Insight into 2-phenylpyrazolo[1,5-a]pyrimidin-3-yl acetamides as peripheral benzodiazepine receptor ligands: Synthesis, biological evaluation and 3D-QSAR investigation

Article abstract of DOI:10.1016/j.bmc.2005.05.015

The present paper reports the synthesis and binding studies of new 2-phenylpyrazolo[1,5-a]pyrimidin-3-yl acetamides as selective Peripheral Benzodiazepine Receptor (PBR) ligands. The variability of substituents at the 3-position was investigated and a 3D-

Full text of DOI:10.1016/j.bmc.2005.05.015

Bioorganic & Medicinal Chemistry 13 (2005) 4821–4834
Insight into 2-phenylpyrazolo[1,5-a]pyrimidin-3-yl acetamides
as peripheral benzodiazepine receptor ligands: Synthesis,
biological evaluation and 3D-QSAR investigation
Silvia Selleri,a Paola Gratteri,a,* Camilla Costagli,a Claudia Bonaccini,a
Annarella Costanzo,a Fabrizio Melani,a Gabriella Guerrini,a Giovanna Ciciani,a
Barbara Costa,b Francesca Spinetti,b Claudia Martinib and Fabrizio Brunia
a
`
Dipartimento di Scienze Farmaceutiche, Universita di Firenze, Via U. Schiff, 6, 50019 Sesto F.no Firenze, Italy
b
`
Dipartimento di Psichiatria, Neurobiologia, Farmacologia e Biotecnologie, Universita di Pisa, Via Bonanno 6, 56126 Pisa, Italy
Received 18 October 2004; revised 4 May 2005; accepted 10 May 2005
Available online 21 June 2005
Abstract—The present paper reports the synthesis and binding studies of new 2-phenylpyrazolo[1,5-a]pyrimidin-3-yl acetamides as
selective Peripheral Benzodiazepine Receptor (PBR) ligands. The variability of substituents at the 3-position was investigated and a
3D-QSAR model was proposed to evaluate the effect of different substitutions on the acetamide moiety. In addition, a subset of the
novel compounds showing high affinity for PBR was tested for their ability to modulate the steroid biosynthesis in C6 glioma cells.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
The PBR is implied in numerous biological processes,
even if the precise function of this receptor remains an
enigma; however, considering its ubiquitous distribution
and high conservation in the evolutionary scale, it is sug-
gested that PBR may serve as a target to control various
mitochondrial and cell functions and probably to pro-
tect against toxic events.6,7 In addition, binding studies
indicated that PBR density can be modulated under a
variety of physiological and pathological conditions.8
However to date, among the many functions attributed
to the PBR, its involvement in biosynthesis of steroids is
the most investigated.9–13 The PBR is involved in the
transport of cholesterol into mitochondria, which repre-
sents the rate-limiting step of the whole steroid bio-
synthesis process. Steroidogenesis begins with the
conversion of cholesterol to pregnenolone by the cyto-
chrome P-450 side chain cleavage (P450scc), which is
located at the inner mitochondrial membrane. Most
studies report that PBR ligands stimulate steroids pro-
duction.12 Nonetheless, a polypeptide of seven amino
acids was identified as a new PBR ligand able to antag-
onize the steroid biosynthesis and it could represent a
useful lead for synthesizing new antagonist PBR ligands,
the target of the current research.14
The peripheral benzodiazepine receptor (PBR) is a pro-
tein of 18 kDa, which can form a trimeric complex with
the voltage dependent anion channel VDAC (32 kDa)
and adenine nucleotide carrier ANC (30 kDa).1 Two
additional proteins (10 kDa and PRAX-1) are believed
to be involved in modulating PBR functions.2 The
18 kDa subunit was identified, cloned and expressed,
initially in rat and subsequently in mouse, bovine and
human species. Among these, the 18 kDa protein, con-
sisting of 169 amino acids, shows about 80% amino acid
conservation.3
The PBR was found in the majority of peripheral tis-
sues, in blood cells and also in astrocytes, even if the
density of the PBR is the highest in endocrine tissues
such as adrenal gland, testis and ovary.4 This receptor
is mainly located in the outer membrane of mitochon-
dria, but there is evidence of its presence elsewhere in
the cell.5
Keywords: Peripheral benzodiazepine receptor; Pyrazolo[1,5-a]pyrim-
idines; 3D QSAR; Steroidogenesis assay.
*
Nowadays, a wide variety of endogenous molecules with
affinity for the PBR have been identified, such as DBI
Corresponding author. Tel.: +39 55 457302; fax: +39 55 4573671;
e-mail: paola.gratteri@unifi.it
0968-0896/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2005.05.015

4822
S. Selleri et al. / Bioorg. Med. Chem. 13 (2005) 4821–4834
(Diazepam Binding Inhibitor), which is a 11 kDa poly-
peptide of 86 amino acids, and porphyrins (protopor-
phyrin IX, mesoporphyrin IX, deuteroporphyrin IX
and hemin).2
In addition, the PBR exhibits nanomolar affinity to ben-
zodiazepine Ro5-4864 (40-chlorodiazepam), but very low
affinity to other benzodiazepines; moreover, isoquino-
lines (PK11195), imidazopyrimidines, indole derivatives,
pirrolobenzoxazepines and phenoxyphenyl acetamides
have been recognized as exogenous PBR ligands. Partic-
ularly, it is reported that isoquinolines, such as PK11195,
interact specifically with the 18 kDa protein component,
whereas PBR-specific benzodiazepines seem to require
the interaction of all three protein components
(18 kDa, VDAC, ANC) for binding. Indeed, the well-
known ligands PK11195 and Ro5-4864 (Chart 1) com-
pete with each other in binding experiments, suggesting
overlapping but not identical binding sites.15–17
In addition to site-direct mutagenesis studies,15 useful
information on the PBR are supplied by several compu-
tational analyses.18–21 These try to rationalize the varia-
tion in the binding affinity of the large variety of ligands
and to identify the crucial amino acids involved in the
binding of synthetic and endogenous ligands. All exist-
ing QSAR studies in the literature agree with the need
for two hydrophobic interaction areas and a H-bond
donating site in the PBR receptor as essential require-
ments for the ligand binding. Among the proposed
models only a few are derived starting from series of
structurally different compounds;18 most of them are ob-
tained on congeneric series of molecules.19–21 However,
with some limitations (which mainly refer to the local
application on the series from which they are derived),
these models represent a reliable tool in both the
description of the molecules used to build the model
and in the prediction of the biological activities of ana-
logues not yet synthesized.
Chart 2.
Mainly, the 5 and 7 positions were investigated, and
from preliminary SAR studies it emerged that substitu-
tions on the pyrimidine moiety seem to be the key factor
promoting the PBR versus CBR (Central Benzodiaze-
pine Receptor) selectivity. Recently, as an extension of
our studies on pyrazolo[1,5-a]pyrimidines as PBR li-
gands, our attention has been focused on the 3 position
to clarify the role of the amide moiety and the main
structural features concerning the number and the
length of the alkyl substituents on the amide nitrogen.
Indeed, it has been reported18–21 that hydrogen bond
interaction of a polar group with the receptor protein
seems to be of paramount importance for the binding
to the PBR. In this paper, we report the synthesis of
new 2-phenyl-5,7-dimethylpyrazolo[1,5-a]pyrimidin-3-yl
acetamides and their biological evaluation in order to
investigate PBR affinity and PBR versus CBR selectiv-
ity. In addition, a 3D-QSAR model using the GRID/
Golpe23 procedure is proposed to evaluate the effect of
the substitution on acetamide chain. Previously pub-
lished compounds are also included (Table 1).
In a previous paper,22 we described the results of our re-
search on PBR ligands, which led to discovery of a new
class of 2-arylpyrazolo[1,5-a]pyrimidin-3-yl acetamides,
azaisosters of Alpidem, endowed with high affinity and
PBR selectivity (Chart 2).
A subset of the synthesized compounds showing high
affinity and selectivity for PBR was also evaluated for
their ability to modulate the steroid biosynthesis in C6
glioma cells, since this test is currently the only reliable
method to prove the affinity and/or intrinsic activity of
PBR ligands.
2. Chemistry
The synthesis of compounds 3b and 4a–q (Scheme 1)
started from the common intermediate, benzoylacetonit-
rile, which reacted in basic medium (LiOHÆH2O)
with iodoacetic acid or ethyl iodoacetate to give the
Chart 1.

