Chiral aryloxyalkylamines: Selective 5-HT1B/1D activation and analgesic activity

Article abstract of DOI:10.1002/cmdc.200900530

A series of chiral 2,3-dichlorophenoxy and 1-naphthyloxy alkylamines were synthesized, and their binding affinities towards 5-HT_1D and h5-HT_1B receptors were evaluated. In the naphthyloxy series, the (R)-prolinol derivative was the most selective 5-HT_1D ligand, while (S)-N-methyl-2-(1-naphthyloxy)propan-1-amine showed the highest selectivity for h5-HT_1B. Both compounds performed as 5-HT_1D agonists in the isolated guinea pig assay and showed higher analgesic activity than both sumatriptan and the achiral analogue 20b in the mouse hotplate test. Neither ligand displayed any affinity for nicotinic ACh receptors present in mouse brain membranes, thus indicating that their analgesic activity does not arise through interaction with these receptors.

Full text of DOI:10.1002/cmdc.200900530

MED
DOI: 10.1002/cmdc.200900530
Chiral Aryloxyalkylamines: Selective 5-HT1B/1D Activation
and Analgesic Activity
Alessia Carocci,[a] Giovanni Lentini,*[a] Alessia Catalano,[a] Maria Maddalena Cavalluzzi,[a]
Claudio Bruno,[a] Marilena Muraglia,[a] Nicola Antonio Colabufo,[a] Nicoletta Galeotti,[b]
Filomena Corbo,[a] Rosanna Matucci,[b] Carla Ghelardini,[b] and Carlo Franchini[a]
A series of chiral 2,3-dichlorophenoxy and 1-naphthyloxy alkyl-
amines were synthesized, and their binding affinities towards
5-HT1D and h5-HT1B receptors were evaluated. In the naphthyl-
oxy series, the (R)-prolinol derivative was the most selective
5-HT1D ligand, while (S)-N-methyl-2-(1-naphthyloxy)propan-1-
amine showed the highest selectivity for h5-HT1B. Both com-
pounds performed as 5-HT1D agonists in the isolated guinea
pig assay and showed higher analgesic activity than both su-
matriptan and the achiral analogue 20b in the mouse hot-
plate test. Neither ligand displayed any affinity for nicotinic
ACh receptors present in mouse brain membranes, thus indi-
cating that their analgesic activity does not arise through inter-
action with these receptors.
Introduction
Serotonin (5-HT) receptors are divided into seven classes (5-
HT1–7) based upon pharmacological, structural, and signaling
properties.[1–4] 5-HT1 receptors are further divided into several
populations that include the 5-HT1B and 5-HT1D subtypes.[5] 5-
HT1B receptors are found in rodent brain but not in human
brain, whereas the reverse is true for 5-HT1D receptors.[1] Two
distinct populations of human 5-HT1D receptors have been
identified: 5-HT1Da and 5-HT1Db receptors. In an attempt to
unify rather confusing nomenclature, these subtypes are now
commonly named h5-HT1D and h5-HT1B, respectively. The latter
is believed to represent the species homologue of the rat 5-
HT1B receptor.[5–7] Receptor mapping studies[8] showed that the
h5-HT1B receptor is widely distributed in both neural and vas-
cular tissues of the central nervous system (CNS), whereas the
h5-HT1D receptor is restricted to neural tissues including the tri-
geminal ganglion. Both receptor subtypes are believed to play
a pivotal role in both vascular and injured nerve-pain modula-
tion. 5-HT1B/1D receptors are localized on peptidergic nocicep-
tors in the so-called trigeminocervical complex—a group of
cells located in the rostral sections of the spinal cord that are
supposed to provide a possible anatomical substrate for migra-
neous pain. Finally, 5-HT1B/1D receptors seem to modulate vas-
cular nociceptive stimuli processed in the thalamus.[9]
the clinical efficacy of sumatriptan, the first of a well-known
class of therapeutic agents for the treatment of acute migraine
headaches, is mediated through action on both h5-HT1D and
h5-HT1B receptor subtypes.[13] Activation of the h5-HT1B recep-
tor may cause coronary artery constriction, which precludes
the use of sumatriptan in patients with concomitant heart dis-
ease.[14–16] These findings have suggested the development of
selective h5-HT1D receptor agonists as potential second-genera-
tion antimigraine agents, which may lack the vasoconstrictor
action of unselective compounds, such as sumatriptan, thus
presenting an improved side-effect profile.
Most h5-HT1D ligands are tryptamine-based and, hence, are
likely to suffer from related drawbacks including poor adsorp-
tion and low bioavailability.[17] As a result, non-tryptamine
agents with selectivity for h5-HT1D over h5-HT1B receptors are
5-HT1B/1D receptor subtypes are also involved in the seroto-
nergic descending pain modulating system,[10] and 5-HT1B acti-
vation was demonstrated to produce a selective blockade of
the spinal nociceptive processing.[11] Both subtypes have been
implicated in the control of several sensory circuits, and the ac-
tivation of 5-HT1B and 5-HT1D receptors have opposing actions
on some nerves receiving afferent input.[12] Thus, selective h5-
HT1D and h5-HT1B agonists could be useful to clarify the role of
each subtype in the processing of nociceptive stimuli and
might drive the research into new agents for the treatment of
migraine and chronic pain. Indeed, it is generally accepted that
[a] Dr. A. Carocci, Prof. G. Lentini, Dr. A. Catalano, Dr. M. M. Cavalluzzi,
Dr. C. Bruno, Dr. M. Muraglia, Prof. N. A. Colabufo, Prof. F. Corbo,
Prof. C. Franchini
Dipartimento Farmaco-Chimico, Universitꢀ degli Studi di Bari
Via Orabona 4, 70125 Bari (Italy)
Fax: (+39)080-5442724
E-mail: glentini@farmchim.uniba.it
[b] Dr. N. Galeotti, Dr. R. Matucci, Prof. C. Ghelardini
Dipartimento di Farmacologia Preclinica e Clinica, Universitꢀ di Firenze
Viale G. Pieraccini 5/6, 50139 Firenze (Italy)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.200900530.
696
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 696 – 704

Chiral Aryloxyalkylamines
attractive targets for development. Studies focused on the
identification of novel non-tryptaminergic 5-HT1D ligands
showed that aryloxyalkylamines, such as 2-(2,3-dichlorophen-
oxy)-N-methylethanamine (20a) and N-methyl-2-(1-naphthyl-
oxy)ethanamine (20b), bind to h5-HT1D and h5-HT1B receptors
with high affinity. Indeed, some naphthyl derivatives display
approximately tenfold selectivity for h5-HT1D over h5-HT1B re-
ceptors.[18,19] Starting from the observation that both trypta-
minic[20,21] and non-indole[22] chiral 5-HT1D/1B ligands show ste-
reochemistry-dependant selectivity in terms of affinity and effi-
cacy, we speculated that both efficacy and selectivity towards
h5-HT1B and h5-HT1D might be modulated by exploiting chirali-
ty. Chiral aryloxyalkylamines 21a,b and 22a,b were designed
by the introduction of a methyl group onto the ethylenic
chain of the parent compounds 20a,b. Two conformationally
constrained analogues were designed by including part of the
ethylenic chain in a pyrrolidine ring (31a,b). Here we report
the synthesis of both series of compounds and the preliminary
investigation into their pharmacological profiles.
Scheme 1. Reagents and conditions: a) TPP, DIAD, anhyd THF, RT; b) potas-
sium phthalimide, anhyd DMF, RT; c) Et3N, toluene, reflux; d) hydrazine hy-
drate (55%), AcOH, CH3OH, reflux; e) (CF3CO)2O, anhyd THF, 08C!RT; f) MeI,
K2CO3, anhyd DMF, reflux; g) K2CO3, CH3OH/H2O, reflux; h) aq HCl.
Chemistry
14a,b–16a,b with methyl iodide gave 17a,b–19a,b, which in
turn were hydrolyzed[26] to afford 20a,b–22a,b. The free
amines were converted into the corresponding hydrochlorides
(20a,b·HCl–22a,b·HCl) by treatment with aqueous HCl and
azeotropic removal of water.
