Monocyclic β-lactams as antibacterial agents: Facing antioxidant activity of N-methylthio-azetidinones

Article abstract of DOI:10.1016/j.ejmech.2012.12.024

A series of N-methylthio-β-lactams with antibacterial activity were thoroughly evaluated as antioxidants. We found that only the presence of a polyphenolic moiety anchored to the β-lactam ring ensured an adequate antioxidant potency. New compounds, efficiently combining in one structure antioxidant and antibacterial activity, may provide a promising basis for the development of new leads useful in adverse clinical conditions such as in cystic fibrosis patients, in whom colonization by MRSA and epithelial damage by chronic pulmonary oxidative stress take place.

Full text of DOI:10.1016/j.ejmech.2012.12.024

European Journal of Medicinal Chemistry 60 (2013) 340e349
Contents lists available at SciVerse ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Monocyclic
b-lactams as antibacterial agents: Facing antioxidant
activity of N-methylthio-azetidinones
,
a
a
b
b
Rinaldo Cervellati a , Paola Galletti , Emanuela Greco , Clementina E.A. Cocuzza , Rosario Musumeci ,
**
Luca Bardini a, Francesco Paolucci a, Matteo Pori a, Roberto Soldati a, Daria Giacomini a
,
*
a Department of Chemistry “G. Ciamician”, University of Bologna, Via Selmi 2, 40126 Bologna, Italy
b Department of Surgery and Interdisciplinary Medicine, University of Milano-Bicocca, Via Cadore 48, 20900, Monza, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of N-methylthio-
We found that only the presence of a polyphenolic moiety anchored to the
adequate antioxidant potency. New compounds, efficiently combining in one structure antioxidant and
antibacterial activity, may provide a promising basis for the development of new leads useful in adverse
clinical conditions such as in cystic fibrosis patients, in whom colonization by MRSA and epithelial
damage by chronic pulmonary oxidative stress take place.
b
-lactams with antibacterial activity were thoroughly evaluated as antioxidants.
-lactam ring ensured an
Received 25 October 2012
Received in revised form
29 November 2012
Accepted 8 December 2012
Available online 20 December 2012
b
Ó 2012 Elsevier Masson SAS. All rights reserved.
Dedicated to Prof. Gianfranco Cainelli on the
occasion of his 80th birthday.
Keywords:
Beta-lactams
Antioxidants
Antibacterial
Cystic fibrosis
Polyphenols
Azetidinones
1. Introduction
resistance to b-lactam antibiotics. This is particularly important in
patients with chronic diseases such as cystic fibrosis (CF) where
persistent colonization by pathogenic bacteria occurs and the
repeated use of antibacterial agents selects for specific resistant
strains. A rise in S. aureus infections has been reported in CF
patients, with an increase in the prevalence of highly virulent,
methicillin-resistant S. aureus strains (MRSA) [4].
CF is characterized by a chronic inflammatory process even in
the absence of bacterial pathogens, and by the recruitment of
activated neutrophils. Under pathophysiological conditions, acti-
vated neutrophils and epithelial cells release highly reactive
molecules in the extracellular space, like reactive oxygen species
(ROS) and reactive nitrogen species (RNS) in order to attack and
eliminate invasive pathogens [5]. However, in CF, several lines of
evidence show that the defence systems are ineffective. In CF an
increased production of ROS may be associated with cell dysfunc-
tion and with disease progression [6]. Thus, the presence of
oxidative stress in CF due to an increased production of ROS and to
an impaired antioxidant status, particularly during the chronic
pulmonary infections, points to new therapeutic possibilities in
targeting anti-oxidant pathways [7].
b
-Lactam antibiotics are still the main class of agents used to
treat bacterial infections [1]. However, notwithstanding the pivotal
role of penicillins and cephalosporins as first choice antibiotics, the
treatment of bacterial infections has been hindered by the
increasing emergence of multidrug-resistant microorganisms,
which today represents one of the greatest challenges in the clinical
management of infectious diseases [2]. Whereas antibiotic resis-
tance was once largely confined to hospitals and long-term care
facilities, it has now emerged in community settings and represents
one of the most pressing global public health concerns [3].
Staphylococcus aureus is a major human pathogen causing
significant morbidity and mortality in both community- and
hospital-acquired infections. Many staphylococcal infections are
associated with multiple recurrences and with the development of
* Corresponding author. Tel.: þ39 051 209 9528; fax: þ39 051 209 9456.
** Corresponding author. Tel.: þ39 051 209 9467; fax: þ39 051 209 9456.
E-mail addresses: rinaldo.cervellati@unibo.it (R. Cervellati), daria.giacomini@
unibo.it (D. Giacomini).
0223-5234/$ e see front matter Ó 2012 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ejmech.2012.12.024

R. Cervellati et al. / European Journal of Medicinal Chemistry 60 (2013) 340e349
341
As part of an interdisciplinary research project, involving the
synthesis of new monocyclic -lactams specifically designed to
selective towards Staphylococcus spp. and Bacillus spp. Moreover,
unlike penicillins, which inhibit cell wall crosslinking enzymes,
N-thiolated lactams are characterized by a bacteriostatic activity
and act through a different mechanistic model [14]. Recently some
monocyclic N-methylthio-azetidinones were even reported as
selective inhibitors of Histone deacetylases (HDACs) [15]. In that
case the presence of an N-methylthio group had a key role in
b
target resistant Gram-positive bacterial strains, a series of new
azetidinones with polyphenolic side chains were evaluated for both
their antibacterial [8] and antioxidant activities [9]. As a general
trend, to activate the antioxidant potency of the tested compounds
to a significant extent the presence of phenolic OH resulted
necessary, with the exception of 4-alkylidene-thiophenol deriva-
tive (1) which had no phenolic residues but retained a certain
antioxidant activity.
More recently a series of N-methylthio-azetidinones were eval-
uated in vitro against Gram-positive and Gram-negative clinical
isolates and in particular on MRSA and MSSA strains isolated from
patientswith cystic fibrosis[10]. The combination of an N-SMe group
and a benzyl ester on the 4-alkylidene-side chain, or electron-
providing to the new
2.1. Synthesis of -lactams
All 15 monocyclic -lactams investigated here are depicted in
b-lactams a stringent isoform selectivity.
b
b
Fig. 1. Compounds 1, 2, 3, 4, 5, 7, 8, 9, and 10 were prepared
according to previously reported procedures [10,16]. N-Methyl-
sulfoxide-4-acetoxy-azetidin-2-one 6 (Scheme 1) was prepared
starting from commercial 4-acetoxy-azetidin-2-one which was
treated with LiHMDSA followed by addition of mesylchloride.
Analogously, N-methylthio-azetidin-2-one 11 was obtained treat-
ing commercial azetidin-2-one with LiHMDSA and then S-methyl
methanethiosulfonate (Scheme 1).
Compounds 12e15 were prepared starting from the commercial
(3R,4R)-4-acetoxy-3-[(1R)-1-(tert-butyldimethylsilyloxy)-ethyl]-az-
etidin-2-one. The starting compound was N-thio methylated to
afford compound 3 using Et3N as a base because LiHMDSA or
LDA led to the formation of significant amounts of by-products.
Compound 3 was then deprotected on the 3-[O-(tert-butyldi-
methylsilyl)]-ethyl side chain using BF3$Et2O in CH3CN to give
the corresponding alcohol 4 (Scheme 2) [16]. Treatment of 4
with the appropriate O-protected benzoic acid 19, 21 or 22, DCC
or EDC as coupling reagents and in the presence of a catalytic
amount of DMAP in dichloromethane, gave compounds 12, 16
and 17.
withdrawing groups, such as OAc, on the C-4 position of the b-lac-
tam ring, appeared to strengthen the potency against Gram-positive
bacteria. This study reports the evaluation of physicochemical
parameters, such as the antioxidant activity and redox potentials, of
some N-methylthio-azetidinones with antibacterial potency. The
aim of this ongoing project was in fact to develop dual targeting
molecules with elements incorporated in their structure that would
allowdualantibacterialandantioxidantactivitiesinordertocontrast
unfavourable clinical conditions, such as those resulting from
bacterialcolonizationof thelungassociatedwithextensiveepithelial
damage brought about by chronic pulmonary oxidative stress.
2. Results and discussion
Beside bicyclic b-lactam antibiotics such as penicillins, cepha-
losporins, and carbapenems, monocyclic compounds (azetidi-
nones) emerged for their multiple and interesting biological
activities [11]. The discovery of monobactams (Aztreonam)
demonstrated for the first time that a conformationally constrained
bicyclic structure was not necessary for antibacterial activity [12].
