Fruitful adrenergic α2C-Agonism/α2A- antagonism combination to prevent and contrast morphine tolerance and dependence (1)

Article abstract of DOI:10.1021/jm100977d

The functional in vitro study of the enantiomers of imidazolines 4-7 highlighted the role played by the nature of the ortho phenyl substituent in determining the preferred α_2C-AR configuration. Indeed, the (S) enantiomers of 4-6 or (R) enantiom

Full text of DOI:10.1021/jm100977d

J. Med. Chem. 2010, 53, 7825–7835 7825
DOI: 10.1021/jm100977d
Fruitful Adrenergic r2C-Agonism/r2A-Antagonism Combination to Prevent and
Contrast Morphine Tolerance and Dependence1,†
Fabio Del Bello,‡ Laura Mattioli,‡ Francesca Ghelfi,‡ Mario Giannella,‡ Alessandro Piergentili,‡ Wilma Quaglia,‡
,^
Claudia Cardinaletti,‡ Marina Perfumi,‡ Russell J. Thomas,§ Ugo Zanelli,§ Carla Marchioro,
Michele Dal Cin, ,^ and Maria Pigini*,‡
‡
ꢀ
Scuola di Scienze del Farmaco e dei Prodotti della Salute, Universita di Camerino, via S. Agostino 1, 62032 Camerino, Italy,
§Siena Biotech SpA, Strada del Petriccio e Belriguardo 35, 53100 Siena, Italy, and GlaxoSmithKline Medicines Research Center,
via Fleming 4, 37135 Verona, Italy. ^ Current address: Aptuit Verona, via Fleming 4, 37135 Verona, Italy
Received July 30, 2010
The functional in vitro study of the enantiomers of imidazolines 4-7 highlighted the role played by the
nature of the ortho phenyl substituent in determining the preferred R2C-AR configuration. Indeed, the
(S) enantiomers of 4-6 or (R) enantiomer of 7 behave as eutomers and activate this subtype as full
agonists; the corresponding distomers are partial agonists. Because in clinical pain management with
opioids R2C-AR agonists, devoid of the R2A-AR-mediated side effects, may represent an improvement
over current therapies with clonidine like drugs, 4 and its enantiomers, showing R2C-agonism/R2A
-
antagonism, have been studied in vivo. The data suggest that partial R2C-activation is compatible with
effective enhancement of morphine analgesia and reduction both of morphine tolerance acquisition and
morphine dependence acquisition and expression. On the contrary, full R2C-activation appears
advantageous in reducing morphine tolerance expression. Interestingly, the biological profile displayed
by 4 (allyphenyline) and its eutomer (S)-(þ)-4 has been found to be very unusual.
Introduction
Chart 1
R2-Adrenoreceptors (R2-ARsa), belonging to the superfam-
ily of G-protein coupled receptors,2 consist of three subtypes
(R2A, R2B, and R2C) and are considered attractive therapeutic
targets for the treatment of a wide range of diseases.3 Because
of the lack of R2-AR subtype selective agonists, the attribution
of physiological functions to a given R2-AR subtype has proven
challenging in vivo. Although genetic manipulations present
some limitations, the use of mice carrying the D79N R2A-AR,
mice that are singly (R2A-AR-/-, R2B-AR-/-, R2C-AR-/-),
doubly (R2AC-AR-/-), or triply (R2ABC-AR-/-) deficient in a
particular subtype, or mice overexpressing the R2C-AR (R2C
-
ARþ/þOE) yielded important information about the subtype-
specific functions.4 These studies demonstrated that R2A-AR
subtype mediates hypotension, sedation, analgesia, hypother-
mia, antiepileptogenesis, and inhibition of monoamine release
and metabolism in the brain. The R2B subtype mediates hyper-
tension and initial peripheral hypertensive responses to R2-AR
agonists. Finally, the R2C-AR subtype appears to be involved in
many CNS processes, such as the startle reflex, stress response,
and locomotion as well as feedback inhibition of adrenal
catecholamine release. Moreover, it has been highlighted that
the R2C subtype can contribute to adrenergic opioid synergy and
spinal R2-agonist-mediated analgesia.5 The discovery and the
development of ligands endowed with preferential R2-AR re-
cognition and able to produce individual subtype activation
have been the focus of our research over the years.6 Our studies
demonstrated that conservative chemical modifications in the
aromatic moiety (Ar) of the R2-AR antagonist l (Chart 1, Table
1), which lacks subtype selectivity, caused an important mod-
ulation of receptor interaction and also affected the subtype
selectivity. Indeed, the introduction of a pendant phenyl sub-
stituent into the ortho position provided the interesting R2A-and
†
ꢁ
This article is dedicated to the memory of Dr. Herve Paris.
*To whom correspondence should be addressed. Phone: þ39 0737
402257. Fax: þ39 0737 637345. Email: maria.pigini@unicam.it.
a Abbreviations: R2-ARs, R2-adrenoreceptors; DEAD, diethyl azodicar-
boxylate; [(þ)-(R)-MTPA], (R)-(þ)-R-methoxy-R-(trifluoromethyl)phenyl-
acetic acid; ADME, absorption, distribution, metabolism, excretion; PAM-
PA, parallel artificial membrane permeability assay; CHO, chinese hamster
ovary; MDCK, Madin Darby Canine Kidney cell line.
R
2C-AR agonist biphenyline (2). Its enantiomer, (S)-(-)-2, was
endowed with enhanced long-lasting antinociceptive potency.7,8
r
2010 American Chemical Society
Published on Web 10/06/2010
pubs.acs.org/jmc

7826 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21
Del Bello et al.
Table 1. Affinity (pKia), Antagonist Potency (pKbb), Agonist Potency (pEC50b), and Intrinsic Activity (iab) on Human R2-AR Subtypes
R2A
R2B
R2C
compd
pKi
pEC50 (pKb)
ia
pKi
pEC50 (pKb)
ia
pKi
pEC50 (pKb)
ia
1
7.57 ( 0.09
7.32 ( 0.08
6.73 ( 0.04
7.04 ( 0.08
7.22 ( 0.03
7.09 ( 0.08
7.14 ( 0.06
7.24 ( 0.11
7.00 ( 0.08
7.28 ( 0.05
7.64 ( 0.20
7.35 ( 0.10
7.68 ( 0.12
7.34 ( 0.02
6.78 ( 0.13
7.35 ( 0.08
7.56 ( 0.02
7.40 ( 0.09
7.01 ( 0.07
(7.01 ( 0.10)
6.94 ( 0.06
4.54 ( 0.07
6.78 ( 0.13
6.30 ( 0.07
5.90 ( 0.03
6.23 ( 0.08
6.18 ( 0.08
6.13 ( 0.07
5.80 ( 0.06
6.47 ( 0.20
6.25 ( 0.12
6.40 ( 0.09
6.51 ( 0.13
6.25 ( 0.10
6.58 ( 0.08
6.38 ( 0.04
6.15 ( 0.07
6.35 ( 0.12
6.45 ( 0.01
6.50 ( 0.04
6.00 ( 0.09
(6.20 ( 0.18)
6.19 ( 0.01
4.00 ( 0.05
6.16 ( 0.10
(5.90 ( 0.10)c
(6.07 ( 0.09)
(5.93 ( 0.12)
NA
6.58 ( 0.12
6.70 ( 0.04
6.01 ( 0.05
6.52 ( 0.09
6.63 ( 0.08
6.59 ( 0.09
6.27 ( 0.08
7.07 ( 0.14
6.75 ( 0.11
7.15 ( 0.09
7.10 ( 0.16
6.90 ( 0.13
7.20 ( 0.10
6.88 ( 0.07
6.50 ( 0.12
6.90 ( 0.08
7.75 ( 0.08
7.50 ( 0.10
6.95 ( 0.11
(6.85 ( 0.15)
7.24 ( 0.01
6.95 ( 0.08
7.73 ( 0.14
7.00 ( 0.09
8.00 ( 0.04
6.13 ( 0.01
7.30 ( 0.09
6.73 ( 0.11
7.60 ( 0.14
7.40 ( 0.13
6.90 ( 0.12
7.45 ( 0.16
7.03 ( 0.28
7.00 ( 0.21
8.20 ( 0.18
7.33 ( 0.15
7.60 ( 0.11
5.60 ( 0.09
7.24 ( 0.14
(()-2 biphenyline
(R)-(þ)-2
(S)-(-)-2
(()-3m-nitrobiphenyline
(R)-(þ)-3
(S)-(-)-3
(()-4
(R)-(-)-4
(S)-(þ)-4
(()-5
(R)-(-)-5
(S)-(þ)-5
(()-6
(R)-(-)-6
(S)-(þ)-6
(()-7
0.70
0.40
0.80
0.50
0.40
0.65
0.80
0.70
0.90
0.60
0.80
0.45
0.90
0.50
0.90
0.90
0.50
0.90
0.85
0.70
0.90
1.15
1.00
0.65
0.60
7.13 ( 0.03
(6.80 ( 0.05)c
(7.11 ( 0.15)
(7.35 ( 0.16)
(7.40 ( 0.06)
(7.40 ( 0.09)
(7.80 ( 0.13)
(7.00 ( 0.09)
(6.90 ( 0.07)
(7.70 ( 0.11)
7.12 ( 0.08
NA
6.00 ( 0.09
5.50 ( 0.15
NA
0.65
0.70
5.50 ( 0.13
NA
NA
0.65
0.70
0.40
0.80
5.50 ( 0.09
7.20 ( 0.13
(6.20 ( 0.11)c
(6.40 ( 0.08)
(6.00 ( 0.12)
8.08 ( 0.12
6.00 ( 0.11
5.28 ( 0.17
(5.67 ( 0.12)
(5.50 ( 0.10)
5.93 ( 0.27
0.60
0.50
(R)-(þ)-7
(S)-(-)-7
clonidine
0.90
0.45
a pKi values were calculated from [3H]RX 821002 on membrane preparations from CHO cells expressing individually each human R2-AR subtype
(R2A, R2B, R2C). b pKb, pEC50 and intrinsic activity (ia) values were determined by applying the Cytosensor microphysiometry system to the same cell
models. Intrinsic activity of the tested compounds is expressed as the fraction of that of the full agonist (-)-noradrenaline taken as equal to 1. The data
are expressed as means ( SEM of three-six separate experiments. Compounds exhibiting ia of <0.3 were considered not active (NA). c Determined in
the present study.
