New 4-[(5-arylidene-2-arylimino-4-oxo-3-thiazolidinyl)methyl]benzoic acids active as protein tyrosine phosphatase inhibitors endowed with insulinomimetic effect on mouse C2C12 skeletal muscle cells

Article abstract of DOI:10.1016/j.ejmech.2012.02.012

In pursuing our research targeting the identification of potent inhibitors of PTP1B and LMW-PTP, we have identified new 4-[(5-arylidene-2-arylimino-4-oxo- 3-thiazolidinyl)methyl]benzoic acids endowed with interesting in vitro inhibitory profiles. Most com

Full text of DOI:10.1016/j.ejmech.2012.02.012

European Journal of Medicinal Chemistry 50 (2012) 332e343
Contents lists available at SciVerse ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
New 4-[(5-arylidene-2-arylimino-4-oxo-3-thiazolidinyl)methyl]benzoic acids
active as protein tyrosine phosphatase inhibitors endowed with insulinomimetic
effect on mouse C2C12 skeletal muscle cells
,
a
a
b
c
Rosaria Ottanà a , Rosanna Maccari , Simona Amuso , Gerhard Wolber , Daniela Schuster ,
*
Sonja Herdlinger c, Giampaolo Manao d, Guido Camici d 1, Paolo Paoli d
,
a Dipartimento Farmaco-chimico, University of Messina, Polo Universitario dell’Annunziata, Viale SS. Annunziata, 98168 Messina, Italy
b Freie Universität Berlin, Institut für Pharmazie, Königin-Luisestr., 2þ4, 14195 Berlin, Germany
c University of Innsbruck, Instiute of Pharmacy and Centre for Molecular Biosciences, Innrain 52, 6020 Innsbruck, Austria
d Dipartimento di Scienze Biochimiche, University of Firenze, Vl. Morgagni 50, 50134 Firenze, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
In pursuing our research targeting the identification of potent inhibitors of PTP1B and LMW-PTP, we have
identified new 4-[(5-arylidene-2-arylimino-4-oxo-3-thiazolidinyl)methyl]benzoic acids endowed with
interesting in vitro inhibitory profiles. Most compounds proved to be inhibitors of PTP1B and LMW-PTP
isoform IF1. The tested inhibitors also showed selectivity towards PTP1B over the closely related TC-PTP.
These compounds were found to activate the insulin-mediated signalling on mouse C2C12 skeletal
muscle cells by increasing the phosphorylation levels of the insulin receptor and promoting cellular 2-
deoxyglucose uptake.
Received 2 October 2011
Received in revised form
5 February 2012
Accepted 6 February 2012
Available online 21 February 2012
Keywords:
Interestingly, 4-{[5-(4-benzyloxybenzylidene)-2-(4-trifluoromethylphenylimino)-4-oxo-3-thiazolidinyl]
methyl}benzoic acid (7d), the best in vitro inhibitor of PTP1B and the isoform IF1 of LMW-PTP, provided the
highest activation level of the insulin receptor and was found to be endowed with an excellent insulino-
mimetic effect on the selected cells. This compound therefore represents an interesting lead compound for
developing novel PTP1B and LMW-PTP inhibitors which could be achieved by improving both its pharma-
cological profile and its potentiating effects on insulin signalling.
Enzyme inhibitor
Diabetes mellitus
Obesity
Protein tyrosine phosphatase
Insulinomimetic agent
Molecular docking
Ó 2012 Elsevier Masson SAS. All rights reserved.
1. Introduction
defective or inappropriate changes in the activity of either PTKs or
PTPs and/or their expression are critically involved in a variety of
Protein tyrosine phosphatases are a superfamily of enzymes
involved in crucial cellular signalling mechanisms by controlling
the phosphorylation levels of specific tyrosine residues in peptides
and proteins which regulate many cellular functions, such as cell
proliferation, survival, metabolism, adhesion and migration [1e4].
The balanced and opposing actions of protein tyrosine kinases
(PTKs) and protein tyrosine phosphatases (PTPs) regulate the phys-
iological tyrosine phosphorylation levels of proteins. Therefore,
human diseases including diabetes, obesity and cancer [5e8].
In the past, substantial efforts were made to elucidate the
physiopathological role of PTKs which led to the development of
drugs targeted at PTKs, such as antireceptor antibodies or small
inhibitors, which are currently available in therapeutic regimes for
the treatment of cancer [9]. The possible involvement of PTPs in the
development of common and serious diseases was understood
much later and recently both academic and industrial research
groups have shown significant interest in PTPs as potential drug
targets [8,10e13].
Among the PTPs so far identified, the cytosolic protein tyrosine
phosphatase 1B (PTP1B) was the first phosphatase to be isolated
and is therefore the most widely studied enzyme of the family. The
role of PTP1B in diabetes and obesity is well-known [5,6,8,13e16].
A large body of experimental data acquired from in vivo studies
on humans and PTP1B-knockout mice has identified PTP1B as
a major negative regulator of both insulin and leptin signals
[14,15,17e19]. Insulin-mediated activation of the insulin receptor
Abbreviations: DM2, type
2 diabetes mellitus; 2-DOG, 2-deoxyglucose; IR,
insulin receptor; IRS, insulin receptor substrate; JAK, Janus kinase; LMW-PTP, low
molecular weight-protein tyrosine phosphatase; PTK, protein tyrosine kinase;
PTP1B, protein tyrosine phosphatase 1B; STAT, signal transducer and activator of
transcription; TC-PTP, T-cell protein tyrosine phosphatase.
* Corresponding author. Tel.: þ39 90 6766408; fax: þ39 90 6766402.
E-mail address: rottana@unime.it (R. Ottanà).
1
Center of Excellence for Scientific Research DENOTHE, Viale Pieraccini 6, 50139
Firenze, Italy.
0223-5234/$ e see front matter Ó 2012 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2012.02.012

R. Ottanà et al. / European Journal of Medicinal Chemistry 50 (2012) 332e343
333
(IR), via the autophosphorylation of specific tyrosine residues of the
cytoplasmatic portion of the transmembrane IR subunits, triggers
COOH
Ph
+
NH
b
O
a cascade of events involved in the regulation of numerous meta-
bolic functions. PTP1B controls the insulin signal through the
dephosphorylation of phosphotyrosine (pTyr) residues of not only
activated IR but also different downstream substrate proteins such
as IRS-1 [5,6,20,21].
It has been shown that the pathological overexpression of PTP1B
is involved in the insulin resistance of tissues sensitive to this
hormone, including skeletal muscle and liver, thus contributing to
the development of type 2 diabetes mellitus (DM2), a cluster of
disorders characterised mainly by high blood levels of glucose due
to defects in both insulin action and secretion.
The silencing of the PTP1B gene in mice as well as the lack of the
enzyme in PTP1B-knockout mice result in increasing insulin
sensitivity and resistance to weight gain on a high-fat diet without
causing any abnormality in the animals [14,15,22,23]. The admin-
istration of PTP1B antisense oligonucleotides to ob/ob and db/db
mice positively modulates downstream events in insulin signalling
pathways in targeted tissues [22]. In humans, they increase insulin
sensitivity without inducing hypoglycemia [17].
-
+
OSO
Na
+
NH
NH
+
S
Br
trodusquemine
ertiprotafib
X
S
COOH
N
O
Ar
1
2
X = O
X = NPh
Ar
Obesity is an endocrine disorder characterized by leptin resis-
tance. Leptin is a circulating hormone produced primarily by
adipose tissue, and it is responsible for energy homeostasis through
the control of food intake and energy consumption. Compelling
evidence indicates that PTP1B is involved in the regulation of leptin
signalling by dephosphorylating the leptin receptor-associated
Janus kinase 2 (JAK2). In vivo studies on leptin-deficient mice
lacking PTP1B have shown moderate weight gain with normal food
intake [5,6,18,19,24,25].
OPh
OPh
OCH Ph
2a
2c
2d
2e
2b
OCH Ph
Obesity is also closely linked to the onset of insulin resistance in
peripheral tissues. This condition is mediated by adipose-derived
cytokines and hormones (among them leptin) that influence
insulin action in liver, fat and skeletal muscle [26,27].
Fig. 1. Chemical structures of some PTP1B inhibitors and selected compounds 2aee
[55].
Considering the alarming rate and rapid spread of DM2 and
obesity in industrialized countries [28], the identification of potent
PTP1B inhibitors as drug candidates represents an attractive novel
approach to the treatment of these pathologies. Numerous efforts
have been thwarted, mainly due to poor selectivity and unfavourable
pharmacokinetic profiles of candidates [13,25,29e31]. Of the potent
PTP1B inhibitors developed so far, ertiprotafib (Fig. 1) was the first
compound to progress to clinical trials for DM2 treatment and the
only drug candidate to reach phase 2 clinical trials. However, dose-
related adverse effects led to its withdrawal from trials [32].
Currently, trodusquemine (Fig. 1), a selective PTP1B inhibitor, has
reached phase 1 clinical trials and shows interesting preclinical
results as antidiabetic agent and appetite suppressant [33,34].
Low molecular weight PTP (LMW-PTP) has also been recognized
to negatively control insulin signalling by blocking the cascade of
events following insulin-mediated receptor activation since it
dephosphorylates pTyr specific residues of IR [5,35e38]. LMW-PTP
and PTP1B share the same phosphate binding loop structure,
although the two enzymes differ as regards tissue specificity. In
fact, LMW-PTP knockdown has been found to improve the insulin
response in liver and adipose tissue, while PTP1B exerts significant
control over the metabolic signalling pathway, especially in liver
and skeletal muscle tissue. Thus, the search for dual inhibitors of
PTP1B and LMW-PTP could be considered an intriguing challenge
for drug discovery destined to treat DM2.
mice die by 3e5 weeks of birth as a result of severe anaemia,
lymphocytopenia and several inflammatory disorders, while TC-
PTPþ/ꢀ heterozygous mice have normal life expectancy and unaf-
fected immune system [41].
Studies performed with recombinant TC-PTP catalytic domain
indicate that the enzyme can dephosphorylate critical pTyr units of
IR similarly to PTP1B [42]. Nevertheless, several studies indicate
that both enzymes are not redundant PTPs but regulate the IR
activation as well as the insulin signalling in a coordinated manner
[42,43]. Although TC-PTP and PTP1B possess high similarity in their
structure, they show an high degree of substrate selectivity as well
as tissue specificity. In vivo studies carried out on liver-specific TC-
PTPþ/ꢀ heterozygous mice underline that the enzyme can reduce
both hepatic gluconeogenesis and fasting blood glucose by regu-
lating STAT-3-mediated gluconeogenic gene expression [44].
In obesity, high hypothalamic levels of TC-PTP contribute to
leptin resistance condition that is triggered by aberrations in
cellular and molecular processes that attenuate the leptin signal-
ling [45].
Taking into account the lethality associated with TC-PTP null
mice, its partial inhibition could represent an useful strategy to
reduce gluconeogenesis and fasting hyperglicaemia in the control
of insulin resistance and obesity [44,46].
The crystal structure of PTP1B provides a valid support for the
structure-based design of potent and selective inhibitors [5,16]. The
presence of a single preserved catalytic domain characterized by
a unique signature motif typifies the PTP family. It represents the
phosphate recognition site which accommodates the double
negatively charged pTyr moiety of the substrate at physiological pH
values. Consequently, in order to mimic the substrate-active-site-
directed interactions, most PTP1B competitive inhibitors possess
Among PTPs involved in the regulation of IR, the T-cell protein
tyrosine phosphatase (TC-PTP) is the PTP that shares with PTP1B
the highest sequence identity in catalytic domain (74%) and
substantial similarity in the active site [39,40]. TC-PTP is ubiqui-
tously expressed, although is present in higher amount in haema-
topoietic cells. Gene-targeting studies indicate that TC-PTP null