N-Methylaryloxyalkylamines 20a,b–22a,b were prepared via
the synthetic routes shown in Scheme 1. Compounds 20a,b
were prepared starting from commercially available phthalimi-
do ethanol 4, which was condensed with the appropriate aro-
matic hydroxy compound (1a,b) under Mitsunobu condi-
tions[23] to give the corresponding aryl ethyl ethers (8a,b). N-
Methyl-3-aryloxy-2-propylamines 21a,b were prepared starting
from phthalimido alcohol 7 (prepared as reported in the litera-
ture),[24] which was condensed with compounds 1a,b under
Mitsunobu conditions to give the corresponding aryl alkyl
ethers (9a,b). N-Methyl-2-aryloxy-1-propylamines 22a,b were
prepared from compounds 1a,b, which were O-alkylated with
3-bromo-2-propanol under Mitsunobu conditions. The result-
ing bromo derivatives (3a,b) underwent a Gabriel reaction
with potassium phthalimide to give the corresponding aryl
propyl ethers (10a,b). Amines 11 a,b–13a,b, obtained by hy-
drazinolysis[25] of the corresponding phthaloylated intermedi-
ates 8a,b–10a,b, were converted into the corresponding tri-
fluoroacetamido derivatives 14a,b–16a,b by treatment with
trifluoroacetic anhydride in anhydrous THF. N-Alkylation of
The homochiral forms of 22b were prepared according to
Scheme 2, which differs from Scheme 1 only in the first three
steps. Thus, 1b was condensed with commercially available
methyl (R)-(+)-lactate [(R)-23] and ethyl (S)-(ꢀ)-lactate [(S)-24]
according to Mitsunobu procedure to give the corresponding
esters (S)-25 and (R)-26, respectively. (S)- and (R)-27, obtained
by reduction of (S)-25 and (R)-26 respectively, were reacted
with phthalimide under Mitsunobu conditions to give the
phthaloylated intermediates (S)- and (R)-10b.
2-(Aryloxymethyl)pyrrolidines 31a,b were prepared accord-
ing to the synthetic route described in Scheme 3. In the first
step, the amino groups of the homochiral pyrrolidin-2-ylmeth-
anols (R)- and (S)-28 were Boc protected. The obtained N-Boc
aminoalcohols (R)- and (S)-29 were reacted with 1a,b accord-
ing to Mitsunobu procedure to give the corresponding aryl
alkyl ethers (R)- and (S)-30a,b. The removal of the Boc protect-
ChemMedChem 2010, 5, 696 – 704
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
697

G. Lentini et al.
MED
(Table 1); pKa, LogP, and LogD values for all of these com-
pounds were experimentally determined in order to evaluate a
possible correlation between their physicochemical properties
and biological activity. All aryloxyalkylamines presented higher
LogD values than sumatriptan. Naphthyloxy derivatives (b
series) were more basic and showed higher affinity for h5-HT1B
than the corresponding 2,3-dichlorophenoxy analogues (a
series). With the exception of (R,S)-21 and (S)-31, the same re-
lationships were also observed in 5-HT1D affinities. Compounds
21a,b, in which a methyl group has been formally introduced
in the 1-position of the ethylamino side chain, were poor 5-
HT1D and h5-HT1B ligands (when compared to reference com-
pounds 20a,b).
The introduction of a methyl group in 2-position of the alkyl
chain (22a,b) offers no advantage over the parent compounds
(20a,b), if racemates are considered. However, when homochi-
ral forms are concerned, chirality conferred stereochemistry-de-
pendent selectivity in favor of h5-HT1B. In fact, the addition of
a methyl group did not markedly alter h5-HT1B affinity regard-
less of stereochemistry (cf. (R)- and (S)-22b with 20b) but dra-
matically lowered (S)-22b affinity for 5-HT1D (two orders of
magnitude). Derivative 22b exhibits high stereo-differentiation
in terms of receptor binding, especially for the 5-HT1D subtype;
the (R)-isomer is the eutomer (Ki distomer/Ki eutomer=84).
Thus, (S)-22b was about ten times more selective for h5-HT1B
then the reference compound 20b.
A reversal of receptor subtype selectivity was observed
when constraining the nitrogen atom into a pyrrolidine ring to
give restricted analogues 31a,b. While h5-HT1B affinity was
lowered (cf. 20b with (R)- and (S)-31a,b), 5-HT1D affinity was
slightly improved in (R)-31b, which performed as the most se-
lective 5-HT1D ligand of both series (h5-HT1B Ki/5-HT1D Ki =26).
Derivative (S)-31b showed about 20-times lower affinity. When
comparing 31a and 31b affinities for 5-HT1D receptors, a rever-
sal of the stereoselectivity pattern was observed, with (S)- and
(R)-isomer being the eutomers, respectively.
Scheme 2. Reagents and conditions: a) TPP, DBAD, anhyd THF, RT; b) LiAlH4,
anhyd THF, reflux!RT; c) phthalimide, TPP, DBAD, anhyd THF, RT; d) hydra-
zine hydrate (55%), AcOH, CH3OH, reflux; e) (CF3CO)2O, anhyd THF, 08C!RT;
f) MeI, K2CO3, anhyd DMF, reflux; g) K2CO3, CH3OH/H2O, reflux; h) aq HCl.
In order to establish their intrinsic activity, the most selective
compounds for h5-HT1B and 5-HT1D, (S)-22b and (R)-31b, re-
spectively, together with sumatriptan as a reference com-
pound, were evaluated in the isolated guinea pig ileum assay,
where 5-HT1D receptor agonists inhibit the substance P evoked
contractions in a dose-dependent fashion.[29] All compounds
displayed agonistic activity in the same rank order (EC50 values
from 12 mm to 30 mm) as depicted in Figure 1. In order to es-
tablish for each compound whether this effect was 5-HT1D
mediated, the assay was repeated after desensitizing 5-HT1D re-
ceptors with sumatriptan at high concentration (Figure 2). The
results showed that the sumatriptan-mediated relaxation effect
was largely due to 5-HT1D contribution. By contrast, the relaxa-
tion effects of (S)-22b and (R)-31b are partially 5-HT1D mediat-
ed (60% and 50%, respectively).
Scheme 3. Reagents and conditions: a) Boc2O, NaOH, THF, RT; b) 1a or 1b,
TPP, DBAD, anhyd THF, RT; c) 98% HCOOH, 08C!RT; d) aq HCl.
ing group with formic acid[20] afforded the aryloxymethylpyrrol-
idines (R)- and (S)-31a,b, which were then converted into the
corresponding hydrochlorides [(R)- and (S)-31a,b·HCl].
Agonists (S)-22b and (R)-31b, together with sumatriptan
and 20b as an in vivo baseline, were evaluated in the hot-
plate test (Table 2). Both compounds, administered (i.p.) at a
dose of 30 mgkgꢀ1, were able to increase the pain threshold in
mice in the presence of a thermal stimulus in a statistically sig-
nificant manner. The analgesic effect induced by (S)-22b and
Results
Aryloxyalkylamines 20a,b and their analogues were evaluated
for their binding affinity towards 5-HT1D and h5-HT1B receptors
698
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 696 – 704

Chiral Aryloxyalkylamines
Table 1. Chemical and binding properties of aryloxyalkylamines.
[a]
[a]
Compd
Ar
R1
R2
R3
n
pKa
LogP[a]
LogD7.4
Affinity[b]; Ki ꢁSEM [nM]
Ratio[c]
5-HT1D
h5-HT1B
20a
2,3-dichlorophenyl
2,3-dichlorophenyl
2,3-dichlorophenyl
1-naphthyl
1-naphthyl
1-naphthyl
1-naphthyl
1-naphthyl
2,3-dichlorophenyl
2,3-dichlorophenyl
1-naphthyl
H
H
CH3
H
H
CH3
CH3
CH3
H
H
H
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
H
H
CH3
H
H
CH3
H
H
H
CH2
CH2
CH2
CH2
0
0
0
0
0
0
0
0
2
2
2
2
8.98ꢁ0.01
9.11ꢁ0.03
8.99ꢁ0.03
9.19ꢁ0.01
9.35ꢁ0.01
9.37ꢁ0.03
–
2.70ꢁ0.01
2.98ꢁ0.02
2.88ꢁ0.01
2.84ꢁ0.01
3.29ꢁ0.01
3.20ꢁ0.01
–
1.14
1.28
1.30
1.11
1.37
1.26
–
–
1.42
–
57.9ꢁ1.5[d]
410ꢁ30
180ꢁ20
18.1ꢁ0.5[g]
962ꢁ50
29.6ꢁ2.3
21.3ꢁ0.2
1800ꢁ300
370ꢁ25
67.0ꢁ1.5
11ꢁ0.50
231ꢁ12
5ꢁ3[j]
420ꢁ20[e]
>1000 (40%)[f]
130ꢁ15
7
>2
0.7
3
0.7
2.8
0.6
0.1
7
18
25.7
0.7
4
(R,S)-21a
(R,S)-22a
20b
55.8ꢁ2.1[h]
640ꢁ25
(R,S)-21b
(R,S)-22b
(R)-22b
(S)-22b
(R)-31a
(S)-31a
(R)-31b
(S)-31b
sumatriptan
83ꢁ7
12.0ꢁ0.5
171ꢁ30
–
–
9.67ꢁ0.02
–
9.84ꢁ0.01
–
3.61ꢁ0.02
–
3.65ꢁ0.06
–
2650ꢁ150
1230ꢁ150
283ꢁ20
H
H
H
1.30
–
ꢀ1.50[i]
1-naphthyl
H
169ꢁ27
22ꢁ5[j]
[a] pKa and LogP values were experimentally determined using a Sirius pKa analyzer; LogD values were calculated on the basis of experimental pKa and
LogP values. [b] All values represent the mean of two or three independent determinations. [c] h5-HT1B/h5-HT1D Ki values. [d] Literature value=45ꢁ16.[19]
[e] Literature value=96ꢁ22.[19] [f] Full Ki value not obtained; percentage inhibition at the concentration shown given in parentheses. [g] Literature
values=34ꢁ9;[18] 31ꢁ3.[19] [h] Literature values=26ꢁ6;[18] 74ꢁ8.[19] [i] See Reference [27]. [j] see Reference [28].
