12- to 22-membered bridged β-lactams as potential penicillin-binding protein inhibitors

Article abstract of DOI:10.1002/asia.201100732

As potential inhibitors of penicillin-binding proteins (PBPs), we focused our research on the synthesis of non-traditional 1,3-bridged β-lactams embedded into macrocycles. We synthesized 12- to 22-membered bicyclic β-lactams by the ring-closing metathesis (RCM) of bis-I-alkenyl-3(S)- aminoazetidinone precursors. The reactivity of 1,3-bridged β-lactams was estimated by the determination of the energy barrier of a concerted nucleophilic attack and lactam ring-opening process by using ab initio calculations. The results predicted that 16-membered cycles should be more reactive. Biochemical evaluations against R39 DD-peptidase and two resistant PBPs, namely, PBP2a and PBP5, revealed the inhibition effect of compound 4 d, which featured a 16-membered bridge and the N-tert-butyloxycarbonyl chain at the C3 position of the β-lactam ring. Surprisingly, the corresponding bicycle, 12 d, with the PhOCH₂CO side chain at C3 was inactive. Reaction models of the R39 active site gave a new insight into the geometric requirements of the conformation of potential ligands and their steric hindrance; this could help in the design of new compounds. Bridging inhibition: A series of 12- to 22-membered bicyclic bridged β-lactams were synthesized with the aim of developing new inhibitors of penicillin-binding proteins and feature a planar amide function and no carboxy group (see picture; Boc=tert-butyloxycarbonyl).

Full text of DOI:10.1002/asia.201100732

DOI: 10.1002/asia.201100732
12- to 22-Membered Bridged b-Lactams as Potential Penicillin-Binding
Protein Inhibitors
Aline Sliwa,[a] Georges Dive,*[b] and Jacqueline Marchand-Brynaert*[a]
Abstract: As potential inhibitors of
penicillin-binding proteins (PBPs), we
focused our research on the synthesis
of non-traditional 1,3-bridged b-lac-
tams embedded into macrocycles. We
synthesized 12- to 22-membered bicy-
clic b-lactams by the ring-closing meta-
thesis (RCM) of bis-w-alkenyl-3(S)-
aminoazetidinone precursors. The reac-
tivity of 1,3-bridged b-lactams was esti-
mated by the determination of the
energy barrier of a concerted nucleo-
philic attack and lactam ring-opening
process by using ab initio calculations.
The results predicted that 16-mem-
bered cycles should be more reactive.
Biochemical evaluations against R39
DD-peptidase and two resistant PBPs,
namely, PBP2a and PBP5, revealed the
inhibition effect of compound 4d,
which featured a 16-membered bridge
and the N-tert-butyloxycarbonyl chain
at the C3 position of the b-lactam ring.
Surprisingly, the corresponding bicycle,
12d, with the PhOCH2CO side chain at
C3 was inactive. Reaction models of
the R39 active site gave a new insight
into the geometric requirements of the
conformation of potential ligands and
their steric hindrance; this could help
in the design of new compounds.
Keywords: ab initio calculations ·
enzymes · inhibitors · macrocycles ·
metathesis
Introduction
sporins, carbapenems, and penems)[2] still remains that most
prescribed. Initially the activity of b-lactam antibiotics (i.e.,
penicillins) was attributed to strain in the four-membered
ring and the twisted amide bond of the 2-azetidinone func-
tionality. Therefore, the search for new b-lactam antibiotics
has been focused on strained 1,4-fused bicyclic structures,
which are intended to improve the so-called “acylating
power” of the b-lactam ring versus target serine proteases
(i.e., bacterial DD-peptidases).[3] Tebipenem, tomopenem,
or razupenem (Scheme 1), which are three carbapenems cur-
rently in phase II clinical studies, represent examples of this
strategy.[4]
The introduction of penicillins to the therapeutic arsenal, in
the early 1940s, was the starting point for the antibiotic era,
which saved millions of people from potentially fatal infec-
tious diseases. However, since the initial use of penicillins as
chemotherapeutic agents, the phenomenon of bacterial re-
sistance has been reported.[1] Owing to antibiotic pressure,
the occurrence of resistant bacteria is increasing and leading
to a worrisome situation with regards to antibiotics efficien-
cy, so worrying, in fact, that the World Health Organization
decided to dedicate World Health Day in 2011 to microbial
resistance.
However, this traditional model of reactivity seems to be
overevaluated because there is still no clear relationship be-
Despite the discovery of several other classes of antibiot-
ics, the b-lactam class (exemplified nowadays by cephalo-
[a] A. Sliwa, Prof. Dr. J. Marchand-Brynaert
Institute of Condensed Matter and Nanosciences (IMCN)
Molecules, Solids and Reactivity (MOST)
Universitꢀ catholique de Louvain
Bꢁtiment Lavoisier, Place Louis Pasteur L4.01.02
1348 Louvain-la-Neuve (Belgium)
Fax : (+32)10-47-41-68
E-mail: jacqueline.marchand@uclouvain.be
[b] Dr. G. Dive
Centre dꢂingꢀnierie des Protꢀines (CIP)
Universitꢀ de Liꢃge
Bꢁtiment B6, Allꢀe de la Chimie
4000 Sart-Tilman (Liꢃge) (Belgium)
Fax : (+32)43-66-33-64
E-mail: gdive@ulg.ac.be
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.201100732. It includes structural
data for compounds 2a–c, 2e, 2 f, 3b, 3c, 3e, 3 f, 4b, 4c, 4e, 4 f, 10a–
c, 11b, 11c, 12b, and 12c, as well as 13C NMR spectra of all new com-
pounds.
Scheme 1. Examples of carbapenems under development.
Chem. Asian J. 2012, 7, 425 – 434
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
425

FULL PAPERS
tween structural characteristics and biological activity. An
alternative model of reactivity was proposed by our group
some years ago, based on 1,3-bridged, planar b-lactam
motifs embedded into macrocycles. The aim was to possibly
decrease the activation barrier for cleavage of the 2-azetidi-
À
none N C(O) bond by increasing the conformational adapt-
ability involved in the atom reorganization during the for-
mation of an acyl–enzyme intermediate.[5] An initial series
of compounds A, structurally related to carbapenems, were
previously reported and evaluated against bacterial enzymes
with mitigated success attributed to an inadequate configu-
ration of the C3 chiral center. To further explore our hy-
pothesis, we decided to synthesize the series of compounds
B structurally related to cephalosporins and penicillins, with
an amino substituent and inversion of configuration at C3.
Because nucleophilic attack by serine enzymes occurs nor-
mally on the a face of the b-lactam, the macrocycle in B
that surrounds the b face would not hamper the serine
enzyme processing.
Scheme 2. Retrosynthetic strategy for the synthesis of series B com-
pounds.
used model for bacterial enzymes, and against two resistant
PBPs, namely, PBP2a from Staphylococcus aureus[8] and
PBP5 from Enterococcus faecium.[9]
Results and Discussion
Synthesis
The Boc family was the first series of bis-alkylated bicycles
B synthesized (X=Y=H,H; R=Boc). The starting material
was (S)-3-(tert-butyloxycarbonyl)amino-2-azetidinone (1),
which was readily prepared from commercially available
Boc-l-serine, as previously described.[6]
The ring-closing metathesis (RCM) reaction was chosen
as the key step for the formation of the bridging macrocy-
cles. The precursors were chiral azetidin-2-ones C with w-al-
À
kenoyl or w-alkenyl chains on the N1 and C3 N positions.
The starting chirons D, which were derivatives of (S)-3-ami-
noazetidin-2-one, came from l-serine.
w-Alkenyl bromides of various lengths were used to
access various sizes of bicycles. All of the w-alkenyl bro-
mides used were commercially available reagents. The bis-
alkyl derivatives 2a–f were synthesized in one step, with
moderate yields, by using 2 equivalents of NaH and
2.2 equivalents of w-alkenyl bromides (step a in Scheme 3).
