QSAR study for a novel series of ortho monosubstituted phenoxy analogues of α1-adrenoceptor antagonist WB4101

Article abstract of DOI:10.1016/j.bmc.2005.01.034

A number of (S)- and (R)-2-[(2-phenoxyethyl)aminomethyl]-1,4-benzodioxanes unsubstituted or ortho monosubstituted at the phenoxy moiety were synthesized and tested in binding assays on the α_1a-AR, α_1b-AR, α_1d-AR and the 5-

Full text of DOI:10.1016/j.bmc.2005.01.034

Bioorganic & Medicinal Chemistry 13 (2005) 2547–2559
QSAR study for a novel series of ortho monosubstituted
phenoxy analogues of a1-adrenoceptor antagonist WB4101
Laura Fumagalli,a Cristiano Bolchi,a Simona Colleoni,b Marco Gobbi,b
Barbara Moroni,a Marco Pallavicini,a Alessandro Pedretti,a Luigi Villa,a
Giulio Vistolia and Ermanno Valotia,*
a
`
Istituto di Chimica Farmaceutica e Tossicologica, Universita di Milano, viale Abruzzi 42, I-20131 Milano, Italy
bIstituto di Ricerche Farmacologiche ‘Mario Negri’, via Eritrea 62, I-20157 Milano, Italy
Received 5October 2004; revised 14 January 2005; accepted 21 January 2005
Abstract—A number of (S)- and (R)-2-[(2-phenoxyethyl)aminomethyl]-1,4-benzodioxanes unsubstituted or ortho monosubstituted
at the phenoxy moiety were synthesized and tested in binding assays on the a1a-AR, a1b-AR, a1d-AR and the 5-HT1A receptor. The
affinity values of the new compounds 1–16 were compared with those of the enantiomers of the 2,6-dimethoxyphenoxy analogue, the
well-known a1 antagonist WB4101, finding that the unsubstituted derivative (S)-1 and the o-methyl, the o-t-butyl, the o-fluoro and
the o-methoxy derivatives, (S)-2, (S)-4, (S)-8 and (S)-16, respectively, display a significantly specific 5-HT1A affinity, very close, with
the exception of (S)-4, to the almost nanomolar one of (S)-WB4101. Otherwise, sensible affinity decreases were recorded for the
three a1-AR subtypes. A classical quantitative structure–activity relationship (Hansch) analysis was successfully applied to com-
pounds (S)-1 to (S)-16 and (S)-WB4101 to rationalize such binding data.
Ó 2005 Elsevier Ltd. All rights reserved.
1. Introduction
ity for a1A- and, to a minor degree, for a1D-ARs with re-
spect to a1B-AR and 5-HT1A receptor.5,6 Such a lead
molecule has been the subject of intense investigations
aimed at identifying its pharmacophoric features and
improving its affinity and selectivity.6–20 As a result, a
variety of analogues have been studied over the past dec-
ades and, more recently, some of them have been char-
acterized for their a1 subtype and 5-HT1A affinity.
Already in the seventies, Kapur et al. pointed out the
contribution of the 2,6-dimethoxyphenyl residue to the
antagonistic activity of WB4101 suggesting ꢀa possible
aromatic subsite through which the dimethoxy benzene
may interactꢁ.21 In particular, they hypothesized ꢀthat
the 2,6-dimethoxy substituents play an important role
in the drug–receptor interaction either indirectly
through a conformational influence (on the phenyl ring
orientation) or directly through a receptor subsite inter-
actionꢁ.21 This would be demonstrated by the fact that
the antagonistic potency is adversely affected by the re-
moval of one or both ortho-methoxy substituents as well
as by their repositioning around the phenyl ring or their
replacement by methyl groups. More recently (1998), the
same authors carried out a QSAR study on WB4101
derivatives enclosing methoxy- and methylphenyl
The existence of multiple a1-adrenoceptor (a1-AR) sub-
types, realized within recent years,1,2 has renewed the
interest of medicinal chemists in ligands for such recep-
tors and, more specifically, in subtype selective a1 antag-
onists. In fact, current evidences, based on the
distribution of a1-ARs, indicate that subtype selective
a1 antagonists could be very useful in the treatment of
lower urinary tract symptoms.3
One of the oldest and best-known a1 antagonists is
represented by WB4101, a 2-aminomethyl-1,4-benzodi-
oxane derivative first described in 1965by Green and
co-workers4 which exhibits about nanomolar affinity to-
wards the three different a1-ARs (a1a, a1b and a1d) and
the 5-HT1A serotoninergic receptor with a slight selectiv-
Keywords: a1-Antagonist; WB4101; WB4101 Analogues; a1-Adrener-
gic receptor subtypes; 5-HT1A Serotoninergic receptor; Binding affin-
ity; QSAR analysis.
*
Corresponding author. Tel.: +39 2 50317553; fax: +39 2 50317565;
e-mail: ermanno.valoti@unimi.it
0968-0896/$ - see front matter Ó 2005Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2005.01.034

