.04.018

In response to the widespread use of β-lactam antibiotics bacteria have evolved drug resistance mechanisms that include the production of resistant Penicillin Binding Proteins (PBPs). Boronic acids are potent β-lactamase inhibitors and have been shown to display some specificity for soluble transpeptidases and PBPs, but their potential as inhibitors of the latter enzymes is yet to be widely explored. Recently, a (2,6-dimethoxybenzamido) methylboronic acid was identified as being a potent inhibitor of Actinomadura sp. R39 transpeptidase (IC₅₀: 1.3 μM). In this work, we synthesized and studied the potential of a number of acylaminomethylboronic acids as inhibitors of PBPs from different classes. Several derivatives inhibited PBPs of classes A, B and C from penicillin sensitive strains. The (2-nitrobenzamido)methylboronic acid was identified as a good inhibitor of a class A PBP (PBP1b from Streptococcus pneumoniae, IC₅₀ = 26 μM), a class B PBP (PBP2xR6 from Streptococcus pneumoniae, IC₅₀ = 138 μM) and a class C PBP (R39 from Actinomadura sp., IC₅₀ = 0.6 μM). This work opens new avenues towards the development of molecules that inhibit PBPs, and eventually display bactericidal effects, on distinct bacterial species.

Full text of DOI:10.1016/j.bmc.2012.04.018

Bioorganic & Medicinal Chemistry 20 (2012) 3915–3924
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis and evaluation of boronic acids as inhibitors of Penicillin
Binding Proteins of classes A, B and C
Astrid Zervosen a, , André Bouillez b, Alexandre Herman a, Ana Amoroso c, Bernard Joris c, Eric Sauvage c,
⇑
Paulette Charlier c, André Luxen a
a Centre de Recherches du Cyclotron, B30, Université de Liège, Sart-Tilman, B-4000 Liège, Belgium
b Laboratoire de Chimie Organique de Synthèse, Département de Chimie, B6a, Université de Liège, Sart-Tilman, B-4000 Liège, Belgium
c Centre d’Ingénierie des Protéines, Institut de Chimie, B6a, Université de Liège, Sart-Tilman, B-4000 Liège, Belgium
a r t i c l e i n f o
a b s t r a c t
Article history:
In response to the widespread use of b-lactam antibiotics bacteria have evolved drug resistance mecha-
nisms that include the production of resistant Penicillin Binding Proteins (PBPs). Boronic acids are potent
b-lactamase inhibitors and have been shown to display some specificity for soluble transpeptidases and
PBPs, but their potential as inhibitors of the latter enzymes is yet to be widely explored. Recently, a (2,6-
dimethoxybenzamido)methylboronic acid was identified as being a potent inhibitor of Actinomadura sp.
Received 7 February 2012
Revised 5 April 2012
Accepted 7 April 2012
Available online 16 April 2012
R39 transpeptidase (IC50: 1.3 lM). In this work, we synthesized and studied the potential of a number of
acylaminomethylboronic acids as inhibitors of PBPs from different classes. Several derivatives inhibited
PBPs of classes A, B and C from penicillin sensitive strains. The (2-nitrobenzamido)methylboronic acid
Keywords:
Penicillin Binding Proteins
Boronic acids
Antibiotics
Antibiotic-resistance
Transpeptidase-inhibition
was identified as a good inhibitor of a class A PBP (PBP1b from Streptococcus pneumoniae, IC50 = 26
a class B PBP (PBP2xR6 from Streptococcus pneumoniae, IC50 = 138 M) and a class C PBP (R39 from Acti-
nomadura sp., IC50 = 0.6 M). This work opens new avenues towards the development of molecules that
inhibit PBPs, and eventually display bactericidal effects, on distinct bacterial species.
Ó 2012 Elsevier Ltd. All rights reserved.
lM),
l
l
1. Introduction
as do those of class A PBPs.4 Class C PBPs are low-molecular mass
PBPs. R39, a soluble D,D-peptidase from Actinomadura, is a class C
type-4 PBP2,5 with an endopeptidase activity. The TP domains
mostly catalyze the cross-linking of stem peptides via an acyl-en-
zyme intermediate formed by nucleophilic attack of a serine resi-
The widespread use of b-lactam antibiotics resulted in the
worldwide appearance of drug-resistant strains. Bacteria have
developed resistance to b-lactams by three main mechanisms:
the production of b-lactamases that catalyze the hydrolysis of b-
lactams, the production of low-affinity, drug resistant Penicillin
Binding Proteins (PBPs) and the overexpression of resistant PBPs.
PBPs are generally membrane-associated proteins. The PBP family
is divided in three classes.1,2 The high-molecular mass PBPs of clas-
ses A and B contain two catalytic domains. Class A PBPs such as
PBP1b from Streptococcus pneumoniae catalyze the last two steps
of the biosynthesis of peptidoglycan (PG): the glycosyltransfer
(GT, the formation of glycan chain (GlcNAc-MurNAc-peptide-)n)
and the transpeptidation (TP, cross-linking of stem peptides).3
The N-terminal domains of class B PBPs such as PBP2xR6 from
Streptococcus pneumoniae and PBP2x5204 from a penicillin resis-
tant pneumococcal strain are believed to play a role in cell mor-
phogenesis, while their C-terminal domains exhibit a TP activity
due on the amide carbonyl group of the penultimate D-Ala of the
stem peptide. The TP domains of PBPs are the main targets of b-lac-
tam antibiotics, due to the structural similarity between penicillin
and the stem-peptide. b-Lactams are suicide substrates of the TP
domains of PBPs forming a stable, covalent adduct with the serine
residue of the active site (Fig. 1).
Partly in response to the emergence and spread of b-lactamases,
many b-lactam derivatives have been developed. For example li-
braries of penams, which preserve the b-lactam core of the drug
but explore diverse R functionalities on the carboxyamido side
chain at the C6 position of the penicillin ring, have been synthe-
sized (Fig. 2). Different R side chains confer different antibacterial
activities, levels of resistance to b-lactamases and kinetics of PBP
inhibition3,6–10 (Fig. 2, Table 1). Another approach has been the
development of non-b-lactam inhibitors, with the objective of
attempting to stall the development of b-lactam resistance. Vari-
ous non-b-lactam inhibitors of PBPs are described in the litera-
ture.3,11–13
Abbreviations: GT, glycosyltransfer; TP, transpeptidation; PG, peptide glycan;
PBP, Penicillin Binding Proteins; RA, residual activity.
⇑
Corresponding author. Tel.: +32 4 3663383; fax: +32 4 3662946.
E-mail address: azervosen@ulg.ac.be (A. Zervosen).
0968-0896/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2012.04.018

3916
A. Zervosen et al. / Bioorg. Med. Chem. 20 (2012) 3915–3924
Table 1
E-Ser-OH
R
O
OH
Acylation rate constants k2/K of different PBPs by penicillin
G as found in the
S
literature3,6–10
PBP
R
B
N
OH
Benzylpenicillin
20,0006
110.000,3 58 0008
300.0007
1043
-
PBP1b
PBP2xR6
R39
CO2
PBP2x5204
PBP2a
PBP5fm
179
2010
R
S
OH
R
currently in clinical use.15 Furthermore, boronic acids have been
shown to inhibit b-lactamases like AmpC and TEM16–19 as well as
PBPs.20–25 Boronic acids form reversible covalent adducts, which
mimic the tetrahedral intermediates in the PBP-catalyzed acylation
reactions22 (Fig. 1) and are thus interesting starting points for the
development of non-b-lactam inhibitors of PBPs. Crystal structures
of adducts formed by boronic acid analogs and R39 described here
(4, 11, 14, 19, Table 2) have already been solved and an unexpected
tricovalent binding mode of these molecules within the active site
of R39 was observed.25 In addition, crystal structures of the pneu-
mococcal PBP1b complexed to different alkyl and aryl boronic acid
analogs (4, 7, 11, 13, 14, Table 2) have been used in combination
with other structures of ethylboronic acids in the structure-guided
design of PBP inhibitors that overcome b-lactam resistance in
Staphylococcus aureus (MRSA).20 (2,6-Dimethoxybenzamido)meth-
ylboronic acid is an inhibitor of b-lactamases from Bacillus cereus
and Pseudomonas aeruginosa.26 This boronic acid was identified
as a potent inhibitor of R39 and was used in this work as a lead-
structure for the synthesis of various acylaminomethylboronic
acids. We show that a number of methylboronic acid analogs
inhibit PBPs from all three classes and thus constitute promising
leads for the development of new inhibitors of peptidoglycan
biosynthesis.
-O
N
B
OH
H
E-Ser-O
E-Ser-O
-
CO2
tetrahedral intermediates
R
O
S
"Acyl-enzyme adduct"
Deacylation
E-Ser-O
HN
-
CO2
H2O
PBP: slow
-lactamase: fast
β
R
O
S
OH HN
-
CO2
Figure 1. b-Lactams and boronic acids form tetrahedral intermediates with the
serine residue of the active site of PBPs and b-lactamases (E).