The further insertion of polar functions into the meta position of
the pendant phenyl substituent induced a significant and selec-
tive R2C-AR activation (for example meta-nitrobiphenyline, 3)8
(Chart 1, Table 1). This biological profile modulation from
antagonist to agonist was attributed to the favorable interaction
of the ortho pendant substituent with the amino acid residues
comprising the highly conserved aromatic cluster (defined by
Trp6.48, Phe6.51, and Phe6.52) of the sixth transmembrane
domain of the R2-AR cavity.8,9 Subsequent studies also demon-
strated that the effects found for the ortho phenyl substituent
were also induced by its oximino-methyl bioisoster, whereas
cyclopentyl and cyclohexyl groups triggered indiscriminate
activation of all three R2-AR subtypes.10 The introduction in
the Ar group of ortho substituents characterized by moderate
steric hindrance, such as allyl or cyclopropyl groups, induced
modulation of the biological profile of the antagonist l, only at
results obtained from genetically engineered mice.5 In addition,
at doses producing significant morphine analgesia enhance-
ment, 4 and 5 were devoid of sedative side effects.6 To increase
our understanding of the interesting biological profile of these
compounds, an assessment of the physicochemical and in vitro
ADME properties of 4 was determined. Additionally, the
presence of a stereocenter in 4 and 5 prompted us to investigate
the stereochemical requirements for a possible improvement of
their biological profile. Therefore, both enantiomers of 4 and 5
were prepared and evaluated by binding and functional assays
on CHO cells expressing recombinant human R2-AR and R1-
AR subtypes. Because the lipophilic (π) and electronic (σ) effects
of the ortho phenyl substituent might contribute in determining
the preferred R2-AR subtype configurations, for a useful com-
parison, we also included in this investigation the preparation
and in vitro study of the enantiomeric pairs of 6and 78 (Chart1).
For the in vivo study, the modulation of morphine analgesia on
mouse tail-flick test was evaluated at the dose of 0.05 mg/kg for
both enantiomeric pairs of 4 and 5. In addition, the effects of 4
and its enantiomers on the acquisition and expression of
morphine tolerance and dependence were determined; also these
experiments were carried out at the dose of 0.05 mg/kg.
R
2C-AR subtype,6,10 similarly to what observed for 3. Indeed,
compounds 4 and 5 were potent and selective R2C-agonists,
while efficiently antagonizing the R2A-AR subtype (Chart 1,
Table 1). Also 3 and its enantiomers11 activated the R2C- and
simultaneously antagonized especially the R2A-AR subtype
(Table 1), but unlike 4 and 5,6 they proved able to interact with
good affinity with the R1a-AR subtype (pKi values between 7.0
and 7.8); their affinity for R1b- andR1d-AR subtypes is negligible
(data not published). On the basis of the assumption that in
clinical pain management, R2C selective agonists, devoid of the
sedative side effects associated with R2A-AR stimulation, might
represent alone or in combination with opioid analgesics an
improvement over current therapies with clonidine-like drugs, 4
and 5 coadministered with morphine were tested in the algesio-
metric paradigm mouse tail-flick test.6 From in vivo dose-
response studies, it emerged that the low experimental dose (0.05
mg/kg) of both compounds caused a significant increase in
morphine analgesia and this effect was comparable to that
obtained with 0.5 mg/kg dose of clonidine. The potentiation
of opioid analgesia displayed by 4 and 5 confirmed previous
Chemistry
The enantiomers (S)-(þ)- and (R)-(-)-4, (S)-(þ)- and
(R)-(-)-5, and (S)-(þ)- and (R)-(-)-6 were prepared according
to Scheme 1. (S)-(-)-and(R)-(þ)-2-(2-Allylphenoxy)propionic
acid methyl ester [(S)-(-)-8 and (R)-(þ)-8, respectively],
(S)-(-)- and (R)-(þ)-2-(2-cyclopropylphenoxy)propionic acid
methyl ester [(S)-(-)-9 and (R)-(þ)-9, respectively], and
(S)-(þ)- and (R)-(-)-2-(2-thiophen-3-yl-phenoxy)-propionic
acid methyl ester [(S)-(þ)-10 and (R)-(-)-10, respectively] were
obtained by Mitsunobu reaction from the appropriate
phenol12,13 and methyl (R)-(þ)- or (S)-(-)-lactate, respectively,
with inversion of configuration.14 The reaction of (S)-(-)-8,
(S)-(-)-9, (S)-(þ)-10, (R)-(þ)-8, (R)-(þ)-9, or (R)-(-)-10

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21 7827
with ethylenediamine in the presence of (CH3)3Al yielded the
corresponding imidazolines (S)-(þ)-4, (S)-(þ)-5, (S)-(þ)-6,
(R)-(-)-4,(R)-(-)-5, and(R)-(-)-6, whose enantiomeric purity,
determined by 1H NMR spectroscopy or HPLC, was <98%.
Therefore, the two (S)-(þ)-6 and (R)-(-)-6 enantiomers were
subsequently resolved by fractional crystallization of the
hydrogen dibenzoyl-L- and hydrogen dibenzoyl-D-tartrate salts,
respectively. The enantiomeric purity of amines (S)-(þ)-6
and (R)-(-)-6, determined by HPLC of their corresponding
diastereomeric ureidic derivatives 11 and 12, obtained by
reaction with (R)-(þ)-R-methylbenzyl isocyanate, was found to
1
confirmed by H NMR spectroscopy in the presence of the
be >98% (detection limit). The enantiomeric purity was further
chiral shift reagent (S)-(þ)-2,2,2-trifluoro-1-(9-anthryl)-etha-
1
nol. In fact, the H NMR spectrum of racemic compound
(()-6 showed a double doublet at δ 1.44 ppm for the methyl
protons, whereas only one doublet was observed for (R)-(-)-6
and (S)-(þ)-6 at δ 1.44 and δ 1.46 ppm, respectively. Attempts
to obtain the pure enantiomers (S)-(þ)-4, (S)-(þ)-5, (R)-(-)-4,
and (R)-(-)-5 by fractional crystallization were unsuccessful.
Therefore, resolution of the racemic imidazolines (()-4and (()-
56 was performed by HPLC using Chiralcel OD-H (25 cm ꢀ
0.46 cm) as the chiral stationary phase and n-hexane/2-propanol
90/10 v/v or n-hexane/2-propanol 85/15 v/v as the mobile phase
for (()-4 and (()-5, respectively (Figure 1). The enantiomeric
purity of amines (þ)-4, (-)-4, (þ)-5, and (-)-5, determined by
Scheme 1a
1
HPLC and further confirmed by H NMR spectroscopy in
comparison with the spectra of racemic compounds (()-4 or
(()-5 and on addition of the chiral shift reagent (S)-(þ)-2,2,2-
trifluoro-1-(9-anthryl)-ethanol, was found to be >98%
(detection limit) for all enantiomers. In fact, the spectrum of
(()-4 showed a double doublet at δ 1.34 ppm for the methyl
protons, whereas only one doublet was observed for (þ)-4 and
(-)-4 at δ 1.32 and δ 1.36 ppm, respectively. Similarly, for the
enantiomers (þ)-5 and (-)-5, one doublet was observed at δ
1.44 and δ 1.40 ppm, respectively, whereas in the spectrum
of (()-5, the double doublet for the methyl protons was at
δ 1.42 ppm.
The absolute configuration of the stereocenter in the bridge
of enantiomers (þ)-4 and (þ)-5 was determined by comparing
the sign of their optical rotations with those of the enantio-
mers obtained by stereospecific synthesis. Consequently, an S
absolute configuration was assigned to the dextrorotatory
enantiomers (þ)-4 and (þ)-5.