&
Figure 2. Maximal inhibition of substance P-induced contractions exerted by
Figure 1. Dose–response curves for sumatriptan ( ; EC50 =30ꢁ2 mm), (S)-
~
!
sumatriptan, (S)-22b, and (R)-31b in physiological conditions (&) and in 5-
HT1D-desensitized receptors (&).
22b ( ; EC50 =12ꢁ1 mm), and (R)-31b ( ; EC50 =16ꢁ2 mm) as 5-HT1D recep-
tor agonists (% inhibition of substance P-induced contractions).
(R)-31b reached a maximum 30 min after injection, (R)-31b
being more active. The analgesic effect of (S)-22b and (R)-31b
was persistent for up to 75 and
ported as nicotinic acetylcholine receptor (nAChR) ligands,[30]
and nicotinic mechanisms are involved in nociception,[31] (S)-
60 min, respectively. Sumatriptan
Table 2. Effect of (S)-22b and (R)-31b in the mouse hot-plate test (52.58C).[a]
was significantly less active than
both (S)-22b and (R)-31b, while
20b was inactive. At the maxi-
mal dose employed, both (S)-
22b and (R)-31b did not alter
animal behavioral parameters,
such as motor coordination and
spontaneous motility (data not
shown). This observation ruled
out possible toxicity to both
brain areas and serotonergic
muscle reflex pathways.[6]
Compd
Dose[b]
Licking latency [s]
[mgkgꢀ1
]
before
treatment
after treatment
15 min
30 min
45 min
60 min
75 min
saline
20b
–
30
3
10
30
10
30
10
30
14.9ꢁ0.5
14.3ꢁ0.8
15.6ꢁ0.6
14.6ꢁ0.4
15.5ꢁ0.4
14.8ꢁ0.8
15.4ꢁ0.7
14.8ꢁ1.1
14.3ꢁ1.3
15.1ꢁ0.5
19.2ꢁ0.9
15.1ꢁ1.0
23.0ꢁ1.5*
29.7ꢁ2.5*
16.9ꢁ0.5
37.2ꢁ1.8*
21.1ꢁ1.3*
25.2ꢁ1.4*
15.2ꢁ0.5
21.4ꢁ1.3*
16.7ꢁ1.0
23.2ꢁ0.9*
32.6ꢁ2.0*
17.6ꢁ0.7
36.4ꢁ2.5*
14.9ꢁ2.0
22.8ꢁ2.1*
15.3ꢁ0.5
18.1ꢁ1.5
14.6ꢁ0.7
19.2ꢁ1.6
29.6ꢁ2.3*
16.0ꢁ0.9
28.7ꢁ2.9*
17.3ꢁ1.9
21.3ꢁ1.6*
14.9ꢁ0.9
17.5ꢁ1.4
–
15.7ꢁ0.5
27.2ꢁ2.7*
–
20.1ꢁ1.5^
15.6ꢁ1.7
17.5ꢁ1.4*
14.9ꢁ0.7
16.0ꢁ1.3
–
(S)-22b
(S)-22b
(S)-22b
(R)-31b
(R)-31b
sumatriptan
sumatriptan
–
23.7ꢁ2.5*
–
18.5ꢁ2.2
16.3ꢁ1.5
15.9ꢁ1.3
[a] *P<0.01, ^P<0.05 versus saline treated mice. Each value represents at least eight mice. [b] Animals dosed
by i.p. administration.
Since 2-(aryloxymethyl)-azacy-
clic compounds have been re-
ChemMedChem 2010, 5, 696 – 704
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
699

G. Lentini et al.
MED
22b and (R)-31b, together with nicotine as a reference com-
pound, were tested for their binding affinities at nAChRs pres-
ent in mouse brain membranes labeled by [3H]cytisine
(Table 3). Only nicotine was able to inhibit the specific binding
of [3H]cytisine, showing an IC50 value of 6.76ꢁ0.16 nm (pIC50 =
8.17ꢁ0.14), while the two newly synthesized compounds
failed to bind to these receptors; (S)-22b and (R)-31b were
only partially able to decrease [3H]cytisine-specific binding,
with IC50 values of 0.692ꢁ0.015 mm (pIC50 =3.16ꢁ0.07) and
44.9ꢁ0.8 mm (pIC50 =4.35ꢁ0.08), respectively.
guinea pig ileum assay, and displayed a more potent analgesic
effect than both sumatriptan and the achiral analogue 20b.
When evaluated in a nAChR binding assay, compounds (S)-22b
and (R)-31b showed no significant affinity towards nAChRs
(Table 3), indicating that the analgesic effect of (S)-22b and (R)-
31b does not benefit from direct nicotinic mechanisms. Other
mechanisms, both centrally- (e.g., noradrenergic modulating
system)[10] and peripherally- (e.g., sodium-channel blockade)[33]
localized, might contribute and, thus, explain why no clear re-
lationship between h5-HT1B/5-HT1D affinities and in vivo analge-
sia was found.
Table 3. Binding affinities of the tested compounds at nAChR in mouse
brain membranes.
Compd
[3H]cytisine[a]
pIC50
[3H]cytisine
IC50 ꢁSEM [mm]
Conclusions
Effective alleviation of pain has still to be achieved. Currently
available therapeutic agents are unsatisfactory for about 50%
of patients and may be accompanied by several side effects.[31]
The last 20 years have witnessed a rise in interest towards the
so-called adjuvant analgesic drugs, which have primary indica-
tions other than pain but may be analgesics under certain cir-
cumstances.[34] In this heterogeneous group of alternative
drugs, h5-HT1B/1D-selective agonists acting on the serotonergic
pain-modulating system may offer new perspectives. The aryl-
oxyalkylamines reported herein present higher LogD7.4 values
than sumatriptan. In principal, they should allow pharmacolog-
ical investigation of parts of the serotonergic pain-modulating
system generally unreachable by sumatriptan because of its
low LogD7.4 values.[27,28]
nicotine
(S)-22b
(R)-31b
8.17ꢁ0.14
3.16ꢁ0.07
4.35ꢁ0.08
0.00676ꢁ0.00016
692ꢁ15
44.9ꢁ0.8
[a] Values are expressed as pIC50 ꢁSEM of at least three independent ex-
periments.
Discussion
In agreement with previously reported data,[19] our study con-
firms that the naphthyl and 2,3-dichlorophenyl rings could be
considered as bioisosteric nuclei. Both aryloxy moieties can
also be considered as bioisosteres of the indole nucleus of
tryptaminergic 5-HT1B/1D ligands. Similar observations have
been made in the melatoninergic system, where we obtained
potent melatonin analogues by replacing the melatonin 5-me-
thoxyindol-3-yl moiety with a m-methoxyphenoxyl group.[32]
However, when homochiral bioisosteres 31a and 31b are con-
sidered, a reversal of the stereochemistry-dependent 5-HT1D af-
finity patterns was observed. Thus, it may be argued that the
two bioisosteric aryloxy moieties do not bind exactly in the
same mode to the active site. The naphthyloxy derivatives (b
series) generally show slightly higher affinities than the 2,3-di-
chlorophenoxy analogues (a series), presumably as a conse-
quence of higher basicities. The introduction of a methyl
group in the a position to the nitrogen of the ethylamino side
chain (compounds 21a,b) was detrimental. The above observa-
tions are in agreement with the hypothesis that an aspartic
acid residue (Asp118) in the h5-HT1D sequence interacts with
the protonated forms of ligands.[21] In fact, while higher pKa
values lead to an increase in the molar ratios of the protonated
forms in the biophase, steric hindrance introduced by substitu-
tion might hamper the interaction with the anionic binding
site and/or prevents the homologues adopting the active con-
formation. However, these detrimental steric factors did not
completely abolish affinity in the pyrrolidine derivatives series
(31a,b), probably due to compensating effects of basicity en-
hancement and/or chain strain imposed by inclusion of the
a-substituent in the pyrrolidine ring.