H
N
R
2. Results and discussion
S
O
N
2.1. Identification of acylaminomethylboronic acids as R39
inhibitors
O
-
CO2
COOH
OCH3
The overall fold of the transpeptidase domains of all PBPs is very
similar to that of class A b-lactamases.1,2 The active site is at the
interface of two subdomains and is defined by three motifs com-
mon to all PBPs and serine-b-lactamases as shown by Sauvage5
for the R39 D,D-peptidase and the class A b-lactamase from Bacillus
licheniformis. PBPs form long-lived acylenzymes with b-lactams. A
loop in the active site of class A b-lactamases bearing an asparagine
and a glutamic acid is responsible for the rapid deacylation of most
b-lactam antibiotics. Because of the structural similarity with b-
lactamases it is not surprising that some powerful b-lactamase
inhibitors like 6-b-iodopenicillanate5 have been shown to inacti-
vate PBPs however rather poorly. (2,6-Dimethoxybenzamido)-
methylboronic acid 4 was identified as a good inhibitor of R39
OCH3
Carbenicillin
3 600M-1s-1
Benzylpenicillin
58 000M-1s-1
Methicillin
1 400M-1s-1
O
N
S
N
COOH
O
O
(IC50: 1.3 lM, Table 2) by an initial screening of a library containing
CH3
Ticarcillin
several b-lactamase inhibitors. Two fragments of molecule 4, the
aminomethylboronic acid 5 and 2-(2,6-dimethoxybenzamido)ace-
tic acid 6, were not or only poor inhibitors at a concentration of
1 mM (Fig. 3) underlining the importance of the presence of the
2,6-dimethoxybenzoyl residue, the side chain of methicillin
(Fig. 2), and of the aminomethylboronic acid function of 4 for the
inhibition of R39. Powerful b-lactamase inhibitors have been syn-
thesized by introducing the R side chains of different penams
(Fig. 2) in acylaminomethylboronic acids.16–19 Here, in addition
to the R side chains of methicillin 4 and benzylpenicillin 19 new
side chains were introduced and some potent inhibitors were
found.
O
4 900M-1s-1
Oxacillin
H3C
21 000M-1s-1
Piperacillin
53000M -1s-1
Figure 2. Acylation rate constants k2/K for commonly used penams of PBP2xR6.8
Boronic acids have been shown to be potent and specific inhib-
itors of serine proteases and of other enzymes.14 The dipeptidyl
boronic acid derivative Bortezomib, a proteasome inhibitor, is

A. Zervosen et al. / Bioorg. Med. Chem. 20 (2012) 3915–3924
3917
Table 2
Screening of lead-structure 4 analogs performed after a 1 h preincubation
O
OH
X
N
H
B
X:
OH
R
R
R
R
R
R
: R = OCH
9
4: R = OCH3
7: R = F
: R = OCH
3
8
3
R
(CH2)n
(CH2)n
R
R
22: R = OCH3, n = 0
21: R = OCH3
10
: R = OCH3, n = 0
23: R = Cl,
24
n = 0
n = 1
11: R = Cl,
12: R = Cl,
13: R = F,
n = 0
n = 1
: R = Cl,
n = 0
S
14
: R = NO2, n = 0
CH
3
15: R = NO2, n = 1
16: R = CH3, n = 0
17: R = CF3, n = 0
27
18
: R = H,
n = 0
n = 1
n = 2
25
26
19: R = H,
20: R = H,
Compound
R39 % residual activity
0 (IC50: 1.3 0.05)
PBP2xR6 % residual activity
PBP1b % residual activity
67 3b
4
7
8
9
26 2 (IC50: 278 17)
40 1 (IC50: 906 47)
69 1c
6
6 (IC50: 35 7)
7
3(100
l
M) (IC50: 7 0.1)cb
69 6c
18 ca; 76 5 (100
90 17ca
l
M)cb
81
43
2
1
82
86
3
1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
1cb; 67 8 (100
l
M)cb
15 1c (IC50: 85 25)
31 0c (IC50: 262 25)
26 1 (IC50: 649 30)
20 1(100
l
M)cb (IC50: 27 1)cb
24ca; 96 7 (100 M)cb
14 4 (100
M)cb (IC50: 16 1)cb
57
3
l
30 1 (IC50: 505)
56
1
l
0c (IC50: 0.6 0.07)
21 1c (IC50: 138 13)
31 2 (IC50: 339 45)
24 1(100
52 7 (100
27 1 (100
69ca
l
M)cb (IC50: 26 3)cb
52
45
5
9
1
l
l
M)cb
70
4
M)ca (IC50: 18 1)cb
1 (IC50: 15 2)
44 2 (IC50: 481 29)
80 1c
78
57
71
55
81
89
3
2
2
2
4
2
56 7ca
29 1 (IC50: 320)
56 3cb
51
50
50
73
71
71
0
2
2
9
3
4
0ca; 95 4 (100 M)ca
l
23 2ca (IC50: 79 16)ca
38 2ca; 82 1 (100
29 3ca; 70 1 (100
20 1ca; 69 1 (100
59 2ca
l
l
l
M)cb
M)cb
M)cb
32 2 (IC50: 376 61)
79
66 4c
100
7
73 1c
74
nd
5
1
87 3cb
The concentration of acylaminomethylboronic acid derivatives, if not specifically mentioned, was 1 mM. IC50 values are given in
lM.
a
Assay A.
Assay B.
b
c
0.01% Triton-X-100.
2.2. Synthesis of acylaminomethylboronic acids
2.3. Activity of acylaminomethylboronic acids
The synthesis described in Scheme 1 was used for the prepara-
tion of acylaminomethylboronic acid derivatives using methods
described in the literature. The diisopropyl-N,N-bis(trimethyl-
silyl)aminomethyl boronate 3 was synthesized in two steps from
dibromomethane.27,28 Compound 3 and acylchlorides were used
for the synthesis of various acylaminomethylboronic acids.26 Low
chemical yields reflected the reactivity of different acylchlorides
and the synthesis has yet to be optimized.
All compounds were tested for inhibition of PBP1b, a class A
PBP, two class PBPs, PBP2xR6 (penicillin sensitive) and
B
PBP2x5204 (penicillin resistant, see Supplementary data) all from
Streptococcus pneumoniae, the penicillin resistant PBP2a from
Staphylococcus aureus and PBP5fm from Enterococcus faecium as
well as R39 from Actinomadura sp. strain, a class C PBP. Thioesters
were used as reporter substrates in the presence of 5,50-dithio-
bis(2-nitrobenzoic acid) (DTNB) for the determination of the

3918
A. Zervosen et al. / Bioorg. Med. Chem. 20 (2012) 3915–3924
OH
K
k3
k2
H3N
B
Cl
5
E + P
E-I
[EI]
E + I
OH
Scheme 2. Model of the inactivation of PBPs by penams. The deacylation rate
constant k3 is very small.
Residual activity: 99 % (1 mM)
OCH3
O
OCH3
O
fluoresceyl ampicillin was used to determine the residual activity.
Inhibition of PBPs by boronic acids is time dependent (data not
shown).25 Residual activities were measured after pre-incubations
of PBPs with acylaminomethylboronic acid for 1 h (PBPs from
penicillin sensitive bacteria strains) or 4 h (PBPs from penicillin
resistant bacteria strains, see Supplementary data). Inhibition
was monitored in the presence of 0.01% Triton-X-100.
OH
N
H
OH
N
H
B
6
O
OCH3
OH
4
OCH3
IC50 = 1.3µM
Residual activity: 87 % (1 mM)
COOH
Several acylaminomethylboronic acids were inhibitors of PBPs
(Table 2). The acylaminomethylboronic acid derivative 14 with a
2-nitrobenzoyl side chain was a potent inhibitor of all PBPs of clas-
OCH3
O
N
H
B(OH)2
28
ses A (PBP1b: IC50 = 26
lM), B (PBP2xR6: IC50 = 138 lM) and C
OCH3
Residual activity: 84 % (1 mM)
Figure 3. (2,6-Dimethoxybenzamido)methylboronic acid 4 was identified as a good
(R39: IC50 = 0.6 M) from penicillin sensitive strains. Some of the
l
boronic acids also displayed low activities against class B PBPs
from penicillin resistant strains (PBP2x5204, PBP2a and PBP5fm)
(Table 1, Supplementary data).
inhibitor of R39 (IC50: 1.3 M). Two fragments of 4: the aminomethylboronic acid 5
l
The results summarized in Table 2 show that the nature of the X
side chain significantly influences the inhibitory ability of acyl-
aminomethylboronic acids. This observation is similar to the re-
sults obtained by kinetic studies with penams bearing different
side chains8 (Fig. 2). The second order rate constant k2/K, which
characterizes the formation of the E-I acyl-enzyme (Scheme 2), de-
pends on the nature of the side chain R (Fig. 2).
and 2-(2,6-dimethoxybenzamido)acetic acid 6, and 3-(dihydroxyboryl)-benzoic
acid 28 described by Inglis21 were not or only poor inhibitors at a concentration of
1 mM.