334
R. Ottanà et al. / European Journal of Medicinal Chemistry 50 (2012) 332e343
a high charge density which generally leads to a lack of acceptable
cell permeability. The challenge posed for research of effective
inhibitors of PTP1B is the identification of small molecules
endowed with appropriate physicochemical properties and in vivo
efficacy.
PTP1B selectivity represents yet another challenge in the search
for potent and safe drug-like molecules for the treatment of dia-
betes and obesity [29,31,47]. Although the catalytic site is widely
conserved throughout the PTP family, it is surrounded by residues
that contribute significantly to the recognition of the substrate and
these residues are unique for each PTP. A secondary aryl phosphate
binding site demarcated by Arg24 and Arg254 has been identified
in a PTP1B subpocket adjacent to the active site and represents
a good target in order to obtain more selective and potent bidentate
inhibitors [5,48e50].
The structure of LMW-PTP is less complex than that of PTP1B.
The conserved catalytic pocket is surrounded by the so-called
variable loop which contains amino acid residues that determine
the selectivity of the two active isoforms of the human enzyme, IF1
and IF2, towards both substrates and inhibitors. The opposite edge
of the catalytic cavity is formed by a mobile loop containing the
catalytic Asp129 and the residues Tyr131 and Tyr132 that are
involved in substrate recognition [51,52].
phosphate binding site. Moreover, these studies suggest that the
introduction of an additional aromatic ring on position 2 of thia-
zolidinone moiety may be useful not only to increase lipophilicity,
potentially improving bioavailability, but also to enhance binding to
Phe182 of the WPD loop situated in proximity to the PTP1B catalytic
site. In the catalytically active enzyme, the WPD loop is in an open
conformation that, upon the binding of the ligand, undergoes
a conformational modification in which Phe182 folds onto the
substrate thereby favouring pTyr hydrolysis. In fact, the insertion of
a 2-phenylimino moiety in the 4-thiazolidinone scaffold provides
4-[(5-arylidene-2-phenylimino-4-oxo-3-thiazolidinyl)methyl]ben-
zoic acids (2, Fig. 1) which have proven to be interesting inhibitors
of both PTP1B and the IF1 isoform of LMW-PTP as well as being
generally more active than the 2-oxo analogues 1 [54,55].
The SARs so far acquired for this class of compounds clearly
indicate that the 5-arylidene moiety plays an important role in the
interaction with the target enzymes and, in particular, with the
secondary aryl phosphate binding site of PTP1B. An additional
aromatic ring in the para or meta position of the 5-benzylidene
moiety gives the best enzyme inhibition levels. Phenoxy-
benzylidene, benzyloxybenzylidene and 2-naphthyl groups in
position 5 of the 4-thiazolidinone scaffold are beneficial to the
affinity for both PTP1B and LMW-PTP.
The search for PTP1B and LMW-PTP inhibitors has been pursued
in our laboratory for some years and 4-[(5-arylidene-2,4-dioxo-3-
thiazolidinyl)methyl]benzoic acids (1, Fig. 1) have been identified
as a scaffold of effective bidentate inhibitors of both PTP1B and
LMW-PTP. Interesting results have been achieved obtaining potent
PTP inhibitors endowed with IC50 values at submicromolar levels.
They have been shown to act as competitive and reversible PTP
inhibitors with significant selectivity for PTP1B and the IF1 isoform
of LMW-PTP [53,54].
Molecular docking into the catalytic site of PTP1B indicates that
p-methylbenzoic acid residue can interact with the catalytically
essential Cys215 of the active site acting as a pTyr-mimetic group,
while the 5-arylidene moiety binds to the secondary aryl
On the basis of the SARs so far acquired, further optimization of
compounds
2 prompted the design and synthesis of 4-[(5-
arylidene-2-arylimino-4-oxo-3-thiazolidinyl)methyl]benzoic acids
7 and 8 (Scheme 1) presented herein in order to ascertain the effect
produced by the introduction of substituents on the para position of
the 2-phenylimino moiety which were selected on the basis of their
different electronic and lipophilic nature. As regards the 5-
arylidene moiety of compounds 7 and 8, we inserted the substit-
uents that proved to be beneficial for the enzyme/inhibitor inter-
action (Scheme 1).
The inhibitory activity of compounds 7 and 8 was investigated
against human PTP1B, the two active isoforms of human LMW-PTP
(IF1 and IF2) and TC-PTP. Moreover, the effects of the newly
S
a
COOH
R
NCS
+
COOH
R
N
H
N
H
H2N
3, 4
b
R
N
S
COOH
R
N
S
COOH
c
N
N
O
O
Ar
5, 6
7a-e
8a-e
OPh
OCH2Ph
OCH2Ph
Ar =
c
d
e
a
3, 5, 7 R = CF3
4, 6, 8 R = OCH3
b
OPh
Scheme 1. Reagents and conditions: (a) EtOH, Δ; (b) ClCH2COCl, Et3N, EtOH, Δ; (c) ArCHO, C5H11N, EtOH, Δ.