Turos and co-workers reported for the first time the synthesis
and the biological evaluation as antibacterial agents of some
As regards the benzoic acids, methyl 4-(ethoxymethoxy)-3,5-
dimethoxybenzoic acid 19 and (E)-3-(3,4-bis(ethoxymethoxy)
phenyl)acrylic acid 21 were prepared starting from syringic methyl
ester or caffeic methyl ester, respectively through protection of the
phenolic oxygen atoms with chloromethylethylether followed by
alkaline hydrolysis of methyl esters 18 and 20 (Scheme 3). Finally,
compounds 12 and 17 were treated with trifluoroacetic acid (TFA)
monocyclic
b-lactams with an alkylthio group on the b-lactam
nitrogen atom [13]. These compounds have been found to be highly
Fig. 1.
b-Lactam compounds tested for the antioxidant activity.

342
R. Cervellati et al. / European Journal of Medicinal Chemistry 60 (2013) 340e349
TEAC values for compounds 14 and 15 are in line with those of
some polyphenolic- -lactams we previously reported, ranging
b
from 0.40 to 2.0 mM equiv Trolox [9]. The DPPH and FRAP values
are in satisfactory agreement with the TEAC ones taking into
account the different chemistry of the methods. For comparison,
the TEAC values for ascorbic acid (vitamin C) and
a-tocopherol
(vitamin E) are 1.05 and 0.97 respectively [20a], the DPPH values
are 1.1 and 0.5 respectively [20b], and the FRAP values for the two
vitamins is 2.0 [23]. As expected compound 12 doesn’t shows
antioxidant activity with these three methods because it does not
contain phenolic OH groups in the molecule (negative control).
Compound 13 is a very low active antioxidant than the other two
phenolic substituted molecules, a possible interpretation is given in
next paragraph.
Scheme 1. Synthesis of b-lactams 6, and 11.
to eliminate the phenolic protection and to give compounds 13 and
15. The ethoxymethylether as protective group was not effective in
the synthesis of 3,4-dihydroxybenzoic ester 14 which was
successfully obtained using t-butyldimethylsilyl to protect the
phenolic positions in compound 16.
The BR values for all compounds are surprisingly higher than
those of some polyphenolic-
b-lactams previously reported, ranging
from 0.021 to 0.28 M equiv Re [9]. We suspect that these unusual
m
results are due to parallel reaction of oxidation of sulphur in the N-
SMe group by acidic iodate that compete with the HOOꢀ scavenging
reaction. On the contrary this parallel reaction doesn’t occur with
2.2. Antioxidant activity
the usual N-unsubstituted b-lactams.
Antioxidant activity determination based on different
approaches was carried out, in light of the importance of a multi-
2.3. Structureeactivity relationships
dimensional evaluation of the antioxidant activity [17]. The b-lac-
tams were then subjected to four antioxidant assays: the Briggse
Rauscher (BR) oscillating reaction [18,19], the Trolox Equivalent
Antioxidant Capacity (TEAC) assay [20a], the DPPH (2,2-diphenyl-
1-picrylhydrazyl DPPHꢀ stable radical) test [21,22] and the Ferric
Reducing Antioxidant Power (FRAP) method [23] (see Experimental
section and Supplementary data for details).
The majority of the synthesized b-lactams did not show appre-
ciable antioxidant activity with BR, TEAC, DPPH, or FRAP methods;
moreover compounds 2, 3, and 10 were also tested with the Foline
As a general trend, to activate the antioxidant potency of the
tested compounds, the presence of two or three phenolic OH
groups are necessary [25]. A theoretical method to calculate the
Bond Dissociation Enthalpies (BDE) for molecules belonging to the
class of polyphenols and to correlate them with their free radical
scavenging activities has been reported by Wright et al. [26]. These
authors also proposed empirical additivity rules that take into
account the electronic, H-bond and conjugation effects of substit-
uents in the phenol parent molecule to evaluate the BDE of a given
Ciocalteu (FC) reagent [24] to check if the N-methylthio-
skeleton has some reductive properties, but the result was negative.
N-unsubstituted compounds 8 and 9 gave also negative results to
b
-lactam
phenolic OH group. In this way a value of BDE or
D
BDE ¼ (BDE-
comp ꢁ BDEF-OH) can easily be calculated from the data reported in
Ref. [26] Table 4. It was found that the number of phenolic OH is not
so important and that it is the strategic placing of these groups with
respect to other substituents that determine the strength of the OH
antioxidant tests. Only the N-methylthio-b-lactams 12e15 with
phenolic moieties were active and showed values in Table 1.
Scheme 2. Synthesis of
b-lactams 13, 14, 15, 16, and 17.

R. Cervellati et al. / European Journal of Medicinal Chemistry 60 (2013) 340e349
343
Scheme 3. Synthesis of acids 19, 22 and 21.
group and then the antioxidant activity. From the data reported in
Ref. [26] Table 4, we calculated the BDE values for two model
compounds A and B in comparison with the free radical scavenging
activity for compounds 13 and 14 (Fig. 2).
As can be seen from Fig. 2, results for the model match well with
the experimental data, thus justifying the very low free radical
scavenger potency of compound 13 with respect to that of 14.
ability of such model compounds to undergo oxidation to evaluate
their antioxidant activity through the measure of the correspond-
ing standard potentials. We also submitted to cyclic voltammetry
the thioester 1 which has no antibiotic activity and that in
a previous work [9] showed some antioxidant activity both in the
BR or TEAC tests (acidic and neutral pH method, respectively), and
explained with a possible generation of an intermediate radical
cation R2Sꢀþ. The voltammograms are reported in Fig. 3.
D
2.4. Voltammetry
All species displayed in acetonitrile a very similar CV behaviour
showing, in the positive potential region, a main anodic peak at
about 0.9 V attributed to the reversible oxidation of the pristine
species. Weak-to-strong adsorption effects were observed in the CV
pattern, leading in most cases to fouling of the electrode surface
and thus to a highly irreversible behaviour. In the case of gold
electrodes (Fig. 3), however, a more reproducible behaviour was
observed, although adsorption was still present and was respon-
sible for the asymmetric pattern of the CV curve and the apparent
larger anodic currents (forward scan) with respect to the cathodic
ones (reverse scan). This was particularly evident in the case of
compound 1. In such a case, in fact, the growth of an intense anodic
It is well known that thioethers can be oxidized to sulfoxides
and further to sulfones by H2O2 or by many other oxidizing agents
[27]. The negative results of the antioxidant assays gave evidence
that in the experimental conditions the N-methylthio group was
not prone to be oxidized. To interpret these results we decided to
perform some cyclic voltammetry experiments on the two mole-
cules 9 and 11 chosen as models. In the first molecule the bivalent
sulphur atom is bonded to C4 atom of the ring while in the second
molecule the bivalent sulphur atom is bonded to the nitrogen atom
of the beta lactam ring. Aim of the investigation was to measure the
Table 1
Results of antioxidant and antibacterial tests.
Compd.
BR (
Resorcinol)
m
M equiv
TEAC (mmol equiv
TROLOX)
DPPH (mmol equiv
TROLOX)
FRAP (mmol equiv Fe)
Antimicrobial activity on MRSA from CFa
MIC range (mg/L)
MIC50 (mg/L)
MIC90 (mg/L)
4
5
8
9
10
12
13
14
15
FOTe
CTXe
nab
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
na
32->128c
64c
32c
64c
32c
16c
nd
64
128
64
>128c
32e64c
64c
16e128c
16e64c
128c
64c
na
na
16e32c
32c
0.70d
Negligible
0.0091 ꢂ 0.0002
1.96 ꢂ 0.13
1.23 ꢂ 0.03
ndb
nd
128
>128
64
1.8 ꢂ 0.3
0.38 ꢂ 0.02
1.95 ꢂ 0.06
0.037 ꢂ 0.001
1.23 ꢂ 0.02
0.98 ꢂ 0.03
0.19 ꢂ 0.02
1.91 ꢂ 0.07
1.98 ꢂ 0.04
16e128
32 to >128
16e128
1 to >128
1 to >128c
128
>128
8c
>128c
a
S. aureus American Type Culture Collection ATCC 29213, and ATCC 43300 were used as control strains.
b
c
na ¼ no activity; nd ¼ not determined.
See Ref. [10].
d
e
Result of only one measure at 14.7
mM in mixture.
FOT ¼ cefotaxime; CTX ¼ ceftriaxone were used as reference antibiotics.