S. Selleri et al. / Bioorg. Med. Chem. 13 (2005) 4821–4834
4823
Table 1. Structures and PBR/CBR affinity values of all compounds reported in this paper
Compound
n
X
Y
Ar
R5
R6
R7
R0
R00
PBR Ki (nM)a CBR Ki (nM)b
3fc
3gc
3ic
3jc
3kc
3lc
3mc
3nc
3oc
3pc
3qc
3rc
3sc
3tc
3uc
3vc
3xc
3yc
4d
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
O
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
S
Ph
Ph
Me
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Me
H
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Me
40
92
518
Me
Me
Me
Me
Me
Me
H
155
29
Ph
Me
Me
Me
Me
251
4Cl–Ph
4F–Ph
4Me–Ph
2.4
9.2
0.8
4.7
25
>10.00
>10.00
>10.00
>10.00
85
4MeO–Ph Me
4F–Ph
Me
H
4Cl–Ph
4Me–Ph
4Me–Ph
Me
8.1
0.9
1
207
Me
CF3
CF3 Et
Me Et
CF3 Et
5800
6900
4MeO–Ph CF3
16
6260
4Cl–Ph
4Cl–Ph
4Cl–Ph
4Cl–Ph
4Me–Ph
Ph
Me
Ph
Et
2.4
3.4
6.1
8.4
57
>10.00
>10.00
>10.00
>10.00
>10.00
>10.00
255
Me
Me
Et
Me
Me
H
Et
Et
COOEt Me
Ph
H
H
H
H
H
Me
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Et
Ph
H
Et
225
1192
1610
710
0.8
89
4Me-Ph
Ph
Ph
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
H
Me
Et
3b
4a
4b
4c
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
>10.00
>10.00
>10.00
10.00
10.00
10.00
10.00
10.00
>10.00
10.00
10.00
10.00
10.00
10.00
>10.00
>10.00
10.00
10.00
10.00
>10.00
Me
n-Pr
i-Pr
n-But
Me
Ph
n-Pr
i-Pr
Ph
Ph
4d
4e
n-But
4.5
8
Ph
n-pentyl n-pentyl
4f
Ph
n-hexyl
n-octyl
Bz
n-hexyl
n-octyl
Bz
23
4g
4h
4i
Ph
Ph
1124
40
Ph
Et
i-Pr
28
4j
Ph
Et
Ph
0.8
7.3
22
4k
4l
Ph
Ph
Et
Me
Bz
(R)-CHCH3Ph
(S)-CHCH3Ph
–(CH2)4–
–(CH2)5–
4m
4n
4o
4p
4q
5
Ph
Me
156
1154
304
Ph
Ph
Ph
–CH2CH2N(Me)CH2CH2– >10,000
376
Ph
n-Pr
Et
H
Ph
Ph
Et
Et
44
270e
8
O
Et
Italic entries represent compounds in the training set.
a Ki values represent means SEM derived from three independent experiments, conducted in triplicate using 0.6 nM [3H]PK11195 as the
radioligand.
b Percent inhibition values of specific [3H]Ro15-1788 binding at 0.2 nM concentration are means SEM of five separate experiments, each done in
triplicate.
c See Ref. 22.
d See Ref. 35.
e Affinity values were calculated using 1 nM [3H]Ro 5-4864 as the radioligand.
corresponding acid (1a) and ester (1b), respectively. The
achievement of compounds 1a and b was performed by
exploiting the base-promoted reactivity of methylene
group of the benzoylacetonitrile, which reacts with elec-

4824
S. Selleri et al. / Bioorg. Med. Chem. 13 (2005) 4821–4834
Scheme 1. Reagents and conditions: (i) iodoacetic acid/LiOHÆH2O/EtOH 80% (method A), ethyl iodoacetate/thallium ethylate/tol (method B);
(ii) N2H4ÆH2O/EtOH/AcOH; (iii) 2,4-pentanedione/EtOH; (iv) NEt3/ClCOOEt/THF/suitable amines reported in Experimental section.
tron-poor carbon atom to give 1,4-dicarbonyl
compounds.
phosphine, which generates triarylphosphonium chlo-
ride,27 and the method which uses N-halosuccinimides
(i.e., NCS and NBS) in combination with triphenylphos-
phine.28 These methods represent two valid approaches
for converting pyrazolo[1,5-a]pyrimidin-3-yl acetic acids
to the corresponding amides; nevertheless, the best
yields were obtained applying the method via mixed
anhydride.
Compound 1a was obtained in good yield (70%) using
LiOHÆH2O as base in 80% EtOH (method A), whereas
compound 1b was synthesized employing ethyl iodoace-
tate in the presence of thallium ethoxide in toluene
(method B).24 A previously published method for the
synthesis of the ester derivative was followed,25 using
Hunigꢀs base (DIEA, diisopropylethylamine) and anhy-
drous magnesium chloride in dichloromethane. Never-
theless, following this procedure the yield of
compound 1b resulted unsatisfactorily; therefore the
alternative method B was used.
Compound 5 was synthesized from N,N-diethyl-5,7-di-
methyl-2-phenylpyrazolo[1,5-a]pyrimidin-3-yl acetam-
ide22 by reaction with Lawessonꢀs reagent (Scheme 2);
compound 8 (Scheme 3) was prepared starting from
5,7-dimethyl-2-phenylpyrazolo[1,5-a]pyrimidine-3-car-
bonitrile29 by treatment with H2SO4 to provide the cor-
responding amide (6), which was further hydrolyzed to
the corresponding carboxylic acid. Finally, compound
7 was converted to the N,N-diethylcarboxyamide (8) fol-
lowing the above mentioned procedure for the synthesis
of compounds 4a–q.
In the second step of this synthetic route, the intermedi-
ates 1a and b reacted in ethanol at reflux with hydrazine
hydrate, in the presence of acetic acid, to give the corre-
sponding 3-amino pyrazoles (2a and b) in good yield and
purity.
In the third step, the condensation of compounds 2a
and b with 2,4-pentanedione was carried out, which
led to the closure of the pyrimidine ring, resulting in
the intermediate 3a and the final compound 3b. Subse-
quently, the acid 3a was converted into a mixed anhy-
dride with ethyl chloroformate, and this intermediate
reacted with a large number of amines affording the
corresponding amides (4a–q) in good yields and in a
short time.
3. Results and discussion
The ability of compounds 3b, 4a–q, 5 and 8 to interact
with the PBR was investigated by a binding assay using
[3H]PK11195 as the radioligand and membranes from
Several attempts were made to convert the acid 3a into
its chlorocarbonyl derivative. However, thionyl chloride
and diethylchlorophosphate,26 which are generally used
for preparing acyl chlorides from carboxylic acids, can-
not be used in this case due to the generation of strong
acid conditions during the reaction, which probably de-
stroy the bicyclic nucleus. Therefore, we considered the
reaction of trichloroacetonitrile (TCA) with triphenyl-
Scheme 2.

S. Selleri et al. / Bioorg. Med. Chem. 13 (2005) 4821–4834
4825
Scheme 3.
rat kidney tissues as receptor source; the tested com-
pounds display a broad range of binding affinities as
shown in Table 1. The newly synthesized PBR ligands
have also been examined for their receptor selectivity
by evaluating their ability to displace [3H]Ro15-1788
from Bz/GABAA receptors and none of the compounds
were found to exhibit any significant binding affinity for
the CBR. The lack of CBR recognition observed in all
the synthesized compounds seems to confirm our previ-
ous hypothesis that the substitution on the pyrimidine
moiety could be the key in promoting the PBR selectiv-
ity.22 In fact, in designing the new PBR ligands, we
started from the lead compound, N,N-diethyl-2-phen-
yl-5,7-dimethylpyrazolo[1,5-a]pyrimidin-3-yl acetamide
(3i),22 and we decided to maintain the phenyl ring at
the 2-position and two methyl groups at the 5- and 7-po-
sitions of the pyrazolo[1,5-a]pyrimidine nucleus because
this di-substitution on pyrimidine ring resulted in
favouring the PBR versus CBR selectivity. In this paper,
our attention is focused on the 3-position of the bicyclic
nucleus to investigate the role of the amide moiety and
the requirements concerning the number and the length
of the alkyl substituents on the amide nitrogen.
amides led to compounds which exhibit low affinity
(4n and 4o) or even lack of recognition at the PBR
(4p). On the other hand, the introduction of phenyl ring
on the amide nitrogen positively contributes to increase
the binding affinity (4j), while the presence of spacers be-
tween amide nitrogen and the aromatic ring results in
decreased affinity values (4h, 4k, 4l and 4m). Interesting-
ly, the insertion of one asymmetric carbon atom in the
acetamide moiety determines a very strong difference
in binding affinity between the stereoisomers 4l and 4m
and this evidence suggests that suitably oriented lipo-
phylic substituents play a substantial role in binding
affinity. This observation is in good agreement with
the only available studies on stereoselectivity of
PK11195 published between the end of the 1980s and
199432–34 in which the R-stereoisomer was observed to
be approximately 2.5-fold more potent than its enantio-
mer. This stereoselectivity was observed in all the tissues
tested (brain, heart, kidney and adrenals) and it was
found also in vivo, suggesting that the use of R-enantio-
mer would have advantages over the use of the racemate
currently employed.
In the series of pyrazolo[1,5-a]pyrimidines, removal of
the methylene linker between the pyrazolopyrimidine
nucleus and the amide group (compound 8) as well as
the replacement of this latter with the ester function
(3b), resulted in a notable decrease in binding potency.31
At first glance, this result seems to confirm the essential
role of the 3-acetamide moiety for PBR interaction.
Nevertheless, the maintenance of good affinity value ex-
erted by compound 5, bearing the tioacetamide moiety
which is unable to stabilize hydrogen bond interaction,
forces revision of the role of the acetamido group at
the 3-position. It could be hypothesized that the acetam-
ide moiety might not be essential in stabilizing the
H-bond with the receptor protein, but it represents an
optimal carrier because its length and orientation allow
lipophylic groups to fit with the proper lipophylic
Analyzing the binding results, it emerged that the chain
length of alkyl groups on the amide nitrogen is impor-
tant for binding affinity, as can be observed in
compounds 4a (methyl), 3i (ethyl),22 4b (n-propyl), 4d
(n-butyl), 4e (n-pentyl), 4f (n-hexyl) and 4g (n-octyl).
Among these compounds, the best result was shown
by compound 4b bearing three carbon atoms in the
amide moiety. Branching of the chain was unfavourable,
as is shown by compounds 4c and 4i. Moreover, the di-
substitution of the amide nitrogen is clearly preferred;
indeed 4q, bearing a single n-propyl group, shows a
low PBR binding affinity. Comparable results have been
obtained with the introduction of cyclic amines.30,31 In
fact, the attempts to explore stiffening of alkyl substitu-
ents on the amide nitrogen with the synthesis of cyclic