R. Ottanà et al. / European Journal of Medicinal Chemistry 50 (2012) 332e343
335
Table 1
synthesized compounds on the activation of the insulin signal
pathway were evaluated on mouse C2C12 skeletal muscle cells by
measuring the phosphorylation levels of IR as well as the cellular 2-
deoxyglucose uptake compared to insulin. In addition, the effect of
corresponding 2-phenylimino analogues 2aee (Fig. 1) [55] on
tyrosine phosphorylation of IR was evaluated in the same C2C12
cells.
In vitro inhibition activity of compounds 7aee and 8aee against human PTP1B,
isoforms IF1 and IF2 of human LMW-PTP and TC-PTP.
Compd
IC50
(
m
M)a
PTP1B
IF1
IF2
TC-PTP
7a
7b
7c
7d
7e
8a
8b
8c
8d
8e
10.4 ꢂ 0.6
2.2 ꢂ 0.2
5.6 ꢂ 0.1
1.4 ꢂ 0.1
2.4 ꢂ 0.1
10.9 ꢂ 0.3
2.5 ꢂ 0.5
5.1 ꢂ 0.9
3.0 ꢂ 0.2
9.0 ꢂ 0.3
13.7 ꢂ 0.7
7.3 ꢂ 0.2
12.0 ꢂ 0.4
4.0 ꢂ 0.1
5.8 ꢂ 0.3
12.0 ꢂ 0.4
11.2 ꢂ 0.3
9.6 ꢂ 0.4
6.4 ꢂ 0.1
11.6 ꢂ 0.3
>150
20.6 ꢂ 3.8
15.6 ꢂ 1.0
>100
33.2 ꢂ 1.0
>150
Molecular docking gave useful insight into the binding mode of
inhibitors 7 and 8.
25.4 ꢂ 0.7
15.2 ꢂ 0.7
21.8 ꢂ 0.4
18.0 ꢂ 0.4
17.5 ꢂ 1.3
9.1 ꢂ 0.2
11.7 ꢂ 0.4
2.2 ꢂ 0.1
20.2 ꢂ 0.8
20.5 ꢂ 0.5
7.1 ꢂ 0.2
11.6 ꢂ 0.4
9.5 ꢂ 0.5
27.0 ꢂ 0.9
2. Chemistry
The synthesis of inhibitors 7aee and 8aee was carried out
according to a previously reported procedure which is depicted in
Scheme 1 [54,55].
a
IC50 values were determined by regression analyses and expressed as
means ꢂ SE of three replicates.
The intermediate 4-[(3-arylthioureido)methyl]benzoic acids (3,
4) were obtained by the reaction of appropriately substituted ary-
lisothiocyanates with the commercially available 4-(aminomethyl)
benzoic acid in refluxing ethanol. The thiazolidinone core structure
was synthesized by the reaction of chloroacetyl chloride with
thioureas 3, 4, under basic conditions, which provided 4-[(2-
arylimino-4-oxo-3-thiazolidinyl)methyl]benzoic acids (5, 6) in
good yields. In proper experimental conditions, the condensation of
thioureas with chloroacetyl chloride can provide two isomers from
the two possible eneethiol intermediates produced by electronic
delocalization of the lone pair of the nitrogen atoms linked to the
thiocarbonyl group. As already reported, when the reaction was
performed in refluxing ethanol starting from N-alkyl-N0-aryl-
substituted thioureas only the 2-arylimino isomer, which was more
stable than the 2-alkylimino isomer, was obtained [56]. The reac-
tion of suitable benzaldehydes and 2-arylimino-4-thiazolidinones
5, 6 afforded 4-[(5-arylidene-2-arylimino-4-oxo-3-thiazolidinyl)
methyl]benzoic acids (7aee and 8aee, Scheme 1) via Knoevenagel
condensation. The reaction time was remarkable reduced when it
was performed by microwaves irradiation (80 ꢁC, max 150 psi, max
300 W, 6 h) although the yields were reduced (data not shown). As
demonstrated by the X-ray crystallographic analysis of previously
synthesized analogues [56,57], all compounds 7 and 8 were ob-
tained solely as Z isomers. In fact, NMR spectra showed in all cases
only one set of signals and no isomerization at the C]C exocyclic
double bond was subsequently observed [58,59].
Indeed, both 2-(4-trifluoromethylphenylimino) substituted
derivatives 7aee and 2-(4-methoxyphenylimino) analogues 8aee
exhibited appreciable inhibitory effects against PTP1B with low
micromolar IC50 values. These results suggest that in this series of
inhibitors the trifluoromethyl group on the 2-arylimino moiety is
generally more favourable than the methoxy group resulting in
higher affinity for the enzyme. In fact, compounds 7a and 7b were
slightly more active than analogues 8a and 8b, while 7d and 7e
were two to threefold more effective than 8d and 8e.
Among the 5-arylidene-2,4-thiazolidinediones
1 and 2-
phenylimino analogues 2 previously reported to be inhibitors of
both PTP1B and LMW-PTP, the substituents on the 5-arylidene
moiety play a determining role in the interaction with PTP1B and
correlated enzymes [53e55]. The phenoxy and benzyloxy groups
on the 5-arylidene moiety bring about enhanced PTP1B inhibitory
activity compared to non-aromatic substituents, improving the
stability of the enzyme/inhibitor complex, especially when inserted
in the para position [53,55]. The inhibitory ability of compounds 7
and 8 against selected PTPs proved to be in close agreement with
the structureeactivity relationships already acquired for this class
of inhibitors. In this new series, the insertion of the 4-phenoxy (7b
The structure of synthesized compounds was assigned on the
basis of spectroscopic and analytical data. 1H and 13C NMR spectra
were critical for the attribution of structures. In 1H NMR spectra of
compounds 5 and 6, the presence of two singlets at 3.87e3.90 ppm
and 5.10 ppm, attributable to the 5-CH2 and NeCH2 protons, as well
as the typical para-disubstituted systems were diagnostic for the
attribution of the structure.
The disappearance of 5-CH2 signal and the presence of the
methylidene proton resonance (7.66e8.16 ppm) proved to be
significant for the attribution of the 5-arylidene-2-arylimino-4-
thiazolidinone (7, 8) structures.
3. Results and discussion
The newly synthesized 4-[(5-arylidene-2-arylimino-4-oxo-3-
thiazolidinyl)methyl]benzoic acids 7aee and 8aee were tested
for their ability to inhibit the recombinant human PTP1B, the two
active isoforms (IF1 and IF2) of human LMW-PTP and TC-PTP
(Table 1). p-Nitrophenylphosphate was used as substrate.
All compounds proved to be interesting as reversible and
competitive inhibitors of PTP1B (Figs. 2 and 3). They were also
shown to inhibit both isoforms of LMW-PTP and TC-PTP, showing
an overall significant preference for PTP1B.
Fig. 2. Inhibition reversibility assay of compounds 7aee and 8aee. Aliquots of PTP1B
were incubated in the presence of at least 25-fold molar excess of each compound for
1 h at 37 ꢁC. Then, the enzyme was diluted 200-fold with the assay solution to measure
the residual activity (37 ꢁC, 5 mM pNPP). Control experiments were carried out adding
DMSO. Data represent the mean ꢂ SE of three replicates.

336
R. Ottanà et al. / European Journal of Medicinal Chemistry 50 (2012) 332e343
Fig. 3. Left panel: LineweavereBurk plots at differing concentrations of 7d, the best inhibitor of PTP1B. The used concentrations of inhibitor were: 0 mM (C); 0.5 mM (B); 1.0 mM
(:); 3 mM (,); 4 mM (A). Right panel: re-plot of the apparent Km values measured at differing concentrations of 7 d against the concentrations of the inhibitor. Data represent the
mean ꢂ SE of three replicates.
and 8b) and 4-benzyloxy (7d and 8d) groups provided more potent
PTP1B inhibitors than the corresponding 3-phenoxy (7a and 8a)
and 3-benzyloxy (7c and 8c) substituted derivatives, determining
a gain in potency which was significant for 7b (fivefold), 7d and 8b
(fourfold) compared to the meta substituted isomers (Table 1).
With the same substituent on the 2-phenylimino group, the
benzyloxy group on the 5-arylidene moiety was found to exert
a generally more favourable effect than the phenoxy one for
interactions with PTP1B regardless of its position on the benzene
ring.
preferential inhibitors of the IF1 isoform achieving from threefold to
more than tenfold higher inhibitory activity levels towards IF1 over
isoform IF2. The ascertained preferential inhibition of IF1 isoform
over IF2 of this class of inhibitors (1, 2, 7 and 8) can be considered
favourable for the control of insulin resistance, since the isoenzyme
IF1 appears to be the main LMW-PTP isoform associated with
hyperglycemia in diabetic patients [60].
Out of both compounds 7 and 8, the benzyloxy group, in either
the meta or para position of the 5-arylidene moiety, proved to be the
most favourable substituent, also for the interaction with LMW-PTP.
Within the same series of compounds, it can be inferred that the
para position is once again the most favourable as 7b, 7d and 8b, 8d
were more active than 7a, 7c and 8a, 8c towards both isoforms of
LMW-PTP.
Compound 7d, the best PTP1B inhibitor of this series of
compounds, proved to be up to eightfold more effective than the
other tested compounds, showing a Ki value of 1.2 ꢂ 0.2
mM.
Compounds 7e and 8e, 5-naphtalen-2-ylmethylidene substituted
derivatives, also exhibited good affinity for PTP1B. Once again, 4-
trifluoromethylphenylimino derivative 7e was more active against
PTP1B than the corresponding 4-methoxyphenylimino analogue 8e.
The comparison between the in vitro activity of derivatives
7aee and 8aee and the previously reported corresponding 2-
phenylimino analogues (2aee, Fig. 1) [55] highlighted that the
introduction of the selected substituents in para position of the 2-
phenylimino moiety generally brought about a clear-cut increase of
IC50 values against all tested PTPs, regardless of the electronic
nature of the substituents; however they maintained a good
inhibitory ability against PTP1B. This was particularly evident for
Among
the
5-naphtalen-2-ylmethylidene
substituted
compounds, the analogous p-trifluoromethylimino derivative 7e
showed clear selectivity towards the IF1 isoform, while 8e did not
determine any difference in the levels of inhibitory ability of both
isoenzymes.
The inhibitory effects of compounds 7 and 8 towards TC-PTP are
reported in Table 1. Compounds 7 and 8 were shown to be inhibi-
tors of TC-PTP, even though none of compounds achieved the
interesting levels of affinity towards PTP1B. Overall, the substituent
on 2-arylimino moiety was not critical since compounds 7 (p-tri-
fluorophenylimino substituted) were generally as active as
compounds 8 (p-methoxyphenylimino substituted).
compounds 7a and 8a (IC50 10.4 and 10.9
proved to be fivefold less active than the corresponding 2-
phenylimino analogue 2a (Fig. 1, IC50 1.9 M) [55]. However, the
most effective inhibitor of the new series, compound 7d, yielded an
interesting IC50 value (1.4 M) very close to that of the corre-
sponding 2-phenylimino 2d analogue (Fig. 1, IC50 1.1 M) [55]. The
mM, respectively), which
In both series of inhibitors (compounds 7 and 8) the presence of
the selected substituents on the para position of 5-benzylidene
group (7b, d and 8b, d) resulted to be favourable in comparison
m
m
with the meta substitution (7a, c and 8a, c). Within tri-
m
fluoromethylphenylimino substituted derivatives, compound 7d,
bearing a p-benzyloxy group on the 5-phenylidene moiety, proved
to be the best inhibitor of TC-PTP which resulted to be about
sevenfold more active than 4-phenoxy analogous 7b. Out of
compounds 8, the introduction of p-benzyloxy and p-phenoxy
groups on 5-phenylidene moiety provided inhibitors endowed
with comparable inhibitory activity levels towards TC-PTP. The
introduction of 5-naphtalen-2-ylmethylidene group (compounds
7e and 8e) proved to be detrimental for the interaction with TC-PTP
by determining a clear-cut decrease from threefold (8e) to tenfold
(7e) of inhibitory ability towards the enzyme. In all cases, the tested
compounds showed to be from twice to more than twenty-fold
selective towards PTP1B over TC-PTP. Taking into account the
high similarity between the catalytic domains of both enzymes, the
significant selectivity of p-trifluoromethylphenylimino substituted
compounds 7b, 7c and 7e can be considered an appreciable goal in
detrimental effect resulting from the introduction of CF3 and OCH3
groups could be attributable to a steric hindrance that prevented
effective interactions with the residues of the loops surrounding
the PTP1B catalytic site. On the basis of these results, it seems
plausible to suggest that the introduction of substituents on the 2-
phenylimino group, although endowed with different electronic or
lipophilic features, might hinder inhibitors from making beneficial
contacts with the amino acid residues flanking the catalytic site of
the enzyme.
The inhibition profiles of the tested compounds towards LMW-
PTP isoforms IF1 and IF2 are reported in Table 1. All tested
compounds were shown to inhibit both isoenzymes, without,
however, achieving the inhibitory levels against PTP1B. Like previ-
ously reported derivatives 1 and 2, 2-arylimino-5-arylidene-4-
thiazolidinones
7 and, to a lesser extent, 8 proved to be