344
R. Cervellati et al. / European Journal of Medicinal Chemistry 60 (2013) 340e349
(TPTZ)2Fe3þ can thermodynamically be reduced by compounds
with lower reduction potential. This is likely applicable to all
N-methylthio- -lactams synthesized here, except for compounds
b
13, 14 and 15. These compounds in fact contain phenolic moiety in
the side chain, and many phenolic compounds have lower oxidation
potential thus reacting with ABTSꢀþ, DPPHꢀ and (TPTZ)2Fe3þ [30].
Finally, the low antioxidant activity of compound 1 observed in
a previous work [9] cannot be attributed, on the basis of the stan-
dard potential value reported above, to the occurrence of outer-
sphere electron transfer between the sulphur atom (donor) and
the antioxidant test reagents (HOOꢀ, ABTSꢀþ, (TPTZ)2Fe3þ acceptors)
since the energy driving force would be largely unfavourable. On
the other hand, in the actual chemical system the formation of
a complex between the b-lactam and the test reagents can take
place. Thus, an intimate interaction of some sort between the
partners might render the electron transfer a thermodynamically
more accessible process [31] in a fashion akin to the inner-sphere
mechanism previously evidenced in the reactions of Grignard
reagents with benzophenone [32].
In line with such a hypothesis, compound 1 displays the
strongest adsorption behaviour on the gold surface among the
investigated species and, because of the free energy of adsorption,
its oxidation (inner-sphere process) becomes significantly easier
than that involving the free species in solution (outer-sphere
process).
Fig. 2.
DBDE for two model compounds A and B in comparison with the free radical
scavenging activity for compounds 13 and 14.
shoulder located at less positive potentials than the main peak
suggests relatively strong adsorption of the product of the anodic
oxidation onto the gold surface [28]. Functional groups within the
molecule in question, which may give rise to this phenomenon are
many. In addition to sulphur, phenyl groups may effectively
promote the adsorption on gold, as well as the carboxyl groups,
thus making identification of the exact nature of the process rather
difficult. In spite of the adsorption effects on the CV curve
morphology, an evaluation of the standard potential for all species
was possible: 0.82, 0.86 and 0.88 V for compounds 9, 11 and 1,
respectively.
On the basis of such results, and assuming that potentials
determined under the aprotic and relatively apolar conditions of CV
(Fig. 3) may be directly used for comparisons in the aqueous media,
all the investigated compounds should not show appreciable anti-
oxidant activity towards TEAC, DPPH or FRAP antioxidant methods.
The redox potential of the couples ABTSꢀþ/ABTS (TEAC); DPPHꢀ/
DPPH, (TPTZ)2Fe3þ/(TPTZ)2/Fe2þ (FRAP) are in fact 0.68 V [17],
0.28 V [29], and 0.77 V [30], respectively, then ABTSꢀþ, DPPHꢀ,
2.5. Antibacterial activity
In order to evaluate the potential antibacterial activity of the
new
b-lactam derivatives with anti-oxidant properties, the
minimum inhibitory concentrations (MICs) of compounds 13e15
were determined in vitro against 45 clinical strains of MRSA iso-
lated from CF patients. Preliminary results (see Table 1) on the
antibacterial activity demonstrated compounds 13 and 15 to be the
most active with MIC values ranging from 16 to 128 mg/L against
the tested isolates; in particular compound 15 showed MIC50 and
MIC90 values equal to 64 mg/L.
Table 1 summarizes the antioxidant and antibacterial data of
exemplificative new
b-lactams. Compounds 4, 5, 8, 9, and 10
notwithstanding the potency against resistant S. aureus strains [10]
and the presence of a sulphur atom in the structure did not present
a sufficient antioxidant activity. The dual activity is present in case
of compounds 13, 14, 15 in which the presence of phenolic residues
on the hydroxyethyl-side chain switched on the antioxidant
potency thus conjugated with an encouraging result of antibacte-
rial activity.
3. Conclusions
In summary, the antioxidant behaviour of a series of monocyclic
N-methylthio-b-lactams was deeply investigated with several
methods: BR, TEAC, DPPH, and FRAP. Cyclic voltammetry
measurements of three model azetidinones with sulphur substit-
uents revealed redox potential in the range 0.82e0.88 eV thus
explaining the impossibility of an N-methylthio group of an azeti-
dinone to undergo oxidative reactions with the methods above
indicated. However the presence of a phenolic moiety on a side
chain on the b-lactams 13e15 activated the ability to be radical
scavenger against HOOꢀ, ABTSꢀþ, and DPPHꢀ.
Preliminary results on the antibacterial activity of these new
compounds against clinical strains of MRSA isolated from CF
patients gave promising results in terms of MIC values.
These encouraging results prompt us to currently work on the
synthesis of a library of antioxidant N-methylthio-4-acetoxy-aze-
tidinones with polyphenolic residues on the side chain which will
Fig. 3. Current potential diagrams for compounds 1, 9, and 11.

R. Cervellati et al. / European Journal of Medicinal Chemistry 60 (2013) 340e349
345
be tested on methicillin-resistant strains of S. aureus isolated from
CF patients. Establishment of a structureeactivity relationship
(SAR) profile will add to the development of new leads capable of
counteracting the adverse conditions in CF patients due to persis-
tent colonization by resistant bacteria and extensive epithelial
damage by chronic pulmonary oxidative stress. We are obviously
aware that no chemical test can mimic what happens in the human
body, however these tests may provide useful information in view
of a possible development of new compounds as drugs with two
synergistic pharmacological properties: antibacterial and antioxi-
dant activity.
4.8 Hz, CHCHH). dC (100 MHz, CDCl3) 20.7, 42.5, 45.7, 75.6, 161.0,
169.3. HPLCeMS (ESI): Rt ¼ 2.47 min, m/z: 225 [M þ H2O]þ, 230
[M þ Na]þ, 246 [M þ K]þ, 437 [2M þ Na]þ.
4.2.2. 1-(Methylthio)azetidin-2-one (11)
LiHMDSA (2.2 mL 1 M solution in THF, 2.2 mmol) was added to
a solution of 2-azetidinone (142 mg, 2 mmol) in THF (14 mL)
at ꢁ78 ꢃC under inert atmosphere, followed shortly by S-methyl
methanethiosulfonate (514 mL, 5 mmol). The solution was allowed
to warm to room temperature and was monitored by TLC. After 3 h,
the reaction was quenched with aqueous NH4Cl (15 mL) and
extracted with EtOAc (3 ꢄ 10 mL). The organic extracts were dried
over Na2SO4 and the residue was purified by flash-chromatography
(cyclohexane/ethylacetate: 75/25) to afford product 11 (187 mg,
80%) as a pale yellow oil. Found C, 40.83; H, 5.98; N, 11.87; S, 27.15%;
C4H7NOS requires C, 41.00; H, 6.02; N, 11.95; S, 27.37%. nmax/cmꢁ1
2970, 2917, 1757 and 1161. dH (200 MHz, CDCl3) 2.41 (3H, s, SMe),
3.04 (2H, t, J ¼ 4.8 Hz, CH2), 3.40 (2H, t, J ¼ 4.8 Hz, CH2). dC
(100 MHz, CDCl3) 21.9, 38.7, 43.0, 170.9. HPLC-MS (ESI):
Rt ¼ 1.97 min, m/z: 118 [M þ H]þ, 135 [M þ H2O]þ, 140 [M þ Na]þ,
257 [2M þ Na]þ. GCeMS (EI, 70eV): Rt ¼ 8.04 min, m/z: 117 (40%,
Mþ), 75 (100, M ꢁ COCH2), 60 (50, M ꢁ COCH2 ꢁ Me).
4. Experimental
4.1. General
All reactions were performed under an inert atmosphere (N2).
Commercial reagents (reagent grade, >99%) were used as received
without additional purification. Anhydrous solvents (CH3CN,
CH2Cl2, THF) were obtained commercially.
HClO4 was analyzed by titration versus a standard 0.1 M NaOH
solution (from Merck). H2O2 was standardized daily by mangano-
metric analysis. All stock solutions were prepared with doubly
distilled, deionized water.
4.2.3. (R)-1-((2R,3R)-2-Acetoxy-1-(methylthio)-4-oxoazetidin-3-
yl)ethyl-4-(ethoxymethoxy)-3,5-dimethoxybenzoate (12)
1H and 13C NMR values were recorded on an INOVA 400, or
a GEMINI 200 instrument with a 5 mm probe. All chemical shifts
To a solution of 4 (117 mg, 0.53 mmol), in CH2Cl2 (20 mL) at 0 ꢃC
under inert atmosphere, the acid 19 (216 mg, 0.84 mmol), DMAP
(13 mg, 0.11 mmol) and DCC (174 mg, 0.84 mmol) were added. After
10 min the mixture was allowed to warm to room temperature.
