Synthesis, activity and molecular modeling of a new series of chromones as low molecular weight protein tyrosine phosphatase inhibitors

Article abstract of DOI:10.1016/j.bmc.2009.02.060

Protein tyrosine phosphatases (PTP) are crucial elements in eukaryotic signal transduction. Several reports suggested that the LMW-PTP family has oncogenic relevance. Moreover, LMW-PTP has been recognized as a negative regulator of insulin-mediated mitotic and metabolic signaling. Thus, inhibition of the LMW-PTP can be considered an attractive approach for the design of new therapeutic agents for the treatment of type II diabetes and for new antitumoral drugs. To date very few (and weak) inhibitors of LMW-PTP have been identified. On the basis of the reported weak activity of some flavonoids on phosphatases, we discovered a lead that originated a new class of highly active LMW-PTP inhibitors; these compounds inhibit also PTP-1B and are active in cellular assays. Docking experiments and SAR highlighted the possible binding mode of these compounds to the enzyme, putting the background for the future optimization of their inhibitory activity and selectivity towards the closely related enzyme PTP-1B.

Full text of DOI:10.1016/j.bmc.2009.02.060

Bioorganic & Medicinal Chemistry 17 (2009) 2658–2672
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis, activity and molecular modeling of a new series of chromones
as low molecular weight protein tyrosine phosphatase inhibitors
Marco Forghieri a, Christian Laggner b, Paolo Paoli c, Thierry Langer d, Giampaolo Manao c,
Guido Camici e, Lucia Bondioli f, Fabio Prati a, Luca Costantino f,
*
a Department of Chemistry, University of Modena and R.E., Via G. Campi 183, 41100 Modena, Italy
b Department of Pharmaceutical Chemistry, Institute of Pharmacy and Center for Molecular Biosciences (CMBI), University of Innsbruck, Innrain 52c, A-6020 Innsbruck, Austria
c Department of Biochemical Sciences, University of Florence, Viale Morgagni 50, 50134 Florence, Italy
d Prestwick Chemical Inc., Bld. Gonthier d’Andernach, 67400 Strasbourg-Illkirch, France
e Center of Excellence for Scientific Research DENOThe, Viale Morgagni 44, 50134 Florence, Italy
f Department of Pharmaceutical Sciences, University of Modena and R.E., Via G. Campi 183, 41100 Modena, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Protein tyrosine phosphatases (PTP) are crucial elements in eukaryotic signal transduction. Several
reports suggested that the LMW-PTP family has oncogenic relevance. Moreover, LMW-PTP has been rec-
ognized as a negative regulator of insulin-mediated mitotic and metabolic signaling. Thus, inhibition of
the LMW-PTP can be considered an attractive approach for the design of new therapeutic agents for
the treatment of type II diabetes and for new antitumoral drugs. To date very few (and weak) inhibitors
of LMW-PTP have been identified. On the basis of the reported weak activity of some flavonoids on phos-
phatases, we discovered a lead that originated a new class of highly active LMW-PTP inhibitors; these
compounds inhibit also PTP-1B and are active in cellular assays. Docking experiments and SAR high-
lighted the possible binding mode of these compounds to the enzyme, putting the background for the
future optimization of their inhibitory activity and selectivity towards the closely related enzyme PTP-1B.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 3 January 2009
Revised 25 February 2009
Accepted 28 February 2009
Available online 5 March 2009
Keywords:
Low molecular weight protein tyrosine
phosphatases
Protein tyrosine phosphatase 1B
Diabetes
Chromones
Antitumoral drugs
1. Introduction
catalysis pathway, the cysteinyl residue of the signature sequence
executes a nucleophilic attack upon the phosphate moiety of the
Tyrosine phosphorylation and dephosphorylation are crucial
elements in eukaryotic signal transduction.1–3 One major group
of enzymes responsible for hydrolyzing the phosphotyrosyl resi-
dues of protein substrates is that of protein tyrosine phosphatases
(PTPs).
substrate, leading to the formation of a thio-phosphate intermedi-
ate. Two isoenzymes of the LMW-PTP, HPTP-A and HPTP-B, corre-
sponding to the fast and the slow electrophoretic forms, are
expressed in humans. They are produced by an alternative splicing
mechanism, and thus display identical sequences except for resi-
dues 40–73.7 Couples of LMW-PTP isoenzymes have been isolated
from other vertebrate sources, termed isoform-1 (IF1) and isoform-
2 (IF2), respectively. The differing amino acid sequence as well as
the substrate specificity and the differing sensitivity to inhibitors
suggest that the two isoenzymes have different function in the cell.
Several reports in the literature have suggested that the LMW-
PTP family has oncogenic relevance. In particular, Chiarugi et al. re-
ported that the LMW-PTP oncogenic potential is mediated by its
EphA2 tyrosine dephosphorylating activity and that dephosphoryl-
ated EphA2 overexpression is highly associated with most human
cancers.8 The overexpression of LMW-PTP is sufficient to transform
epithelial cells.9 High levels of LMW-PTP has been detected in
some types of human tumors.9,10 Moreover, LMW-PTP has been
recognized as a negative regulator of insulin-mediated mitotic
and metabolic signalling.11,12 Thus, inhibition of the LMW-PTP
can be considered an attractive approach for the design of new
The PTP superfamily is composed by at least three families:4,5 (i)
‘classical PTPs’, which exist both as transmembrane forms (such as
LAR or PTP-a) and non transmembrane forms (such as PTP-1B or
TC-PTP). They share at least one conserved catalytic domain of
240 residues and are phosphotyrosine specific; (ii) the dual-speci-
ficity phosphatases, which are able to dephosphorylate both phos-
photyrosine and phosphothreonine in specific sequence contexts;
(iii) the low molecular weight phosphotyrosine protein phospha-
tases (LMW-PTPs) that have a molecular mass of 18 kDa, and are
phosphotyrosine specific. A minimal signature sequence for the
phosphate binding site, CXXXXXR, is highly conserved among all
the PTPs.6 They share a common catalytic mechanism: in the
* Corresponding author. Tel.: +39 059 2055125; fax: +39 059 2055131.
E-mail address: luca.costantino@unimore.it (L. Costantino).
0968-0896/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2009.02.060

M. Forghieri et al. / Bioorg. Med. Chem. 17 (2009) 2658–2672
2659
therapeutic agents for the treatment of type II diabetes and for new
antitumoral drugs.
HO
OH
OH
OH
HO
O
Another PTP, PTP-1B, is involved in diabetes. This phosphatase
negatively regulates insulin receptor and insulin receptor sub-
strate-1 phosphorylation. Mice that lack the PTP-1B gene have in-
creased insulin sensitivity with resistance to weight gain on a high
fat diet and are otherwise normal. This enzyme is object of an in-
tense search for new compounds able to treat type II diabetes
and obesity.13
HO
O
O
OH
OH
OH
O
OH
Morin
Quercetin
LMW-PTP possesses a phosphate binding loop that has both a
similar amino acid sequence (CLGNICR vs CSAGIGR) and a virtually
identical backbone structure as that of PTP-1B. Apart from this
short sequence, no sequence homology between high and low
molecular weight PTPs has been found. The rest of the two binding
sites is build up by different structural elements, although both
binding sites share one aspartic acid and one tyrosine (or trypto-
phane in the case of LMP-PTP isoform II) residue at similar posi-
tions. Figure 1 shows an overlay of human LMW-PTP isoform I
with human PTP-1B (PDB structures 5PNT14 and 1G7G15). A short
discussion about binding site similarities is given in Ref. 14.
To date very few (and weak) inhibitors of LMW-PTP have been
found; among these, a series of thiazolidinediones were shown to
be able to inhibit LMW-PTP besides their primary target PTP-
1B.16 Moreover, a virtual screening method selected, from a data-
base of almost 500,000 commercially available compounds, a sub-
set of 34 compounds; nine of these were experimentally confirmed
to be competitive inhibitors of bovine LMW-PTP (this enzyme
shares a 94% identity with human LMW-PTP-B), two of them show-
Chart 1. Structures of quercetin and morin.
at positions C30 and C40 (quercetin) or on C20 and C40 (morin) of the
2-phenyl residue (Chart 1).18 While the flavonoids were tested in a
screening program, phosphonic acids were designed on the basis of
the activation of HCPTP-B and inhibition of HCPTP-A exerted by
adenine.20
It is also known that flavonoids are inhibitors of cyclic AMP
phosphodiesterase; SAR studies together with the observed com-
petitive kinetics for cAMP suggests that flavonoids compete with
cAMP for a nucleotide binding site at which stacking occurs.21
Quantum chemical studies showed that the charge distribution
of the pyrone ring of flavone resembles that of the pyrimidine ring
of the adenine present in cAMP.21 Both ring systems are planar,
aromatic, and of similar size. Furthermore, we found that 3-hydro-
xy-flavones and adenine residues often share similar positions and
hydrogen bonding patterns in certain kinases (phosphoinositide 3-
kinase, CDK2, and Pim-1; compare PDB structures 1E8W with
1E8X, 2DUV with 1HCK, and 2O3P with 1YI4), with the 3-hydroxy
and 4-carbonyl oxygen atoms of the flavones aligning with the 6-
amino nitrogen atom and the ring nitrogen atom at position 1 of
the adenine structures, respectively.
ing Ki values around 10 l
M.17 No data were reported for PTP-1B
inhibition exerted by these compounds. Other inhibitors with
IC50 values in the millimolar range include flavones,18 the buffer
constituent HEPES19 and phosphonic acids.20 The flavonoids were
tested on the bovine kidney LMW-PTP, and the results obtained
highlighted the importance of the position of the hydroxy groups
in the structure for their activity—in particular the hydroxy groups
Among the tested flavonoids, quercetin behaves as an activator
of the enzyme. It is yet unclear whether the mechanism is similar
to that proposed by Wang for the activation of LMW-PTP by ade-
nine, where a water molecule participates at the active site to im-
prove catalysis.22 Furthermore, quercetin and related compounds
have been reported to inhibit the closely related enzyme PTP-1B
(IC50 = 23.3 l
M).23
On the basis of these findings, we decided to screen several in-
house flavonoids with in vitro enzyme inhibition assays against
both LMW-PTP isoforms in order to detect a lead for further opti-
mization. This led to the discovery of compound 1 (Table 1) which
in turn led to the preparation and testing of a small initial series of
compounds (2–8, Table 1).
Docking studies were then conducted in order to understand
the binding mode of these compounds. On the basis of the initial
results obtained, two series of compounds (Chart 2, compounds
9–15 and 18–20 of general formula A, (Tables 1 and 2) and 16,
17, general formula B, Table 1) were prepared and their inhibitory
activity against the two LMW-PTP isoforms assessed. The results
from the binding assays were then used to reevaluate the docking
results in order to gain insight into their possible binding modes,
allowing the development of more active compounds in the future.
Moreover, the activity of representative compounds of this class
were tested for their ability to inhibit the phosphatases in cells,
and thus activate the insulin signaling pathway.
2. Results and discussion
2.1. Chemistry
The substituted 4H-1-benzopyran-4-one derivatives 1,5–7 were
synthetized from the appropriate 4-substituted-20-hydroxy-40-
phenyldihydrochalchones 23a–b by reaction with the appropriate
ester in the presence of NaH/pyridine (Scheme 1), following the
Figure 1. Overlay of ligand binding sites of human LMW-PTP isoform 1 (PDB code
5PNT; yellow and red) and human PTP-1B (1G7G; cyan, blue and purple). The
phosphate binding loop and the conserved Tyr and Asp residues are marked in red
and blue, respectively. Conserved residues are labeled.