R. Ottanà et al. / European Journal of Medicinal Chemistry 50 (2012) 332e343
337
the development of new agents for the treatment of obesity and
DM2. In fact, the inhibitory profile of these compounds can be
considered a favourable feature since the partial inhibition of TC-
PTP can contribute to the control of insulin resistance without
affecting immune system [44,46].
The discovery of pharmaceutically useful PTP1B inhibitors for
the treatment of insulin resistance in obesity and DM2 remains
a great challenge to develop small pTyr-mimetic molecules into
orally available drugs endowed with appropriate physicochemical
properties and in vivo effectiveness.
Therefore, in order to evaluate the efficacy of all tested
compounds in cells, in vitro inhibitors 7, 8 and, for comparison,
compounds 2aee were assayed on cultures of differentiated mouse
C2C12 myotubes which constitutively express both PTP1B and IR.
Even at the highest dose of 25 mM, compounds 7 and 8, did not
interfere with the viability of cells, which maintained their ability
to reduce the MTT to formazan (Fig. 4).
Incubating the tested compounds (25 mM final dose) with C2C12
cells highlighted their ability to influence IR activation (Fig. 5). All
compounds 7 and 8, except derivative 8d, induced a notable
increase in tyrosine phosphorylation levels of the IR
b subunits
compared to controls. Although it was not possible to draw clear
SARs, it is worth underlining that most compounds (7cee, 8aec) at
25 mM concentration in the medium were able to activate the IR to
a similar extent as that produced by insulin (10 nM). Interestingly,
compound 7d, the best inhibitor of both PTP1B and the IF1 isoform
of LMW-PTP, induced the highest degree of IR phosphorylation.
Moreover, all compounds 7 and 8, with the exception of 8d, were
shown to be more effective than corresponding 2-phenylimino
substituted compounds 2, even when these latter were tested at
higher dose (see supplementary material). Interestingly, these
results indicated that the introduction of the selected substituents
on the para position of the 2-phenylimino moiety was shown to
enhance the tyrosine phosphorylation of the IR, although
compounds 7 and 8 resulted to be less potent PTP1B inhibitors than
analogues 2.
Fig. 5. Effect of compounds 7aee and 8aee and insulin on tyrosine phosphorylation of
IR in differentiated C2C12 cells. Cells were incubated at 37 ꢁC with the compounds
(25 mM final concentration) for 90 min as well as with insulin (10 nM final concen-
tration) for 30 min. Proteins of the cell lysates were separated by SDS-PAGE and then
transferred to polyvinylidene membranes, which were probed with anti-phospho-
einsulin receptor
anti-actin -actin) antibodies. Data are representative of at least three separate
experiments. Top panel, data normalized against actin. Bottom panel: western blotting.
b-subunit (a-pIRb) antibodies. Then membrane was reprobed with
(
a
levels determined by compounds 7a, cee at 25 mM concentration in
the medium were comparable to that produced by insulin (10 nM).
These results suggest that the compounds are cell-permeable and
able to strengthen insulin signalling by interfering with down-
stream components of the insulin signal cascade.
Compounds 7 and 8 (25 mM final dose) were also incubated in
C2C12 cell cultures for 90 min in order to evaluate the cellular
uptake of 2-deoxyglucose (2-DOG). Compounds 7a, cee were
shown to increase intracellular 2-DOG uptake significantly whereas
analogues 8 did not elicit any effect (Fig. 6). The 2-DOG uptake
Fig. 6. 2-Deoxyglucose (2-DOG) uptake in C2C12 myotubes. Cells were serum-starved
for 20 h and then treated with 7aee and 8aee (25
for 90 min, or insulin (10 nM) for 30 min. Subsequently, 0.1
m
M final concentration), or DMSO
Fig. 4. MTT responses in differentiated C2C12 cell cultures treated with compounds
m
Ci 2-deoxy-D-1,2-[3H-
7aee and 8aee (25
m
M final concentration in the culture medium). The MTT assay was
glucose] in RingereHepes was added to each well, and glucose uptake was allowed at
37 ꢁC for 15 min. Aliquots of the lysates were counted in a liquid scintillation counter.
Data represent the mean ꢂ SE of three replicates, and were normalized against protein
concentration.
used to evaluate toxicity of the tested compounds after 2 h of exposure. Each point is
the mean absorbance at 570 nm converted to percentage of control values (DMSO).
Data represent the mean ꢂ SE of three replicates.

338
R. Ottanà et al. / European Journal of Medicinal Chemistry 50 (2012) 332e343
The observed structureeactivity relationships indicate that
solvent (Fig. 7). The phenyl ring in the 4-phenoxybenzylidene
the trifluoromethylphenylimino group (7) can significantly
contribute to promoting the insulinomimetic activity and
bioavailability of this class of compounds, probably due to the
greater lipophilicity of the CF3 group compared to the methoxy
group (8).
moiety of compound 8b allows for an orientation towards Arg254
enabling a cation-p interaction, while in 8d the benzyl ring reaches
out to the hydrophobic Ala27, which might provide an explanation
for the higher activities of these two compounds in series 8 (Fig. 7).
Overall, the binding of compounds 7 and 8 is dominated by
hydrophobic interactions involving residues Tyr46, Phe182, Ile219
and Met258 (Fig. 8 shows compound 7e as a representative
example in detail).
In order to gain insight into the interaction of inhibitors with
the enzyme binding site molecular docking studies were per-
formed. PDB [61] entry 2QBS [62] was selected as template for
human PTP1B. 5PNT [63] and 1XWW [64] were selected as
structures for human LMW-PTP IF1 and IF2, respectively. The
selected protein structures were processed with CCDCs software
GOLD version 5 [65] for docking experiments. After docking,
structures were minimized in the binding site using LigandSc-
out’s [66] MMFF94 implementation, and prioritized using a 3D
pharmacophore model that explains the most plausible docking
poses. The conformations obtained were all oriented with the
carboxylic acid towards the catalytic centre. For compounds 7,
the trifluoromethylphenylimino moiety is oriented towards the
solvent, while the phenyl ring of the residue of p-methylbenzoic
The lower activity of the tested compounds towards IF1 and IF2
may be explained by the smaller and less deep binding pockets
compared to human PTP1B. For IF1, an additional weak hydrogen
bond with Tyr49 could be proposed with either the ether oxygen of
the 5-arylidene moiety (7aed and 8aed) or the methoxy oxygen of
the 2-arylimino moiety. In Fig. 9, compound 7d docked into LMW-
PTP isoform IF1 is reported as representative of this binding mode.
In this pose, the carboxylate group of the p-methylbenzoic acid
moiety potentially interacts with residues of the catalytic site
through charge transfer (Arg18) and a network of hydrogen bonds
(Ile16, Cys17 and Arg18). In addition, a weak hydrogen bond of the
benzyloxy oxygen atom with Tyr49 is observed. For IF2, no plau-
sible docking poses could be found for compounds 7c and 7d. The
3-benzyloxybenzylidene moiety of 7c is trapped between the tri-
fluoromethyl group and Trp49, and therefore the compound would
be forced in a highly strained conformation upon putative binding,
which might account for the inactivity of compound towards IF2
(data not shown).
acid shows a
p-stacking interaction with Phe182 located in the
WPD loop. However, the trifluoromethyl moiety of compounds
7b, 7d and 7e is oriented in such a way that an electrostatic
interaction of one of the fluorine atoms with Arg24 becomes
possible (Figs. 7 and 8), which may explain the higher activity of
those compounds.
For compounds 8, we observed poses where the methoxy
moiety either points to the side of the WPD loop or towards the
4. Conclusions
In continuing the search for inhibitors of PTP1B and LMW-PTP
through targeted structural modifications of previously reported
analogues, new 5-arylidene-2-aryilimino-4-thiazolidinones, highly
active inhibitors of both PTP1B and the isoform IF1 of LMW-PTP,
were identified. In all cases, the tested compounds showed to be
from moderately to significantly selective towards PTP1B over
closely related TC-PTP.
Moreover, most tested compounds were endowed with cellular
activity. They were able to activate the IR on mouse C2C12 skeletal
muscle cells since their addition to the culture medium brought
about an increase in the level of IR
b tyrosine phosphorylation.
Interestingly, compounds 7cee and 8aec induced IR activation
levels similar to those of insulin. In addition, compounds 7 and 8
resulted to enhance IR tyrosine phosphorylation to a greater
extent than analogues 2, indicating that the introduction of the
selected substituents on the para position of the 2-phenylimino
moiety can favour the effect of these compounds on the IR
activation.
At the tested doses, compounds 7a, cee were also able to
promote the intracellular uptake of 2-DOG to an extent that is
comparable to that induced by insulin, therefore they can be
considered insulinomimetic agents. Overall, these results indi-
cate that they are able to strengthen the insulin signal likely by
inhibiting PTP1B activity; this causes the enhancement of
insulin receptor phosphorylation and thus activates other
downstream components of the insulin signalling pathway
including those of the final pathway that induce the movement
of GLUT4 vesicles to the plasma membrane. However, we found
that compounds 8aec, e were able to stimulate IR phosphory-
lation but were unable to increase cellular 2-DOG uptake. These
findings suggest that they not only inhibit PTP1B but may also
negatively interfere with downstream components of the
insulin signal cascade.
Fig. 7. Compounds 7aee (top) and 8aee (bottom) docked into human PTP1B (PDB
entry 2QBS). Molecular surface shows hydrophobic regions in yellow, hydrophilic
regions in blue. Visualization was performed with LigandScout 3.02 [66]. (For inter-
pretation of the references to colour in this figure legend, the reader is referred to the
web version of this article.)
As
regards
structureeactivity
relationships,
2-(4-
trifluoromethylphenylinimo) substituted analogues 7 proved to be

