Diphenylcyclohexylamine derivatives as new potent multidrug resistance (MDR) modulators

Article abstract of DOI:10.1016/j.bmc.2004.11.043

A series of compounds with a diphenylmethyl cyclohexyl skeleton, loosely related to verapamil, has been synthesized and tested as MDR modulators on anthracycline-resistant erythroleukemia K 562 cells. Their residual cardiovascular action (negative inotrop

Full text of DOI:10.1016/j.bmc.2004.11.043

Bioorganic & Medicinal Chemistry 13 (2005) 985–998
Diphenylcyclohexylamine derivatives as new potent
multidrug resistance (MDR) modulators
Silvia Dei,a,* Roberta Budriesi,b Paiwan Sudwan,c,ꢀ Marta Ferraroni,d
Alberto Chiarini,b Arlette Garnier-Suillerot,c Dina Manetti,a
Cecilia Martelli,a Serena Scapecchia and Elisabetta Teodoria
a
`
Dipartimento di Scienze Farmaceutiche, Universita di Firenze, via U. Schiff 6, 50019 Sesto Fiorentino (FI), Italy
b
`
Dipartimento di Scienze Farmaceutiche, Universita di Bologna, via Belmeloro 6, 40126 Bologna, Italy
c
`
´
Laboratoire de Physicochimie Biomoleculaire et Cellulaire (UMR 7033), Universite Paris 13,
74 rue Marcel Cachin, 93017 Bobigny, France
`
Dipartimento di Chimica, Universita di Firenze, via della Lastruccia 5, 50019 Sesto Fiorentino (FI), Italy
d
Received 16 July 2004; accepted 23 November 2004
Available online 19 December 2004
Abstract—A series of compounds with a diphenylmethyl cyclohexyl skeleton, loosely related to verapamil, has been synthesized and
tested as MDR modulators on anthracycline-resistant erythroleukemia K 562 cells. Their residual cardiovascular action (negative
inotropic and chronotropic activity as well as vasorelaxant activity) was evaluated on guinea-pig isolated atria preparations and on
guinea-pig aortic strip preparations. Most compounds of the series possess a good MDR-reverting activity together with a low car-
diovascular action. Among them, compounds 3a1, 7a, and 8a are more potent than verapamil as MDR reverters and lack any car-
diovascular action; they can represent useful leads for the development of new safe MDR reversing drugs.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
ety of proteins that act as ATP-dependent extrusion
pumps. At the moment, the best-known extrusion pro-
teins are P-glycoprotein (P-gp) and MRP1;5 both belong
to the ABC superfamily of transporters. It is important
to notice that, in addition to their role in cancer cell
resistance, these proteins seem to have multiple physio-
logical functions as well6,7 since they are expressed also
in many important non-tumoural tissues, such as the
blood–brain barrier (BBB), intestinal epithelium and
hepatic cells, and similar transporting proteins of the
ABC superfamily are largely present in procariotic
organisms.8
Drug resistance is a phenomenon that frequently im-
pairs proper treatment of infections and cancer with
chemotherapeutics. Multidrug resistance (MDR) is a
kind of acquired drug resistance of cancer cells to multi-
ple classes of chemotherapic drugs that can be structur-
ally and mechanistically unrelated.1,2 MDR can be the
result of a variety of mechanisms that are not fully
understood, but the most widely implicated mechanism
is concerned with altered membrane transport in tumour
cells.3 This mechanism is often referred to as classical
MDR4 and is related to a lower cell concentration of
cytotoxic drugs associated with accelerated efflux of
antitumour agents, due to the over expression of a vari-
Due to the unprecedented variety of substrates extruded,
the mechanism of action of these transporters is still
controversial. As a matter of fact, a number of models
have been proposed to explain the involvement of these
extrusion pumps in MDR:9 the dominant ꢁdrug pumpꢂ
model has been questioned but, at present, it seems that
P-gp and related proteins are indeed directly involved in
the extrusion of multiple drugs.10–12
Keywords: Multidrug resistance (MDR); MDR modulators; Chemo-
sensitizer; Diphenylcyclohexylamine derivatives.
*
Corresponding author. Tel.: +39 055 4573689; fax: +39 055
4573671; e-mail: silvia.dei@unifi.it
ꢀ Present Address: Laboratory of Physical Chemistry, Molecular and
Cellular Biology, Faculty of Science, Burapha University, Bangsaen,
Chonburi 20131, Thailand.
Whatever the mechanism of action of P-gp, inhibition of
its functions has been rapidly recognized as a possible
0968-0896/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.11.043

986
S. Dei et al. / Bioorg. Med. Chem. 13 (2005)985–998
approach to circumvent MDR and drugs possessing
inhibitory properties have been and are actively being
sought.13,14 Although verapamil, the first drug able to
reverse MDR, was recognized more than 20 years ago
as a chemosensitizer, a potent, specific and safe drug
able to reverse multidrug resistance and to assist in the
chemotherapy of cancer is still lacking. In fact, most
of the molecules initially found active (the so called ꢁfirst
generationꢂ chemosensitizers) were known drugs with a
definite pharmacological action that induced unwanted
side effects. Therefore, the problem of the specific design
of chemosensitizers was approached, and a second gen-
eration of drugs able to revert MDR appeared on the
stage, but the fairly heterogeneous chemical structure
of the compounds found active in reverting MDR has
made difficult to establish structure–activity relation-
ships, and made the design of new candidates difficult.13
Indeed, the broad substrate selectivity of P-gp suggests
that there might be more than a single binding mode
within a large binding site15–17 and it is not surprising
that the efforts to establish SAR has led only to qualita-
tive, generic indications.18–20 Also the role of lipophili-
city, a characteristic that is generally considered
important for MDR-reverting properties, is quite elu-
sive, since till now no significant quantitative relation-
ships have been established between partition
coefficients and activity,21–23 and successful correlations
were found only for highly homogeneous sets of
molecules.24,25
As pointed out above, verapamil, a calcium channel
antagonist (Chart 1), was one of the most studied first
generation chemosensitizers, both in vitro and in vivo,
but a limiting factor of its therapeutic use are the pro-
nounced cardiovascular effects, which occur at the plas-
ma concentrations required to efficiently block P-gp
transport. From a clinical point of view, it is therefore
important to find analogues with low calcium channel
blocking activity and high MDR-reverting action. In-
deed, verapamil has been used as the lead molecule in
several attempts to identify more potent and selective
drugs. In a previous study,21 we identified some rigid
verapamil analogues that were, as chemosensitizers,
slightly more potent than the lead and, at the same time,
showed lower activity on the cardiovascular system in
some cases. On the basis of these results, for the last
few years we have been synthesizing different sets of
compounds, in which the backbone of our derivatives
has been substituted with several kinds of large and lipo-
philic amines.26–28 Our research has led to the discovery
H3CO
H3CO
CH3
N
CN
OCH3
OCH3
verapamil
R
N
CN
R1
1-8
OCH3
OCH3
OCH3
OCH3
CH3
R
H
N
N
=
(CH2)2
N
(CH2)2
R1
2
1
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
H
N
N
N
H
3
4
5
OCH3
OCH3
H
H
H
N
N
S
N
(CH2)2 N
8
6
7
Chart 1.