After 70 h the reaction mixture was washed with cool water and
extracted with CH2Cl2 (3 ꢄ 15 mL). The collected organic phases
were dried on Na2SO4 and evaporated. The residue was treated
with EtOAc and filtered. The solid was discarded whereas the
solvent was evaporated and purified by flash-chromatography
(cyclohexane/EtOAc: 70/30) obtaining product 12 (177 mg, 73%) as
have been quoted relative to deuterated solvent signals,
d in parts
per million and J in hertz. FT-IR: Nicolet 380 measured as films
between NaCl plates, wavenumbers reported in cmꢁ1. TLC: Merck
60 F254. Column chromatography: Merck silica gel 200e300 mesh.
GCeMS: Agilent Technologies, column HP5 5% Ph-Me Silicon. MS:
Agilent Technologies MSD1100 single-quadrupole mass spectrom-
eter, EI voltage 70 eV, gradient from 50 ꢃC to 280 ꢃC in 30 min.
HPLCeMS, HPLC: Agilent Technologies HP1100, column ZOBRAX-
Eclipse XDB-C8 Agilent Technologies, mobile phase: H2O/CH3CN,
gradient from 30% to 80% of CH3CN in 8 min, 80% of CH3CN until
25 min, 0.4 mL/min MS: Agilent Technologies MSD1100 single-
quadrupole mass spectrometer, full-scan mode from m/z 50 to
m/z 2600, scan time 0.1 s in positive ion mode, ESI spray voltage
4500 V, nitrogen gas 35 psi, drying gas flow 11.5 mL/min, frag-
mentor voltage 20 V. Elemental analysis was performed on
a Thermo Flash 2000 CHNS/O Analyzer, and they were within
ꢂ0.4% of the theoretical values. A Shimadzu UV-1601 PC spectro-
photometer was used for spectrophotometric measurements.
a white syrup. Found C, 52.58; H, 5.94; N, 3.11; S, 6.89%; C20H27NO9S
22
requires C, 52.51; H, 5.95; N, 3.06; S, 7.01%. [
a]
ꢁ3.25 (c 1.6 in
D
CHCl3). nmax/cmꢁ1 3323, 2926, 2850, 1789, 1755, 1716 and 1126. dH
(400 MHz, CDCl3) 1.17 (3H, t, J ¼ 7.2 Hz, CH3CH2), 1.48 (3H, d,
J ¼ 6.4 Hz, CH3CHOCO), 2.14 (3H, s, CH3CO2), 2.47 (3H, s, SCH3), 3.46
(1H, dd, J ¼ 1.2 and 6.4 Hz, CHCHO), 3.84 (2H, q, J ¼ 7.2 Hz, CH3CH2),
3.87 (6H, s, 2 ꢄ OCH3), 5.20 (2H, s, OCH2O), 5.43 (1H, quintet,
J ¼ 6.4 Hz, CH3CHO), 6.21 (1H, d, J ¼ 1.2 Hz, CHOAc), 7.22 (2H, s,
arom.). dC (50 MHz, CDCl3) 14.8, 18.2, 20.8, 22.8, 56.1, 64.1, 64.9,
66.9, 81.5, 96.4, 106.7, 125.1, 138.9, 153.0, 164.9, 167.3, 169.6. HPLCe
MS (ESI): Rt ¼ 8.62 min, m/z: 475 [M þ H2O]þ.
4.2. Synthesis of b-lactams
Azetidinones 1 [33], and 2, 3, 4, 5, 7, 8, 9, 10, were prepared as
previously reported [10].
4.2.4. (R)-1-((2R,3R)-2-Acetoxy-1-(methylthio)-4-oxoazetidin-3-
yl)ethyl-4-hydroxy-3,5-dimethoxy benzoate (13)
A stirred solution of 12 (76 mg, 0.17 mmol) in CH2Cl2 under inert
4.2.1. 1-(Methylsulfonyl)-4-oxoazetidin-2-yl acetate (6)
atmosphere was treated with aliquots of trifluoroacetic acid (12 mL,
A 1 M solution of LiHMDSA (1.1 mL, 1.1 mmol) was added to
a solution of 4-acetoxy-2-azetidinone (129 mg, 1.0 mmol) in THF
(10 mL) at ꢁ78 ꢃC under inert atmosphere, followed by meth-
0.17 mmol) every 15 min until the disappearing of the starting
material. After the completion of the reaction, the solvent and the
trifluoroacetic acid were evaporated to obtain 13 (55 mg, 81%) as
anesulfonyl chloride (194
mL, 2.5 mmol). The solution was allowed
a yellow oil. Found C, 51.03; H, 5.37; N, 3.44; S, 7.89%; C17H21NO8S
22
to warm to room temperature and was monitored by TLC. After 3 h
the reaction was quenched with aqueous NH4Cl (15 mL) and
extracted with EtOAc (3 ꢄ 10 mL). The organic extracts were dried
over Na2SO4 and the residue was purified by flash-chromatography
(cyclohexane/ethylacetate: 60/40) to afford product 6 (42 mg, 20%)
as a pale yellow oil. Found C, 34.92; H, 4.51; N, 6.90; S, 15.32%;
C6H9NO5S requires C, 34.78; H, 4.38; N, 6.76; S, 15.48%. nmax/cmꢁ1
2970, 2917, 1757 and 1161. dH (400 MHz, CDCl3) 2.17 (3H, s, CH3CO2),
3.10 (1H, dd, J ¼ 2.0 and 16.4 Hz, CHCHH), 3.23 (3H, s, SO2CH3), 3.54
(1H, dd, J ¼ 4.8 and 16.4 Hz, CHCHH), 6.47 (1H, dd, J ¼ 2.0 and
requires C, 51.12; H, 5.30; N, 3.51; S, 8.03%. [
a
]
ꢁ36.3 (c 0.52 in
D
CHCl3). nmax/cmꢁ1 3412, 2938, 2850, 1783, 1756, 1711 and 1116. dH
(400 MHz, CDCl3) 1.50 (3H, d, J ¼ 6.4 Hz, CH3CHOCO), 2.16 (3H, s,
CH3CO2), 2.49 (3H, s, SCH3), 3.48 (1H, dd, J ¼ 1.2 and 6.0 Hz,
CHCHOCO), 3.94 (6H, s, 2 ꢄ OCH3), 5.44 (1H, quintet, J ¼ 6.4 Hz,
CH3CHO), 5.97 (1H, s, OH), 6.23 (1H, d, J ¼ 1.2 Hz, CHOAc), 7.26 (2H,
s, arom.). dC (100 MHz, CDCl3) 18.3, 20.8, 22.9, 56.4, 64.2, 66.7, 81.5,
106.7, 120.6, 139.5, 146.6, 165.2, 167.5, 169.7. HPLC-MS (ESI):
Rt ¼ 5.68 min, m/z: 417 [M þ H2O]þ, 422 [M þ Na]þ, 821
[2M þ Na]þ.

346
R. Cervellati et al. / European Journal of Medicinal Chemistry 60 (2013) 340e349
4.2.5. (R)-1-((2R,3R)-2-Acetoxy-1-(methylthio)-4-oxoazetidin-3-
yl)ethyl-3,4-bis((tert-butyldimethylsilyl)oxy)benzoate (16)
94.1, 115.5, 115.7, 116.0, 123.4, 128.4, 145.3, 147.5, 149.5, 165.9, 167.5,
169.8. HPLCeMS (ESI): Rt ¼ 9.9 min, m/z: 515 [M þ H2O]þ, 520
[M þ Na]þ, 1017 [2M þ Na]þ.
To a solution of 4 (121 mg, 0.55 mmol) in CH2Cl2 (8.2 mL),
compound 22 (210 mg, 0.55 mmol) and DMAP (67 mg, 0.55 mmol)
were added. The mixture was then cooled to 0 ꢃC, EDC (105 mg,
0.55 mmol) was added and the system was allowed to reach rt in
10 min. After 30 h, the reaction was quenched with water and some
drops of HCl (1 M), extracted with CH2Cl2, dried on Na2SO4 and
evaporated. The silylated intermediate 16 was obtained after flash-
chromatography (CH2Cl2/Et2O 99/1) in 23% yield (73 mg) as an oil.