R. Ottanà et al. / European Journal of Medicinal Chemistry 50 (2012) 332e343
339
Fig. 8. Compound 7e docked into human PTP1B (PDB entry 2QBS). Interactions were automatically detected and visualized in 2D (left) and 3D (right) by LigandScout [66]. Yellow
spheres indicate lipophilic regions, pink disks cation- interactions, and red arrows hydrogen bonds or electrostatic fluorine-donor interactions. (For interpretation of the references
p
to colour in this figure legend, the reader is referred to the web version of this article.)
endowed with excellent insulinomimetic effect in mouse skeletal
muscle cells, despite the fact that the insertion of a substituent in the
para position of the 2-phenylimino ring was found to be either
indifferent or unfavourable for the in vitro inhibition of PTP1B and
LMW-PTP.
Of note among these is compound 7d, the best inhibitor of both
PTP1B and the isoform IF1 of LMW-PTP, which determined the
highest degree of phosphorylation of IR and also promoted a high
level of glucose uptake. In addition, it showed a modest selectivity
towards PTP1B over TC-PTP.
It is also worth noting that, out of the tested compounds, 7c is
the most selective inhibitor of PTP1B and LMW-PTP IF1 isoform
over both IF2 and TC-PTP and is endowed with a potent insulino-
mimetic activity on C2C12 cells. Overall, a relatively low inhibition
of TC-PTP associated with PTP1B inhibition is considered to be
useful to cooperate in the improvement of insulin resistance in
DM2 and obesity without interfering with haematopoietic and
immune functions [44,46]. Compounds 7, which are endowed of
promising activity profiles, can therefore be recruited as lead
compounds for the design of new drugs for the treatment of DM2
and obesity.
5. Experimental section
5.1. Chemistry
Melting points were recorded on a Kofler hot-stage apparatus
and are uncorrected. Reactions were monitored by thin layer
chromatography (TLC) on Merck 0.2 mm precoated silica gel plates
aluminium sheets (60 F 254 Merck) and visualized by irradiation
with a UV lamp.
Elemental analyses (C, H, N) were performed on a C. Erba mod.
1106 elem. Analyzer and were within ꢂ0.4% of the theoretical
values. 1H and 13C NMR spectra were recorded on a Varian 300
magnetic resonance spectrometer (300 MHz for 1H and 75 MHz for
13C). Chemical shifts are given in
d units (ppm) relative to internal
standard Me4Si in CDCl3 or DMSO-d6 used as solvents. Coupling
constants (J) are given in hertz (Hz). 13C NMR spectra were deter-
mined by Attached Proton Test (APT) experiments and the reso-
nances were always attributed by protonecarbon heteronuclear
chemical shift correlation. Reactions under microwaves were per-
formed in a microwave oven Discover (CEM). Unless stated other-
wise, all materials were obtained from commercial suppliers and
Fig. 9. Complementary chemical features of compound 7d docked into LMW-PTP isoform IF1 (PDB entry 5PNT) analyzed in 3D (left) and 2D (right) using LigandScout [66]. Yellow
spheres indicate lipophilic regions, red arrows hydrogen bond acceptors, and red rays the ionic interaction if the carboxylic acid moiety with Arg18. (For interpretation of the
references to colour in this figure legend, the reader is referred to the web version of this article.)

340
R. Ottanà et al. / European Journal of Medicinal Chemistry 50 (2012) 332e343
used without further purification. All tested compounds possessed
a purity of at least 95% as verified by elemental analyses by
comparison with the theoretical values.
5.1.3.1. 4-{[2-(Trifluoromethylphenylimino)-4-oxo-5-(3-phenoxyben-
zylidene)-3-thiazolidinyl]methyl}benzoic acid (7a). Yield 27%; mp
214e216 ꢁC; 1H NMR (CDCl3):
d 5.22 (s, 2H, NCH2); 6.99e7.04 (m,
3H, arom); 7.08e7.16 (m, 2H, arom); 7.29e7.41 (m, 6H, arom);
7.55e7.66 (m, 4H, arom); 7.75 (s, 1H, CH methylidene); 8.03 (m, 2H,
arom); Anal. calcd for C31H21F3N2O4S: C, 64.80 H, 3.68; N, 4.88.
Found: C, 64.69; H, 3.51; N, 4.73.
5.1.1. General procedure for the synthesis of 4-[(3-arylthioureido)
methyl]benzoic acids (3, 4)
A mixture of appropriate arylisothiocyanate (2.6 mmol) and 4-
(aminomethyl)benzoic acid (0.51 g, 3.4 mmol) in ethanol (50 ml)
was refluxed for 24 h. The solvent was evaporated under vacuo and
the solid residue was crystallized from methanol providing reaction
products 3 and 4.
5.1.3.2. 4-{[2-(4-Trifluoromethylphenylimino)-4-oxo-5-(4-phenoxyben-
zylidene)-3-thiazolidinyl]methyl}benzoic acid (7b). Yield 36%; mp
256e258 ꢁC; 1H NMR (CDCl3):
d 5.24 (s, 2H, NCH2); 6.99e7.07 (m, 6H,
arom); 7.19 (m, 1H, arom); 7.36e7.43 (m, 4H, arom); 7.61e7.65 (m, 4H,
arom); 7.78 (s, 1H, CH methylidene); 8.08 (m, 2H, arom); Anal. calcd
for C31H21F3N2O4S: C, 64.80; H, 3.68; N, 4.88. Found: C, 64.62; H, 3.39;
N, 4.57.
5.1.1.1. 4-{[3-(4-Trifluoromethylphenyl)thioureido]methyl}benzoic
acid (3). Yield 74%; mp 265e268 ꢁC; 1H NMR (DMSO-d6):
d 4.83 (d,
J ¼ 5.4 Hz, 2H, NCH2); 7.36 (m, 2H, arom); 7.43e7.88 (m, 4H, arom);
7.91 (m, 2H, arom); 8.66 (brs, 1H, NHCH2); 10.15 (brs, 1H, NH); 11.38
(brs, 1H, COOH). Anal. calcd for C16H13F3N2O2S: C, 54.23; H, 3.70; N,
7.91. Found: C, 54.01; H, 3.55; N, 7.79.
5.1.3.3. 4-{[5-(3-Benzyloxybenzylidene)-2-(4-trifluoromethylphenyli-
mino)-4-oxo-3-thiazolidinyl]methyl}benzoic acid (7c). Yield 37%;
242e244 ꢁC (dec.); 1H NMR (DMSO-d6):
d 5.11 (s, 2H, NCH2); 5.13 (s,
5.1.1.2. 4-{[3-(4-Methoxyphenyl)thioureido]methyl}benzoic
acid
2H, OCH2); 7.10 (m, 2H, arom); 7.21 (m, 2H, arom); 7.30e7.40 (m, 8H,
arom); 7.77e7.87 (m, 6H, arom and CH methylidene); Anal. calcd for
C32H23F3N2O4S: C, 65.30; H, 3.94; N, 4.76. Found: C, 65.21; H, 3.79;
N, 4.63.
(4). Yield 93%; mp 245e248 ꢁC; 1H NMR (CDCl3):
d
3.67 (s, 3H,
OCH3); 4.78 (s, 2H, NCH2); 6.78 (m, 2H, arom); 6.86 (brs, 1H, NH);
7.11 (m, 2H, arom); 7.25 (m, 2H, arom); 7.87 (m, 2H, arom); 8.54 (s,
1H, NHCH2); 9.98 (s, 1H, COOH); Anal. calcd for C16H16N2O3S: C,
60.74; H, 5.10; N, 8.85. Found: C, 60.55; H, 5.21; N, 8.73.
5.1.3.4. 4-{[5-(4-Benzyloxybenzylidene)-2-(4-trifluoromethylphenyli-
mino)-4-oxo-3-thiazolidinyl]methyl}benzoic acid (7d). Yield 62%; mp
5.1.2. General method for the synthesis of 4-[(2-arylimino-4-oxo-3-
thiazolidinyl)methyl]benzoic acids (5, 6)
278e280 ꢁC; 1H NMR (CDCl3):
d 5.01 (s, 2H, NCH2); 5.11 (s, 2H,
OCH2); 6.90e6.98 (m, 4H, arom); 7.29e7.55 (m, 11H, arom); 7.66 (s,
1H, CH methylidene); 7.93 (m, 2H, arom); Anal. calcd for
C32H23F3N2O4S: C, 65.30; H, 3.94; N, 4.76. Found: C, 65.19; H, 3.71;
N, 4.59.
A mixture of appropriate 4-[(3-arylthioureido)methyl]benzoic
acid (3, 4) (1.7 mmol) and triethylamine (0.52 g, 5.1 mmol) in
ethanol (50 ml) was maintained at reflux for 30 min. After cooling
to room temperature, a solution of chloroacetyl chloride (0.28 g,
2.5 mmol) in ethanol (10 ml) was added and the mixture was
refluxed for 24 h. After evaporation of the solvent under reduced
pressure, the crude solid was dissolved in CHCl3 and the solution
was washed with water. The crude mixture was purified by
chromatography on silica gel column (diethyl ether/n-hexane, 6/
4). The crystallization from methanol provided pure compounds 5
and 6.
5.1.3.5. 4-{[2-(4-Trifluoromethylphenylimino)-5-(naphthalen-2-ylme-
thylidene)-4-oxo-3-thiazolidinyl] methyl]}benzoic acid (7e). Yield
35%; mp dec. starting from 240 ꢁC; 1H NMR (DMSO-d6):
d 5.19 (s, 2H,
NCH2); 7.22 (m, 2H, arom); 7.52e7.60 (m, 5H, arom); 7.63 (m, 2H,
arom); 7.76e8.01 (m, 6H, arom); 8.16 (s, 1H, CH methylidene); Anal.
calcd for C29H19F3N2O3S: C, 65.41; H, 3.60; N, 5.26. Found: C, 65.29;
H, 3.38; N, 5.01.
5.1.2.1. 4-{[2-(4-Trifluoromethylphenylimino)-4-oxo-3-thiazolidinyl]
methyl}benzoic acid (5). Yield 76%; yellow oil; 1H NMR (CDCl3):
5.1.3.6. 4-{[2-(4-Methoxyphenylimino)-4-oxo-5-(3-phenoxybenzyli-
dene)-3-thiazolidinyl]methyl}benzoic acid (8a). Yield 50%; mp
d
3.91 (s, 2H, 5-CH2); 5.10 (s, 2H, NCH2); 7.02 (m, 2H, arom);
196e198 ꢁC; 1H NMR (CDCl3):
d 3.86 (s, 3H, OCH3); 5.22 (s, 2H,
7.56e7.61 (m, 4H, arom); 8.09 (m, 2H, arom); Anal. calcd for
C18H13F3N2O3S: C, 54.82; H, 3.32; N, 7.10. Found: C, 54.52; H, 3.43;
N, 6.94.
NCH2); 6.91e7.03 (m, 6H, arom); 7.12e7.19 (m, 3H, arom);
7.29e7.36 (m, 4H, arom); 7.60 (m, 2H, arom); 7.70 (s, 1H, CH
methylidene); 8.08 (m, 2H, arom); 13C NMR (DMSO-d6):
d 47.3
(NCH2); 55.6 (OCH3); 115.0, 119.7, 119.9, 124.4, 125.1, 127.8, 129.9,
130.1, 130.4, 131.2 (CH arom); 127.9 (C5); 135.4, 140.7, 154.7, 156.1,
157.3, 158.1 (Cq); 166.1, 168.1 (CO); Anal. calcd for C31H24N2O5S: C,
69.39; H, 4.51; N, 5.22. Found: C, 69.28; H, 4.39; N, 5.34.
5.1.2.2. 4-{[2-(4-Methoxyphenylimino)-4-oxo-3-thiazolidinyl]methyl}
benzoic acid (6). Yield 77%; mp 157e160 ꢁC; 1H NMR (CDCl3):
d 3.81
(s, 3H, OCH3); 3.87 (s, 2H, 5-CH2); 5.10 (s, 2H, NCH2); 6.90 (m, 4H,
arom); 7.58 (m, 2H, arom); 8.07 (m, 2H, arom); 9.98 (s, 1H, COOH);
Anal. Calcd for C18H16N2O4S: C, 60.66; H, 4.53; N, 7.86. Found: C,
60.49; H, 4.61; N, 7.76.
5.1.3.7. 4-{[2-(4-Methoxyphenylimino)-4-oxo-5-(4-phenoxybenzyli-
dene)-3-thiazolidinyl]methyl}benzoic acid (8b). Yield 50%; mp
243e245 ꢁC; 1H NMR (CDCl3):
d 3.83 (s, 3H, OCH3); 5.23 (s, 2H, NCH2);
5.1.3. General method for the synthesis of 4-[(5-arylidene-2-
arylimino-4-oxo-3-thiazolidinyl)methyl]benzoic acids (7aee, 8aee)
A mixture of appropriate 4-[(2-arylimino-4-oxo-3-thiazolidinyl)
methyl]benzoic acids (5, 6) (1.5 mmol), opportune aldehyde
(1.5 mmol) and piperidine (0.26 g, 3 mmol) in ethanol (50 ml) was
refluxed until the starting materials disappeared (72e96 h, TLC
analysis). The solution was poured in water acidified with glacial
acetic acid. The crude solid was washed with H2O, dried and
purified by chromatography on silica gel column (diethyl ether/n-
hexane, 8/2). The crystallization from methanol provided pure
acids 7 and 8.
6.93 (m, 5H, arom); 7.00 (m, 2H, arom); 7.04 (m, 2H, arom); 7.36 (m,
2H, arom), 7.43 (m, 2H, arom); 7.62 (m, 2H, arom); 7.75 (s, 1H, CH
methylidene); 8.08 (m, 2H, arom); 13C NMR (DMSO-d6):
d 40.3
(NCH2); 55.4 (OCH3); 114.9, 118.6, 119.7, 124.6, 127.8, 129.6, 129.8,
130.2, 130.3, 130.4, 132.2 (CH arom); 119.8 (C5); 127.8, 140.7, 141.2,
149.3, 156.1, 158.7, 158.8 (Cq); 166.0, 168.1 (CO); Anal. calcd for
C31H24N2O5S: 69.39; H, 4.51; N, 5.22. Found: C, 69.45; H, 4.43; N, 5.31.
5.1.3.8. 4-{[5-(3-Benzyloxybenzylidene)-2-(4-methoxyphenylimino)-
4-oxo-3-thiazolidinyl]methyl}benzoic acid (8c). Yield 38%; mp
205e207 ꢁC; 1H NMR (CDCl3):
d 3.84 (s, 3H, OCH3); 5.08 (s, 2H,