The superior analgesic effect of the newly proposed homo-
chiral bioisosteres of sumatriptan, (S)-22b and (R)-31b, com-
pared to that of both the parent compound and the achiral
analogue 20b and despite comparable affinity for 5-HT1B/1D re-
ceptors, might be ascribed to concurrent unknown mecha-
nisms and/or advantageous pharmacokinetics. Our preliminary
results suggest that nicotinic systems should be ruled out.
Work has been undertaken to explore the possible participa-
tion of nerve sodium-channel blockade.
Since aryloxyalkylamines can be obtained through facile and
versatile synthetic routes reported in this paper and else-
where,[24,31,35,36] the compounds described herein might help to
reach the goal described above. The requirements for stereo-
chemistry-specific binding of an aryloxyalkylamine to 5-HT1D
are generally more stringent than those for h5-HT1B binding.
These observations show that it may be possible to design
more selective h5-HT1B ligands by modulating the volume of
the substituent in the 2-position of 20b alkyl chain in order to
reduce 5-HT1D affinity, as observed for (S)-22b. On the other
hand, the conformationally constrained analogue (R)-31b may
well be considered as the starting point for future SAR studies
aimed at improving affinity. Finally, it should be noted that
chiral aryloxyalkylamines, despite their relatively low-molecular
weight, possess molecular complexity (i.e., relatively high pro-
portion of sp3-hybridized carbons and chirality) and this should
increase the probability that their study will lead to useful
drugs.[37,38]
Both (S)-22b and (R)-31b performed as 5-HT1D agonists,
showing higher efficacy than sumatriptan in the isolated
700
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 696 – 704

Chiral Aryloxyalkylamines
Anal. calcd for C10H13NOCl2·0.67HCl: C 46.48, H 5.33, N 5.42, found:
C 46.32, H 5.33, N 5.64.
Experimental Section
Chemistry
(R,S)-2-(2,3-Dichlorophenoxy)-N-methylpropan-1-amine Hydro-
chloride (22a·HCl): The title compound was prepared from 19a
using the procedure described for the preparation of 20a·HCl (See
Supporting Information). Compound 22a (70%): MS (70 eV): m/z
(%): 234 [M+1]+ (<1), 44 (100). Salt 22a·HCl (57%): mp: 175–
1768C (EtOH/Et2O); 1H NMR (300 MHz, [D6]DMSO): d=1.25 (d, J=
5.9 Hz, 3H, CH3CH), 2.60 (s, 3H, CH3NH), 3.18–3.30 (m, 2H, CH2),
4.90–5.05 (m, 1H, CH), 7.20–7.40 (m, 3H, Ar), 9.26 ppm (br s, 2H,
NH2); 13C NMR (75 MHz, CD3OD): d=15.9 (1C), 33.2 (1C), 53.4 (1C),
72.5 (1C), 115.5 (1C), 123.6 (1C), 124.0 (1C), 128.0 (1C), 133.8 (1C),
153.7 ppm (1C); Anal. calcd for C10H13NOCl2·HCl·0.33H2O: C 43.42, H
5.34, N 5.06, found: C 43.71, H 5.02, N 5.07.
Chemicals were purchased from Sigma–Aldrich or Lancaster in the
highest quality commercially available. Solvents were RP grade
unless otherwise indicated. Compound 7 was prepared as previ-
ously described.[24] (+)-(S)-O-p-Chlorophenylmandelic acid, used as
a chiral solvating agent, was prepared in our laboratories.[39] Yields
refer to purified products and were not optimized. Yields of hydro-
chlorides refer to the recrystallized products. The structures of the
compounds were confirmed by routine spectrometric analyses.
Only spectra for compounds not previously described are given.
Melting points were determined on a Gallenkamp melting point
apparatus in open glass capillary tubes and are uncorrected. The
infrared spectra were recorded on a Perkin–Elmer Spectrum One
FT spectrophotometer and band positions are given in reciprocal
centimeters (cmꢀ1). 1H NMR spectra were recorded on either a
Varian XL 200 spectrometer (90 MHz) using TMS as an internal stan-
dard and CDCl3 as the solvent or a Varian Mercury-VX spectrometer
operating at 300 MHz using CDCl3 or [D6]DMSO (where indicated)
as the solvent. Chemical shifts (d) are reported in ppm relative to
either TMS or the residual nondeuterated solvent resonance
(CDCl3, d=7.26 ppm; [D6]DMSO, d=2.50 ppm; CD3OD, d=
3.30 ppm) as internal references. Coupling constants (J) are given
in Hz. 13C NMR spectra were recorded on a Varian Mercury-VX spec-
trometer operating at 75 MHz using CD3OD as solvent (d=47.8).
(R,S)-N-Methyl-1-(1-naphthyloxy)propan-2-amine Hydrochloride
(21b·HCl): The title compound was prepared from 18b using the
procedure described in the preparation of 20a·HCl (See Supporting
Information). Compound 21b (56%): 1H NMR (300 MHz): d=1.26
(d, J=6.5 Hz, 3H, CH3C), 2.54 (s, 3H, CH3N), 3.17 (sextet, 1H, CH),
3.98–4.17 (m, 2H, CH2), 6.82 (d, J=7.3 Hz, 1H, Ar), 7.36 (t, 1H, Ar),
7.40–7.52 (m, 4H, Ar, NH), 7.75–7.85 (m, 1H, Ar), 8.20–8.30 ppm (m,
1H, Ar); IR (neat): n˜ =3331 cmꢀ1 (NH); MS (70 eV): m/z (%): 215
[M]+ (7), 58 (100). Salt 21b·HCl (95%): mp: 146–1488C (abs EtOH/
Et2O); 1H NMR (300 MHz, [D6]DMSO): d=1.43 (d, J=6.7 Hz, 3H,
CH3C), 2.63 (s, 3H, CH3N), 3.68 (br s, 1H, CH), 4.30 (dd, J=10.6,
5.5 Hz, 1H, CHH), 4.41 (dd, J=10.6, 3.7 Hz, 1H, CHH), 6.98 (d, J=
7.6 Hz, 1H, Ar), 7.41 (t, 1H, Ar), 7.45–7.55 (m, 3H, Ar), 7.80–7.90 (m,
1H, Ar), 8.35–8.45 (m, 1H, Ar), 9.4 ppm (br d, 2H, NH2); 13C NMR
(75 MHz, CD3OD): d=12.6 (1C), 30.3 (1C), 54.7 (1C), 67.9 (1C), 105.4
(1C), 121.3 (1C), 121.7 (1C), 125.3 (1C), 125.5 (1C), 125.7 (1C), 126.5
(1C), 127.4 (1C), 134.9 (1C), 153.7 ppm (1C); Anal. calcd for
C14H17NO·0.50HCl·0.75H2O: C 68.07, H 7.75, N 5.67, found: C 67.70,
H 7.68, N 5.61.
1
Enantiomeric excess (ee) values were determined by chiral H NMR
analyses, which were performed on free amines recovered by ex-
traction of a sample of the corresponding hydrochloride salts and
dissolved with 1.5 equiv of chiral solvating agent in CDCl3. (+)-(S)-
O-Acetylmandelic acid was used to determine the ee values of (R)-
and (S)-22b, while (+)-(S)-O-p-chlorophenylmandelic acid was used
for (R)- and (S)-31a,b. Spectra were recorded at 258C and the split-
ting of the signal of the aromatic HC-2 for (R)- and (S)-31a (Dd=
~0.069), (R)- and (S)-31b (Dd=~0.089), and HC-6 for (R)- and (S)-
22b (Dd=~0.077) were observed. EIMS spectra were recorded on
a Hewlett–Packard 6890–5973 MSD gas chromatograph/mass spec-
trometer at low resolution. The molecular ion is given as [M]+. Ele-
mental analyses were performed on a Eurovector Euro EA 3000 an-
alyzer and the data for C, H, N were within ꢁ0.4 of theoretical
values. Optical rotations were measured on a Perkin–Elmer (Nor-
walk, CT) Mod 341 spectropolarimeter; concentrations are ex-
pressed in g100 mLꢀ1, and the cell length was 1 dm, thus [a]2D0
values are given in units of 10ꢀ1 deg cm2 gꢀ1. Chromatographic
separations were performed on silica gel columns by flash chroma-
tography (Kieselgel 60, 0.040–0.063 mm, Merck, Darmstadt, Germa-
ny) using the technique described by Still et al.[40] TLC analyses
were performed on precoated silica gel on aluminum sheets (Kie-
selgel 60 F254, Merck). TLC plates were visualized under UV light
and/or in a iodine chamber. The purity of the final compounds was
determined by elemental analysis.