4826
S. Selleri et al. / Bioorg. Med. Chem. 13 (2005) 4821–4834
pockets on the receptor protein. Moreover, it should be
considered that the N4 atom may represent an alterna-
tive H-bond acceptor group in this series, if the group
at the 3-position is unable to form this interaction.
affinity (a low pKi value), while negative PLS coeffi-
cients (blue zones) correlate with a decrease in affinity
(a high pKi value).
3.1.1. N1 probe. In the whole dataset, there are no signif-
icant variations in the capacity to accept H-bond, since
the number, nature and position of these groups are not
so different from one molecule to another. Nevertheless,
the N1 probe groups the dataset molecules in three main
classes describing them in similar ways depending on the
basis of the flexible chain in the 3-position: this observa-
tion could be interpreted as a raw steric effect of the sub-
stitution in this position that influences the ability of the
heteroatoms to interact with the N1 probe. Both the ali-
phatic (higher than methyl without distinction) and aro-
matic amides, forming the most populated group, are
represented in Figure 1c by molecule 4b which has the
highest affinity value. The less active compounds are
divided into two other groups: one, represented in Fig-
ure 1d by molecule 3b, comprising the monosubstituted
amide (4q), the ester (3b) and the dimethylamides (4, 4a)
and the other for the cyclized amides (4n, 4o and 4p) in
Figure 1e by molecule 4p, the least active compound of
the whole dataset.
3.1. 3D-QSAR analysis
All the 38 pyrazolopyrimidines reported in Table 1 (19
compounds presented in this paper plus 18 already pub-
lished,22 together with compound 435) fulfill the require-
ments for the PBR recognition, that is, the ability to
interact by a H-bond with the receptor protein and the
presence of hydrophobic groups. This series of PBR
ligands was investigated according to a Grid/Golpe
procedure focusing our attention on evaluating the
effects of the substitutions on the acetamide chain.
Probes C3 and N1 were selected for the Grid/Golpe
analysis. The DRY probe provides almost the same
information as the C3 probe, nevertheless it was in-
cluded in the model derivation since it accounts for
the overall aromaticity of this class of PBR ligands.
Figures 1a–e and 2a–e show contour maps of the PLS
(Partial Least Squares) coefficients and the molecular
interaction fields (MIFs) for N1 and C3 probes.
Due to the elimination of the repulsive (positive)
interactions (see Section 5), only negative values cor-
responding to attractive interactions are present in
our model. In the figures, blue and yellow areas cor-
respond to negative and positive values of the para-
meter (PLS coefficient or MIFs, respectively. Positive
areas (yellow) in the PLS coefficient plot indicate re-
gions where a negative favourable interaction with
the probe (blue MIFs) correlates with an increase in
PLS coefficients contour plots highlight two blue (A, B)
and one yellow region (C), which indicate interesting
areas for the description of activity variations within
the series. The wide blue region A corresponds to wide
attractive interactions with the probe in the case of
low affinity compounds 3b and 4p (region A in Figs.
1d and e), while this region has limited dimensions in
the MIF of the active compound 4b (Fig. 1c). The
favourable interaction in this zone, which is mainly
due to the presence of the heteroatom adjacent to the
Figure 1. Probe N1 contour plots for negative (a) and positive (b) PLS coefficients together with the molecular interaction fields (MIFs) for molecules
4b (c), 3b (d), and 4p (e). A–C refer to areas pointed out by the model as important for affinity.

S. Selleri et al. / Bioorg. Med. Chem. 13 (2005) 4821–4834
4827
Figure 2. Probe C3 contour plots for negative (a) and positive, (b) PLS coefficients together with the molecular interaction fields (MIFs) for
molecules 4b (c), 4g (d), and 4p (e). A–E refer to areas pointed out by the model as important for affinity.
carbonyl group, is caused by the shielding effect of the
alkyl substituents on the heteroatom: the more extended
they are, the more active the compound is.
If molecules occupy the A region in C3-MIFs (com-
pounds 4b and 4g in Figs. 2c and d, respectively), they
prevent an unfavourable interaction with the probe.
The model indicates a limit to the length of the chain
in this position, fixed between the butylic and the pen-
tylic chain, since higher homologues show attractive
interactions in region C, which is again not favourable
for the activity.
The extension of the B regions in N1-MIFs is directly re-
lated to the steric hindrance of the alkyl substituents on
the acetamidic moiety and it is inversely related to the
PBR affinity. The shielding effect of dialkyl substitutions
greater than diethyl group (Fig. 1c) partially prevents
the interaction between the carbonyl oxygen and the
N1 probe, which is wider for the less active compounds
bearing short dialkyl and monoalkyl chains (Fig. 1d) as
well as cyclic amides (Fig. 1e).
On the contrary, the occupation of area B in C3-MIFs
by the molecule prevents favourable interactions with
the probe (compound 4g in Fig. 2d) in a zone of positive
PLS coefficients (Fig. 2b). This could explain the pro-
gressive reduction of affinity passing from 4b (propyl)
to 4g (octyl), since it is possible for alkyl groups higher
than pentyl to interact with the C3 probe in area D,
which is associated to negative values of the PLS
coefficients.
As concerns the PLS coefficient region C, this is specif-
ically occupied by the less active compounds that are
completely unable to interact with the probe at this level
(Figs. 1d and e); molecules bearing more extended ali-
phatic or aromatic amide chains (Fig. 1c) that are able
to widen as a consequence of the steric repulsion always
leave this area empty, thus making the interaction with
the N1 probe possible. Such weak interactions are not
shown in the figures due to the high energy threshold
used to represent MIFs. The use of a lower energy cutoff
would have made plot interpretation difficult. However,
more insight into this aspect will be considered in the C3
probe discussion.
Region E of Figure 2b corresponds to region C in Figure
1b. Due to the average lower energy values associated to
the C3-probe, the energy cutoff used for the representa-
tion of the C3-MIFs shows significant polyhedra in cor-
respondence of area E for molecules 4b and 4g (Figs. 2c
and d, respectively), which were not visible in Figure 1c
and are not present in the MIFs of compound 4p (Figs.
1e and 2e).
3.1.2. C3 probe. With respect to N1 probe, C3 is more
able to discriminate molecules within the aforemen-
tioned classes. In particular, it seems to describe a sort
of difference in amide chain features, supporting the
hypothesis of the presence of cavities in PBR binding
site with different steric tolerance, where the two alkyl
substituents could be accommodated.36,21 Such conclu-
sions can be derived comparing both regions A and B
of the PLS coefficient plots (Figs. 2a and 2b,
respectively).
Branching (compound 4c) or cyclization of the alkyl
chains (compound 4p, Fig. 2e) results in a decreased
PBR affinity due to the presence of the interaction area
A in C3-MIFs, associated to negative PLS coefficients
(Fig. 2a), together with the absence of favourable inter-
actions in regions B and E (Fig. 2e).
The 3D-QSAR analysis provides an interpretation of
the significant difference in the activity of the enantio-
mers 4l (Fig. 3a) and 4m (Fig. 3b): in fact, the more

4828
S. Selleri et al. / Bioorg. Med. Chem. 13 (2005) 4821–4834
Figure 4. The effect of 2-arylpyrazolo[1,5-a]pyrimidin-3-yl)acetamides
on the pregnenolone accumulation in C6 glioma rat cells. All the
compounds were used at the same concentration (40 lM); at the end of
the incubation period (2 h), the amount of pregnenolone was quan-
tified by radio immunoassay (RIA), using an antibody obtained from
ICN Biochemical Inc., CA, USA. The values are means SEM of at
least three determinations (see Section 5).
in C6 glioma cells and they were previously shown to
elicit similar effects in MA10 Leyding cells and Y-1
cells.38 Among the tested compounds, in spite of compa-
rable binding affinity, a different effect on steroids pro-
duction can be observed.
Figure 3. Probe C3 molecular interaction fields (MIFS) for molecules
4l (a) and 4m (b). A, B and E refer to areas pointed out by the model as
important for affinity.
In particular 4b, which does not increase the level of
pregnenolone, warrants further investigation in order
to understand its possible antagonist profile. On the
other hand, compounds 4d and 4j appear to function
better than PK11195 and Ro5-4864 in regard to stimu-
lation of pregnenolone production.
active R enantiomer presents extended B and E areas,
associated to an increase in activity, while S enantiomer
shows a wide area of interaction mainly in region A,
unfavourable for affinity.
3.2. Steroidogenesis assay
4. Conclusions
Nowadays, investigation of PBR ligands activity is basi-
cally restricted to steroidogenesis assays,37–39 even if it is
well known that the PBR is involved in many cellular
functions related to mitochondrial mechanisms (oxida-
tive processes, cellular proliferation and apoptosis).
Considering this observation, the inadequacy of the bio-
logical and pharmacological methods for evaluating the
PBR ligands functionalities is revealed. In addition to
this gap, the scanty use of PBR ligands different from
the classic benzodiazepines (Ro5-4864) and PK11195
in steroidogenesis in vivo assays, which have already
shown many discrepancies in acute and chronic treat-
ments in rodents, should be kept in mind.12 Therefore,
the individuation of new PBR ligands endowed with dif-
ferent intrinsic activity in steroidogenesis assays in com-
parison to the well known Ro5-4864 and PK11195 may
afford useful new opportunities to investigate the PBR
functionality.The most potent compounds (4b, 4d and
4j) were examined for their ability to modulate steroid
biosynthesis in C6 glioma rat cells (Fig. 4), which can
represent a preliminary evaluation of the intrinsic activ-
ity of the PBR ligands. As shown in Figure 4, both
PK11195 and Ro5-4864 stimulate steroid biosynthesis
The series described in this paper represents a useful
extension of pyrazolo[1,5-a]pirimidin-3-yl acetamides
as PBR ligands, which offer a wide margin of modelling
to identify a PBR pharmacophore. In addition to pro-
viding a pharmacophore model, we have focused our
attention on the structural differences present on a
molecular fragment within the series of the studied pyr-
azolopyrimidines, namely the acetamide chains. Accord-
ing to the suggestions provided by the GRID/Golpe
model, work is currently in progress to further investi-
gate these aspects through the design of new compounds
bearing different substitutions on the amide nitrogen.
Moreover, the GRID/Golpe model is able to explain
in terms of hydrophobic interactions also the features
leading to stereoselectivity at PBR binding site, thus
supporting the idea of an asymmetric site with two quite
different cavities.
Furthermore, the PBR ligands 4d and j presented in this
paper could be added to the classical benzodiazepines
and isoquinolinecarboxyamides as tools in the investiga-
tion of PBR functionality. It is also noteworthy that our
PBR ligand, 4b, may present itself as an antagonist