R. Ottanà et al. / European Journal of Medicinal Chemistry 50 (2012) 332e343
341
NCH2); 5.23 (s, 2H, OCH2); 6.94e7.08 (m, 6H, arom); 7.33e7.39 (m,
electrophoresis according to Laemmli [72]. Recombinant TC-PTP
was obtained from New England Biolabs.
7H, arom); 7.62 (m, 2H, arom); 7.73 (s, 1H, CH methylidene); 8.10
(m, 2H, arom); 13C NMR (DMSO-d6):
d 45.9 (NCH2); 55.6 (OCH3);
69.7 (OCH2); 115.1, 116.7, 117.2, 121.8, 122.1, 122.2, 122.5, 127.9, 128.0,
128.3, 128.8, 130.3 (CH arom); 121.2 (C5); 132.6, 140.8, 157.1, 159.0,
159.4 (Cq); 166.0, 167.4 (CO); Anal. calcd for C32H26N2O5S: C, 69.80;
H, 5.09; N, 4.76. Found: C, 69.69; H, 5.29; N, 4.66.
5.3.1. Phosphatase assay and inhibition experiments
PTP1B and LMW-PTP assay was carried out at 37 ꢁC using p-
nitrophenylphosphate as substrate and the final volume was 1 ml.
The assay buffer (pH 7.0) contained 0.075 M b,b-dimethylglutarate
buffer, 1 mM EDTA and 5 mM dithiothreitol. The reactions were
initiated by addition of aliquots of the enzyme preparations and
stopped at appropriate times by adding 4 ml of 1 M KOH. The
released p-nitrophenolate ion was determined by reading the
absorbance at 400 nm (ε ¼ 18,000 Mꢀ1 cmꢀ1). The main kinetic
parameters (Km and Vmax) were determined by measuring the
initial rates at different substrate concentrations. Experimental
data were analysed using the Michaelis equation and a non-linear
fitting program FigSys [73]. Inhibition constants were determined
measuring initial hydrolysis rates at differing substrate and inhib-
itor concentrations. The apparent Km values measured at the
various inhibitor concentrations were plotted against concentra-
tion of the inhibitor to calculate the Ki values. All measurements of
initial rate were carried out in triplicate. For each inhibitor, IC50 was
determined by measuring the initial hydrolysis rate under fixed p-
nitrophenylphosphate concentration, equal to the Km value of each
tested PTP. Data were fitted to the following equation using the
FigSys program:
5.1.3.9. 4-{[5-(4-Benzyloxybenzylidene)-2-(4-methoxyphenylimino)-
4-oxo-3-thiazolidinyl]methyl}benzoic acid (8d). Yield 40%; mp
225e227 ꢁC; 1H NMR (CDCl3):
d 3.84 (s, 3H, OCH3); 5.10 (s, 2H,
NCH2); 5.22 (s, 2H, OCH2); 6.94 (m, 4H, arom); 7.01 (m, 2H, arom);
7.40 (m, 7H, arom); 7.61 (m, 2H, arom); 7.73 (s, 1H, CH methyl-
idene); 8.07 (m, 2H, arom); 13C NMR (DMSO-d6):
d 40.4 (NCH2);
55.6 (OCH3); 69.8 (OCH2); 115.0, 115.9, 126.2, 127.8, 128.7, 129.9,
130.0, 132.2 (CH arom); 118.3 (C5); 126.3, 130.3, 140.8, 141.2, 156.8,
160.1 (Cq); 166.2 (CO); Anal. calcd for C32H26N2O5S: C, 69.80; H,
5.09; N, 4.76. Found: C, 69.64; H, 5.25; N, 4.59.
5.1.3.10. 4-{[2-(4-Methoxyphenylimino)-5-(naphthalen-2-ylmethyli-
dene)-4-oxo-3-thiazolidinyl]methyl]}benzoic acid (8e). Yield 41%;
mp 250e252 ꢁC; 1H NMR (DMSO-d6):
d 3.75 (s, 3H, OCH3); 5.19 (s,
2H, NCH2); 6.98 (s, 4H, arom); 6.98 (m, 4H, arom); 7.49e7.65 (m,
5H, arom); 7.92e8.01 (m, 6H, arom); 8.15 (s, 1H, CH methylidene);
Anal. calcd for C29H22N2O4S: C, 70.43; H, 4.48; N, 5.66. Found: C,
70.27; H, 4.37; N, 5.51.
Max ꢀ Min
y ¼
slope þ Min
ꢀ
ꢁ
5.2. Molecular modelling
x
1 þ
IC50
Docking studies including protein and ligand preparations were
carried out using the software GOLD 5 (http://www.ccdc.cam.ac.
uk/products/life_sciences/gold/) [65,67] using default parameters
(GOLDScore, 100% search efficiency). PDB [61] entries 2QBS [62]
(human PTP1B), 5PNT [63] (LMW-PTP IF1) and 1XWW [64]
(LMW-PTP IF2) were selected as protein templates. The active site
was determined by selecting all residues with any atom within 7 Å
from the outside atoms the co-crystallized ligand for 2QBS, and the
analogue residues for the other two PDB structures using the
sequence alignment and structure overlay functionality of the
software tool MOE [68] provided by the Chemical Computing
Group (http://www.chemcomp.com/). Docking was performed
without any constraints and all compounds were minimized using
LigandScout’s [66,69,70] general purpose MMFF94 implementation
after docking. In order to get the most probable binding poses, a 3D
pharmacophore was developed using LigandScout. For 2QBS the
co-crystallized ligand was used, for 5PNT and 1XWW analog 3D
pharmacophores were created using the docking poses of the most
active ligands synthesized in this work. These 3D pharmacophores
were then utilized to select the docking poses that formed the basis
for the analysis described. LigandScout was also used for visuali-
zation and analysis.
where y ¼ vi/vo, i.e. the ratio between the activity measured in the
presence of the inhibitor (vi) and the activity of the control without
the inhibitor (vo). The parameter “x” is the inhibitor concentration.
To verify if compounds 7aee and 8aee were reversible inhibi-
tors, appropriate aliquots of PTP1B were incubated in the presence
of at least 25-fold molar excess of inhibitor for 1 h at 37 ꢁC. Control
experiments were performed adding DMSO instead of inhibitor.
After this interval time, the enzyme solutions were diluted 200-fold
and the residual enzyme activity was assayed.
The assay of TC-PTP activity was performed as follows: an
appropriate aliquot of enzyme (0.02 mg) was diluted to 0.75 ml with
the same assay buffer used for the other PTPs, containing p-nitro-
phenylphosphate at the final concentration 2.8 mM (a value cor-
responding to that of the enzyme Km). After the appropriate
incubation time, reactions were stopped by adding 0.25 ml of 0.4 M
KOH. The released p-nitrophenolate ion was measured reading the
absorbance at 400 nm. The IC50 values were determined as
described above.
5.3.2. Cellular assays
C2C12 cells were obtained from the European Collection of Cell
Cultures (Salisbury, UK). Assays were carried out using differenti-
ating mouse C2C12 myoblasts which constitutively express both
PTP1B and IR. Cells were grown in DMEM, supplemented with 10%
5.3. Enzyme section
FBS, 2 mM L-glutamine, 100 U/ml penicillin, and 100 mg/ml strep-
The complete coding sequences of human IF1, IF2 LMW-PTPs
and human PTP1B were cloned in frame with the sequence of the
glutathione S-transferase in the pGEX-2T bacterial expression
vector. Enzyme expression and purification were achieved in the
Escherichia coli TB1 strain [71]. Briefly, the recombinant fusion
proteins were purified from bacterial lysate using a single step
affinity chromatography on glutathione-Sepharose. The solution
containing purified fusion proteins was treated with thrombin for
3 h at 37 ꢁC. Then the enzymes were purified from GST and
thrombin by gel filtration on Superdex G-75. The purity of protein
preparations were analyzed by SDSepolyacrylamide gel
tomycin at 37 ꢁC in 5% CO2 humidified atmosphere. To induce
differentiation, the medium of 90% subconfluent cells was replaced
by DMEM supplemented with 2% horse serum, and cells were
incubated at 37 ꢁC. After 2 days, the medium was replaced by fresh
differentiation medium, and cells were incubated for additional 2
days at the same conditions. The obtained myotubes were starved
for 20 h before experiments. Compounds 7aee and 8aee (25 mM
final concentration) or DMSO (1.2% final concentration) were added
to the culture medium and the cells were incubated for 90 min at
37 ꢁC. The positive control with insulin (10 nM final concentration)