residual activity of PBP1b, PBP2x and R39. PBP2a and PBP5fm
from penicillin resistant bacterial strains do not hydrolyze the
thioesters 2-(2-benzamidopropanoylthio)acetic acid (S2d) and 2-
(2-(2-phenylacetamido)propanoylthio)propanoic acid (PATP) (see
Supplementary data). Thus counter labeling of active PBPs with
Compound 4 was a good inhibitor of R39 (IC50 = 1.3
mediocre inhibitor of PBP2xR6 (IC50 = 278 M) and a poor inhibitor
of PBP1b (IC50 >1000 M) (Table 2). Similar observations can
be made by comparing the results of inhibition experiments
lM), a
l
l
O
Br
Si
B
O
O
B
O
O
N
Si
O
2)
1)
B
O
2
1
3
3) or 3')
OH
B
O
H
R
N
O
OH
R
N
B
H
O
E
O
1) n-BuLi in hexane, CH2Br2, THF, -78°C, 90 min, 79 %
2) n-BuLi, HMDS, THF, -78°C, 30 min, then to rt overnight, 58 %
3) 5 equiv. RCOCl, CH Cl , -78°C
rt, 120 min; MeOH, -78°C
rt, 30 min, H O (method A)
→
→
2
2
3') 1 equiv MeOH, -78°2C
rt, 2 h ; 1 equiv RCOCl, CH Cl , -78°C
rt, 17 h; H O (method B)
→
→
2
2
2
Scheme 1. Synthesis of acylaminomethylboronic acids.

A. Zervosen et al. / Bioorg. Med. Chem. 20 (2012) 3915–3924
3919
performed with compounds 7 and 13. In contrast to compound 7,
which carries a 2,6-difluorobenzoyl-side chain, compound 13, with
a single fluoro substituent, was a poor inhibitor of R39 (13:
amino acids side chains in the R39 active site (Y147, L349) disfa-
voring the formation of the tricovalent adduct that is probably
the species responsible for the strong inhibition of R39 by amin-
omethylboronic acids. Thus 8 with methoxy groups in positions 2
and 4 was a poor inhibitor. Compound 10 was expected to be a
good inhibitor but its methoxy group might be oriented to the
exterior of the active site. A similar observation was made with
IC50 = 505
of PBP1b (7 and 13: IC50 <20
l
M, 7: IC50 = 35
l
M) while both were good inhibitors
M). These results reflect the struc-
l
tural differences between the active sites of PBPs as do the very dif-
ferent k2/K second order rate constants observed for penicillin G
with various PBPs (Table 1).
13 (IC50 = 505
35 M), which bears a 2,6-difluorobenzoyl side chain.
In the presence of ethylboronic acids, which mimic the
lM), which was also less potent than 7 (IC50
=
The nature of the substituent on the boron atom (aryl or alky)
l
also influences the inhibition.
A
compound with
a
2,6-
D
-ala-
dimethoxybenzoylamino side chain in the meta position of 3-
(dihydroxyboryl) benzoic acid (such as 2821 (Fig. 3)) was a poor
inhibitor of R39 while 4, with a aminomethylboronic acid group,
was a potent inhibitor. 3-(Dihydroxyboryl)benzoic acid derivatives
with a 2-methoxybenzoylamino, a 2-phenyacetylamino or a ben-
zoylamino side chains were described as good inhibitors of R39
nine residue, the natural amino acid that acylates the active serine
of PBPs, only monocovalent complexes (Fig. 4) were observed
with R39.24,25 The IC50-value of 29 (Fig. 4), bearing
a
2,6-
M24, IC50
M24 was observed for
the ethylboronic acid analogs of 7 (IC50 = 35 M) indicating that
dimethoxybenzoyl side chain, was increased (IC50 = 33
(4) = 1.3 M), while an IC50-value of 3.3
l
l
l
l
(IC50 <100
l
M)21 while the corresponding aminomethylboronic
the 2,6-difluorobenzoyl side chain probably fits better in the
monocovalent complex than the 2,6-dimethoxybenzoyl group.
The co-crystal structures of the monocovalent complexes of
PBP1b were classified into two ‘families’ depending on the
occupation of the pockets by the aryl groups.20 The aryl groups of
glycine analogs (7, 11, 13 and 14) and the R-methyl-substituted
analog of 29 occupied pocket 1 while the side chains of the
glycine analog 4 and of the S-methyl substituted analogs were
acid derivatives 10 (RA = 43% at 1 mM), 19 (IC50: 320 lM) and 18
(RA = 80% at 1 mM) were less potent.
Crystal structures of adducts of R39 with 4, 11, 14 and 1925 and
PBP1b with 4, 7, 11, 13, and 1420 were solved and allowed an inter-
pretation of the results. No crystal structure of a complex of
PBP2xR6 with a boronic acid has been described yet.
An unexpected tricovalent-binding mode was observed in the
active site of R39 with 4, 11, 14 and 19 (Fig. 4). In the tricovalent
adduct specific interactions of ortho substituents on the phenyl
group of the aminomethylboronic acids within the active site of
R39 explain the decrease of Ki25 and IC50-values of 4, 7, 11, 14
and 17 when compared with compounds without ortho substitu-
found in pocket 2 indicating that the C-a chirality may direct the
side chain to a specific pocket. The C-6/C-7 amide side chains of
cephalosporins and penicillins, including methicillin, which has a
2,6-dimethoxybenzoyl side chain (Fig. 2) also fill pocket 2.20
Compound 4 was a poor inhibitor of PBP1b (IC50 >1000 lM) indicat-
ents like 19 (IC50 = 320
was weak in the case of 7 (fluoro, IC50 = 35
IC50 = 85 M, Ki = 42
M25) and 17 (trifluoromethyl, IC50 = 15
but strong with 4 (methoxy, IC50 = 1.3 M, Ki = 1.5
M25) and 14
(nitro, IC50 = 0.6 M, Ki = 0.36
M25). Furthermore the introduction
l
M, Ki = 125
l
M25). The improvement
M), 11 (chloro,
M)
ing that its side chain does not fit very well in pocket 2. S-Methyl
substituted analogs with a hydrophobic group in the ortho position
of the aryl side chains such as the 2-fluoro-6-phenylbenzoyl
group (E1020), fit better and display good inhibitory activity against
l
l
l
l
l
l
l
l
PBP1b (IC50 of E10 = 20
Compounds 7, 11, 13, 14 and 16 are good inhibitors of PBP1b
(IC50: 7–27 M). No co-crystal structure with 16 has been solved
lM).
of a methylene group between the 2-chlorophenyl or the 2-nitro-
phenyl-side chains and the aminomethylboronic acid (12 and 15,
respectively), presumably reducing the prevalence of the tricova-
lent adduct in the inhibition of R39 led to an important increase
l
but results of inhibition studies indicate that this boronic acid
probably occupies pocket 1. The introduction of a methylene group
between the 2-chlorophenyl or the 2-nitrophenyl side chains and
the aminomethylboronic acid (12 and 15, respectively), increased
of the IC50-value (IC50 >1000
14 were 85 and 0.6 M, respectively. 20, with a 3-phenylpropanoyl
side chain (RA: 51% at 1 mM), was also less active than 19, with a
2-phenylacetyl side chain (IC50 = 320 M). In the tricovalent ad-
lM) while the IC50-values of 11 and
l
the IC50-values (12: RA = 97% at 100
RA = 52% at 100 M) while the IC50-values of 11 and 14 were 27
and 26 M, respectively. An increase of IC50-values could reflect
lM and 24% at 1 mM, 15:
l
l
duct formed between R39 and 4, one methoxy group occupies
l
the hydrophobic pocket of PBPs that is thought to accommodate
that molecules fit poorly in pocket 1 and that they could occupy
the methyl group of the penultimate
D
-alanine of the natural sub-
-Ala). A methoxy group
pocket 2 like compound 4.
strate (a pentapeptide ending with -Ala-
D
D
In conclusion, glycine analogs can form unexpected complexes
with PBPs (R39: tricovalent complex,25 PBP1b: 7, 11, 13 and 14
in position 3 (21, 9) or 4 (22, 9) of the phenyl group would certainly
not allow such hydrophobic interactions and could clash with
in pocket 120) while crystal structures with
D-alanine analogs (S-
methyl-substituted boronic acids) were more similar to those
formed with b-lactams (R39: monocovalent24; PBP1b: pocket 220).
Compounds 4, 7, 11, 13 and 14 showed no antibacterial activity
when tested on a broad range of bacterial strains (data not shown).
The (S)-1-(2-fluoro-6-phenylbenzamido)ethaneboronic acid E10
described before was active against pathogens including a methi-
cillin-resistant S. aureus (MRSA).20 Compounds 4 and 19 are ser-
ine-b-lactamase inhibitors.16,26 Inhibition studies of b-lactamases
with the acylaminomethylboronic acids presented in this study
are underway. The utilization of acylaminomethylboronic acids
in combination with a b-lactam can potentially broaden the spec-
trum of antibacterial activity.16
Lys410
OCH3
O
NH
OH
N
H
B
Ser298
O
B
O
Ser49
OH
R
4
OCH3
tricovalent
IC50: 1.3µM
OCH3
O
Me
OH
OH
N
H
B
HO
B
O
Ser49
OH
R
OCH3
29
IC50: 33µM
monocovalent
3. Conclusion
After the identification of (2,6-dimethoxybenzamido)methylbo-
Figure 4. Crystal structures of adducts of R39 with 4 and 29.24,25 An unexpected
tricovalent binding mode was observed with glycine analog 4 while the ethylbo-
ronic acid 4 as a potent R39 inhibitor (IC50 = 1.3
lM) analogs of 4
ronic acid 2924 that mimics the
D-alanine residue formed a monocovalent complex.
were prepared. Inhibition studies with PBPs of classes A, B and C

3920
A. Zervosen et al. / Bioorg. Med. Chem. 20 (2012) 3915–3924
from penicillin sensitive strains showed the importance of the nat-
ure of the side chain for the inhibition properties. The (2-nitroben-
zamido)methylboronic acid 14 was a potent inhibitor of all studied
PBPs from penicillin sensitive strains.
distillated thionyl chloride (3.4 mmol) was added and the solution
was stirred at room temperature overnight. The solvent and
thionyl chloride were removed under vacuum. The acid chloride
was dissolved with dichloromethane (15 mL) under nitrogen
atmosphere and directly used for the synthesis of the boronic acid.
4. Experimental
4.3. General procedure for the preparation of boronic acids
(method A)
All chemicals and reagents were either purchased p.A. from
commercial suppliers. All solvents used were HPLC grade. The HPLC
chain consists of a pump (Waters 600) and an UV detector (PDA
Waters 996), (200–400 nm). Analytical HPLC analysis was per-
For the preparation of boronic acids a protocol described by
Crompton26 was used. The previously prepared solution of acid
chloride (11.6 mmol) in dichloromethane (15 mL) under nitrogen
atmosphere (see above) was slowly added to diisopropyl-N,N-
bis(trimethylsilyl)aminomethylboronate (3, 5 mL, 2.2 mmol) at
ꢂ78 °C. After stirring for 2 h at room temperature the solution
was cooled to ꢂ78 °C and methanol (2.5 mL) was added. After stir-
ring for 30 min at room temperature, water (10 mL) was added and
the solution was stirred for 1 h. Organic solvents were removed
and the product was suspended in water (15 mL). After extraction
with ethyl acetate (3 ꢀ 15 mL) the aqueous layer was lyophilized
and the products were collected. Further purification was per-
formed by chromatography on Sep-Pak C18 Cartridges. Crude
products were suspended in water, put on the top of a Sep-Pak col-
umn and the column was washed with water containing 0.1% TFA.
The products were eluted using water with 0.1% TFA/acetonitrile
50/50 v/v. After the removing of acetonitrile under vacuum, aque-
ous solution was lyophilized.
formed on a XTerra RP18 (150 ꢀ 4.6 mm, 3.5
lm) column. The fol-
lowing protocol was used: solvent A: water (MilliporeQ) containing
0.1% TFA v/v and solvent B: acetonitrile. Flow: 0.7 mL/min, gradi-
ent: 0–30 min: 0–100% B, 30–35 min: 100% B, 35–36 min: 0–100%
A, 36–56 min: 100% A. The injection volumes were 20 lL.
Manual chromatography was performed with Sep-Pak C18 Car-
tridges from Waters. 1H NMR and 13C NMR spectra were recorded
at room temperature on either Bruker 250 or 400 MHz. Chemical
shifts d are given in ppm. The employed abbreviations for the mul-
tiplicities are the following ones: s = singlet, d = doublet, t = triplet,
q = quadruplet, p = quintet, m = multiplet, br = broad. The coupling
constant J is given in Hz. Spectra were recorded as solutions in d6-
DMSO (dH at 2.5 ppm, dC at 39.5 ppm) or D2O (dH at 4.79 ppm)
which were used as internal references. No 11B NMR spectra were
recorded because
a-amido boronic acids have different chemical
shifts at different pH values corresponding to different possible
conformations.29 TMSP (trimethylsilyl propionate) was used as
external reference for 13C NMR spectra done with D2O. MS analysis
was made on a device TSQ 7000 Thermoquest Finnigan equipped
4.4. General procedure for the preparation of boronic acids
(method B)
with an electrospray source. The co-solvent (injected in 200 lL/
To a solution of diisopropyl-N,N-bis(trimethylsilyl)aminometh-
ylboronate (3, 1.7 mmol) in dichloromethane (15 mL) under nitro-
gen atmosphere 2 mL of a dichloromethane solution containing
0.068 mL (1.7 mmol) methanol at ꢂ78 °C were added. After stirring
for 2 h at room temperature, the solution was cooled to ꢂ78 °C and
a solution (2 mL) of acid chloride (1.7 mmol) in dichloromethane
was added. After stirring overnight at room temperature the sol-
vent was removed under vacuum and the product was resolved
in water (15 mL) and upload to a Sep-Pak as described earlier in
method A.
min) is a mixture 50/50: H2O/acetonitrile containing 0.1% acetic
acid v/v. FT-MS were done on a device 9,4 Tesla Apex-QeFTICR
(Bruker Daltonics, Billerica, MA) with an electrospray source using
a ESI-Positive mode. The conditions of injection were: solvent is a
mixture 50/50: H2O/acetonitrile containing 0.1% formic acid v/v
containing ꢁ25
l
M of test compound. FT-MS⁄ were done on a SYN-
APT HDMS (Waters) with an internal calibration (72 m/z to 1285
m/z, error <2 ppm) using a nano ESI-Positive mode. Test com-
pounds were injected at a concentration of 10 lM in a solution
of 50/50: H2O/acetonitrile. Determination of an exact melting point
of boronic acids was not possible.
4.5. Preparation of various molecules
4.1. Preparation of diisopropyl-N,N-bis(trimethylsilyl)-
aminomethylboronate 3
4.5.1. (2,6-Dimethoxybenzamido)methylboronic acid 4 (method
A)
Three hundred and twenty milligrams (1.3 mmol) of a white so-
lid (yield: 88%) were obtained from 2,6-dimethoxybenzoyl chloride
and 3 (1.5 mmol). NMR-data were in agreement with Crompton.26
HPLC: Rt = 12.6 min (210 nm). MS (ESI, positive ion) m/z: [221
(24%), 222 (100%), 223 (12%)] (MꢂH2O+H). FT-MS: calculated
(MꢂH2O+H): 222.0932, observed: 222.0932.
17.5 g Diisopropyl (bromomethyl) boronate 2 were prepared
from dibromomethane as described in literature (2, 17.5 g, 79%).28
For the second step, an identical protocol as described by Martic-
honok27 was used. After the synthesis of lithiohexamethyldisilaz-
ane (LiN(TMS)2) from hexamethyldisilane (16.5 mL, 79 mmol) in
100 mL THF and n-BuLi (49 mL of 1.6 M solution in hexane,
78 mmol)27, the resulting solution was cooled (ꢂ78 °C) and diiso-
4.5.2. Aminomethylboronic acid 5
propyl (bromomethyl) boronate
2 (17.5 g, 78 mmol) in THF
A solution of 3 (5 mL) was mixed with a solution of methanol/
HCl 6 N v/v (20 mL) at 0 °C for 1 h. After another hour at room tem-
perature 10 mL of water was added. After washing with diethyl
ether (3 ꢀ 20 mL) the water phase was lyophilized. Eight hundred
and fifty milligrams of aminomethylboronic acid hydrochloride
were obtained. 1H NMR (400 MHz, D2O): d (ppm) = 2.56 (s, CH2).
MS (ESI, positive ion) m/z: [76 (100%), 75 (25%)] (MꢂH2O+H).
(100 mL) was slowly added. The reaction mixture was then allowed
to warm to 20 °C and stirred overnight. THF was removed under
vacuum and the residue Kugelrohr distilled to give diisopropyl-
N,N-bis(trimethylsilyl)-aminomethylboronate 3 (14.5 g, 46 mmol,
59%). 1H NMR-data were in agreement with the literature.30
4.2. General procedure for the preparation of a solution of acid
chlorides in CH2Cl2
4.5.3. 2-(2,6-Dimethoxybenzamido)acetic acid 6
2,6-Dimethoxybenzoic acid (1.83 g, 10 mmol) in 20 mL DMF
was mixed with the glycine methyl ester (HCl, 1.39 g, 11 mmol).
Under nitrogen atmosphere the solution was cooled at 0 °C.
To a solution of carboxylic acids (1.7 mmol) in dry DMF (100
and dry toluene (15 mL) under nitrogen atmosphere, freshly
lL)

A. Zervosen et al. / Bioorg. Med. Chem. 20 (2012) 3915–3924
3921
2.6 mL Diphenylphosphoryl azide (DPPA) (10 mmol) and 3.75 mL
di-isopropyl ethylamine (0.02 mol) were slowly added. After 4 h
at 0 °C the solution was stirred overnight at room temperature.
The reaction was followed by TLC (silica gel 60, ethyl acetate, RF
(product): 0.62). After the addition of 100 mL dichloromethane
the solution was washed first with 2 N HCl and then with NaHCO3
(5% w/v) and dried over Na2SO4. After filtration and the solvents re-
moval the product was solubilized in a small amount of dichloro-
methane and purified by filtration through silica gel using ethyl
acetate/petroleum ether 40–60, 4:1. After the elimination of the
solvent under vacuum a white solid (1.65 g, 0.0065 mol, yield:
65%) was obtained. The product was mixed with 2 equiv LiOH
solved in THF/water 1:1 v/v (20 mL). The pH of the solution was
adjusted to 1 with 6 N HCl. After the extraction of the product with
ethyl acetate the solution was washed with water and dried over
Na2SO4. After filtration the solvent was removed under vacuum
and a white solid (1.52 g, 6.4 mmol, yield: 98%) was obtained.
Mp: 110 °C. HPLC: Rt: 11.6 min (210 nm). 1H NMR was in agree-
ment with Chavez.31 13C NMR (100 MHz, DMSO): d (ppm) = 171.1
(COOH), 164.9 (CO), [156.9, 130.2, 116.4, 104.4 (Ar)], 55.8 (OCH3),
40.9 (CH2). MS (ESI, positive ion): m/z: [240 (100%), 241 (12%),
242 (2%)], (M+1). FT-MS: calculated (M+Na): 262.0691, observed:
262.0691.
CH2-B), 4.01 (s, 3H, OCH3), 7.18 (t, 3J = 7.6 Hz, 1H, Ar-H), 7.24 (d,
3J = 7.6 Hz, 1H, Ar-H), 7.69 (t, 3J = 7.2 Hz, 1H, Ar-H), 8.01 (d,
3J = 7.6 Hz, 1H, Ar-H). 13C NMR (62 MHz, D2O): d (ppm) = 28.8 (br,
CH2-B), 58.7 (OCH3), [114.9, 115.2, 123.7, 134.2, 139.1, 162.1
(Ar)], 172.7 (C@O). HPLC: Rt = 13.5 min (210 nm). MS (ESI, positive
ion) m/z: [191 (25%), 192 (100%), 193 (10%)] (MꢂH2O+H). FT-MS:
calculated (MꢂH2O+H): 192.0827, observed: 192.0826.
4.5.8. (2-Chlorobenzamido)methylboronic acid 11 (method A)
Two hundred and six milligrams (0.97 mmol) of a white solid
(yield: 44%) were obtained from 2-chlorobenzoyl chloride and 3
(2.2 mmol). 1H NMR (250 MHz, D2O): d (ppm) = 2.67 (s, 2H, CH2-
B), 7.36–7.52 (m, 4H, Ar-H). 13C NMR (62 MHz, D2O):
d
(ppm) = 33.5 (br, CH2-B), [130.1, 132.1, 133.0, 133.6, 134.2, 135.3
(Ar)], 173.8 (C@O). HPLC: Rt = 12 min (210 nm). MS (ESI, positive
ion) m/z: [195 (24%), 196 (100%), 197 (20%), 198 (34%)]
(MꢂH2O+H). FT-MS: calculated (MꢂH2O+H): 196.0331, observed:
196.0331.
4.5.9. (2-(2-Chlorophenyl)acetamido)methylboronic acid 12
(method B)
One hundred milligrams (0.44 mmol) of a white solid (yield:
26%) were obtained from 2-(2-chlorophenyl)acetyl chloride and 3
(1.7 mmol). 1H NMR (400 MHz, D2O): d (ppm) = 2.44 (s, 2H, CH2-
B), 3.88 (s, 2H, Ar-CH2), 7.35 (m, 3H, Ar-H), 7.48 (m, 1H, Ar-H).
13C NMR (62 MHz, D2O): d (ppm) = 34.8 (br, CH2-B), 37.3 (Ar-
CH2), [130.3, 132.4, 134.0, 134.8, 137.0 (Ar)], 178.3 (C@O). HPLC:
Rt = 14.5 min (210 nm). MS (ESI, positive ion) m/z = [209, 210
(64%), 211, 212] (MꢂH2O+H). FT-MS: calculated (MꢂH2O+H):
210.0488, observed: 210.0488.
4.5.4. (2,6-Difluorobenzamido)methylboronic acid 7 (method A)
One hundred and sixty-two milligrams (0.75 mmol) of a white
solid (yield: 34%) were obtained from 2,6-difluorobenzoyl chloride
and 3 (2.2 mmol). 1H NMR was in agreement with Inglis.32 13C
NMR (62 MHz, D2O):
d (ppm) = 31.6 (br, CH2-B), 113.9 (t,
4
2
2JCF = 18.6 Hz, C-1), 114.8 (dd, JCF = 3.2 Hz, JCF = 22 Hz, C-3 and
3
1
C-5), 136.0 (t, JCF = 10.7 Hz, C-4), 162.2 (dd, JFC = 247 Hz,
3JCF = 6.6 Hz, C-2 and C-6), 166.9 (C@O). HPLC: Rt = 9.9 min
(210 nm). MS (ESI, positive ion) m/z: [197 (25%), 198 (100%), 199
(10%)] (MꢂH2O+H). FT-MS: calculated (MꢂH2O+H): 198.0532, ob-
served: 198.0532.
4.5.10. (2-Fluorobenzylamido)methylboronic acid 13 (method
B)
One hundred and eighty milligrams (0.91 mmol) of a white so-
lid (yield: 54%) were obtained from 2-fluorobenzoyl chloride and 3
(1.7 mmol). 1H NMR (400 MHz, D2O): d (ppm) = 2.66 (s, 2H, CH2-B),
3
3
4.5.5. (2,4-Dimethoxybenzamido)methylboronic acid 8 (method
B)
7.29 (dd, JHH = 8.6 Hz, JHF = 11.8 Hz 1H, Ar-H), 7.35 (t,
3JHH = 7.6 Hz, 1H, Ar-H), 7.66 (m, 1H, Ar), 7.87 (td, JHH = 7.6 Hz,
3
Forty two milligrams (0.18 mmol) of a white solid (yield: 10%)
4JHF = 1.6 Hz, 1H, Ar-H). 13C NMR (62 MHz, D2O): d (ppm) = 36.1
2
2
were obtained from 2,4-dimethoxybenzoyl chloride and
3
(br, CH2-B), 119.2 (d, JCF = 12 Hz, Ar), 119.3 (d, JCF = 21.7 Hz, Ar),
(1.7 mmol). 1H NMR (250 MHz, D2O): d (ppm) = 2.52 (s, 2H, CH2-
B), 3.88 (s, 3H, OCH3), 3.91 (s, 3H, OCH3), 6.6 (s, 1H, Ar-H), 6.68
(dd, 3J = 8.8 Hz, 4J = 2.3 Hz, 1H, Ar-H), 7.91 (d, 3J = 9 Hz, 1H, Ar-H).
13C NMR (62 MHz, D2O): d (ppm) = 39.0 (br, CH2-B), 58.5 (OCH3),
[100.7, 108.2, 109.4, 136.1, 164.0, 168.2 (Ar)], 172.2 (C@O). HPLC:
Rt = 18.5 min (210 nm). MS (ESI, positive ion) m/z: [221 (24%),
222 (100%), 223 (11%)] (MꢂH2O+H). FT-MS⁄: calculated (M+Na):
262.0868, observed: 262.0864.
127.7 (d, JCF = 3.2 Hz, Ar), 133.4 (d, JCF = 1.8 Hz, Ar), 138.1 (d,
4
3
3JCf = 9.3 Hz, Ar), 163.4(d, JCF = 249 Hz, Ar), 170.9 (d, JCF = 2.2 Hz,
C@O). HPLC: Rt = 12.0 min (210 nm). MS (ESI, positive ion) m/z:
[179 (25%), 180 (100%), 181 (14%)] (MꢂH2O+H). FT-MS: calculated
(MꢂH2O+1): 180.0627, observed: 180.0627.
1
3
4.5.11. (2-Nitrobenzamido)methylboronic acid 14 (method A)
Two hundred and twenty-one milligrams (0.099 mmol) of a
white solid (yield: 45%) were obtained from 2-nitrobenzoyl chlo-
ride and 3 (2.2 mmol). 1H NMR (250 MHz, D2O): d (ppm) = 2.77
(s, 2H, CH2-B), 7.52 (m, 1H, Ar-H), 7.62 (m, 1H, Ar-H), 7.72 (m,
4.5.6. (3,4-Dimethoxybenzamido)methylboronic acid 9 (method
A)
Two hundred and eighty-three milligrams (1.2 mmol) of a
white solid (yield: 54%) were obtained from 3,4-dimethoxybenzoyl
chloride and 3 (2.2 mmol). 1H NMR (400 MHz, D2O): d (ppm) = 2.54
(s, 2H, CH2-B), 3.85 (s, 3H, OCH3), 3.88 (s, 3H, OCH3), 7.05 (d,
3J = 8.4 Hz, 1H, Ar-H), 7.37 (s, 1H, Ar-H), 7.51 (d, 3J = 8.4 Hz, 1H,
Ar-H). 13C NMR (62 MHz, D2O): d (ppm) = 37.8 (br, CH2-B), 58.5
(OCH3), 58.6 (OCH3), [113.3, 114.1, 121.9, 125.4, 150.8, 155.6
(Ar)], 174 (C@O). HPLC: Rt = 13.2 min (210 nm). MS (ESI, positive
ion) m/z: [221 (23%), 222 (100%), 223 (17%)] (MꢂH2O+H). FT-MS:
calculated (MꢂH2O+H): 222.0932, observed: 222.0932.