2660
M. Forghieri et al. / Bioorg. Med. Chem. 17 (2009) 2658–2672
Table 1
Structure and enzyme inhibition data of compounds 1–17
2
R7
R6
O
O
n
4'
OH
R3
Compound
R6
R7
R3
n
IF I,
l
M (SD)a
IF II,
l
M (SD)a
PTP-1B, l
M (SD)a
1
2
3
H
H
H
OH
OH
OH
0
1
1
5.7 (0.1)
13.9 (0.4)
4% inh. at 15
0% inh. at 15
0% inh. at 15
lM
lM
lM
H
10% inh. at 15
10% inh. at 15
l
M
M
10% inh. at 15
10% inh. at 15
l
M
M
l
l
4
5
6
7
8
H
OH
OH
OH
H
1
2
3
1
0
5.3 (0.2)
17.5 (0.5)
17.3 (0.4)
10% inh. at 15
17.2 (0.4)
18.4 (0.1)
10% inh. at 15
20% inh. at 15
l
l
M
H
6.5 (0.2)
M
H
10% inh. at 15
8.6 (0.4)
lM
lM
0% inh. at 15
20% inh. at 15
17.0 (0.1)
lM
OH
H
lM
OH
8.2 (0.4)
OH
O
9
10
11
H
H
H
OH
OH
H
0
0
0
11.4 (0.1)
7.4 (0.1)
26.8 (0.6)
12.4 (0.1)
12.6 (0.1)
6.2 (0.1)
7.7 (0.1)
5.0 (0.2)
OH
OH
OH
OH
14.3 (0.4)
OH
OH
12
13
H
H
OH
H
0
0
1.3 (0.02)
0.7 (0.05)
2.2 (0.1)
1.0 (0.3)
OH
O
OH
5.7 (0.04)
0.9 (0.04)
OH
O
OH
14
H
OH
0
60% inh. at 20
lM
25% inh. at 20
lM
50% inh. at 20 lM
OH

M. Forghieri et al. / Bioorg. Med. Chem. 17 (2009) 2658–2672
2661
Table 1 (continued)
Compound
R6
H
R7
R3
n
IF I,
l
M (SD)a
IF II,
l
M (SD)a
PTP-1B,
l
M (SD)a
O
OH
OH
15
16
OH
0
5 % inh. at 20
l
M
10% inh. at 20
l
M
50% inh. at 20 lM
OH
OH
OH
OH
0
0
1.1 (0.10)
3.0 (0.06)
2.0 (0.08)
17
H
25% inh. at 20
l
M
10% inh. at 20
lM
10% inh. at 20 lM
a
The enzyme inhibition data are expressed as IC50 values (SD, Standard deviation) or % inhibition at a given concentration.
R
OH
General Formula A
9-15, 18-20
HO
O
O
R
8
R
HO
HO
O
O
General Formula B
R
16, 17
OH
Chart 2. Development of compounds of general formula A and B.
general procedure described in literature.24 The intermediate 1,3-
diketones were then cyclized with acetic acid/sulphuric acid, to ob-
tain 24a–d. Then the compounds were deprotected to give 1,5–7
by treatment with BBr3. Owing to the very low yield obtained in
the presence of the dihydrochalchone 23e following Route A, a dif-
ferent synthetic procedure (Route B) was followed for compounds
8–16. The dihydrochalchones 23c–f were acylated in the presence
of trietylamine and dimetylaminopyridine (25a–d). Claisen con-
densation with NaH in DMSO at rt and subsequent cyclization with
acetic acid/sulphuric acid afforded the corresponding compounds
27a–d. The bromo derivatives 27b–d were subjected to Suzuki
coupling with (substituted)phenylboronic acids in order to obtain
the biphenyl derivatives 28a–h. The fully protected derivatives
were deprotected with the aid of BBr3 to obtain 8–16. Suzuki cou-
pling, followed by BBr3 deprotection was also applied for the syn-
thesis of compounds 18–20 (Scheme 2).
1 showed to be somewhat selective against PTP-1B (Ki = 2.5 lM
for LMW-PTP A-form (Isoform I, IF I), Ki = 29.1 lM for PTP-1B,
Table 3).
The lead was then modified at the chain at position 2 (com-
pounds 4–6) and at the position of the hydroxy group at position
7 (compound 7) of the benzopyrone nucleus. The highest activity
and selectivity towards PTP-1B was observed for n = 0 and 1, while
no difference in activity was observed for the change of position of
the hydroxy group from position 7 to 6 (compound 4 and 7, respec-
tively). Moreover, the introduction of a hydroxy group at position
40 of the biphenyl moiety (compound 8) influenced the inhibitory
activity of compound 1 only slightly.
2.3. Molecular modeling
To gain some insight into the possible binding modes of our
compounds, molecular docking simulations were performed. Since
the binding site of the LMW-PTPs is lined by multiple side-chain
residues that possess some conformational flexibility, we chose
the Induced-Fit Docking (IFD) method by Schrödinger, LLC.25 Initial
docking experiments of compounds 1, 4, 7 and 8 in the active site
of LMW-PTP, A-form or Isoform I (PDB code 5PNT14), suggested
that these compounds can bind to the enzyme in two binding
modes, depending on the presence of the hydroxyl group at 40 po-
sition of the biphenyl moiety: while in the obtained poses for com-
pounds 1, 3 and 7, the chromanone ring was located inside the
binding pocket, for compound 8 the biphenyl was placed inside
the catalytic pocket. Stacking was predicted to occur between Tyr
2.2. Initial screens
Screening of an in-house collection of natural and synthetic
flavonoids led to the discovery of compound 1 as a lead for the
development of LMW-PTP inhibitors. This compound was the only
one that we found active, and it was the only one among the com-
pounds examined that possesses a biphenyl moiety at position 3 of
the flavone ring. The importance of this moiety for the enzyme
inhibition was checked by the examination of compounds 2 and
3; their lack of activity underlines the role of this moiety on the
PTPases inhibitory activity of compound 1. Moreover, compound

2662
M. Forghieri et al. / Bioorg. Med. Chem. 17 (2009) 2658–2672
O
O
7
6
n
HO
OH
1) (c,d): 24a-d
2) (e): 1,5-7
Route A (from 23a,23b)
4
OH
O
R
4'
4'
5'
4
5
OH
O
OH
O
4
3
a
R
b
R
R
R
3
5'
21a: R4 = OCH3
21b: R5 = OCH3
22a:R4' =OCH3; R4 = C6H5
22b: R5' =OCH3; R4 = C6H5
23a: R4' =OCH3; R4 = C6H5
23b: R5' =OCH3; R4 = C6H5
21c: R4 = R5 = OCH3
21d: R = H
22c: R4' = R5' = OCH3; R4 = C6H5
22d: R4' = OCH3; R3 = Br
22e: R4' = OCH3;R4 = Br
22f: R = H, R4 = Br
23c: R4' = R5' = OCH3; R4 = C6H5
23d: R4' = OCH3; R3 = Br
23e: R4' = OCH3;R4 = Br
23f: R = H; R4 = Br
Route B (from 23c-f)
f
3
OCH3
4'
OH
O
R
4
R
OCH3
O
5'
7
O
O
4'
5'
O
O
O
4
3
d
R
R
g
R
6
H3CO
25a: R4' = R5' = OCH3, R4 =
C6H5 (from 23c)
3'
R
26a: R4' = R5' = OCH3, R4 = C6H5 (from 25a)
26b: R4' = OCH3, R3 = Br (from 25b)
26c: R4'= OCH3, R4 = Br (from 25c)
26d: R = H, R4= Br (from 25d)
4'
27a: R6 = R7 = OCH3, R4' = C6H5
(from 26a)
25b: R4' = OCH3, R3 = Br
(from 23d)
27b: R7 = OCH3, R3' = Br (from 26b)
27c: R7 = OCH3, R4' = Br (from 26c)
27d: R = H; R4'=Br (from 26d)
25c: R4' = OCH3, R4 = Br
(from 23e)
OH
25d: R = H; R4 = Br (from
23f)
7
O
O
HO
e
6
1) (h): 28 a-h
2) (e): 8-15
3'
16 (from 27a)
4'
3"
R
4"
Scheme 1. Synthesis of compounds 1,5–16: (a) (substituted)benzaldehyde, Ba(OH)2, methanol, reflux; (b) Pd/C, EtOAc; (c) R-COOCH3, NaH, Py; (d) CH3COOH/H2SO4; (e)BBr3;
(f) 4-methoxybenzoyl chloride, TEA/DMAP; (g) NaH, DMSO; (h) Suzuki coupling, (substituted)phenylboronic acid.
49 and the aromatic residues that were not inside the binding
pocket (details not shown).
OH
B
2
We noticed, however, that the results obtained depended
strongly on the presented protonation state of the docked com-
pounds. Most of the PTP binding molecules have acidic groups like
carboxylic acid moieties that have pKa values well below 4.5. How-
ever, Malamas et al. reported micromolar IC50 values against PTP-
1B for a series of benzofurane and benzothiophene biphenyls,
including compounds 31–34 (Chart 3) which bear the same biphe-
nyl residue as compound 8 and lack any other acidic group.26 For
our 7-hydroxy flavones, a pKa of about 7.0–7.5 can be assumed27
3
a
HO
2
R
R
4
3
4
29a-c
30a-c
b
18-20
Scheme 2. Synthesis of compounds 18-20: (a) Suzuki coupling, C6H5Br; (b) BBr3.

M. Forghieri et al. / Bioorg. Med. Chem. 17 (2009) 2658–2672
2663
Table 2
nature of the binding site, in addition to the pH value of the
cytosol.
Structure and enzyme inhibition data of compounds 18–20
Available X-ray structures of PTPs showed that the residues of
the phosphate binding loop are able to make multiple hydrogen
bonds to oxygen atoms of phophate or sulfonate groups at the bot-
tom of the pocket. Based on these results, we docked both 6,7-
dihydroxy flavones and 30,40-dihydroxy biphenyl derivatives,
which were both predicted to bind favorably into the binding site.
Thus, on the basis of the docking results, we decided to synthet-
ize and test the biphenyl derivatives of general formula A (Chart 2)
that might be able to bind to the enzyme active site via the biphe-
nyl moiety. Based on the docking results, we prepared compounds
bearing one or two hydroxy and sometimes also one carboxy group
at positions 20, 30 and 40, resulting in compounds 9–15 and the
biphenyl derivatives 18–20 lacking the flavone residue (Tables 1
and 2, resp.). Moreover, to investigate the possible binding mode
via the flavone moiety, we prepared the chromones 16 and 17 of
general formula B (Chart 2, Table 1).
R2
R3
R4
IF Ia
a
Compound R2
R3
R4
IF II
PTP-1Ba
18
19
20
H
OH OH
0% inh. at
0% inh. at
20% inh. at
20
OH COOH 0% inh. at
20
0% inh. at
20
l
M
20
0% inh. at
20
0% inh. at
20
l
M
20
5% inh. at
20
10% inh. at
20
lM
H
l
M
l
M
lM
OH
H
OH
l
M
l
M
lM
a
Enzyme inhibition data are expressed as % inhibition at a given concentration.
Among the compounds examined, compounds 12 and 16
showed to be the best, with an inhibitory activity (IC50) towards
LMW-PTP of around 1 lM. However, all these new compounds
Table 3
Ki values of selected compounds
are able to inhibit LMW-PTPs and PTP-1B on approximately the
same extent. Lack of activity of compounds 18–20 on all the three
enzymes underlines the fact that the biphenyl, while it is very
important for the activity of compound 1, is by itself not sufficient
to provide binding of the compounds. The binding affinities of 7–
10 towards LMW-PTPs are virtually the same, suggesting a binding
pose where substitution at the distal phenyl ring has little effect on
the binding affinity, while a decreasing selectivity towards PTP-1B
could be observed for every additional hydroxy or carboxy group.
On the other hand, introduction of a carboxylic group at position
R04 of compound 10 (compound 12) gave an increase in affinity
by a factor of more than 6, albeit in connection with an even higher
affinity for PTP-1B. Compounds 14 and 15 showed only weak
affinities.
As predicted by the docking study, the introduction of the hy-
droxy group at position 6 of compound 1 (compound 16) increased
the inhibitory activity by almost an order of magnitude, but, simi-
larly to compound 12, with lack of selectivity towards PTP-1B. Even
in this case, the biphenyl moiety showed to be necessary for the
activity since compound 17, lacking this group, is practically de-
void of inhibitory activity.
Compound
IF I,
l
M (S.D.)a
IF II,
l
M (S.D.)a
PTP-1B, l
M (S.D.)a
1
10
16
2.5 (0.1)b
7.8 (0.7)b
29.1 (2.2)b
N.D.d
2.23 (0.3)b
0.27 (0.03)c
2.68 (0.1)b
0.34 (0.06)c
N.D.d
a
SD, Standard deviation.
Mixed type inhibition.
Competitive inhibition.
ND, Not determined.
b
c
d
and an even higher value of 8–10, depending on possible conjuga-
tion with the flavone ring, for the phenolic hydroxy groups at the
phenyl and biphenyl residues. Thus, at physiological pH, these res-
idues can be expected to be only partially deprotonated. On the
other hand, we recently performed studies on aldose reductase
(ALR2) inhibitors where the assumption of deprotonated phenolic
groups led to reasonable binding poses,28 and was in good agree-
ment with high-resolution X-ray crystal structures that suggested
a hydantoin-derivative with a calculated pKa of about 9.3 to be
deprotonated at the ALR2 binding site.29 These results showed that
the protonation state of a ligand may be strongly influenced by the
X
O
R2
O
R1
X
R1
H
R2
IC50 (µM) PTP-1B
31
32
33
34
-
nBu
0.74
-
H
4-OH-Bz 1.08
CH2
H
nBu
1.19
2.19
-
CH2COOH nBu
Chart 3. Inhibitory activity of some biphenyls towards PTP-1B.26

2664
M. Forghieri et al. / Bioorg. Med. Chem. 17 (2009) 2658–2672
2.4. Docking studies
that are not conjugated with the flavone ring. We decided to dock
the singly and—if a second pKa value below 9.0 could be expected—
doubly deprotonated form for the investigated compounds, with
the exception of 18, for which, because of its low acidity, we
docked both the neutral and singly deprotonated form. A negative
charge of a deprotonated 7-hydroxy flavone group can also be
transferred to the oxygen atom at position 3 via mesomerism,
but since these structures could not meet the main requirement
of known inhibitors—a negative charge at the bottom of the narrow
binding site—such mesomeric forms were not further investigated.
PDB structures 5PNT (human LMW-PTP isoform I14) and 1XWW
(human LMW-PTP isoform II32) were selected as template struc-
tures for the docking calculations. Overlay of the two isoforms
showed that the phosphate binding site is identical in both pro-
teins. The nearest amino acid variation which may thus affect li-
gand selectivity is at the opening of the binding site at position
49 (Tyr/Trp for isoform I/II), followed by 50 (Glu/Asn) and 53
(Asn/Arg). The selected protein structures were processed in Mae-
stro (Schrödinger, LLC)25 using the Protein Preparation Wizard,
which performs the following steps: assigning of bond orders,
addition of hydrogens, optimization of hydrogen bonds by flipping
amino side chains, correction of charges, and minimization of the
protein complex. All water molecules were removed from the pro-
tein structure. The IFD protocol developed by Schrödinger, LLC, was
used for docking our compounds. This method takes into account
the flexibility of the binding site. The method is described in more
detail elsewhere.28,33 This approach seemed reasonable because of
the flexible side chains of multiple residues reaching into the bind-
ing site, especially those of Leu 13, Ile 16, Tyr/Trp 49, Tyr 131, and
Tyr 132. The size of the box, into which the compounds are docked,
is usually centred around the selected co-crystallized ligand. In the
case of 5PNT and 1XWW we had (a) two co-crystallized ligands (2-
(N-morpholino)-ethanesulfonic acid (MES) and a sulfate ion) of dif-
ferent sizes which (b) were both considerably smaller than most of
the investigated ligands. To use comparably sized binding sites
that can accomodate all of our ligands, we thus chose to overlay
both proteins and to copy the ligand MES from 5PNT into
1XWW. Both binding site boxes where then defined by the center
of the MES ligand, with an enclosing box size set to 30 Å. All other
values of the IFD protocol where left at their default values.
For each compound, the poses obtained for the two different
isoforms I and II were highly similar. This is in good agreement
with the rather constant IC50 and Ki ratio of 1.7–3.12 (Tables 1
and 3) between the affinities for isoforms I and II, suggesting that
all investigated compounds share similar interactions with the
amino acid residues that are different between the isoforms. One
pose was predominantly found for most of our compounds (1, 4,
8, 10, 14, and 16) and is depicted in Figure 2 (top row), exemplified
for compound 16 in 5PNT. The flavone ring is buried inside the
binding pocket, with the deprotonated 7-hydroxy group of the fla-
vone moiety forming a charged interaction with the protonated
Arg 18 residue, accepting hydrogen bonds from the same group
as well as from the backbone NH group of Leu 13.
Compounds 1, 4, 8, 10, 12, 14, 16 and 18 were chosen for dock-
ing to give a representative sampling of our series. Additionally,
quercetin and morin (Chart 1) were included.
The obtained binding profiles of the initial set of compounds
suggested that binding via the 400-hydroxyl group of the biphenyl
moiety did not play a role (compounds 1 and 8 possess the same
IC50 values, see Table 1), which appeared to be in contrast to the
results obtained by Malamas et al. discussed above.26
Thus, given the great importance in the acidity of the inhibitors
of the closely related PTP-1B for the enzyme inhibition, we calcu-
lated the pKa values of the different hydroxy and carboxy groups,
using the freely accesible web tool SPARC on line calculator
v4.130 to gain some insights about the acidity of the different res-
idues of the docked compounds.
This tool has been previously evaluated against experimental
pKa values of eight flavonoids—including the compounds quercetin
and morin—obtained by capillary electrophoresis, and has proved
to be quite successful for most predictions, with the exception that
the lowest pKa value of morin was predicted too high by 1.87
units.31 It should be noted that all values were calculated with
the intention to find out which group would be the most acidic,
thus on the assumption that no other residues were already depro-
tonated in the molecule. Depending on the distance between the
investigated residues, this simplification may have either a small
or a big effect on the pKa values of additional groups: For com-
pound 12, deprotonation of the most acidic carboxylic acid group
does not influence the calculated value for the 7-OH position of
the flavone ring (7.36) but strongly changes the value for the
neighboring hydroxy group at 300 (13.24 vs 7.31). A summary of
the obtained values is given in Table 4.
The obtained values suggest that the most acidic group is—ex-
cept for the carboxylic acid group in 12—always located at position
7 of the flavone moiety. The next groups to be deprotonated would
be hydroxy groups at 2-phenyl residues, preferentially at positions
20 or 40, followed by hydroxy groups at phenyl or biphenyl groups
Table 4
Calculated and experimental pKa values for the determination of the most acidic
groups of selected docking compounds, obtained with the SPARC on line calculator30
Compound
Flavone ring
Residues at
position 2
Residues at
position 3
1
4
8
10
7-OH: 7.36
7-OH: 7.20
7-OH: 7.37
7-OH: 7.37
40-OH: 8.61
40-OH: 9.84
40-OH: 8.62
40-OH: 8.62
400-OH: 9.60
400-OH: 9.09
300-OH: 9.30
400-COOH: 3.17
300-OH: 7.31
400-OH: 8.99
300-OH: 9.21
12
7-OH: 7.36
7-OH: 7.38
40-OH: 8.61
40-OH: 8.63
40-OH: 8.65
14
16
7-OH: 6.85
6-OH: 8.63
4-OH: 8.93
3-OH: 9.31
7-OH: 7.08
5-OH: 8.42
3-OH: 10.30
5.06, 8.64, 10.62
7-OH: 7.04
5-OH: 8.41
3-OH: 10.18
7.19, 9.36, 11.56
7-OH: 7.39
18
In the case of 16, the second hydroxy group at position 6 of the
flavone ring can accept additional hydrogen bonds from the sur-
rounding backbone NH groups of Ile 16 and Cys 17, an both donate
and accept a hydrogen bond from the thiol group of Cys 12. The fla-
vone moiety is buried in the pocket lined by the side chains of Leu
13, Glu 15, Ile 16, Asp 129, and Tyr 131. While the carbonyl group
at position 4 is important to raise the acidity of the 7-hydroxy
group, no hydrogen bonds with the binding site residues could
be observed. The 2-(40-hydroxyphenyl) residue makes lipophilic
interactions with Leu 13 and Tyr 131.
Morin
20-OH: 8.18
40-OH: 8.48
Exp. values31
Quercetin
40-OH: 8.26
30-OH: 8.58
Exp. values31
7-Hydroxy-2-phenyl-
chromen-4-one
Exp. values27
For compound 4, this residue bends towards Tyr 49, forming
hydrogen bonds between the two phenolic hydroxy groups of li-
gand and amino acid. Lastly, the biphenyl residue bends in such
Calculations were performed on the assumption that the investigated group is the
first one to be deprotonated (see text for discussion).

M. Forghieri et al. / Bioorg. Med. Chem. 17 (2009) 2658–2672
2665
Figure 2. Docking conformations proposed for 16 (top) and 12 (bottom) in LMW-PTP isoform I (PDB structure 5PNT14). Left and middle: 3D representations, from two
39
different angles, right: 2D representation. Pharmacophoric interactions were automatically detected by LIGANDSCOUT
.
Yellow sphere: hydrophopic feature, green arrow:
hydrogen bond donor, red arrow: hydrogen bond acceptor, red star: negative ionizable feature, purple ring and arrow: aromatic feature.
a way that both phenyl rings have hydrophobic interactions with
the tyrosine phenyl ring. For a classical stacking, both rings are
slightly shifted. Similar interactions can be observed for the indol
ring of Trp 49 in the isoform II structure 1XWW (not shown). This
binding pose is in good agreement with the binding data for 1, 4, 8,
10, and 16: the compounds have similar IC50 values, suggesting
that the structural variations occurred at positions where they
had no big influence on the binding site interactions. Indeed, the
presented binding pose predicts no interactions between the pro-
tein and hydroxy groups at the distal end of the biphenyl moiety,
and the residues at position 2 are placed at an open part of the
binding pocket, presenting possibilities for both hydrophobic and
hydrogen bond interactions with residues of variable length. How-
ever, binding data for the 2-(40-hydroxybenzyl) derivatives 4, 5 and
6 suggest that this residue may play an important role for selectiv-
ity against PTB-1B. Likewise, additional hydroxy groups at the dis-
tal ring of the biphenyl moiety abolish selectivity against PTP-1B.
For the less active compound 14, the distal phenyl ring is shifted
slightly away from the aromatic ring and towards the backbone
carbonyl group of Tyr/Trp 49.
Poses obtained for 12, doubly deprotonated at both the carbox-
ylic group and the 7-hydroxy group of the flavone, point towards a
different possible binding mode for this compound (Fig. 2): Here
the biphenyl moiety is buried in the catalystic pocket, with charged
interactions between the carboxylic group and Arg 18, supported
by hydrogen bonds donated from backbone NH groups of Ile 16,
Cys 17, and Arg 18. The 300-hydroxy group forms additional hydro-
gen bonds with Arg 18 and Asp 129. The biphenyl rings have
hydrophobic contacts with Leu 13, Ile 16, Tyr/Trp 49, and Tyr
131, and the flavone ring stacks with the aromatic ring of Tyr/Trp
49. Lastly, the hydroxy group of the 2-(4-hydroxyphenyl) residue
can donate a hydrogen bond to the carboxylate group of Glu 50
(isoform I) the guanidine group of Arg 53 (isoform II), or the back-
bone carbonyl group of Trp 49. Despite this alternate binding
mode, a similar affinity pattern can be observed for 12 and 16.
The biphenyl ring of 18 was placed at a similar position as that
of 12. Furthermore, the position of the 4- and 3-hydroxy groups
overlaid well with that of the 7- and 6-hydroxy groups of the bind-
ing pose of 16 discussed above.
Finally, docking of morin and quercetin produced multiple dif-
ferent poses. In one of the two most prominent poses, the flavone
ring is located in a similar way as discussed above for 16, while in
another one the flavone ring is flipped, with the carbonyl group
facing the opposite direction.
When 1, 10, 12 and 16 where docked in the doubly deproto-
nated form, bearing an additional negative charge at the 2-(40-
hydroxyphenyl) residue, also poses that placed this second nega-
tive charge at the catalystic centre were obtained. However, we
were unable to correlate these poses with our binding affinity in
a similarly plausible way than for the example discussed above.
Depending on the assumed protonation state, other poses
where the 2-phenyl ring is buried inside the catalystic centre were

2666
M. Forghieri et al. / Bioorg. Med. Chem. 17 (2009) 2658–2672
obtained as well. Because of the multiple possible tautomeric/mes-
omeric forms, broad distribution of hydroxy groups that are capa-
ble of forming hydrogen bonds, and the sterically undemanding
character of these two molecules, no clear preference for one dis-
tinct pose could be found. While we could thus show that both
morin and quercetin dock well into the investigated binding sites,
we were unable to find an indication why morin is an inhibitor and
quercetin is an activator of bovine LMW PTP (which, based on com-
parison of PDB complexes 1DG934 with 1XWW, has a binding site
identical to that of the human isoform II).
The GlideScore values produced by the docking runs were sim-
ilarly low for both the active and inactive molecules (lower nega-
tive values should indicate stronger binding interactions). For
isoform I, the lowest value obtained was ꢀ9.61, while for isoform
II, this value was even lower with ꢀ11.83 (morin). The problem
of correlating the GlideScore values with the binding affinity can
be shown by the values obtained for 18: depending on whether
the protonated or deprotonated form of the molecule was docked,
the lowest GlideScore values obtained were either ꢀ9.19/ꢀ9.51
(deprotonated) or ꢀ6.75/ꢀ5.92 (protonated) for isoforms I and II,
respectively. While in retrospect, this might indicate that the phe-
nolic groups of 18 are not acidic enough to be deprotonated at the
binding site, such assumptions on the protonation state—and thus
on the binding affinity—are more difficult to make in prospective
screening when little data is available as to which protonation
states should to be expected.
2.5. Cellular assays
The identification of cell-active compounds has been a major
challenge in the closely related PTP-1B drug discovery programs,
owing to the low cell permeability of most phosphotyrosine mi-
metic. Therefore the most in vitro active compounds (compounds
9, 10, 12, 13 and 16) have been assayed using a human hepatocel-
lular liver carcinoma cell line (HepG2) which constitutively express
the human insulin receptor, in order to test the compound’s ability
to exert an in vivo inhibition on PTPs, thus activating the insulin
signalling pathway.
HepG2 cells were incubated for 90 min with 9, 10, 12, 13 and 16
(10
l
M final concentration) as well as with insulin (10 nM). DMSO
M) were used as negative
(0.5%) and sodium orthovanadate (10
l
and positive controls, respectively (Fig. 3). The cell lysates were
subjected to SDS-PAGE, transferred to polyvinyldene membrane,
and followed by Western blot with a phospho-specific anti-IRb
antibodies (
a-pIRb) (Fig. 3A, top panel).
Among tested compounds, 10, 12 and 16 at 10
lM concentra-
tion stimulated insulin receptor phosphorylation comparable to
that exerted by insulin (Fig. 3A and B). Moreover, since Akt acts
downstream to the receptor in insulin signalling and thus it is
essential for the insulin-stimulated Glut4 translocation to the plas-
ma membrane, agents that enhance activity of Akt should enhance
cellular glucose uptake. Thus blots were also re-probed with a
phospho-specific anti-Akt antibody (
panel) shows that compounds 10 and 12 at 10
a
-pAkt). Figure 3A (middle
M concentration
Despite the considerable number of flexible protein side chains
reaching into the binding pocket, most of the obtained binding site
conformations showed a high similarity with the crystal structures.
Compared to 5PNT, Tyr 49 was usually rotated around the bond
between Cb and the phenyl ring to improve stacking with aromatic
ligand residuesin thecalculatedstructuresof isoformI. Additionally,
Tyr 131 and Tyr 132 showed some flexibility, especially the later,
where the whole side chain can rotate away from the binding site.
l
are capable to increase the phosphorylation level of Akt with an ex-
tent very close to that exerted by 10 nM insulin. In addition, the
same blots were re-probed with an anti-phosphotyrosine antibody.
Figure 3C demonstrates that insulin and the above compounds in-
crease tyrosine phosphorylation of identical protein bands,
whereas 10
lM orthovanadate, a general PTP inhibitor, causes pro-
miscuous tyrosine phosphorylation (Fig. 3D).
Figure 3. Effect of 9, 10, 12, 13 and 16 and insulin on tyrosine phosphorylation of IR in HepG2 cells. Subconfluent HepG2 cells were incubated at 37 °C with these compounds
at 10 M final concentration as well as with 10 nM insulin for 90 min. The cell lysates were subjected to SDS-PAGE and the resolved polypeptides transferred to
polyvinylidene membranes, which were probed with anti-phospho-insulin receptor b-subunit ( -pIRb). Then membrane was stripped and reprobed with anti-phospho-Akt
-Akt), anti-phosphotyrosine ( -pTyr), and anti-actin ( -Actin) antibodies. Data are representative of at least three separate experiments.
l
a
(a
a
a

M. Forghieri et al. / Bioorg. Med. Chem. 17 (2009) 2658–2672
2667
J 15.5), 7.80–7.59 (7H, m), 7.56–7.35 (4H, m), 7.16 (1H, dd, J 3.9, J
9.1), 7.00 (1H, d, J 9.1), 3.89 (3H, s). Anal. (C22H18O3) C, H.
3. Conclusions
The docking poses presented in the previous section appear well
fit to explain the SAR of our compounds on a qualitative level. How-
ever, the difference in acidity of the phenolic hydroxyls is too little
in order to decide unequivocally which is the binding mode of the
compounds inside the active site; the two most probably binding
poses (Fig. 2) will be useful for the design of new compounds in or-
der to investigate the accurateness of the presented models. More-
over, while we were yet unable to obtain substantial selectivity for
one of the two human isoforms of LMW-PTP, the docking results
provide a good starting point to further derivative our compounds
to let them interact with the subtype diverse side chain residues
at amino acid positions 49, 50, and 53. Some of the most active com-
pounds in in vitro assays were shown to be active on cells, being
able to increase the extent of IRb tyrosine phosphorylation in
HepG2 cells and activate Akt (Fig. 3); the lack of activity of 13, which
is active in vitro, is at the moment unexplained.
4.1.1.3. 20-Hydroxy-4-phenyl-40,50-dimethoxychalchone
(22c). Compound 22c was synthesized by reaction between 20-hy-
droxy-40,50-dimethoxyacetophenone and 4-biphenylcarboxalde-
hyde: yield 71%; mp 140–2 °C; 1H NMR (CDCl3): 13.38 (1H, broad
s), 7.94 (1H, d, J 15.4), 7.84–7.23 (11H, m), 6.55 (1H, s), 3.96 (3H,
s), 3.94 (3H, s). Anal. (C23H20O4) C, H.
4.1.1.4. 3-Bromo-20-hydroxy-40-methoxychalchone (22d). Com-
pound 22d was synthesized by reaction between 20-hydroxy-40-
methoxyacetophenone and 3-bromobenzaldehyde: yield 77%, mp
98–100 °C; 1H NMR (CDCl3): 13.31 (1H, broad s), 7.85 (3H, m),
7.60 (3H, m), 7.32 (1H, d, J 7.1), 6.51 (1H, dd, J 2.5, J 8.7), 6.49
(1H, s), 3.88 (3H, s). Anal. (C16H13BrO3) C, H.
4.1.1.5. 4-Bromo-20-hydroxy-40-methoxychalchone (22e). Com-
pound 22e was synthesized by reaction between 20-hydroxy-40-
methoxyacetophenone and 4-bromobenzaldehyde: yield 56%, mp
138–40 °C; 1H NMR (CDCl3): 13.35 (1H, broad s), 7.89–7.78 (2H,
m), 7.64–7.40 (5H, m), 6.50 (1H, dd, J 2.5, J 7.3), 6.48 (1H, s), 3.88
(3H, s). Anal. (C16H13BrO3) C, H.
In conclusion, a new class of highly active LMW-PTP inhibitors
was discovered. Docking experiments and experimental results
highlighted the possible binding mode of these compounds to
the enzyme, providing the background for the future optimization
of their inhibitory activity and selectivity towards the closely re-
lated enzyme PTP-1B.
4.1.2. General synthesis of compounds 23a–f
A solution of the appropriate chalchone 22a–f (3.0 mmol) and
Pd/C (0.6 g) in EtOAc (70 mL) was hydrogenated at 4 atm. for 1 h.
The reaction mixture was filtered through Celite and the filtrate
was evaporated under reduced pressure to give a solid, which
was crystallized (MeOH, unless otherwise stated).
4. Experimental
4.1. Chemistry
Melting points were determined on a Buchi 510 melting point
apparatus and are uncorrected. NMR spectra were recorded on a
Bruker DPX200 or Avance400 spectrometers (Centro Interdiparti-
mentale Grandi Scrumenti, Modena University). Chemical shifts
are reported in ppm from tetramethylsilane. J values are given in
Hertz. Microanalyses were carried out in the microanalysis labora-
tory of the Dipartimento di Scienze Farmaceutiche, Modena Uni-
versity. Analyses indicated by the symbols of the elements were
within 0.4% of the theoretical values. TLC were performed on pre-
coated silica gel F254 plates (Merck). Silica gel (Merck, 70–230
mesh) was used for column chromatography. Mass spectra were
obtained on a Finnigan MAT SSQ 7000 spectrometer (EI, 70 eV).
Compounds 2–4 and 17 were available from previous researches.24
Chalchones 22a24, 22f35 and 23a24 were synthesized as described.
4.1.2.1. 20-Hydroxy-4-phenyl-40-methoxydihydrochalchone
(23a). Yield 92%, mp 100–2 °C (EtOH) (103–4 °C)36
;
1H NMR
(CDCl3): 7.72–7.29 (10H, m), 6.47 (1H, s), 6.44 (1H, dd, J 2.5,
J
7.3), 3.73 (3H, s), 3.30 (2H, m), 3.15 (2H, m). Anal.
(C22H20O3) C, H.
4.1.2.2. 4-Phenyl-20-hydroxy-50-methoxydihydrochalchone
(23b). Yield 88%; mp 76–8 °C; 1H NMR (CDCl3): 7.63–7.30 (9 H, m),
7.20 (1H, d, J 3.2), 7.14 (1H, dd, J 3.2, J 8.9), 6.96 (1H, d, J 8.9), 3.80
(3H, s), 3.38 (2H, m), 3.15 (2H, m). Anal. (C22H20O3) C, H.
4.1.2.3. 4-Phenyl-20-hydroxy-40,50-dimethoxydihydrochalchone
(23c). Yield 79%; mp 83–5 °C; 1H NMR (CDCl3): 12.72 (1H, broad
s), 7.65–7.30 (9H, m), 7.08 (1H, s), 6.49 (1H, s), 3.93 (3H, s), 3.82
(3H, s), 3.30 (2H, m), 3.15 (2H, m). Anal. (C23H22O4) C, H.
4.1.1. General synthesis of compounds 22a–e
To
a solution of the appropriate acetophenone 21a–d
4.1.2.4. 3-Bromo-20-hydroxy-40-methoxydihydrochalchone
(23d). Yield 81%; mp 61–3 °C; 1H NMR (CDCl3): 12.71 (1H, broad
s), 7.63 (1H, d, J 9.6), 7.45–7.10 (4H, m), 6.49–6.38 (2H, m), 3.85
(3H, s), 3.23 (2H, m), 3.05 (2H, m). Anal. (C16H15BrO3) C, H.
(18.0 mmol) and the appropriate (substituted) benzaldehyde
(18.0 mmol) in CH3OH (100 mL) a solution of Ba(OH)2 (3.42 g,
18.0 mmol) was added dropwise during 15 min under stirring at
rt; the solution thus obtained was heated to reflux for 2 h; after
cooling, the solvent was removed under reduced pressure and
the residue was diluted with water (100 mL), acidified (pH 5.0)
with 1 N HCl, and extracted with EtOAc (3 ꢁ 50 mL). The organic
phase was then dried (Na2SO4), the solvent removed under re-
duced pressure and the residue crystallized from EtOH.
4.1.2.5. 4-Bromo-20-hydroxy-40-methoxydihydrochalchone
(23e). Yield 89%, mp 44–6 °C; 1H NMR (CDCl3): 12.73 (1H, broad s),
7.62 (2H, d, J 9.5), 7.42 (2H, d, J 8.3), 7.12 (2H, d, J 8.3), 6.45 (2H, m),
3.85 (3H, s), 3.21 (2H, m), 3.01 (2H, m). Anal. (C16H15BrO3) C, H.
4.1.1.1. 20-Hydroxy-4-phenyl-40-methoxychalchone (22a). Yield
90%, mp 103–5 °C (EtOH); 1H NMR (CDCl3): 13.50 (1H, broad s),
7.97 (1H, d, J 15.5), 7.89 (2H, m), 7.80–7.40 (9H, m), 6.55 (1H, dd,
J 8.2, J 2.1), 6.53 (1H, s), 3.96 (3H, s). Anal. (C22H18O3) C, H.
4.1.2.6. 4-Bromo-20-hydroxydihydrochalchone (23f). Yield 86%;
mp70–2 °C; 1H NMR (CDCl3): 7.70 (1H, dd, J 2.1, J 8.0); 7.56–7.38
(3H, m), 7.16 (2H, m), 7.00 (1H, dd, J 1.4, J 8.0), 6.90 (1H, m),
3.32 (2H, m), 3.05 (2H, m). Anal. (C15H13BrO2) C, H.
4.1.1.2. 20-Hydroxy-4-phenyl-50-methoxychalchone (22b). Com-
pound 22b was synthesized by reaction between 20-hydroxy-50-
methoxyacetophenone and 4-biphenylcarboxaldehyde: yield 66%;
mp 118–20 °C; 1H NMR (CDCl3): 12.40 (1H, broad s), 7.97 (1H, d,
4.1.3. General synthesis of compounds 24a–d
A
solution of the appropriate dihydrochalchone 23a–b
(3.01 mmol) and the appropriate methyl ester (3.31 mmol,
1.1 equiv) in anhyd pyridine (9 mL) was added dropwise to a

2668
M. Forghieri et al. / Bioorg. Med. Chem. 17 (2009) 2658–2672
well-stirred suspension of NaH (60% dispersion in mineral oil)
(9.03 mmol, 3.0 equiv) in anh. pyridine (7 mL). When the reaction
subsided, the mixture was heated at 90 °C for 15 min. After cooling,
the mixture was decomposed in 2 N HCl and extracted with meth-
ylene chloride (2 ꢁ 60 mL). The combined organic layers were
washed with 1 N HCl (2 ꢁ 30 mL) and water (30 mL), then dried
(Na2SO4), and the solvent was removed under reduced pressure.
The residue was purified by means of column chromatography
(cyclohexane/ethyl acetate 8.5/1.5); the oily product thus obtained
showed an Rf 0.15–0.20 (cyclohexane/ethyl acetate 8.5/1.5). The
product was then dissolved in acetic acid (20 mL) and concentrated
sulphuric acid (0.1 mL). The resulting solution was then heated at
100 °C for 1 h. After cooling, the solution was neutralized with
NaHCO3 2 N then extracted with ethyl acetate (2 ꢁ 30 mL), the or-
ganic layer was washed again with NaHCO3 0.5 N and satd NaCl
solution, dried (Na2SO4) and the solvent was removed under re-
duced pressure. The residue was then washed with cold water
and collected by filtration.
6.94 (2H, d, J 8.9), 6.71 (1H, s), 3.94 (6H, s), 3.83 (3H, s), 3.23
(2H, m), 3.01 (2H, m). Anal. (C31H28O6) C, H.
4.1.4.2. 40-Methoxyphenyl-2-[3-(3-bromophenyl)propionyl]-4-
methoxy phenyl ester (25b) (from 23d). Yield 79%; mp 40–3 °C,
1H NMR (CDCl3): 8.15 (2H, m), 7.87 (1H, d, J 8.8), 7.30–6.92 (6H,
m), 6.86 (1H, dd,
J 2.5, J 8.8), 6.73 (1H, d, J 2.5), 3.92
(3H, s), 3.86 (3H, s), 3.15 (2H, m), 2.94 (2H, m). Anal. (C24H21BrO5)
C, H.
4.1.4.3. 40-Methoxyphenyl-2-[3-(4-bromophenyl)propionyl]-4-
methoxy phenyl ester (25c) (from 23e). Yield 88%; mp 65–7 °C,
1H NMR (CDCl3): 8.12 (2H, m), 7.85 (1H, d, J 8.7), 7.30 (2H, m),
7.03–6.90 (4H, m), 6.87 (1H, dd, J 2.3, J 8.7), 6.73 (1H, d, J 2.3),
3.92 (3H, s), 3.88 (3H, s), 3.88 (3H, s), 3.15 (2H, m), 2.92 (2H, m).
Anal. (C24H21BrO5) C, H.
4.1.4.4. Phenyl-2-[3-(4-bromophenyl)propionyl]-4-methoxy
phenyl ester (25d) (from 23f). Yield 77%; mp 108–10 °C; 1H NMR
(CDCl3):8.10 (2H, m), 7.76 (1H, dd, J 2.3, J 8.1), 7.57 (1H, m), 7.40–
7.20 (4H, m), 7.06–6.90 (4H, m), 3.91 (3H, s), 3.20 (2H, m), 2.94 (2H,
m). Anal. (C23H19BrO4) C, H.
4.1.3.1. 7-Methoxy-2-(40-methoxyphenyl)-3-[(1,10-biphenyl-4-
yl)methyl]-4H-1-benzopyran-4-one (24a). Compound 24a was
synthesized by reaction between 23a and 4-methoxyphenylbenzo-
ic acid methyl ester. Yield 12%; 1H NMR (CDCl3): 8.18 (1H, d, J 8.9),
7.60–7.22 (11H, m), 7.00 (2H, d, J 8.9), 7.00 (1H, dd, J 2.4, J 8.9), 6.87
(1H, d, J 2.4), 4.03 (2H, s), 3.91 (3H, s), 3.87 (3H, s). Anal. (C30H24O4)
C, H.
4.1.5. General synthesis of 4H-1-benzopyran-4-one derivatives
(27a–d)
To a stirred suspension of NaH (60%, 6 mmol, 3 equiv) in DMSO
(2 mL) at 20 °C was added dropwise a solution of 25a–d (2.0 mmol)
in DMSO (3.5 mL). The reaction mixture was stirred at 20 °C for 3 h,
then slowly poured into a stirred slurry of ice in an oxalic acid solu-
tion. The aqueous mixture was extracted with ethyl acetate which
in turn was successively washed with water and dried (Na2SO4)
and concentrated under reduced pressure to give an oil that was
purified by column chromatography (cyclohexane/ethyl acetate
7.5/2.5) (26a–d). The 1H NMR spectra (CDCl3) showed the charac-
teristic presence of a triplet (1H, t, J = 6.9) at 5.30 ppm. The oil was
then dissolved in CH3COOH (10 mL) and H2SO4 (0.1 mL), then it
was heated at 95 °C for 1 h. After cooling, the solution was neutral-
ized with NaHCO3 2 N and then extracted with ethyl acetate
(2 ꢁ 20 mL). The organic layer was rewashed with NaHCO3 0.5 N
and saturated NaCl solution, then dried (Na2SO4) and the solvent
was removed under reduced pressure.
4.1.3.2. 7-Methoxy-2-(40-methoxyphenylethyl)-3-[(1,10-biphen-
yl-4-yl)methyl]-4H-1-benzopyran-4-one (24b). Compound 24b
was synthesized by reaction between 23a and 4-methoxyphenyl-
propanoic acid methyl ester. Yield 18%; 1H NMR (CDCl3): 8.16
(1H, d, J 8.9), 7.60–7.21 (9H, m), 7.04 (2H, d, J 8.4), 6.97 (1H, dd, J
1.9, J 8.9), 6.83 (1H, s), 6.81 (2H, d, J 8.4), 3.92 (3H, s), 3.88 (2H,
s), 3.78 (3H, s), 2.94 (4H, m). Anal. (C32H28O4) C, H.
4.1.3.3. 7-Methoxy-2-(40-methoxyphenylproyl)-3-[(1,10-biphen-
yl-4-yl)methyl]-4H-1-benzopyran-4-one (24c). Compound 24c
was synthesized by reaction between 23a and 4-methoxyphenylb-
utanoic acid methyl ester. Yield 22%; 1H NMR (CDCl3): 8.15 (1H, d, J
8.9), 7.60–7.18 (9H, m), 7.05 (2H, d, J 8.4), 6.96 (1H, dd, J 1.8, J 8.9),
6.81 (2H, d, J 8.4), 6.80 (1H, s), 3.90 (6H, s), 3.78 (2H, s), 2.72 (2H, t, J
7.7), 2.62 (2H, t, J 7.5), 1.95 (2H, m). Anal. (C33H30O4) C, H.
4.1.5.1. 2-(4-Methoxyphenyl)-3-(biphenylmethyl)-6,7-dime-
thoxy-4H-1-benzopyran-4-one (27a) (from 26a). Yield 55%; mp
200–2 °C; 1H NMR (DMSO-d6): 7.70–7.30 (10H, m), 7.24 (1H, s),
7.16 (2H, d, J 8.4), 7.08 (2H, m), 3.92 (3H, s), 3.87 (3H, s), 3.82
(3H, s), 3.30 (2H, s). Anal. (C31H26O5) C, H.
4.1.3.4. 6-Methoxy-2-(40-methoxybenzyl)-3-[(1,10-biphenyl-4-
yl)methyl]-4H-1-benzopyran-4-one (24d). Compound 24d was
synthesized by reaction between 23b and 4-methoxyphenylacetic
acid methyl ester. Yield 25%; 1H NMR (CDCl3): 7.65–7.18 (12H, m),
7.10 (2H, d, J 8.7), 6.82 (2H, d, J 8.7), 4.09 (2H, s), 4.03 (2H, s), 3.90
(3H, s), 3.78 (3H, s). Anal. (C31H26O4) C, H.
4.1.5.2. 2-(4-Methoxyphenyl)-3-(3-bromobenzyl)-7-methoxy-
4H-1-benzopyran-4-one (27b) (from 26b). Yield 49%; mp 60–
3 °C; 1H NMR (CDCl3): 8.16 (1H, d,
7.25 (2H, m), 7.10 (2H, m), 7.05–6.95 (3H, m), 6.88 (1H, d,
J 2.1), 3.94 (2H, s), 3.91 (3H, s), 3.88 (3H, s). Anal. (C24H19BrO4)
C, H.
J 8.4), 7.49 (2H, m),
4.1.4. General synthesis of the phenylesters (25a–d)
To a stirred solution of the appropriate dihydrochalchone 23c–f
(3.0 mmol), triethylamine (6.0 mmol, 2 equiv) and dimethylamino-
pyridine (0.3 mmol, 0.1 equiv) in 5 mL of methylene chloride at
0 °C was added dropwise 4-methoxybenzoyl chloride (2 mL,
3.3 mmol, 1.1 equiv). The reaction mixture was allowed to warm
to rt over 1 h, then the solution was diluted with methylene chlo-
ride and washed successively with 1 N HCl and water, then dried
(Na2SO4) and concentrated under reduced pressure. The residue
thus obtained was triturated with diethyl ether (5 mL), and the res-
idue was collected by filtration.
4.1.5.3. 2-(4-Methoxyphenyl)-3-(4-bromobenzyl)-7-methoxy-
4H-1-benzopyran-4-one (27c) (from 26c). Yield 45%; mp 145–
7 °C; 1H NMR (CDCl3): 8.15 (1H, d, J 8.9), 7.48 (2H, d, J 8.9), 7.35
(2H, d, J 8.4), 7.10–6.93 (5H, m), 6.87 (1H, d, J 2.3), 3.92 (5H, s),
3.88 (3H, s). Anal. (C24H19BrO4) C, H.
4.1.5.4. 2-(4-Methoxyphenyl)-3-(4-bromobenzyl)-4H-1-benzo-
pyran-4-one (27d) (from 26d). Yield 49%; mp 108–10 °C; 1H NMR
(CDCl3): 8.25 (1H, dd, J 2.1, J 8.6), 7.68 (1H, m), 7.55–7.30 (6H, m),
7.10–6.95 (4H, m), 3.95 (2H, s), 3.81 (3H, s). Anal. (C23H17BrO3) C,
H.
4.1.4.1. 40,50-Dimethoxyphenyl-2-[3-(biphenyl)propionyl]-4-
methoxyphenyl ester (25a) (from 23c). Yield 85%; 1H NMR
(CDCl3): 8.13 (2H, d, J 8.9), 7.60–7.29 (8H, m), 7.12 (2H, d, J 8.1),

M. Forghieri et al. / Bioorg. Med. Chem. 17 (2009) 2658–2672
2669
4.1.6. General synthesis of compounds 28a–h
4.1.6.7. 2-(40-Methoxyphenyl)-3-[(300,400-dimethoxy-1,10-biphe-
nyl-4-yl)methyl]-4H-1-benzopyran-4-one (28g). Compound 28g
was synthesized by reaction between 27d and 3,4-dimethoxy-
phenyl boronic acid: yield 77%; mp 45–47 °C; 1H NMR (CDCl3):
8.27 (1H, dd, J 1.9, J 8.1), 7.68 (1H, m), 7.58 (2H, m), 7.53–7.35
(4H, m), 7.22 (2H, m), 7.16–6.90 (5H, m), 4.06 (2H, s), 3.93 (3H,
s), 3.91 (3H, s), 3.87 (3H, s). Anal. (C31H26O5) C, H.
To a suspension of Pd(PPh3)4 (0.039 mmol, 0.05 equiv) in anhy-
drous DME (2 mL) was added the compound 27b (for compounds
14 and 15), 27c (for compounds 9, 10 and 12) and 27d (for com-
pounds 11 and 13) (0.35 g, 0.78 mmol) in DME (5.5 mL) and the
solution thus obtained was stirred for 20 min at rt. Then the appro-
priate phenylboronic acid (1.17 mmol, 1.5 equiv) in DME (2.5 mL)
was added and the mixture became dark and cloudy. After adding
K2CO3 (1.56 mmol, 2 equiv) as solid, the reaction was refluxed
overnight. After cooling, the suspension was diluted with water
(60 mL) and saturated NaCl solution (30 mL), then extracted with
EtOAc (2 ꢁ 60 mL). The combined organic layers were washed with
saturated NaCl solution and dried over MgSO4, then concentrated
under reduced pressure. The residue was purified by column chro-
matography (n-hexane/EtOAc 6/4).
4.1.6.8. 2-(40-Methoxyphenyl)-3-[(400carboxy-300-methoxy-1,10-
biphenyl-4-yl)methyl]-4H-1-benzopyran-4-one methyl ester
(28h). Compound 28h was synthesized by reaction between 27d
and
4-carboxy-3-methoxyphenyl
boronic
acid
methyl
ester: yield 42%; mp 48–50 °C; 1H NMR (CDCl3): 8.28 (1H, dd, J
2.2, J 8.0), 7.86 (1H, d, J 8.0), 7.69 (1H, m), 7.58 (2H, m), 7.50
(2H, m), 7.42 (1H, m), 7.32–7.12 (5H, m), 7.01 (2H, m), 4.06
(2H, s), 3.96 (3H, s), 3.91 (3H, s), 3.89 (3H, s). Anal. (C32H26O6)
C, H.
4.1.6.1. 7-Methoxy-2-(40-methoxyphenyl)-3-[(400-methoxy-1,10-
biphenyl-4-yl)methyl]-4H-1-benzopyran-4-one
(28a). Com-
pound 28a was synthesized by reaction between 27c and 4-
methoxyphenyl boronic acid: yield 30%; 1H NMR (CDCl3): 8.18
(1H, d, J 8.9), 7.60–7.40 (6H. m), 7.21 (2H, d, J 8.3), 7.05–6.92
(5H, m), 6.87 (1H, d, J 2.3), 4.02 (2H, s), 3.90 (3H, s), 3.86 (3H, s),
3.83 (3H, s). Anal. (C31H26O5) C, H.
4.1.7. Synthesis of the biphenyl derivatives 30a–c
A mixture of Na2CO3 (212 mg, 2 mmol), the appropriate phenyl-
boronic acid 29a–c (1.5 mmol), Pd(OAc)2 (1.1 mg, 0.005 mmol) and
bromobenzene (105 lL, 1 mmol) in distilled water/acetone (3.5
and 3 mL, respectively) was stirred for 1 h at 35 °C in a water bath.
The reaction was then extracted with EtOAc (2 ꢁ 10 mL). The com-
bined organic layers were washed with saturated NaCl solution
and dried over MgSO4, then concentrated under reduced pressure.
The residue was purified by column chromatography (n-hexane/
EtOAc 6/4).
4.1.6.2. 7-Methoxy-2-(40-methoxyphenyl)-3-[(300-carboxy-1,10-
biphenyl-4-yl)methyl]-4H-1-benzopyran-4-one
ethyl
ester
(28b). Compound 28b was synthesized by reaction between 27c
and 3-carboxyphenyl boronic acid ethyl ester: yield 46%; 1H NMR
(CDCl3): 8.25–7.35 (8H, m), 7.30–6.85 (7H, m), 4.41 (2H, q, J 7.1),
4.03 (2H, s), 3.91 (3H, s), 3.88 (3H, s), 1.42 (3H, t, J 7.1). Anal.
(C33H28O6) C, H.
4.1.7.1. 30,40-Dimethoxy-1,10-biphenyl (30a). Compound 30a was
synthesized by reaction between bromobenzene and 3,4-dime-
thoxyphenyl boronic acid: yield 75%; mp 58–60 °C; 1H NMR
(CDCl3): 7.62–7.53 (2H, m), 7.50–7.28 (3H, m), 7.22–7.10 (2H, m),
6.96 (1H, d, J 8.1), 3.97 (3H, s), 3.94 (3H, s). Anal. (C14H14O2) C, H.
4.1.6.3. 7-Methoxy-2-(40-methoxyphenyl)-3-[(300,400-dimethoxy-
1,10-biphenyl-4-yl)methyl]-4H-1-benzopyran-4-one (28c). Com-
pound 28c was synthesized by reaction between 27c and 3,4-
dimethoxyphenyl boronic acid: yield 68%; mp 153–5 °C; 1H NMR
(CDCl3): 8.18 (1H, d, J 8.9), 7.56 (2H, d, J 8.9), 7.43 (2H, d, J 8.2),
7.22 (2H, d, J 8.2), 7.18–6.85 (7H, m), 4.01 (2H, s), 3.94 (3H, s),
3.91 (6H, s), 3.88 (3H, s). Anal. (C32H28O6) C, H.
4.1.7.2. 40-Carboxy-30-methoxy-1,10-biphenyl methyl ester
(30b). Compound 30b was synthesized by reaction between bro-
mobenzene and 4-carboxy-3-methoxyphenyl boronic acid methyl
ester: yield 74%; mp 45–7 °C; 1H NMR (CDCl3): 7.90 (1H, d, J 7.9),
7.68–7.58 (2H, m), 7.54–7.40 (3H, m), 7.25–7.16 (2H, m), 4.01
(3H, s), 3.93 (3H, s). Anal (C15H14O3) C, H.
4.1.6.4. 7-Methoxy-2-(40-methoxyphenyl)-3-[(400carboxy-300-
methoxy-1,10-biphenyl-4-yl)methyl]-4H-1-benzopyran-4-one
methyl ester (28d). Compound 28d was synthesized by reaction
between 27c and 4-carboxy-3-hydroxyphenyl boronic acid methyl
ester: yield 36%; mp 140–2 °C; 1H NMR (CDCl3): 8.16 (1H, d, J 8.9),
7.86 (1H, d, J 7.9), 7.55 (2H, d, J 8.8), 7.49 (2H, d, J 8.2), 7.30–6.95
4.1.7.3. 20,40-Dimethoxy-1,10-biphenyl (30c). Compound 30c was
synthesized by reaction between bromobenzene and 2,4-dime-
thoxyphenyl boronic acid: yield 93% (oil); 1H NMR (CDCl3): 7.60–
7.26 (6H, m), 6.58 (2H, m), 3.88 (3H, s), 3.83 (3H, s). Anal.
(C14H14O2) C, H.
(7H, m), 6.88 (1H, d,
J
2.4), 4.03 (2H, s), 3.96
(3H, s), 3.92 (3H, s), 3.91 (3H, s), 3.88 (3H, s). Anal. (C33H28O7)
C, H.
4.1.8. Synthesis of compounds1, 5–16 and biphenyl derivatives
18–20
4.1.6.5. 7-Methoxy-2-(40-methoxyphenyl)-3-[(300,400-dimethoxy-
1,10-biphenyl-3-yl)methyl]-4H-1-benzopyran-4-one (28e). Com-
pound 28e was synthesized by reaction between 27b and 3,4-
dimethoxyphenyl boronic acid: yield 71%; mp 105–7 °C; 1H NMR
(CDCl3): 8.16 (1H, d, J 8.8), 7.53 (2H, d, J 7.6), 7.40–6.80 (11H, m),
4.05 (2H, s), 3.93 (3H, s), 3.92 (6H, s), 3.85 (3H, s). Anal.
(C32H28O6) C, H.
A solution of the protected compounds 24a–d, 27a, 28a–h and
30a–c (0.5 mmol) in methylene chloride (10 mL) at 0 °C was trea-
ted with BBr3 (1.0 M solution in CH2Cl2) (2 equiv for each methoxy-
group). The resulting solution was stirred at rt for 24 h, then cooled
at 0 °C, water and ice were added, and the precipitate thus formed
was filtrated and washed with water. Column chromatography
(CH2Cl2/CH3OH 9:1 for compounds 1, 5–8, 10, 11, 14 and 16,
CH2Cl2/CH3OH 9:1 followed by CH3OH 100% for compounds
9,12,13 and 15, cyclohexane/ethyl acetate 8:2 for 18 and 20) al-
lowed the obtainment of the desired compounds.
4.1.6.6.
7-Methoxy-2-(40-methoxyphenyl)-3-[(400-carboxy-300-
methoxy-1,10-biphenyl-3-yl)methyl]-4H-1-benzopyran-4-one
methyl ester (28f). Compound 28f was synthesized by reaction
between 27b and 4-carboxy-3-methoxyphenyl boronic acid
methyl ester: yield 21%; mp 53–5 °C; 1H NMR (CDCl3): 8.16 (1H,
d, J 8.8), 7.83 (1H, d, J 7.8), 7.53 (2H, d, J 8.4), 7.40–6.90 (9H, m),
6.86 (1H, d, J 2.0), 4.06 (2H, s), 3.93 (3H, s), 3.90 (6H, s), 3.85 (3H,
s). Anal. (C33H28O7) C, H.
4.1.8.1.
7-Hydroxy-2-(40-hydroxyphenyl)-3-[(1,10-biphenyl-4-
yl)methyl]-4H-1-benzopyran-4-one (1) (from 24a). Yield 68%;
mp 247–50 °C dec (Methanol); 1H NMR (DMSO-d6): 10.71 (1H,
broad s), 10.0 (1H, broad s), 7.90 (1H, d, J 8.6), 7.67–7.30 (9H, m),
7.15 (2H, d, J 8.2), 6.90 (4H, m), 3.88 (2H, s). Anal. (C28H20O4) C, H.

2670
M. Forghieri et al. / Bioorg. Med. Chem. 17 (2009) 2658–2672
4.1.8.2. 7-Hydroxy-2-(40-hydroxyphenyletyl)-3-[(1,10-biphenyl-
4-yl)methyl]-4H-1-benzopyran-4-one (5) (from 24b). Yield 71%;
mp 195–7 °C (dec.); 1H NMR (400 MHz, DMSO-d6): 7.86 (1H, d, J
8.7), 7.59 (2H, d, J 7.7), 7.50 (2H, d, J 8.1), 7.42 (2H, dd, J 7.4, J
7.7), 7.32 (1H, t, J 7.4), 7.19 (2H, d, J 8.1), 6.94 (2H, d, J 8.4), 6.90
(1H, dd, J 2.1, J 8.7), 6.84 (1H, d, J 2.1), 6.64 (2H, d, J 8.4), 3.75
(2H, s), 2.94 (2H, dd, J 7.2, J 7.6), 2.79 (2H, dd, J 7.2, J 7.6); 13C
NMR (100 MHz, DMSO-d6): 176.3, 165.4, 162.9, 157.7, 156.2,
140.5, 139.9, 138.3, 130.7, 129.7, 129.3, 129.0, 127.6, 127.4,
127.0, 126.9, 119.7, 115.61, 115.56, 115.3, 102.4, 34.1, 32.0, 28.9.
Anal. (C30H24O4) C, H.
4.1.8.8.
nyl-4-yl)methyl]-4H-1-benzopyran-4-one
2-(40-Hydroxyphenyl)-3-[(300,400-dihydroxy-1,10-biphe-
(11) (from
28g). Yield 49%; mp 250–2 °C (Methanol); 1H NMR (400 MHz,
DMSO-d6): 10.20 (1H, broad s), 9.00 (1H, broad s), 8.06 (1H, dd, J
1.3, J 7.9), 7.81 (1H, m), 7.68 (1H, d, J 8.3), 7.52 (2H, d, J 8.7),
7.56–7.46 (1H, m), 7.38 (2H, d, J 8.2), 7.10 (2H, d, J 8.2), 6.99 (1H,
d, J 2.0), 6.91 (2H, d, J 8.7), 6.88 (1H, d, J 2.0, J 8.2), 6.78 (1H, d, J
8.2), 3.90 (2H, s); 13C NMR (100 MHz, DMSO-d6): 177.5, 163.3,
160.1, 156.1, 150.0, 145.4, 138.5, 134.5, 131.8, 130.7, 128.6,
126.3, 125.63, 125.56, 123.7, 122.7, 119.0, 118.8, 117.9, 116.5,
115.9, 114.2, 30.9. Anal. (C28H20O5) C, H.
4.1.8.3. 7-Hydroxy-2-(40-hydroxyphenylpropyl)-3-[(1,10-biphe-
nyl-4-yl)methyl]-4H-1-benzopyran-4-one (6) (from 24c). Yield
75%; mp 166–8 °C; 1H NMR (400 MHz, DMSO-d6): 10.70 (1H, s),
9.15 (1H, s), 7.90 (1H, d, J 8.4), 7.60 (2H, d, J 7.6), 7.52 (2H, d, J
8.0), 7.43 (2H, t), 7.33 (1H, t, J 7.0), 7.20 (2H, d, J 8.0), 6.93 (2H, d,
J 8.2), 6.91 (1H, dd, J 1.7, J 8.2), 6.82 (1H, d, J 1.7), 6.65 (2H, d, J
8.2), 3.80 (2H, s), 2.67 (2H, dd, J 7.4, J 7.7), 2.49 (2H, dd, J 7.4, J
7.7), 1.77 (2H, m); 13C NMR (100 MHz, DMSO-d6): 176.3, 166.3,
162.9, 157.7, 155.9, 140.5, 140.2, 138.3, 131.7, 129.6, 129.3,
129.0, 127.6, 127.4, 127.1, 126.9, 119.5, 115.6, 115.5, 115.2,
102.4, 34.3, 31.5, 29.3, 29.0. Anal. (C31H26O4) C, H.
4.1.8.9. 7-Hydroxy-2-(40-hydroxyphenyl)-3-[(400carboxy-300-
hydroxy-1,10-biphenyl-4-yl)methyl]-4H-1-benzopyran-4-one
(12) (from 28d). Yield 64%, mp > 280 °C; 1H NMR (400 MHz,
DMSO-d6): 7.89 (1H, d, J 8.7), 7.59 (2H, d, J 8.6), 7.46 (2H, d, J
8.5), 7.21–7.12 (3H, m), 7.09 (1H, s), 6.88 (2H, d, J 8.5), 6.97–6.84
(3H, m), 3.89 (2H, s); 13C NMR (100 MHz, DMSO-d6): 176.7,
164.0, 163.0, 162.7, 160.9, 159.9, 159.8, 157.9, 140.9, 137.3,
131.0, 130.6, 128.8, 127.3, 127.2, 123.9, 118.3, 118.2, 116.0,
115.9, 115.6, 115.4, 115.2, 102.6, 30.9. Anal. (C29H20O7) C, H.
4.1.8.10. 2-(40-Hydroxyphenyl)-3-[(400carboxy-300-hydroxy-1,10-
biphenyl-4-yl)methyl]-4H-1-benzopyran-4-one
(13)
(from
4.1.8.4.
6-Hydroxy-2-(40-hydroxybenzyl)-3-[(1,10-biphenyl-4-
28h). Yield 55%; mp 267–70 °C; Ms (Q-TOF Global Ultima,
Micromass, Waters, MALDI, CH3OH solution) (m/z) [M+H]+:
465.11; 1H NMR (400 MHz, DMSO-d6): 10.26 (1H, broad s),
8.06 (1H, d, J 8.0), 7.82 (1H, m), 7.73 (1H, d, J 8.4), 7.68 (1H,
d, J 8.4), 7.55–7.44 (5H, m), 7.15 (2H, d, J 8.3), 6.92 (2H, d, J
8.5), 6.92–6.84 (2H, m), 3.96 (2H, s); 13C NMR (100 MHz,
DMSO-d6): 177.5, 172.2, 163.43, 163.37, 160.2, 156.1, 143.7,
139.8, 138.3, 134.5, 131.0, 130.7, 128.6, 127.1, 125.6, 125.5,
123.6, 122.7, 118.94, 118.87, 118.8, 115.9, 115.1, 114.0, 31.0.
Anal. (C29H20O6) C, H.
yl)methyl]-4H-1-benzopyran-4-one (7) (from 24d). Yield 62%;
mp 135–7 °C; 1H NMR (DMSO-d6): 9.90 (1H, broad s), 9.89 (1H,
broad s), 9.22 (1H, broad s), 7.70–7.23 (11H, m), 7.17 (1H, dd, J
3.0, J 9.0), 7.00 (2H, d, J 8.5), 6.65 (2H, d, J 8.5), 3.99 (4H, s); 13C
NMR (100 MHz, DMSO-d6): 176.7, 163.0, 159.8, 157.9, 157.4,
138.8, 138.2, 131.3, 130.6, 130.4, 128.6, 128.0, 127.4, 126.3,
123.9, 118.5, 116.1, 115.8, 115.6, 115.4, 102.6, 30.8. Anal.
(C29H22O4) C, H.
4.1.8.5.
biphenyl-4-yl)methyl]-4H-1-benzopyran-4-one
7-Hydroxy-2-(40-hydroxyphenyl)-3-[(400-hydroxy-1,10-
(8) (from
4.1.8.11. 7-Hydroxy-2-(40-hydroxyphenyl)-3-[(300,400-dihydroxy-
1,10-biphenyl-3-yl)methyl]-4H-1-benzopyran-4-one (14) (from
28e). Yield 66%; mp 138–40 °C (dec.); Ms (m/z): 452 [M+]; 1H NMR
(400 MHz, DMSO-d6): 10.83 (1H, broad s), 10.09 (1H, broad s), 9.00
(2H, broad s), 7.90 (1H, d, J 8.8), 7.48 (2H, d, J 8.6), 7.34–7.19 (4 H,
m), 6.89 (2H, d, J 8.6), 7.02–6.74 (5H, m), 3.88 (2H, s); 13C NMR
(100 MHz, DMSO-d6): 176.7, 163.0, 162.6, 159.9, 157.9, 146.0,
145.6, 141.2, 140.9, 132.1, 130.6, 129.2, 127.4, 126.2, 125.9,
124.0, 118.5, 118.1, 116.5, 115.8, 115.6, 115.4, 114.3, 102.6, 31.2.
Anal. (C28H20O6) C, H.
28a). Yield 71%; mp 183–5 °C (dec.); Ms (m/z): 436 [M+]; 1H
NMR (400 MHz, DMSO-d6): 10.74 (1 H, broad s), 10.07 (1H, broad
s), 9.48 (1H, broad s), 7.89 (1H, d, J 8.7), 7.47 (2H, d, J 8.6), 7.43
(2H, d, J 8.1), 7.42 (2H, d, J 8.5), 7.10 (2H, d, J 8.1), 6.97–6.84 (4H,
m), 6.82 (2H, d, J 8.5), 3.85 (2H, s); 13C NMR (100 MHz, DMSO-
d6): 176.7, 163.0, 159.8, 157.9, 157.4, 138.8, 138.2, 131.3, 130.6,
130.4, 128.6, 128.0, 127.4, 126.3, 123.9, 118.5, 116.1, 115.8,
115.6, 115.4, 102.6, 30.8. Anal. (C28H20O5) C, H.
4.1.8.6.
biphenyl-4-yl)methyl]-4H-1-benzopyran-4-one
7-Hydroxy-2-(40-hydroxyphenyl)-3-[(300-carboxy-1,10-
(9) (from
4.1.8.12. 7-Hydroxy-2-(40-hydroxyphenyl)-3-[(400carboxy-300-
hydroxy-1,10-biphenyl-3-yl)methyl]-4H-1-benzopyran-4-one
(15) (from 28f). Yield 69%; mp >280 °C; 1H NMR (400 MHz,
DMSO-d6): 10.81 (1H, broad s), 10.11 (1H, broad s), 7.90 (1H, d, J
8.8), 7.73 (1H, d J 8.4), 7.49 (2H, d, J 8.6), 7.39 (1H, d, J 7.9), 7.34–
7.24 (2H, m), 7.03 (1H, d, J 8.0), 6.90 (2H, d, J 8.6), 6.96–6.77 (4H,
m), 3.90 (2H, s); 13C NMR (100 MHz, DMSO-d6): 176.7, 172.2,
163.4, 162.9, 162.4, 159.9, 158.1, 144.0, 141.3, 140.6, 131.0,
130.6, 129.2, 127.32, 127.30, 126.7, 124.9, 123.9, 119.8, 118.6,
115.9, 115.6, 115.3, 115.2, 114.2, 102.6, 31.1. Anal. (C29H20O7) C, H.
28b). Yield 66%; mp 280–3 °C (dec.); 1H NMR (400 MHz,
DMSO-d6): 10.85 (1H, broad s), 10.09 (1H, broad s), 8.14 (1H,
s), 7.94–7.84 (3H, m), 7.62–7.43 (5H, m), 7.28–7.15 (2H, m),
6.98–6.83 (4H, m), 3.89 (2 H, s); 13C NMR (100 MHz, DMSO-
d6): 176.7, 167.7, 163.1, 162.7, 159.9, 157.9, 140.6, 137.3,
131.9, 131.3, 130.6, 129.7, 128.9, 128.5, 127.5, 127.4, 127.2,
123.9, 118.6, 118.3, 116.6, 115.8, 115.6, 102.6, 30.9. Anal.
(C29H20O6) C, H.
4.1.8.7. 7-Hydroxy-2-(40-hydroxyphenyl)-3-[(300,400-dihydroxy-
1,10-biphenyl-4-yl)methyl]-4H-1-benzopyran-4-one (10) (from
28c). Yield 65%; mp 212–15 °C; Ms (m/z): 452 [M+]; 1H NMR
(400 MHz, DMSO-d6): 10.80 (1H, broad s), 10.05 (1H, broad s),
7.89 (1H, d, J 8.7), 7.46 (2H, d, J 8.5), 7.37 (2 H, d, J 7.7), 7.09 (2H,
d, J 7.7), 6.98 (1H, s), 6.96–6.84 (5H, m), 6.78 (1H, d, J 8.2), 3.86
(2H, s); 13C NMR (100 MHz, DMSO-d6): 176.7, 163.0, 162.6,
159.8, 157.9, 146.0, 145.4, 138.8, 138.5, 131.9, 130.6, 128.6,
127.4, 126.3, 123.9, 118.4, 117.9, 116.5, 115.8, 115.6, 115.4,
114.2, 102.6, 30.8. Anal. (C28H20O6) C, H.
4.1.8.13. 6,7-Dihydroxy-2-(40-hydroxyphenyl)-3-[(1,10-biphenyl-
4-yl)methyl]-4H-1-benzopyran-4-one (16) (from 27a). Yield
44%; mp >280 °C; Ms (m/z): 436 [M+]; 1H NMR (400 MHz, DMSO-
d6): 7.63 (2H, d, J 7.4), 7.54 (2H, d, J 8.2), 7.49–7.39 (4H, m), 7.33
(1H, t, J 7.4), 7.15 (2H, d, J 8.2), 7.04 (1H, s), 6.88 (2H, d, J 8.9),
6.79 (1H, s), 3.89 (2H, s); 13C NMR (100 MHz, DMSO-d6): 176.7,
161.6, 159.6, 157.9, 152.5, 150.4, 140.6, 140.4, 138.1, 130.5,
129.3, 128.7, 127.6, 127.0, 126.9, 124.4, 117.6, 115.8, 115.3, 99.8,
97.3, 31.0. Anal. (C28H20O5) C, H.

M. Forghieri et al. / Bioorg. Med. Chem. 17 (2009) 2658–2672
2671
4.1.8.14. 300,40-Dihydroxy-1,10-biphenyl (18) (from 30a). Yield
44%; mp 126–8 °C (137–9 °C37); Ms (m/z): 186 [M+]; 1H NMR
(CDCl3): 7.55–7.46 (2H, m), 7.43–7.22 (3H, m), 7.02 (1H, d, J 2.2),
6.92 (1H, dd, J 2.2, J 8.2), 6.80 (1H, d, J 8.2). Anal. (C12H10O2) C, H.
10% bovine calf serum in a 5% CO2 humidified atmosphere. Subcon-
fluent cells were serum-starved for 20 h before receiving inhibitors
or insulin. Inhibitors ((10 lM final concentration) or insulin (10 nM
final concentration) were added to the culture medium and the
cells were incubated for 90 min at 37 °C. Cells were then lysed with
SDS Laemmly sample buffer and analyzed by SDS-PAGE and Wes-
tern Blot (anti-phosphorylated insulin receptor, anti-Akt and anti-
actin antibodies were purchased from Santa Cruz Biotechnology
Inc.; monoclonal anti-phosphotyrosine antibody was purchased
from Upstate Biotechnology Inc.).
4.1.8.15. 40-Carboxy-30-Hydroxy-1,10-biphenyl
(19)
(from
30b). Yield 56%; mp 202–4 °C (207–8 °C38); 1H NMR (DMSO-d6):
7.82 (1H, d, J 8.0), 7.70 (2H, d, J 7.7), 7.56–7.32 (3H, m), 7.24–
7.08 (2H, m). Anal. (C13H10O3) C, H.
4.1.8.16. 20-40-Dihydroxy-1,10-biphenyl (20) (from 30c). Yield
65%; mp 127–9 °C (145 °C39); Ms (m/z): 186 [M+]; 1H NMR (CDCl3):
7.60–7.30 (5H, m), 7.10 (1H, d, J 9.2), 6.50 (2H, m). Anal. (C12H10O2)
C, H.
Acknowledgement
Work supported in part by the Italian MIUR (FIRB-2003 project
RBNE03FMCJ_007).
4.2. Molecular modeling
All protein alignments, protein and ligand preparations, and
docking studies were performed using the Schrödinger Suite
2007, including MAESTRO 8.0, GLIDE 4.5, and PRIME 1.6,25 on a PC
equipped with a 2.13 GHz Core 2 Duo processor and 3 GB of
RAM, running Fedora Core 8 Linux. Figure 1 was created with PYMOL
v0.99.40 Pharmacophore feature detection was performed with
LIGANDSCOUT v2.01,41 which was also used to create the graphics in
Figure 2.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2009.02.060.
References and notes
1. Fischer, E. H. Adv. Enzyme Regul. 1999, 39, 359.
2. Hubbard, S. R.; Till, J. H. Annu. Rev. Biochem. 2000, 69, 373.
3. Hunter, T. Cell 2000, 100, 113.
4.3. Enzyme section
4. Neel, B. G.; Tonks, N. K. Curr. Opin. Cell Biol. 1997, 9, 193.
5. Andersen, J. N.; Mortensen, O. H.; Peters, G. H.; Drake, P. G.; Iversen, L. F.; Olsen,
O. H.; Jansen, P. G.; Andersen, H. S.; Tonks, N. K.; Moller, N. P. Mol. Cell Biol.
2001, 21, 7117.
6. Caselli, A.; Marzocchini, R.; Camici, G.; Manao, G.; Moneti, G.; Pieraccini, G.;
Ramponi, G. J. Biol. Chem. 1998, 273, 32554.
7. Wo, Y. Y. P.; McCormack, A. L.; Shabanowitz, J.; Hunt, D. F.; Davis, J. P.; Mitchell,
G. L.; Van Etten, R. L. J. Biol. Chem. 1992, 267, 10856.
8. Chiarugi, P.; Taddei, M. L.; Schiavone, N.; Papucci, L.; Giannoni, E.; Fiaschi, T.;
Capaccioli, S.; Raugei, G.; Ramponi, G. Oncogene 2004, 23, 3905.
9. Kikawa, K. D.; Vidale, D. R.; van Etten, R. L.; Kinch, M. S. J. Biol. Chem. 2002, 277,
39274.
The complete coding sequences of human LMW-PTP (IF1 and
IF2 isoforms) and human PTP-1B were cloned in frame with the se-
quence of the glutathione S-transferase (GST) in the p-GEX-2T bac-
terial expression vector. Enzyme expression and purification were
achieved in the E. coli TB1 strain.42 Briefly, the recombinant fusion
proteins were purified from bacterial lysate using a single-step
affinity chromatography. The solution containing purified fusion
proteins was treated with thrombin for 3 h at 37 °C. Then the en-
zymes were purified from GST and thrombin by gel filtration.
The purity of proteins preparations was analyzed by SDS-poly-
acrylamide gel electrophoresis according to Laemmli.43
10. Malentacchi, F.; Marzocchini, R.; Gelmini, S.; Orlando, C.; Serio, M.; Ramponi,
G.; Raugei, G. Biochem. Biophys. Res. Commun. 2005, 334, 875.
11. Chiarugi, P.; Cirri, P.; Marra, F.; Raugei, G.; Camici, G.; Manao, G.; Ramponi, G.
Biochem. Biophys. Res. Commun. 1997, 238, 676.
(a) Enzymatic assay and inihibition experiments. The enzy-
matic assay was carried out at 37 °C using p-nitrophenylphos-
phate as substrate; the final volume was 1 mL. The assay buffer
(pH 7.0) contained 0.075 M of 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-nitrophenol was determined by reading the absorbance at
12. Taddei, M. L.; Chiarugi, P.; Cirri, P.; Talini, D.; Camici, G.; Manao, G.; Raugei, G.;
Ramponi, G. Biochem. Biophys. Res. Commun. 2000, 270, 564.
13. Sarabu, R.; Tilley, J. Ann. Rep. Med. Chem. 2005, 40, 167.
14. Zhang, M.; Stauffacher, C. V.; Lin, D.; Van Etten, R. L. J. Biol. Chem. 1998, 273,
21714.
15. Bleasdale, J. E.; Ogg, D.; Palazuk, B. J.; Jacob, C. S.; Swanson, M. L.; Wang, X. Y.;
Thompson, D. P.; Conradi, R. A.; Mathews, W. R.; Laborde, A. L.; Stuchly, C. W.;
Heijbel, A.; Bergdahl, K.; Bannow, C. A.; Smith, C. W.; Svensson, C.; Liljebris, C.;
Schostarez, H. J.; May, P. D.; Stevens, F. C.;Larsen, S. D. Biochemistry 2001, 40, 5642.
16. Maccari, R.; Paoli, P.; Ottanà, R.; Jacomelli, M.; Ciurlo, R.; Manao, G.; Steindl, T.;
Langer, T.; Vigorita, M. G.; Camici, G. Bioorg. Med. Chem. 2007, 15, 5137.
17. Vidal, D.; Blobel, J.; Perez, Y.; Thormann, M.; Pons, M. Eur. J. Med. Chem. 2007,
42, 1102.
18. Miranda, M. A.; Okamoto, A. K.; Ferreira, C. V.; Silva, T. L.; Granjeiro, J. M.;
Aoyama, H. J. Enzyme Inhib. Med. Chem. 2006, 21, 419.
19. Wang, S.; Tabernero, L.; Zhang, M.; Harms, E.; Van Etten, R. L.; Stauffacher, C. V.
Biochemistry 2000, 39, 1903.
20. Zabell, A. P. R.; Corden, S.; Helquist, P.; Stauffacher, C. V.; Wiest, O. Bioorg. Med.
Chem. 2004, 12, 1867.
21. Ferrell, J. E.; Chang Sing, P. D.; Loew, G.; King, R.; Mansour, J. M.; Mansour, T. E.
Mol. Pharmacol. 1979, 16, 556.
22. Wang, S.; Stauffacher, C. V.; Van Etten, R. L. Biochemistry 2000, 39, 1234.
23. Chen, R. M.; Hu, L. H.; An, T. Y.; Li, J.; Shen, Q. Bioorg. Med. Chem. Lett. 2002, 12,
3387.
24. Costantino, L.; Rastelli, G.; Gamberini, M. C.; Vinson, J. A.; Bose, P.; Iannone, A.;
Staffieri, M.; Antolini, L.; Del Corso, A.; Mura, U.; Albasini, A. J. Med. Chem. 1999,
42, 1881.
25. Schrödinger Suite 2007; Schrödinger, LLC; New York, NY, USA, 2007.
26. Malamas, M. S.; Sredy, J.; Moxham, C.; Katz, A.; Xu, W.; McDevitt, R.; Adebayo,
F. O.; Sawicki, D. R.; Seestaller, L.; Sullivan, D.; Taylor, J. R. J. Med. Chem. 2000,
43, 1293.
27. Wolfbeis, O. S.; Leiner, M.; Hochmuth, P.; Geiger, H. Ber. Bunsenges. Phys. Chem.
1984, 88, 759.
400 nm (e
= 18,000 Mꢀ1 cmꢀ1). Km and Vmax were determined by
measuring the initial rates at different substrate concentrations.
Experimental data were analysed using the Michaelis equation
and a nonlinear fitting program (FIGSYS).
Inhibition constants were determined measuring initial hydro-
lysis rates at differing substrate and inhibitor concentrations. The
apparent Km values measured at the various inhibitor concentra-
tions were plotted against concentration of the inhibitor to calcu-
late the Ki values. All initial rate measurements were carried out
in triplicate. For each inhibitor, IC50 was determined by measuring
the initial hydrolysis rate under fixed p-nitrophenylphosphate con-
centration, equal to the Km value of the considered PTP. Data were
fitted to the following equation using the FIGSYS program:
Vi/V° = IC50/IC50 + [I]), where Vi is the reaction velocity when the
inhibitor concentration is [I], V° is the reaction velocity with no
inhibitor and IC50 = Ki + Ki[S]/Km. Therefore, when the substrate
concentrations [S] is equal to Km, IC50 = 2Ki.
(b) Cell-based Assays. HepG2 cells were routinely cultured at
28. Maccari, R.; Ottanà, R.; Ciurleo, R.; Rakowitz, D.; Matuszczak, B.; Laggner, C.;
Langer, T. Bioorg. Med. Chem. 2008, 16, 5840.
37 °C in Dulbecco0s modified Eagle0s medium supplemented with

2672
M. Forghieri et al. / Bioorg. Med. Chem. 17 (2009) 2658–2672
29. Steuber, H.; Zentgraf, M.; Podjarny, A.; Heine, A.; Klebe, G. J. Mol. Biol. 2006,
356, 45.
30. (a) Said, H.; Karickhoff, S. W.; Carreira, L. A. Quant. Struct.-Act. Relat. 1995, 14,
348; (b) SPARC on line calculator v4.1, http://ibmlc2.chem.uga.edu/sparc/,
accessed Jan. 2008.
36. Bargellini, G.; Monti, L. Gazz. Chim. Ital. 1914, 44, 25.
37. Lin, P. H.; Sangaiah, R.; Ranasinghe, A.; Ball, L. M.; Swenberg, J. A.; Gold, A. Org.
Biomol. Chem. 2004, 2, 2624.
38. Walker, J. J. Chem. Soc. 1939, 120.
39. Suter, C. M.; Perrin, G. S. J. Am. Chem. Soc. 1939, 61, 166.
40. PYMOL Molecular Graphics System v0.99, DeLano Scientific, Palo Alto, CA, USA,
2006.
31. Herrero-Martínez, J. M.; Sanmartin, M.; Rosés, M.; Bosch, E.; Ràfols, C.
Electrophoresis 2005, 26, 1886.
32. Zabell, A. P.; Schroff, A. D.; Bain, B. E.; Van Etten, R. L.; Wiest, O.; Stauffacher, C.
V. J. Biol. Chem. 2006, 281, 6520.
33. Sherman, W.; Day, T.; Jacobson, M. P.; Friesner, R. A.; Farid, R. J. Med. Chem.
2006, 49, 534.
41. (a) Wolber, G.; Dornhofer, A.; Langer, T. J. Comput. Aided Mol. Des.
2006, 20, 773; (b) Wolber, G.; Langer, T. J. Chem. Inf. Comput. Sci.
2005, 45, 160–169; (c) LIGANDSCOUT 2.0; Inte:Ligand GmbH, Vienna,
Austria, 2008.
34. Zhang, M.; Zhou, M.; Van Etten, R. L.; Stauffacher, C. V. Biochemistry 1997, 36, 15.
35. Zhao, P. L.; Liu, C. L.; Huang, W.; Wang, Y. Z.; Yang, G. F. J. Agric. Food Chem.
2007, 55, 5697.
42. Marzocchini, R.; Bucciantini, M.; Stefani, M.; Taddei, N.; Thunnissen, M. G. M.
M.; Nordlund, P.; Ramponi, G. FEBS Lett. 1998, 426, 52.
43. Laemmli, U. K. Nature 1970, 227, 680.