4.2.8. (E)-(R)-1-((2R,3R)-2-Acetoxy-1-(methylthio)-4-oxoazetidin-
3-yl)ethyl-3-(3,4-dihydroxyphenyl)acrylate (15)
To a solution of 17 (94 mg, 0.19 mmol) in CH2Cl2 (23 mL) under
inert atmosphere aliquots of TFA (64.5 mL, 0.869 mmol) were added
every 15 min until the disappearing of the starting material (TLC
monitoring). The solution was evaporated and the crude was
purified by flash-chromatography (cyclohexane/ethylacetate 40/
60). The product obtained from the collected fractions was finally
triturated with pentane to afford the desired product 15 as a white
solid in 90% yield (65 mg). Found C, 53.89; H, 5.13, N 3.55; S, 8.37%;
[a]
22 ꢁ13.8 (c 0.5 in CH2Cl2). nmax/cmꢁ1 3427, 2930, 2858, 1793, 1758,
D
1719 and 1120. dH (400 MHz, CDCl3) 0.22 (6H, s, SitBuMe2), 0.23 (6H, s,
SitBuMe2), 0.99 (9H, s, SitBuMe2), 1.00 (9H, s, SitBuMe2), 1.47 (3H, d,
J ¼ 6.0 Hz, CH3CHOCO), 2.17 (3H, s, CH3CO2), 2.47 (3H, s, SCH3), 3.47
(1H, dd, J ¼ 1.6 and 6.0 Hz, CHCHO), 5.45 (1H, quintet, J ¼ 6.0 Hz,
CH3CHO), 6.26 (1H, d, J ¼ 1.2 Hz, CHOAc), 6.82e6.84 (1H, m, Ar),
7.46e7.49 (2H, m, Ar). dC (100 MHz, CDCl3) ꢁ4.2, ꢁ4.1, ꢁ4.1, ꢁ4.0,
18.3, 18.4, 18.5, 20.9, 22.8, 25.8, 25.9, 64.1, 66.3, 81.0, 120.4, 122.3,
122.9, 123.6, 146.8, 152.0, 165.0, 167.5, 169.8. HPLCeMS (ESI):
Rt ¼ 26.6 min, m/z: 601 [M þ H2O]þ, 606 [M þ Na]þ,1190 [2M þ Na]þ.
C17H19NO7S requires C, 53.53; H, 5.02; N, 3.67; S, 8.41%; mp: 53e
22
58 ꢃC. [
a
]
ꢁ19.3 (c 1.11 in CH2Cl2). nmax/cmꢁ1 3391, 2933, 1755,
D
1716, 1605, 1515, 1445, 1397, 1260 and 1056. dH (400 MHz, CDCl3)
1.44 (3H, d, J ¼ 6.0 Hz, CH3CHOCO), 2.19 (3H, s, CH3CO2), 2.52 (3H, s,
SCH3), 3.44 (1H, dd, J ¼ 1.2 and 5.2 Hz, CHCHO), 5.38 (1H, quintet,
J ¼ 6.0 Hz, CH3CHO), 6.18 (1H, d, J ¼ 16.0 Hz, CH]CHCO), 6.23 (1H,
d, J ¼ 1.2 Hz, CHOAc), 6.87 (1H, d, J ¼ 8.0 Hz, arom), 6.97 (1H, dd,
J ¼ 2.0 and 8.4 Hz, arom), 7.06 (1H, d, J ¼ 1.6 Hz, arom), 7.54 (1H, d,
J ¼ 16.0 Hz, CH]CHCO). dC (100 MHz, CDCl3) 18.2, 20.9, 22.7, 63.8,
65.9, 80.8, 114.1, 114.3, 115.4, 122.4, 126.9, 144.2, 146.1, 147.0, 166.5,
168.5, 170.1. HPLCeMS (ESI): Rt ¼ 5.6 min, m/z: 399 [M þ H2O]þ,
404 [M þ Na]þ, 785 [2M þ Na]þ.
4.2.6. (R)-1-((2R,3R)-2-Acetoxy-1-(methylthio)-4-oxoazetidin-3-
yl)ethyl3,4-dihydroxy benzoate (14)
To a solution of 16 (47 mg, 0.08 mmol) in anhydrous CH3CN
(1.64 mL) under inert atmosphere and at 0 ꢃC BF3$Et2O (23
mL,
0.182 mmol) was added dropwise. After 30 min the ice-bath was
removed, and after 45 min at rt the reaction was quenched with
a pH ¼ 6 phosphate buffer solution 0.1 M, extracted with CH2Cl2,
dried on Na2SO4 and evaporated. The crude was finally triturated to
afford the desired product 14 in 46%yield (13 mg) as a light yellow
oil. Found C, 50.96; H, 4.91; N, 3.96; S, 8.95%; C15H17NO7S requires C,
4.2.9. Methyl 4-(ethoxymethoxy)-3,5-dimethoxybenzoate (18)
A solution of methyl 4-hydroxy-3,5-dimethoxybenzoate (1.24 g,
5.9 mmol) in THF (10 mL) was added dropwise to a suspension of
NaH (60% in mineral oil, 306 mg, 7.6 mmol) in THF (20 mL) at 0 ꢃC
50.70; H, 4.82; N, 3.94; S, 9.02%. [
a
]
22 ꢁ13.7 (c 0.25 in CH2Cl2). nmax
/
under inert atmosphere. After 10 min a solution of chlor-
D
cmꢁ1 3371, 2983, 2929, 1781, 1761, 1713, 1602 and 1294. dH
(400 MHz, CDCl3) 1.47 (3H, d, J ¼ 6.0 Hz, CH3CHOCO), 2.18 (3H, s,
CH3CO2), 2.47 (3H, s, SCH3), 3.49 (1H, dd, J ¼ 1.2 and 4.8 Hz, CHCHO),
5.46 (1H, quintet, J ¼ 6.0 Hz, CH3CHO), 6.28 (1H, d, J ¼ 1.2 Hz,
CHOAc), 6.86 (1H, J ¼ 8.4 Hz, Ar), 7.44 (1H, dd, J ¼ 2.0 and 8.4 Hz,
Ar), 7.50 (1H, d, J ¼ 2.0 Hz, Ar). dC (100 MHz, CDCl3) 18.3, 20.9, 22.7,
64.0, 66.3, 80.8, 114.8, 116.5, 121.8, 123.5, 143.4, 149.2, 165.3, 168.2,
170.1. HPLCeMS (ESI): Rt ¼ 5.0 min, m/z: 373 [M þ H2O]þ, 378
[M þ Na]þ, 733 [2M þ Na]þ.
omethylethylether (1.1 mL, 11.75 mmol) in THF (7 mL) was added
dropwise and the reaction mixture was allowed to warm to room
temperature and monitored by TLC. After 2 h the reaction was
quenched with aqueous NH4Cl (30 mL) and extracted with EtOAc
(3 ꢄ 20 mL). The organic extracts were dried over Na2SO4 and
concentrated to obtain product 18 (1.46 g, 92%) as pale yellow solid.
Found C, 57.62; H, 6.80%; C13H18O6 requires C, 57.77; H, 6.71%; mp:
52.7e55.6. nmax/cmꢁ1 2926, 1720, 1592 and 1128. dH (200 MHz,
CDCl3) 1.21 (3H, t, J ¼ 7.0 Hz, CH3CH2), 3.88 (2H, q, J ¼ 7.0 Hz,
CH3CH2), 3.90 (6H, s, 2 ꢄ OCH3), 3.92 (3H, s, CO2CH3), 5.23 (2H, s,
OCH2O), 7.31 (2H, s, arom). dC (100 MHz, CDCl3) 14.6, 51.8, 55.8,
64.6, 96.2, 106.4, 125.3, 138.4, 152.8, 166.3. HPLCeMS (ESI):
Rt ¼ 7.81 min, m/z: 271 [M þ H]þ, 293 [M þ Na]þ, 309 [M þ K]þ.
4.2.7. (E)-(R)-1-((2R,3R)-2-Acetoxy-1-(methylthio)-4-oxoazetidin-
3-yl)ethyl-3-(3,4-bis(ethoxymethoxy)phenyl)acrylate (17)
To a solution of 4 (68 mg, 0.31 mmol) in CH2Cl2 (11.6 mL),
compound 21 (145 mg, 0.49 mmol) and DMAP (7.6 mg, 0.06 mmol)
were added. The mixture was then cooled to 0 ꢃC, DCC (101 mg,
0.49 mmol) was added and the system was allowed to reach rt in
15 min. After 65 h, the reaction was quenched with water, extracted
with CH2Cl2, dried on Na2SO4, evaporated and triturated with
AcOEt to separate dicycloexylurea precipitate. The solution was
then evaporated and purified by flash-chromatography (cyclo-
hexane/ethylacetate 90/10 to 80/20) to afford the product in 61%
yield (94 mg) as an oil. Found C, 55.78; H, 6.32; N, 2.68; S, 6.30%;
4.2.10. Methyl 4-(ethoxymethoxy)-3,5-dimethoxy benzoic acid (19)
A stirred solution of 18 (730 mg, 2.7 mmol) in 10 mL of a mixture
of THF/MeOH: 90/10 was treated with 5 M NaOH (2.7 mL). The
reaction mixture was heated at 40 ꢃC for 4 h and monitored by TLC.