Other strong bases were tested (namely, lithium hexame-
First, we tried to synthesize the bicyclic family B with bis-
acylated chains (X=Y=O). Unexpectedly, when applying
the RCM reaction to the bis-acylated precursors C (X=Y=
O, Scheme 2), we recovered exclusively cyclodimers, except
in one case (R=tert-butyloxycarbonyl (Boc); n=2).[6] Due
to the presence of amide and imide functionalities in the
precursors C, the conformers leading to the desired cycliza-
tions are strongly disfavored. This should not be the case for
the bis-alkylated precursors (X=Y=H,H). Therefore, we
kept the same retrosynthetic strategy to synthesize the 1,3-
bridged bicyclic b-lactam compounds B with bis-alkylated
chains (X=Y=H,H).
Herein, we describe the successful synthesis of target mol-
ecules B in the bis-alkylated family (X=Y=H,H). All of
these compounds were investigated from a theoretical point
of view and their reactivity in a model of a penicillin-bind-
ing protein (PBP) cavity was studied. Theoretical predic-
tions could be experimentally confirmed by in vitro evalua-
tions against R39 DD-peptidase,[7] which is the commonly
Scheme 3. Synthesis of bis-alkylated azetidinones 2 followed by a RCM
À
À
reaction:
a) NaH,
Br
(CH2)n CH=CH2,
N,N-dimethylformamide
(DMF), 0 to 208C, 12 h; b) Grubbs II catalyst (2ꢅ5 mol%), CH2Cl2,
408C, 12 h; c) H2, Pd/C catalyst, MeOH, 208C, 3 h.
426
www.chemasianj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2012, 7, 425 – 434

Penicillin-Binding Protein Inhibitors
thyldisilazide (LiHMDS), KOH in the presence of
Bu4NHSO4, and NaI), without improvement in the yield.
The use of an excess of base led to racemization of the bis-
alkyl derivatives 2. The bis-alkyl derivative 2 with n=2 was
not obtained, probably because HBr elimination occurred
from 1-butenyl bromide. With these precursors in hand, we
could validate the RCM strategy: using the second-genera-
tion Grubbs catalyst, macrocycles 3b–f were readily formed
and isolated in good yields after chromatographic purifica-
tion (step b in Scheme 3), thereby leading to 12- to 22-mem-
bered rings. The bis-alkyl derivative 2a did not cyclize; the
expected 8-membered ring bicycle is too small. Catalytic hy-
drogenation led to the corresponding saturated macrocycles
4b–f in high yields (step c in Scheme 3).
droxylamine and N,N’-dicyclohexylcarbodiimide (DCC), in
87% yield. Intramolecular cyclization by the method pro-
posed by Miller et al.[12] afforded b-lactam 8 in 64% yield.
Subsequent hydrogenation in the presence of Raney-Ni
gave the desired chiron 9 in quantitative yield.
Similar to the Boc family, the bis-alkyl derivatives 10
were synthesized in one step, from 9 in moderate yields.
Under the conditions used for the RCM reaction, precursors
10b–d (but not 10a) cyclized into bicycles 11b–d in good
yields, thus affording 12- to 16-membered rings. Catalytic
hydrogenation led to the saturated bicycles 12b–d.
(Scheme 5)
As the N-unprotected derivatives B (X=Y=H,H; R=H)
were desirable for biological evaluation and/or further deri-
vatization with the side chain of penicillin, we attempted to
remove the Boc group from precursors 2 and bicycles 3 and
4. Several conditions were tested (trifluoroacetic acid (TFA)
in CH2Cl2, HCl 2m solution, ceric ammonium nitrate (CAN)
in CH3CN at reflux,[10] or trimethylsilyl chloride (TMSCl) in
the presence of NaI[11]) without success; in all cases degrada-
tion of the b-lactam ring was observed (IR and NMR spec-
troscopy analyses).
We tried to replace the Boc protecting group with the
penicillin V side chain (V=PhOCH2CO) in situ by direct
acylation (with phenoxyacetyl chloride and NEt3 or ethyldii-
sopropylamine (DIEA)) of the crude mixtures resulting
from the treatment of 2, 3, and 4 under the above-men-
tioned deprotection conditions. Again, the results were dis-
appointing with the recovery of intractable mixtures. Hence,
to access the bis-alkylated bicycles B with the V side chain
(X=Y=H,H; R=PhOCH2CO) we restarted the total syn-
thesis from l-serine (Scheme 4).
Scheme 5. Synthesis of bis-alkylated azetidinones 10 followed by a RCM
À
À
reaction: a) NaH, Br ACTHNUTRGNEUN(G CH2)n CH=CH2, DMF, 08C to 208C, 12 h;
b) Grubbs II catalyst (2ꢅ5 mol%), CH2Cl2, 408C, 12 h; c) H2, Pd/C cata-
lyst, MeOH, 208C, 3 h.
All precursors (2, 10) and bicycles (3–4, 11–12) were char-
acterized by IR and NMR spectroscopy and MS (see the
Experimental Section and the Supporting Information). In
particular, MS was useful for detecting the possible occur-
rence of side products that resulted from (cyclo)oligomeriza-
tions and/or double-bond migrations (in precursors), leading
to cyclic products with the formal extrusion of one CH2
unit. Only the 18- and 22-membered-ring bicycles, these are,
compounds 3e–f, and consequently, 4e–f, were contaminat-
ed with small amounts of lower homologues (17- and 21-
membered-ring bicycles, respectively).
The NMR spectra confirmed our hypothesis that 1,3-
bridged b-lactams embedded into large rings were endowed
with certain conformational adaptability, as discussed in the
next section.
Computational Chemistry: Conformational Study—Heat of
Formation
Scheme 4. Synthesis of chiron 9: a) phenoxyacetyl chloride, a saturated
aqueous solution of NaHCO3, CH3CN, RT, 12 h; b) N,N-dicyclohexylcar-
bodiimide (DCC), NH2OBn (Bn=benzyl), tetrahydrofuran (THF), 08C
to RT, 12 h; c) PPh3, CCl4, NEt3, CH3CN, 08C to RT, 12 h; d) H2, Raney-
Ni, MeOH/EtOAc, RT, 12 h. V=PhOCH2CO.
Many conformers can exist for the monocyclic molecules 2
and 10, as well as for bicyclic molecules 3, 4, 11, and 12. In
solution, the coexistence of conformers for these compounds
was experimentally detected by NMR spectroscopy;
13C NMR spectra of the compounds were recorded in
[D2]1,1,2,2-tetrachloroethane at different temperatures. At
308C, many signals were visible because some carbon atoms
gave rise to splitting of the signals. Increasing the tempera-
ture to 1208C led to signals coalescence, and thus, allowed
structural assignment. For example, for compound 12d, the
quaternary carbon atoms (Cq) at d=167.8 (C=O), 165.3
l-Serine (5) was acylated under Schotten–Baumann con-
ditions to give (S)-3-hydroxy-2-(2-phenoxyacetamido)propa-
noic acid (6) in 71% yield. This compound was converted
into the corresponding hydroxamate 7, using O-benzylhy-
Chem. Asian J. 2012, 7, 425 – 434
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
427

G. Dive, J. Marchand-Brynaert, and A. Sliwa
FULL PAPERS
Table 1. Relative energies (E) of the i and ii conformations of the precursors/bicycles and respective heats of
formation (H) resulting from cyclization.
(C=O), and 158.0 ppm (PhO) at
1208C appeared as multiple sig-
nals at 308C. The methylene
peak of the V side chain at d=
67.9 ppm at 1208C presented
two distinct signals at 308C.
The C3 carbon of the b-lactam
ring at d=62.2 ppm at 1208C
gave two very well spaced sig-
nals at 308C. (The 13C spectra
of compound 12d, as well as
10d and 11d, can be found in
the Supporting Information.)
The geometry of all mole-
cules was fully optimized at the
RHF level by using the minimal
basis set MINI-1’.[13]
Precursor
Product
Geometry
Eprecursors [kcalmolÀ1
]
Ebicycles [kcalmolÀ1
]
Hbicycles [kcalmolÀ1
]
i
ii
i
ii
i
ii
2b
3b
a
b
a
b
a
b
a
b
a
b
a
b
a
b
a
b
1.39
0
0.31
0
6.54
7.61
5.66
6.05
8.30
7.50
11.50
11.32
5.67
5.37
6.47
7.14
4.19
1.78
7.94
7.78
2c
3c
2.16
2.17
2.95
3.01
2.95
2.95
0.01
0.00
0.10
0.00
2.47
2.99
3.79
3.49
0.44
0.00
0.94
0.00
2.92
0.00
2.84
3.23
2.66
1.93
7.83
7.66
0.33
0.00
0.00
0.58
1.88
0.00
1.45
0.99
0.29
0.00
0.00
0.18
0.00
0.27
4.83
4.98
7.64
8.76
3.69
6.88
2d
3d
2e
3e
2 f
3 f
10b
10c
10d
11b
11c
11d
0.96
0.00
2.54
0.00
1.11
0.22
0.68
0.00
4.92
4.99
8.42
10.28
In the case of bis-alkylated
precursors 2 and 10 (X=Y=
H,H), the molecules are confor-
mationally less constrained than the bis-acylated compounds
(X=Y=O)[6] because the carbonyl groups are replaced with
methylene groups, which can accommodate more conforma-
tions. Depending on the nature of the N substituent (Boc or
V), a large number of local minima could be trapped.
For all of the compounds studied, two conformations, “i”
and “ii”, of the bridging cycle (of compounds 3, 11, 4, and
12) have been located with respect to the b-lactam ring.
Conformation i expands the cycle to the upper-right corner
of the b-lactam C4; in conformation ii, the cycle is more ori-
entated to the carbonyl C2 of the b-lactam. For each confor-
mation, the carbonyl of the side chain (Boc or V) can rise
above the four-membered ring (“a” conformation) or below
(“b” conformation) (Figure 1). For unsaturated compounds
3 and 11, the trans configuration of the substituted ethylene
has been considered.
ring has a significant impact on the conformations for the
smallest (12-membered ring) and the largest ones (22-mem-
bered ring). For compounds 3b and 11b, only the i confor-
mation can be trapped. In 3 f the ring is so large that it ex-
pands on both sides of the b-lactam in a pseudo-i conforma-
tion.
As the bis-alkylated precursors 2 and 10 are highly flexi-
ble, their cyclization can easily lead to desired compounds 3
and 11. Similar reactions were not possible for the bis-acy-
lated precursors.[6]
Reactivity versus Serine Enzyme Models
The reactivity of the bridged molecules was studied by using
a simple model of the PBP cavity (Scheme 6) at the RHF/
MINI-1’ level (Table 2).
In this model, the ring open-
ing of the b-lactam occurs
through a concerted process:
the nucleophilic serine is mim-
icked by 2-(formyl)amino-1-eth-
anol, interacts with methyla-
mine, which acts as a proton
relay to methanol, and transfers
Scheme 6. Model of concerted
nucleophilic attack on the b-
lactam ring.
the proton to the b-lactam ni-
trogen.[14]
The
formamide
moiety mimics oxyanion hole
Figure 1. Conformers ib and iia of compound 4d.
stabilization. At the transition state (TS), this pseudo eight-
membered ring is described by the reaction coordinate asso-
ciated to the negative curvature of the second-derivative
energy matrix.
The results obtained with Boc and V side chains present
some common features concerning the lowest energy barri-
ers: the optimum size of the cycle is a 16-membered ring
(i.e., 3d–4d and 11d–12d), while the shorter and the bigger
ones enhance the energy barrier calculated with the model.
In most of the cases, the activation energy is also higher for
The heat of formation of 3 and 11 has been computed
with respect to the open precursor with the same conforma-
tion as that of the corresponding cyclized molecule. For the
Boc and V side chains, the 4 conformations lie in the same
range of stability; the relative energies are less than 8 kcal
molÀ1 in the case of the ii conformations, which are often
more stable than those of i type (Table 1). The size of the
428
www.chemasianj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2012, 7, 425 – 434

Penicillin-Binding Protein Inhibitors
Table 2. Activation energy of concerted nucleophilic attack (MINI-1’).
open precursor (2d; Table 3,
entry 4). Some activities on
PBP2a were also observed for
other compounds of the Boc
family (Table 3, entries 2, 3, 11,
15). On the other hand, none of
the V side-chain molecules had
a significant activity on the R39
DD-peptidase or on PBP2a and
PBP5. Only the saturated 16-
membered ring of the Boc
family (4d) had a high activity
on the R39 DD-peptidase. To
understand this phenomenon,
Unsaturated
compound
Saturated
compound
Geometry
DE unsaturated [kcalmolÀ1
]
DE saturated [kcalmolÀ1
]
i
ii
i
ii
3b
4b
a
b
a
b
a
b
a
b
a
b
a
b
a
b
a
b
27.86
27.18
22.79
20.89
18.90
16.86
20.84
19.13
28.82
28.62
21.96
26.36
26.87
23.95
21.47
15.53
9.33
28.37
31.78
25.15
21.86
19.48
17.96
20.19
19.17
28.80
28.86
23.42
30.87
29.36
23.11
21.22
16.52
3c
4c
25.12
23.21
20.47
20.55
19.66
17.71
26.31
25.28
23.76
22.41
20.55
18.43
3d
4d
3e
4e
3 f
4 f
11b
11c
11d
PenG
12b
12c
12d
25.35
22.99
18.95
18.95
24.53
21.14
19.82
18.38
more
elaborate
reactivity
models of the active site were
built.
the saturated cycle with respect to the corresponding unsa-
turated one (with the trans configuration at the C=C bond).
Table 3. Evaluation of bis-alkylated azetidinones against R39 DD-pepti-
dase, PBP5, and PBP2a.
Entry
n
R39
PBP5
PBP2a
RA [%]
RA [%]
RA [%]
Inhibition of R39, PBP2a, and PBP5
1
2
3
4
5
6
7
8
2a
2b
2c
2d
2e
2 f
10a
10b
10c
10d
3b
3c
3d
3e
3 f
11b
11c
11d
4b
4c
4d
4e
4 f
12b
12c
12d
1
3
4
5
6
8
1
3
4
5
3
4
5
6
8
3
4
5
3
4
5
6
8
3
4
5
101Æ4
>100
81
94
69
96
71
68
59
95
100
96
95
98
91
53
96
58
96
69
89
97
93
98
88
61
89
90
97
100
100
All products were evaluated for their potential inhibition
effect on bacterial serine enzymes. R39 from Actinomadura
is a model serine enzyme of low-molecular-weight DD-pep-
tidases, which are usually considered for preliminary screen-
ing of penicillin-like compounds. R39 and the tested b-
lactam (100 mm) were incubated (1 h, 258C). Then the
enzyme residual activity (RA) was determined by observing
the hydrolysis of the thioester S2d substrate[15] in the pres-
ence of 5,5’-dithiobis-(2-nitrobenzoic acid) (DTNB), which
labeled the formed thiol, and reading at 412 nm. The results
are given in Table 3 as percentages of RA. The activity in
the absence of inhibitor was set at 100%, and therefore, low
values indicated a very active compound because the bacte-
rial enzyme was inhibited by the tested compound, and con-
sequently, could not hydrolyze its substrate. A tested com-
pound was considered a “hit” (i.e., potential inhibitor) when
RA was less than 80%. All of the compounds were also
evaluated against two high-molecular-weight DD-peptidases
responsible for bacterial resistance to b-lactam antibiotics:
PBP2a from methicillin-resistant S. aureus (MRSA) and
PBP5 from E. faecium. The tested b-lactams (1 mm) were
incubated with the PBPs (4 h, 308C), then fluorescein-la-
beled ampicillin was added to detect the residual activity.
This reagent is an inhibitor that forms a stable acyl–enzyme
intermediate. After denaturation and sodium dodecyl sulfate
polyacrylamide gel electrophoresis (SDS-PAGE) separation
of the acylated enzyme from the reagent band, fluorescence
was measured. The fluorescence intensity was proportional
to the residual active protein, that is, protein not acylated by
the tested compound.
97Æ11
80Æ4
97Æ2
101Æ1
103Æ8
97Æ3
101Æ1
100Æ4
>100
103Æ9
83Æ4
98Æ1
97Æ3
102Æ4
101Æ3
104Æ4
>100
96Æ11
52Æ3
95Æ1
96Æ1
101Æ3
97Æ4
105Æ9
80
100
100
100
90
100
96
68
78
74
99
72
95
100
100
96
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
89
61
100
86
100
100
99
Building the Models
The R39 active site is composed of the three conserved
motifs found in PBP and b-lactamases, as highlighted from
X-ray data.[7] The first motif connects Ser49 (a nucleophilic
serine) and Lys52 with Asn50 and Met51. Remarkably, the
conformation of the backbone is stabilized by a hydrogen
bond between the carbonyl group of the Ser49 backbone
and the NH group of Lys52, thus allowing the lysine residue
to extend in such a way that the amino group Nz lies in the
vicinity of the Og of Ser49. The second motif is formed by
Interestingly, the lowest RA values were found for Boc
molecules with saturated (4d; Table 3, entry 21) and unsatu-
rated (3d; Table 3, entry 13) 16-membered rings, and their
Chem. Asian J. 2012, 7, 425 – 434
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
429

G. Dive, J. Marchand-Brynaert, and A. Sliwa
FULL PAPERS
Ser298, Asn299, and Asn300. Ser298 is connected to the
amino group of Lys52 residue. Owing to turns in the confor-
mations of Asn299 and Asn300, the NH of Asn300 interacts
with the ligand carbonyl side chain. The third motif is
formed by Lys410, Thr411, Gly412, and Thr413. Several in-
teractions stabilize its conformation. The carbonyl backbone
of Thr413 makes a hydrogen bond with the NH ligand side
chain.
The NH backbone of Thr413 interacts with the oxyanion
of the ligand, while the amino group of the Lys410 side
chain and the OH group of Thr411 stabilize the carboxylic
group of penicillin-type antibiotics. Gly416 starts the b4
sheet with Val417 and Ser418 parallel to the b3 one. The
bottom of the cavity is delimited by Gly348, Leu349, Ser350,
and Arg351 on one side and by Ala146, Tyr147, and Ser148
on the other. As depicted in the 2D drawing (Scheme 7),
the side chains of Arg351, Leu349, and Tyr147 could interact
with the side chain of the ligand. They could also give rise
to steric hindrance with part of the ligand.
Figure 2. a) Pen in the first model. b) Pen in the second model. c) Pen in
the third model. d) Compound 4d in conformation ib in the second
model. Legend: C of Pen or 4d in purple, others in yellow; H in gray; O
in red; N in blue; hydrogen bonds as thin lines.
complete protein, but to analyze the geometrical constraints
resulting from the models. From a geometric point of view,
the position of the thiazolidine ring of penicillin (or the tet-
rahydrothiazine ring of cephalosporin; results not shown)
lies at the entrance of the cavity. This feature could be relat-
ed to the fact that many DD-peptidases can easily accom-
modate large b-lactam antibiotics, such as tricyclic carbape-
nems.[16] A second important geometry constraint is related
to the conformation of the third motif, which defines the ac-
cessible volume above the b-lactam ring.
The three TS structures have been located as first-order
saddle points, for which the imaginary frequency describes
the motion of hydrogen between Ser49 and Lys52, Lys52
and Ser298, and Ser298 to the nitrogen of the b-lactam ring.
Remarkably, the three equilibrium structures are nearly su-
perimposable (cf. the simple model in Scheme 6) when con-
sidering the eight-membered ring formed by the proton
shuttle.
Additional calculations for 4d were performed and the
TS structures were located for the second (143 atoms, 407
basis functions) and third (201 atoms, 573 basis functions)
models when using ib and iib conformations as starting geo-
metries. It appears that conformation iib has important
steric hindrance with the third motif and that only confor-
mation ib could accommodate the geometry of the active
site model (Figure 2d). The goodness of fit between this
structure and that obtained with Pen could partially explain
the RA value of 52% observed with 4d on R39.
Scheme 7. The bottom of the R39 cavity; the fragment in gray is part of
the ligand with a side chain on C3.
Three models have been built by increasing their com-
plexity to locate the transition structure for penicillin with a
side chain limited to a formamide group (referred to as
Pen). The first model contains the 49 to 52 amino acids of
the first motif and methanol, which mimics Ser298 (82
atoms, 250 basis functions) in the simple model (Figure 2a).
In the second model, the second motif has been added with
298 to 300 amino acids (110 atoms, 342 basis functions; Fig-
ure 2b). Finally, the inclusion of 410 to 413 amino acids of
the third motif constitutes the third model formed by 168
atoms and 508 basis functions (Figure 2c; hydrogen atoms
have been deleted for clarity). Model three, with hydrogen
atoms and hydrogen bonds, is presented in the Supporting
Information. For clarity, the point of view is also rotated
around the y axis to show that motif three lies above the b-
lactam ring.
The aim of these calculations is not, at the present stage,
to determine an energy barrier that could be representative
of the energy involved in the enzymatic reaction with the
430
www.chemasianj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2012, 7, 425 – 434

Penicillin-Binding Protein Inhibitors
One question remains with regards to the lack of activity
of 12d. This molecule can adopt two conformations of the
phenoxy side chain, for which the ib conformation of the
bridged cycle differs only by 1.52 kcal. Both conformations
have been superimposed on conformation ib of 4d
(Figure 3). In one conformation, the phenoxy group has
steric hindrance with Tyr147 phenol; in the other, the phen-
oxy group is in contact with Leu349 and Arg351 side chains
at a distance that is too short. This geometric feature, occur-
ring in both conformations, could be related to the biologi-
cal results because 12d cannot accommodate the active-site
model geometry, whereas 4d can.
tionality only plays a role in stabilizing the conformation of
penicillin-like molecules by hydrogen bonds to residues of
the third motif.
Experimental Section
General
Experiments were performed under an argon atmosphere in flame-dried
glassware. All solvents, including anhydrous solvents, and reagents were
purchased from Acros Organics, Alfa Aesar, Fluka, Sigma–Aldrich, or
VWR, and used without any further purification. TLC analyses were per-
formed on aluminum plates coated with silica gel 60F254 (Merck) and vi-
sualized with a solution of KMnO4 and UV (254 nm) detection. Column
chromatography was performed on silica gel (40–63 or 63–200 mm) pur-
chased from Rocc. Melting points were determined on a Bꢆchi B-540 ap-
paratus calibrated with caffeine, vanillin, and phenacetin. [a]D values
were measured on Perkin–Elmer 241 MC polarimeter at 208C. Concen-
trations are given in g/100 mL. NMR (1H and 13C) spectra were recorded
at 300 MHz for proton and 75 MHz for carbon (Bruker Avance 300 spec-
trometer) or 500 MHz for proton and 125 MHz for carbon (Bruker
Avance 500 spectrometer). Chemical shifts are reported in parts per mil-
lion relative to the residual solvent signal from the deuterated solvent or
relative to peak of tetramethylsilane (TMS). NMR coupling constants (J)
are reported in hertz. IR spectra were recorded by using an FTIR-8400S
Shimadzu apparatus. Products were analyzed as thin films deposited on
an Se–Zn crystal by evaporation from solutions in CH2Cl2. Low-resolu-
tion mass spectra were obtained by using ThermoFinnigan LCQ Quan-
tum or FinniganMat TSQ7000 spectrometers. High-resolution mass spec-
trometry (HRMS) analyses were performed at University College
London (UK).
Figure 3. Compounds 4d (black) and 12d (gray) in conformation ib.
Bis-alkyl Precursor 2d
NaH (60% dispersion in oil, 279 mg, 6.98 mmol) was added to a stirred
solution of 1 (0.65 g, 3.49 mmol) in dry DMF (41 mL) at 08C. Alkyl bro-
mide (1.17 mL, 7.68 mmol) was added after 30 min and the mixture was
then slowly warmed to room temperature and stirred overnight. The mix-
ture was quenched with water (30 mL) at 08C and then the medium was
diluted with brine (50 mL). The resulting mixture was extracted with
EtOAc (3ꢅ30 mL) and the organic layers were washed with brine
(70 mL). After drying over MgSO4 and removing the solvent under re-
duced pressure, the residue was purified by flash column chromatography
(hexane/EtOAc 4:1) to provide 2d as a pale-yellow oil (515 mg, 39%).
Rf =0.63 (hexane/EtOAc 3:2); [a]D =+3.0 (c=1.7 in CH3OH); 1H NMR
(300 MHz, CDCl3, 208C): d=5.84–5.71 (m, 2H), 5.01–4.90 (m, 4H+
0.4H; rotamer), 4.52 (brs, 0.6H; rotamer), 3.46–3.43 (m, 1H), 3.26–3.08
(m, 5H), 2.07–2.00 (m, 4H), 1.57–1.24 ppm (m, 21H); 13C (125 MHz,
CDCl3, 208C): d=167.1, 154.7, 138.9, 138.7, 114.7, 114.5, 80.9, 80.5, 63.2,
62.9, 47.9, 47.1, 46.7, 41.7, 33.8, 33.7, 29.6, 29.0, 28.7, 28.5, 27.5, 26.5,
26.4 ppm; IR: n˜ =2974–2856, 1753, 1693, 1639, 1456 cmÀ1; MS (ESI): m/z
(%): 779 (12) [2M+Na]+, 401 (100) [M+Na]+; HRMS (ESI): m/z calcd
for C22H38N2O3Na: 401.2780 [M+Na]+; found: 401.2778.
Conclusion
In the bis-acylated family previously reported (X=Y=O;
Scheme 2), cyclodimers were obtained when applying the
RCM reaction as a key step for macrocyclization. Herein,
this synthetic strategy was successful and led to cyclomono-
mers of the bis-alkylated family (X=Y=H,H) owing to the
high conformational adaptability of the bis-alkylated precur-
sors. We synthesized 12- to 22-membered, 1,3-bridged b-lac-
tams for the Boc series and 12- to 16-membered bicyclic b-
lactams for the V side-chain series.
The reactivity of the 1,3-bridged b-lactams in the active
site of a serine enzyme depends greatly on the ligand con-
formation and explains why compounds with a V side chain
do not fit into the geometry of the catalytic cavity. This fea-
ture was analyzed by using different models of the R39
active site, including the three conserved motifs of this
enzyme family. Of the synthesized bicycles, compound 4d
was a good inhibitor of the R39 enzyme (as theoretically
predicted) and some compounds also exhibited significant
activity against resistant PBP2a from MRSA. This activity
could be related to the flexibility of the bicycles, which can
be accommodated into the enzyme active site.
Unsaturated Bicycle 3d
Grubbs II catalyst (56 mg, 66.09 mmol) was added to a stirred solution of
2d (0.50 g, 1.32 mmol) in dry CH2Cl2 (264 mL) and the solution was
stirred at reflux under argon for 4 h. Then a second addition of Grubbs II
catalyst (56 mg, 66.09 mmol) was made and the mixture was additionally
stirred at reflux for 20 h. The solvent was removed under reduced pres-
sure and the crude product was purified three times by column chroma-
tography (hexane/EtOAc 4:1) to provide 3d as
a pale-brown solid
(416 mg, 90%). Rf =0.43 (hexane/EtOAc 3:2); m.p. 90.1–90.98C; [a]D =
À24.4 (c=1.2 in CHCl3); 1H NMR (500 MHz, CDCl3, 208C): d=5.51
(brs, 0.6H; rotamer), 5.34–5.18 (m, 2H+0.4H; rotamer), 3.73–3.67 (m,
1H), 3.42–3.35 (m, 1H), 3.25–2.93 (m, 3H), 2.81–2.76 (m, 1H), 2.13–1.94
(m, 4H), 1.69–1.17 ppm (m, 21H); 13C (125 MHz, CDCl3, 208C): d=
167.6, 167.4, 155.1, 154.3, 131.9, 131.7, 130.4, 130.2, 80.6, 80.3, 62.5, 61.9,
The activity of the 1,3-bridged macrocycles derived from
bis-w-alkenyl-3(S)-aminoazetidinone precursors suggested a
way to design new lactam antibiotics with a planar amide
bond and devoid of carboxylic group. Indeed, this acid func-
Chem. Asian J. 2012, 7, 425 – 434
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
431

G. Dive, J. Marchand-Brynaert, and A. Sliwa
FULL PAPERS
46.9, 45.4, 45.1, 41.2, 41.0, 40.8, 32.2, 30.5, 28.8, 28.4, 28.1, 27.9, 27.2, 27.1,
26.9, 26.7, 26.2, 26.1, 25.1, 25.0, 24.8, 24.5 ppm; IR: n˜ =2924–2852, 1755,
1697, 1454 cmÀ1; MS (ESI): m/z (%): 373 (100) [M+Na]+, 317 (23)
[M+Na-C4H8]+, 295 (25) [M+H-C4H8]+; HRMS (ESI): m/z calcd for
C20H34N2O3Na: 373.2467 [M+Na]+; found: 373.2455.
(2.50 g, 7.26 mmol) in dry CH3CN (20 mL), containing CCl4 (0.77 mL,
7.98 mmol) and anhydrous triethylamine (1.52 mL, 10.89 mmol). The re-
sultant reaction mixture was stirred for 2 h at 08C and then overnight at
room temperature. After completion of the reaction, the obtained white
precipitate (OPPh3) was separated by filtration and the resulting clear re-
action mixture was concentrated under reduced pressure. The residue
was dissolved with EtOAc (50 mL), washed with a saturated aqueous so-
lution of NH4Cl (2ꢅ40 mL) and brine (40 mL). After drying over MgSO4
and removing the solvent under reduced pressure, the residue was puri-
fied by flash column chromatography (hexane/EtOAc 1:1) to provide 8
as a white solid (1.51 g, 64%). Rf =0.41 (hexane/EtOAc 4:6); m.p. 138.8–
139.48C; [a]D =À6.2 (c=1.0 in (CH3)2CO); 1H NMR (300 MHz,
(CD3)2CO, 208C): d=8.19 (brd, J=7.0 Hz, 1H), 7.49–7.28 (m, 7H), 7.01–
6.96 (m, 3H), 4.98 (s, 2H), 4.91–4.85 (m, 1H), 4.53 (brd, J=1.4 Hz, 2H),
3.72–3.69 (m, 1H), 3.53 ppm (dd, J=2.4, 4.3 Hz, 1H); 13C (75 MHz,
(CD3)2CO, 208C): d=169.1, 163.4, 158.7, 136.5, 130.4 (2C), 130.0 (2C),
129.5, 129.3 (2C), 122.4, 115.6 (2C), 77.9, 67.9, 52.6, 52.4 ppm; IR: n˜ =
3363–3193, 1776, 1677, 1598, 1531, 1495 cmÀ1; MS (ESI): m/z (%): 365 (8)
[M+K]+, 349 (100) [M+Na]+, 327 (10) [M+H]+; HRMS (ESI): m/z
calcd for C18H18N2O4Na: 349.1164 [M+Na]+; found: 349.1151.
Saturated Bicycle 4d
10% Pd/C (10 mg) was added to a stirred solution of 3d (110 mg,
0.31 mmol) in methanol (10 mL). After being stirred under a hydrogen
atmosphere (P=1 atm) for 3 h at room temperature, the mixture was fil-
tered through a short pad of Celite and concentrated under reduced pres-
sure. The residue was purified by column chromatography (hexane/
EtOAc 4:1) to provide 4d as a white solid (108 mg, 98%). Rf =0.48
(hexane/EtOAc 3:2); m.p. 55.4–56.28C; [a]D =À24.9 (c=1.0 in CHCl3);
1H NMR (500 MHz, CDCl3, 208C): d=5.44 (brs, 0.6H; rotamer), 5.18
(brs, 0.4H; rotamer), 3.68 (brs, 1H), 3.43 (brs, 1H), 3.26–2.99 (m, 3H),
2.84–2.79 (m, 1H), 1.75–1.69 (m, 1H), 1.59–1.20 ppm (m, 28H); 13C
(125 MHz, CDCl3, 208C): d=167.4, 167.1, 155.2, 155.1, 80.5, 62.6, 62.1,
48.0, 47.7, 45.1, 41.9, 29.1, 28.5, 28.2, 26.9, 26.8, 26.6, 26.5, 26.0, 25.4 ppm;
IR: n˜ =2924–2854, 1757, 1697, 1458 cmÀ1; MS (ESI): m/z (%): 375 (100)
[M+Na]+, 319 (10) [M+Na-C4H8]+, 297 (26) [M+H-C4H8]+; HRMS
(ESI): m/z calcd for C20H36N2O3Na: 375.2624 [M+Na]+; found: 375.2621.
N-[(3S)-2-Oxoazetidin-3-yl]-2-phenoxyacetamide (9)
Compound 8 (1.00 g, 3.06 mmol) was dissolved in methanol (18 mL) and
EtOAc (18 mL) and placed under H2 (1 atm) at room temperature in the
presence of Raney-Ni (50% in water) catalyst for 12 h. Then the mixture
was filtered through a pad of Celite and concentrated under reduced
pressure to provide 9 as a white solid (0.67 g, 100%), which was used
without further purification. Rf =0.42 (EtOAc/MeOH 95:5); m.p. 155.7–
156.68C; [a]D =À19.1 (c=1.0 in CH3OH); 1H NMR (500 MHz, CD3OD,
208C): d=7.32–7.28 (m, 2H), 7.01–6.97 (m, 3H), 5.04 (dd, J=2.7, 5.4 Hz,
1H), 4.54 (brd, J=1.2 Hz, 2H), 3.58–3.56 (m, 1H), 3.39 ppm (dd, J=2.6,
5.5 Hz, 1H); 13C (125 MHz, CD3OD, 208C): d=171.5, 170.8, 159.2, 130.6
(2C), 122.9, 115.9 (2C), 68.1, 57.8, 44.1 ppm; IR: n˜ =3281–3020, 1776,
ACHTUNGTRENNUNG
Phenoxyacetyl chloride (6.57 mL, 47.58 mmol) was added to a solution of
5
(5.00 g, 47.58 mmol) in saturated aqueous solution of NaHCO3
a
(200 mL) and MeCN (40 mL). The reaction mixture was stirred vigorous-
ly overnight at room temperature and the aqueous phase was extracted
with diethyl ether (2ꢅ200 mL). The aqueous solution was acidified to
pH 2–3 with HCl 36% and extracted with EtOAc (4ꢅ200 mL). The or-
ganic layer was dried over MgSO4 and concentrated under reduced pres-
sure. The crude product was dissolved in a minimum of EtOAc and
hexane was added until the appearance of a precipitate. After 5–6 h the
suspension was filtered, washed with hexane, and dried under reduced
pressure to provide 6 as a white solid (8.08 g, 71%); m.p. 30.1–130.98C;
[a]D = +30.4 (c=1.1 in (CH3)2CO); 1H NMR (300 MHz, (CD3)2CO,
208C): d=7.62–7.60 (brd, J=6.5 Hz, 1H), 7.35–7.30 (m, 2H), 7.04–6.98
(m, 3H), 4.65–4.60 (m, 1H), 4.55 (s, 2H), 4.06–4.00 (dd, J=4.1, 11.1 Hz,
1H), 3.95–3.90 ppm (dd, J=3.6, 11.1 Hz, 1H); 13C (75 MHz, (CD3)2CO,
208C): d=171.7, 168.6, 158.7, 130.5 (2C), 122.5, 115.7 (2C), 68.0, 62.7,
55.0 ppm; IR: n˜ =3393–3060, 2754–2469, 1731, 1643, 1539, 1495 cmÀ1; MS
(ESI): m/z (%): 262 (100) [M+Na]+, 240 (38) [M+H]+, 222 (52)
[M+HÀH2O]+; HRMS (CI): m/z calcd for C11H14NO5: 240.08720
[M+H]+; found: 240.08649.
1703, 1664, 1628, 1556, 1498 cmÀ1
; MS (ESI): m/z (%): 243 (100)
[M+Na]+; HRMS (CI): m/z calcd for C11H13N2O3: 221.09262 [M+H]+,
found: 221.09253.
Bis-alkyl Precursor 10d
Compound 10d was prepared by the same procedure as that described
for 2d. Yield: 30%; Rf =0.47 (hexane/EtOAc 1:1); [a]D =+3.1 (c=1.0 in
1
CHCl3); H NMR (500 MHz, C2D2Cl4, 208C): d=7.33 (t, J=7.7 Hz, 2H),
7.03 (t, J=7.3 Hz, 1H), 6.93 (brd, J=8.0 Hz, 2H), 5.84–5.75 (m, 2H),
5.12–4.95 (m, 4H+0.3H; rotamer), 4.76–4.62 (m, 2H+0.7H; rotamer),
3.53–3.10 (m, 6H), 2.07–2.03 (m, 4H), 1.71–1.52 (m, 4H), 1.44–1.25 ppm
(m, 8H); 13C (125 MHz, C2D2Cl4, 908C): d=167.9, 165.1, 157.9, 138.3,
129.4, 121.7, 114.8, 114.4, 67.6, 62.9, 46.8, 41.9, 33.2, 28.3, 28.2, 27.1, 26.3,
26.1 ppm; IR: n˜ =3003–2854, 1748, 1668, 1599, 1587, 1497 cmÀ1; MS
(ESI): m/z (%): 435 (100) [M+Na]+, 413 (6) [M+H]+, 288 (10); HRMS
(ESI): m/z calcd for C25H36N2O3Na: 435.2624 [M+Na]+; found: 435.2625.