(R,S)-N-Methyl-2-(1-naphthyloxy)propan-1-amine Hydrochloride
(22b·HCl): The title compound was prepared from 19b using the
procedure described in the preparation of 20a·HCl (See Supporting
Information). Compound 22b (97%): 1H NMR (300 MHz): d=1.40
(d, J=6.1 Hz, 3H, CH3C), 1.69 (s, 1H, NH), 2.51 (s, 3H, CH3N), 2.86
(dd, J=12.4, 3.6 Hz, 1H, CHH), 3.04 (dd, J=12.4, 7.7 Hz, 1H, CHH),
4.72–4.85 (m, 1H, CH), 6.90 (d, J=7.7 Hz, 1H, Ar), 7.37 (t, 1H, Ar),
7.38–7.54 (m, 3H, Ar), 7.76–7.84 (m, 1H, Ar), 8.22–8.32 ppm (m, 1H,
Ar); MS (70 eV): m/z (%): 215 [M]+ (20), 72 (100). Salt 22b·HCl
(54%): mp: 203–2058C (abs EtOH/Et2O); 1H NMR (300 MHz, CD3OD):
d=1.40 (d, J=6.3 Hz, 3H, CH3C), 2.82 (s, 3H, CH3N), 3.38–3.56 (m,
2H, CH2), 4.98–5.10 (m, 1H, CH), 7.07 (d, J=7.4 Hz, 1H, Ar), 7.42 (t,
1H, Ar), 7.45–7.55 (m, 3H, Ar), 7.78–7.86 (m, 1H, Ar), 8.26–8.34 ppm
(m, 1H, Ar); 13C NMR (75 MHz, CD3OD): d=15.8 (1C), 33.1 (1C), 53.6
(1C), 70.3 (1C), 107.6 (1C), 121.4 (1C), 122.0 (1C), 125.3 (1C), 125.7
(1C), 126.4 (1C), 126.7 (1C), 127.4 (1C), 135.1 (1C), 152.2 ppm (1C);
Anal. calcd for C14H17NO·HCl·0.50H2O: C 64.48, H 7.34, N 5.37,
found: C 64.69, H 6.94, N 5.55.
(R,S)-1-(2,3-Dichlorophenoxy)-N-methylpropan-2-amine Hydro-
chloride (21a·HCl): The title compound was prepared from 18a
using the procedure described for the preparation of 20a·HCl (See
Supporting Information). Compound 21a (72%): MS (70 eV): m/z
(%): 232 [Mꢀ1]+ (<1), 58 (100). Salt 21a·HCl (60%): mp: 138–
(+)-(S)-N-Methyl-2-(1-naphthyloxy)propan-1-amine Hydrochlo-
ride [(+)-(S)-22b·HCl]: The title compound was prepared from (S)-
19b using the procedure described in the preparation of 20a·HCl
(See Supporting Information). Compound (+)-(S)-22b (83%):
[a]2D0 =+53 (c=2.5, CHCl3); 1H NMR (300 MHz): d=1.39 (d, J=
6.0 Hz, 3H, CH3C), 1.67 (br s, 1H, NH), 2.51 (s, 3H, CH3N), 2.86 (dd,
J=12.4, 3.6 Hz, 1H, CHH), 3.04 (dd, J=12.4, 7.7 Hz, 1H, CHH), 4.72–
4.88 (m, 1H, CH), 6.90 (dd, J=7.4, 0.8 Hz, 1H, Ar), 7.35–7.55 (m, 4H,
Ar), 7.75–7.85 (m, 1H, Ar), 8.20–8.30 ppm (m, 1H, Ar); MS (70 eV):
m/z (%): 215 [M]+ (14), 72 (100). Salt (+)-(S)-22b·HCl (58%): mp:
1
1408C (abs EtOH/Et2O); H NMR (300 MHz, [D6]DMSO): d=1.37 (d,
J=6.7 Hz, 3H, CH3C), 2.63 (s, 3H, CH3N), 3.54–3.67 (m, 1H, CH), 4.34
(d, J=4.9 Hz, 2H, CH2), 7.20 (d, J=8.1 Hz, 1H, Ar), 7.26 (d, J=
7.8 Hz, 1H, Ar), 7.35 (t, 1H, Ar), 9.33 ppm (br s, 2H, NH2); 13C NMR
(75 MHz, CD3OD): d=12.7 (1C), 30.7 (1C), 54.7 (1C), 69.4 (1C), 112.7
(1C), 123.5 (1C), 128.1 (1C), 129.1 (1C), 133.6 (1C), 155.0 ppm (1C);
ChemMedChem 2010, 5, 696 – 704
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
701

G. Lentini et al.
MED
183–1848C (abs EtOH/Et2O); 92% ee (1H NMR); [a]D20 =+73 (c=2,
CH3OH); 1H NMR (300 MHz, [D6]DMSO): d=1.32 (d, J=6.0 Hz, 3H,
CH3C), 2.48 (br q, 1H, NH), 2.62 (s, 3H, CH3N), 3.25–3.45 (m, 2H,
CH2), 4.95–5.15 (m, 1H, CH), 7.09 (d, J=7.7 Hz, 1H, Ar), 7.38–7.58
(m, 4H, Ar), 7.82–7.90 (m, 1H, Ar), 8.25–8.35 (m, 1H, Ar), 9.12 ppm
(br s, 1H, NH); 13C NMR (75 MHz, CD3OD): d=15.8 (1C), 33.0 (1C),
53.6 (1C), 70.3 (1C), 107.6 (1C), 121.5 (1C), 122.0 (1C), 125.3 (1C),
125.7 (1C), 126.4 (1C), 126.7 (1C), 127.4 (1C), 135.1 (1C), 152.2 ppm
(1C); Anal. calcd for C14H17NO·HCl: C 66.79, H 7.21, N, 5.56, found:
C 66.63, H 7.37, N 5.58.
1H, Ar), 7.11 ppm (t, 1H, Ar); spectrometric data were in agree-
ment with those reported for the (S)-enantiomer. Salt (ꢀ)-(R)-
31a·HCl (67%): mp: 191–1928C (abs EtOH/Et2O); 98% ee (1H NMR);
[a]2D0 =ꢀ23 (c=2.5, CH3OH); 1H NMR (300 MHz, CD3OD): d=1.92–
2.40 (m, 4H, NCH2CH2CH2), 3.32–3.48 (m, 2H, CH2N), 4.08–4.18 (m,
1H, CH), 4.23 (dd, J=10.5, 7.2 Hz, 1H, CHHO), 4.44 (dd, J=10.5,
3.0 Hz, 1H, CHHO), 7.10 (dd, J=8.2, 1.5 Hz, 1H, Ar), 7.19 (dd, J=
8.2, 1.5 Hz, 1H, Ar), 7.29 ppm (t, 1H, Ar); 13C NMR (75 MHz, CD3OD):
d=24.1 (1C), 26.5 (1C), 46.3 (1C), 58.8 (1C), 68.4 (1C), 111.8 (1C),
121.4 (1C), 123.1 (1C), 128.1 (1C), 133.6 (1C), 155.0 ppm (1C); Anal.
calcd for C11H13NOCl2·HCl: C 46.75, H 4.99, N 4.96, found: C 46.51, H
4.97, N 4.96.
(ꢀ)-(R)-N-Methyl-2-(1-naphthyloxy)propan-1-amine Hydrochlo-
ride [(ꢀ)-(R)-22b·HCl]: The title compound was prepared from (R)-
19b using the procedure described in the preparation of 20a·HCl
(See Supporting Information). Compound (ꢀ)-(R)-22b (93%):
[a]2D0 =ꢀ62 (c=2.6, CHCl3); 1H NMR (300 MHz): d=1.40 (d, J=
6.0 Hz, 3H, CH3C), 1.92 (br s, 1H, NH), 2.51 (s, 3H, CH3N), 2.86 (dd,
J=12.4, 3.5 Hz, 1H, CHH), 3.05 (dd, J=12.4, 7.8 Hz, 1H, CHH), 4.72–
4.88 (m, 1H, CH), 6.90 (d, J=7.4 Hz, 1H, Ar), 7.35–7.55 (m, 4H, Ar),
7.75–7.85 (m, 1H, Ar), 8.20–8.30 ppm (m, 1H, Ar); MS (70 eV): m/z
(%): 215 [M]+ (14), 72 (100). Salt (ꢀ)-(R)-22b·HCl (52%): mp: 184–
1858C (abs EtOH/Et2O); 97% ee (1H NMR); [a]D20 =ꢀ70 (c=2,
CH3OH); 1H NMR (300 MHz, CD3OD): d=1.39 (d, J=6.0 Hz, 3H,
CH3C), 2.82 (s, 3H, CH3N), 3.38–3.55 (m, 2H, CH2), 4.98–5.12 (m, 1H,
CH), 7.08 (d, J=7.4 Hz, 1H, Ar), 7.42 (t, 1H, Ar), 7.45–7.55 (m, 3H,
Ar), 7.78–7.88 (m, 1H, Ar), 8.28–8.36 ppm (m, 1H, Ar); 13C NMR
(75 MHz, CD3OD): d=15.8 (1C), 33.1 (1C), 53.6 (1C), 70.3 (1C), 107.6
(1C), 121.4 (1C), 122.0 (1C), 125.3 (1C), 125.7 (1C), 126.4 (1C), 126.7
(1C), 127.4 (1C), 135.1 (1C), 152.2 ppm (1C); Anal. calcd for
C14H17NO·HCl·0.17H2O: C 66.00, H 7.25, N 5.50, found: C 66.14, H
6.92, N 5.61.
(+)-(S)-2-[(1-Naphthyloxy)methyl]pyrrolidine Hydrochloride [(+)-
(S)-31b·HCl]: The title compound was prepared from (S)-30b using
the procedure described in the preparation of (+)-(S)-31a·HCl.
1
Compound (+)-(S)-31b (93%): [a]2D0 =+6.0 (c=2.1, CHCl3); H NMR
(300 MHz): d=1.60–2.20 (m, 4H, NCH2CH2CH2, 1H, NH), 2.90–3.20
(m, 2H, CH2N), 3.67 (quintet, 1H, CH), 4.03–4.15 (m, 2H, CH2O), 6.82
(dd, J=7.3, 0.9 Hz, 1H, Ar), 7.36 (t, 1H, Ar), 7.38–7.55 (m, 3H, Ar),
7.75–7.85 (m, 1H, Ar), 8.24–8.34 ppm (m, 1H, Ar); spectrometric
data were in agreement with the (R)-enantiomer and with the liter-
ature.[32] Salt (+)-(S)-31b·HCl (67%): mp: 227–2298C (abs EtOH/
Et2O); >98% ee (1H NMR); [a]D20 =+52 (c=2.2, CH3OH); 1H NMR
(300 MHz, CD3OD): d=1.92–2.42 (m, 4H, NCH2CH2CH2), 3.43 (t, J=
7.2 Hz, 2H, CH2N), 4.14–4.25 (m, 1H, CH), 4.30 (dd, J=10.5, 8.5 Hz,
1H, CHHO), 4.51 (dd, J=10.5, 3.3 Hz, 1H, CHHO), 6.97 (d, J=7.5 Hz,
1H, Ar), 7.40 (t, 1H, Ar), 7.45–7.56 (m, 3H, Ar), 7.78–7.87 (m, 1H,
Ar), 8.28–8.36 ppm (m, 1H, Ar); 13C NMR (75 MHz, CD3OD): d=23.9
(1C), 26.5 (1C), 46.0 (1C), 59.4 (1C), 67.1 (1C), 105.0 (1C), 121.2 (1C),
121.6 (1C), 125.3 (1C), 125.4 (1C), 125.7 (1C), 126.4 (1C), 127.4 (1C),
134.9 (1C), 153.7 ppm (1C); Anal. calcd for C15H17NO·HCl: C 68.30, H
6.88, N 5.31, found: C, 67.98, H 6.63, N 5.24.
(+)-(S)-2-[(2,3-Dichlorophenoxy)methyl]pyrrolidine Hydrochlo-
ride [(+)-(S)-31a·HCl]: A solution of (S)-30a (1.10 g, 3.18 mmol) in
98% formic acid (26 mL) was stirred at 08C for 3 h and then at RT
for 2.5 h. The solvent was removed in vacuo and the residue neu-
tralized by addition of saturated aq K2CO3 and extracted with
EtOAc (3ꢁ30 mL). The combined extracts were dried (Na2SO4), fil-
tered and concentrated in vacuo to give (+)-(S)-31a as a yellow oil
(0.56 g, 72%): [a]2D0 =+6.4 (c=2.6, CHCl3); IR (neat): n˜ =3346 cmꢀ1
(ꢀ)-(R)-2-[(1-Naphthyloxy)methyl]pyrrolidine Hydrochloride [(ꢀ)-
(R)-31b·HCl]: The title compound was prepared from (R)-30b
using the procedure described in the preparation of (+)-(S)-
31a·HCl. Compound (ꢀ)-(R)-31b (92%): [a]2D0 =ꢀ9.0 (c=2.2,
1
CHCl3); H NMR (300 MHz): d=1.64–2.12 (m, 4H, NCH2CH2CH2, 1H,
NH), 2.95–3.17 (m, 2H, CH2N), 3.68 (quintet, 1H, CH), 4.03–4.16 (m,
2H, CH2O), 6.82 (d, J=7.4 Hz, 1H, Ar), 7.36 (t, 1H, Ar), 7.38–7.54 (m,
3H, Ar), 7.76–7.84 (m, 1H, Ar), 8.22–8.32 ppm (m, 1H, Ar); MS
(70 eV): m/z (%): 227 [M]+ (4), 70 (100). Salt (ꢀ)-(R)-31b·HCl (80%):
mp: 226–2278C (abs EtOH/Et2O); >98% ee (1H NMR); [a]D20 =ꢀ59
1
(NH); H NMR (300 MHz): d=1.55–2.05 (m, 4H, NCH2CH2CH2), 2.11
(s, 1H, exch D2O, NH), 2.90–3.16 (m, 2H, NCH2), 3.50–3.68 (m, 1H,
CH), 3.88–4.08 (m, 2H, CH2O), 6.78–6.92 (m, 1H, Ar), 7.02–7.20 ppm
(m, 2H, Ar); MS (70 eV): m/z (%): 244 [Mꢀ1]+ (<1), 70 (100). The
salt [(+)-(S)-31a·HCl] was obtained by dissolving the free base in a
small amount of aq HCl (2n) and removing H2O azeotropically. The
crude solid was recrystallized from EtOH/Et2O to afford (+)-(S)-
31a·HCl as a white solid (0.49 g, 55%): mp: 192–1938C (abs EtOH/
Et2O); 98% ee (1H NMR); [a]D20 =+23 (c=2.3, CH3OH); 1H NMR
(300 MHz, CD3OD): d=1.92–2.40 (m, 4H, NCH2CH2CH2), 3.34–3.50
(m, 2H, NCH2), 4.05–4.18 (m, 1H, CH), 4.23 (dd, J=10.5, 7.2 Hz, 1H,
CHHO), 4.45 (dd, J=10.5, 3.3 Hz, 1H, CHHO), 7.10 (dd, J=8.3,
1.4 Hz, 1H, Ar), 7.19 (dd, J=8.3, 1.4 Hz, 1H, Ar), 7.29 ppm (t, 1H,
Ar); 13C NMR (75 MHz, CD3OD): d=24.1 (1C), 26.5 (1C), 46.3 (1C),
58.8 (1C), 68.4 (1C), 111.8 (1C), 121.4 (1C), 123.1 (1C), 128.1 (1C),
133.6 (1C), 155.0 ppm (1C); Anal. calcd for C11H13NOCl2·HCl: C 46.75,
H 4.99, N 4.96, found: C 46.70, H 4.86, N 4.99.
1
(c=2.2, CH3OH); H NMR (300 MHz, CD3OD): d=1.92–2.42 (m, 4H,
NCH2CH2CH2), 3.43 (t, J=7.2 Hz, 2H, CH2N), 4.14–4.25 (m, 1H, CH),
4.30 (dd, J=10.5, 8.5 Hz, 1H, CHHO), 4.52 (dd, J=10.5, 3.3 Hz, 1H,
CHHO), 6.98 (d, J=7.4 Hz, 1H, Ar), 7.40 (t, 1H, Ar), 7.45–7.56 (m,
3H, Ar), 7.78–7.87 (m, 1H, Ar), 8.28–8.36 ppm (m, 1H, Ar); 13C NMR
(75 MHz, CD3OD): d=23.9 (1C), 26.5 (1C), 46.0 (1C), 59.4 (1C), 67.1
(1C), 105.0 (1C), 121.2 (1C), 121.5 (1C), 125.3 (1C), 125.4 (1C), 125.7
(1C), 126.4 (1C), 127.4 (1C), 134.9 (1C), 153.7 ppm (1C); Anal. calcd
for C15H17NO·HCl: C 68.30, H 6.88, N 5.31, found: C 68.29, H 7.03, N
5.32.
Physicochemical data
(ꢀ)-(R)-2-[(2,3-Dichlorophenoxy)methyl]pyrrolidine Hydrochlo-
ride [(ꢀ)-(R)-31a·HCl]: The title compound was prepared from (R)-
30a using the procedure described in the preparation of (+)-(S)-
31a·HCl. Compound (ꢀ)-(R)-31a (97%): [a]2D0 =ꢀ8.0 (c=3, CHCl3);
1H NMR (300 MHz): d=1.53–2.06 (m, 4H, NCH2CH2CH2), 2.11 (s, 1H,
NH), 2.88–3.10 (m, 2H, NCH2), 3.50–3.62 (m, 1H, CH), 3.87–4.03 (m,
2H, CH2O), 6.82 (dd, J=8.0, 1.6 Hz, 1H, Ar), 7.05 (dd, J=8.2, 1.6 Hz,
Physicochemical data of compounds shown in Table 1 were ob-
tained by a pH-metric technique using a GlpKa apparatus (Sirius
Analytical Instruments Ltd., Forrest Row, East Sussex, UK).[41–44] Be-
cause of the low solubility of the investigated compounds in aque-
ous medium, methanol was used as a cosolvent for pKa measure-
ments. Three separate solutions of test compound in CH3OH/H2O
(10–30% w/w) were prepared (concentration ~10ꢀ5 m) and subse-
702
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 696 – 704

Chiral Aryloxyalkylamines
[3H]Cytisine binding
quently acidified to pH 4 using aq HCl (0.5m). The solutions were
then titrated with aq KOH (0.5m) to pH 12. Initial pKa values, which
are the apparent ionization constants relative to the mixture of the
solvents, were obtained by Bjerrum Plot, that is, the curve ob-
tained by the difference between the curve of titration of the ioniz-
able substance and that of the blank solution. These values were
then optimized by a weighted nonlinear least-squares procedure
(Refinement Pro 1.0 software) to obtain pKa values in the absence
of cosolvent, by extrapolation using the Yasuda–Shedlovsky equa-
tion.[45]
To obtain LogP data, at least three separate titrations were per-
formed on each compound. The concentration of the analyte was
approximately 10ꢀ5 m, in mixtures of H2O (7.5 mL) and n-octanol,
(0.1–10 mL). The biphasic solutions were acidified to pH 4 with aq
HCl (0.5m) and then titrated with aq KOH (0.5m) to pH 12. The re-
sults were optimized as described above, and the average of these
data gave the LogP value for each compound.[42,43] All titrations
were carried out at 25ꢁ0.18C under N2 atmosphere to exclude
CO2.
[3H]Cytisine binding assays were performed as described by Ji
et al.[47] with few adjustments. Samples containing 100–200 mg of
protein from frozen mouse brains, 0.75 nm [3H]cytisine and the test
compounds were incubated in 50 mm Tris-HCl buffer (containing
120 mm NaCl, 5 mm KCl, 2 mm CaCl2, 2 mm MgCl2, pH 7.4) in a final
volume of 500 mL for 75 min at 48C. For equilibrium competition
binding assays, the concentration of unlabeled compounds varied
from 10ꢀ10 m to 10ꢀ3 m. Bound radioactivity was isolated by
vacuum filtration onto presoaked GF/B using a 96-well filtration ap-
paratus, and were then rapidly rinsed with 2 mL of the same ice-
cold buffer. MicroScint-20 scintillation cocktail was added to each
well and radioactivity determined using a Perkin–Elmer TopCount
instrument. The nonspecific binding was measured in the presence
of 10 mm nicotine bitartrate.
Isolated guinea pig ileum assay
Guinea pigs were anesthetized, decapitated and the proximal
ileum removed. The intestine was carefully flushed several times
with warm Krebs–Henseleit solution (118 mm NaCl, 25 mm
NaHCO3, 4.7 mm KCl, 0.6 mm MgSO4, 1.2 mm KH2PO4, 1.2 mm CaCl2,
11.2 mm glucose, pH 7.4). Whole ileal segments (~3 cm in length)
were suspended under 1.0 g tension in Krebs solution gassed with
95% O2 and 5% CO2 and maintained at 378C. Following a litera-
ture procedure[29] with minor modification, the bath organ medium
contained 1 mm atropine to antagonize cholinergically-mediated
contractions due to activation of 5-HT3 and 5-HT4 receptors, 1 mm
ketanserin to block 5-HT2A receptors, 1 mm prazosin to block a1
adrenoreceptors and 1 mm indomethacin to preserve the organ in-
tegrity. Changes in tension of the tissue were recorded by Fort 10
Original WPI isometric transducer (2Biological Instruments; Besoz-
zo, Italy) connected to a PowerLab/400 workstation. Tissue re-
sponses were recorded as gram changes in isometric tension and
expressed as percentage of reduction in the height of the contrac-
tion. Tissue was contracted by substance P (200 nm). This value
was preliminary determined by concentration–response curves (1–
200 nm). 200 nm substance P elicited 80% of maximum contrac-
tion. The reference compound, sumatriptan, was added 5 min
before substance P addition and noncumulative concentration–
response curves were constructed (0.5–5 mm). Sumatriptan induced
relaxation with maximal response (52%) at 3 mm. Analogously,
each test compound (0.5–3 mm) was added 5 min before substan-
ce P addition and a noncumulative concentration–response curve
was constructed. The study of agonists was consequently repeated
after an equilibration time (75 min) in the presence of 3 mm suma-
triptan in order to obtain the desensitization of 5-HT1D receptors
and to establish the contribution of other receptors.
Pharmacology
Human cloned 5-HT1B receptor proteins and [3H]-5-CT were pur-
chased from PerkinElmer Life Science (Monza, Italy), 5-CT was ob-
tained from Tocris (Bristol, UK), [3H]Sumatriptan was purchased
from Amersham (Piscataway, USA), sumatriptan was purchased
from Kemprotec Ltd (Middlesbrough, UK).
Guinea pigs used in the isolated guinea pig ileum assay were ob-
tained from Harlan (San Pietro al Natisone, Italy). Each animal
weighed 250–300 g. Male Swiss albino mice (24–26 g) from Morini
(San Polo d’Enza, Italy) were used in the hot-plate assay. All experi-
ments were carried out in accordance with the NIH Guide for the
Care and Use of Laboratory animals. All efforts were made to mini-
mize animal suffering, and to reduce the number of animals used.
Serotonergic h5-HT1B receptor binding assay
This experiment was carried out as described by Sternfeld et al.
with minor modifications.[20] The test compound at several concen-
trations (from 10ꢀ10 m to 10ꢀ5 m), [3H]-5-CT (0.56 nm), and human
cloned receptor proteins were placed in incubation buffer (50 mm
Tris·HCl, EDTA 1 mm, 12.5 mm MgCl2, 0.1% ascorbic acid, pH 7.7) to
give a final volume of 500 mL. The samples were incubated for
60 min at 258C. The suspension was filtered on GF/B filters (pre-
soaked in PEI for 30 min) washing twice with the same incubation
buffer. The nonspecific binding was determined in the presence of
10 mm 5-CT.
Serotonergic 5-HT1D receptor binding assay
Guinea-pig striatum membranes and experimental conditions were
as described by Audinot et al. with minor modifications.[46] The test
compound at several concentrations (from 10ꢀ10 m to 10ꢀ5 m),
[3H]sumatriptan (1.80 nm), and striatum membranes were placed in
incubation buffer (50 mm Tris·HCl, 10 mm Pargylin, 4 mm CaCl2,
0.1% ascorbic acid, pH 7.7) to give a final volume of 500 mL. The
samples were incubated for 1 h at 258C in the presence of 8-OH-
DPAT (1 mm) and mesulergine (10 mm) to mask 5-HT1A, D2 and 5-HT2
receptors, respectively. The suspension was filtered through glass
fiber filter plates (GF/B), presoaked in PEI for 30 min, washing twice
with the same incubation buffer. The nonspecific binding was de-
termined in the presence of 10 mm sumatriptan.
Hot-plate test[48]
Fifteen male Swiss albino mice (24–26 g) were housed per cage.
The cages were placed in the experimental room 24 h before the
test for acclimatization. The animals were fed a standard laboratory
diet and tap water ad libitum and kept at 23ꢁ18C with a 12 h
light/dark cycle (lights on at 07:00). Mice were placed inside a
stainless steel container, which was set thermostatically at 52.5ꢁ
0.18C in a precision water bath (KW Mechanical Workshop, Siena,
Italy). Reaction times (s) were measured with a stopwatch before
and 15, 30, 45, and 60 min after administration of the analgesic
drug. The endpoint used was the licking of the fore or hind paws.
ChemMedChem 2010, 5, 696 – 704
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
703

G. Lentini et al.
MED
[21] R. C. Glen, G. R. Martin, A. P. Hill, R. M. Hyde, P. M. Woollard, J. A. Salmon,
J. Buckingham, A. D. Robertson, J. Med. Chem. 1995, 38, 3566–3580.
[22] G. Van Lommen, M. De Bruyn, M. Schroven, W. Verschueren, W. Jans-
sens, J. Verrelst, J. Leysen, Bioorg. Med. Chem. Lett. 1995, 5, 2649–2654.
[23] O. Mitsunobu, Synthesis 1981, 1–28.
[24] A. Carocci, A. Catalano, F. Corbo, A. Duranti, R. Amoroso, C. Franchini, G.
Lentini, V. Tortorella, Tetrahedron: Asymmetry 2000, 11, 3619–3634.
[25] E. Roubini, R. Laufer, C. Gilon, Z. Selinger, B. P. Roques, M. Cohrev, J.
Med. Chem. 1991, 34, 2430–2438.
[26] P. Depreux, D. Lesieur, H. A. Mansour, P. Morgan, H. E. Howell, P. Renard,
D.-H. Caignard, B. Pfeiffer, P. Delagrange, B. Guardiola, S. Yous, A. De-
marque, G. Adam, J. Andrieux, J. Med. Chem. 1994, 37, 3231–3239.
[27] D. Rance, N. Clear, L. Dallman, E. Llewellyn, J. Nuttall, H. Verrier, Head-
ache [abstract] 1997, 37, 328.
Those mice scoring less than 12 and more than 18 s in the pretest
were rejected (30%). An arbitrary cut-off time of 45 s was adopted.
Statistical analysis
All experimental results are given as the mean ꢁSEM (standard
error of the mean). Analysis of variance ANOVA, followed by Fish-
er’s protected least significant difference procedure for post-hoc
comparison, were used to verify significance between two means.
Data were analyzed with the StatView software for Macintosh.
P values of less than 0.05 were considered significant.
[28] T. A. Barf, P. de Boer, H. Wikstrçm, J. Med. Chem. 1996, 39, 4717–4726.
[29] M. Leopoldo, E. Lacivita, P. De Giorgio, C. Fracasso, S. Guzzetti, S. Caccia,
M. Contino, N. A. Colabufo, F. Berardi, R. Perrone, J. Med. Chem. 2008,
51, 5813–5822.
[30] R. L. Elliott, H. Kopecka, D. E. Gunn, N. H. Lin, D. S. Garvey, K. B. Ryther,
M. W. Holladay, D. J. Anderson, J. E. Campbell, J. P. Sullivan, M. J. Buckley,
K. L. Gunther, A. B. O’Neill, M. W. Decker, S. P. Arneric’, Bioorg. Med.
Chem. Lett. 1996, 6, 2283–2288.
Acknowledgements
This work was accomplished thanks to the financial support of
the Ministero dell’Istruzione, dell’Universitꢀ e della Ricerca (MIUR).
The authors thank Prof. Roberto Perrone for helpful input in the
early stages of the project.
[31] M. Williams, E. A. Kowaluk, S. P. Arneric, J. Med. Chem. 1999, 42, 1481–
1500.
Keywords: analgesia · chirality · receptors · serotonin · triptans
[32] A. Carocci, A. Duranti, G. Lentini, N. Tangari, B. Stankov, V. Lucini, R.
Nonno, F. Fraschini, V. Tortorella, Book of Abstracts, Italian Chemical So-
ciety (SCI) Meeting, Torino (Italy), September 1997, A15.
[33] T. Anger, D. J. Madge, M. Mulla, D. Riddall, J. Med. Chem. 2001, 44, 115–
137.
[1] N. M. Barnes, T. Sharp, Neuropharmacology 1999, 38, 1083–1152.
[2] D. Hoyer, J. P. Hannon, G. R. Martin, Pharmacol. Biochem. Behav. 2002,
71, 533–554.
[3] F. G. Boess, I. L. Martin, Neuropharmacology 1994, 33, 275–317.
[4] D. Hoyer, D. E. Clarke, J. R. Fozard, P. R. Hartig, G. R. Martin, E. J. Myle-
charane, P. R. Saxena, P. P. Humphrey, Pharmacol. Rev. 1994, 46, 157–
203.
[5] P. R. Hartig, D. Hoyer, P. P. A. Humprey, G. R. Martin, Trends Pharmacol.
Sci. 1996, 17, 103–105.
[34] R. K. Portenoy, J. Pain Symptom Manage. 2000, 19 (Suppl. 1), 16–20.
[35] K. Huang, M. Ortiz-Marciales, V. Stepanenko, M. De Jesuꢃs, W. Correa, J.
Org. Chem. 2008, 73, 6928–6931.
[36] D. G. Loughhead, L. A. Flippin, R. J. Weikert, J. Org. Chem. 1999, 64,
3373–3375.
[37] K. Kingwell, Nat. Rev. Drug Discovery 2009, 8, 931.
[6] M. W. Hamblin, M. A. Metcalf, Mol. Pharmacol. 1991, 40, 143–148.
[7] H. Jin, D. Oksenberg, A. Ashkenazi, S. J. Peroutka, A. M. Duncan, R. Roz-
mahel, Y. Yang, G. Mengod, J. M. Palacios, B. F. O’Dowd, J. Biol. Chem.
1992, 267, 5735–5738.
[8] I. Bouchelet, Z. Cohen, B. Case, P. Sꢂguela, E. Hamel, Mol. Pharmacol.
1996, 50, 219–223.
[9] P. J. Goadsby, Trends Mol. Med. 2007, 13, 39–44.
[10] I. Szelenyi, Drug News Perspect. 1989, 2, 45–49.
[11] J. A. Lopez-Garcia, Curr. Top. Med. Chem. 2006, 6, 1987–1996.
[12] R. D. Jeggo, Y. Wang, D. Jordan, A. G. Ramage, Br. J. Pharmacol. 2007,
150, 987–995.
[38] F. Lovering, J. Bikker, C. Humblet, J. Med. Chem. 2009, 52, 6752–6756.
[39] G. Carbonara, A. Carocci, G. Fracchiolla, C. Franchini, G. Lentini, F. Loio-
dice, P. Tortorella, ARKIVOC 2004, Part (v), 5–25.
[40] W. C. Still, M. Kahn, A. Mitra, J. Org. Chem. 1978, 43, 2923–2925.
[41] J. E. Comer, K. Y. Tam in Pharmacokinetic Optimization in Drug Research
(Eds.: B. Testa, H. van de Waterbeemd, G. Folkers, R. H. Guy), Wiley-VCH,
Weinheim, 2001, pp. 275–304.
[42] A. Avdeef, Quant. Struct.-Act. Relat. 1992, 11, 510–517.
[43] A. Avdeef, J. Pharm. Sci. 1993, 82, 183–190.
ˇ
[44] A. Avdeef in Lipophilicity in Drug Action and Toxicology (Eds.: V. Pilska, B.
Testa, H. van de Waterbeemd), VCH Publishers, Weinheim, 1996,
pp. 109–139.
[13] E. Hamel, Serotonin 1996, 1, 19–29.
[14] A. Ferro, J. Longmore, R. G. Hill, M. J. Brown, Br. J. Clin. Pharmacol. 1995,
40, 245–251.
[15] J. Longmore, C. M. Boulanger, B. Desta, R. G. Hill, W. N. Schofield, A. A.
Taylor, Br. J. Clin. Pharmacol. 1996, 42, 431–441.
[45] A. Avdeef, K. J. Box, J. E. Comer, M. Gilges, M. Hadley, C. Hibbert, W. Pat-
terson, K. Y. Tam, J. Pharm. Biomed. Anal. 1999, 20, 631–641.
[46] V. Audinot, S. Lochon, A. Newman-Tancredi, G. Lavielle, M. J. Millan, Eur.
J. Pharmacol. 1997, 327, 247–256.
[16] P. P. A. Humphrey, P. J. Goadsby, Cephalalgia 1994, 14, 401–410.
[17] M. Wilkinson, V. Pfaffenrath, J. Schoenen, H.-C. Diener, T. J. Steiner, Ceph-
alalgia 1995, 15, 337–357.
[47] J. Ji, W. H. Bunnelle, D. J. Anderson, C. Faltynek, T. Dyhring, P. K. Ahring,
L. E. Rueter, P. Curzon, M. J. Buckley, K. C. Marsh, A. Kempf-Grote, M. D.
Meyer, Biochem. Pharmacol. 2007, 74, 1253–1262.
[18] A. M. Ismaiel, M. Dukat, H. Law, R. Kamboj, E. Fan, D. K. H. Lee, L. Maz-
zocco, D. Buekschkens, M. Teitler, M. E. Pierson, R. A. Glennon, J. Med.
Chem. 1997, 40, 4415–4419.
[48] J. P. O’Callaghan, S. G. Holtzman, J. Pharmacol. Exp. Ther. 1976, 197,
533–544.
[19] M. Bondarev, T. Iwamura, D. Hynd, L. Mazzocco, D. K. Lee, M. Dukat,
R. A. Glennon, Med. Chem. Res. 1998, 8, 333–342.
[20] F. Sternfeld, A. R. Guiblin, R. A. Jelley, V. G. Matassa, A. J. Reeve, P. A.
Hunt, M. S. Beer, A. Heald, J. A. Stanton, B. Sohal, A. P. Watt, L. J. Street,
J. Med. Chem. 1999, 42, 677–690.
Received: December 22, 2009
Published online on March 16, 2010
704
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 696 – 704