At completion, EtOAc (10 mL) was added and the organic phase was
separated and discarded. The aqueous phase was then cooled to 0 ꢃC
and adjusted to pH 6 with aqueous NH4Cl. The mixture was extracted
with EtOAc (3 ꢄ 15 mL), dried on Na2SO4 and concentrated to afford
19 (432 mg, 63%) as a white solid. Found C, 56.18; H, 6.34%; C12H16O6
requires C, 56.24; H, 6.29%; mp: 107.7e109.4 ꢃC. nmax/cmꢁ1 3398,
2924,1716,1589 and 1120. dH (400 MHz, CDCl3) 1.22 (3H, t, J ¼ 7.2 Hz,
CH3CH2), 3.89 (2H, q, J ¼ 7.2 Hz, CH3CH2), 3.92 (6H, s, 2 ꢄ OCH3), 5.26
(2H, s, OCH2O), 7.37 (2H, s, arom). dC (50 MHz, CDCl3) 14.8, 56.0, 64.9,
96.3, 107.1, 124.5, 139.3, 153.0, 171.5. HPLCeMS (ESI): Rt ¼ 1.53 min,
m/z: 279 [M þ Na]þ, 535 [2M þ Na]þ.
C23H31NO9S requires C, 55.52; H, 6.28; N, 2.82; S, 6.44%. [
a
]
22 ꢁ7.4 (c
D
0.84 in CH2Cl2). nmax/cmꢁ1 3327, 2977, 2930, 1789, 1757, 1711, 1635,
1599, 1510, 1437, 1251 and 1159. dH (400 MHz, CDCl3) 1.21e1.28 (6H,
m, 2 ꢄ OCH2CH3), 1.44 (3H, d, J ¼ 6.4 Hz, CH3CHO), 2.17 (3H, s,
CH3CO2), 2.51 (3H, s, SCH3), 3.42 (1H, dd, J ¼ 1.6 and 6.0 Hz, CHCHO),
3.74e3.81 (4H, m, 2 ꢄ OCH2CH3), 5.30 (2H, s, OCH2O), 5.31 (2H, s,
OCH2O), 5.37 (1H, quintet, J ¼ 6.0 Hz, CH3CHO), 6.22 (1H, d,
J ¼ 1.6 Hz, CHOAc), 6.25 (1H, d, J ¼ 16.0 Hz, CH]CHCO), 7.13 (1H, dd,
J ¼ 2.0 and 8.0 Hz, arom), 7.19 (1H, d, J ¼ 8.0 Hz, arom), 7.37 (1H, d,
J ¼ 2.0 Hz, arom), 7.53 (1H, d, J ¼ 16.0 Hz, CH]CHCO). dC (100 MHz,
CDCl3) 15.0, 15.1, 18.3, 20.9, 22.7, 64.1, 64.5, 64.6, 66.1, 80.9, 93.8,
4.2.11. (E)-Methyl 3-(3,4-bis(ethoxymethoxy)phenyl)acrylate (20)
A solution of caffeic acid methyl ester [34] (540 mg, 2.78 mmol)
in THF (6 mL) was added dropwise into a solution of NaH

R. Cervellati et al. / European Journal of Medicinal Chemistry 60 (2013) 340e349
347
(60% dispersion in oil, 289 mg, 7.2 mmol) in THF (15 mL) under inert
tinhib; r.a.c. is expressed as
mM resorcinol equivalents [18]. Oscilla-
atmosphere and at 0 ꢃC. After 10 min a solution of chloromethyl
ethyl ether (1.01 mL, 11.1 mmol) in THF (7 mL) was added dropwise.
After 2 h the solution was allowed to rt and quenched by adding
a saturated NH4Cl solution, extracted with CH2Cl2, dried over
Na2SO4 and evaporated. The product 20 was obtained in 90% yield
(776 mg) after flash-chromatography (cyclohexane/ethylacetate
90/10) as a colourless oil. Found C, 62.12; H, 7.38%; C16H22O6
requires C, 61.92; H, 7.15%. nmax/cmꢁ1 2977, 2900, 1717, 1634, 1600,
1582, 1511, 1436, 1393, and 1105. dH (400 MHz, CDCl3) 1.21e1.27
(6H, m, 2 ꢄ OCH2CH3), 3.74e3.81 (4H, m, 2 ꢄ OCH2CH3), 3.81
(3H, s, OCH3), 5.30 (1H, s, OCH2O), 5.31 (1H, s, OCH2O), 6.33 (1H, d,
J ¼ 16.0 Hz, CH]CHCO), 7.15 (1H, dd, J ¼ 2.0 and 8.0 Hz, arom), 7.19
(1H, d, J ¼ 8.4 Hz, arom), 7.39 (1H d, J ¼ 1.6 Hz, arom), 7.63 (1H, d,
tory behaviours V(Pt) were followed potentiometrically by using
the couple bright platinum electrode (Hamilton 238945) e Ag/AgCl
double junction electrode (Ingold 373-90-WTE-ISE-S7) connected
to a multimeter (WTW pH 540 GLP), controlled by a PC. Reacting
mixtures were thermostated at 25.0 ꢂ 0.1 ꢃC. The straight line of
the standard was checked before each series of measurements.
More details on the experimental procedure are reported in
Refs. [18,19,35].
4.3.2. TEAC assay [20a]
The radical cation is preformed by reaction between ABTS and
K2S2O8 in PBS medium, pH ¼ 7.4. The mixture was kept in the dark
until the reaction was complete and absorbance at 734 nm stable.
Spectroscopic measurements were performed in triplicate at four
J ¼ 16.0 Hz, CH]CHCO). dC (100 MHz, CDCl3)
d 14.9, 15.0, 51.4, 64.3,
64.4, 93.6, 94.0, 115.5, 115.9, 116.0, 123.1, 128.5, 144.4, 147.4, 149.2,
167.4. HPLCeMS (ESI): Rt ¼ 9.2 min, m/z: 311 [M þ H]þ, 333
[M þ Na]þ, 349 [M þ K]þ.
concentrations of the
of the standard (Trolox) were treated in the same way. In brief,
3.0 mL of diluted ABTSꢀþ solution and 30
L of sample were mixed
b-lactams in DMSO. Suitably diluted solutions
m
in a photometric cuvette and absorbance was measured at 734 nm
at exactly 6 min after the mixing of the reagents (T ¼ 30.0 ꢂ 0.1 ꢃC).
A blank with DMSO was measured in the same way. The difference
4.2.12. (E)-3-(3,4-Bis(ethoxymethoxy)phenyl)acrylic acid (21)
A stirred solution of 20 (310 mg, 1 mmol) in 3.8 mL of a mixture
of THF/MeOH 90/10 was treated with 1 mL NaOH 5 M, warmed at
40 ꢃC and stirred for 4 h. After completion, AcOEt (6 mL) was added
and discharged. The aqueous phase was treated with HCl (1 M),
extracted with AcOEt, the organic phase was dried over Na2SO4 and
evaporated to afford the desired product in 96% yield (284 mg) as
a white solid. Found C, 60.52; H, 6.95%; C15H20O6 requires C, 60.80;
H, 6.80%; mp: 101e103 ꢃC. nmax/cmꢁ1 2976, 1683, 1626, 1596, 1514,
between the absorbance of the blank and the sample gave
DE6
(E6blank ꢁ E6sample ¼
D
E6). Data ( E6 vs conc. (mM)) are well
D
fitted by straight lines through the origin (following the Lamberte
Beer law). Then, the relative antioxidant activity with respect to
Trolox (TEAC) was obtained by the ratio:
TEAC ¼ mðsmpÞ=mðTroloxÞ
1418, 1244 and 1114. dH (400 MHz, CDCl3)
d 1.22e1.29 (6H, m,
2 ꢄ OCH2CH3), 3.75e3.85 (4H, m, 2 ꢄ CH3CH2O), 5.31 (1H, s,
OCH2O), 5.33 (1H, s, OCH2O), 6.35 (1H, d, J ¼ 16.0 Hz, CH]CHCO),
7.18 (1H, dd, J ¼ 1.6 and 8.8 Hz, 1H, arom), 7.22 (1H, d, J ¼ 8.8 Hz,
arom), 7.42 (1H, d, J ¼ 1.6 Hz, arom), 7.73 (1H, d, J ¼ 16.0 Hz, CH]
CHCO). dC (50 MHz, CDCl3) 15.0, 15.1, 64.5, 64.6, 93.8, 94.2, 115.5,
115.8, 116.0, 123.7, 128.3, 146.8, 147.5, 149.7, 172.1. HPLCeMS (ESI):
Rt ¼ 1.90 min, m/z: 297 [M þ Na]þ, 319 [2M þ Na]þ, 335 [M þ K]þ.
where m(smp) and m(Trolox) are the slopes of the straight lines of
the sample and the standard respectively. Standard error of this
ratio is calculated in the usual way.
Here also the straight line of the standard was always checked
before each series of measurements.
4.3.3. DPPH assay
The principle of this method is the decolorization of the stable
radical 2,2-diphenyl-1-picrylhydrazyl (DPPHꢀ) by antioxidants
[21,22]. The DPPHꢀ is intensely purple coloured (lmax ¼ 512e
530 nm) due to the presence of the picric group (1,3,5-
nitrobenzene) and is stabilized by either the steric hindrance of
the aromatic substituents and the delocalization of the unpaired
electron on the whole structure. The disappearance of this radical
due to the addition of an antioxidant H-donor, may be evaluated
spectrophotometrically, at its absorption maximum in methanol
(515 nm). The measured decrease in absorbance at 515 nm is an
expression of the antioxidant capacity of a sample [36]. The extent
of the decolorization is a function of concentration and time. Trolox
was used as the standard. In practice 2.9 mL of DPPHꢀ methanolic
solution was placed in a cuvette and the initial absorbance, A0, was
4.2.13. 3,4-Bis((tert-butyldimethylsilyl)oxy)benzoic acid (22)
To
a solution of 3,4-dihydroxy-benzoic acid (500 mg,
3.24 mmol) in DMF (5.8 mL) under inert atmosphere, imidazole
(1.983 g, 29.16 mmol) and then TBSCl (2.200 g, 14.6 mmol) were
added. After 66 h the reaction was quenched at 0 ꢃC by adding
water, and then extracted with Et2O (10 ꢄ 4 mL). The organic
phases were washed twice with water and brine, dried over Na2SO4
and evaporated. To the persilylated intermediate (2.049 g,
4.12 mmol) in a 6 mL MeOH/THF 40/60, a solution of K2CO3
(231 mg, 1.67 mmol) in water (2.2 mL) was added. After 12 h the
mixture was quenched a 0 ꢃC by adding a saturated solution of citric
acid and extracted with Et2O. The organic phases were washed with
water and brine, dried over Na2SO4 and evaporated. The product
was finally obtained by flash-chromatography (CH2Cl2/MeCN: 90/
10) in 75% yield (930 mg) as a white solid. Found C, 60.12; H, 9.10%;
C19H34O4Si2 requires C, 59.64; H, 8.96%; mp: 152e154 ꢃC. dH
measured. Then 100 mL was added and mixed; after 15 min the
final absorbance, Af, was measured. The percentage of inhibition
was calculated according to the following equation (%)
Inhibition ¼ (1 ꢁ Af/A0) ꢄ 100. The same procedure was repeated
with the standard. Data (%inhib vs conc. (mM)) are well fitted by
straight lines through the origin (following the LamberteBeer law).
Then, the relative antioxidant activity with respect to Trolox was
obtained by the ratio: DPPH ¼ m(smp)/m(Trolox).
(400 MHz, CDCl3)
d 0.24 (6H, s, SitBuMe2), 0.25 (6H, s, SitBuMe2),
1.00 (9H, s, SitBuMe2), 1.01 (9H, s, SitBuMe2), 6.88 (1H, d, J ¼ 8.0 Hz,
arom), 7.60 (1H, d, J ¼ 2.0 Hz, arom), 7.63 (1H, dd, J ¼ 1.6 and 8.8 Hz,
1H, arom). dC (50 MHz, CDCl3) ꢁ4.1, ꢁ4.0, 18.4, 18.5, 25.8, 25.9,
120.5, 122.4, 122.7, 124.4, 146.8, 152.5, 171.2.
4.3. Antioxidant activity methods
4.3.4. FRAP assay [23]
2,4,6-Tris(2-pyridyl)-s-triazine-Fe3þ (FRAP reagent) was
prepared by mixing suitable amounts of TPTZ and FeCl3 solutions in
4.3.1. BR assay
Relative antioxidant activity (r.a.c.) with respect to a substance
chosen as standard, resorcinol (Re) in our case, is determined on the
basis of concentrations of sample and resorcinol that give the same
aqueous acetate buffer. In brief, 3.0 mL of FRAP reagent and 100 mL
of suitably diluted sample were mixed in a photometric cuvette and
absorbance was measured after exactly 4 min after the mixing

348
R. Cervellati et al. / European Journal of Medicinal Chemistry 60 (2013) 340e349
(T ¼ 25.0 ꢂ 0.1ꢃ). The absorbance of the blue Fe2þ complex is
measured at 593 nm (Shimadzu UV-1601 PC spectrophotometer).
Four or five different sample concentrations were tested and the
straight line Abs vs. conc. was then compared with that of the
standard (FeSO4). The LamberteBeer law is followed, then the ratio:
FRAP ¼ m(smp)/m(Fe2þ) gives the relative ferric reducing activity
(mmol equiv Fe2þ).
# 11/2011 with the contribution of FC delegation of Catania, “Il
sorriso di Jenny”, and Associazione Trentina FC.
Appendix A. Supplementary data
Supplementary data associated with this article can be found in
the online version, at http://dx.doi.org/10.1016/j.ejmech.2012.12.024.
4.4. Cyclic voltammetry
References
All electrochemical experiments were conducted in acetonitrile
from SigmaeAldrich (>99.5% over molecular sieves) after careful
deaeration of the reaction mixture with argon; in all instances
a silver wire was used as a pseudo-reference electrode, while
a platinum coil was used as a counter electrode. Voltammetric
analysis was performed with platinum, gold and glassy carbon
working electrodes (Ø ¼ 2 mm) using a function generator AMEL
mod 568 equipped with a digital oscilloscope.
[1] (a) ESAC
e European Surveillance of Antimicrobial Consumption, Final
Management Report 2009e2010.
(b) ECOC e European Centre for Disease Prevention and Control, Annual
Epidemiological Report on Communicable Diseases in Europe 2010. http://
www.ecdc.europa.eu.
[2] C. Walsh, Antibiotics: Actions, Origins, Resistance, American Society for
Microbiology (ASM) Press, Washington, DC, 2003.
[3] (a) L. Diaz Högberg, A. Heddini, O. Cars, Trends Pharmacol. Sci. 31 (2010)
509e515;
(b) H.W. Boucher, G.H. Talbot, J.S. Bradley, J.E. Edwards, D. Gilbert, L.B. Rice,
M. Scheld, B. Spellberg, J. Bartlett, Clin. Infect. Dis. 48 (2009) 1e12;
(c) EMEA: European Medicines Agency, The Bacterial Challenge: Time to
React, Technical Report EMEA/576176/2009, London, 2009. http://www.
ema.europa.eu.
4.5. Bacterial strains selection and antibacterial activity evaluation
4.5.1. Bacterial isolates
[4] (a) S. Razvi, L. Quittell, A. Sewall, H. Quinton, B. Marshall, L. Saiman, Chest 136
(2009) 1554e1560;
(b) A.L. Prunier, B. Malbruny, M. Laurans, J. Brouard, J.F. Duhamelo, R. Leclerc,
J. Infect. Dis. 187 (2003) 1709e1716.
[5] (a) G. Folkerts, J. Kloek, R.B. Muijsers, F.P. Nijkamp, Eur. J. Pharmacol. 429
(2001) 251e262;
A collection of 45 recently isolated and well characterized for
the antimicrobial susceptibility phenotype MRSA strains recovered
from CF patients was tested. All the strains were isolated at the
Department of Pediatrics, CF Center, Fondazione IRCCS Cà Granda
Ospedale Maggiore Policlinico, University of Milano.
(b) M.C. Martınez, R. Andriantsitohaina, Antioxid. Redox Signal. 11 (2009)
669e702.
[6] M. Rottner, S. Tual-Chalot, H.A. Mostefai, R. Andriantsitohaina, J.M. Freyssinet,
4.5.2. Antimicrobial susceptibility determinations
M.C. Martinez, PLoS One
journal.pone.0024880.
6 (2011) e24880. http://dx.doi.org/10.1371/
The in vitro susceptibility of methicillin-resistant isolates of
S. aureus to new antimicrobial compounds was evaluated by means
of the broth microdilution method for the determination of the
minimum inhibitory concentrations (MIC), in accordance with the
Clinical and Laboratory Standards Institute (CLSI; formerly National
Committee for Clinical Laboratory Standards) guidelines [37].
Briefly, serial two-fold dilutions were made of the compounds in
96-well plates in order to obtain concentrations ranging from of
0.06e128 mg/L in cation-adjusted MullereHinton broth (DIFCO).
An equal volume of 1 ꢄ 106 CFU/mL (Colony Forming Unit/mL)
bacterial inoculum was added to each well of the microtitre plate
containing 0.05 mL of the serial antibiotic concentrations. The
microtitre plate was then incubated overnight at 37 ꢃC and subse-
quently analysed for the presence of visible bacterial growth. MIC
was defined as the lowest concentration of the tested compound
able to inhibit visible growth of the microorganism after overnight
incubation. All the tested compounds were solubilized in 100%
DMSO and appropriate dilutions were set up to prevent its toxicity.
Positive strain controls without antimicrobial compounds, controls
with DMSO and uninoculated media were run parallel to the tested
compounds under the same conditions.
[7] P.Ø. Jensen, J. Lykkesfeldt, T. Bjarnsholt, H.P. Hougen, N. Høiby, O. Ciofu, Basic
Clin. Pharmacol. Toxicol. 110 (2012) 353e358.
[8] F. Broccolo, G. Cainelli, G. Caltabiano, C.E.A. Cocuzza, C.G. Fortuna, P. Galletti,
D. Giacomini, G. Musumarra, R. Musumeci, A. Quintavalla, J. Med. Chem. 49
(2006) 2804e2811.
[9] G. Cainelli, C. Angeloni, R. Cervellati, P. Galletti, D. Giacomini, S. Hrelia,
R. Sinisi, Chem. Biodivers. 5 (2008) 811e829.
[10] P. Galletti, C.E.A. Cocuzza, M. Pori, A. Quintavalla, R. Musumeci, D. Giacomini,
ChemMedChem 6 (2011) 1919e1927.
[11] For a recent review see for instance: P. Galletti, D. Giacomini Curr. Med. Chem.
18 (2011) 4265e4283.
[12] (a) C. Walsh, Nature 406 (2000) 775e781;
(b) C.M. Cimarusti, R.B. Sykes, Med. Res. Rev. 4 (1984) 1e24.
[13] (a) E. Turos, M.I. Konaklieva, R.X.F. Ren, H.C. Shi, J. Gonzalez, S. Dickey,
D.V. Lim, Tetrahedron 56 (2000) 5571e5578;
(b) E. Turos, T.E. Long, M.I. Konaklieva, C. Coates, J.Y. Shim, S. Dickey, D.V. Lim,
A. Cannons, Bioorg. Med. Chem. Lett. 12 (2002) 2229e2231.
[14] K.D. Revell, B. Heldreth, T.E. Long, S. Jang, E. Turos, Bioorg. Med. Chem. 15
(2007) 2453e2467.
[15] P. Galletti, A. Quintavalla, C. Ventrici, G. Giannini, W. Cabri, S. Penco, G. Gallo,
S. Vincenti, D. Giacomini, ChemMedChem 4 (2009) 1991e2001.
[16] G. Cainelli, P. Galletti, S. Garbisa, D. Giacomini, L. Sartor, A. Quintavalla, Bioorg.
Med. Chem. 11 (2003) 5391e5399.
[17] K. Schlesier, M. Harwat, V. Böhm, R. Bitsch, Free Radical Res. 36 (2002) 177e187.
[18] R. Cervellati, K. Höner, S.D. Furrow, C. Neddens, S. Costa, Helv. Chim. Acta 84
(2001) 3533e3547.
[19] R. Cervellati, K. Höner, S.D. Furrow, F. Mazzanti, F.S. Costa, Helv. Chim. Acta 87
S. aureus American Type Culture Collection (ATCC) 29213
(MSSA) and ATCC 43300 (MRSA) were used as control strains.
Cefotaxime sodium salt was used as the reference antibiotic for MIC
experiments validation. Breakpoints for resistance were those
recommended by the CLSI [38].
(2004) 133e155.
[20] (a) R. Re, N. Pellegrini, A. Proteggente, A. Pannala, M. Yang, C. Rice-Evans, Free
Radic. Biol. Med. 26 (1999) 1231e1237;
(b) A. Prakhash, F. Rigelhof, E. Miller, Antioxidant Activity. www.medlabs.
com/Downloads/Antiox_acti_.pdf.
[21] M.S. Blois, Nature 181 (1958) 1199e1200.
[22] W. Brand-Williams, M.E. Cuvelier, C. Berset, Lebensm. Wiss. Technol/Food Sci.
Technol. 28 (1995) 25e30.
Acknowledgements
[23] I.F.F. Benzie, J.J. Strain, Anal. Biochem. 239 (1996) 70e76.
[24] V.L. Singleton, J.A. Rossi, Am. J. Enol. Vitic. 16 (1965) 144e148.
[25] C.A. Rice-Evans, N.J. Miller, G. Paganga, Free Radic. Biol. Med. 20 (1996) 933e956.
[26] J.S. Wright, E.R. Johnson, G.A. Di Labio, J. Am. Chem. Soc. 123 (2001) 1173e1183.
[27] M.B. Smith, J. March, March’s Advanced Organic Chemistry, Wiley Inter-
science, New Jersey, 2007, p. 1780, ch. 19.
The authors would like to thank on clinical isolates furnished by
Dr. Laura Garlaschi and Dr. Lisa Cariani at the Department of
Pediatrics, CF Center, University of Milan. We would like to thank
Mrs. Elisa Bellicchi, Francesca Lattanzio, and Francesca Casoni for
technical assistance. This work was supported by MIUR, University
of Bologna, and by the Italian Cystic Fibrosis Foundation, grant FFC
[28] A.J. Bard, L.R. Faulkner, Electrochemical Methods: Fundamentals and Appli-
cations, second ed., Wiley, 2001, p. 598.
ꢀ
ꢀ
[29] Milardovic, D. Ivekovic, V. Rumenjak, B.S. Grabaric, Electroanalysis 17 (2005)
1847e1853.

R. Cervellati et al. / European Journal of Medicinal Chemistry 60 (2013) 340e349
349
[30] K.M. Schaich, Cracking the code on antioxidant testing-sorting your ORAC
from your FRAP, in: Nutraingredients Antioxidant Conference, Brussels, June
30, 2010, slides 21.
[31] H. Lund, K. Daasbjerg, T. Lund, S.U. Pedersen, Acc. Chem. Res. 28 (1995) 313e319.
[32] (a) T. Holm, J. Am. Chem. Soc. 115 (1993) 916e918;
(b) T. Holm, Acta Chem. Scand., Ser. B 42 (1998) 685e689;
(c) T. Lund, D. Ohlrich, P. Borling, Acta Chem. Scand. 53 (1999) 932e937.
[33] G. Cainelli, D. Giacomini, P. Galletti, A. Quintavalla, Eur. J. Org. Chem. (2003)
1765e1774.
[35] R. Cervellati, C. Renzulli, M.C. Guerra, E. Speroni, J. Agric. Food Chem. 50
(2002) 7504e7509.
[36] P. Molyneux, Songklanakarin J. Sci. Technol. 26 (2004) 211e219.
[37] [CLSI] Clinical and Laboratory Standards Institute, Methods for Dilution
Antimicrobial Susceptibility Tests for Bacteria that Grow Aerobically.
Approved StandarddEighth Edition, CLSI document M07-A8, Clinical and
Laboratory Standards Institute, Wayne, PA, 2008.
[38] [CLSI] Clinical and Laboratory Standards Institute, Performance Standards for
Antimicrobial Susceptibility Testing: 22nd Informational Supplement. CLSI
document M100-S22, Clinical and Laboratory Standards Institute, Wayne, PA,
2012.
ꢁ
ꢁ
[34] P. Dzubák, M. Hajdúch, R. Gazák, A. Svobodová, J. Psotová, D. Walterová,
ꢁ
P. Sedmera, V. Kren, Bioorg. Med. Chem. 14 (2006) 3793e3810.