342
R. Ottanà et al. / European Journal of Medicinal Chemistry 50 (2012) 332e343
was carried out incubating cells for 30 min a 37 ꢁC. The cells were
then lysed with SDS Laemmly sample buffer and analyzed by SDS-
PAGE and Western Blot using both anti-phosphorylated insulin
receptor and anti-actin antibodies (Santa Cruz Biotechnology Inc.).
To perform the cytotoxicity assays, DMEM was supplemented
with MTT (0.5 mg/ml final concentration) and then myotubes were
incubated at 37 ꢁC. After 2 h, the medium was removed and cells
were lysed with DMSO. The absorption levels of formazan were
read at 570 nm.
[23] M.M. Swarbrick, P.J. Havel, A.A. Levin, A.A. Bremer, K.L. Stanhope, M. Butler,
S.L. Booten, J.L. Graham, R.A. Mckay, S.F. Murray, L.M. Watts, B.P. Monia,
S. Bhanot, Endocrinology 150 (2009) 1670e1679.
[24] T.O. Johnson, J. Ermolieff, M.R. Jirousek, Nat. Rev. Drug Discov. 1 (2002)
696e709.
[25] B.P. Montalibet, B.P. Kennedy, Drug Discov. Today Ther. Strateg. 2 (2005)
129e135.
[26] A.M. Hennige, N. Stefan, K. Kapp, R. Lehmann, C. Weigert, A. Beck, K. Moeschel,
J. Mushack, E. Schleicher, H.-U. Haring, FASEB J. 20 (2006) E381eE389.
[27] G. Sesti, Best Pract. Res. Cl. En. 20 (2006) 665e679.
[28] J.E. Shaw, R.A. Sicree, P.Z. Zimmet, Diabetes Res. Clin. Pract. 87 (2010) 4e14.
[29] A.P. Combs, J. Med. Chem. 53 (2010) 2333e2344.
To determine glucose uptake, C2C12 cells were seeded in 24-
well plates, differentiated as described above, and serum-starved
for 20 h. Then the myotubes were treated with compounds 7aee
[30] S. Lee, Q. Wang, Med. Res. Rev. 27 (2007) 553e573.
[31] S. Zhang, Z.Y. Zhang, Drug Discov. Today 12 (2007) 373e381.
[32] D.V. Erbe, S. Wang, Y.L. Zhang, K. Harding, L. Kung, M. Tam, L. Stolz, Y. Xing,
S. Furey, A. Qadri, L.D. Klaman, J.F. Tobin, Mol. Pharmacol. 67 (2005) 69e77.
[33] N. Takahashi, Y. Qi, H.R. Patel, R.S. Ahima, J. Hepatol. 41 (2004) 391e398.
[34] K.A. Lantz, S.G. Hart, S.L. Planey, M.F. Roitman, I.A. Ruiz-White, H.R. Wolfe,
M.P. McLane, Obesity 18 (2010) 1516e1523.
[35] S.K. Pandey, X.X. Yu, L.M. Watts, M.D. Michael, K.W. Sloop, A.R. Rivard,
T.A. Leedom, V.P. Manchem, L. Samadzadeh, R.A. McKay, B.P. Monia, S. Bhanot,
J. Biol. Chem. 282 (2007) 14291e14298.
[36] P. Chiarugi, P. Cirri, F. Marra, G. Raugei, G. Camici, G. Manao, G. Ramponi,
Biochem. Biophys. Res. Commun. 238 (1997) 676e682.
[37] M.L. Taddei, P. Chiarugi, P. Cirri, D. Talini, G. Camici, G. Manao, G. Raugei,
G. Ramponi, Biochem. Biophys. Res. Commun. 270 (2000) 564e569.
[38] R. Maccari, R. Ottanà, J. Med. Chem. 55 (2012) 2e22.
and 8aee (25 mM final concentration) or 1.2% DMSO for 90 min.
The positive control with insulin (10 nM final concentration) was
carried out incubating cells for 30 min a 37 ꢁC.
Glucose transport was assayed as follows: after inhibitor or
other treatments, 0.1
mCi of 2-deoxy-D
-1,2-[3H]-glucose (20 Ci/
mmol) in RingereHepes buffer was added to each well and glucose
uptake was allowed at 37 ꢁC for 15 min. Cells were subsequently
washed with cold PBS and lysed in 0.2 M NaOH. Aliquots of the
lysates were counted with a b-scintillation counter, and data were
normalized against protein concentration.
[39] L.F. Iversen, K.B. Møller, A.K. Pedersen, G.H. Peters, A.S. Petersen,
H.S. Andersen, S. Branner, S.B. Mortensen, N.P.H. Møller, J. Biol. Chem. 277
(2002) 19982e19990.
[40] A. Bourdeau, N. Dubé, M.L. Tremblay, Curr. Opin. Cell Biol. 17 (2005) 203e209.
[41] K.M. Heinon, K.M. Heinonen, F.P. Nestel, E.W. Newell, G. Charette,
T.A. Seemayer, M.L. Tremblay, W.S. Lapp, Blood 103 (2004) 3457e3464.
[42] S. Galic, C. Hauser, B.B. Kahn, F.G. Haj, B.G. Neel, M.K. Tonks, T. Tiganis, Mol.
Cell. Biol. 25 (2005) 819e829.
[43] S. Galic, M. Klingler-Hoffmann, M.T. Fodero-Tavoletti, M.A. Puryer, T.-C. Meng,
M.K. Tonks, T. Tiganis, Mol. Cell. Biol. 23 (2003) 2096e2108.
[44] A. Fukushima, K. Loh, S. Galic, B. Fam, B. Shields, F. Wiede, M.L. Tremblay,
M.J. Watt, S. Andrikopoulos, T. Tiganis, Diabetes 59 (2010) 1906e1914.
[45] K. Loh, A. Fukushima, X. Zhang, S. Galic, D. Briggs, P.J. Enriori, S. Simonds,
F. Wiede, A. Reichenbach, C. Hauser, N.A. Sims, K.K. Bence, S. Zhang, Z.-
Y. Zhang, B. Kahn, B.G. Neel, Z.B. Andrews, M.A. Cowley, T. Tiganis, Cell Metab.
14 (2011) 684e698.
Acknowledgements
Work supported in part by the ‘Ente Cassa di Risparmio di
Firenze’ and by University of Messina.
Appendix. Supplementary material
Supplementary material associated with this article can be
found, in the online version, at doi:10.1016/j.ejmech.2012.02.012.
[46] A.J. Nichols, R.D. Mashal, B. Balkan, Drug Develop. Res. 67 (2006) 559e566.
[47] C. Li, X.-P. He, Y.-J. Zhang, Z. Li, L.-X. Gao, X.-X. Shi, J. Xie, J. Li, G.-R. Chen,
Y. Tang, Eur. J. Med. Chem. 46 (2011) 4212e4218.
[48] L. Tabernero, A.R. Aricescu, E.Y. Jones, S.E. Szedlacsek, FEBS J. 275 (2008)
867e882.
[49] G. Scapin, S.B. Patel, J.W. Becker, Q. Wang, C. Desponts, D. Waddleton,
K. Skorey, W. Cromlish, C. Bayly, M. Therien, J.Y. Gauthier, C.S. Li, C.K. Lau,
C. Ramachandran, B.P. Kennedy, E. Asante-Appiah, Biochemistry 42 (2003)
11451e11459.
[50] Y.A. Puius, Y. Zhao, M. Sullivan, D.S. Lawrence, S.C. Almo, Z.Y. Zhang, Proc. Natl.
Acad. Sci. USA 94 (1997) 13420e13425.
[51] P. Cirri, T. Fiaschi, P. Chiarugi, G. Camici, G. Manao, G. Raugei, G. Ramponi,
J. Biol. Chem. 271 (1996) 2604e2607.
[52] A.P.R. Zabell, A.D. Schroff, B.E. Bain, R.L. Van Etten, O. Wiest, C.V. Stauffacher,
J. Biol. Chem. 281 (2006) 2604e2607.
[53] R. Maccari, P. Paoli, R. Ottanà, M. Jacomelli, R. Ciurleo, G. Manao, T. Steindl,
T. Langer, M.G. Vigorita, G. Camici, Bioorg. Med. Chem. 15 (2007) 5137e5149.
[54] R. Maccari, R. Ottanà, R. Ciurleo, P. Paoli, G. Manao, G. Camici, C. Laggner,
T. Langer, ChemMedChem 4 (2009) 957e962.
[55] R. Ottanà, R. Maccari, R. Ciurleo, P. Paoli, M. Jacomelli, G. Manao, G. Camici,
C. Laggner, T. Langer, Bioorg. Med. Chem. 17 (2009) 1928e1937.
[56] R. Ottanà, R. Maccari, M.L. Barreca, G. Bruno, A. Rotondo, A. Rossi, G. Chiricosta,
R. Di Paola, L. Saubetin, S. Cuzzocrea, M.G. Vigorita, Bioorg. Med. Chem. 13
(2005) 4243e4252.
[57] G. Bruno, L. Costantino, C. Curinga, R. Maccari, F. Monforte, F. Nicolò, R. Ottanà,
M.G. Vigorita, Bioorg. Med. Chem. 10 (2002) 1077e1084.
[58] Y. Momose, K. Meguro, H. Ikeda, C. Hatanaka, S. Oi, T. Sohada, Chem. Pharm.
Bull. 39 (1991) 1440e1445.
[59] T. Ispida, Y. In, M. Inoue, C. Tanaka, N. Hamanaka, J. Chem. Soc. Perkin Trans. 2
(1990) 1085e1091.
[60] U. Iannaccone, A. Bergamaschi, A. Magrini, G. Marino, N. Bottini, P. Lucarelli,
E. Bottini, F. Gloria-Bottini, Metab. Clin. Exp. 54 (2005) 891e894.
[61] H.L. Berman, J. Westbrook, Z. Feng, G. Gilliland, T.N. Bhat, H. Weissig,
I.N. Shindyalov, P.E. Bourne, Nucl. Acids Res. 28 (2000) 235e242.
[62] D.P. Wilson, Z.-K. Wan, W.-X. Xu, S.J. Kirincich, B.C. Follows, D. Joseph-
McCarthy, K. Foreman, A. Moretto, J. Wu, M. Zhu, E. Binnun, Y.-L. Zhang,
M. Tam, D.V. Erbe, J. Tobin, X. Xu, L. Leung, A. Shilling, S.Y. Tam, T.S. Mansour,
J. Lee, J. Med. Chem. 20 (2007) 4681e4698.
[63] M. Zhang, C.V. Stauffacher, D. Lin, R.L. Van Etten, J. Biol. Chem. 273 (1998)
21714e21720.
[64] A.P.R. Zabell, A.D. Schroff Jr., B.E. Bain, R.L. Van Etten, O. Wiest,
C.V. Stauffacher, J. Biol. Chem. 281 (2006) 6520e6527.
References
[1] A. Alonso, J. Sasin, N. Bottini, I. Friedberg, A. Osterman, A. Godzik, T. Hunter,
J. Dixon, T. Mustelin, Cell 117 (2004) 699e711.
[2] T. Hunter, Cell 100 (2000) 113e127.
[3] B.G. Neel, N.K. Tonks, Curr. Opin. Cell Biol. 9 (1997) 193e204.
[4] N.K. Tonks, Nat. Rev. Mol. Cell Biol. 7 (2006) 833e846.
[5] S. Koren, I.G. Fantus, Best Pract. Res. Cl. En. 21 (2007) 621e640.
[6] N. Dubè, M.L. Tremblay, Biochim. Biophys. Acta 1754 (2005) 108e117.
[7] A. Ostman, C. Hellberg, F.D. Böhmer, Nat. Rev. Cancer 6 (2006) 307e320.
[8] V.V. Vintonyak, A.P. Antonchick, D. Rauh, H. Waldmann, Curr. Opin. Chem.
Biol. 13 (2009) 272e283.
[9] J. Zang, P.L. Yang, N.S. Gray, Nat. Rev. Cancer 9 (2009) 28e39.
[10] V.R.H. Huijsduijnen, A. Bombrum, D. Swinnen, Drug Discov. Today 7 (2002)
1013e1019.
[11] M.A.T. Blaskovich, Cur. Med. Chem. 16 (2009) 2095e2176.
[12] P. Heneberg, Curr. Med. Chem. 16 (2009) 706e733.
[13] V.V. Vintonyak, H. Waldmann, D. Rauh, Bioorg. Med. Chem. 19 (2011)
2145e2155.
[14] M. Elchebly, P. Payette, E. Michaliszyn, W. Cromlish, S. Collins, A.L. Loy,
D. Normandin, A. Cheng, J. Himms-Hagen, C. Chan, C. Ramachandran,
M.J. Gresser, M.L. Tremblay, B.P. Kennedy, Science 283 (1999) 1544e1548.
[15] L.D. Klaman, O. Boss, O.D. Peroni, J.K. Kim, J.L. Martino, J.M. Zabolotny,
N. Moghal, M. Lubkin, Y.B. Kim, A.H. Sharpe, A. Stricker-Krongrad,
G.I. Shulman, B.G. Neel, B.B. Kahn, Mol. Cell. Biol. 20 (2000) 5479e5489.
[16] N.K. Tonks, FEBS Lett. 546 (2003) 140e148.
[17] G. Liu, Curr. Opin. Mol. Ther. 6 (2004) 331e336.
[18] A. Cheng, N. Uetani, P.D. Simoncic, V.P. Chaubey, A. Lee-Loy, J. McGlade,
B.P. Kennedy, M.L. Tremblay, Dev. Cell 2 (2002) 497e503.
[19] J.M. Zabolotny, K.K. Bence-Hanulec, A. Stricker-Krongrad, F. Haj, Y. Wang,
Y. Minokoshi, Y.B. Kim, J.K. Elmquist, L.A. Tartaglia, B.B. Kahn, B.G. Neel, Dev.
Cell 2 (2002) 489e495.
[20] T. Tiganis, A.M. Bennett, Biochem. J. 402 (2007) 1e15.
[21] K.A. Kenner, E. Anyanwu, J.M. Olefsky, J. Kusari, J. Biol. Chem. 33 (1996)
19810e19816.
[22] B.A. Zinker, C.M. Rondinone, J.M. Trevillyan, R.J. Gum, J.E. Clampit, J.F. Waring,
N. Xie, D. Wilcox, P. Jacobson, L. Frost, P.E. Kroeger, R.M. Reilly, S. Koterski,
T.J. Opgenorth, R.G. Ulrich, S. Crosby, M. Butler, S.F. Murray, R.A. McKay,
S. Bhanot, B.P. Monia, M.R. Jirousek, Proc. Natl. Acad. Sci. USA 17 (2002)
11357e11362.