S. Dei et al. / Bioorg. Med. Chem. 13 (2005)985–998
987
of some promising derivatives that are more potent as
MDR reverters, with respect to the lead verapamil,
and are endowed with reduced cardiovascular activity.
Among them compound MM36 is active at a nanomolar
concentration on anthracycline-resistant erythroleu-
kemia K 562 cell lines.26
tetraline and 6,7-dimethoxy-1,2,3,4-tetrahydro-2-iso-
quinoline can be considered restricted flexibility
analogues of the previous amines, and have already been
successfully inserted in another series of MDR modula-
tors.21 The 9-aminomethylanthracene and 9-amino-
methylthioxantene moieties are present in the previously
mentioned MM3630 and in its thioxantene analogue,28
which is almost completely devoid of cardiovascular
activity. Finally, compound 8 was synthesized to insert
in this series some of the features of XR 9576,31 a potent
modulator of MDR.
Since the cyclohexyl moiety present in some of our rigid
derivatives seemed to reduce the residual cardiovascular
action of verapamil-derived compounds, we have
decided to synthetize and study the series of compounds
shown in Chart 1. There are indications that weak polar
interactions such as those produced by the overlapping
of p orbitals of aromatic rings can play an important
role in stabilizing the binding of MDR-reverting agents
to P-gp protein:29 on this basis, all the newly designed
derivatives bear a diphenyl group on the carbon in posi-
tion 1 of the cyclohexane, instead of the dimethoxyphen-
ylisopropyl moiety characteristic of verapamil and
present in the original derivatives. This general structure
was modulated by inserting different lipophilic amines in
position 3 (Chart 1). N-Methylhomoveratrylamine was
chosen because it is present in the lead compound verap-
amil; also the corresponding desmethyl amine was uti-
lized in order to evaluate the activity of secondary
amine derivatives that often result more effective21 in
comparison with the corresponding N-methylated ones.
5,6-Dimethoxy-2-aminoindane, 6,7-dimethoxy-2-amino-
2. Chemistry
The synthetic pathways used to obtain the desired com-
pounds are shown in Schemes 1 and 2. The first step of
the procedure is the Michael addition of diphenylaceto-
nitrile to 2-cyclohexen-1-one, leading to compound 9
(Scheme 1). This compound is among the many claimed
in a Japanese patent;32 no details on the synthesis have
been, however, reported. We performed this reaction
both using butyllithium and LDA (lithium diisopropyl-
amide) at a low temperature. The best yield (47%) was
obtained with butyllithium at ꢀ78 ꢁC, but even in this
case side reactions could not be eliminated, and the reac-
tion of butyllithium with the CN group gave rise to the
by-product 1,1-diphenyl-2-hexanone.
CN
H
b
CN
O
9
CN
OH
CN
c
OSO2CH3
11a cis
10a + 10b (cis + trans, 9:1)
NC
Br
NC
Ar
N
12a cis (33%)
d
e
CH3
Br
NC
2a cis
12b trans (67%)
OCH3
OCH3
Ar = CH2CH2
Scheme 1. Reagents and conditions: (a) BuLi, 2-cyclohexen-1-one, ꢀ78 ꢁC; (b) NaBH4; (c) CH3SO2Cl; (d) LiBr, chromatographic separation; (e) N-
methylhomoveratrylamine.

988
S. Dei et al. / Bioorg. Med. Chem. 13 (2005)985–998
a
CN
CN
N
R
O
OCH3
OCH3
9
13 R = CH3,
CH CH
14 R =
2
2
c
b
R
N
CN
Ti (OiPr)3
NRR1
CN
O
d
R1
1-8, 15a
OCH3
OCH3
OCH3
CH3
H
(CH2)2
OCH3
N
(CH2)2
N
2
1
R
OCH3
OCH3
N
=
OCH3
OCH3
OCH3
OCH3
R1
H
N
N
N
H
3
4
5
OCH3
OCH3
H
H
H
N
N
S
N
(CH2)2 N
8
6
7
H
N
15
CH3
Scheme 2. Reagents and conditions: (a) 13: methylamine, toluene, p-toluenesulfonic acid, 14: homoveratrylamine, benzene, p-toluenesulfonic acid;
(b) 15a: NaBH4 in MeOH, 1a: H2, Pd/C in EtOH or NaBH4 in MeOH; (c) titanium(IV) isopropoxide, the suitable amine; (d) NaBH3CN.
In a first attempt, looking for a general procedure that
could be applied to all our derivatives, cyclohexanone
9, was reduced with NaBH4 yielding two isomeric alco-
hols (10a cis and 10b trans, Scheme 1) in a 9:1 ratio. The
alcohol mixture was transformed into the corresponding
mesyl derivatives 11 in almost the same isomeric ratio;
crystallization of this mixture yielded the cis isomer
11a, which was reacted with LiBr. Bromination oc-
curred with partial inversion, giving the two isomeric
bromides 12b (trans) and 12a (cis) in a 2:1 ratio. This
mixture was separated by chromatography. The stereo-
chemistry of these compounds was attributed on the
basis of the NMR characteristics of the C3 proton (see
next section). However, both isomers 12a and 12b, when
reacted with N-methylhomoveratrylamine, gave only the
most stable isomer 2a (cis).
bases obtained reduced with NaBH4 or H2/Pd/C to ob-
tain compounds 15 and 1. From both 1 and 15 it was
possible, using the proper reagents (methylamine and
2-(3,4-dimethoxyphenyl)ethyl bromide, respectively) to
obtain compound 2. However, using this procedure
yields were very poor and only the cis isomers 1a and
15a were obtained.
The desired amines 1–8 were eventually obtained
in a good yield, according to the Mattson procedure33
by reductive alkylation of 9 with the suitable amine,
using titanium(IV) isopropoxide as Lewis acid cata-
lyst and NaBH3CN as reducing agent (Scheme 2).
The amines used are commercially available (6,7-
dimethoxy-1,2,3,4-tetrahydroisoquinoline.HCl, 2-(3,4-
dimethoxyphenyl)ethylamine or homoveratrylamine,
2-[(3,4-dimethoxyphenyl)ethyl]methylamine) or N-meth-
ylhomo-veratrylamine or synthesized according to
the literature ((9H-thioxanthen-9-yl)methylamine),28
4-[2-(1,2,3,4-tetrahydro-6,7-dimethoxy-2-isoquinolyl)-
ethyl]benzenamine,34 anthracen-9-ylmethylamine,26 5,6-
Therefore, in order to obtain both isomers, we decided
to explore the reduction of the Schiff bases according
to Scheme 2. Accordingly, compound 9 was reacted with
methylamine and homoveratrylamine and the Schiff

S. Dei et al. / Bioorg. Med. Chem. 13 (2005)985–998
989
dimethoxyindan-2-ylamine,35 6,7-dimethoxy-1,2,3,4-tet-
rahydronaphthalen-2-ylamine, procedure slightly modi-
fied with the addition of molecular sieves.36 This
procedure made it possible to obtain each diastereoiso-
mer in a sufficient quantity to perform the pharmacolog-
ical and biochemical tests. In fact, the compounds of the
series possess two stereogenic centres (three in the case
of 3) and, therefore, four optical isomers are possible
(eight for 3) as two diastereoisomeric racemic mixtures.
At this stage of the research, also considering the modest
impact of stereochemistry in the MDR activity of verap-
amil-like compounds,37 we decided to postpone this
problem until complete evaluation of their pharmaco-
logical profiles.
As regards the final compounds 1–8, structure attribu-
tion was less simple, because in some cases the diagnos-
tic signals are partially obscured (Table 1). Also here,
the proton in position 1 always shows axial characteris-
tics, suggesting that the bulky substituent prefers an
equatorial position. As regards the protons in 3 how-
ever, while for the couples of compounds 1, 5, 6, 7
and 8 attribution is quite straightforward, in the case
of 2a and 2b the coupling constant of the less abundant
isomer (2b) is too high to safely attribute a trans stereo-
chemistry. To solve this problem, we performed X-ray
crystallography (see Fig. 1), which confirmed that the
most abundant compound, 2a, has a cis geometry, leav-
ing to compound 2b a trans stereochemistry. In the case
of couple 4, where the diagnostic signals are obscured,
the attribution was based mainly on the fact that in this
series of compounds, the cis isomer is consistently found
to be the most abundant one in the mixture. This attri-
bution is also supported by the d values of the C3 pro-
tons. In fact, their signals are at higher field in
compound 4a, with respect to the corresponding b iso-
mers, suggesting a cis configuration for the former,
and a trans geometry for the latter. As regards com-
pound 3, which presents three stereocentres, NMR spec-
troscopy allowed us to establish that both the
diastereoisomeric mixtures isolated (3a1 and 3a2) possess
a cis geometry with respect to the cyclohexane ring (H3:
3a1 d = 3.16 ppm, Jaa = 11.4 Hz; 3a2 d = 3.05 ppm, w/2 =
25.8 Hz).
In the case of 3 (the only compound that presents three
stereocentres), two out of the four possible racemates
were isolated in good yield, the others being present only
in traces. NMR spectroscopy indicated that the two dia-
stereoisomers isolated possess a cis geometry with re-
spect to the cyclohexane ring.
3. Stereochemistry
The relative stereochemistry of the substituents in posi-
tion 1,3 on the cyclohexane ring of the synthesized com-
pounds was attributed on the basis of the 1H NMR
characteristics of the C1 and C3 protons. In fact, in most
cases, it was possible to extract the Jaa (and sometimes
the Jae and Jee) constants; when the signal does not
allow extraction of the coupling constants, the chemical
shift, as well as the half-height amplitude of the signal
(w/2),27,38 allow confident attribution of their equatorial
or axial nature. Moreover, the chemical shift of the sig-
nals is diagnostic, because axial protons resonate at
higher fields, while the corresponding equatorial ones
are less shielded.
4. Pharmacology
4.1. MDR-reverting activity
The ability of the compounds to revert MDR was evalu-
ated on anthracycline-resistant erythroleukemia K 562
cells, measuring the uptake of THP-adriamycin (piraru-
bicin) by continuous spectrofluorometric signal of the
anthracycline at 590 nm (kex = 480 nm) after incubation
of the cells, following the protocols reported in previous
papers.21,26,27 Greater detail is provided in the experi-
mental part of the present paper. MDR-reverting activ-
ity is described by (i) a, which represents the fold
increase in the nuclear concentration of pirarubicin in
the presence of the MDR-reverting agent and varies be-
tween 0 (in the absence of the inhibitor) and 1 (when the
amount of pirarubicin in resistant cells is the same as in
sensitive cells); (ii) amax, which expresses the efficacy of
MDR-modulator and is the maximum increase that
can be obtained in the nuclear concentration of piraru-
In the case of the two isomeric alcohols (10a cis and 10b
trans), the proton in position 1 in both compounds
shows axial characteristics (d = 2.55 ppm, Jaa = 11.4 Hz,
Jae not detectable), suggesting that, as expected, the
bulky substituent prefers an equatorial position. On
the other hand, the proton in position 3 shows prevalent
axial characteristics (d = 3.60 ppm, w/2 = 26.5 Hz) in the
most abundant isomer (10a), as compared with that of
the minor isomer (10b) which shows equatorial charac-
teristics (d = 4.20 ppm, w/2 = 8.2 Hz). This indicates that
10a has the hydroxy group in an equatorial position,
implying that the bulky diphenylacetonitrile residue
and the hydroxy group are cis to each other. Accord-
ingly, 10b will have the bulky residue and the hydroxy
group trans to each other. The same applies to the bro-
mo derivatives 12b (trans) and 12a (cis). Also in this
case, in both compounds, the C1 protons show
axial characteristics (d = 3.20 ppm, Jaa = 11.6 Hz,
Jae = 3.2 Hz for 12b and d = 2.58 ppm, Jaa = 12.0 Hz,
Jae = 4.0 Hz for 12a), whereas the C3 protons show
equatorial characteristics in the case of the trans isomer
(d = 4.77 ppm, Jae = 2.8 Hz) and axial character-
istics for the cis one (d = 4.00 ppm, Jaa = 11.8 Hz,
Jae = 4.2 Hz).
bicin in resistant cells with a given inhibitor; (iii) [i]0.5
,
which measures the potency of MDR-reverting agent
and represents the concentration of the inhibitor that
causes a half-maximal increase in nuclear concentration
of pirarubicin at a = 0.5 (see Table 2).
4.2. Cardiovascular activity
Inotropic and chronotropic activities were tested on
guinea-pig isolated atria preparations, and vasodilator
activity was tested on guinea-pig aortic strip prepara-
tions following standard procedures,39 details of which

990
S. Dei et al. / Bioorg. Med. Chem. 13 (2005)985–998
Table 1. 1H NMR signals due to C1 and C3 protons at 200 MHz
R
N
H3
H1
NC
R1
Compound
H1
Axial H3 (cis)
Equatorial H3 (trans)
1a
d = 2.58 ppma,b
Jaa = 11.0 Hz
Jae = 4.0 Hz
d = 2.60 ppma,b
Jaa = 12.0 Hz
Jae = 4.0 Hz
1b
d = 2.82–2.70 ppma,c
d = 3.08 ppma
w/2 = 18.0 Hzd
2a
2b
d = 2.76–2.53 ppmc
d = 2.76–2.54 ppmc
d = 2.76–2.53 ppmc
d = 3.00 ppm
J = 10.4 Hz
3a1
3a2
d = 2.96–2.47 ppmc
d = 2.95–2.38 ppmc
d = 3.16 ppm
Jaa = 11.4 Hz
d = 3.05 ppm
w/2 = 25.8 Hzd
d = 2.68–2.56 ppmc
4a
4b
5a
d = 2.68–2.56 ppmc
d = 3.16–3.02 ppmc
d = 2.70–2.45 ppmc
d = 3.16–3.02 ppmc
d = 3.20 ppm
Jaa = 11.5 Hz
5b
6a
6b
7a
7b
8a
d = 2.83–2.58 ppmc
d = 2.71–2.49 ppmc
d = 2.60–2.21 ppmc
d = 3.75 ppm
w/2 = 12.0 Hzd
d = 2.87 ppm
Jaa = 15.6 Hz
d = 3.10 ppmc
w/2 = 11.0 Hzd
d = 2.43 ppmb
Jaa = 11.0 Hz
d = 2.95–2.60 ppmc
d = 2.49 ppmb
Jaa = 10.8 Hz
d = 2.99 ppm
w/2 = 10.0 Hzd
d = 2.43 ppm
Jaa = 13.5 Hz
Jae = 2.2 Hz
d = 2.43 ppm
Jaa = 11.6 Hz
d = 3.00–2.60 ppmc
8b
d = 3.14 ppm
Jae = 4.2 Hz
a Evaluated on the 400 MHz 1H NMR spectra.
b The attribution of H1 and H3 signal can be interchanged.
c Signal obscured.
d The use of w/2 (half-height width) to estimate the equatorial or axial properties of cyclohexane protons is a simple way of evaluating the size of
coupling constants when the exact value cannot be obtained (see text).
Figure 1. Thermal ellipsoid plot (30% ellipsoids) of compound 2a oxalate.

Table 2. Cardiovascular and chemosensitizer activity of compounds 1–8
Cardiovascular activity
Negative chronotropy
MDR activity
Negative inotropy
Vasorelaxant activity
Comp
Ia %a ( SEM)
EC50 (lM)b
(95% cl)
Ia %c ( SEM)
EC30 (lM)b
(95% cl)
Ia %d ( SEM)
IC50 (lM)b
(95% cl)
[i]0.5 lM
a max
1a
1b
50 3.6
46 2.4
63 3.5
33 1.7
46 1.4
40 0.8f
81 3.6
45 1.4
75 2.6f
61 1.5
56 1.2
54 3.4
38 1.5
32 2.3f
41 1.4e
16 0.4g
84 2.1f
55 0.2
68 3.4
67 2.5
90 1.4
73 2.5e
49 3.8
68 3.4
75 4.5f
73 3.2
21 1.9
65 2.7
18 0.8
22 1.8
24 1.7
14 1.1f
41 2.6
31 2.4
94 3.4g
65 4.3
2.07
1.42
3.75
1.67
(1.82–2.15)
(1.31–1.62)
(3.39–4.41)
(1.31–1.92)
40 2.5
15 0.7
41 2.3
23 1.6
20 1.3
26 1.5
31 2.1
25 1.5
15 1.1
22 0.2
0.44 0.09
0.34 0.07
0.10 0.02
0.68 0.14
0.36 0.07
6.2 1.2
0.86
0.88
0.87
0.85
0.90
0.80
0.90
0.90
0.83
0.86
0.85
ꢁ1
2a
2b
1.49
(1.06–2.11)
3a1
3a2
4a
1.57
1.68
6.09
(1.25–1.90)
(1.45–2.02)
(5.81–6.38)
1.84
(1.36–2.91)
0.14 0.03
0.29 0.06
1.30 0.26
0.50 0.1
1.5 0.3
4b
5a
0.41
1.58
1.02
0.83
(0.27–0.60)
(1.12–2.23)
(0.79–1.21)
(0.62–0.98)
5b
6a
4.97
(4.66–4.72)
6
5
2
2
0.3
0.4
0.2
0.1
6b
7a
50 10
0.90 0.18
1.70 0.2
0.54 0.11
1.40 0.30
1.6 0.3
0.83
0.77
0.88
ꢁ1
7b
8a
37 1.3
19 0.3
95 1.7g
29 2.3
8b
VRP
MM36
0.61
1.11
(0.40–0.80)
(0.85–1.45)
0.07
0.86
(0.05–0.10)
(0.76–1.00)
0.38
(0.20–0.70)
0.70
0.70
0.05 0.01
a Intrinsic activity: decrease in the developed tension in isolated guinea-pig left atrium at 5 · 10ꢀ5 M, expressed as percent changes from the control (n = 4–6). The left atria were driven at 1 Hz. The
5 · 10ꢀ5 M concentration gave the maximum effect for most compounds.
b Calculated from log concentration–response curves (Probit analysis by Litchfield and Wilcoxon with n = 5–6). When the maximum effect was <50%, the EC50 ino., EC30 chrono., IC50 values were not
calculated.
c Intrinsic activity: decrease in the atrial rate on guinea-pig spontaneously beating isolated right atrium at 5 · 10ꢀ5 M, expressed as percent changes from the control (n = 6–7). The pretreatment heart rate
ranged from 165 to 190 beats/min. 5 · 10ꢀ5 M gave the maximum effect for most compounds.
d Intrinsic activity: percent inhibition of calcium-induced contraction on K+-depolarized guinea-pig aortic strip at 5 · 10ꢀ5 M (n = 5–6). The 5 · 10ꢀ5 M concentration gave the maximum effect for most
compounds.
e At the 5 · 10ꢀ6 mol/L concn.
f At the 10ꢀ5 mol/L concn.
g At the 10ꢀ6 mol/L concn.

992
S. Dei et al. / Bioorg. Med. Chem. 13 (2005)985–998
are reported in the experimental section. Potency of the
drug is defined as EC50 (negative inotropic activity),
IC50 (vasodilator activity), and EC30 (negative chrono-
tropic activity). Activity is defined as the percent de-
crease in developed tension on isolated left atrium
(negative inotropic activity), percent decrease in atrial
rate of spontaneously beating isolated right atrium (nega-
tive chronotropic activity), and percent inhibition of
calcium-induced contraction on K+-depolarized aortic
strips (vasodilator activity) at the concentrations indi-
cated in the footnotes of Table 2.
secondary amines.21 For example, compounds 1 and 2
show more or less the same profile, with [i]0.5 ranging
from 0.10 (2a) to 0.68 (2b) lM, and with a similar car-
diovascular activity. Restricting the conformational
freedom of the amine moiety does not ameliorate the
pharmacological profile of the compounds and does
not seem to induce major changes, as shown by the sim-
ilar profiles of 3, 4, 5 for both MDR and cardiovascular
activity, with the notable exception of 3a1, that com-
pletely loose any cardiovascular action. At the contrary,
the introduction of flat and bulky groups on the nitro-
gen atom, as happens in compounds 6, 7, and 8 pro-
duces interesting effects: compound 6 shows only a
residual negative inotropic activity and, as previously
observed in other series,28 introduction of the thioxan-
tene moiety (7) does abolish cardiovascular activity.
The same happens for compound 8. These results con-
firm the importance of lipophilic moiety to separate
cardiovascular and anti-MDR activity.
5. Results and discussion
The results of MDR-reverting activity and cardiovascu-
lar effect of our compounds are reported in Table 2
where the corresponding values for verapamil and
MM36 are reported as reference. In general, all com-
pounds show good MDR-reverting activity with a po-
tency that is comparable to that of verapamil and, in
several cases, higher. Only 3a2 and 6b show a decrease
in the chemosensitizing effect with respect to the refer-
ence compound. On the contrary, cardiovascular activ-
ity is, as expected, always lower. Actually, some
compounds show very low intrinsic activity and may
be considered inactive, both on myocardial and vascular
preparations. As a consequence, when the maximum ef-
fect, that is intrinsic activity (evaluated up to the
5 · 10ꢀ5 M concentration), was lower than 50%, the cor-
responding EC50 inotropic, EC30 chronotropic, and IC50
values were not calculated.
In accordance with previously reported results, stereo-
chemistry does not seem to play a major role.21 In fact,
cis isomers seem to be more effective as MDR antago-
nists with respect to the trans ones, but the differences
are not so impressive, the largest difference being found
between compounds 6a and 6b. The tetrahydronaphtha-
lene derivatives are an exception that will require further
studies: 3a1 is 20 times more potent than 3a2 as chemo-
sensitizer and completely lacks cardiovascular activity,
while its less MDR-potent isomer maintains a residual
negative chronotropic action.
In conclusion, we have synthesized a new series of po-
tent MDR reverters that are only loosely related to
verapamil. Three compounds of the series (3a1, 7a,
8a), which are more potent than verapamil and lack
any cardiovascular action, deserve further studies and
can be useful leads for the development of new MDR
reversing drugs.
Considering this new series of derivatives, compounds
2a ([i]0.5 = 0.10 lM) and 4a ([i]0.5 = 0.14 lM) show the
highest chemosensitizing effect, with a [i]0.5 that is one
order of magnitude lower with respect to the reference
compound verapamil. They lack vasorelaxant activity,
but maintain a residual cardiac effect, both on inotropy
and chronotropy. Compounds 1a ([i]0.5 = 0.44 lM), 1b
([i]0.5 = 0.34 lM) and 2b ([i]0.5 = 0.68 lM) also possess
good anti-MDR activity showing only a modest chrono-
tropic effect. The best results were obtained with com-
pound 4b ([i]0.5 = 0.29 lM) showing only a residual
negative chronotropic effect (100 times lower than that
6. Experimental
6.1. Chemistry
of verapamil) and with compounds 7a ([i]0.5
0.90 lM), 8a ([i]0.5 = 0.54 lM) and 3a1 ([i]0.5
=
=
All melting points were taken on a Buchi apparatus and
¨
are uncorrected. Infrared spectra were recorded with a
Perkin–Elmer 681 or a Perkin–Elmer Spectrum RX I
FT-IR spectrophotometer in Nujol mull for solids and
neat for liquids. Mass spectra were acquired by a triple
quadrupole mass spectrometer API 365 equipped with a
TurboIon Spray interface from Applied Biosystems.
Spectra were recorded in positive ion mode in the mass
range from 120 to 620 Th, at unit mass resolution. GC–
MS spectra were acquired by a Perkin–Elmer 8420 cap-
illary gas chromatograph, connected to a Perkin–Elmer
Ion Trap detector. Unless otherwise stated, NMR spec-
tra were recorded on a Gemini 200 spectrometer
0.36 lM), that lack any cardiovascular effect. Therefore,
the first result to mention is that, as hypothesized, the
diphenylmethyl cyclohexane moiety decreases cardio-
vascular activity without affecting MDR-reverting prop-
erties which, on the contrary, are in most cases exalted
in the new derivatives.
Analysis of MDR structure–activity relationships does
not give univocal answers. A first aspect to evaluate is
the influence of the amine moieties on the pharmacolog-
ical profile of our derivatives. Comparing the activity of
compounds 1–5, there is no apparent trend relating
MDR-reverting activity and selectivity with the number
of alkyl substituents on the nitrogen, contrary to the
suggestions of previous works which indicate a higher
MDR activity and a lower cardiovascular activity for
1
(200 MHz for H NMR, 50.3 MHz for 13C), and chro-
matographic separations were performed on a silica
gel column by gravity chromatography (Kieselgel 40,
0.063–0.200 mm; Merck) or flash chromatography (Kie-
selgel 40, 0.040–0.063 mm; Merck). Although the IR

S. Dei et al. / Bioorg. Med. Chem. 13 (2005)985–998
993
spectra data are not always included, they were obtained
for all compounds reported and are consistent with the
assigned structures. Yields are given after purification,
unless otherwise stated. Where analyses are indicated
by symbols, the analytical results are within 0.4% of
the theoretical values. Compounds were named follow-
ing IUPAC rules as applied by Beilstein-Institut Auto-
Nom (version 2.1), a software for systematic names in
organic chemistry. When reactions were performed in
anhydrous conditions, the mixtures were maintained
under nitrogen.
6.1.3. cis Methanesulfonic acid 3-(cianodiphenylmethyl)-
cyclohexyl ester (11a). A solution of 10 (as the mixture
of the cis and the trans isomer) (890 mg, 3.06 mmol)
and triethylamine (0.6 mL, 440 mg, 4.32 mmol) in anhy-
drous CH2Cl2 was treated with 0.36 mL (500 mg,
4.32 mmol) of methanesulfonyl chloride dissolved in
CH2Cl2 at ꢀ20 ꢁC and kept at this temperature for
2 h. The mixture was then allowed to warm to room
temperature overnight. The solution was washed with
water, dried, and evaporated under vacuum to give a
solid that was purified by crystallization from ethanol
(mp 90–92 ꢁC). Yield: 1.06 g (2.66 mmol, 87%). TLC
and GC–MS indicated a small amount of the trans
isomer that remained in the mater liquor.
6.1.1. (3-Oxocyclohexyl)diphenylacetonitrile (9). A solu-
tion of butyllithium (12 mL of a 1.6 M solution in hex-
ane, 19.2 mmol) was added to
a
solution of
diphenylacetonitrile (3 g, 15.52 mmol) in anhydrous
THF at ꢀ78 ꢁC. The mixture turned yellow and was
kept at ꢀ78 ꢁC for 1 h. Then 1.8 mL (1.79 g, 18.59
mmol) of 2-cyclohexen-1-one were added, and the reac-
tion was left to reach room temperature overnight. To
the crude reaction mixture, kept at 0 ꢁC, a saturated
aqueous NH4Cl solution was added, the layers were sep-
arated, the aqueous solution was re-extracted with
diethyl ether, and the two organic extracts were
combined, washed with water and dried over Na2SO4.
The solvent was removed in vacuo, and the crude prod-
uct was purified by flash chromatography on silica gel,
using cyclohexane/ethyl acetate (70:30) as the eluent.
Yield 2.11 g (7.30 mmol, 47%).
1H NMR d 7.52–7.23 (m, 10H, aromatics), 4.69 (tt,
J = 4.6 Hz, J = 11.0 Hz, 1H, CHOSO2) 2.94 (s, 3H,
CH3SO2) 2.62 (tt, J = 3.2 Hz, J = 12.0 Hz, 1H, CH–C–
CN), 2.30–1.85 (m, 3H, cyclohexane protons), 1.74–
1.22 (m, 5H, cyclohexane protons); GC–MS of the mix-
ture: first eluted compound 11a (cis): 399 (M+Å), 274
(M+Åꢀ125), 247 (M+Åꢀ152), 193 (M+Åꢀ206, 100%), sec-
ond eluted compound 11b (trans): 399 (M+Å), 274
(M+Åꢀ125), 247 (M+Åꢀ152), 193 (M+Åꢀ206, 100%).
Anal. (C21H23NO3S) C, H, N.
6.1.4. cis (3-Bromocyclohexyl)diphenylacetonitrile (12a)
and trans (3-bromocyclohexyl)diphenylacetonitrile (12b).
Compound 11a (650 mg, 1.63 mmol) was dissolved in
dry acetone, and a solution of LiBr (984 mg, 7 equiv,
11.41 mmol) in dry acetone was added. The mixture
was heated to reflux for 38 h, then the solvent was re-
moved and the residue dissolved in CH2Cl2, washed
with water and dried. Evaporation of the solvent gave
a thick oil which was a mixture of the two isomers
(TLC) that were separated by column chromatography
using ethyl acetate/cyclohexane (10:90) as eluting
system.
1H NMR (CDCl3) d 7.52–7.25 (m, 10H, aromatics),
3.04–2.86 (m, 1H, CH–C–CN), 2.50–2.20 (m, 3H, cyclo-
hexane protons), 2.18–2.04 (m, 1H, cyclohexane pro-
ton), 2.00–1.60 (m, 4H, cyclohexane protons); MS: 289
(M+Å), 271 (M+Åꢀ16), 193 (M+Åꢀ96, 100%). Anal.
(C20H19NO) C, H, N.
6.1.2. (3-Hydroxycyclohexyl)diphenylacetonitrile (10). A
portion of 9 (1.11 g, 3.84 mmol) in anhydrous THF
was added dropwise to a suspension of 436 mg (3 equiv,
11.52 mmol) of NaBH4 in anhydrous THF at 0 ꢁC. The
mixture was kept 2 h at 0 ꢁC, then was allowed to warm
to room temperature overnight. The mixture was then
cooled at 0 ꢁC and 5 mL of a 30% solution of NH4OH
added. The organic layer was separated, dried and evap-
orated to give a white solid (mp 48–50 ꢁC). TLC and
GC–MS showed the presence of two isomers. The mix-
ture was then column chromatographed, using ethyl
acetate/cyclohexane (30:70) as the eluent. The chroma-
tography yielded 810 mg (2.78 mmol, 73% yield) of a
mixture of the two isomers, which could not be
separated.
The trans isomer 12b was eluted first (240 mg, 42%
1
yield). H NMR d 7.57–7.21 (m, 10H, aromatics), 4.77
(t, J = 2.8 Hz, 1H, CHBr) 3.20 (tt, J = 3.2 Hz,
J = 11.6 Hz, 1 H, CH–C–CN), 2.40–1.20 (m, 8H, cyclo-
hexane protons); MS: 355/353 (1:1, M+Å), 329–327 (1:1,
M+Åꢀ26), 273 (M+Åꢀ82/80), 193 (M+Åꢀ162/160, 100%).
Anal. (C20H20BrN): C, H, N.
The second fraction was the cis isomer 12a (130 mg, 23%
1
yield). H NMR d 7.56–7.21 (m, 10H, aromatics), 4.00
(tt, J = 4.2 Hz, J = 11.8 Hz, 1H, CHBr) 2.58 (tt,
J = 4.0 Hz, J = 12.0 Hz, 1H, CH–C–CN), 2.36–1.32
(m, 8H, cyclohexane protons); MS: 355/353 (1:1, M+Å),
329/327 (1:1, M+Åꢀ26), 273 (M+Åꢀ82/80), 193
(M+Åꢀ162/160, 100%). Anal. (C20H20BrN) C, H, N.
1H NMR d 7.50–7.20 (m, 10H, aromatics), 4.20 (br s,
w/2 = 8.2 Hz, 0.1H), 3.60 (br s, w/2 = 26.5 Hz, 0.9H)
(CH–O), 2.55 (t, J = 11.4 Hz, 1H, CH–C–CN), 2.10–
1.77 (m, 3H, OH and cyclohexane protons), 1.70–1.47
(m, 2H, cyclohexane protons), 1.43–1.15 (m, 4H, cyclo-
hexane protons); GC–MS first eluted compound 10b
(trans): 291 (M+Å), 274 (M+Åꢀ17), 247 (M+Åꢀ44), 193
(M+Åꢀ98, 100%), second eluted compound 10a (cis):
291 (M+Å), 274 (M+Åꢀ17), 247 (M+Åꢀ44), 193 (M+Åꢀ98,
100%). Anal. (C20H21NO) C, H, N.
6.1.5. cis (3-Methylaminocyclohexyl)diphenylacetonitrile
(15a). A 20% toluene solution of methylamine (2 mL), 9
(0.5 g, 1.73 mmol) and p-toluenesulfonic acid monohy-
drate (150 mg) was kept 24 h at 110 ꢁC in a small steel
bomb. Evaporation of the solvent gave an oil that was
dissolved in CH2Cl2, washed with water and dried. Re-
moval of the solvent gave 420 mg of the Schiff base (3-
methyliminocyclohexyl)diphenylacetonitrile (13) as an