S. Selleri et al. / Bioorg. Med. Chem. 13 (2005) 4821–4834
4829
against PK 11195 and Ro5-4864 regarding
steroidogenesis.
5.4. General procedure for the synthesis of compounds
(2a–b)
To a solution of 1a–b (10 mmol) in EtOH (50 mL),
hydrazine hydrate (20 mmol, 0.97 mL) and acetic acid
(1 mL) were added and the reaction mixture was re-
fluxed for 6 h. After cooling, evaporation of the solvent,
under reduced pressure, gave a residue that was purified
by silica gel column chromatography [CHCl3/MeOH,
10:1 v/v, as eluent].
5. Experimental
5.1. Chemistry
The structures of all compounds were supported by
their H NMR data (measured with a Varian Gemini
1
at 200 MHz, chemical shifts expressed in d (ppm)
using DMSO-d6 or CDCl3 as solvent). The following
abbreviations are used: b, broad; d, doublet; m, multi-
plet; q, quartet; t, triplet; and s, singlet. Melting points
were determined with a Gallenkamp apparatus an-
dwere uncorrected. Elemental analyses were performed
by the laboratories of Dipartimento Farmaco-Chim-
5.5. 3-Amino-5-phenylpyrazol-4-yl acetic acid (2a)
Ivory crystals from water; obtained from compound 1a;
yield 38%; mp 200–201 ꢁC. 1H NMR (DMSO-d6) d: 3.38
(s, 2H, CH2), 4.87 (br s, 2H, NH2 exchangeable), 7.32–
7.59 (m, 5H, Ph). Anal. Calcd for C11H11N3O2: C, 60.82;
H, 5.10; N, 19.34. Found: C, 60.70; H, 4.97; N, 19.15.
`
ico-Tecnologico of Universita di Siena, Italy, with a
Perkin-Elmer model 240C, elemental analyzer, and
their results are within 0.4% of theoretical values.
The purity of samples was determined by means of
TLC, which was performed using Machery-Nagel Du-
ren, Alugram silica gel plates. Silica gel 60 (Merck 70–
230 mesh) was used for column chromatography.
5.6. Ethyl 3-amino-5-phenylpyrazol-4-yl acetate (2b)
Yellow liquid obtained from compound 1b, yield
41%. 1H NMR (CDCl3) d: 1.30 (t, 3H, OCH2CH3),
3.46 (s, 2H, CH2), 4.20 (q, 2H, OCH2CH3), 5.80
(br s, 2H, NH2 exchangeable), 7.42–7.54 (m, 3H,
Ph), 8.09–8.12 (m, 2H, Ph). Anal. Calcd for
C13H15N3O2: C, 63.66; H, 6.16; N, 17.13. Found:
C, 63.39; H, 5.97; N, 17.15.
5.2. 3-Cyano-4-oxo-4-phenyl butanoic acid (1a)
To a suspension of LiOHÆH2O (10 mmol) in EtOH
(50 mL) was added 3-oxo-3-phenylpropanenitrile
(10 mmol) previously dissolved in EtOH (30 mL) and
subsequently a solution of iodoacetic acid (12.5 mmol)
and LiOHÆH2O (12.5 mmol) in 70% EtOH (50 mL)
dripped into the reaction mixture; it was refluxed under
magnetic stirring for 6 h. After cooling, evaporation of
the solvent under reduced pressure gave a residue, that
was treated with ice-water (100 mL) and the resulting
mixture was washed with diethyl ether (60 mL ·3). Acid-
ification of the aqueous phase with concentrated hydro-
chloric acid causes the separation of a solid product,
which was isolated by filtration. Ivory crystals from
5.7. General procedure for the synthesis of compounds 3a
and b
To a solution of 2a and b (1 mmol) in EtOH (5 mL),
2,4-pentanedione (1 mmol) was added and the mix-
ture was refluxed under stirring for 2 h. The progress
of reaction was monitored by TLC. The solvent was
evaporated under reduced pressure and the residue
was purified by silica gel column chromatography
to give the corresponding pyrazolopyrimidines 3a
and b.
1
H2O; yield 32%; mp 143–144 ꢁC; H NMR (CDCl3) d:
3.00 (dd, 1H, CH2), 3.29 (dd, 1H, CH2), 4.70–4.81 (m,
1H, CH), 7.42–7.77 (m, 3H, Ph), 8.0–8.15 (m, 2H, Ph),
8.70 (br s, 1H, OH exchangeable). Anal. Calcd for
C11H9NO3: C, 65.02; H, 4.46; N, 6.89. Found: C,
65.13; H, 4.18; N, 6.55.
5.8. 5,7-Dimethyl-2-phenylpyrazolo[1,5-a]pyrimidin-3-yl
acetic acid (3a)
This compound was obtained from 2a and purified by
column chromatography (CHCl3/MeOH, 20:3 v/v as
eluent); yield 83%; white crystals; mp 231–232 ꢁC. H
NMR (DMSO-d6) d: 2.46 (s, 3H, 5-CH3), 2.78 (s, 3H,
7-CH3), 3.80 (s, 2H, CH2), 6.95 (s, 1H, H-6), 7.41–7.48
(m, 3H, Ph), 7.78–7.82 (m, 2H, Ph), 12.4 (br s, 1H,
OH exchangeable). Anal. Calcd for C16H15N3O2: C,
68.31; H, 5.37; N, 14.94. Found: C, 68.68; H, 5.58; N,
14.99.
1
5.3. Ethyl 3-cyano-4-oxo-4-phenyl butanoate (1b)
To
a
solution of 3-oxo-3-phenylpropanenitrile
(10 mmol) in toluene (80 mL) were added ethyl iodoac-
etate (15 mmol) and TlOEt (10.5 mmol) under magnetic
stirring. The reaction mixture was maintained at 70 ꢁC
for 3 h. After cooling, the inorganic materials were fil-
tered off and the evaporation of the filtrate, under re-
duced pressure, gave a residue that was purified by
silica gel column chromatography [toluene/ethyl acetate,
5.9. Ethyl 5,7-dimethyl-2-phenylpyrazolo[1,5-a]pyrimidin-
3-yl acetate (3b)
1
8/3 (v/v), as eluent]. Orange liquid; yield 30%; H NMR
This compound was obtained from 2b; white crystals
from EtOH; yield 25%; mp 104–105 ꢁC. 1H NMR
(CDCl3) d: 1.21 (t, 3H, OCH2CH3), 2.62 (s, 3H, 5-
CH3), 2.81 (s, 3H, 7-CH3), 4.01 (s, 2H, CH2), 4.18 (q,
2H, OCH2CH3), 6.61 (s, 1H, H-6), 7.39–7.57 (m, 3H,
Ph), 7.78–7.84 (m, 2H, Ph). Anal. Calcd for
(CDCl3) d: 1.30 (t, 3H, OCH2CH3), 2.97 (dd, 1H, CH2),
3.30 (dd, 1H, CH2), 4.19 (q, 2H, OCH2CH3), 4.79–4.83
(m, 1H, CH), 7.43–7.77 (m, 3H, Ph), 8.01–8.19 (m, 2H,
Ph). Anal. Calcd for C13H13NO3: C, 67.52; H, 5.67; N,
6.06. Found: C, 67.47; H, 5.64; N, 6.38.

4830
S. Selleri et al. / Bioorg. Med. Chem. 13 (2005) 4821–4834
C18H19N3O2: C, 69.88; H, 6.19; N, 13.58. Found: C,
70.03; H, 6.32; N, 13.65.
5.14. N,N-Di-n-butyl-(5,7-dimethyl-2-phenylpyrazolo-
[1,5-a]pyrimidin-3-yl)acetamide (4d)
5.10. General procedure for the synthesis of amides (4a–q)
This compound was obtained using N,N-di-n-butyl-
amine and purified by column chromatography
(CHCl3/MeOH, 10:1 v/v, as eluent). White crystals; yield
17%; mp 95–96 ꢁC; H NMR (CDCl3) d (ppm): 0.82–
0.98 (m, 6H, N[(CH2)3CH3]2), 1.21–1.38 (m, 4H,
N[(CH2)3CH3]2), 1.47–1.63 (m, 4H, N[(CH2)3CH3]2),
2.59 (s, 3H, 5-CH3), 2.79 (s, 3H, 7-CH3), 3.29–3.44 (m,
4H, N[(CH2)3CH3]2), 3.97 (s, 2H, CH2), 6.58 (s, 1H, 6-
H), 7.39–7.44 (m, 3H, Ph), 7.79–7.86 (m, 2H, Ph). Anal.
Calcd for C24H32N4O: C, 73.43; H, 8.22; N, 14.27.
Found: C, 73.13; H, 8.02; N, 14.01.
The 5,7-dimethyl-2-phenylpyrazolo[1,5-a]pyrimidin-3-yl
acetic acid (3a) (1 mmol) and triethylamine (3.5 mmol)
were dissolved in 5 mL of dry THF and cooled to
ꢀ10 ꢁC. After stirring for 30 min, to this solution was
added ethyl chloroformate (1.1 mmol) and then the
reaction suspension was stirred for 1 h at ꢀ10 ꢁC before
the addition of 1.1 mmol of the suitable amine. The sus-
pension was allowed to warm to room temperature un-
der magnetic stirring for 5 h. The reaction was quenched
by introducing 5 mL of H2O, and the resulting mixture
was extracted with ether (2 · 10 mL); the ethereal ex-
tracts were washed successively with aqueous 5% HCl
(5 mL) and dried over Na2SO4. The crude acetamide
(4a–q) was obtained by removing the solvent under re-
duced pressure and the residue was purified by column
chromatography.
1
5.15. N,N-Di-n-pentyl-(5,7-dimethyl-2-phenylpyrazolo-
[1,5-a]pyrimidin-3-yl)acetamide (4e)
This compound was obtained using N,N-di-n-pentyl-
amine and purified by column chromatography
(CHCl3/MeOH, 20:1 v/v, as eluent). White crystals;
1
yield 34%; mp 92–93 ꢁC; H NMR (CDCl3) d (ppm):
5.11. N,N-Dimethyl-(5,7-dimethyl-2-phenylpyrazolo-
[1,5-a]pyrimidin-3-yl)acetamide (4a)
0.80–0.98 (m, 6H, N[(CH2)4CH3]2), 1.17–1.38 (m, 6H,
N[(CH2)4CH3]2), 1.53–1.64 (m, 6H, N[(CH2)4CH3]2),
2.58 (s, 3H, 5-CH3), 2.78 (s, 3H, 7-CH3), 3.23–3.42 (m,
4H, N[(CH2)4CH3]2), 3.94 (s, 2H, CH2), 6.58 (s, 1H, 6-
H), 7.40–7.52 (m, 3H, Ph), 7.78–7.87 (m, 2H, Ph). Anal.
Calcd for C26H36N4O: C, 74.25; H, 8.63; N, 13.32.
Found: C, 74.15; H, 8.52; N, 13.11.
This compound was obtained using N,N-dimethylamine
and purified by column chromatography (CHCl3/
MeOH, 10:1 v/v, as eluent). Light pink crystals; yield
1
46%; mp 162–163ꢁC; H NMR (CDCl3) d (ppm): 2.58
(s, 3H, 5-CH3), 2.78 (s, 3H, 7-CH3), 3.00 (s, 3H,
N(CH3)2), 3.18 (s, 3H, N(CH3)2), 3.95 (s, 2H, CH2),
6.56 (s, 1H, H-6), 7.40–7.53 (m, 3H, Ph), 7.80–7.85 (m,
2H, Ph). Anal. Calcd for C19H22N4O: C, 70.11; H,
6.54; N, 18.17. Found: C, 69.91; H, 6.29; N, 18.42.
5.16. N,N-Di-n-hexyl-(5,7-dimethyl-2-phenylpyrazolo-
[1,5-a]pyrimidin-3-yl)acetamide (4f)
This compound was obtained using N,N-di-n-hexylamine
and purified by column chromatography (CHCl3/MeOH,
1
20:1 v/v, as eluent). Uncoloured liquid; yield 34%. H
5.12. N,N-Di-n-propyl-(5,7-dimethyl-2-phenylpyrazolo-
[1,5-a]pyrimidin-3-yl)acetamide (4b)
NMR (CDCl3)
d
(ppm): 0.82–0.97 (m, 6H,
N[(CH2)5CH3]2), 1.20–1.28 (m, 10H, N[(CH2)5CH3]2),
2.58 (s, 3H, 5-CH3), 2.78 (s, 3H, 7-CH3), 3.26–3.42 (m,
4H, N[(CH2)5CH3]2), 3.92 (s, 2H, CH2), 6.58 (s, 1H, 6-
H), 7.38–7.48 (m, 3H, Ph), 7.79–7.89 (m, 2H, Ph). Anal.
Calcd for C28H40N4O: C, 74.96; H,8.99; N, 12.49. Found:
C, 74.75; H, 8.62; N, 12.11.
This compound was obtained using N,N-di-n-propyl-
amine and purified by column chromatography
(CHCl3/MeOH, 10:1 v/v, as eluent). White crystals; yield
54%; mp 129–130 ꢁC; 1H NMR (CDCl3) d (ppm): 0.82–
0.98 (m, 6H, N(CH2CH2CH3)2), 1.43–1.78 (m, 4H,
N(CH2CH2CH3)2), 2.58 (s, 3H, 5-CH3), 2.78 (s, 3H,
7-CH3), 3.28–3.47 (m, 4H, N(CH2CH2CH3)2), 3.98 (s,
2H, CH2), 6.57 (s, 1H, H-6), 7.39–7.51 (m, 3H, Ph),
7.78–7.82 (m, 2H, Ph). Anal. Calcd for C22H28N4O: C,
72.50; H, 7.74; N, 15.37. Found: C, 72.20; H, 7.50; N,
15.43.
5.17. N,N-Di-n-octyl-(5,7-dimethyl-2-phenylpyrazolo-
[1,5-a]pyrimidin-3-yl)acetamide (4g)
This compound was obtained using N,N-di-n-octylamine
and purified by column chromatography (CHCl3/
MeOH, 10:0.3 v/v, as eluent). White waxed solid; yield
41%. 1H NMR (CDCl3) d (ppm): 0.85–0.90 (m, 6H,
N[(CH2)7CH3]2), 1.14–1.38 (m, 20H, N[(CH2)5CH3]2),
1.46–1.61 (m, 4H, N[(CH2)5CH3]2), 2.56 (s, 3H, 5-CH3),
2.75 (s, 3H, 7-CH3), 3.27–3.40 (m, 4H, N[(CH2)5CH3]2),
3.94 (s, 2H, CH2), 6.52 (s, 1H, 6-H), 7.37–7.82 (m, 5H,
Ph). Anal. Calcd for C32H48N4O: C, 76.15; H, 9.59; N,
11.10. Found: C, 76.45; H, 9.62; N, 11.37.
5.13. N,N-Di-i-propyl-(5,7-dimethyl-2-phenylpyrazolo-
[1,5-a]pyrimidin-3-yl)acetamide (4c)
This compound was obtained using N,N-di-i-propyl-
amine and purified by column chromatography
(CHCl3/MeOH, 10:1 v/v, as eluent). White crystals; yield
22%; mp 177–178 ꢁC; 1H NMR (CDCl3) d (ppm): 1.17–
1.36 (m, 12H, N(CHCH3CH3)2), 2.62 (s, 3H, 5-CH3),
2.79
(s,
3H,
7-CH3),
3.01–3.20
(m,
2H,
5.18. N,N-Dibenzyl-(5,7-dimethyl-2-phenylpyrazolo-
[1,5-a]pyrimidin-3-yl)acetamide (4h)
N(CHCH3CH3)2), 4.01 (s, 2H, CH2), 6.58 (s, 1H, H-
6), 7.42–7.51 (m, 3H, Ph), 7.80–7.84 (m, 2H, Ph). Anal.
Calcd for C22H28N4O: C, 72.50; H, 7.74; N, 15.37.
Found: C, 72.71; H, 7.83; N, 15.15.
This compound was obtained using N,N-dibenzyl and
purified by column chromatography (EtOAc as eluent).

S. Selleri et al. / Bioorg. Med. Chem. 13 (2005) 4821–4834
4831
White waxed solid; yield 48%. 1H NMR (CDCl3) d
(ppm): 2.55 (s, 3H, 5-CH3), 2.72 (s, 3H, 7-CH3), 4.07
(s, 2H, CH2), 4.61 (s, 2H, CH2Ph), 4.64 (s, 2H, CH2Ph),
6.51 (s, 1H, 6-H), 7.12–7.45 (m, 13H, Ph), 7.78–7.85 (m,
2H, Ph). Anal. Calcd for C30H28N4O: C, 78.23; H, 6.13;
N, 12.16. Found: C, 78.36; H, 6.25; N, 11.98.
C25H26N4O: C, 75.35; H, 6.58; N, 14.06. Found: C,
75.50; H, 6.88; N, 13.95.
5.23. (S)-(ꢀ)-N,a-Dimethylbenzyl-(5,7-dimethyl-2-phenyl-
pyrazolo[1,5-a]pyrimidin-3-yl)acetamide (4m)
This compound was obtained using (S)-(-)-N,a-Dimeth-
ylbenzylamine and purified by column chromatography
(CHCl3/MeOH, 20:1 v/v, as eluent). White crystals; yield
51%; mp 104—106 ꢁC; 1H NMR (CDCl3) d (ppm): 1.53
(d, 3H, CH3), 2.70–2.90 (m, 9H: 5-CH3, 7-CH3, N-CH3),
4.01–4.22 (m, 2H, CH2), 6.10 (q, 1H, CH), 6.63 (s, 1H,
6-H), 7.22–7.38 (m, 5H, Ph), 7.41–7.58 (m, 3H, Ph),
7.78–7.91 (m, 2H, Ph). Anal. Calcd for C25H26N4O: C,
75.35; H, 6.58; N, 14.06. Found: C, 75.67; H, 6.43; N,
14.26.
5.19. N-Ethyl-N-i-propyl-(5,7-dimethyl-2-phenylpyrazolo-
[1,5-a]pyrimidin-3-yl)acetamide (4i)
This compound was obtained using N-ethyl-N-i-propyl-
amine and purified by column chromatography (CHCl3/
MeOH, 20:1 v/v, as eluent). Light yellow crystals; yield
35%; mp 130–131 ꢁC; 1H NMR (CDCl3) d (ppm):
1.10–1.36 (m, 9H: 3H, N(CH2CH3); 6H, N(CH(CH3)2),
2.58 (s, 3H, 5-CH3), 2.78 (s, 3H, 7-CH3), 3.20–3.56 (m,
2H, N(CH2CH3)), 3.90–4.00 (m, 2H, CH2), 4.36–4.80
(m, 1H, N(CH(CH3)2), 6.57 (s, 1H, 6-H), 7.38–7.50
(m, 3H, Ph), 7.78–7.85 (m, 2H, Ph). Anal. Calcd for
C21H26N4O: C, 71.97; H, 7.48; N, 15.99. Found: C,
72.06; H, 7.27; N, 15.90.
5.24. 5,7-Dimethyl-2-phenyl-3-[(pyrrolidinocarbonyl)-
methyl]pyrazolo[1,5-a]pyrimidine (4n)
This compound was obtained using pyrrolidine and
purified by column chromatography (CHCl3/MeOH,
20:1 v/v, as eluent). White crystals; yield 85%; mp 191–
5.20. N-Ethyl-N-phenyl-(5,7-dimethyl-2-phenylpyrazolo-
[1,5-a]pyrimidin-3-yl)acetamide (4j)
1
192 ꢁC; H NMR (CDCl3) d (ppm): 1.81–2.03 (m, 4H,
N(CH2)4), 2.58 (s, 3H, 5-CH3), 2.78 (s, 3H, 7-CH3),
3.43–3.70 (m, 4H, N(CH2)4), 43.86 (s, 2H, CH2), 6.56
(s, 1H, 6-H), 7.39–7.50 (m, 3H, Ph), 7.83–7.91 (m, 2H,
Ph). Anal. Calcd for C20H22N4O: C, 71.83; H, 6.63; N,
16.75. Found: C, 71.98; H, 6.57; N, 16.60.
This compound was obtained using N-ethyl-N-phenyla-
mine and purified by column chromatography (CHCl3/
MeOH, 20:1 v/v, as eluent). White crystals; yield 24%;
mp 153–154 ꢁC; 1H NMR (CDCl3) d (ppm): 1.10 (t,
3H, N(CH2CH3)), 2.58 (s, 3H, 5-CH3), 2.68 (s, 3H, 7-
CH3), 3.65–3.81 (m, 4H: 2H, CH2; 2H, N(CH2CH3)),
6.56 (s, 1H, ,6-H), 7.30–7.56 (m, 8H: 3H, Ph; 5H,
N-Ph), 7.70–7.80 (m, 2H, Ph). Anal. Calcd for
C24H24N4O: C, 74.97; H, 6.29; N, 14.57. Found: C,
75.16; H, 6.27; N, 14.87.
5.25. 5,7-Dimethyl-2-phenyl-3-[(piperidinocarbonyl)meth-
yl]pyrazolo[1,5-a]pyrimidine (4o)
This compound was obtained using piperidine and
purified by column chromatography (CHCl3/MeOH,
20:1 v/v, as eluent). Ivory crystals; yield 33%; mp
177–178 ꢁC; 1H NMR (CDCl3) d (ppm): 1.44–1.72
(m, 6H, N(CH2)5), 2.59 (s, 3H, 5-CH3), 2.79 (s, 3H,
7-CH3), 3.50–3.63 (m, 4H, N(CH2)5), 3.98 (s, 2H,
CH2), 6.58 (s, 1H, 6-H), 7.37–7.53 (m, 3H, Ph), 7.79–
7.90 (m, 2H, Ph). Anal. Calcd for C21H24N4O: C,
72.39; H, 6.94; N, 16.08. Found: C, 72.67; H, 7.10;
N, 15.97.
5.21. N-Benzyl-N-ethyl-(5,7-dimethyl-2-phenylpyrazolo-
[1,5-a]pyrimidin-3-yl)acetamide (4k)
This compound was obtained using N-benzyl-N-ethyl-
amine and purified by column chromatography (CHCl3/
MeOH, 20:1 v/v, as eluent). White crystals; yield 20%;
1
mp 141–142 ꢁC; H NMR (CDCl3) d (ppm): 1.10–1.30
(m, 3H, N(CH2CH3)), 2.84–2.89 (m, 6H: 3H, 5-CH3;
3H, 7-CH3), 3.40–3.60 (m, 2H, N(CH2CH3)), 4.31 (d,
2H, CH2), 4.73 (d, 2H, NCH2Ph), 6.60–6.75 (m, 1H,
6-H), 7.20–7.27 (m; 5H, N CH2Ph), 7.40–7.58 (m, 3H,
Ph), 7.70–7.83 (m, 2H, Ph). Anal. Calcd for
C25H26N4O: C, 75.35; H, 6.58; N, 14.06. Found: C,
75.24; H, 6.98; N, 13.89.
5.26. 5,7-Dimethyl-3-[(4-methylpiperazinocarbonyl)meth-
yl]-2-phenyl-pyrazolo[1,5-a]pyrimidine (4p)
This compound was obtained using N-methylpiperazine
and purified by column chromatography (CHCl3/
MeOH, 10:1 v/v, as eluent). White crystals; yield 13%;
mp 156–157 ꢁC; 1H NMR (CDCl3) d (ppm): 2.36 (s,
3H, NCH3), 2.38–2.44 (m, 4H, NN(CH2)4), 2.58 (s,
3H, 5-CH3), 2.78 (s, 3H, 7-CH3), 3.07 (m, 4H,
NN(CH2)4), 3.96 (s, 2H, CH2), 6.58 (s, 1H, H-6),
7.42–7.48 (m, 3H, Ph), 7.80–7.84 (m, 2H, Ph). Anal.
Calcd for C21H25N5O: C, 69.40; H, 6.93; N, 19.27.
Found: C, 69.15; H, 7.05; N, 19.59.
5.22. (R)-(+)-N,a-Dimethylbenzyl-(5,7-dimethyl-2-phenyl-
pyrazolo[1,5-a]pyrimidin-3-yl)acetamide (4l)
This compound was obtained using (R)-(+)-N,a-
Dimethylbenzylamine and purified by column chro-
matography (CHCl3/MeOH, 20:1 v/v, as eluent). White
1
crystals; yield 56%; mp 104–106 ꢁC; H NMR (CDCl3)
5.27. N-Propyl-(5,7-dimethyl-2-phenylpyrazolo[1,5-a]-
pyrimidin-3-yl)acetamide (4q)
d (ppm): 1.50 (d, 3H, CH3), 2.6–2.84 (m, 9H: 5-CH3,
7-CH3, N-CH3), 4.01–4.22 (m, 2H, CH2), 6.10 (q, 1H,
CH), 6.62 (s, 1H, 6-H), 7.22–7.58 (m, 8H: 5H, Ph; 3H,
Ph), 7.78–7.91 (m, 2H, Ph). Anal. Calcd for
This compound was obtained using propylamine and
purified by column chromatography (CHCl3/MeOH,

4832
S. Selleri et al. / Bioorg. Med. Chem. 13 (2005) 4821–4834
10:1 v/v, as eluent). Ivory crystals; yield 57%; mp 188–
189 ꢁC; 1H NMR (CDCl3) d (ppm): 0.82 (t, 3H,
N-CH2CH2CH3), 1.26–1.34 (m, 2H, N-CH2CH2CH3),
2.60 (s, 3H, 5-CH3), 2.81 (s, 3H, 7-CH3), 3.16–3.24 (m,
2H, N-CH2CH2CH3), 3.83 (s, 2H, CH2), 6.61 (s, 1H,
H-6), 6.81 (br s, 1H, NH exchangeable), 7.42–7.58 (m,
3H, Ph), 7.86–8.01 (m, 2H, Ph). Anal. Calcd for
C19H22N4O: C, 70.78; H, 6.88; N, 17.38. Found: C,
70.63; H, 6.59; N, 17.64.
COOH). Anal. Calcd for C17H21N3O2: C, 68.20; H,
7.07; N, 14.04. Found: C, 68.10; H, 7.18; N, 13.92.
5.31. N,N-Diethyl-5,7-dimethyl-2-phenylpyrazolo[1,5-a]-
pyrimidin-3-carboxamide (8)
Compound 7 (1 mmol) and triethylamine (3.5 mmol)
were dissolved in 5 mL of dry THF and cooled to
ꢀ10 ꢁC. After stirring for 30 min, to this solution was
added ethyl chloroformate (1.1 mmol) and then the
reaction suspension was stirred for 1 h at ꢀ10 ꢁC before
the addition of 1.1 mmol of the suitable amine. The sus-
pension was allowed to warm to room temperature un-
der magnetic stirring for 5 h. The reaction was quenched
by introducing 5 mL of H2O, and the resulting mixture
was extracted with ether (2 · 10 mL); the ethereal ex-
tracts were washed successively with aqueous 5% HCl
(5 mL) and dried over Na2SO4. The crude amide 8 was
obtained by removing the solvent under reduced pres-
sure and purified by column chromatography (CHCl3/
MeOH, 20:1 v/v, as eluent). Yellow crystals; yield
5.28. N,N-Diethyl-(5,7-dimethyl-2-phenylpyrazolo[1,5-a]-
pyrimidin-3-yl)tioacetamide (5)
To a solution of N,N-diethyl-(2-phenyl-5,7-dimethyl-
pyrazolo[1,5-a]pyrimidin-3-yl)acetamide (3i)18 (1 mmol)
in toluene (25 mL), Lawessonꢀs reagent (2 mmol) was
added and the reaction mixture was refluxed under mag-
netic stirring for 2 h. The resulting solution was treated
with 5% NaHCO3 and the organic phase was washed
with water and then evaporated under reduced pressure;
the residue was purified by silica gel column chromato-
graphy (CHCl3/MeOH, 20:1 v/v, as eluent). White crys-
1
35%; mp 180–181 ꢁC; H NMR (CDCl3) d (ppm): 1.02
1
(t, 3H, N(CH2CH3)2), 1.37 (t, 3H N(CH2CH3)2), 2.60
(s, 3H, 5-CH3), 2.81 (s, 3H, 7-CH3), 3.26 (d, 2H,
N(CH2CH3)2) 3.76 (d, 2H, N(CH2CH3)2), 6.68 (s, 1H,
6-H), 8.10–8.20 (m, 3H, Ph), 8.23–8.35 (m, 2H, Ph).
Anal. Calcd for C21H30N4O: C, 71.15; H, 8.53; N,
15.80. Found: C, 71.10; H, 8.19; N, 15.72.
tals; yield 30%; mp 136–137 ꢁC; H NMR (CDCl3) d
(ppm): 1.26–1.32 (m, 6H, N(CH2CH3)2), 2.55 (s, 3H,
5-CH3), 2.77 (s, 3H, 7-CH3), 3.82 (q, 2H, N(CH2CH3)2),
4.07 (q, 2H, N(CH2CH3)2), 4.28 (s, 2H, CH2), 6.54 (s,
1H, H-6), 7.42–7.46 (m, 3H, Ph), 7.85–7.88 (m, 2H,
Ph). Anal. Calcd for C20H24N4S: C, 68.15; H, 6.86; N,
15.89. Found: C, 68.01; H, 7.09; N, 15.72.
5.32. 3D QSAR analysis
5.29. 5,7-Dimethyl 2-phenylpyrazolo[1,5-a]pyrimidin-3-
carboxamide (6)
Calculations were performed on a AMD Athlon XP
1800+ personal computer. All molecular modelling
and QSAR studies were performed on a SGI O2
R10000 workstation. The structures of all studied com-
pounds were generated using fragment libraries and/or
the Builder module of the InsightII2000 package.40
The energies of the molecules were minimized with con-
jugate gradient procedure using cvff forcefield (FF) and
Discover module of InsightII 2000. The biological activ-
ity of the compounds was homogeneously expressed as
binding affinity (Ki) and transformed to the (ꢀlog Ki)
values for using in the QSAR analysis. The conforma-
tional search was performed using a simulated annealing
(SA) procedure which was started using cvff FF, distant
The 5,7-dimethyl-2-phenylpyrazolo[1,5-a]pyrimidin-3-
carbonitrile25 (2.62 mmol) was dissolved in H2SO4
(5 mL) and the resulting solution was maintained at
room temperature for 1 h, under magnetic stirring. To
the reaction mixture was added ice water and compound
6 was isolated by filtration and copiously washed with
water. The crude amide was purified by column chro-
matography (CHCl3/MeOH, 10:1 v/v, as eluent). White
crystals; yield 89%; mp 252–253 ꢁC; 1H NMR (CDCl3) d
(ppm): 2.64 (s, 3H, 5-CH3), 2.82 (s, 3H, 7-CH3), 5.57
(s, 1H, NH2), 6.77 (s, 1H, 6-H), 7.42–7.57 (m, 3H,
Ph), 7.92–8.17 (m, 2H, Ph), 8.57 (s, 1H, NH2). Anal.
Calcd for C17H22N4O: C, 68.43; H, 7.43; N, 18.78.
Found: C, 68.10; H, 7.19; N, 18.72.
dependent dielectric constant and a convergence crite-
ꢀ1. The molecular dynamic
˚
A
rion of 0.001 kcalmolꢀ1
(MD) simulation took 1000 fs to reach 900 K. The sys-
tem was then kept at 900 K for 2000 fs and subsequently
cooled down at 400 K, at regular intervals, by decreas-
ing the simulation temperature. Molecules were then al-
lowed to minimize in order to reach the final minimum
conformation. The described cycle was repeated 100
times so that 100 energy-minimized conformers for each
PBR inhibitor were obtained. A cluster analysis was per-
formed using an ꢁin houseꢀ developed routine written in
Visual Basic programming language version 6.0 (Micro-
soft Corporation) that groups molecules on the basis of
the root mean square (rms) value of the distance
between atoms. The alignment of PBR inhibitors was
accomplished by superposing the common bicyclic core.
Single conformers of each molecule were obtained so
that all of them present the same orientation of the ace-
5.30. 5,7-Dimethyl-2-phenylpyrazolo[1,5-a]pyrimidin-3-
carboxylic acid (7)
A saturated solution of NaNO2 was added slowly to a
solution of compound 6 (1.3 mmol) in H2SO4 (5 mL)
kept at ꢀ5 ꢁC. The resulting suspension was stirred for
30 min and then extracted with ether (2· 10 mL); the
ethereal extracts were dried over Na2SO4. The crude
acid 7 was obtained by removing the solvent under re-
duced pressure and purified by column chromatography
(CHCl3/MeOH, 20:3 v/v, as eluent). Ivory crystals; yield
27%; mp: 188 ꢁC; 1H NMR (DMSO-d6) d (ppm): 2.59 (s,
3H, 5-CH3), 2.71 (s, 3H, 7-CH3), 7.19 (s, 1H, 6-H), 7.60–
7.82 (m, 3H, Ph), 7.99–8.38 (m, 2H, Ph), 12.29 (s, 1H,

S. Selleri et al. / Bioorg. Med. Chem. 13 (2005) 4821–4834
4833
tamidic group and most of them correspond to lower en-
5.34. [3H]Ro15-1788 binding to rat cerebral cortex
membranes
ergy state. Molecules of the training set were selected
according to statistical molecular design principles car-
ried out in the space of the principal component scores
of the Volsurf41 descriptors so that the resulting 27 se-
lected compounds provided actually an adequate cover-
age of the whole structural space. Molecular interaction
fields (MIFs) were computed with GRID programme36
using C3, DRY and N1 probes and the following
Rat cerebral cortex membranes were prepared as previ-
ously described.44 After differential centrifugation, the
obtained crude membrane fraction was subjected to
washing procedures to remove endogenous GABA.45
The washed membranes were incubated with 0.2 nM
[3H]Ro 15-1788 for 90 min at 0 ꢁC in 500 lL of buffer,
50 mM Tris–citrate buffer, pH 7.4, as previously
described.46
˚
parameters: 1 A grid spacing, 26 · 30 · 20 grid box
dimension. PCA (Principal Component Analysis) and
PLS (Partial Least Squares) algorithm were used as
implemented in Golpe programme23 on the matrix de-
rived from the unfolding of the probe-ligand interac-
tions into vectors (one for each ligand considered).
This matrix was submitted to Golpe advanced pretreat-
ment including max cutoff = 0 kcal, zeroing = 0.01 kcal,
5.35. Cell culture
Rat glioma C6 cells were cultured in Dulbeccoꢀs modi-
fied Eagleꢀs medium supplemented with 10% FBS,
2
mM L-glutamine, 100 units/mL penicillin and
˚
Min sd = 0.02 A, N-level = 2, 3 and 4. The resulting
10,016 variables were normalized according to the
BUW procedure. The smart region definition (number
100 lg/mL streptomycin. Cultures were maintained in
a humidified atmosphere of 5% CO2/95% air at 37ꢁC.
˚
of seed = 1757, critical distance cutoff = 1.0 A, collaps-
˚
ing cutoff = 2.0 A) in combination with the fractional
5.36. Steroid biosynthesis
factorial design was used in the variable selection.
PCA based on C3 and N1 probes was carried out: no
outliers were found. The optimal dimensionality of the
PLS model was determined according to leave-one-out
(LOO) and random groups cross-validation procedure.
C6 cells were seeded in 24-well plates at a density of
ꢁ1 · 106 cells/well in a final volume of 1 mL. Prior to
measurement of pregnenolone production, the cells were
washed three times with a simple salts medium consist-
ing of 140 mM NaCl, 5 mM KCl, 1.8 mM CaCl2,
1 mM MgSO4, 10 mM glucose, 10 mM HEPES/NaOH,
pH 7.4, plus 0.1% BSA. During experiments, cells were
incubated with this simple salts medium in air incubator
at 37 ꢁC. In order to measure pregnenolone secreted into
the medium, its further metabolism was blocked by the
addition of trilostane (25 lM) and SU 10603 (10 lM)
(inhibitors of 3b-hydroxysteroid dehydrogenase and
17a-hydroxylase, respectively) to the simple salts med-
ium, as previously described.19 The addition of the novel
compounds and of PK 11195, Ro5-4864, or clonazepam
to the C6 cells was made by the complete change of the
simple salts medium to medium containing the appro-
priate concentration (40 lM) of compound. The final
concentration of ethanol was constant for all the wells
within each experiment and did not exceed 0.5% (v/v),
a concentration which on its own had no effect on ste-
roid production. At the end of the incubation period
(2 h), the cell medium was retained and centrifuged at
1500 g for 10 min. The amount of pregnenolone secreted
into the medium was quantified by radio immunoassay
(RIA), using an antibody obtained from ICN Biochem-
ical Inc., CA, U.S.A., under the conditions recom-
mended by the supplier. Cell protein concentration
was measured according to the method of Lowry et al.43
5.33. Binding to rat kidney mitochondrial membranes
For binding studies, mitochondria were prepared as pre-
viously described31,19 with minor modification as de-
scribed below, from kidneys of Male Wistar rats killed
by cervical dislocation. Kidneys were homogenized
in 20 volumes of ice-cold 50 mM Tris/HCl, pH 7.4,
0.32 M sucrose and 1 mM EDTA (buffer A), containing
protease inhibitors (160 lg/mL benzamidine, 200 lg/mL
bacitracine and 20 lg/mL soybean trypsin inhibitor)
with a Teflon pestle in a glass homogenizer and centri-
fuged at 600g for 10 min at 4 ꢁC. The resulting superna-
tant was centrifuged at 10,000g for 10 min at 4 ꢁC. The
pellet was then resuspended in 20 volumes of ice-cold
buffer A, and centrifuged again at 10,000g for 10 min
at 4 ꢁC. The crude mitochondrial pellet was frozen at
ꢀ20 ꢁC until the time of assay or incubated with either
0.6 nM [3H]PK11195 or 1 nM [3H]Ro 5-4864 in
50 mM Tris/HCl, pH 7.4 (buffer B), with a range of con-
centrations of the tested compounds (0.1 nM to 10 lM)
in a total volume of 0.5 mL for 90 min at 4 ꢁC. The incu-
bation was terminated by dilution to 5 mL with ice-cold
buffer B, followed immediately by rapid filtration
through glass fiber Whatman GF/C filters. The filters
were then washed (2· 5 mL) with buffer B and the
amount of radioactivity retained on the filters was deter-
mined by Packard 1600 TR liquid scintillation counter
at 66% efficiency. Non-specific binding was estimated
in each case in the presence, respectively, of unlabeled
1 lM PK 11195 or Ro5-4864. For the active com-
pounds, the IC50 values were determined and Ki values
were derived according to the equation of Cheng and
Prusoff.42 Protein concentration was estimated by the
method of Lowry et al.43 with bovine serum as
standard.
Acknowledgment
This research was supported by funds of the Italian
Ministry of University and Scientific Research
(MURST).
References and notes
1. McEnery, M. W.; Snowman, A. M.; Trifiletti, R. R.;
Snyder, S. H. Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 3170.

4834
S. Selleri et al. / Bioorg. Med. Chem. 13 (2005) 4821–4834
2. Casellas, P.; Galiegue, S.; Basile, A. S. Neurochem. Int.
2002, 40, 475.
3. Krueger, K. E. Biochim. Biophys. Acta 1995, 1241, 453.
4. Gavish, M.; Bachman, I.; Shoukrun, R.; Katz, Y.;
Veenman, L.; Weisinger, G.; Weizman, A. Pharmacol.
Rev. 1999, 51, 629.
5. OꢀBeirne, G. B.; Williams, D. C. Eur. J. Biochem. 1988,
175, 413.
6. Galiegue, S.; Tinel, N.; Casellas, P. Curr. Med. Chem.
2003, 10, 1563.
25. Mansour, T. S. Tetrahedron Lett. 1988, 29, 3437.
26. Ansell, M. F. In Preparation of Acyl Halides; Interscience:
London, Engl., 1972, pp 35–68.
27. Jang, D. O.; Park, D. J.; Kim, J. Tetrahedron Lett. 1999,
40, 5323.
28. Frøyen, P. Synth. Commun. 1995, 25, 959.
29. Ghozlan Said, A. S. Pharmazie 1991, 46, 352.
30. Romeo, E.; Auta, J.; Kozikowski, A. P.; Ma, D.;
Papadopoulos, V.; Puia, G.; Costa, E.; Guidotti, A.
J. Pharmacol. Exp. Ther. 1992, 262, 971.
7. Veenman, L.; Leschiner, S.; Spanier, I.; Weisinger, G.;
Weizman, A.; Gavish, M. J. Neurochem. 2002, 80, 917.
8. Avital, A.; Richter-Levin, G.; Leschiner, S.; Spanier, I.;
Veenman, L.; Weizman, A.; Gavish, M. Neuropsycho-
pharmacology 2001, 25, 669.
9. Kelly-Hershkovitz, E.; Weizman, R.; Spanier, I.; Lesch-
iner, S.; Lahav, M.; Weisinger, G.; Gavish, M. J. Biol.
Chem. 1998, 273, 5478.
10. Lang, S. Curr. Med. Chem. 2002, 9, 1411.
11. Hauet, T.; Liu, J.; Li, H.; Gazouli, M.; Culty, M.;
Papadopoulos, V. Endocr. Res. 2002, 28, 395.
12. Lacapere, J. J.; Papadopoulos, V. Steroids 2003, 68, 569.
13. Liu, J.; Li, H.; Papadopoulos, V. J. Steroid. Biochem. Mol.
Biol 2003, 85, 275.
14. Gazouli, M.; Han, Z.; Papadopoulos, V. J. Pharmacol.
Exp. Ther. 2002, 303, 627.
15. Farges, R.; Joseph-Liauzun, E.; Shire, D.; Caput, D.; Le
Fur, G.; Ferrara, P. Mol. Pharmacol. 1994, 46, 1160.
16. Yeliseev, A. A.; Kaplan, S. J. Biol. Chem. 1999, 274,
21234.
17. Papadopoulos, V.; Amri, H.; Boujrad, N.; Cascio, C.;
Culty, M.; Garnier, M.; Hardwick, M.; Li, H.; Vidic, B.;
Brown, A. S.; Reversa, J. L.; Bernassau, J. M.; Drieu, K.
Steroids 1998, 62, 21.
18. Cinone, N.; Ho¨ltje, H.-D.; Carotti, A. J. Comput. Aided
Drug Des. 2000, 14, 753.
31. Trapani, G.; Franco, M.; Ricciardi, L.; Latrofa, A.;
Genchi, G.; Sanna, E.; Tuberi, F.; Cagetti, E.; Biggio, G.;
Liso, G. J. Med. Chem. 1997, 40, 3109.
32. Dubroeucq, M. C.; Benavides, J.; Doble, A.; Guillox, F.;
Allam, D.; Vaucher, N.; Bertran, P.; Gueremy, C.;
Renault, C.; Uzan, A. Eur. J. Pharmacol. 1986, 128,
269.
33. Newman, A. H.; Lueddens, H. W.; Skolnick, P.; Rice, K.
C. J. Med. Chem. 1987, 30, 1901.
34. Shah, F.; Hume, S. P.; Pike, V. W.; Ashworth, S.;
McDermott, J. Nucl. Med. Biol. 1994, 21, 573.
35. Selleri, S.; Bruni, F.; Costagli, C.; Costanzo, A.; Guerrini,
G.; Ciciani, G.; Gratteri, P.; Besnard, F.; Costa, B.;
Martini, C.; Fohlin, J.; De Siena, G.; Malmberg Aiello, P.;
J. Med. Chem., submitted for publication.
36. (a) Goodford, P. J. J. Med. Chem. 1985, 28, 849;
(b) GRIDv.20 Molecular Discovery Ltd (http://
www.moldiscovery.com).
37. Trapani, G.; Franco, M.; Latrofa, A.; Ricciardi, L.;
Carotti, A.; Serra, M.; Sanna, E.; Biggio, G.; Liso, G.
J. Med. Chem. 1999, 42, 3934.
38. Campiani, G.; Ramunno, A.; Fiorini, I.; Nacci, V.;
Morelli, E.; Novellino, E.; Goegan, M.; Mennini, T.;
Sullivan, S.; Zisterer, D. M.; Williams, C. D. J. Med.
Chem. 2002, 45, 4276.
39. Primofiore, G.; Da Settimo, F.; Taliani, S.; Simorini, F.;
Patrizi, M. P.; Novellino, E.; Greco, G.; Abignente, E.;
Costa, B.; Chelli, B.; Martini, C. J. Med. Chem. 2002, 45,
4276.
40. InsightII 2000, Accelrys Inc. (www.accelrys.com).
41. (a) Volsurf v.3.0.11. (http://www.moldiscovery.com); (b)
Cruciani, G.; Crivori, P.; Carrupt, P.-A.; Testa, B. J. Mol.
Struct. Theochem. 2000, 503, 17.
19. Campiani, G.; Nacci, V.; Fiorini, I.; De Filippis, M. P.;
Garofalo, A.; Ciani, S. M.; Greco, G.; Novellino, E.;
Williams, D. C.; Zisterer, D. M.; Woods, M. J.; Mihai, C.;
Manzoni, C.; Mennini, T. J. Med. Chem. 1996, 3435.
20. Cappelli, A.; Anzini, M.; Vomero, S.; De Benedetti, P. G.;
Menziani, M. C.; Giorgi, G.; Manzoni, C. J. Med. Chem.
1997, 40, 2910.
21. Anzini, M.; Cappelli, A.; Vomero, S.; Seeber, M.; Men-
ziani, M. C.; Langer, T.; Hagen, B.; Manzoni, C.;
Bourguignon, J.-J. J. Med. Chem. 2001, 44, 1134.
22. Selleri, S.; Bruni, F.; Costagli, C.; Costanzo, A.; Guerrini,
G.; Ciciani, G.; Costa, B.; Martini, C. Bioorg. Med. Chem.
2001, 9, 2661.
23. GOLPE 4.5. Multivariate Infometric Analysis Srl. (http://
www.miasrl.com).
24. Taylor, E. C.; Hawks, G. H.; McKillop, A. J. Am. Chem.
Soc. 1968, 90, 2421.
42. Cheng, Y.; Prusoff, W. H. Biochem. Pharmacol. 1973, 22,
3099.
43. Lowry, O. H.; Rosenbrough, N. J.; Farr, A. L.; Randall,
R. J. J. Biol. Chem. 1951, 193, 265.
44. Squires, R. F.; Braestrup, C. Nature 1997, 266, 732.
45. Karobath, M.; Sperk, G. Proc. Natl. Acad. Sci. U.S.A.
1979, 76, 1004.
46. Costanzo, A.; Guerrini, G.; Ciciani, G.; Bruni, F.; Selleri,
S.; Costa, B.; Martini, C.; Lucacchini, A.; Aiello, P. M.;
Ipponi, A. J. Med. Chem. 1999, 42, 2218.