ACHTUNGTRENNUNG(2S)-N-(Benzyloxy)-3-hydroxy-2-(2-phenoxyacetamido)propanamide (7)
A solution of DCC (2.72 g, 13.17 mmol) in THF (10 mL) was added
dropwise, at 08C, to a well-stirred solution of 6 (3.00 g, 12.54 mmol) and
O-(phenylmethyl)hydroxylamine (1.46 mL, 12.54 mmol) in THF
(120 mL). The reaction mixture was stirred for 1 h at 08C and overnight
at room temperature. The obtained white precipitate was separated by
filtration and the resulting clear reaction mixture was concentrated under
reduced pressure. After addition of Et2O (100 mL), the precipitate was
filtered, washed with Et2O, and dried under reduced pressure to provide
Unsaturated Bicycle 11d
Compound 11d was prepared by the same procedure as that described
for 3d. Yield: 67%; Rf =0.42 (hexane/EtOAc 3:7); m.p. 86.7–87.38C;
[a]D =À2.8 (c=1.9 in CHCl3); 1H NMR (500 MHz, C2D2Cl4, 208C): d=
7.34–7.30 (m, 2H), 7.02 (t, J=7.3 Hz, 1H), 6.93–6.91 (m, 2H), 5.84–5.74
(m, 0.4H; rotamer), 5.37–5.16 (m, 2H+0.6H; rotamer), 4.79–4.62 (m,
2H), 3.72–3.66 (m, 1H), 3.55–3.08 (m, 4H), 2.88–2.66 (m, 1H), 2.16–1.85
(m, 4H), 1.73–1.19 ppm (m, 12H); 13C (125 MHz, C2D2Cl4, 1208C): d=
167.8, 165.5, 158.0, 131.8, 131.1, 130.3, 130.1, 129.4, 121.7, 115.0, 67.9,
62.3, 49.1, 47.1, 46.7, 45.2, 44.2, 41.8, 33.7, 31.7, 31.5, 31.3, 30.5, 28.4, 28.2,
28.0, 27.8, 27.4, 27.2, 27.1, 26.5, 26.1, 25.5, 25.2, 24.6 ppm; IR: n˜ =3003–
2853, 1740, 1664, 1497 cmÀ1; MS (ESI): m/z (%): 407 (100) [M+Na]+,
385 (89) [M+H]+, 357 (21) [M+HÀCO]+; HRMS (ESI): m/z calcd for
C23H32N2O3Na: 407.2311 [M+Na]+; found: 407.2322.
7
as a white solid (3.76 g, 87%). Rf =0.30 (CH2Cl2/MeOH 95:5);
m.p. 145.0–145.78C; [a]D =À1.5 (c=1.0 in (CH3)2SO); 1H NMR
(300 MHz, (CD3)2SO, 208C): d=11.33 (brs, 1H), 8.01 (brd, J=7.9 Hz,
1H), 7.39–7.28 (m, 7H), 6.99–6.95 (m, 3H), 5.08 (brs, 1H), 4.78 (s, 2H),
4.55 (s, 2H), 4.23–4.29 (m, 1H), 3.59 ppm (brd, J=5.5 Hz, 2H); 13C
(75 MHz, (CD3)2SO, 208C): d=167.6, 166.7, 152.7, 135.8, 129.5 (2C),
128.9 (2C), 128.3 (3C), 121.2, 114.7 (2C), 76.9, 66.6, 61.4, 52.8 ppm; IR:
n˜ =3313–2999, 1650, 1556, 1499 cmÀ1; MS (ESI): m/z (%): 367 (100)
[M+Na]+, 345 (30) [M+H]+, 222 (12) [M+HÀNH2OBn]+; HRMS (CI):
m/z calcd for C18H21N2O5: 345.14505 [M+H]+; found: 345.14649.
N-[(3S)-1-(Benzyloxy)-2-oxoazetidin-3-yl]-2-phenoxyacetamide (8)
Saturated Bicycle 12d
A solution of PPh3 (2.09 g, 7.98 mmol) in dry MeCN (35 mL) was added
dropwise, at 08C and under an Ar atmosphere, to a stirred solution of 7
Compound 12d was prepared by the same procedure as that described
for 4d. Yield: 93%; Rf =0.33 (hexane/EtOAc 1:1); m.p. 85.2–85.78C;
432
www.chemasianj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2012, 7, 425 – 434

Penicillin-Binding Protein Inhibitors
1
[a]D =À1.0 (c=1.7 in CHCl3); H NMR (500 MHz, C2D2Cl4, 1208C): d=
Acknowledgements
7.40–7.36 (m, 2H), 7.10–7.03 (m, 3H), 5.46 (brs, 1H), 4.83–4.74 (m, 2H),
4.05 (brs, 0.4H; rotamer), 3.81–3.75 (m, 1H), 3.60–3.50 (m, 3H+0.6H;
rotamer), 3.38–3.36 (m, 1H), 2.93–2.89 (m, 1H), 2.03–1.47 ppm (m,
19H); 13C (125 MHz, C2D2Cl4, 1208C): d=167.8, 165.3, 158.0, 129.3,
121.7, 115.0, 67.9, 62.2, 49.1, 47.7, 45.0, 42.1, 33.7, 28.8, 27.7, 27.0, 26.9,
26.8, 26.3, 26.2, 26.1, 25.6, 25.5, 24.6 ppm; IR: n˜ =3028–2856, 1740, 1662,
1437 cmÀ1; MS (ESI): m/z (%): 409 (100) [M+Na]+, 387 (25) [M+H]+,
359 (7) [M+HÀCO]+; HRMS (ESI): m/z calcd for C23H34N2O3Na:
409.2467 [M+Na]+; found: 409.2439.
This work was supported by the Interuniversity Attraction Pole (IAP P6/
19 PROFUSA), F.R.S.-FNRS, UCL, and ULg (computational facilities).
G.D. and J.M.-B. are senior research associates of the F.R.S.-FNRS (Bel-
gium). Dr. Astrid Zervosen is acknowledged for R39 testing and Dr. Ana
Amoroso and Olivier Verlaine for PBP2a and PBP5 testing. Raoul Ro-
zenberg and Dr. Cꢀcile Le Duff contributed to structural analyses.
[1] a) E. P. Abraham, E. Chain, Nature 1940, 146, 837; b) C. Walsh,
Nature 2000, 406, 775–781; c) D. M. Livermore, J. Antimicrob. Che-
mother. 2009, 64 (suppl 1), i29–i36; d) G. L. French, Int. J. Antimi-
crob. Agents 2010, 36S3, S3–S7.
[2] C. Hubschwerlen in Comprehensive Medicinal Chemistry, 2nd edi-
tion (Eds.: J. B. Taylor, D. J. Triggle), Elsevier, Oxford, 2007, chap-
ter 7.17, pp. 479–517.
[3] a) J. L. Strominger, Antibiotics 1967, 1, 705–713; b) R. B. Woodward
in The Chemistry of Penicillin (Eds.: H. T. Clarke, J. R. Johnson, R.
Robinson), Princeton University Press, Princeton, New Jersey, 1949,
pp. 440–454; c) B. Alcaide, C. Aragoncillo, P. Almendros in Com-
prehensive Heterocyclic Chemistry III, Vol. 2 (Ed.: C. Stevens),
Elsevier, Oxford, 2008, chapter 2.02, pp. 111–171; d) J. Marchand-
Brynaert, C. Brulꢀ in Comprehensive Heterocyclic Chemistry III,
Vol. 2 (Ed.: C. Stevens), Elsevier, Oxford, 2008, chapter 2.03,
pp. 173–237.
Computational Chemistry
All of the calculations were performed with the Gaussian 03 suite of pro-
grams.[17] The geometry was optimized by analytical gradient energy min-
imization. The nature of the located extrema was defined by the inertia
of the second-derivative energy (Hessian matrix). For the minima, all of
the eigenvalues were positive; in the case of a first-order saddle point as
the TS structures, the first eigenvalue was negative and was associated to
the imaginary frequency. In the models studied, the related eigenvector
components were the geometric variables involved in hydrogen transfer
in the pseudo-eight-membered ring.