The synthesis of the two enantiomers (S)-(-)-7 and
(R)-(þ)-7 is reported in Scheme 2. The imidazolines (S)-(þ)-
13 and (R)-(-)-13 were obtained by catalytic hydrogenation
over Pd/C of the nitro-derivatives (S)-(-)-3 and (R)-(þ)-3.11
Diazotization and subsequent hydrolysis of the correspond-
ing arenediazonium salts yielded the imidazolines (S)-(-)-7
and (R)-(þ)-7. The enantiomeric purity of (S)-(-)-7 and
(R)-(þ)-7, determined by 1H NMR spectroscopy in compar-
ison with the spectrum of racemic compound (()-7 and on
addition of the chiral shift reagent (R)-(þ)-R-methoxy-
R-(trifluoromethyl)phenylacetic acid [(þ)-(R)-MTPA], was
found to be >98% (detection limit): the 1H NMR spectrum
of the racemic compound (()-7 showed a double doublet at δ
1.43 ppm for the methyl protons, whereas only one doublet
was observed for (S)-(-)-7 and (R)-(þ)-7 at δ 1.42 and 1.44
ppm, respectively.
a Reagents: (a) DEAD, PPh3, dry THF; (b) (CH3)3Al, NH2CH2-
CH2NH2, dry toluene, Δ; (c) R-(þ)-R-methylbenzyl isocyanate.
Figure 1. HPLC spectra of (()-4, (()-5, and their corresponding enantiomers.

7828 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21
Del Bello et al.
Results and Discussion
pEC50 = 7.60 and 7.45, respectively), while the corresponding
distomers (R)-(-)-4 and (R)-(-)-5 behaved as partial agonists
(Figure 2). The R2B-AR subtype was moderately activated by
(S)-(þ)-4 and (S)-(þ)-5 (ia = 0.65; pEC50 = 6.00 and 5.50,
respectively); (R)-(-)-4 and (R)-(-)-5 were found to be
inactive. Less affected by chirality, the significant R2A-AR
antagonism of 4 and 5 was maintained by (R)-(-)-4 and
(R)-(-)-5 and moderately enhanced by (S)-(þ)-4 and
(S)-(þ)-5 (Figure 3). We previously reported11 the surprising
reversal of enantioselectivity observed for the structurally
similar compounds 2 and 3, whose (S) and (R) forms,
respectively, were the preferred R2C-AR configurations. The
higher R2C-potencies displayed by (S)-(þ)-4 and (S)-(þ)-5
confirmed our previous results according to which the inter-
actions between the aromatic cluster of the R2-AR binding
cavity and the ortho pendant moiety depended on the phys-
icochemical characteristics of this latter.8 Indeed, the pendant
phenyl, allyl, and cyclopropyl substituents of (S)-(-)-2,
(S)-(þ)-4, and (S)-(þ)-5, respectively, endowed with the same
þπ and -σ character and spatial orientation, and presum-
ably, recognizing common amino acidic residues, gave more
efficient lipophilic interactions. On the contrary, in the case of
3, the polar interaction determined by the meta nitro group
(-π and þσ), was more favorably obtained by the inverse
configuration. Further support for this hypothesized pathway
has been provided by the examination of the functional
behavior of the enantiomeric pairs of 6 and 7, prepared in
this study (Table 1). The R2C-AR preferred configurations
The affinity values (pKi), antagonist properties (pKb),
agonist potencies, and intrinsic activities (pEC50 and ia,
respectively) of the novel compounds on R2-AR subtypes
are reported in Table 1. The biological profiles of the lead 1,
clonidine and imidazolines 2 and 3, and their enantiomers
were also included for useful comparison. Analogously to
what was observed for the enantiomers of 2 and 3,11 both
enantiomers of 4 and 5 displayed comparable binding affinity
for a given R2-AR subtype. Moreover, analogously to their
corresponding racemic compounds,6 they showed moderate
R1a-AR binding affinity (pKi e 6.5) and negligible intrinsic
activity (ia <0.3). The functional results, demonstrating the
role played by chirality, highlighted that for both pairs of the
enantiomers, the (S) form behaved as eutomer and produced
better R2C-AR agonist activity; indeed, (S)-(þ)-4 and (S)-(þ)-
5 fully activated this subtype with high potency (ia = 0.90;
Scheme 2a
[(S)-(þ)-6 (pEC50 = 8.2; ia = 0.90) or (R)-(þ)-7 (pEC50
=
7.60; ia = 1.0)] were expected, because, in agreement with the
peculiar nature of the thiophen-3-yl or meta-hydroxy phenyl
pendant moieties, characterized by þπ and -σ or -π and þσ
conditions, respectively. On the other hand, it has already
been reported that modifications on ligands belonging to the
same series can produce changes in their binding mode.15
Moreover, the enantiomers of 6 and 7 displayed R2A- and R2B
-
AR profiles similar to those of the corresponding enantiomers
of the parents 2 or 3, respectively. Anyway, the ability of the
stereocenter in the bridge in affecting significantly the R2-AR
agonism or negligibly the R2-AR antagonism is observed for
all compounds of this series.
The data from the in vivo studies showed that, surprisingly,
the pretreatment with (S)-(þ)-4 or (R)-(-)-4 enhanced to the
same extent morphine analgesia with an effect comparable to
that of the racemic compound 4.6 This effect was rapid (30
min), potent, and long lasting (duration 240 min). Analogous
behavior was observed for the (S)-(þ)-5 and (R)-(-)-5 pair
(Figure 4A,B). The R2C-AR potencies shown by the distomers
a Reagents: (a) H2/Pd, (b) H2SO4, H2O, NaNO2; H2O, Cu(NO3)2 3H2O,
3
Cu2O.
Figure 2. Agonist potencies of clonidine, (()-4, (()-5, and their corresponding enantiomers determined by Cytosensor microphysiometer in
CHO cells stably expressing the human R2C-AR subtype.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21 7829
Figure 3. Antagonist potencies of (()-4, (()-5, and their corresponding enantiomers determined by Cytosensor microphysiometer in CHO
cells stably expressing the human R2A-AR subtype.
Figure 5. Effect of (S)-(þ)-4 and (R)-(-)-4 on morphine-induced an-
algesia after yohimbine pretreatment in the tail-flick test. The reaction
latencies were expressed as a percent of the maximum possible effect (%
MPE). Each point represents the media ( SEM of 7-8 animals. Signi-
ficant differences: *p < 0.05, **p < 0.01 compared to morphine group.
Figure 4. Effect of a single administration of (()-4 (A) or (()-5
(B) and their corresponding enantiomers on morphine-induced
analgesia in the tail-flick test. The reaction latencies were expressed
as a percent of the maximum possible effect (%MPE). Each point
represents the media ( SEM of eight animals. Significant differ-
ences: *p < 0.05, **p < 0.01 compared to morphine group.
(R)-(-)-4 and (R)-(-)-5 (ia = 0.50; pEC50 = 6.73 and 6.90,
respectively), comparable to that of clonidine (pEC50 R2C
=
7.24; ia = 0.60), may explain their significant morphine analge-
sia enhancement. As with the corresponding racemic com-
pounds, the novel derivatives had no analgesic effect when
given alone (data not shown). The fact that pretreatment with
the R2-AR antagonist yohimbine blocked the effects of both
(S)-(þ)-4 and (R)-(-)-4 unambiguously demonstrated the in-
volvement of the R2C-AR activation in the positive morphine
analgesia modulation of both enantiomers (Figure 5). Surpris-
ingly, a full R2C-AR activation appeared advantageous to
contrast the expression of morphine tolerance. Indeed, while
Figure 6. Effect of a single administration of (()-4, (S)-(þ)-4, and
(R)-(-)-4 on expression of tolerance to morphine-induced analge-
sia. Antinociceptive effect (% MPE) on control (n = 10) or
morphine (n = 36) treated mice was evaluated on day 1 and day
5. Significant differences: **p < 0.01, compared to control group;
þþp < 0.01, compared to related-morphine group on day 1; ꢀꢀp <
0.01, compared to morphine group on day 5.

7830 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21
Del Bello et al.
Figure 8. Effect of (()-4, (S)-(þ)-4, and (R)-(-)-4 on pentobarbital
sodium-induced sleep time. Mice were treated with 50 mg/kg
pentobarbital sodium (Pent; n = 40) 15 min after the administration
of 0.05 mg/kg (()-4 and its enantiomers. The sleeping time was
expressed in seconds. No significant differences were observed.
were devoid of sedative effects (Figure 8). Moreover, unlike
clonidine,16,17 the repeated treatment with 4 or its enantiomers,
before each morphine administration, decreased acquisition of
tolerance and dependence. Significant enhancement of the
morphine analgesia both on day 1 and on day 5 was observed
(Figure 9); in addition, the naloxone-induced behavioral with-
drawal symptoms proved reduced. Total signs were decreased
by about 40%, while jumping was almost completely blocked by
4; in this case, the observed reduction was 80% (Figure 10A,B).
The positive effects on dependence are important because
withdrawal symptoms are believed to have a major role in
relapse to drug-taking behavior after detoxification.18
Conclusion
In summary, the study of 4 and its enantiomers, confirming
the involvement of the R2C-AR in morphine analgesia,5,6 sug-
gests that the partial activation of this subtype proved to be
compatible with (i) sharp enhancement of morphine analgesia
and (ii) effective reduction of acquisition of morphine tolerance
as well as acquisition and expression of morphine dependence.
The distomer (R)-(-)-4 displayed these effects similarly to the
full agonists 4and (S)-(þ)-4. On the other hand, the full R2C-AR
activation, as that triggered by 4 and (S)-(þ)-4, appeared
advantageous in reducing the expression of morphine tolerance.
Therefore, 4 and (S)-(þ)-4, enhancing morphine analgesia and
decreasing both acquisition and expression of morphine toler-
ance and dependence, displayed an unusual biological profile
and could become novel promising therapeutic tools in the
management of chronic pain and opioid addiction. In addition,
the assessment of the physicochemical and in vitro ADME
properties of 4 showed it to be highly soluble in water at
physiological pH and have a high level of metabolic stability
in both mouse and human microsomal preparations. It also
exhibits high passive diffusion in an artificial phospholipid
membrane (PAMPA)19 and an MDCK cell line, with no issues
of P-glycoprotein (PGP)-mediated efflux. Although an in vivo
PK and brain-plasma distribution was not carried out, the in
vitro profile of this compound bodes well for significant brain
exposure and is in line with the observed central effects in the
pharmacology models.
Figure 7. Effect of a single administration of (()-4, (S)-(þ)-4, and
(R)-(-)-4 on the expression of morphine dependence. Naloxone-
precipitated withdrawal symptoms, both in control (n = 10) and
morphine (n = 36) treated mice, are given as a measure of the
frequency of jumping (A) and somatic signs (B) expressed as
summary of rearing, forepaw tremors, and teeth chatter. Significant
differences: **p<0.01, compared to vehicle; þþp<0.01, compared
to morphine group.
(R)-(-)-4 was ineffective, the acute administration of a 0.05 mg/
kg dose of 4 or its enantiomer (S)-(þ)-4, 15 min before the last
morphine treatment on the test day, reduced the analgesic effect
only of 22% compared to 74% observed for alone morphine-
treated group (Figure 6). The acute administration of 4 or its
enantiomers, 15 min before naloxone injection, significantly
decreased the expression of naloxone-precipitated withdrawal
syndrome: the frequencies of jumping and other somatic signs,
such as rearing, paw tremor, and teeth chatter, were halved
(Figure 7A,B). Clonidine also appears to attenuate the expres-
sion of morphine tolerance and withdrawal symptoms, but at
higher doses (2 and 5 mg/kg for tolerance and dependence
expression, respectively) and with associated sedation and
hypotension side effects.16 Instead, 4, (S)-(þ)-4, and (R)-(-)-4
Finally, considering that (i) clonidine is ineffective to prevent
the development of morphine tolerance,16,17 but is clinically

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21 7831
Figure 9. Effect of repeated administration of (()-4, (S)-(þ)-4, and (R)-(-)-4 on the acquisition of tolerance to morphine-induced analgesia.
Antinociceptive effect (% MPE) on control (n = 30) or morphine (n = 41) treated mice was evaluated on day 1 and day 5. Significant
differences: **p < 0.01, compared to control group; þþp < 0.01, compared to related-morphine group on day 1; ꢀꢀp < 0.01, compared to
morphine group on day 5.
effective at reducing the intensity of withdrawal symptoms,
increasing treatment duration and completion,20,21 (ii) mor-
Enantiomeric Separation of Racemic Compounds (()-4 and (()-5.
The enantiomers of the title compounds6 were separated by chiral
HPLC by using a chiral column Chiralcel OD-H (25 cm ꢀ 0.46
cm, 5 μm particle size), mobile phase: n-hexane/2-propanol 90/10
v/v for (()-4 or n-hexane/2-propanol 85/15 v/v for (()-5; flow
rate 1 mL/min; detection was monitored at a wavelength of
220 nM: run time 15.0 min for compound (()-4 and 18.0 min
for compound (()-5 (Figure 1). Retention times: (þ)-4 6.51 min;
(-)-4 10.52 min; (þ)-5 7.54 min; (-)-5 14.00 min; ee g 99%.
The free bases were transformed into the hydrogen oxalate
phine withdrawal-induced jumping was attenuated in R2A
-
AR KO mice,17 and (iii) the chronic application of yohim-
bine attenuated naloxone-induced opioid withdrawal in
mice22 and partially prevented opioid withdrawal signs in
methadone maintained patients,23 we might attribute the
unique profiles of 4 (named allyphenyline) and (S)-(þ)-4 to
the synergistic combination of their potent R2C-AR agonism/
R2A-AR antagonism.
salts, which were recrystallized from EtOH:
1
(þ)-4: [R]20 = þ2.56ꢀ (c 1, MeOH); mp 154-155 ꢀC. H
D
NMR (DMSO) δ 1.53 (d, 3, CHCH3), 3.42 (m, 2, CH2CH), 3.88
(s, 4, NCH2CH2N), 5.06 (dd, 2, CHdCH2), 5.40 (q, 1, OCH),
5.95 (m, 1, CHdCH2), 6.92-7.26 (m, 4, ArH), 7.81 (br s, 1, NH,
Experimental Protocols
Chemistry. Melting points were taken in glass capillary tubes
on a Buchi SMP-20 apparatus and are uncorrected. IR and
NMR spectra were recorded on Perkin-Elmer 297 and Varian
EM-390 instruments, respectively. Chemical shifts are reported in
parts per million (ppm) relative to tetramethylsilane (TMS), and
spin multiplicities are given as s (singlet), d (doublet), t (triplet), q
(quartet), or m (multiplet). IR spectral data (not shown because of
the lack of unusual features) were obtained for all compounds
reported and are consistent with the assigned structures. The
microanalyses were performed by the Microanalytical Laboratory
of the department of Chemical Sciences of the University of
Camerino. The elemental composition of the compounds agreed
to within (0.4% of the calculated value. Optical rotation was
measured at a 1 g/100 mL concentration (c = 1) with a Perkin-
Elmer 241 polarimeter (accuracy (0.002ꢀ). Chromatographic
separations were performed on silica gel columns (Kieselgel 40,
0.040-0.063 mm, Merck) by flash chromatography. Compounds
were named following IUPAC rules as applied by ChemBioDraw
Ultra (version 11.0) software for systematically naming organic
chemicals. HPLC analyses of (S)-(þ)-6 and (R)-(-)-6 were re-
corded on an Hewelett Packard 1090 I series chromatograph on a
Hypersil 200 mm ꢀ 2.1 mm stainless steel column (packed with 5
μm particles, Hewlett-Packard). The mobile phase used was
hexane/AcOEt (60/40), and the flow rate was set at 1.0 mL/min.
The purity of the new compounds was determined by combustion
analysis and was g95%.
exchangeable with D2O). Anal. (C14H18N2O H2C2O4) C, H, N.
3
(-)-4: [R]20 = -2.67ꢀ (c 1, MeOH); mp 154-155 ꢀC. H
1
D
NMR (DMSO) δ 1.53 (d, 3, CHCH3), 3.42 (m, 2, CH2CH), 3.88
(s, 4, NCH2CH2N), 5.06 (dd, 2, CHdCH2), 5.40 (q, 1, OCH),
5.95 (m, 1, CH=CH2), 6.92-7.26 (m, 4, ArH), 7.81 (br s, 1, NH,
exchangeable with D2O). Anal. (C14H18N2O H2C2O4) C, H, N.
3
1
(þ)-5: [R]20 = þ0.68ꢀ (c 1, MeOH); mp 141-142 ꢀC. H
D
NMR (DMSO) δ 0.51-0.98 (m, 4, CH2CH2), 2.58 (d, 3,
CHCH3), 2.24 (m, 1, CH), 3.91 (s, 4, NCH2CH2N), 5.37 (q, 1,
OCH), 6.83-7.18 (m, 4, ArH), 9.51 (br s, 1, NH, exchangeable
with D2O). Anal. (C14H18N2O H2C2O4) C, H, N.
3
(-)-5: [R]20 = -0.63ꢀ (c 1, MeOH); mp 141-142 ꢀC. H
1
D
NMR (DMSO) δ 0.51-0.98 (m, 4, CH2CH2), 2.58 (d, 3,
CHCH3), 2.24 (m, 1, CH), 3.91 (s, 4, NCH2CH2N), 5.37 (q, 1,
OCH), 6.83-7.18 (m, 4, ArH), 9.51 (br s, 1, NH, exchangeable
with D2O). Anal. (C14H18N2O H2C2O4) C, H, N.
3
(S)-(-)-Methyl 2-(2-allylphenoxy)propanoate [(S)-(-)-8]. Asolu-
tion of DEAD (3.01 g, 14.9 mmol) in dry THF (10 mL) was added
dropwise to a mixture of methyl (R)-(þ)-lactate (1.36 g, 13.1 mmol),
2-allylphenol (1.70 g, 12.7 mmol), and triphenylphosphine (3.34 g,
12.7 mmol) in THF (20 mL). The mixture was stirred at room
temperature under nitrogen atmosphere overnight. The solvent was
evaporated, and diethyl ether/hexane solution was added. The
precipitate triphenylphosphine oxide was filtered off, and the re-
moval of the dried solvent gave a residue that was purified by flash
chromatography eluting with cyclohexane/AcOEt (95:5) (2.37 g,

7832 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21
Del Bello et al.
The mixture was heated to 65 ꢀC for 3 h, cooled to 0 ꢀC, and
quenched cautiously with MeOH (0.8 mL), followed by H2O
(0.2 mL). After addition of CHCl3 (5 mL) and filtration, the organic
layer was extracted with 2 N HCl. The aqueous layer, made basic
with 10% NaOH, was extracted with CHCl3. Removal of dried
solvent gave the free base (S)-(þ)-4, which was purified by flash
chromatography using cyclohexane/AcOEt/MeOH/33% NH4OH
(6:3:1:0.1) (63% yield). The enantiomeric purity, determined by
HPLC by using a chiral column Chiralcel OD-H (25 cm ꢀ 0.46
cm), eluent n-hexane/2-propanol 90/10 v/v, was about 85%.
1
[R]20 = þ1.75ꢀ (c 1, MeOH). H NMR (DMSO) δ 1.53 (d, 3,
D
CHCH3), 3.42 (m, 2, CH2CH), 3.88 (s, 4, NCH2CH2N), 5.06 (dd, 2,
CHdCH2), 5.40 (q, 1, OCH), 5.95 (m, 1, CHdCH2), 6.92-7.26 (m,
4, ArH), 7.81 (br s, 1, NH, exchangeable with D2O).
(R)-(-)-2-(1-(2-Allylphenoxy)ethyl)-4,5-dihydro-1H-imidazole
[(R)-(-)-4]. This was prepared as described for (S)-(þ)-4 starting
from (R)-(þ)-8 (59% yield). The enantiomeric purity, determined
by HPLC by using a chiral column Chiralcel OD-H (25 cm ꢀ 0.46
cm), eluent n-hexane/2-propanol 90/10 v/v, was about 85%.
[R]20D = -1.64ꢀ (c 1, MeOH). 1H NMR (DMSO) δ 1.53 (d, 3,
CHCH3), 3.42 (m, 2, CH2CH), 3.88 (s, 4, NCH2CH2N), 5.06 (dd,
2, CHdCH2), 5.40(q, 1, OCH), 5.95(m, 1, CHdCH2), 6.92-7.26
(m, 4, ArH), 7.81 (br s, 1, NH, exchangeable with D2O).
N-((R)-1-Phenylethyl)-2-((S)-1-(2-(thiophen-3-yl)phenoxy)ethyl)-
4,5-dihydro-1H-imidazole-1-carboxamide (11). (R)-(þ)-R-Methyl-
benzyl isocyanate (0.14 g, 0.95 mmol) was added to a solution
of (S)-(þ)-6 (0.26 g, 0.95 mmol) in dry CH2Cl2. After 4 h at
room temperature, the solvent was removed and the residue
was purified by flash chromatography using cyclohexane/
AcOEt (5:5). 1H NMR (CDCl3) δ 1.40 (d, 3, OCHCH3),
1.60 (d, 3, CH3CHN), 3.82 (m, 4, NCH2CH2N), 5.05 (m, 1,
CHN), 5.50 (d, 1, NH, exchangeable with D2O), 5.85 (q, 1,
OCH), 6.90-8.75 (m, 12, ArH).
N-((R)-1-Phenylethyl)-2-((R)-1-(2-(thiophen-3-yl)phenoxy)ethyl)-
4,5-dihydro-1H-imidazole-1-carboxamide (12). This was prepared as
1
described for 11 starting from (R)-(-)-6 (53% yield). H NMR
(CDCl3) δ 1.39 (d, 3, OCHCH3), 1.70 (d, 3, CH3CHN), 3.82 (m, 4,
NCH2CH2N), 5.01 (m, 1, CHN), 5.50 (d, 1, NH, exchangeable with
D2O), 5.80 (q, 1, OCH), 6.90-8.75 (m, 12, ArH).
(S)-(þ)-20-(1-(4,5-Dihydro-1H-imidazol-2-yl)ethoxy)biphenyl-
3-amine Hydrogen Oxalate Salt [(S)-(þ)-13]. A solution of
(S)-(-)-311 (1.0 g, 3.2mmol) in MeOH (20 mL) was hydrogenated
for 8 h at room temperature under pressure (40 psi) using Pd/C as
catalyst. After catalyst removal, the evaporation of the solvent
gave the free base which was purified by flash chromatography.
Eluting with cyclohexane/AcOEt/MeOH/33% NH4OH (5:4:1:0.1)
afforded the free base (0.27 g, 30% yield), which was transformed
Figure 10. Effect of repeated administration of (()-4, (S)-(þ)-4,
and (R)-(-)-4 on the acquisition of morphine dependence. Nalox-
one-precipitated withdrawal symptoms, both in control (n = 30)
and morphine (n = 41) treated mice, are given as a measure of the
frequency of jumping (A) and somatic signs (B) expressed as
summary of rearing, forepaw tremors, and teeth chatter. Significant
differences: **p < 0.01, compared to vehicle; þþp < 0.01, com-
pared to morphine group.
into the hydrogen oxalate salt; recrystallized from EtOH: [R]20
=
D
þ6.76ꢀ (c 1, MeOH); mp 158.8-160.5 ꢀC. 1H NMR (DMSO) δ
1.50 (d, 3, CHCH3), 3.82 (s, 4, NCH2CH2N), 5.10 (q, 1, OCH),
6.50-7.40 (m, 8, ArH), 8.92 (br s, 2, NH2 exchangeable with
D2O)10.5 (br s, 1, NH, exchangeable with D2O).
(R)-(-)-20-(1-(4,5-Dihydro-1H-imidazol-2-yl)ethoxy)biphenyl-
3-amine Hydrogen Oxalate Salt [(R)-(-)-13]. This was prepared
as described for (S)-(þ)-13 starting from (R)-(þ)-311 (39% yield).
The free base was transformed into the hydrogen oxalate
85%yield):[R]20D =-7.18ꢀ (c1, CHCl3).1HNMR(CDCl3) δ1.61
(d, 3, CHCH3), 3.44 (m, 2, CH2CH), 3.76 (s, 3, OCH3) 4.78 (q, 1,
OCH), 5.08 (dd, 2, CHdCH2), 6.01 (m, 1, CHdCH2), 6.71-7.19
(m, 4, ArH).
salt, which was recrystallized from EtOH: [R]20 = -6.25ꢀ
D
1
(R)-(þ)-Methyl 2-(2-allylphenoxy)propanoate [(R)-(þ)-8]. This
(c 1, MeOH); mp 155.6-157.4 ꢀC. H NMR (DMSO) δ 1.50
was prepared as described for (S)-(-)-8 starting from methyl
D
(d, 3, CHCH3), 3.82 (s, 4, NCH2CH2N), 5.10 (q, 1, OCH),
6.50-7.40 (m, 8, ArH), 8.92 (br s, 2, NH2 exchangeable with
D2O), 10.5 (br s, 1, NH, exchangeable with D2O).
1
(S)-(-)-lactate (81% yield): [R]20 = þ7.33ꢀ (c 1, CHCl3). H
NMR (CDCl3) δ 1.61 (d, 3, CHCH3), 3.44 (m, 2, CH2CH), 3.76
(s, 3, OCH3) 4.78 (q, 1, OCH), 5.08 (dd, 2, CHdCH2), 6.01 (m, 1,
CHdCH2), 6.71-7.19 (m, 4, ArH).
(S)-(-)-20-[1-(4,5-Dihydro-1H-imidazol-2-yl)-ethoxy]-biphenyl-
3-ol Hydrochloride Salt [(S)-(-)-7]. Crushed ice was added to a
solution of (S)-(þ)-13 (0.3 g, 1.07 mmol) in 26% sulfuric acid
(24 mL). The stirred mixture was maintained at 0-5 ꢀC, and a
solution of sodium nitrite (0.07 g, 1.01 mmol) in water (0.86 mL)
was slowly added. After the addition, the mixture was left at the
same temperature for 15 min. Urea was added to remove the
excessofsodiumnitrite. Asolutionofcopper(II) nitrate trihydrate
(2.54 g, 10.5 mmol) in water (20.4 mL) was added to the freshly
(S)-(þ)-2-(1-(2-Allylphenoxy)ethyl)-4,5-dihydro-1H-imidazole
[(S)-(þ)-4]. A solution of ethylenediamine (0.42 mL, 6.28 mmol)
in dry toluene (4 mL) was added dropwise to a mechanically
stirred solution of 2 M trimethylaluminum (3.2 mL, 6.28 mmol)
in dry toluene (4 mL) at 0 ꢀC under nitrogen atmosphere. After
1 h, the solution was cooled to 0 ꢀC, and a solution of (S)-(-)-8
(0.69 g, 3.14 mmol) in dry toluene (8 mL) was added dropwise.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21 7833
prepared solution of diazonium salt maintained at 0 ꢀC, then
copper(I) oxide (0.12 g, 0.84 mmol) was added to the vigorously
stirred mixture, and gas evolution being observed. The reaction
mixture was left at room temperature for 30 min, then basified
with NaHCO3 and extracted with dichloromethane. Removal of
dried solvent gave the free base (S)-(-)-7, which was purified by
flash chromatography using cyclohexane/AcOEt/MeOH/33%
NH4OH (6:4:1.5:0.1): (0.1 g, 33% yield). The free base was
subtypes or R1A-AR were seeded into 12 mm capsule cups at a
density of 3 ꢀ 105 cells/cup and incubated at 37 ꢀC under 5% CO2
atmosphere for 24 h. The capsule cups were loaded into the sensor
chambers of the instrument and perfused with a running medium
(bicarbonate-free DMEM containing 0.584 g/L glutamine and 2.59
g/L NaCl) at a flow rate of 100 μL/min. Agonists were diluted into
running medium and injected through a second fluid path. Valves
directed the flow from either fluid path to the sensor chamber. For
each 90 s pump cycle, the pump was on for 60 s and was then
switched off for the remaining 30 s and the pH value was recorded
for 20 s (from second 68-88). Cells were exposed to agonists for
4 min, and consecutive agonist exposures were separated by a
30 min washing period. This stimulation protocol was validated in
preliminary experiments with four known agonists, (-)-noradrena-
line, clonidine, UK 14304, and BHT 920 for R2-AR CHO cells and
(-)-noradrenaline and cirazoline for R1A-AR CHO cells. The rate
of acidification of the chamber was calculated by the Cytosoft
program (Molecular Devices). Changes in the rate of acidification
were calculated as the difference between the maximum effect after
agonist addition, and the average of three measurements taken
prior to agonist addition. For antagonist studies, a control con-
centration-response curve was first obtained with clonidine and
the cells were then exposed to antagonist for at least 30 min prior to
construction of another clonidine concentration-effect curve in the
presence of the antagonist. Therefore, each chamber acted as its
own control. Data were analyzed as the ability of the antagonist to
shift the agonist concentration-effect curve and defined as Kb.
Statistical Analysis. The values of Ki and EC50 and the extent
of maximal response (Emax) were calculated from the computer
analysis of binding inhibition data and dose-response curves
using the program GraphPad Prism (GraphPad Software, San
Diego, CA, USA). The Kb values were calculated as M/concen-
tration ratio-1, where concentration ratio is the EC50 obtained in
the presence of the antagonist divided by that obtained in the
absence of the antagonist.26 Data were expressed as pKb
[-log10(Kb)] and reported values are means ( SEM of three-six
separate experiments.
In Vivo Assays. Animals. Male CD-1 mice (Harlan SRC, Milan,
Italy) weighing 25-35 g were used. The mice were kept in a
dedicated room, with a 12:12 h light/dark cycle (lights on at
09:00), a temperature of 20 to 22 ꢀC, and a humidity of 45-55%.
They were provided with free access to tap water and food pellets
(4RF18, Mucedola, Settimo Milanese, Italy). Each mouse was used
in only one experimental session. All of the procedures were
conducted in adherence to the European Community Directive
for the Care and Use of Laboratory Animals. Ethical guidelines for
the investigation of experimental pain in conscious animals were
followed. The procedures were carried out according to the EEC
ethical regulations for animal research (ECC Council 86/609; D.
Lgs. 27/01/1992, no. 116), and tests have been approved by the local
ethical committee.
Chemicals and Drugs. Morphine hydrochloride was pur-
chased from Salars Spa (Como, Italy) and yohimbine hydro-
chloride and naloxone hydrochloride from Sigma (St. Louis,
MO). Drugs were dissolved in distilled water immediately before
use, with the exception of morphine hydrochloride, which was
dissolved in saline. Compounds 4, 5, and their enantiomers (0.05
mg/kg), yohimbine hydrochloride (1.250 mg/kg) and naloxone
(5 mg/kg), were injected intraperitoneally (ip), while morphine
hydrochloride was administered subcutaneously (sc) at the dose
of 5.0 or 10 mg/kg according to the tests paradigm.
transformed into the hydrochloride salt (oil). [R]20 = -2.30ꢀ
D
(c 1, MeOH). 1H NMR (DMSO) δ 1.40 (d, 3, CHCH3), 3.80 (s, 4,
NCH2CH2N), 4.95 (q, 1, OCH), 6.70-7.38 (m, 8, ArH), 7.68 (brs,
1, OH, exchangeable with D2O), 10.17 (br s, 1, NH, exchangeable
with D2O). Anal. (C17H18N2O2 HCl) C, H, N.
3
(R)-(þ)-20-[1-(4,5-Dihydro-1H-imidazol-2-yl)-ethoxy]-biphenyl-
3-ol Hydrochloride Salt [(R)-(þ)-7]. This was prepared as des-
cribed for (S)-(-)-7 starting from (R)-(-)-13 (39% yield). The free
base was transformed into the hydrochloride salt (oil). [R]20
=
D
þ2.15ꢀ (c 1, MeOH). 1H NMR (DMSO) δ 1.40 (d, 3, CHCH3),
3.80 (s, 4, NCH2CH2N), 4.95 (q, 1, OCH), 6.70-7.38 (m, 8, ArH),
7.68 (br s, 1, OH, exchangeable with D2O), 10.17 (br s, 1, NH,
exchangeable with D2O). (C17H18N2O2 HCl) C, H, N.
Biological Experiments. Culture of CHO Clones Expressing
3
r2-AR Subtypes. The plasmids used to stably express the human
R
2A-, R2B-, or R2C-AR in CHO cells have been previously
described.24 Briefly, each vector contains an expression cassette
comprising the CMV promoter/enhancer, the coding region of
R2C10-, R2C2-, or R2C4-gene, the IRES derived from the
encephalomyocarditis virus, the neomycin phosphotransferase
gene, and a fragment from the rabbit β-globin gene containing
an intron and a polyadenylation signal. Cells were transfected
using the calcium phosphate method, and clones were selected in
the presence of G418-sulfate (1 mg/mL). CHO-R2A, CHO-R2B
,
and CHO-R2C were routinely subcultured in Dulbecco’s Mod-
ified Eagle’s Medium (DMEM) supplemented with 10% fetal
calf serum. The clones used in this study respectively express
1.45 ( 0.3, 4.2 ( 0.8, and 2.7 ( 0.5 pmol of receptor/mg of
membrane protein.
Binding Experiments. Binding studies were performed on
crude cell membrane preparations using the selective R2-antagonist,
[3H]RX 821002, as radioligand. Briefly, membranes were incubated
in a final volume of 400 μL of Tris-Mg buffer (50 mM Tris-HCl,
0.5 mM MgCl2, pH 7.5) containing the radioligand at a concentra-
tion corresponding to twice its Kd value for the considered subtype
(3nMforCHO-R2A,10nMforCHO-R2B, and6nMforCHO-R2C
)
and increasing concentrations (10-9 to 10-4 M) of the competitor
to be tested. After 30 min of incubation at room temperature,
membrane bound radioactivity was collected by filtration through a
fiberglass filter (Whatman GFC) using a Skatron cell harvester.
Filters were rapidly washed, transferred to scintillation vials, and
counted for radioactivity. Inhibition data were analyzed using the
GraphPad Prism computer program (GraphPad Software Inc.,
San Diego, CA, USA), allowing nonlinear regression analysis
according to a one- or a two-site inhibition model.
Binding to cloned human R1-AR subtypes was performed in
membranes from CHO cells transfected by electroporation
with DNA expressing the gene encoding each R1-AR subtype.
Cloning and stable expression of the human R1-AR gene were
performed as described in ref 25. CHO cell membranes (30 μg
proteins) were incubated in 50 mM Tris-HCl, pH 7.4, with
0.1-0.4 nM [3H]prazosin, in a final volume of 1.02 mL for
30 min at 25 ꢀC, in the absence or presence of competing drugs
(1 pM to 10 μM). Nonspecific binding was determined in the
presence of 10 μM phentolamine. The incubation was stopped
by addition of ice-cold Tris-HCl buffer and rapid filtration
through 0.2% polyethyleneimine pretreated Whatman GF/B
or Schleicher & Schuell GF52 filters.
Nociceptive Test. Nociception was evaluated by the radiant
heat tail-flick test: briefly, it consists of the irradiation of the
lower third of the tail with an IR source (Ugo Basile, Comerio,
Italy). The basal predrug latency, ranging between 2 and 3 s, was
calculated as the mean of two trials performed at 30 min interval.
The mice then received tested drugs or related vehicle 15 min
before morphine or saline administration. The antinociceptive
activity was evaluated 30, 60, 90, 120, and 240 min after
Functional Assays: Cytosensor Microphysiometry. Extracellular
acidification was measured using an eight-channel Cytosensor
microphysiometry instrument (Molecular Devices, Menlo Park,
CA, USA). CHO cells expressing individually human R2-AR

7834 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21
Del Bello et al.
morphine injection. A cutoff latency of 12 s was established to
minimize tissue damage.
morphine increased the analgesia on day 1, as compared to
related-morphine group on day 1 (p < 0.01). Repeated admin-
istration of 10 mg/kg morphine sc twice daily to the mice
induced a significant decrease in the analgesia latency in the
tail-flick test on day 5 [(F(1.15) = 78.275; p < 0.001]. Pretreat-
ment of the mice with 4 or its enantiomers 15 min before each
morphine injection inhibited the development of tolerance to
morphine analgesia [(F(7.65) = 15.233; p < 0.001]. Finally,
repeated administration of 4 or its enantiomers to nondepen-
dent mice (Control) did not modify the analgesia latency in these
mice (p < 0.005) (Figure 9).
To demonstrate the involvement of R2C-AR subtype on the
enhancement of morphine antinociception induced by novel
compounds, the R2-AR antagonist yohimbine was administered
15 min before the enantiomers (S)-(þ)-4 and (R)-(-)-4. The
antinociceptive effects are expressed as percentage of the max-
imum possible effect (%MPE), according to the following
formula: %MPE = (postdrug latency - baseline latency)/(cut-
off value - baseline latency) ꢀ 100, where the postdrug latency
was the tail-flick latency 30 min after the last morphine dose.
The analgesic effect of morphine (5 mg/kg, sc) was maximal
30 min after administration and then decreased progressively so
that it became ineffective 4 h after injection. Pretreatment with
(S)-(þ)-4 and (R)-(-)-4 induced a rapid (30 min), strong, and
long lasting analgesic effect (until 240 min). The overall analysis
of variance revealed statistically significant treatment [(F(3.28) =
15.609; p<0.001] and time effects [(F(4.12) = 30.801; p < 0.001]
and interaction time ꢀ treatment effects were seen for analgesia
latency [(F(12.112) = 2.063; p < 0.05]. Also, pretreatment with
(S)-(þ)-5 and (R)-(-)-5 significantly increased the morphine-
induced antinociception as confirmed by ANOVA [(F(3.28) =
4.043; p < 0.05] (Figure 4A,B). Previous treatment of mice with
yohimbine (devoid of analgesic effect when given alone) pre-
vented the increase of morphine analgesia produced by the
enantiomers of 4 (Figure 5).
Induction of Morphine Tolerance. In accordance with the
literature,27,28 tolerance was induced in two groups of mice by the
injection of morphine (10 mg/kg, sc) twice daily at 12 h intervals for
5 days. Tolerance was evaluated by testing the antinociceptive
response to morphine in the tail flick test on the fifth day, 30 min
after the last morphine injection. In the experiments on acquisition
(development) of morphine tolerance, one group of mice was orally
administered with 0.05 mg/kg of 4, (S)-(þ)-4, and (R)-(-)-4 or
vehicle alone, twice daily, 15 min before each morphine treatment.
To determinate the effects on the expression phases of morphine
tolerance, another group of morphine-treated mice received acute
administration of 4, (S)-(þ)-4, and (R)-(-)-4 (0.05 mg/kg, ip) or
vehicle, only 15 min before the last morphine injection on the test
day (day 5). The development and expression of morphine toler-
ance were evaluated by testing the analgesic effects of morphine in
the tail-flick test, as previously described.
For morphine tolerance expression, the overall analysis of
variance revealed statistically significant treatment [(F(4.41) =
24.499; p < 0.001] and time effects [(F(1.4) = 59.158; p < 0.001]
and interaction time ꢀ treatment effects were seen for analgesia
latency [(F(4.41) = 9.536; p < 0.001]. The mice treated with
morphine showed a maximal antinociceptive effect (%MPE) on
day 1 ( p < 0.01) between control and morphine groups. Repeated
administration of 10 mg/kg morphine sc twice daily to morphine-
treated mice induced a significant decrease in the analgesia latency
in the tail-flick test, which resulted in a marked reduction on day 5,
as compared with day 1 (p < 0.01). Acute administration of 4 or its
enantiomers (0.05 mg/kg) 15 min before morphine injection on the
test day produced significant decrease in the expression of morphine
tolerance, as compared with the morphine vehicle group (p < 0.01).
The post hoc analysis revealed a statistically significant effect of 4
and (S)-(þ)-4 compared to the effect of morphine group on day 5
( p < 0.01) and lack of effect for the enantiomer (R)-(-)-4
(Figure 6). Acute administration of 4 and its enantiomers to
nondependent mice (control) did not modify the analgesia latency
in these mice (p > 0.05) (data not shown).
Induction of Morphine Dependence. To induce morphine
dependence, naıve mice were treated with morphine (10 mg/kg; sc)
twice daily at 12 h intervals for 6 days.27,28 Two hours after the
last dose of morphine, the withdrawal syndrome (abstinence), as
an index of morphine dependence, was precipitated by an ip
injection of naloxone (5 mg/kg).27,28 The combination of mor-
phine with high dose of naloxone on day 6 has been demon-
strated to induce more severe symptoms, including autonomic
signs, because naloxone precipitates dose-dependent withdra-
wal symptoms in animals acutely or chronically dependent upon
morphine.27 Ten minutes before naloxone treatment, the mice
were placed in a transparent acrylic cylinder (20 cm diameter, 35
cm high) to habituate them to the new environment. Immedi-
ately after the naloxone challenge, each mouse was again placed
gently into the cylinder and then monitored for 15 min for the
occurrence of withdrawal symptoms (jumping, rearing, forepaw
tremor, teeth chatter). To examine the effects on morphine
dependence, 4 or its enantiomers were given at dose 0.05 mg/
kg ip to one group of mice chronically treated 15 min prior to
each morphine injection (acquisition) or to another group of
mice acutely treated 15 min before naloxone (expression) as
described above. In addition, the effects of 4 or its enantiomers
alone on naloxone-induced withdrawal symptoms were exam-
ined in nondependent mice, where they received single or
repeated administration of these compounds or vehicle. The
assessment of naloxone-precipitated withdrawal symptoms
after the administration of 4 or its enantiomers has been
described above. Repeated administration of morphine pro-
duced physical dependence, as assessed by a summary of
characteristic set of behavioral responses, which included jump-
ing, rearing, forepaw tremor, and teeth chattering following the
naloxone challenge as compared to vehicle (p < 0.001). Acute
administration of 4, (S)-(þ)-4, or (R)-(-)-4 15 min prior to the
naloxone injection significantly decreased both the number of
jumping, a sign featuring the morphine withdrawal syndrome
[(F(4.40) = 15.539; p < 0.001] (Figure 7A), and the frequencies
of others withdrawal manifestations, reported as the sum of the
frequency of rearing, forepaw tremor, and teeth chattering
[(F(4.40) = 16.130; p<0.001] (Figure 7B) in morphine-depen-
dent mice. No differences were seen for the control mice treated
with 4 or its enantiomers as compared with vehicle (p < 0.05)
(data not shown). Following the naloxone challenge, the mice
that received repeated administrations of morphine showed
severe signs of withdrawal as compared with the saline control
group (p < 0.01). Pretreatment of the mice with 4 or its
enantiomers 15 min before each morphine injection significantly
attenuated both the development of jumping [F(7.63) = 30.328;
p < 0.001] (Figure 10A) and the total signs of withdrawal
[F(7.63) = 41.974; p<0.001] (Figure 10B). Particularly, jumping
was almost completely blocked by 4 because its reduction
was 80%. The mice treated with only 4 or its enantiomers
(0.05 mg/kg) did not show any significant differences in their
withdrawal symptoms, as compared with the control group
( p > 0.05).
For morphine tolerance development, the overall analysis of
variance revealed statistically significant treatment [(F(7.65) =
80.026; p < 0.001] and time effects [(F(1.7) = 70.234; p < 0.001]
and interaction time ꢀ treatment effects were seen for analgesia
latency [(F(7.65) = 11.015; p<0.001]. The mice treated with
morphine showed a maximal antinociceptive effect (%MPE) on
day 1 [(F(7.65) = 113.725; p < 0.001]. The post hoc analysis
revealed that acute coadministration of 4 or its enantiomers with
Pentobarbital Sodium-Induced Sleep Time. To rule out a
possible sedative effect elicited by (S)-(þ)-4 and (R)-(-)-4 as
already observed for 4,6 these compounds were evaluated on
pentobarbital sodium-induced sleep time test. Briefly, groups
of mice (n = 10) were injected ip with sodium pentobarbital

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 21 7835
(50 mg/kg) 15 min after tested compounds (0.05 mg/kg) or
vehicle injection. The time between losing and regaining righting
reflex (duration of sleeping) was monitored (in seconds) for each
animal.29,30
All the experiments were replicated independently in separate
rounds of animals, and the results were registered by two
observers who were blind to the treatment conditions. At tested
dose of 0.05 mg/kg, 4 or its enantiomers were not able to increase
the sleeping time induced by pentobarbital as compared to
control [F(3.36) = 1.209, p < 0.05] (Figure 8).
Statistical Analysis. The obtained results from the different
tests are presented as mean ( SEM. Antinociceptive and
morphine tolerance tests were analyzed using two-way split-plot
analysis of variance (ANOVA), with treatment as the between-
subject factor and time as the within-subject factor; when
appropriate, one-way ANOVA was used. The morphine depen-
dence data were analyzed by one-way ANOVA. The differences
between groups were determined by a post hoc comparison
Newman-Keuls test. Statistical significance was set at p < 0.05.
(11) Crassous, P.-A.; Cardinaletti, C.; Carrieri, A.; Bruni, B.; Di Vaira,
M.; Gentili, F.; Ghelfi, F.; Giannella, M.; Paris, H.; Piergentili, A.;
Quaglia, W.; Schaak, S.; Vesprini, C.; Pigini, M. R2-Adrenorecep-
tors profile modulation. 3. (R)-(þ)-m-Nitrobiphenyline, a new
efficient and R2C-subtype selective agonist. J. Med. Chem. 2007,
50, 3964–3968.
~
(12) Arredondo, Y.; Moreno-Manas, M.; Pleixats, R.; Palacin, C.;
ꢁ
Raga, M. M.; Castello, J. M.; Ortiz, J. A. Preparation, antimicrobial
evaluation, and mutagenicity of [2-hydroxyaryl]-[1-methyl-5-nitro-1H-
2-imidazolyl]methanols, [5-tert-butyl-2-methylaminophenyl]-[1-methyl-
5-nitro-1H-2 imidazolyl]methanol, and [2-hydroxyaryl]-[1-methyl-5-ni-
tro-1H-2-imidazolyl] ketones. Bioorg. Med. Chem. 1997, 5, 1959–1968.
(13) Bishop, M. J.; Barvian, K. A.; Berman, J.; Bigham, E. C.; Garrison,
D. T.; Gobel, M. J.; Hodson, S. J.; Irving, P. E.; Liacos, J. A.;
Navas, I.; Saussy, D. L.; Speake, J. D. R1-Adrenoceptor agonists:
the identification of novel R1A subtype selective 20-heteroaryl-
2-(phenoxymethyl)imidazolines. Bioorg. Med. Chem. Lett. 2002,
3, 471–475.
€
(14) Tottie, L.; Baeckstrom, P.; Moberg, C.; Tegenfeldt, J.; Heumann,
A. Molecular sieve controlled diastereoselectivity: effect in the
palladium-catalyzed cyclization of cis-1,2-divinylcyclohexane with
R-oxygen-substituted acids as chiral nucleophiles. J. Org. Chem.
1992, 57, 6579–6587.
€
(15) Bostrom, J.; Hogner, A.; Schmitt, S. Do structurally similar ligands
Acknowledgment. We thank the MIUR (Rome), the Uni-
versity of Camerino, and the Fondazione Monte dei Paschi
(Siena) for financial support.
bind in a similar fashion? J. Med. Chem. 2006, 49, 6716–6725.
€
ꢂ
€
€
(16) Ozdogan, U. K.; Lahdesmaki, J.; Scheinin, M. Influence of prazo-
sin and clonidine on morphine analgesia, tolerance and withdrawal
in mice. Eur. J. Pharmacol. 2003, 460, 127–134.
€
ꢂ
€
€
(17) Ozdogan, U. K.; Lahdesmaki, J.; Hakala, K.; Scheinin, M.
The involvement of R2A-adrenoceptors in morphine analgesia,
tolerance and withdrawal in mice. Eur. J. Pharmacol. 2004, 497,
161–171.
Supporting Information Available: Experimental and spectro-
scopic details of (S)-(þ)- and (R)-(-)-5, (S)-(þ)- and (R)-(-)-6,
(S)-(-)-, and (R)-(þ)-9, (S)-(þ)-, and (R)-(-)-10, solubility and
permeability assays, intrinsic clearance, and elemental analysis
of the final compounds. This material is available free of charge
via the Internet at http://pubs.acs.org.
(18) Koob, G. F.; Le Moal, M. Neurobiological mechanisms for
opponent motivational processes in addiction. Philos. Trans. R.
Soc., B 2008, 363, 3113–3123.
(19) Di, L.; Kerns, E. H.; Fan, K.; McConnell, O. J.; Carter, G. T. High
throughput artificial membrane permeability assay for blood-
-brain barrier. Eur. J. Med. Chem. 2003, 38, 223–232.
(20) Gowing, L.; Farrell, M.; Ali, R.; White, J. M. Alpha2 adrenergic
agonists for the management of opioid withdrawal. Cochrane
Database Syst. Rev. 2009, 2, CD002024.
(21) Veilleux, J. C.; Colvin, P. J.; Anderson, J.; York, C.; Heinz, A. J. A
review of opioid dependence treatment: pharmacological and
psychosocial interventions to treat opioid addiction. Clin. Psychol.
Rev. 2010, 30, 155–166.
(22) El-Kadi, A. O.; Sharif, S. I. The influence of chronic treatment with
clonidine, yohimbine and idazoxan on morphine withdrawal.
Psychopharmacology 1997, 132, 67–73.
References
(1) Part 6 of R2-Adrenoreceptors profile modulation. For part 5, see ref 6.
(2) Hieble, J. P. Adrenoceptor subclassification: an approach to
improved cardiovascular therapeutics. Pharm. Acta Helv. 2000,
74, 163–171.
ꢁ
(3) Crassous, P.-A.; Denis, C.; Paris, H.; Senard, J. M. Interest of
R2-adrenergic agonists and antagonists in clinical practice: back-
ground, facts and perspectives. Curr. Top. Med. Chem. 2007, 7,
187–194.
(4) Tan, C. M.; Limbird, L. E. The R2-adrenergic receptors. In The
Receptors: the Adrenergic Receptors in the 21st Century; Perez, D.,
Ed.; Humana Press Inc.: Totowa, NJ, 2006; pp 241-265.
(5) Fairbanks, C. A.; Stone, L. S.; Kitto, K. F.; Nguyen, H. O.;
Posthumus, I. J.; Wilcox, G. L. R2C-Adrenergic receptors mediate
spinal analgesia and adrenergic-opioid synergy. J. Pharmacol. Exp.
Ther. 2002, 300, 282–290.
(23) Hameedi, F. A.; Woods, S. W.; Rosen, M. I.; Pearsall, H. R.;
Kosten, T. R. Dose dependent effects of yohimbine on methadone
maintained patients. Am. J. Drug Alcohol Abuse 1997, 23, 327–333.
(24) Schaak, S.; Devedjian, J. C.; Paris, H. Use of eukaryotic vectors for
the expression of adrenergic receptors. Methods Mol. Biol. 2000,
126, 189–206.
(6) Cardinaletti, C.; Mattioli, L.; Ghelfi, F.; Del Bello, F.; Giannella,
M.; Bruzzone, A.; Paris, H.; Perfumi, M.; Piergentili, A.; Quaglia,
W.; Pigini, M. Might adrenergic R2C-agonists/R2Aantagonists be-
come novel therapeutic tools for pain treatment with morphine?
J. Med. Chem. 2009, 52, 7319–7322 and references therein.
(7) Gentili, F.; Bousquet, P.; Brasili, L.; Caretto, M.; Carrieri, A.;
Dontenwill, M.; Giannella, M.; Marucci, G.; Perfumi, M.; Piergentili,
A.; Quaglia, W.; Rascente, C.; Pigini, M. R2-Adrenoreceptors profile
modulation and high antinociceptive activity of (S)-(-)2-[1-biphenyl-2-
yloxy]ethyl]-4,5-dihydro-1H-imidazole. J. Med. Chem. 2002, 45, 32–40.
(8) Gentili, F.; Ghelfi, F.; Giannella, M.; Piergentili, A.; Pigini, M.;
Quaglia, W.; Vesprini, C.; Crassous, P.-A.; Paris, H.; Carrieri, A.
R2-Adrenoreceptors profile modulation. 2. Biphenyline analogues
as tools for selective activation of the R2C-subtype. J. Med. Chem.
2004, 47, 6160–6173.
(9) Shi, L.; Javitch, J. A. The binding site of aminergic G protein-
coupled receptors: the transmembrane segments and second extra-
cellular loop. Annu. Rev. Pharmacol. Toxicol. 2002, 42, 437–467.
(10) Gentili, F.; Cardinaletti, C.; Vesprini, C.; Carrieri, A.; Ghelfi, F.;
Farande, A.; Giannella, M.; Piergentili, A.; Quaglia, W.; Laurila,
J. M.; Huhtinen, A.; Scheinin, M.; Pigini, M. R2-Adrenoreceptors
profile modulation. 4. From antagonist to agonist behaviour.
J. Med. Chem. 2008, 51, 4289–4299.
(25) Testa, R.; Taddei, C.; Poggesi, E.; Destefani, C.; Cotecchia, S.;
Hieble, J. P.; Sulpizio, A. C.; Naselsky, D.; Bergsma, D.; Ellis, S.;
Swift, A.; Ganguly, S.; Ruffolo, R. R.; Leonardi, A. Rec 15/2739
(SB 216469): a Novel prostate selective R1-adrenoceptor antago-
nist. Pharmacol. Commun. 1995, 6, 79–86.
(26) Arunlakshana, O.; Schild, H. O. Some quantitative uses of drug
antagonists. Br. J. Pharmacol. Chemother. 1959, 14, 48–58.
(27) Ren, X.; Noda, Y.; Mamiya, T.; Nagai, T.; Nabeshima, T. A
neuroactive steroid, dehydroepiandrosterone sulfate, prevents the
development of morphine dependence and tolerance via c-fos
expression linked to the extracellular signal-regulated protein
kinase. Behav. Brain Res. 2004, 152, 243–250.
(28) Abdel-Zaher, A. O.; Hamdy, M. M.; Aly, S. A.; Abdel-Hady,
R. H.; Abdel-Rahman, S. Attenuation of morphine tolerance and
dependence by aminoguanidine in mice. Eur. J. Pharmacol. 2006,
540, 60–66.
(29) Carvalho, A. A.; Galdino, P. M.; Nascimento, M. V.; Kato, M. J.;
Valadares, M. C.; Cunha, L. C.; Costa, E. A. Antinociceptive and
antiinflammatory activities of grandisin extracted from Virola
surinamensis. Phytother. Res. 2010, 24, 113–118.
ꢁ
(30) Carlini, E. A.; Burgos, V. Screening farmacologico de ansiolıticos:
´
~
metodologia laboratorial e comparac-ao entre diazepam e cloro-
benzepam. Rev. Assoc. Bras. Psiq. 1979, 1, 25–31.