994
S. Dei et al. / Bioorg. Med. Chem. 13 (2005)985–998
oil that was used as such in the following reaction. IR
(neat) m 2250 (CN), 1670 (C@N) cmꢀ1. Then NaBH4
(60 mg) was cautiously added to a solution of the crude
Schiff base 13 in hot methanol (10 mL) over a period of
0.5 h. The mixture was then heated to reflux for 5 h.
After cooling, the mixture was treated with a few drops
of water and the excess solvent removed. The residue
was dissolved in CH2Cl2, washed with a saturated solu-
tion of NaHCO3 and dried. Evaporation of the solvent
gave an oil that was separated by column chromatogra-
phy with chloroform/methanol (97:03) as the eluent,
yielding 40 mg of the title compound.
6.1.7.
cis[3-[2-[2-(3,4-Dimethoxyphenyl)ethyl]methyl-
amino]cyclohexyl]diphenylacetonitrile (2a). Compound
trans 12b (220 mg, 0.62 mmol) were added to 0.11 mL
(120 mg, 0.62 mmol) of N-methylhomoveratrylamine in
triethylamine. The mixture was refluxed for 20 h.
CH2Cl2 was then added, the organic layers were washed
with water and a saturated solution of NaHCO3, and
dried. Evaporation of the solvent gave the crude product
that was purified by column chromatography using
chloroform/methanol (97:03) as the eluent. Yield:
50 mg of isomer 2a (0.11 mmol, 17%).
With the same procedure described above, and starting
from the cis isomer 12a, the same compound 2a was ob-
tained in 11% yield.
1H NMR d 7.60–7.30 (m, 10H, aromatics), 2.52 (s, 3H,
NCH3), 2.46 (br s, w/2 = 22 Hz, 1H, CHN), 2.20–1.60
(m, 10H, cyclohexane protons and NH) ppm. Anal.
(C21H24N2) C, H, N.
6.1.8. General procedure for the synthesis of diphenylcy-
clohexane derivatives (1–8). A mixture of 9 (300 mg,
1.09 mmol), an equimolar amount of the suitable amine
(1.09 mmol) and an excess of titanium(IV) isopropoxide
(0.45 mL, 1.36 mmol) was stirred with a drying tube at
room temperature for 2–10 h and heating at 60 ꢁC if
necessary, depending on the starting compound. When
necessary, a few drops of benzene were added in order
to make the mixture homogeneous. After a suitable
time, the IR spectrum of the mixture showed no ketone
band, and the viscous solution was diluted with abs eth-
anol (2 mL). Sodium cyanoborohydride (60 mg, 0.74
mmol) was added, and the solution was stirred for
20 h. Water (0.5 mL) was then added, and the mixture
was concentrated in vacuo. The crude product was
dissolved in CH2Cl2, filtered to remove the solids,
washed with a solution of NaHCO3 and water, dried
over Na2SO4 and concentrated in vacuo. The crude sub-
stance was then purified by column chromatography.
Total yield ranged from 40% to 85%. The eluents used
and the diastereoisomers yield ratios are reported in
Table 3. The obtained diastereoisomers were trans-
formed into the corresponding oxalates by treatment
of the resulting compounds with an equimolar amount
of oxalic acid in ethyl acetate.
6.1.6.
[3-[2-(3,4-Dimethoxyphenyl)ethylamino]cyclo-
hexyl]diphenylacetonitrile (1). A solution of 9 (530 mg,
1.83 mmol), homoveratrylamine (0.34 mL, 360 mg,
1.99 mmol) and p-toluenesulfonic acid monohydrate
(150 mg) in anhydrous benzene was heated to reflux
for 24 h and water was removed from the reaction with
the aid of a Dean–Stark trap. The solvent was distilled
off and the residue dissolved in CH2Cl2, washed with a
saturated solution of NaHCO3 and dried. Evaporation
of the solvent gave 800 mg of Schiff base [3-[2-(3,4-
dimethoxyphenyl)ethylimino]cyclohexyl]diphenylaceto-
nitrile (14) that is quite unstable and was used as such in
the next reaction. IR (neat): m 2250 (CN), 1670 (C@N)
cmꢀ1. The Schiff base was then reduced by two different
procedures.
6.1.6.1. Procedure A. To a solution of crude 14
(400 mg) in hot methanol (10 mL), NaBH4 (68 mg)
was cautiously added over a period of 0.5 h. The mix-
ture was then heated to reflux for 5 h. After cooling,
the mixture was treated with a few drops of water and
the excess solvent removed. The residue was dissolved
in CH2Cl2, washed with a saturated solution of NaH-
CO3 and dried. Evaporation of the solvent gave
380 mg of an oil that was converted into the oxalate to
eliminate non-basic side products. The mixture of
amines obtained from the oxalate (aqueous NaOH and
CH2Cl2 extraction) (300 mg) was separated by column
chromatography with chloroform/methanol (90:10).
The third fraction constituted about 7% of the chroma-
The chemical and physical characteristics of the com-
pounds 1–8 are reported in Table 3. Their IR spectra
are consistent with the proposed structures; their 1H
NMR spectra are reported in the supplementary mate-
1
rial. The H NMR spectra of the two isomers of cis
{3-[2-(3,4-dimethoxyphenyl)ethylamino]cyclohexyl}di-
phenylacetonitrile (1a) and trans {3-[2-(3,4-dimethoxy-
phenyl)ethylamino]cyclohexyl}diphenylacetonitrile (1b)
were performed also at 400 MHz, and these spectra
are reported as an example.
1
tographed mixture and was the cis isomer 1a. Its H
NMR spectrum is consistent with the proposed struc-
ture and is reported in the supplementary material.
1
6.1.6.2. Procedure B. To a solution of crude 14
(400 mg) in abs ethanol (10 mL), Pd/C (120 mg) was
added and the mixture was maintained for 4 days at
82 psi at room temperature in a Parr apparatus. The
mixture was then filtered. Evaporation of the solvent
gave 350 mg of an oil that was separated by column
chromatography with chloroform/methanol (90:10).
The second fraction (70 mg) was constituted by a mix-
ture of the two isomers 1a cis and 1b trans (TLC), which
was not purified further.
1a: H NMR, 400 MHz (CDCl3) d: 7.51–7.45 (m, 4H,
aromatics); 7.37–7.31 (m, 4H, aromatics); 7.29–7.21
(m, 2H, aromatics); 6.85–6.68 (m, 3H, aromatics); 3.85
(s, 3H, OCH3); 3.84 (s, 3H, OCH3); 2.95–2.69 (m, 4H,
CH2–CH2); 2.60 (tt, J = 12.0 Hz, J = 4.0 Hz, 1H, CH–
N); 2.58 (tt, J = 11.0 Hz, J = 4.0 Hz, 1H, CH–C–CN);
2.20 (br s, 1H, NH); 2.02 (d, J = 12.0 Hz, 1H); 1.95 (d,
J = 16.0 Hz, 1H); 1.88–1.80 (m, 1H); 1.66–1.60 (m,
1H); 1.38–1.18 (m, 3H); 1.12–1.04 (m, 1H) (cyclohexane
protons).

S. Dei et al. / Bioorg. Med. Chem. 13 (2005)985–998
Table 3. Chemical and physical characteristics of derivatives 1–8
995
R1
N
CN
R
OCH
3
OCH
CH
OCH
3
3
3
OCH
3
HN (CH )
OCH
N
B =
(CH )
OCH
3
C =
F =
N
A =
HN
HN
2 2
3
2 2
OCH
3
OCH
OCH
3
3
N
D =
HN
E =
OCH
3
HN
(CH )
2 2
G =
H =
S
OCH
3
HN
OCH
3
Compd
–NRR1
Mass (M+H)+a
Eluent
Isomers order of elution and yield ratio
Mp (ꢁC)b
Analysisc
1b
1a
2b
2a
3a1
3a2
4b
4a
5a
5b
6b
6a
7b
7a
8a
8b
A
A
B
455
455
469
469
481
481
467
467
467
467
481
481
501
501
586
586
CHCl3/MeOH
(95:05)
CHCl3
trans, cis
4:6
235–237
200–201
103–105
179–181
257–260
200–204
233–235
239–241
135–136
155–157
95–97
C32H36N2O6
C32H36N2O6
C33H38N2O6
C33H38N2O6
C34H38N2O6
C34H38N2O6
C33H36N2O6
C33H36N2O6
C33H36N2O6
C33H36N2O6
C37H34N2O4
C37H34N2O4
trans, cis
4:6
B
(100)
CHCl3
C
C
D
D
E
cis, cis
4:4d
(100)
CH2Cl2
(100)
trans, cis
3:7
CHCl3/MeOH
(99:01)
cis, trans
7:3
E
F
CHCl3/petr. ether
(90:10)
CHCl3
trans, cis
4:6
F
155–156
218–220
170–171
180–181
163–164
G
G
H
H
trans, cis
3:7
C36H34N2O4S
C36H34N2O4S
C41H45N3O6
C41H45N3O6
(100)
CHCl3/MeOH
(98:02)
cis, trans
9:1
a As free base. Spectra were recorded in positive ion mode at unit mass resolution (see experimental part for details).
b As oxalate. Crystallization solvent:absolute ethanol.
c As oxalate. All compounds have been analyzed for C, H, N after vacuum drying at a temperature below the melting point; the results obtained
range within 0.4% of the theoretical values.
d The two trans isomers represent 20% of the mixture and were not characterized.
1
1b: H NMR, 400 MHz (CDCl3) d: 7.53–7.43 (m, 4H,
aromatics); 7.38–7.21 (m, 6H, aromatics); 6.85–6.73
(m, 3H, aromatics); 3.89 (s, 6H, OCH3); 3.08 (br s, w/2 =
18 Hz, 1H, CH–N); 2.90 (br s, 1H, NH); 2.82–2.70 (m,
5H, CH–C–CN and CH2–CH2); 1.82–1.61 (m, 4H,
cyclohexane protons); 1.61–1.20 (m, 4H, cyclohexane
protons).
syntheses. One independent compound molecule and
one oxalate molecule were found in the asymmetric unit.
Refinement was carried out on F2 by full-matrix least
square techniques, using the SHELX97 program pack-
age.42 Hydrogen atoms, except for the non-salified
hydrogen of oxalic acid, were added in the model
constrained to idealized positions and refined using a
riding model with riding isotropic displacement param-
eters. All the non-hydrogen atoms were refined
anisotropically.
6.1.9. Crystal structure determination and refinement of
2a oxalate. Diffraction data were collected at 293 K on
CuKa radiation, using a rotating anode source and a
1K CCD detector from Bruker equipped with Go¨bel
mirrors. Data were processed using the program SAINT
from Bruker and intensities were corrected for absorp-
tion (SADABS).40 The crystal data and details of the
data collection and structure refinement are summarized
in Table 4. The crystal structure was solved by direct
methods using the SIR97 program41 which gave the posi-
tion of most of the non-hydrogen atoms. The remaining
atoms were identified by successive Fourier difference
In Figure 1 a thermal ellipsoid representation of 2a oxa-
late is shown. Crystallographic data have been deposited
with the Cambridge Crystallographic Data Centre with
number CCDC 228890.
6.2. Pharmacology
6.2.1. MDR-reverting activity. 6.2.1.1. Drugs and chemi-
cals. Purified pirarubicin was provided by Laboratoire

996
S. Dei et al. / Bioorg. Med. Chem. 13 (2005)985–998
Table 4. Crystal data and structure refinement for 2a oxalate
promising cell viability. All experiments were conducted
in 1 cm quartz cuvettes containing 2 mL of buffer at
37 ꢁC using a circulating thermostated water bath. Cells,
2 · 106, were suspended in 2 mL of glucose containing
HEPES buffer at pH 7.3, under vigorous stirring;
20 lL of the stock anthracycline solution was quickly
added to this suspension yielding an anthracycline con-
centration CT equal to 1 lM. The decrease of the fluo-
rescence intensity F at 590 nm was followed as a
function of time. After about 20 min, the curve F = f(t)
reached a plateau and the fluorescence intensity was
equal to Fn. The drug-cells system was thus in a steady
state and the overall concentration Cn of drug interca-
lated between the base pairs in the nucleus was Cn =
CTÆ(F0 ꢀ Fn)/F0. Once the steady state was reached, the
inhibitor at concentration [i] ([i] was varied from 0.05
to 10 lM) was added and a new steady state was
reached, the fluorescence intensity being F in The overall
concentration Cin of drug intercalated between the base
pairs in the nucleus was then Cin ¼ CT Á ðF 0 ꢀ F ni Þ=F 0.
An aliquot of the solution was then taken away and cell
viability was assessed by trypan blue exclusion. Cell
membranes were then permeabilized by the addition of
0.05% Triton X-100 yielding the equilibrium state which
was characterized by a new value FN of the fluorescence
intensity. The overall concentration CN of drug interca-
lated between the base pairs in the nucleus was then
CN = CTÆ(F0 ꢀ FN)/F0. We checked that tested com-
pounds did not affect the fluorescence of THP-adriamy-
cin. a was measured with the following expression:
a ¼ ðCin ꢀ CnÞ=ðCN ꢀ CnÞ.
Formulaa
Formula weighta
Space group
C33H37N2O6
557.65
P-1
Unit cell dimensions
˚
a, A
˚
9.1353(10)
11.2413(12)
15.0484(17)
83.957(5)
87.607(6)
78.541(6)
1505.80
b, A
˚
c, A
a, deg
b, deg
c, deg
3
˚
Volume, A
Z
2
1.230
0.69
Density (calcd), g/cm3
Abs coeff, mmꢀ1
Radiation
CuKa
4971
Reflns collected
No. of unique reflns
Completeness %
Value of Rint
3193
81.6
0.0284
Final R indices [I > 2sigma(I)]
R indices (all data)
Extinction coeff
R1 = 0.0508
R1 = 0.0647, wR2 = 0.1306
0.001573
0.32, ꢀ0.12
ꢀ3
˚
Largest diff. peak and hole, e A
a The non-salified hydrogen of oxalic acid was not introduced in the
refinement (see the experimental part).
Roger Bellon (France). Concentrations were determined
by diluting stock solutions to approximately 10ꢀ5 M
and using e480 = 11,500 Mꢀ1 cmꢀ1. Stock solutions were
prepared just before use. Buffer solutions were HEPES
buffer containing 5 mM HEPES, 132 mM NaCl,
3.5 mM CaCl2, 5 mM glucose, at pH 7.25.
6.2.2. Cardiovascular activity. The pharmacological pro-
file of compounds was tested on guinea-pig isolated left
and right atria to evaluate their inotropic and chrono-
tropic effects, respectively, and on K+-depolarized gui-
nea-pig aortic strips to assess calcium antagonist
activity. At first all compounds were checked at increas-
ing doses to evaluate the percent decrease on developed
tension on isolated left atrium driven at 1 Hz (negative
inotropic activity), the percent decrease in atrial rate
on spontaneously beating right atrium (negative chrono-
tropic activity) and the percent inhibition of calcium-in-
duced contraction on K+-depolarized aortic strips
(vasorelaxant activity). Data were analyzed by Studentꢂs
t-test. The potency of drugs defined as EC50, EC30 and
IC50 was evaluated from log concentration–response
curves (Probit analysis by Litchfield and Wilcoxon,
n = 6–8) in the appropriate pharmacological prepara-
tions. All data are presented as mean SEM.46
6.2.1.2. Cell lines and cultures. K 562 is a human leuk-
emia cell line.43 Cells resistant to doxorubicin were ob-
tained by continuous exposure to increasing
doxorubicin concentrations and were maintained in
medium containing doxorubicin (400 nM) until 1–4
weeks before experiments. This subline expresses a un-
ique membrane glycoprotein with a molecular weight
of 180,000 Da.44 Doxorubicin-sensitive and -resistant
erythroleukemia K 562 cells were grown in suspension
in RPMI 1640 (Sigma) medium supplemented with L-
glutamine and 10% FCS at 37 ꢁC in a humidified atmo-
sphere of 95% air and 5% CO2. Cultures, initiated at a
density of 105 cells/mL grew exponentially to 8–10 ·
105 cells/mL in three days. For the spectrofluorometric
assays, in order to have cells in the exponential growth
phase, culture was initiated at 5 · 105 cells/mL and cells
were used 24 h later, when the culture had grown to
about 8–10 · 105 cells/mL. Cell viability was assessed
by trypan blue exclusion. The cell number was deter-
mined by Coulter counter analysis.
6.2.2.1. Guinea-pig atrial preparations. Guinea pigs
(300–400 g female) were sacrificed by cervical disloca-
tion. After thoracotomy the heart was immediately re-
moved and washed by perfusion through the aorta
with oxygenated Tyrode solution of the following com-
position (mM): 136.9 NaCl, 5.4 KCl, 2.5 CaCl2, 1.0
MgCl2, 0.4 NaH2PO4 · H2O, 11.9 NaHCO3 and 5.5 glu-
cose. The physiological salt solution (PSS) was buffered
at pH 7.4 by saturation with 95% O2–5% CO2 gas, and
the temperature was maintained at 35 ꢁC. Isolated gui-
nea-pig heart preparations were used, spontaneously
beating right atria and left atria driven at 1 Hz. For each
6.2.1.3. Cellular drug accumulation. The uptake of
pirarubicin cells was followed by monitoring the de-
crease in the fluorescence signal at 590 nm
(kex = 480 nm) following the method previously de-
scribed.45 Using this method it is possible to accurately
quantify the kinetics of the drug uptake by the cells
and the amount of anthracycline intercalated between
the base pairs in the nucleus at the steady state, as incu-
bation of the cells with the drug proceeds without com-

S. Dei et al. / Bioorg. Med. Chem. 13 (2005)985–998
997
2. Mitscher, L. A.; Pillai, S. P.; Gentry, E. J.; Shankel, D. M.
Med. Res. Rev. 1999, 19, 477.
3. Bosch, I.; Croop, J. Biochim. Biophys. Acta 1996, 1288,
F37.
4. Volm, M.; Mattern, J. Crit. Rev. Oncogenesis 1996, 7, 227.
5. Aszalos, A.; Ross, D. D. Anticancer Res. 1998, 18, 2937.
6. Johnstone, R. W.; Ruefly, A. A.; Tainton, K. M.; Smyth,
M. J. Leukemia Lymphoma 2000, 38, 1.
7. Johnstone, R. W.; Ruefli, A. A.; Smyth, M. J. Trends
Biochem. Sci. 2000, 25, 1.
preparation, the entire left and right atria were dissected
from the ventricles, cleaned of excess tissue, and hung
vertically in a 15-mL organ bath containing the PSS
continuously bubbled with 95% O2–5% CO2 gas at
35 ꢁC, pH 7.4. The contractile activity was recorded iso-
metrically by means of a force transducer (FT0.3, Grass
Instruments, Quincy, MA) using Power Lab software
(Basile, Italy). The left atria were stimulated by rectan-
gular pulses of 0.6–0.8 ms duration and about 50%
threshold voltage through two platinum contact elec-
trodes in the lower holding clamp (Grass S88 stimula-
tor). The right atrium was in spontaneous activity.
After the tissue was beating for several minutes, a
length–tension curve was determined, and the muscle
length was maintained at which elicited 90% of maxi-
mum contractile force observed at the optimal length.
A stabilization period of 45–60 min was allowed before
the atria were used to test compounds. During the equil-
ibration period, the bathing solution was changed every
15 min and the threshold voltage was ascertained for the
left atria. Atrial muscle preparations were used to exam-
ine the inotropic and chronotropic activity of the com-
pounds (0.1, 0.5, 1, 5, 10, 50 and 100 lM), first
dissolved in DMF and then diluted with PSS. According
to this procedure, the concentration of DMF in the bath
solution never exceeded 0.3%, a concentration that did
not produce appreciable inotropic and chronotropic ef-
fects. During the construction of cumulative dose–re-
sponse curves, the next higher concentration of the
compounds was added only after the preparation
reached a steady state.
8. Hrycyna, C. A.; Gottesman, M. M. Drug Resistance
Update 1998, 1, 81.
9. Dei, S.; Gualtieri, F.; Scapecchi, S.; Teodori, E.; Garnier-
Suillerot, A. Recent Research Development in Medicinal
Chemistry; TRN/MC: Trivandrum, India, 2001; pp 17–64.
10. Bolhuis, H.; van Veen, E. W.; Poolman, B.; Driessen, A. J.
M.; Konings, W. N. FEMS Microbiol. Rev. 1997, 21, 55.
11. Zhu, B. T. Mol. Carcinogenesis 1999, 25, 1.
12. Roepe, P. D. Curr. Pharmaceut. Des. 2000, 6, 241.
13. Teodori, E.; Dei, S.; Scapecchi, S.; Gualtieri, F. Farmaco
2002, 57, 385.
14. Robert, J. Drug of the Future 1997, 22, 149.
15. Rosemberg, M. F.; Kamis, A. B.; Callaghan, R.; Higgins,
C. F.; Ford, R. C. J. Biol. Chem. 2003, 278, 8294.
16. Lu, L.; Leonessa, F.; Clarke, R.; Wainer, I. W. Mol.
Pharmacol. 2001, 59, 62.
17. Garrigos, M.; Mir, L. M.; Orlowski, S. Eur. J. Biochem.
1997, 244, 664.
18. Pommerenke, E. W.; Osswald, H.; Hahn, E. W.; Volm, M.
Cancer Lett. 1990, 55, 17.
19. Klopman, G.; Srivastava, S.; Kolossvary, I.; Epand, R. F.;
Ahmed, N.; Epand, R. M. Cancer Res. 1992, 52, 4121.
20. Ramu, A.; Ramu, N. Cancer Chemother. Pharmacol. 1994,
34, 423.
21. Pereira, E.; Teodori, E.; Dei, S.; Gualtieri, F.; Garnier-
Suillerot, A. Biochem. Pharmacol. 1995, 50, 451.
22. Litman, T.; Zeuthen, T.; Skovssgard, T.; Stein, W. D.
Biochim. Biophys. Acta-Mol. Bases of Disease 1997, 1361,
159.
23. Matsumoto, T.; Fujii, R.; Sugita, M.; Sumizawa, T.;
Sakai, S.; Takahashi, T.; Sueda, N.; Furukawa, T.;
Akiyama, S.; Nagata, Y. Anti-Cancer Drug Des. 1994, 9,
251.
24. Chiba, P.; Ecker, G.; Schmid, D.; Drach, J.; Tell, B.;
Goldemberg, S.; Gekeler, V. Mol. Pharmacol. 1996, 49,
1122.
25. Chiba, P.; Tell, B.; Jager, W.; Richter, E.; Hitzler, M.;
Ecker, G. Arch. Pharm. 1997, 330, 343.
26. Teodori, E.; Dei, S.; Quidu, P.; Budriesi, R.; Chiarini, A.;
Garnier-Suillerot, A.; Gualtieri, F.; Manetti, D.; Roma-
nelli, M. N.; Scapecchi, S. J. Med. Chem. 1999, 42, 1687.
27. Dei, S.; Teodori, E.; Garnier-Suillerot, A.; Gualtieri, F.;
Scapecchi, S.; Budriesi, R.; Chiarini, A. Bioorg. Med.
Chem. 2001, 9, 2673.
28. Teodori, E.; Dei, S.; Garnier-Suillerot, A.; Scapecchi, S.;
Budriesi, R. Med. Chem. Res. 2001, 10, 563.
29. Klopman, G.; Shi, L. M.; Ramu, A. Mol. Pharmacol.
1997, 52, 323.
30. Teodori, E.; Ettori, D.; Garnier-Suillerot, A.; Gualtieri,
F.; Manetti, D.; Romanelli, M. N.; Scapecchi, S. Bioorg.
Med. Chem. 1999, 7, 1873.
31. Roe, M.; Folkers, A.; Ashworth, P.; Brumwell, J.; Chima,
L.; Hunian, S.; Pretswell, I.; Dangerfield, W.; Ryder, H.;
Charlton, P. Bioorg. Med. Chem. Lett. 1999, 9, 595.
32. Ohno, N.; Nakano, M.; Endoh, J.; Miura, M.; Aizawa,
H.; Fukuzaki, A.; Seida, K. PTC Int. Appl., 1999, WO
9905095.
6.2.2.2. Guinea-pig aortic strips. The thoracic aorta
was removed and placed in a Tyrode solution of the fol-
lowing composition (mM): 118 NaCl, 4.75 KCl, 2.54
CaCl2, 1.20 MgSO4, 1.19 KH2PO4, 25 NaHCO3 and
11 glucose equilibrated with 95% O2–5% CO2 gas at
pH 7.4. The vessel was cleaned of extraneous connective
tissue. Two helicoidal strips (10 mm · 1 mm) were cut
from each aorta beginning from the end most proximal
to the heart. Vascular strips were then tied with surgical
thread (6–0) and suspended in a jacketed tissue bath
(15 mL) containing aerated pharmacological salt solu-
tion (PSS) at 35 ꢁC. Strips were secured at one end to
a force displacement (FT0.3, Grass) transducer for
monitoring changes in isometric contraction. Aortic
strips were subjected to a resting force of 1 g and washed
every 20 min with fresh PSS for 1 h after the equilibra-
tion period; guinea-pig aortic strips were contracted by
washing in PSS containing 80 mM KCl (equimolar sub-
stitution of K+ for Na+). After the contraction reached a
plateau (about 45 min), the compounds (0.1, 0.5, 1, 5,
10, 50 and 100 lM) were added cumulatively to the bath
allowing for any relaxation to obtain an equilibrated
level of force. Addition of the drug vehicle had no
appreciable effect on K+-induced contraction (DMF
for all compounds).
References and notes
33. Mattson, R. J.; Pham, K. M.; Leuck, D. J.; Cowen, K. A.
J. Org. Chem. 1990, 55, 2552.
1. Kane, S. E. Advances in Drug Research; Academic, 1996,
pp 181–252.

998
S. Dei et al. / Bioorg. Med. Chem. 13 (2005)985–998
34. Dodic, N.; Dumaitre, B.; Daugan, A.; Pianetti, P. J. Med.
Chem. 1995, 38, 2418.
35. Cannon, J. G.; Perez, J. A.; Bhatnagar, R. K.; Long, J. P.;
Sharabi, F. M. J. Med. Chem. 1982, 25, 1442.
41. Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G.
L.; Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.;
Polidori, G.; Spagna, R. J. Appl. Cryst. 1999, 32,
115.
36. Cannon, J. G.; Lee, T.; Goldman, H. D.; Costall, B.;
Naylor, R. J. J. Med. Chem. 1977, 20, 1111.
42. Sheldrick, G. M. SHELX97-Programs for Crystal Structure
Analysis (Release 97-2); Institut fur Anorganische Chemie
¨
¨
37. Raderer, M.; Scheithauer, W. Cancer 1993, 72, 3553.
38. Oliveira, P. R.; Wiectzycosky, F.; Basso, E. A.; Goncalves,
R. A. C.; Pontes, R. M. J. Mol. Struct. 2003, 657, 191.
39. Bellucci, C.; Gualtieri, F.; Scapecchi, S.; Teodori, E.;
Budriesi, R.; Chiarini, A. Il Farmaco 1989, 44, 1167.
40. Bruker, S. A. S. S. R. M., Bruker AXS Inc., Madison,
Wisconsin, USA SMARTand SAINTSoftware Reference
Manuals, Bruker AXS Inc., Madison, Wisconsin, USA.
1998.
der Universita¨t, Go¨ttingen (Germany). 1998.
43. Lozio, C. B.; Lozzio, B. B. Blood 1975, 45, 321.
44. Tsuruo, T.; Iida, H.; Kawataba, H.; Oh-hara, T.;
Hamada, H.; Utakoji, T. Jpn. J. Cancer Res. 1986, 77,
682.
45. Mankhetkorn, S.; Garnier-Suillerot, A. Eur. J. Pharmacol.
1998, 343, 313.
46. Tallarida, R. J.; Cowan, A.; Adler, M. W. Life Sci. 1979,
25, 637.