2548
L. Fumagalli et al. / Bioorg. Med. Chem. 13 (2005) 2547–2559
analogues.22 The results of their investigations con-
firmed the importance of the ring substitution though
ꢀit was not possible to elucidate the mechanism by which
substituents on that ring exert their effectꢁ due to the fact
that only methoxy and methyl substituents had been
considered.22
a1b-AR, a1d-AR and the 5-HT1A receptor. Finally, a
classical quantitative structure–activity relationship
(Hansch) study was applied, in particular, to the S iso-
mers of 1–16 in the expectation of finding significant
correlations of each of the four biological affinities with
properly chosen physicochemical parameters of the aro-
matic substituents.
OMe
+
O
O
O
*
N
H
H
2. Chemistry
MeO
Cl-
The enantiomeric pairs of compounds 1–16 were pre-
pared by alkylation of ortho substituted 2-phenoxyethyl-
amines with the enantiomers of 2-iodomethyl- or
2-mesyloxymethyl-1,4-benzodioxane or, vice versa, by
alkylation of the enantiomers of 2-aminomethyl-1,4-
benzodioxane with ortho substituted 2-phenoxyethyl
halides or mesylates.
WB 4101
The hypothesized involvement of an aromatic subsite in
the a1 antagonist interaction was successively confirmed
by Perez mutagenesis studies,23,24 which demonstrated
that two phenylalanine residues (Phe-312 and Phe-308)
in transmembrane domain 7 (TM7) of the a1-ARs are
major sites of a1-AR antagonist binding. Consistently
with these results, our recent researches,25 based on
the automated docking of WB4101 into the a1a-AR built
by comparative modeling, showed that the phenoxy
moiety of the ligand is inserted in an aromatic pocket
lined by Phe-288, Phe-289 (TM6), Phe-308, Phe-312
and Tyr-316 (TM7). In addition, our interaction model
supported the original thesis of WB4101 devisers high-
lighting a hydrogen bond between a glutamine residue
(Gln-177) in extracellular loop 2 (EL2) and one ligand
methoxy group, which mimics the norepinephrine a-hy-
droxyl. Furthermore, it is interesting to observe that our
docked WB4101 conformer displays, relatively to the 2-
phenoxyethyhlamino framework, a pronounced analogy
to the two candidate bioactive forms of the ligand we
had previously identified on the basis of a thorough
study of the conformational properties of WB4101 and
of its fitting with the more constrained structure of cory-
nanthine.19 In fact, these forms show, like the docked
conformer, the phenyl ring perpendicular, or nearly, to
the oxyethyl plane and a gauche disposition of the nitro-
gen with respect to the phenoxy oxygen giving rise to an
intramolecular hydrogen bond between this latter and
one of the ammonium protons. Therefore, a double role
would be played by the 2,6-dimethoxy substituents: (a) a
specific interaction with an aminoacidic residue of the
receptor and (b) a determinant influence on the phenyl
orientation and the nitrogen–phenoxy oxygen reciprocal
disposition.
+
O
O
O
R
*
N
H H
Cl
-
1
2
3
4
5
6
7
8
R = H
= Me
= Et
= t-Bu
= CN
= CF3
= COMe 15
= F
9
10
11
12
13
14
R = Cl
= SMe
= SOMe
= SO2Me
= NO2
= NHCOMe
= NH2
16
= OMe
As outlined in Scheme 1, the syntheses of various 2-
phenoxyethylamines and 2-phenoxyethyl halides and
sulfonates invariably started from the corresponding
ortho substituted phenols. The synthetic pathways con-
sisted in the following steps: (a) 2-hydroxyethylation
of the phenolic moiety with ethylene carbonate; (b)
tosylation of the o-ethyl, o-thiomethyl and o-acetyl
phenoxyethyl alcohols 17, 19 and 20; (c) conversion of
this latter into 2-(2-iodoethoxy)acetophenone (25); (d)
mesylation of the o-trifluoromethyl and o-nitro interme-
diates 18 and 21; (e) N-alkylation of phthalimide with
the tosyl esters 22 and 23 and the mesyl ester 26; (f)
hydrazinolysis of the N-alkylphthalimides 28–30 to give
the target 2-phenoxyethylamines 31–33; (g) oxidation of
the thioamine 33 to the corresponding sulfoxide 34; (h)
oxidation of this latter to the sulfone 35. As shown in
the same scheme, 2-(2-nitrophenoxy)ethanol (21) was
also used to synthesize 2-(2-acetamidophenoxy)ethyl-
amine (40) by the following steps: (i) reduction to the
aniline 36; (j) N-acetylation; (d) O-mesylation; (k)
aminolysis with dibenzylamine; (l) hydrogenolytic de-
benzylation to yield 40. 2-Phenoxyethylamine, the other
ortho substituted 2-phenoxyethylamines (o-CN, o-F,
o-Cl, o-MeO, o-t-BuO) and 2-(2-chloroethoxy)anisole
were prepared according to the literature methods,
while the enantiomers of 2-iodomethyl-, 2-mesyloxy-
methyl- and 2-aminomethyl-1,4- benzodioxane were ob-
tained practicing our previously reported synthetic
methods.6,19,26
In this context, where the phenoxy group and its substi-
tution pattern are generally recognized as key structural
determinants for the WB4101 affinity, it is rather sur-
prising that no ring substituents alternative to methoxyl
and methyl have been ever considered. Therefore, it
seemed worthwhile to investigate, in the first instance,
a wide series of WB4101 analogues differently ortho
monosubstituted at the phenoxy residue in order to get
a better insight into the nature of the receptor subsite
interaction of this molecular portion and, hopefully, to
modulate the affinity profile of the lead compound. To
this end, both the enantiomers of compounds 1–16 were
synthesized and tested in binding assays on the a1a-AR,

L. Fumagalli et al. / Bioorg. