Synthesis and evaluation of N1/C4-substituted β-lactams as PPE and HLE inhibitors

Article abstract of DOI:10.1016/j.bmc.2003.10.009

4-(Alkylamino)carbonyl-1-(alkoxy)carbonyl-2-azetidinones (9-11) have been prepared in five steps from 4-(benzyloxy)carbonyl-1-(t-butyldimethyl)silyl-2- azetidinone (1). The β-lactam reactivity of 9 has been established by ¹H NMR experiment. Compound 11 was a good reversible inhibitor of PPE and HLE. Based on theoretical design, series of 2-azetidinones (12-17) and 4-(alkoxy)carbonyl-2-azetidinones (18-21) bearing various carbonyl (ester, thiolester, amide) and thiocarbonyl (thioamide) functionalities at position N1 were similarly prepared. In the absence of C4-substituent, the compounds were inactive against elastases. On the other hand, 4-(benzyloxy)carbonyl-1- (ethylthioxy)carbonyl-2-azetidinone (19) and 4-(benzyloxy)carbonyl-1- (benzylamino)thiocarbonyl-2-azetidinone (21) were both good reversible inhibitors, but acting most probably via different mechanisms (enzymic processing of the exocyclic ester function or β-lactam ring opening).

Full text of DOI:10.1016/j.bmc.2003.10.009

Bioorganic & Medicinal Chemistry 12 (2004) 129–138
Synthesis and evaluation of N1/C4-substituted ꢀ-lactams as PPE
and HLE inhibitors
Stephane Gerard,a Moreno Galleni,b Georges Diveb
and Jacqueline Marchand-Brynaerta,*
a
´ ´ ´
Unite de Chimie Organique et Medicinale, Universite catholique de Louvain, Batiment Lavoisier, place Louis Pasteur 1,
B-1348 Louvain-la-Neuve, Belgium
ˆ
b
´
´
´
`
ˆ
´
`
Centre d’Ingenierie des Proteines, Universite de Liege, Batiment B6, Allee de la chimie 17, B-4000 Sart-Tilman, Liege, Belgium
Received 14May 2003; accepted 10 October 2003
Abstract—4-(Alkylamino)carbonyl-1-(alkoxy)carbonyl-2-azetidinones (9–11) have been prepared in five steps from 4-(benzyloxy)-
carbonyl-1-(t-butyldimethyl)silyl-2-azetidinone (1). The b-lactam reactivity of 9 has been established by 1H NMR experiment.
Compound 11 was a good reversible inhibitor of PPE and HLE. Based on theoretical design, series of 2-azetidinones (12–17) and 4-
(alkoxy)carbonyl-2-azetidinones (18–21) bearing various carbonyl (ester, thiolester, amide) and thiocarbonyl (thioamide) function-
alities at position N1 were similarly prepared. In the absence of C4-substituent, the compounds were inactive against elastases. On
the other hand, 4-(benzyloxy)carbonyl-1-(ethylthioxy)carbonyl-2-azetidinone (19) and 4-(benzyloxy)carbonyl-1-(benzylamino)-
thiocarbonyl-2-azetidinone (21) were both good reversible inhibitors, but acting most probably via different mechanisms (enzymic
processing of the exocyclic ester function or b-lactam ring opening).
# 2003 Elsevier Ltd. All rights reserved.
1. Introduction
the active serine residue (Ser-195) and the reactive car-
bonyl function of the substrate/inhibitor; this inter-
mediate is stabilized in the so-called oxy-anion hole.
Porcine pancreatic elastase (PPE)5 is usually considered
as a good model of HLE (the respective aminoacid
sequences share 40% residue identities); this enzyme is
more readily available and crystallographic studies have
demonstrated the structural similarity between the
active sites of both enzymes (conserved catalytic triad,
namely His-57, Asp-102 and Ser-195).
One of the current research lines in medicinal chemistry
for the discovery of novel anti-inflammatory drugs is
based on the design of human leukocyte elastase (HLE)
inhibitors.1 This enzyme, released by neutrophils, exerts
numerous and important biological functions, including
the degradation of connective tissue proteins. The HLE
proteolitic activity is regulated by endogenous inhibi-
tors; but, under pathological conditions, increased pro-
duction or inadequate neutralisation of HLE can lead to
uncontrolled connective tissue degradation, generating
inflammatory diseases such as pulmonary emphysema,
cystic fibrosis and chronic bronchitis.2 Thus, synthetic
HLE inhibitors able to restore the protease/antiprotease
balance are of potential therapeutic interest.3
Nowadays, numerous peptidic and non peptidic HLE
inhibitors of low molecular weight have been reported,
including for example, trifluoromethylketones (TFMK),6
thiazolidinones7 and coumarinic derivatives.8 Lactams
have also been considered as lead-structures for serine
proteases inhibition, such as d-lactams,9 g-lactams,10
and b-lactams.11 Our interest in this last family led us to
propose N-(alkoxy)carbonyl derivatives as potential
suicide-substrates (mechanism-based inhibitors) designed
to operate according to Figure 1 (Y=X=O), but this
mechanism has not actually been demonstrated.
HLE belongs to the superfamily of serine proteases like
chymotrypsin.4 The hydrolysis catalytic mechanism
involves the formation of an acyl-enzyme connecting
Keywords: b-lactam; Serine-protease inhibition; Elastase; N-acylation;
N-thioacylation.
We have previously studied two series of compounds,
namely, 3-halogeno-12 and 4-(alkoxy)carbonyl-1-
* Corresponding author. Tel.: +32-10-472740; fax: +32-10-474168;
e-mail: marchand@chim.ucl.ac.be
0968-0896/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2003.10.009

´
S. Gerard et al. / Bioorg. Med. Chem. 12 (2004) 129–138
130
13
(alkoxy)carbonyl-2-azetidinones
shown in Figure 2.
2. Results and discussion
These b-lactams behaved as reversible inhibitors of PPE
[activities (Ki) in the micromolar range], but their mode
of action was different, depending on the C3/C4sub-
stitution. Indeed, after a transient inhibition of PPE,
enzymic hydrolysis of 3-halogeno-2-azetidinones A
(active configuration: 3R) led to the b-lactam bond
cleavage without expulsion of the OR leaving group,
while PPE processing of 4-(alkoxy)carbonyl-2-azetidi-
nones B (active configuration: 4S) led to the OR0 ester
cleavage, but not to the b-lactam ring opening. Thus, in
molecules B, the ester carbonyl appeared to be more
sensitive towards serine nucleophilic attack than the b-
lactam carbonyl; the same reactivity was also observed
under smooth basic chemical hydrolysis (phosphate
buffer, pH 7.5). On the other hand, in molecules A, the
N-carbamate function did not behave as an isocyanate
precursor under enzymic processing of the b-lactam (see
Fig. 1; Y=X=O). Accordingly, we have designed
structural modifications of b-lactams B in two direc-
tions: (i) the decrease of the reactivity of the C4car-
bonyl substituent; (ii) the increase of the b-lactam
carbonyl reactivity and the subsequent ability of the N1
substituent to expulse a leaving group, after C2–N1
bond cleavage.
2.1. Modification of the C4 substituent
(S)-4-(Benzyloxy)carbonyl-1-(phenethyloxy)carbonyl-2-
azetidinone13 (molecule B with R=CH2CH2Ph and
R0=CH2Ph) is a good reversible inhibitor of PPE
(Ki=10 mM), fitting into the enzymic cavity in such a
manner that Ser-195 selectively attacks the ester car-
bonyl and not the b-lactam carbonyl. However, docking
experiments14 showed that other positioning should be
possible, similar to that occurring in the case of (R)-3-
bromo-1-(benzyloxy)carbonyl-2-azetidinone12a
(mol-
ecule A with R=CH2Ph and X=Br). In order to avoid
competitive enzymic hydrolysis of the C4substituent
with respect to the N1–C2 bond cleavage, we decided to
replace the ester group of compounds B (Fig. 2) with
amide functions, leading to the target-molecules 9–11
(Scheme 1).
We used racemic 1-(t-butyldimethylsilyl)-4-(benzyloxy)-
carbonyl-2-azetidinone (1)15 as the starting material.
Catalytic hydrogenation furnished quantitatively the
corresponding acid 2.15 Activation of 2 with carbo-
nyldiimidazole (CDI)16 followed by addition of benzyl-
amine, N-(methyl)benzylamine and dibenzylamine gave
the amides 3, 4 and 5, respectively, in modest to good
yields after chromatographic purifications. Deprotec-
tion of the silyl group was performed, in moderate yield,
with cesium fluoride in a polar non protic solvent in
order to avoid b-lactam ring opening. The resulting
NH-derivatives 6–8 were directly engaged in the two
step reaction of N-functionalisation, that is deprotona-
tion with lithium hexamethyldisilazanate (LiHMDS) at
low temperature and acylation with phenethyl chloro-
formate, followed a previously established protocol.13
Good yields of 1-(phenethyloxy)carbonyl-4-(alkylamino)-
carbonyl-2-azetidinones (9–11) were recovered after
chromatography on silica gel (Table 1). The compounds
were characterized by the usual spectroscopies (see
In this paper, we describe the synthesis and evaluation
against elastases (PPE and HLE) of 4-(alkylamino)-
carbonyl-1-(alkoxy)carbonyl-2-azetidinones on the one
hand, and series of 2-azetidinones and 4-(alkoxy)-
carbonyl-2-azetidinones bearing various carbonyl and
thiocarbonyl functionalities at position N1 (Fig. 1;
Y=O, S and X=O, S, NH), on the other hand. The
replacement of the oxygen atoms (Y=X=O) in the N1
side chain with other heteroatoms (S, NH) should
impact on several factors (binding interactions with the
enzyme, hydrogen bond formation, lipophilicity, polar-
izability, leaving group ability of XR2,. . .) in a way
which is hardly predictable.
Table 1. Yields of compounds 3–11 and activity of compounds 9–11
against PPE
R1
R2
Compd (yield)
Ki (mM)a (Compd)
H
Me
CH2Ph 3 (68%); 6 (50%); 9 (71%)
CH2Ph 4 (92%); 7 (54%); 10 (81%)
115
290
36
(9)
(10)
(11)
CH2Ph CH2Ph 5 (27%); 8 (55%); 11 (74%)
a Racemic mixture; experiments were made three times.
Figure 1. Design of b-lactamic inhibitors — possible suicide mechanism.
Figure 2. Reversible inhibition of PPE by b-lactams — products of the
enzymic reaction.
Scheme 1. Synthesis of 4-(alkylamino)carbonyl-1-phenethyloxy-
carbonyl-2-azetidinones.

´
S. Gerard et al. / Bioorg. Med. Chem. 12 (2004) 129–138
131
Experimental). In particular, the 1H NMR spectra
showed a typical ABX pattern around 4.50, 3.20 and
3.00 ppm corresponding to protons H4, H3 and H30.
For compound 10, two rotamers (ratio 60/40) were
visible in NMR. The 13C NMR spectra displayed three
carbonyl functions around 168, 162 and 149 ppm
attributed, respectively, to the amide, carbamate and b-
lactam functions. Lastly, the three carbonyl motifs gave
Table 1. Comparatively to the reference B (C4ester
substituent), the amide derivatives 9–11 were less active.
Surprisingly, the most sterically constrained amide 11
1
revealed to be the most potent inhibitor. From our H
NMR experiments, we could reasonably suppose that
the mode of action of compounds 9–11 is different from
that of the reference B. The inhibition should result
from the slow deacylation of the acyl-enzyme inter-
mediate formed with the C2 carbonyl. The experimental
order of activity of tested azetidinones (11>9>10) was
hardly predictable and most probably reflects the best fit
of the C4side chains into the enzymic pocket, rather
than a different electronic activating effect.
IR signals around 1670, 1730 and 1820 cmꢀ1
.
The chemical reactivity of b-lactam 9 has been eval-
uated in a 1H NMR experiment. The evolution of a
10ꢀ3 M solution of 9 in deuteriated phosphate buffer
(pD 7.5; 50 mM) containing 5% of DMSO-d6 was fol-
lowed at 500 MHz in function of time. Within 15 h,
compound 9 was slowly hydrolyzed to give the corre-
sponding b-aminoacid derivative resulting from nucleo-
philic attack onto the b-lactam carbonyl. The small ring
opening was ascertained by a significant deshielding of
the former’s H3 and H30 protons (Fig. 3). Cleavage of
the carbamate function could not be observed (absence
of signals at 2.68 and 3.65 ppm typical of phenethyl
alcohol). Thus, the replacement of benzyloxy residue
with benzylamino residue at C4has well restored a
higher chemical reactivity to the b-lactam ring. Similar
NMR experiments could not be performed with amides
10 and 11 because these derivatives were poorly soluble
in aqueous media at the concentrations required for
NMR spectroscopy. More disappointingly, the PPE
catalyzed hydrolysis of 9 was really too slow to be fol-
lowed by NMR, without significant competition with
chemical hydrolysis.
2.2. Modification of the N1 side chain
The electron withdrawing character of the N1 side chain
should have an influence on the b-lactam reactivity, that
is the sensitiveness of the C2 carbonyl towards nucleo-
philic attack (see Figure 1). This has been theoretically
evaluated at the ab initio level with a minimal basis set,
using a previously established model of the elastase
active site.13,17 The catalytic environment was mimicked
by an imidazole (model of His-57), a water molecule
(transient vehicle of the proton) and N-(b-hydroxy-
ethyl)formamide (model of Ser-195 side chain including
the amide bond of the backbone which is involved in the
oxyanion hole stabilization). The ꢀG values (calculated
energies at the transition states) have been computed by
reference to the isolated partners. The high values thus
obtained give an indication of the complexation effort
produced by the enzyme to generate the Michaelis
complex as the starting point of the reaction path
towards the transition state. On the other hand, the
optimization of the complexes connected by the transi-
tion state could generate non representative structures
as the active site geometric constraints are not taken
into account in the ab initio calculations.
b-Lactams 9–11 were evaluated for their potential inhi-
bitory effect on PPE in competitive experiments with a
good chromophoric substrate. The rates of PPE-cata-
lyzed hydrolysis of N-succinyl-Ala-Ala-Ala-p-nitroani-
lide were measured in the presence of various
concentrations of tested compounds. The variation of
absorbance at 410 nm (p-nitroaniline) was recorded in
function of time. Transient inhibitions were observed.
Plots of V/Vi, corresponding to the ratios of initial rates
of substrate hydrolysis in the absence and in the pre-
sence of inhibitors, versus the concentration of inhibitor
gave a straight line where the slope corresponds to 1/Ki.
The measured inhibition constants (Ki) are collected in
We have considered carbonyl (Y=O) and thiocarbonyl
(Y=S) functions bearing different potential leaving
groups (X-benzyl, where X=O, S, NH). Results of
Table 2 showed a significant difference between the N-
carbonyl (entries 1–3) and the N-thiocarbonyl deriva-
tives (entries 4–6), the later being more reactive
(ꢀꢀG=4.4–6.6 kcal/mol). In each series, the ꢀG values
Table 2. Theoretical evaluation of reactivity
Entry
Y
X
ꢀG (kcal/mol)a,b
1
2
3
O
O
O
O
S
NH
40.1
40.2
41.8
4S
5
6
O
.9
34
S
S
S
NH
33.6
37.4
a Calculated energy at the transition state.
b Azetidinone C–N bond cleavage.
Figure 3. 1H NMR data (d) of chemical hydrolysis.

´
S. Gerard et al. / Bioorg. Med. Chem. 12 (2004) 129–138
132
calculated to reach the transition state of the N1–C2
bond cleavage appeared weakly dependent on the nat-
ure of the potential leaving group (X=O, S); however,
(thio)amide substituents (X=NH) were the less favour-
able. The important activating effect of sulfur-contain-
ing side chains (particularly, entry 5) could be attributed
mainly to a polarizability effect.
d- and e-lactams.19 Reactions of 2-azetidinone with
thiophosgene or thiocarbonyldiimidazole, followed by
treatment with alcohols or thiols (Y=S; X=O, S) also
gave untractable mixtures. Finally, in our hands, the
only readily synthesized compounds were the thio-
carbamate derivatives 16 and 17 (Scheme 2, Table 3)
resulting from the reaction of 2-azetidinone with
LiHMDS and phenyl- or benzylisothiocyanate. Similar
conditions were used by Anderson during the synthesis
of PSA inhibitors.20 The thiocarbonyl function of 16–17
gave a typical line at 176–178 ppm in the 13C NMR
spectra.
We first prepared C3/C4-unsubstituted b-lactams repre-
sentative of the N1 substituted families (Y=O) with
ester (X=O), thiolester (X=S) and amide (X=NH)
groups, respectively. Commercially available 2-azetidi-
none (Scheme 2; R1=H) was first deprotonated with
LiHMDS at low temperature, then acylated with ethyl
chloroformate and ethyl thiochloroformate to furnish b-
lactams 12 and 13 (Table 3). According to a procedure
developed by Aoyama et al. for the synthesis of human
chymase inhibitors,18 we prepared b-lactams 14 and 15
by reacting 2-azetidinone with phenyl- and benzyliso-
cyanate, respectively. Compounds 12–15 were char-
acterized in 1H NMR by two triplets around 3.0 and 3.6
d for H4and H3 protons. In 13C NMR, the b-lactam
carbonyl gave a signal around 164–167 ppm; carbonyls
of carbamate, thiocarbamate and urea functions were
visible at 149, 165 and 148 ppm, respectively.
Starting from racemic 4-(benzyloxy)carbonyl-2-azetidi-
none (Scheme 2, R1=CO2CH2Ph),21 we prepared the
N1-substituted b-lactams 18–21 (Table 3) by using
the protocols validated above. A typical ABX pattern
1
was visible in the H NMR spectra around 3.0 d, 3.3 d
and 4.5 d corresponding to protons H30, H3 and H4.
The chemical reactivity of b-lactam 19 has been assayed
1
by H NMR as described for b-lactam 9 (Fig. 3): the
benzyl ester was hydrolyzed without opening of the b-
lactam ring; the ABX pattern of protons H3, H30 and
H4was still visible. The formed benzyl alcohol gave a
typical singlet at 4.44 d. Thus, the N1 thiolcarbamate
function does not improve the b-lactam carbonyl reac-
tivity versus the exocyclic ester function. The same
result was previously obtained for b-lactam 18.13 Com-
pounds 20 and 21 could not be similarly analyzed
because their water solubility was to low. Enzymic pro-
cessing of b-lactam 19 with PPE, followed by 1H NMR,
showed also the exocyclic ester hydrolysis.
Preparation of the corresponding thiocarbonyl families
(Scheme 2; R1=H and Y=S) was a very difficult task:
the required reagents (chlorothioformates, chlor-
odithioformates; for X=O, S) are not commercially
available, nor easily prepared. Moreover, all our
attempts to quench the anion derived from 2-azetidi-
none with carbondisulfide and an alkylating agent failed
(Y=X=S), though this sequence of reaction has been
successfully applied in the case of aliphatic amides, g-,
b-Lactams 12–21 were evaluated against PPE as descri-
bed for compounds 9–11. Results, collected in Table 3
(Ki values), showed that the C4unsubstituted deriva-
tives are inactive, except the 1-(ethylthio)carbonyl b-
lactam 13 which is very poorly active. On the other
hand, all the compounds bearing a benzyl ester group at
position C4were active against PPE, the more active b-
lactams being the sulfur-containing derivatives 19 and
21. The inhibition was reversible (native enzyme activity
was recovered after 1–2 h); the possible suicide
mechanism outlined in Scheme 1 did not occur, most
probably because the enzyme hydrolyzes the C4ester
group faster than the b-lactam function. This has been
experimentally proved (1H NMR) in the case of b-lac-
tams 1813 and 19, but not for compounds 20 and 21.
However, we could suppose that the same mechanism
operates for compound 20 (Y=O) and that the differ-
ences of Ki values only reflect the goodness of fit of the
N1 side chains into the enzymic pocket. On the other
hand, the mode of action of azetidinone 21 (Y=S)
could not be determined presently.
Table 3. Yields of compounds 12–21 and activity against PPE
R1
Y
X
R2
Compd (yield)
Ki (mM)a
H
H
H
H
H
H
CO2Bn
CO2Bn
CO2Bn
CO2Bn
O
O
O
O
S
O
S
C2H5
C2H5
Ph
CH2Ph
Ph
CH2Ph
C2H5
C2H5
CH2Ph
CH2Ph
12 (76%)
13 (21%)
14 (74%)
15 (96%)
16 (52%)
17 (50%)
18 (78%)
19 (20%)
20 (74%)
21 (20%)
Inactive
>500
Inactive
Inactive
Inactive
Inactive
160
NH
NH
NH
NH
O
S
NH
NH
S
O
O
O
S
7
>500
37
a See Table 1.
2.3. Inhibition of HLE
In view to confirm the inhibitory activities recorded
with PPE, we evaluated representative compounds of
both families (C4modified substitution and N1 mod-
ified substitution) against HLE. The results collected in
Table 4 are expressed as percentages of enzyme residual
Scheme 2. Synthesis of N1-substituted b-lactams with ester, thiolester,
amide and thioamide groups.

´
S. Gerard et al. / Bioorg. Med. Chem. 12 (2004) 129–138
133
activity. Thus, HLE was incubated with a good sub-
strate (namely N-methoxysuccinyl-Ala-Ala-Pro-Val-p-
nitroanilide) and the tested inhibitor. The residual
activity was obtained by comparison of the variation of
the absorbance (410 nm, p-nitroaniline) in function of
time, between this sample and the reference (sample
without inhibitor). For water-solubility reasons, b-lac-
tams 9–11 (first family) were assayed at a concentration
of 6 mM, while the other b-lactams 12–19 (second
family) were tested at 60 mM.
influence the mode of action, nor promote the suicide
mechanism proposed in Figure 1.
According to our theoretical evaluation, N1 sub-
stituents belonging to the thiocarbonyl family (Y=S;
Table 2) should enhance the b-lactam reactivity versus
the C4ester group ( ꢀG of ester cleavage=40.0–42.8
kcal/mol).13 Docking experiments14 showed that the
good positioning of potential inhibitors required for the
N1–C2 bond cleavage, instead of the C4-ester hydro-
lysis, is also well accessible. Actually, one representative
of this family, the thioamide 21 (Y=S, X=NH; Scheme
2, Table 3), has been prepared and found to be active
against PPE (Ki=37 mM). We are now studying novel
synthetic routes allowing the preparation of N1-thio-
noester (Y=S; X=O) and N1-dithioester derivatives
(Y=X=S), since we could not carry out the direct N-
functionalisation of a preformed azetidinone with CS2-
type reagents or synthetic equivalents.
Similar experiments were performed with PPE (sub-
strate=N-succinyl-Ala-Ala-Ala-p-nitroanilide) (Table
4). The results recorded with HLE and PPE were very
similar, confirming that PPE is a valuable model of
elastase. Except in the case of the sulfur-containing
derivatives 13 and 19 (but not 17), all b-lactams were
slightly more active against HLE than PPE. This could
be explained by structural differences in the close envir-
onment of the respective active sites; for instance, resi-
due 192 corresponds to a phenyl-alanine in HLE and a
glutamine in PPE.4,5 This residue appeared to be in
interaction with the N1 side chains when the inhibitors
are placed in the enzymic cavity in such a way that ser-
3. Experimental
3.1. General
14
ine 195 attacks the C4ester substituent.
Reagents and solvents were purchased from Acros chi-
mica, Aldrich or Fluka. Porcine pancreatic elastase
(type 1), human leukocyte elastase and N-succinyl-l-
alanyl-l-alanyl-l-alanyl-p-nitroanilide were obtained
from Sigma Chemical Co. N-Methoxysuccinyl-l-alanyl-
l-alanyl-l-prolin-l-valine-p-nitroanilide was obtained
from Calbiochem. Tetrahydrofuran was dried over
sodium/benzophenone, then distilled. Column chroma-
tographies were carried out with silica gel 60 (70–230
mesh ASTM) supplied by Merck. The IR spectra were
recorded with a Perkin–Elmer 1710 instrument, only the
most significant absorption bands being reported. The
mass spectra were obtained with a Finnigan MAT TSQ-
70 instrument. The microanalyses were performed at the
Christopher Ingold Laboratories of the University Col-
lege, London (Dr A. Stones). The melting points were
determined with an Electrothermal microscope and are
2.4. Conclusion
Our results showed the potential interest of N1-(thio)-
carbonyl-2-azetidinones as reactive structures for the
design of novel elastase inhibitors. However, an elec-
tron-withdrawing substituent (activating group) placed
at position C4(or C3) was systematically required to
reach a good level of enzyme inhibition. The nature of
this group could dramatically change the mode of
action of the azetidinone derivatives: a C4-ester sub-
stituent is more reactive towards nucleophilic attack
than the b-lactam function itself, while a C4-amide
substituent is less reactive. Thus compound 11 (Scheme
1, Table 1) is a good reversible inhibitor of PPE and
HLE, acting most probably like 3-halogeno-1-(alkoxy)
carbonyl-2-azetidinones (structure A in Fig. 2). Com-
pound 19 (Scheme 2, Table 3) is more active against
PPE and HLE, but behaves in the same way as 1,4-
bis[(alkoxy)carbonyl]-2-azetidinones (structure B in Fig.
2). In this case, the nature of the potential leaving group
(S-alkyl instead of O-alkyl) of the N1 side chain did not
1
uncorrected. The H and 13C NMR spectra were recor-
ded on Varian Gemini 300 (at 300 MHz for proton and
75 MHz for carbon) or Brucker AM-500 spectrometers
(at 500 MHz for proton and 125 MHz for carbon); the
chemical shifts are reported in ppm (d) downfield from
tetramethylsilane (internal standard). The high resolu-
tion mass spectra were obtained at the University of
Mons, Belgium (Prof. R. Flammang).
Table 4. Biochemical evaluation
Compd
(concn)
Residual activity (%)
3.2. Synthesis
PPE
HLE
3.2.1. General procedure for 4-(alkylamino)carbonyl-1-
(tert-butyldimethylsilyl)-2-azetidinones. A solution of
215 (equiv) and CDI (1.2 equiv) in dry CH2Cl2 (10 mL/
mmol) was stirred at room temperature for 30 min.
Amine (1.2 equiv) was slowly added and stirring was
continued for 24h. The solution was washed with 1 N
HCl. The aqueous layer was extracted with EtOAc. The
combined extracts were dried over MgSO4, filtered and
concentrated under vacuum. The resulting b-lactam was
9 (6 mM)
87
92
85
100
81
100
98
83
78
63
89
88
73
81
37
10 (6 mM)
11 (6 mM)
12 (60 mM)
13 (60 mM)
15 (60 mM)
17 (60 mM)
18 (60 mM)
19 (60 mM)
51
1429

´
S. Gerard et al. / Bioorg. Med. Chem. 12 (2004) 129–138
134
purified by column chromatography on silica gel
(CH2Cl2/EtOAc, 90/10 to 80/20).
temperature. The crude mixture was concentred and
after addition of EtOAc, the organic layer was washed
twice with water, dried over MgSO4, filtered and con-
centrated under vacuum. The deprotected b-lactam was
directly used in the acylation reaction without purifica-
tion.
3.2.1.1. 4 - (Benzylamino)carbonyl - 1 - (tert - butyldi-
methylsilyl)-2-azetidinone (3). Reaction of 2 (115 mg; 0.5
mmol) with CDI (97 mg; 0.6 mmol) and benzylamine
(66 mL; 0.6 mmol) yielded 3 (108 mg, 68%) as a white
solid; mp 130–131 ꢁC; IR (KBr) 3429, 1744, 1674 cmꢀ1
;
3.2.2.1. 4 - (Benzylamino)carbonyl - 2 - azetidinone (6).
Reaction of 3 (190 mg; 0.6 mmol) and cesium fluoride
(137 mg; 0.9 mmol) yielded the N-deprotected azetidin-
2-one (64mg, 50%) as a colourless oil; IR (film) 3225,
1H NMR (500 MHz; CDCl3) d 0.09 and 0.22 (2ꢂs,
2ꢂ3H, 2ꢂSiCH3), 0.90 (s, 9H, C(CH3)3), 2.98 (dd,
J=15.5 Hz, 3.0 Hz, 1H, H-3), 3.32 (dd, J=15.5 Hz, 6.1
Hz, 1H, H-30), 3.96 (dd, J=6.1 Hz, 3.0 Hz, 1H, H-4),
4.44 (m, 2H, CH2Ph), 6.55 (br s, 1H, NH), 7.34-7.42 (m,
5H, Ph); 13C NMR (125 MHz; CDCl3) d ꢀ6.3 and ꢀ5.7
(SiCH3), 18.5 (C(CH3)3), 26.1 (C(CH3)3), 43.6 and 44.5
(CH2Ph and C-3), 50.3 (C-4), 127.9, 128.5 and 128.7
(CHAr), 137.5 (CAr), 171.1 and 171.7 (2ꢂCO); MS (EI)
m/z 318, 261, 185, 91. Anal. calcd for C17H26N2O2Si: C,
64.11; H, 8.23; N, 8.79. Found: C, 64.25; H, 8.25; N,
8.72%.
1
1734, 1670 cmꢀ1; H NMR (300 MHz; CDCl3) d 3.01
(dd, J=15.3 Hz, 2.8 Hz, 1H, H-3), 3.34(dd, J=15.3 Hz,
6.1 Hz, 1H, H-30), 4.15 (dd, J=6.1 Hz, 2.8 Hz, 1H, H-
4), 4.42 (m, 2H, CH2Ph), 6.30 and 6,65 (2ꢂbr s, 2ꢂ1H,
2ꢂNH), 7.25–7.40 (m, 5H, Ph); 13C NMR (75 MHz;
CDCl3) d 42.8 (C-3), 45.6 (CH2Ph), 47.9 (C-4), 126.2,
127.5 and 128.6 (CHAr), 136.3 (CAr), 166.1 and 169.9
(2ꢂCO).
3.2.2.2. 4-[(N-Methyl)benzylamino]carbonyl-2-azetidi-
none (7). Reaction of 4 (332 mg; 1 mmol) and cesium
fluoride (228 mg; 1.5 mmol) yielded N-deprotected aze-
tidin-2-one (117 mg, 54%) as a colourless oil; IR (film)
3.2.1.2. 4-[(N-Methyl)benzylamino]carbonyl-1-(tert-
butyldimethylsilyl)-2-azetidinone (4). Reaction of 2 (115
mg; 0.5 mmol) with CDI (97 mg; 0.6 mmol) and
methylbenzylamine (78 mL; 0.6 mmol) yielded 4 (152
mg, 92%) as a white solid; mp 80–81 ꢁC; IR (KBr) 1738,
1
3245, 1772, 1654 cmꢀ1; H NMR (300 MHz; CDCl3) d
2.90 (s, 1H, CH3), 3.06 (m, 2H, H-3 and H-30), 4.52 (m,
1
1655, 1263 cmꢀ1; H NMR (500 MHz; CDCl3) d 0.12
3H, H-4and CH Ph), 7.10 (br s, 1H, NH), 7.15–7.48
2
and 0.35 (2ꢂs, 2ꢂ3H, 2ꢂSiCH3), 0.98 (s, 9H, C(CH3)3),
2.86 (s, 3H, CH3), 2.89 (dd, J=14.7 Hz, 3.2 Hz, 1H, H-
3), 3.34(dd, J=14.7 Hz, 6.2 Hz, 1H, H-30), 4.32 (dd,
J=6.2 Hz, 3.2 Hz, 1H, H-4), 4.53 and 4.63 (2ꢂd,
(m, 5H, Ph); 13C NMR (75 MHz; CDCl3) d 33.6 (CH3),
42.9 (C-3), 47.1 (C-4), 51.1 (CH2Ph), 126.2, 127.5 and
128.6 (CHAr), 136.3 (CAr), 166.1 and 169.9 (2ꢂCO).
J=14.8 Hz, 2ꢂ1H, CH2Ph), 7.20–7.31 (m, 5H, Ph); 13
C
NMR (125 MHz; CDCl3) d ꢀ6.5 and ꢀ5.8 (SiCH3),
18.4( C(CH3)3), 26.3 (C(CH3)3), 33.6 (CH3), 43.3 (C-3),
47.8 (C-4), 51.1 (CH2Ph), 127.5, 127.9 and 128.5
(CHAr), 136.5 (CAr), 170.0 and 170.5 (2ꢂCO); MS (EI)
m/z 332, 275, 149, 91. Anal. calcd for C18H28O2N2Si: C,
65.02; H, 8.49; N, 8.42. Found: C, 65.33; H, 8.48; N,
8.22%.
3.2.2.3. 4-(Dibenzylamino)carbonyl-2-azetidinone (8).
Reaction of 5 (111 mg; 0.3 mmol) and cesium fluoride
(60 mg; 0.4mmol) yielded N-deprotected azetidin-2-one
(45 mg, 55%) as a colourless oil; IR (film) 1748, 1685
1
cmꢀ1; H NMR (300 MHz; CDCl3) d 3.01 (dd, J=15.2
Hz, 2.7 Hz, 1H, H-3), 3.19 (dd, J=15.2 Hz, 6.1 Hz, 1H,
H-30), 4.48 (m, 5H, 2ꢂCH2Ph and H-4), 6.25 (br s, 1H,
NH), 7.20–7.31 (m, 10H, Ph); 13C NMR (75 MHz;
CDCl3) d 43.5 (C-3), 47.1 (C-4), 48.8 and 49.3
(2ꢂCH2Ph), 126.4, 127.5, 127.8, 127.9, 128.1 and 128.3
(CHAr), 135.4and 136.5 (2 ꢂCAr), 165.8 and 170.4
(2ꢂCO).
3.2.1.3. 4 - (Dibenzylamino)carbonyl - 1 - (tert - butyldi-
methylsilyl)-2-azetidinone (5). Reaction of 2 (115 mg; 0.5
mmol) with CDI (97 mg; 0.6 mmol) and dibenzylamine
(115 mL; 0.6 mmol) yielded 5 (54mg, 27%) as a colour-
less oil; IR (film) 1749, 1653, 1264 cmꢀ1
;
1H NMR
3.2.3. General procedure for 1-(phenethyloxy)carbonyl-4-
(alkylamino)carbonyl-2-azetidinones.
(300 MHz; CDCl3) d 0.12 and 0.35 (2ꢂs, 2ꢂ3H,
2ꢂSiCH3), 0.98 (s, 9H, C(CH3)3), 3.14(dd, J=15.2 Hz,
3.0 Hz, 1H, H-3), 3.34(dd, J=15.2 Hz, 6.1 Hz, 1H, H-
30), 4.32 (dd, J=6.1 Hz, 3.0 Hz, 1H, H-4), 4.63 (m, 4H,
2ꢂCH2Ph), 7.20–7.31 (m, 5H, Ph); 13C NMR (75 MHz;
A
solution of
LiHMDS (1 equiv) in dry THF (5 mL/mmol) was added
dropwise to a solution of N-deprotected azetidinone (1
equiv) in dry THF (5 mL/mmol), cooled at ꢀ78 ꢁC
under argon atmosphere. The mixture was stirred for 30
min at ꢀ78 ꢁC, then phenethyl chloroformate (1 equiv)
was added with a syringe through a rubber stopper.
Stirring was continued for 1 h at this temperature and
the mixture was allowed to reach room temperature and
further stirred for 45 min at 20 ꢁC. After addition of
water and extraction with CH2Cl2, the organic layer was
washed with brine, dried over MgSO4, filtered and con-
centrated under vacuum. The resulting b-lactam was
purified by column chromatography on silica gel
(CH2Cl2/EtOAc, 90/10).
CDCl3;Me4Si)
d
ꢀ6.6 and ꢀ5.5 (SiCH3), 18.6
(C(CH3)3), 26.1 (C(CH3)3), 43.2 (C-3), 47.9 (C-4), 51.1
and 52.9 (2ꢂCH2Ph), 127.5, 127.8, 127.9 and 128.3
(CHAr), 136.5 and 138.2 (CAr), 170.1 and 170.3 (2ꢂCO);
MS (CI) m/z 409 (C24H32N2O2Si), 302, 91.
3.2.2. General procedure for 4-(alkylamino)carbonyl-2-
azetidinones (N-deprotection). A solution of N-silylated
azetidin-2-one (1 equiv) and cesium fluoride (1.5 equiv)
in dry DMF (5 mL/mmol) was stirred for 2 h at room

´
S. Gerard et al. / Bioorg. Med. Chem. 12 (2004) 129–138
135
3.2.3.1. 1 - (Phenethyloxy)carbonyl - 4 - (benzylamino)-
carbonyl-2-azetidinone (9). Yield from 4-(benzylamino)-
carbonyl-2-azetidinone (64mg; 0.3 mmol), LiHMDS
(72 mg; 0.3 mmol) and phenethyl chloroformate (55 mg;
0.3 mmol): 75 mg (71%) as ꢀa1 colourless oil; IR (film)
carbamate), 161.8 (CO azetidinone), 168.5 (CO amide);
MS (CI) m/z 443 (M+1), 105, 91; HRMS (CI) calcd for
C27H27N2O4: 443.1970. Found: 443.1987.
3.2.4. General procedure for 1-ethyloxy(thio)carbonyl-2-
azetidinones. A solution of LiHMDS (1 equiv) in dry
THF (5 mL/mmol) was dropwise added to a solution of
azetidinone (1 equiv) in dry THF (5 mL/mmol), cooled
at ꢀ78 ꢁC under argon atmosphere. The mixture was
stirred for 30 min at ꢀ78 ꢁC, then ethylchloroformate or
thiochloroformate (1.2 equiv) was added with a syringe
through a rubber stopper. Stirring was continued for 1 h
at this temperature and the mixture was allowed to
reach room temperature and stirred for 45 min at 20 ꢁC.
After addition of water and extraction with CH2Cl2, the
organic layer was washed with brine, dried over MgSO4,
filtered and concentrated under vacuum. The resulting
b-lactam was purified by column chromatography on
silica gel (CH2Cl2/EtOAc, 90/10 to 70/30).
3367, 1819, 1731, 1684cm
;
1H NMR (500 MHz;
CDCl3) d 2.96 (t, J=7.3 Hz, 2H, CH2CH2Ph), 3.19 (dd,
J=16.2 Hz, 6.7 Hz, 1H, H-3), 3.29 (dd, J=16.2 Hz, 3.7
Hz, 1H, H-30), 4.35 (dd, J=6.7 Hz, 3.7 Hz, 1H, H-4),
4.39 (t, J=7.3 Hz, 2H, CH2CH2Ph), 4.42 (dd, J=14.6
Hz, 5.9 Hz, 1H, CH2Ph), 4.47 (dd, J=14.6 Hz, 5.9 Hz,
1H, CH2Ph), 7.03 (br s, 1H, NH), 7.18–7.38 (m, 10H,
Ph); 13C NMR (125 MHz; CDCl3) d 34.8 (CH2CH2Ph),
41.1 (C-3), 43.6 (CH2Ph), 51.3 (C-4), 67.5 (CH2CH2Ph),
126.7, 127.5, 128.5, 128.6 and 128.9 (CHAr), 136.7 and
137.4(C Ar), 149.5 (CO carbamate), 162.8 (CO azetidi-
none), 167.4(CO amide); MS (CI) m/z 353 (M+1), 243,
105; HRMS (CI) calcd for C20H21N2O4: 353.1501.
Found: 353.1489.
3.2.3.2. 1 - (Phenethyloxy)carbonyl - 4 - [(N -
methyl)benzylamino]carbonyl-2-azetidinone (10). Yield
from 4-[(N-methyl)benzylamino]carbonyl-2-azetidinone
(153 mg; 0.6 mmol), LiHMDS (145 mg; 0.6 mmol) and
phenethyl chloroformate (110 mg; 0.6 mmol): 176 mg
(81%) as a colourless oil; IR (film) 1818, 1731, 1662
3.2.4.1. 1 - (Ethyloxy)carbonyl - 2 - azetidinone (12).
Reaction of 2-azetidinone (71 mg; 1 mmol), LiHMDS
(241 mg; 1 mmol) and ethyl chloroformate (96 mL; 1.2
mmol) yielded 12 (109 mg, 76%) as a yellow oil; IR
(film) 2927, 1802, 1717 cmꢀ1 1H NMR (300 MHz,
;
CDCl3) d 1.35 (t, J=7.1 Hz, 3H, CH3), 3.06 (t, J=5.2
Hz, 2H, H-4), 3.64 (t, J=5.2 Hz, 2H, H-3), 4.30 (q,
J=7.1 Hz, 2H, CH2); 13C NMR (75 MHz, CDCl3) d
14.3 (CH3), 36.5 and 37.7 (C-3 and C-4), 62.7 (OCH2),
149.3 (CO carbamate), 164.1 (CO azetidinone); MS (CI)
m/z 144 (M+1, C6H10NO3), 116, 98.
cmꢀ1 1H NMR (500 MHz; CDCl3) major rotamer
;
(values in parentheses correspond to the minor rotamer)
d 2.95 (s, 3H, CH3) (3.05), 3.03 (t, J=7.2 Hz, 2H,
CH2CH2Ph), 3.06 (dd, J=15.4Hz, 3.2 Hz, 1H, H-3),
0
3.24(dd, J=15.4Hz, 6.4Hz, 1H, H-3 ), 4.35 (dd, J=6.4
Hz, 3.2 Hz, 1H, H-4), 4.43 (t, J=7.2 Hz, 2H,
CH2CH2Ph), 4.60 (d, J=14.7 Hz, 1H, CH2Ph) (4.50),
4.65 (d, J=14.7, 1H, CH2Ph), 7.11–7.45 (m, 10H, Ph);
13C NMR (125 MHz; CDCl3) major rotamer (values in
parentheses correspond to the minor rotamer) d 34.0
(CH3) (34.8), 34.9 (CH2CH2Ph), 41.2 (C-3) (41.3), 47.5
(C-4) (47.4), 51.4 (CH2Ph), 67.2 (CH2CH2Ph), 125.9,
126.6, 127.6, 127.9, 128.4, 128.7, 128.9 and 129.0
(CHAr), 136.2 and 136.9 (CAr), 148.5 (CO carbamate),
161.8 (CO azetidinone) (161.9), 167.6 (CO amide)
(168.1); MS (CI) m/z 367 (M+1), 105, 91; HRMS (CI)
calcd for C21H23N2O4: 367.1657. Found: 367.1660.
3.2.4.2. 1-(Ethyloxy)thiocarbonyl-2-azetidinone (13).
Reaction of 2-azetidinone (71 mg; 1 mmol), LiHMDS
(241 mg; 1 mmol) and ethyl thiochloroformate (104 mL;
1.2 mmol) yielded 13 (33 mg, 21%) as a white solid; mp
ꢁ
1
71–72 C; IR (KBr) 2959, 1791, 1653 cmꢀ1; H NMR
(300 MHz, CDCl3) d 1.33 (t, J=7.4Hz, 3H, CH ), 3.01
3
(q, J=7.4Hz, 2H, CH ), 3.09 (t, J=5.3 Hz, 2H, H-4),
2
3.69 (t, J=5.3 Hz, 2H, H-3); 13C NMR (75 MHz,
CDCl3) d 14.7 (CH3), 23.6 (SCH2), 36.1 and 38.1 (C-3
and C-4), 163.8 (CO azetidinone), 165.5 (CO thio-
carbamate); MS (EI) m/z 159 (C6H9NO2S), 98, 89, 70.
3.2.3.3. 1-(Phenethyloxy)carbonyl-4-(dibenzylamino)-
carbonyl-2-azetidinone (11). Yield from 4-(dibenzylami-
no)carbonyl-2-azetidinone (45 mg; 0.15 mmol),
LiHMDS (37 mg; 0.15 mmol) and phenethyl chloro-
formate (29 mg; 0.15 mmol): 49 mg (74%) as a colour-
3.2.4.3. 1-(Ethyloxy)carbonyl-4-(benzyloxy)carbonyl-2-
azetidinone (18). This compound was obtained from 4-
(benzyloxy)carbonyl-2-azetidinone (205 mg; 1 mmol) by
treatment with LiHMDS (241 mg; 1 mmol) and ethyl
chloroformate (96 mL; 1.2 mmol) to yield 18 (215 mg,
less oil; IR (film) 1819, 1732, 1663 cmꢀ1 1H NMR
;
(500 MHz; CDCl3) d 2.93 (dd, J=16.4Hz, 5.7 Hz, 1H,
H-3), 2.99 (dd, J=16.4Hz, 3.7 Hz, 1H, H-3 ), 3.01 (t,
78%) as a yellow oil; IR (film) 2959, 1819, 1730 cmꢀ1
;
0
J=7.4Hz, 2H, CH CH2Ph), 4.40 (d, J=17.0 Hz, 1H,
1H NMR (300 MHz, CDCl3) d 1.28 (t, J=7.2 Hz, 3H,
CH3), 3.03 (dd, 1H, J=15.9 Hz, 3.3 Hz, H-30), 3.31 (dd,
1H, J=15.9 Hz, 6.6 Hz, H-3), 4.26 (q, J=7.2 Hz, 2H,
CH2), 4.48 (dd, 1H, J=6.6 Hz, 3.3 Hz, H-4), 5.27 (AB
pattern, 2H, JAB=13.6 Hz, CH2Ph), 7.36–7.41 (m, 5H,
Ph); 13C NMR (75 MHz, CDCl3) d 14.1 (CH3), 41.6 (C-
3), 49.4 (C-4), 63.1 (OCH2), 67.6 (CH2Ph), 128.3, 128.6
and 128.7 (CHAr), 134.9 (CAr), 148.3 (CO carbamate),
161.6 (CO azetidinone), 168.8 (CO ester); MS (CI) m/z
278 (M+1), 250, 144, 91. Anal. calcd for C14H15NO5:
2
CH2Ph), 4.43 (t, J=7.4Hz, 2H, C H2CH2Ph), 4.54 (d,
J=17.0 Hz, 1H, CH2Ph), 4.64 (dd, J=5.7 Hz, 3.7 Hz,
1H, H-4), 4.65 (d, J=14.6 Hz, 1H, CH2,Ph), 4.72 (d,
J=14.6 Hz, 1H, CH2,Ph), 7.11–7.45 (m, 15H, Ph); 13C
NMR (125 MHz; CDCl3) d 34.9 (CH2CH2Ph), 41.6 (C-
3), 47.5 (C-4), 49.6 (2ꢂCH2Ph), 67.2 (CH2CH2Ph),
126.6, 127.6, 127.9, 128.1, 128.2, 128.4, 128.7, 128.9 and
129.1 (CHAr), 135.6, 136.3 and 136.9 (CAr), 148.4 (CO

´
S. Gerard et al. / Bioorg. Med. Chem. 12 (2004) 129–138
136
C, 60.64; H, 5.45; N, 5.05. Found: C, 60.31; H, 5.51; N,
5.06%.
3.2.5.3.
1-(Benzylamino)carbonyl-4-(benzyloxy)car-
bonyl-2-azetidinone (20). This compound was obtained
from 4-(benzyloxy)carbonyl-2-azetidinone (205 mg; 1
mmol) by treatment with DMAP (16 mg; 0.2 mmol),
benzyl isocyanate (68 mL; 1.1 mmol) and TEA (56 mL; 0.4
mmol) to yield 20 (322 mg, 95%) as a colourless oil; IR
(film): 3383, 3052, 1782, 1715 cmꢀ1; 1H NMR (300MHz,
CDCl3) d 2.98 (dd, 1H, J=10.6 Hz, 2.1 Hz, H-30), 3.25
(dd, 1H, J=10.6 Hz, 4.2 Hz, H-3), 4.48 (m, 3H, H-4 and
NHCH2), 5.21 (AB pattern, 2H, JAB=8.8 Hz, CH2Ph),
6.72 (br s, 1H, NH), 7.22–7.41 (m, 10H, Ph); 13C NMR
(75 MHz, CDCl3) d 40.6 (C-3), 43.3 (NHCH2), 48.3 (C-
4), 67.2 (CH2Ph), 127.1, 127.2, 127.8, 128.1, 128.2 and
128.5 (CHAr), 134.6 and 137.5 (CAr), 149.1 (CO urea),
164.3 (CO azetidinone), 168.8 (CO ester); MS (FAB) m/z
339 (C19H19N2O4), 247, 135, 107, 91.
3.2.4.4. 1 - (Ethyloxy)thiocarbonyl - 4 - (benzyloxy)-
carbonyl-2-azetidinone (19). This compound was
obtained from 4-(benzyloxy)carbonyl-2-azetidinone
(205 mg; 1 mmol) by treatment with LiHMDS (241 mg;
1 mmol) and ethyl thiochloroformate (104 mL; 1.2
mmol) to yield 19 (61 mg, 20%) as a yellow oil; IR (film)
2960, 1796, 1751, 1667 cmꢀ1 1H NMR (300 MHz,
;
CDCl3) d 1.32 (t, J=7.4Hz, 3H, CH ), 2.98 (q, J=7.4
3
Hz, 2H, CH2), 3.01 (dd, 1H, J=16.1 Hz, 3.5 Hz, H-30),
3.31 (dd, 1H, J=16.1 Hz, 6.6 Hz, H-3), 4.54 (dd, 1H,
J=6.6 Hz, 3.5 Hz, H-4), 5.23 (AB pattern, 2H,
JAB=12.3 Hz, CH2Ph), 7.29–7.38 (m, 5H, Ph); 13C
NMR (75 MHz, CDCl3) d 14.1 (CH3), 23.8 (SCH2), 41.0
(C-3), 49.6 (C-4), 67.7 (CH2Ph), 128.2, 128.6 and 128.7
(CHAr), 134.9 (CAr), 161.4(CO azetidinone), 165.0 (CO
thiocarbamate), 168.4(CO ester); MS (CI) m/z 294
(M+1, C14H16NO4S), 232, 91.
3.2.6. General procedure for 1-(alkylamino)thiocarbonyl-
2-azetidinones. A solution of LiHMDS (1 equiv) in dry
THF (5 mL/mmol) was dropwise added to a solution of
azetidinone (1 equiv) in dry THF (5 mL/mmol), cooled
at ꢀ78 ꢁC under argon atmosphere. The mixture was
stirred for 30 min at ꢀ78 ꢁC, then isothiocyanate (1.1
equiv) was slowly added. Stirring was continued for 30
min at this temperature and the mixture was allowed to
reach room temperature and stirred for 4h at 20 ꢁC.
After addition of a solution of saturated NH4Cl and
extraction with CH2Cl2, the organic layer was dried
over MgSO4, filtered and concentrated under vacuum.
The resulting b-lactam was purified by column chroma-
tography on silica gel (CH2Cl2).
3.2.5. General procedure for 1-(alkylamino)carbonyl-2-
azetidinones. To a solution of azetidinone (1 equiv) and
DMAP (0.2 equiv) in dry CH2Cl2 (5 mL/mmol), under
argon atmosphere, were successively added dropwise
isocyanate (1.1 equiv) and TEA (0.4equiv). The mixture
was stirred and heated to reflux for 24h. The crude
mixture was then washed successively with a saturated
solution of NaHCO3, 1.2 M HCl and brine. The organic
layer was dried over MgSO4, filtered and concentrated
under vacuum. The resulting b-lactam was purified by
column chromatography on silica gel (CH2Cl2/EtOAc,
95/5).
3.2.6.1. 1-(Phenylamino)thiocarbonyl-2-azetidinone (16).
Reaction of 2-azetidinone (108 mg; 1.5 mmol),
LiHMDS (360 mg; 1.5 mmol) and phenyl iso-
thiocyanate (197 mL; 1.65 mmol) yielded 16 (160 mg,
52%) as a white solid; mp 85–86 ꢁC; IR (KBr) 3281,
1756, 1326 cmꢀ1; 1H NMR (300 MHz, CDCl3) d 3.09 (t,
J=5.0 Hz, 2H, H-4), 3.88 (t, J=5.0 Hz, 2H, H-3), 7.21–
7.49 (m, 5H, Ph), 10.2 (br s, 1H, NH); 13C NMR
(75 MHz, CDCl3) d 34.4 and 40.2 (C-3 and C-4), 123.3,
126.3 and 128.7 (CHAr), 137.1 (CAr), 165.2 (CO azetidi-
none), 175.7 (CO thiourea); MS (CI) m/z 207 (M+1),
136, 70. Anal. calcd for C10H10N2OS: C, 58.23; H, 4.88;
N, 13.59; S, 15.54. Found: C, 58.19; H, 4.77; N, 14.14;
S, 15.30%.
3.2.5.1. 1-(Phenylamino)carbonyl-2-azetidinone (14).
Reaction of 2-azetidinone (108 mg; 1.5 mmol), DMAP
(24mg; 0.3 mmol), phenyl isocyanate (179 mL; 1.65
mmol) and TEA (84 mL; 0.60 mmol) yielded 14 (210 mg,
74%) as a white solid; mp 112–113 ꢁC; IR (KBr) 3381,
1
2964, 1767, 1714 cmꢀ1; H NMR (300 MHz, CDCl3) d
3.16 (t, J=4.8 Hz, 2H, H-4), 3.76 (t, J=4.8 Hz, 2H, H-
3), 7.05–7.52 (m, 5H, Ph), 8.43 (br s, 1H, NH); 13C
NMR (75 MHz, CDCl3) d 35.9 and 37.2 (C-3 and C-4),
119.4, 124.0 and 128.9 (CHAr), 136.9 (CAr), 147.6 (CO
urea), 167.1 (CO azetidinone); MS (CI) m/z 191 (M+1),
149, 98, 94, 72. Anal. calcd for C10H10N2O2: C, 63.14;
H, 5.30; N, 14.72. Found: C, 62.73; H, 5.17; N, 14.42%.
3.2.6.2. 1-(Benzylamino)thiocarbonyl-2-azetidinone (17).
Reaction of 2-azetidinone (108 mg; 1.5 mmol),
LiHMDS (360 mg; 1.5 mmol) and benzyl isothiocyanate
(219 mL; 1.65 mmol) yielded 17 (156 mg, 50%) as a yel-
low solid; mp 76–77 ꢁC; IR (KBr) 3326, 1759, 1315
cmꢀ1; 1H NMR (300 MHz, CDCl3) d 3.01 (t, J=4.9 Hz,
2H, H-4), 3.81 (t, J=4.9 Hz, 2H, H-3), 4.85 (d, J=5.7
Hz, 2H, CH2Ph), 7.34–7.48 (m, 5H, Ph), 8.72 (br s, 1H,
NH); 13C NMR (75 MHz, CDCl3) d 34.4 and 40.2 (C-3
and C-4), 48.3 (CH2Ph), 127.5, 127.6 and 128.7 (CHAr),
137.8 (CAr), 165.0 (CO azetidinone), 178.1 (CO
thiourea); MS (CI) m/z 221 (M+1), 91, 70. Anal. calcd
for C11H12N2OS: C, 59.98; H, 5.49; N, 12.73; S, 14.55.
Found: C, 59.77; H, 5.39; N, 12.69; S, 14.93%.
3.2.5.2. 1-(Benzylamino)carbonyl-2-azetidinone (15).
Reaction of 2-azetidinone (53 mg; 0.75 mmol), DMAP
(12 mg; 0.15 mmol), benzyl isocyanate (102 mL; 0.85
mmol) and TEA (42 mL; 0.40 mmol) yielded 15 (148 mg,
96%) as a colourless oil; IR (film) 3370, 2985, 1762,
1721 cmꢀ1; 1H NMR (300 MHz, CDCl3) d 3.07 (t, J=5.3
Hz, 2H, H-4), 3.67 (t, J=5.3 Hz, 2H, H-3), 4.49 (d,
J=6.7 Hz, 2H, CH2Ph), 6.82 (br s, 1H, NH), 7.28–7.42
(m, 5H, Ph); 13C NMR (75 MHz, CDCl3) d 36.0 and 37.2
(C-3 and C-4), 43.6 (CH2Ph), 127.5, 127.6 and 128.7
(CHAr), 137.8 (CAr), 149.8 (CO urea), 167.0 (CO azeti-
dinone); MS (CI) m/z 205 (M+1, C11H13N2O2), 91, 70.

´
S. Gerard et al. / Bioorg. Med. Chem. 12 (2004) 129–138
137
3.2.6.3.
1-(Benzylamino)thiocarbonyl-4-(benzyloxy)-
3.5. Assay of HLE (human leukocyte elastase)
carbonyl-2-azetidinone (21). This compound was
obtained from 4-(benzyloxy)carbonyl-2-azetidinone
(205 mg; 1 mmol) by treatment with LiHMDS (241 mg;
1 mmol) and benzyl isothiocyanate (146 mL; 1.1 mmol)
to yield 21 (64mg, 20%) as a yellow oil; IR (film) 3336,
To 5 mL of the solution of substrate [solution of meth-
oxysuccinyl-Ala-Ala-Pro-Val-p-nitroanilide (50 mM in
DMSO)] diluted with 490 mL of buffer (Tris–HCl 0.1 M,
pH 7.5 and NaCl 0.5 M solution) were added 5 mL of
the solution of elastase (in Tris–HCl and NaCl buffer)
and 3 mL of the solution of the tested compound (10ꢀ2
M in DMSO). The appearance of the substrate hydro-
lysis product (p-nitroaniline) was measured (with a Cary
210 spectrophotometer) at 410 nm as a function of time
and the residual activity was obtained by comparison
with the variation of the absorbance of a reference
(sample without inhibitor).
1
1775, 1751, 1269 cmꢀ1; H NMR (300 MHz, CDCl3) d
2.95 (dd, 1H, J=15.7 Hz, 3.2 Hz, H-30), 3.22 (dd, 1H,
J=15.7 Hz, 6.1 Hz, H-3), 4.66 (dd, 1H, J=6.1 Hz, 3.2
Hz, H-30), 4.81 (d, J=5.6 Hz, 2H, NHCH2), 5.28 (AB
pattern, 2H, JAB=12.4Hz, CH 2Ph), 7.26–7.39 (m, 10H,
Ph), 8.52 (br s, 1H, NH); 13C NMR (75 MHz, CDCl3) d
39.5 (C-3), 48.7 (C-4), 51.4 (NHCH2), 67.7 (CH2Ph),
127.8, 127.9, 128.1, 128.2, 128.6 and 128.7 (CHAr), 134.8
and 136.1 (CAr), 163.0 (CO azetidinone), 168.8 (CO
ester), 177.1 (CO thiourea); MS (CI) m/z 355 (M+1,
C19H19N2O3S), 150, 91.
Acknowledgements
This work was supported by UCL, PAI and FNRS
(Belgium). J.M-B. and G.D. are senior research associ-
ates of FNRS (Belgium). The authors thank B. Clamot
and C. Barthelemy for technical assistance.
3.3. Theoretical calculations
All the calculations have been performed at the ab initio
RHF level using the minimal basis set MINI-10. They
were performed with Gaussian 98 on two computers, a
Dec Alpha 8400 8-processor and a Dec Alpha 4100 4-
processor running Digital Unix.22 The starting geome-
tries were sketched drawn by standard fragments and
then completely optimized following all the 3N-6
degrees of freedom either for the minima or the transi-
tion state structures. For each equilibrium structure, the
thermochemistry data are derived from the analytical
frequency calculation at 298.15 K and 1 atm.
References and notes
1. (a) Hastla, D. J.; Pagani, E. D. Ann. Resp. Med. Chem.
1994, 29, 195. (b) Berstein, P. R.; Edwards, P. D.; Wil-
liams, J. C. Prog. Med. Chem. 1994, 31, 59. (c) Anderson,
G. P.; Shinagawa, K. Curr. Opin. Antiinfl. Immun. Invest.
Drugs 1999, 1, 29.
2. (a) Gompertz, S.; Stockley, R. A. Semin Resp. Inf. 2000,
15, 14. (b) Stockley, R. A. Am. J. Respir. Crit. Care Med.
1999, 160, S49. (c) Skiles, J. W.; Jeng, A. Y. Exp. Opin.
Ther. Pat. 1999, 9, 869.
3. (a) Powers, J. C.; Asgian, J. L.; Ekici, O. D.; James, K. E.
Chem. Rev. 2002, 102, 4639. (b) Powers, J.C.; Zimmer-
man, M. In Design of Enzyme Inhibitors as Drugs; Sand-
ler, M., Smith, H. J., Eds.; Oxford University Press:
Oxford; 1989; p 596.
4. (a) Bode, W.; Wei, A.-Z.; Huber, R.; Meyer, E. F., Jr.;
Travis, J.; Neumann, S. EMBO J. 1986, 5, 2453. (b) Wei,
A.-Z.; Mayr, I.; Bode, W. FEBS Lett. 1988, 234, 367. (c)
Navia, M. A.; Springer, J. P.; Lin, T.-Y. Nature 1987, 327,
79. (d) Navia, M. A.; Mc Keever, B. M.; Springer, J. P.
Proc. Natl. Acad. Sci. U.S.A. 1989, 86, 7.
3.4. Assay of PPE (porcine pancreatic elastase)
To 2 mL of the solution of substrate (solution of N-
[2.7
succinyl-l-alanyl-l-alanyl-l-alanyl-p-nitroanilide
mg in 200 mL of N-methyl pyrrolidone (NMP) diluted
with TRIS buffer (20 mL at 100 mM, pH 7.5 solution)]
were added 20 mL of the solution of the tested com-
pound (10ꢀ2–10ꢀ4 M in NMP) and 67 mL of the solu-
tion of elastase [6ꢂ10ꢀ6 M in acetate buffer (50 mM, pH
5)]. The appearance of the substrate hydrolysis product
(p-nitroaniline) was measured (with a Cary 210 spectro-
photometer) at 410 nm as a function of time.
5. (a) Radhakrishnan, R.; Presta, L. G.; Meyer, E. F., Jr.;
Wildonger, R. J. Mol. Biol. 1987, 198, 417. (b) Meyer,
E. F., Jr.; Cole, G.; Radhakrishnan, R.; Epp, O. Acta
Crystallogr. 1988, B44, 26. (c) Bode, W.; Meyer, E. F., Jr.;
Powers, J. C. Biochemistry 1989, 28, 1951. (d) Wurtele,
M.; Hahn, M.; Hilpert, K.; Hohne, W. Acta Crystallogr.
2000, D56, 520.
Plots of V/VI versus [I] (ratios of initial rates of hydro-
lysis in the absence and in the presence of inhibitors)
gave the inhibition constants indicated in Tables 1
and 3. All experiments were performed two or three
times.
6. Cregge, R. J.; Durham, S. L.; Farr, R. A.; Gallion, S. L.;
Hare, C. M.; Hoffman, R. V.; Janusz, M. J.; Kim, H.-O.;
Koehl, J. R.; Mehdi, S.; Metz, W. A.; Peet, N. P.; Pelton,
J. P.; Schreuder, H. A.; Sunder, S.; Tardif, C. J. Med.
Chem. 1998, 41, 2461.
7. Kuang, R.; Epp, J. B.; Ruan, S.; Chong, L. S.; Venka-
taraman, R.; Tu, J.; He, S.; Truong, T. M.; Groutas, W. C.
Bioorg. Med. Chem. 2000, 8, 1005.
The reversibility of the inhibition was controlled by the
incubation/dilution method. The incubation solution
consisted of 5 mL of inhibitor solution (10ꢀ3 M in
DMSO), 34 mL of elastase solution (2.5ꢂ10ꢀ4 M in
acetate buffer, 50 mM, pH 5) and 161 mL of Tris buffer
(100 mM, pH 7.5). At various times (0, 5, 10, 15, 20, 30
min), 10 mL of the incubation solution were dissolved in
2 mL of the substrate solution (10ꢀ4 M in TRIS buffer),
and the enzyme activity was measured as before. No
decrease of activity was recorded as a function of incu-
bation time.
8. Pochet, L.; Doucet, C.; Dive, G.; Wouters, J.; Masereel,
B.; Reboux-Ravaux, M.; Pirotte, B. Bioorg. Med. Chem.
2000, 8, 1489.
9. Siebel, J.; Brown, D.; Amour, A.; MacDonald, S. J.;

´
S. Gerard et al. / Bioorg. Med. Chem. 12 (2004) 129–138
138
Oldham, N. J.; Schofield, C. J. Bioorg. Med. Chem. Lett.
2003, 13, 387.
Chem. 1996, 58, 85. (b) Dive, G.; Dehareng, D. Int. J.
Quant. Chem. 1999, 73, 161.
10. MacDonald, S. J.; Dowle, M. D.; Harrison, L. A.; Spooner,
J. E.; Shah, P.; Johnson, M. R.; Inglis, G.; Clarke, G.;
Belton, D.; Smith, R.; Molloy, C.; Dixon, M.; Murkitt,
G.; Godward, R. E.; Skarzynski, T.; Singh, O.; Kumar,
K. A.; Hodgson, S. T.; McDonald, E.; Hardy, G. W.;
Finch, H. Bioorg. Med. Chem. Lett. 2001, 11, 243.
11. (a) Mascaretti, O. A.; Boschetti, C. E.; Danelon, G. O.;
Mata, E. G.; Roveri, O. A. Curr. Med. Chem. 1995, 1,
441, and references cited therein.. (b) Finke, P. E.; Shah,
S. K.; Fletcher, D. S.; Ashe, K. M.; Maycock, A. L.;
Osinga, D. G.; Underwood, D. J.; Weston, H.; Davies, P.;
Doherty, J. B. J. Med. Chem. 1995, 38, 2449. (c) Vergely,
18. Aoyama, Y.; Uenaka, M.; Kii, M.; Tanaka, M.; Konoike,
T.; Hayasaki-Kajiwara, Y.; Noya, N.; Nokajima, M.
Bioorg. Med. Chem. 2001, 9, 3065.
19. (a) Yokoyama, M.; Imamoto, T. Synthesis 1984, 797. (b)
Sato, M.; Fukada, N.; Kurauchi, M.; Takeshima, T.
Synthesis 1981, 554. (c) Takeshima, T.; Ikeda, M.;
Yokoyama, M.; Fukada, N.; Muraoka, M. J. Chem. Soc.,
Perkin Trans. 1 1979, 692.
20. Anderson, B. A.; Becker, G.; Carty, J.; Harn, N.;
Hatfield, L.; Neubauer, B.; Rizzo, J.; Zhang, T. Y., Patent
WO 9825895 A1 980618.
21. Salzmann, T. N.; Ratcliffe, R. W.; Christensen, B. G.;
Bouffard, F. A. J. Am. Chem. Soc. 1980, 120, 6161.
22. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria,
G. E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.;
Montgomery, J. A., Jr.; Stratmann, R. E.; Burant, J. C.;
Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.;
Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi,
M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.;
Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.;
Cui, Q.; Morokuma, K.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz,
J. V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.; Liashenko,
A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M.; Peng, C. Y.;
Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill,
P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Andres,
J. L.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaus-
sian 98, revision A.7; Gaussian, Inc: Pittsburgh PA, 1998.
I.; Laugaˆ a, P.; Reboud-Ravaux, M. J. Mol. Graph. 1996,
14, 158.
12. (a) Beauve, C.; Tjoens, G.; Touillaux, R.; Lamotte-
Brasseur, J.; Marchand-Brynaert, J.; Fastrez, J. Eur. J.
Org. Chem. 1999, 1441. (b) Gerard, S.; Nollet, G.; Vande
Put, J.; Marchand-Brynaert, J. Bioorg. Med. Chem. 2002,
10, 3955.
13. Gerard, S.; Dive, G.; Clamot, B.; Touillaux, R.; Marchand-
Brynaert, J. Tetrahedron 2002, 58, 2423.
14. Dehareng, D.; Dive, G.; Vigneron, L.; Gerard, S.;
Marchand-Brynaert, J. Unpublished results.
15. Baldwin, J. E.; Adlington, R. M.; Collins, D. W.;
Schofield, C. J. Tetrahedron 1990, 46, 4733.
16. Han, W. T.; Trehan, A. K.; Wright, J. J. K.; Federici,
M. E.; Seiler, S. M.; Meanwell, N. A. Bioorg. Med. Chem.
1995, 3, 1123.
17. (a) Dive, G.; Dehareng, D.; Peeters, D. Int. J. Quant.