R. Ottanà et al. / European Journal of Medicinal Chemistry 50 (2012) 332e343
343
[65] G. Jones, P. Willett, R.C. Glen, J. Mol. Biol. 245 (1995) 43e53.
[66] G. Wolber, T. Langer, J. Chem. Inf. Model. 45 (2005) 160e169.
[67] G. Jones, P. Willett, R.C. Glen, A.R. Leach, R. Taylor, J. Mol. Biol. 267 (1997)
727e748.
[68] Molecular operating environment (MOE), version 2010.20, Chemical
Computing group, 1010 Sherbrooke St. W, Suite 910, Montreal, Quebec, Canada.
[69] T. Seidel, G. Ibis, F. Bendix, G. Wolber, Drug Discov. Today Tecnologies 7 (2010)
e221ee228.
[70] G. Wolber, A.A. Dornhofer, T. Langer, J. Comput. Aid. Mol. Des 20 (2006)
773e788.
[71] R. Marzocchini, M. Bucciantini, M. Stefani, N. Taddei, M.G.M.M. Thunnissen,
P. Nordlund, G. Ramponi, FEBS Lett. 426 (1998) 52e56.
[72] U.K. Laemmli, Nature 227 (1970) 680e685.
[73] FigSysÒ 2003, A Graphing, Charting and Data Analysis Software Package, is
produced by BIOSOFTÒ, 2D Dolphin Way, Stapleford Cambridge, CB2 5DW,
UK.