1H, Ar-H), 8.06 (m, 1H, Ar-H). 13C NMR (62 MHz, D2O):
d
(ppm) = 32.4 (br, CH2-B), [127.6, 131.7, 132.4, 134.3, 137.4, 148.3
(Ar)], 172.9 (C@O). HPLC: Rt = 9.4 min (210 nm). MS (ESI, positive
ion) m/z: [206 (25%), 207 (100%), 208 (17%)] (MꢂH2O+H). FT-MS:
calculated (MꢂH2O+H): 207.0571, observed: 207.0571.
4.5.12. (2-(2-Nitrophenyl)acetamido)methylboronic acid 15
(method B)
One hundred and seventy-five milligrams (0.74 mmol) of a
white solid (yield: 43%) were obtained from 2-(2-nitro-
phenyl)acetyl chloride and 3 (1.7 mmol). 1H NMR (250 MHz,
D2O): d (ppm) = 2.52 (s, 2H, CH2-B), 4.09 (s, 2H, Ar-CH2), 7.55 (dd,
3J = 7.5 Hz, 4J = 1.5 Hz, 1H, Ar-H), 7.63 (td, 3J = 7.5 Hz, 4J = 1.5 Hz,
1H, Ar-H), 7.77 (td, 3J = 7.5 Hz, 4J = 1.5 Hz, 1H, Ar-H), 8.2 (dd,
3J = 8.0 Hz, 4J = 1.5 Hz, 1H, Ar-H). 13C NMR (62 MHz, D2O): d
4.5.7. (2-Methoxybenzamido)methylboronic acid 10 (method A)
One hundred and sixty-one milligrams (0.77 mmol) of a white
solid (yield: 35%) were obtained from 2-methoxybenzoyl chloride
and 3 (2.2 mmol). 1H NMR (400 MHz, D2O): d (ppm) = 2.56 (s, 2H,

3922
A. Zervosen et al. / Bioorg. Med. Chem. 20 (2012) 3915–3924
(ppm) = 34.3 (br, CH2-B), 40.4 (Ar-CH2), [128.3, 131.5, 132.2, 136.7,
137.5, 150.9 (Ar)], 177.7 (C@O). HPLC: Rt = 11.6 min (210 nm). MS
(ESI, positive ion) m/z: [220 (24%), 221 (100%), 222 (16%)]
(MꢂH2O+H). FT-MS⁄: calculated (M+Na): 261.0656, observed:
261.0657.
Ar-H), 7.51 (m, 2H, Ar-H). 13C NMR (62 MHz, D2O):
d
(ppm) = 58.5 (OCH3), [115.9, 122.5, 123.4, 131.9, 133.3, 162.0
(Ar)], 174.5 (C@O). HPLC: Rt = 13.6 min (210 nm). MS (ESI, positive
ion) m/z: [191 (25%), 192 (100%), 193 (10%)] (MꢂH2O+H). FT-MS:
calculated (MꢂH2O+H): 192.0827, observed: 192.0826.
4.5.13. (2-Methylbenzamido)methylboronic acid 16 (method A)
Three hundred and seventy-seven milligrams (1.95 mmol) of a
white solid (yield: 89%) were obtained from 2-methylbenzoyl chlo-
ride and 3 (2.2 mmol). 1H NMR (250 MHz, D2O): d (ppm) = 2.15 (s,
3H, Ar-CH3), 2.44 (s, 2H, CH2-B), 7.03 (m, 2H, Ar-H), 7.25 (m, 2H, Ar-
H).. 13C NMR (62 MHz, D2O): d (ppm) = 22.1 (Ar-CH3), 37.4 (br, CH2-
B), [128.7, 129.2, 131.1, 134.1, 135.4, 140.6 (Ar)], 177.3 (C@O).
HPLC: Rt = 12.4 min (210 nm). MS (ESI, positive ion) m/z: [175
(25%), 176 (100%), 177 (10%)] (MꢂH2O+H). FT-MS: calculated
(MꢂH2O+H): 176.0877, observed: 176.0877.
4.5.19. (4-Methoxybenzamido)methylboronic acid 22 (method
A)
One hundred and eighty-seven milligrams (0.89 mmol) of a
white solid (yield: 40%) were obtained from 4-methoxybenzoyl
chloride and
3 d
(2.2 mmol). 1H NMR (400 MHz, D2O):
(ppm) = 2.52 (s, 2H, CH2-B), 3.88 (s, 3H, OCH3), 7.08 (d,
3J = 8.4 Hz, 2H, Ar-H), 7.86 (d, 3J = 8.4 Hz, 2H, Ar-H). 13C NMR
(62 MHz, D2O): d (ppm) = 35.4 (br, CH2-B), 56.3 (OCH3), [115.1,
118.9, 131.0,, 164.3 (Ar)], 172.2 (C@O). HPLC: Rt = 13.4 min
(210 nm). MS (ESI, positive ion) m/z: [191 (25%), 192 (100%), 193
(10%)] (MꢂH2O+H). FT-MS⁄ calculated (MꢂH2O+H): 192.0827, ob-
served: 192.0833.
4.5.14. (2-(Trifluoromethyl)benzamido)methylboronic acid 17
(method B)
One hundred and seventeen milligrams (0.47 mmol) of a white
solid (yield: 28%) were obtained from 2-(trifluoromethyl)benzoyl
4.5.20. (4-Chlorobenzamido)methylboronic acid 23 (method A)
One hundred and eleven milligrams (0.51 mmol) of a white so-
lid (yield: 39%) were obtained from 4-chlorobenzoyl chloride and 3
(1.3 mmol). 1H NMR (250 MHz, D2O): d (ppm) = 2.61 (s, 2H, CH2-B),
7.53 (d, 3J = 8.5 Hz, 2H, Ar-H), 7.81 (d, 3J = 8.5 Hz, 2H, Ar-H). 13C
NMR (62 MHz, D2O): d (ppm) = 27.0 (br, CH2-B), [126.4, 129.8,
130.0, 140.0 (Ar)], 171.6 (C@O). HPLC: Rt = 15.8 min (210 nm).
MS (ESI, positive ion) m/z: [195 (24%), 196 (100%), 197 (19%), 198
(34%)] (MꢂH2O+H). FT-MS⁄: calculated (MꢂH2O+H): 196.0337, ob-
served: 196.0336.
chloride and
3 d
(1.7 mmol). 1H NMR (400 MHz, D2O):
(ppm) = 2.83 (s, 2H, CH2-B), 7.59 (d, 3J = 6.8 Hz, 1H, Ar-H), 7.71
(m, 2H, Ar-H), 7.84 (d, 3J = 7.2 Hz, 1H, Ar-H). 13C NMR (62 MHz,
1
D2O): d (ppm) = 32.8 (CH2-B(OH)2), 126.3 (q, JCF = 269 Hz, CF3)
3
2
129.4 (q, JCF = 4 Hz, Ar), 129.35 (q, JCF = 32 Hz, Ar), 131.1,
133.5(Ar), 135.0 (m, Ar), 135.2 (m, Ar), 174.4 (C@O). HPLC:
Rt = 15.5 min (210 nm). MS (ESI, positive ion) m/z: [229 (28%),
230 (100%), 231 (8%)] (MꢂH2O+H). FT-MS: calculated (MꢂH2O+H):
230.0595, observed: 230.0595.
4.5.21. (2-(4-Chlorophenyl)acetamido)methylboronic acid 24
(method B)
4.5.15. Benzamidomethylboronic acid 18 (method B)
Two hundred and eleven milligrams (1.2 mmol) of a white solid
(yield: 70%) were obtained from benzoylchloride and 3 (1.7 mmol).
NMR-data were in agreement with Lai.29 HPLC: Rt = 11.3 min
(210 nm). MS (ESI, positive ion) m/z: [161 (24%), 162 (100%), 163
(13%)] (MꢂH2O+H). FT-MS⁄: calculated (M+Na): 202.0651, ob-
served: 202.0651.
One hundred and twenty-five milligrams (0.55 mmol) of a
white solid (yield: 32%) were obtained from 2-(4-chloro-
phenyl)acetyl chloride and 3 (1.7 mmol). 1H NMR (400 MHz,
D2O): d (ppm) = 2.38 (s, 2H, CH2-B), 3.69 (s, 2H, Ar-CH2), 7.25 (d,
3J = 8.8 Hz, 2H, Ar-H), 7.38 (d, 3J = 8.5 Hz, 2H, Ar-H). 13C NMR
(62 MHz, D2O): d (ppm) = 37.5 (Ar-CH2), [129.9, 131.9, 133.3,
134.6 (Ar)], 178.7 (C@O). HPLC: Rt = 17.7 min (210 nm). MS (ESI,
positive ion) m/z: [209 (25%), 210 (100%), 211 (19%), 212 (34%)]
(MꢂH2O+H). FT-MS: calculated (MꢂH2O+H): 210.0488, observed:
210.0488.
4.5.16. (2-Phenylacetamido)methylboronic acid 19 (method B)
One hundred and twenty-eight milligrams (0.66 mmol) of a
white solid (yield: 39%) were obtained from 2-phenylacetyl chlo-
ride and 3 (1.7 mmol). NMR-data were in agreement with Cromp-
ton.26 HPLC: Rt = 15.1 min (210 nm). MS (ESI, positive ion) m/z:
[175 (28%), 176 (100%), 177 (14%)] (MꢂH2O+H). FT-MS: calculated
(MꢂH2O+H): 176.0877, observed: 176.0877.
4.5.22. (3-Cyclopentylpropanamido)methylboronic acid 25
(method B)
One hundred and ninety-seven milligrams (1 mmol) of a white
solid (yield: 58%) were obtained from cyclopentanepropionyl chlo-
ride and 3 (1.7 mmol). 1H NMR (400 MHz, D2O): d (ppm) = 1.09 (m,
2H, cyclopentyle-CH2), 1.63 (m, 6H, cyclopentyle-CH2), 1.75 (m,
3H, cyclopentyle-CH2), 2.34 (s, 2H, CH2-B), 2.42 (t, 3J = 7.2 Hz, 2H,
CH2–C@O). 13C NMR (62 MHz, D2O): d (ppm) = [27.5, 32.8, 33.5,
34.6, 41.7 (cyclopentyle-CH2–CH2)], 37.2 (br, CH2-B), 182.4
(C@O). MS (ESI, positive ion) m/z: [181 (25%), 182 (100%), 183
(12%)] (MꢂH2O+H). FT-MS: calculated (MꢂH2O+H): 182.1347, ob-
served: 182.1347.
4.5.17. (3-Phenylpropanamido)methylboronic acid 20 (method
B)
Two hundred and eighteen milligrams (1.1 mmol) of a white so-
lid (yield: 64%) were obtained from 3-phenylpropanoyl chloride
and 3 (1.7 mmol). 1H NMR (400 MHz, D2O): d (ppm) = 2.30 (s, 2H,
CH2-B), 2.71 (t, 3J = 7.2 Hz, 2H, CH2–C@O), 2.97 (t, 3J = 7.2 Hz, 2H,
Ar-CH2), 7.28-7.40 (m, 5H, Ar-H). 13C NMR (62 MHz, D2O): d
(ppm) = 33.3 (Ar-CH2), 35.7 (CH2–C@O), 36.0 (br, CH2-B), [129.4,
131.2, 131.6, 142.7 (Ar)], 180.6 (C@O). HPLC: Rt = 21.2 min
(210 nm). MS (ESI, positive ion) m/z: [189 (25%), 190 (100%), 191
(10%)] (MꢂH2O+H). FT-MS: calculated (MꢂH2O+H): 190.1034, ob-
served: 190.1034.
4.5.23. (Thiophene-2-carboxamido)methylboronic acid 26
(method B)
One hundred and forty-five milligrams (0.78 mmol) of a white
solid (yield: 46%) were obtained from thiophenyl-2-carbonyl chlo-
rid and 3 (1.7 mmol). 1H NMR (400 MHz, D2O): d (ppm) = 2.63 (s,
2H, CH2-B), 7.22 (t, 3J = 4.4 Hz, 1H), 7.80 (m, 2H). 13C NMR
(62 MHz, D2O): d (ppm) = 35.2 (br, CH2-B), [131.3, 134.4, 134.5,
135.9 (Ar)], 168.7 (C@O). HPLC: Rt = 11.3 min (210 nm). MS (ESI,
positive ion) m/z: 168 (MꢂH2O+H). FT-MS: calculated (MꢂH2O+H):
168.0285, observed: 168.0285.
4.5.18. (3-Methoxybenzamido)methylboronic acid 21 (method
A)
One hundred and sixty-seven milligrams (0.8 mmol) of a white
solid (yield: 36%) were obtained from 3-methoxybenzoyl chloride
and 3 (2.2 mmol). 1H NMR (250 MHz, D2O): d (ppm) = 2.62 (s, 2H,
CH2-B(OH)2), 3.89 (s, 3H, OCH3), 7.29 (m, 1H, Ar-H), 7.44 (m, 1H,

A. Zervosen et al. / Bioorg. Med. Chem. 20 (2012) 3915–3924
3923
4.5.24. Acetamidomethylboronic acid 27 (method A)
One hundred and fifty-nine milligrams (1.36 mmol) of a white
solid (yield: 80%) were obtained from acetyl chloride and 3
(1.7 mmol). 1H NMR (400 MHz, D2O): d (ppm) = 2.06 (s, 3H, CH3)
2.32 (s, 2H, CH2-B), 13C NMR (100 MHz, D2O): d (ppm) = 19.2,
37.8 (br, CH2-B), 179.7 (C@O). MS (ES+): m/z: [100 (100%), 101
(10%), 102 (5%)] (MꢂH2O+H). FT-MS⁄: calculated (MꢂH2O+H):
100.0570, observed: 100.0569.
0.01% Triton-X-100 v/v. As described in the literature promiscuous
inhibitors are slow binding, non-competitive inhibitors. In order to
avoid a detailed kinetic study13, it is possible to identify such com-
pounds by performing tests in the presence of Triton-X-100.37,38
Promiscuous inhibitors show no inhibition in the presence of Tri-
ton-X-100. IC50-values of R39, PBP2xR6 and PBP1b were deter-
mined in the presence of 0.01% Triton-X-100 v/v. RA was
measured over a range of concentrations from which IC50 values
were determined by performing a non-linear regression analysis
using Sigma Plot (Systat software) and fitting the data to the equa-
tion3 y = y0 + (a ꢀ b)/(b + x).
4.6. Biological assays
R39 from Actinomadura was prepared and purified as described
by Granier.33 PBP2x-5204 and, PBP2x-R6 from Streptococcus pneu-
moniae were prepared as described by Carapito4 while PBP1b from
the same organism was purified as described by Di Guilmi.6 Fluo-
rescein labeled ampicillin was prepared as described by Lakaye.34
The thioester 2-(2-benzamidopropanylthio)acetic acid S2d was
prepared as described by Adam35 and Schwyzer.36 The preparation
of 2-(2-(2-phenylacetamido)propanoylthio)propanoic acid PATP
will be described elsewhere.
Acknowledgments
We thank the European Union (European Community Sixth
Framework Programme) via the EUR-INTAFAR project for the
financial support for this research. We thank F. Bouillenne (CIP,
University of Liège, Belgium) for the gift of R39, André Zapun
(IBS, Grenoble, France) for PBP2xR6 and PBP2x5204 and Andréa
Dessen (IBS, Grenoble, France) for PBP1b. We thank Nathalie Teller
for the preparation of S2d and Gabriel Mazzucchelli (LSM-GIGA-
Proteomics, Liège, Belgium) for the FT-MS analysis. The authors
wish to thank Andrea Dessen (IBS) and Jean-Marie Frère (CIP, Uni-
versity of Liège, Belgium) for critical reading of the manuscript.
4.6.1. Inhibition tests of R39 and PBP2xR6
Inhibition experiments with R39 and PBP2xR6 were performed
by monitoring the degree of hydrolysis of the substrate S2d using
microtiter 96 well plates and a Power Wave microtiter plate reader
(Bio-Tek Instruments). Enzyme residual activity (RA) was deter-
mined after pre-incubation of the PBPs in the presence of potential
inhibitors. The initial rate of hydrolysis of S2d and the rate of spon-
taneous hydrolysis of S2d in the presence of the inhibitors was
determined in the presence of 5,50-dithiobis(2-nitrobenzoic acid)
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmc.2012.04.018.
(DTNB, e
412 nm = 14,150 Mꢂ1 sꢂ1). All experiments were performed
References and notes
in triplicate. Activity of PBPs in absence of inhibitors (100% RA) was
measured with six replicates on each plate. Potential inhibitors
were solubilized in DMF. The final concentration of DMF in the as-
says was 1%. Potential inhibitors were incubated with 3.5 nM R39
in 10 mM sodium phosphate buffer (pH 7.2) with 100 mM NaCl,
1. Matteï, P.-J.; Neves, D.; Dessen, A. Curr. Opin. Struct. Biol. 2010, 20, 749.
2. Sauvage, E.; Kerff, F.; Terrak, M.; Ayala, J. A.; Charlier, P. FEMS Microbiol. Rev.
2008, 32, 234.
3. Macheboeuf, P.; Fischer, D. S.; Brown, T.; Zervosen, A.; Luxen, A.; Joris, B.;
Dessen, A.; Schofield, C. J. Nat. Chem. Biol. 2007, 3, 565.
4. Carapito, R.; Chesnel, L.; Vernet, T.; Zapun, A. J. Biol. Chem. 2006, 281,
1771.
100 mM
the case of PBP2xR6, 0.09
D
-alanine and 0.01 mg/mL BSA for 60 min at 25 °C. In
M enzyme was incubated in the pres-
l
´
ence of potential inhibitors in 10 mM sodium phosphate buffer
(pH 7.0) and 0.01 mg/mL BSA for 60 min at 25 °C. After the pre-
incubation RAs were determined by adding S2d (1 mM) and DTNB
(1 mM) and measuring the initial rate of hydrolysis of S2d at
5. Sauvage, E.; Zervosen, A.; Dive, G.; Herman, R.; Amoroso, A.; Joris, B.; Fonze, E.;
Pratt, R. F.; Luxen, A.; Charlier, P.; Kerff, F. d. r. J. Am. Chem. Soc. 2009, 131,
15262.
6. Di Guilmi, A. M.; Dessen, A.; Dideberg, O.; Vernet, T. J. Bacteriol. 2003, 185, 4418.
7. Frère, J. -M.; Nguyen-Distèche, M.; Coyette, J.; Joris, B. In The chemistry of beta-
lactams, Hall, C. A. Ed.; Glasgow, 1992; pp 148–195.
8. Jamin, M.; Hakenbeck, R.; Frere, J.-M. FEBS Lett. 1993, 331, 101.
9. Lu, W.-P.; Sun, Y.; Bauer, M. D.; Paule, S.; Koenigs, P. M.; Kraft, W. G.
Biochemistry 1999, 38, 6537.
10. Zorzi, W.; Zhou, X. Y.; Dardenne, O.; Lamotte, J.; Raze, D.; Pierre, J.; Gutmann,
L.; Coyette, J. J. Bacteriol. 1996, 178, 4948–4957.
412 nm (total test volume: 150 lL).
4.6.2. Inhibition tests of PBP1b
4.6.2.1. Assay A. PBP1b (0.7
lM) was incubated with poten-
tial inhibitors (1 mM) in 10 mM sodium phosphate (pH 7.0) for
60 min at 25 °C. Active PBP1b was then counter labeled with fluo-
resceyl-ampicillin (10 lM) 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). Background fluorescence was subtracted.
11. Miguet, L.; Zervosen, A.; Gerards, T.; Pasha, F. A.; Luxen, A.; Diste`che-Nguyen,
M.; Thomas, A. J. Med. Chem. 2009, 52, 5926.
12. Turk, S.; Verlaine, O.; Gerards, T.; Zivec, M.; Humljan, J.; Sosic, I.; Amoroso, A.;
Zervosen, A.; Luxen, A.; Joris, B.; Gobec, S. PLoS One 2011, 6, e19418.
13. Zervosen, A.; Lu, W.-P.; Chen, Z.; White, R. E.; Demuth, T. P., Jr.; Frere, J.-M.
Antimicrob. Agents Chemother. 2004, 48, 961.
14. Dembitsky, V. M.; Al Quntar, A. A.; Srebnik, M. Chem. Rev. 2011, 111, 209.
15. Kropff, M.; Bisping, G.; Schuck, E.; Liebisch, P.; Lang, N.; Hentrich, M.; Dechow,
T.; Kroger, N.; Salwender, H.; Metzner, B.; Sezer, O.; Engelhardt, M.; Wolf, H. H.;
Einsele, H.; Volpert, S.; Heinecke, A.; Berdel, W. E.; Kienast, J. Br. J. Haematol.
2007, 138, 330.
16. Caselli, E.; Powers, R. A.; Blasczcak, L. C.; Wu, C. Y. E.; Prati, F.; Shoichet, B. K.
Chem. Biol. 2001, 8, 17.
17. Drawz, S. M.; Babic, M.; Bethel, C. R.; Taracila, M.; Distler, A. M.; Ori, C.; Caselli,
E.; Prati, F.; Bonomo, R. A. Biochemistry 2009, 49, 329.
18. Morandi, F.; Caselli, E.; Morandi, S.; Focia, P. J.; Blázquez, J.; Shoichet, B. K.;
Prati, F. J. Am. Chem. Soc. 2002, 125, 685.
19. Wang, X.; Minasov, G.; Blázquez, J.; Caselli, E.; Prati, F.; Shoichet, B. K.
Biochemistry 2003, 42, 8434.
20. Contreras-Martel, C.; Amoroso, A.; Woon, E. C.; Zervosen, A.; Inglis, S.; Martins,
A.; Verlaine, O.; Rydzik, A. M.; Job, V.; Luxen, A.; Joris, B.; Schofield, C. J.; Dessen,
A. ACS Chem. Biol. 2011, 943.
21. Inglis, S. R.; Zervosen, A.; Woon, E. C. Y.; Gerards, T.; Teller, N.; Fischer, D. S.;
Luxen, A.; Schofield, C. J. J. Med. Chem. 2009, 52, 6097.
4.6.2.2. Assay B.
Inhibition experiments with PBP1b were also
performed by monitoring the degree of hydrolysis of the PATP thio-
ester as described before. Enzyme residual activity (RA) was calcu-
lated from initial rates after pre-incubation of PBP1b (0.2
the presence of potential inhibitors in 10 mM sodium phosphate
buffer (pH 7.0), 100 mM -alanine and 0.01 mg/mL BSA for
lM) in
D
60 min at 25 °C. Initial rates of hydrolysis of PATP (5 mM) were
determined in the presence of 5,50-dithiobis(2-nitrobenzoic acid)
(DTNB, 1.0 mM) as described (total test volume: 150 lL).
4.6.3. Screening of ‘false positives’ and determination of IC50
values
False positives (promiscuous inhibitors) were detected by per-
forming assays under the same conditions but in the presence of
22. Nicola, G.; Peddi, S.; Stefanova, M.; Nicholas, R. A.; Gutheil, W. G.; Davies, C.
Biochemistry 2005, 44, 8207.

3924
A. Zervosen et al. / Bioorg. Med. Chem. 20 (2012) 3915–3924
23. Pechenov, A.; Stefanova, M. E.; Nicholas, R. A.; Peddi, S.; Gutheil, W. G.
Biochemistry 2002, 42, 579.
31. Chavez, F.; Kennedy, N.; Rawalpally, T.; Williamson, R. T.; Cleary, T. Org. Process
Res. Dev. 2010, 14, 579.
24. Woon, E. C. Y.; Zervosen, A.; Sauvage, E.; Simmons, K. J.; Zꢀivec, M.; Inglis, S. R.;
Fishwick, C. W. G.; Gobec, S.; Charlier, P.; Luxen, A.; Schofield, C. J. ACS Med.
Chem. Lett. 2011, 2, 219.
25. Zervosen, A.; Herman, R.; Kerff, F.; Herman, A.; Bouillez, A.; Prati, F.; Pratt, R. F.;
Frere, J. M.; Joris, B.; Luxen, A.; Charlier, P.; Sauvage, E. J. Am. Chem. Soc. 2011,
133, 10839.
26. Crompton, I. E.; Cuthbert, B. K.; Lowe, G.; Waley, S. G. Biochem. J. 1988, 251, 453.
27. Martichonok, V.; Bryan Jones, J. Bioorg. Med. Chem. 1997, 5, 679.
28. Michnick, T. J.; Matteson, D. S. Synlett 1991, 631.
29. Lai, J. H.; Liu, Y. X.; Wu, W. G.; Zhou, Y. H.; Maw, H. H.; Bachovchin, W. W.; Bhat,
K. L.; Bock, C. W. J. Org. Chem. 2006, 71, 512.
32. Inglis, S. R.; Strieker, M.; Rydzik, A. M.; Dessen, A.; Schofield, C. J. Anal. Biochem.
2012, 420, 41.
33. 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. Method Enzymol. 1994, 244, 249.
34. Lakaye, B.; Damblon, C.; Jamin, M.; Galleni, M.; Lepage, S.; Joris, B.; Marchand-
Brynaert, J.; Frydrych, C.; Frere, J. M. Biochem. J. 1994, 300, 141.
35. Adam, M.; Damblon, C.; Plaitin, B.; Christiaens, L.; Frere, J. M. Biochem. J 1990,
270, 525.
36. Schwyzer, R. Helv. Chim. Acta 1954, 37, 647.
30. Matteson, D. S.; Singh, R. P.; Sutton, C. H.; Verheyden, J. D.; Lu, J.-h. Heteroat.
Chem. 1997, 8, 487.
37. Feng, B. Y.; Shoichet, B. K. Nat. Protoc. 2006, 1, 550.
38. Shoichet, B. K. Drug Discovery Today 2006, 11, 607.