All absolute energies of the precursors, unsaturated bicycles, saturated bi-
cycles, and all the absolute energies of the TS structures in the simple
model (Scheme 6) computed for the unsaturated bicycles and saturated
bicycles in the selected conformations can be found in the Supporting In-
formation.
[4] a) M. Bassetti, L. Nicolini, S. Esposito, E. Righi, C. Viscoli, Curr.
Med. Chem. 2009, 16, 564–575; b) L. I. Llarrull, S. A. Testero, J. F.
Fisher, S. Mobashery, Curr. Opin. Microbiol. 2010, 13, 551–557;
c) M. Bassetti, F. Ginocchio, M. Mikulska, Crit. Care 2011, 15, 215.
[5] a) A. Urbach, G. Dive, J. Marchand-Brynaert, Eur. J. Org. Chem.
2009, 1757–1770; b) A. Urbach, G. Dive, B. Tinant, V. Duval, J.
Marchand-Brynaert, Eur. J. Med. Chem. 2009, 44, 2071–2080.
[6] a) A. Sliwa, G. Dive, J.-L. Habib Jiwan, J. Marchand-Brynaert, Tet-
rahedron 2010, 66, 9519–9527, and references therein; b) A. Sliwa,
J. Marchand-Brynaert, M. Luhmer, Magn. Reson. Chem. 2011, DOI:
10.1002/mrc.2808.
[7] a) E. Sauvage, R. Herman, S. Petrella, C. Duez, F. Bouillenne, J.-M.
Frꢃre, P. Charlier, J. Biol. Chem. 2005, 280, 31249–31256; b) E.
Sauvage, A. J. Powell, J. Heilemann, H. R. Josephine, P. Charlier, C.
Davies, R. F. Pratt, J. Mol. Biol. 2008, 381, 383–393.
[8] D. Lim, N. C. Strynadka, Nat. Struct. Biol. 2002, 9, 870–876.
[9] E. Sauvage, F. Kerff, E. Fonzꢀ, R. Herman, B. Schoot, J.-P. Mar-
quette, Y. Taburet, D. Prevost, J. Dumas, G. Leonard, P. Stefanic, J.
Coyette, P. Charlier, Cell. Mol. Life Sci. 2002, 59, 1223–1232.
[10] J. R. Hwu, M. L. Jain, S.-C. Tsay, G. H. Hakimelahi, Tetrahedron
Lett. 1996, 37, 2035–2038.
Assay with Resistant PBPs
Purified PBP5 from E. faecium D63r and PBP2a from MRSA ATCC
43300 were used as target proteins to test the inhibitory activity of syn-
thesized b-lactams. Each of the purified PBPs (2.5 mm) were first incubat-
ed with 1 mm potential inhibitor in 100 mm phosphate buffer, 0.01%
Triton X-100,[18] pH 7, for 4 h at 308C. Then, 25 mm fluorescein-labeled
ampicillin[19] was added to detect the residual penicillin binding activity
(RA). The samples were further incubated for 30 min at 378C in a total
volume of 20 mL. Denaturation buffer was added (0.1m Tris/HCl, pH 6.8,
containing 25% glycerol, 2% SDS, 20% b-mercaptoethanol, and 0.02%
bromophenol blue) and the samples were heated to 1008C for 1 min. The
samples were then loaded onto a 10% SDS acrylamide gel (10ꢅ7 cm)
and electrophoresis was performed for 45 min at 180 V (12 mA). Detec-
tion and quantification of the RAs were done with Molecular Image FX
equipment and Quantity One software (BioRad, Hercules, CA, USA).
Three independent experiments were carried out for each inhibitor.
Assay with R39
All assays with R39 were performed in 96-well microtiter plates (Brand,
Wertheim, Germany). 20 mm of the tested compounds were dissolved in
DMF. Finally, 7.5 mL of the solution were used in the assay. The final con-
centration of the compounds in the assays was 100 mm. The final concen-
tration of DMF in the assays was 0.25%. R39 (3.5 nm) was incubated in
the presence of the potential inhibitors in 10 mm sodium phosphate
buffer (pH 7.2) with 100 mm NaCl, 100 mm d-alanine, 0.01 mgmLÀ1 BSA,
and 0.01% Triton for 60 min at 258C. This preincubation was realized so
that slow-binding inhibitors were also detected. After preincubation, the
RA value of R39 was determined by observing the hydrolysis of the thio-
ester S2d substrate, in the presence of DTNB, catalyzed by the uninhibit-
ed enzyme. The initial rate of hydrolysis of 1 mm S2d in the presence of
1 mm DTNB was determined by monitoring the increase in absorbance
at 412 nm (DTNB: e[De]=13600mÀ1 sÀ1) by using a microplate absorb-
ance reader (Power Wave X, Biotek Instruments, Winooski, USA). The
rate of spontaneous hydrolysis of S2d in the presence of the inhibitors
was also determined in absence of R39. All assays were performed three
times. The determination of RA of R39 in absence of inhibitors was per-
formed six times on each plate. To detect false positives, which could be
slow-binding, noncompetitive promiscuous inhibitors, the assays were
performed in the presence of 0.01% Triton-X-100.[20]
[11] G. A. Olah, S. C. Narang, B. G. B. Gupta, R. Malhotra, J. Org.
Chem. 1979, 44, 1247–1251.
[12] M. J. Miller, P. G. Mattingly, M. A. Morrison, J. F. Kerwin, Jr., J.
Am. Chem. Soc. 1980, 102, 7026–7032.
[13] a) H. Tatewaki, S. Huzinaga, J. Comput. Chem. 1980, 1, 205–228;
b) G. Dive, D. Dehareng, J. M. Ghuysen, Theor. Chim. Acta 1993,
85, 409–421.
[14] G. Dive, D. Dehareng, Int. J. Quantum Chem. 1999, 73, 161–174.
[15] N-Benzoyl-d-alanylthioglycolate (S2d); a) R. Schwyzer, C. Hurli-
mann, C. Coenzyle, Helv. Chim. Acta 1954, 37, 155–166; b) M.
Adam, C. Damblon, M. Jamin, W. Zorzi, V. Dusart, M. Galleni, A.
el Kharroubi, G. Piras, B. G. Spratt, W. Keck, J. Coyette, J.-M. Ghuy-
sen, M. Nguyen-Disteche, J.-M. Frꢃre, Biochem. J. 1991, 279, 601–
604.
[16] H. S. Sader, A. C. Gales, Drugs 2001, 61, 553–564.
[17] Gaussian 03, RevisionD.02, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgom-
ery, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S.
Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani,
N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
Chem. Asian J. 2012, 7, 425 – 434
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
433

G. Dive, J. Marchand-Brynaert, and A. Sliwa
FULL PAPERS
J. B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y.
Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg,
V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O.
Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Fores-
man, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski,
B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L.
Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Na-
nayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen,
M. W. Wong, C. Gonzalez, and J. A. Pople, Gaussian, Inc., Walling-
ford CT, 2004.
[18] B. Y. Feng, B. K. Shoichet, Nat. Protoc. 2006, 1, 550–553.
[19] B. Lakaye, C. Damblon, M. Jamin, M. Galleni, S. Lepage, B. Joris, J.
Marchand-Brynaert, C. Frydrych, J.-M. Frꢃre, Biochem. J. 1994, 300,
141–145.
[20] A. Zervosen, W.-P. Lu, Z. Chen, R. E. White, T. P. Demuth, Jr., J.-
M. Frꢃre, Antimicrob. Agents Chemother. 2004, 48, 961–969.
Received: August 29, 2011
Published online: December 8, 2011
434
www.chemasianj.org
ꢄ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2012, 7, 425 – 434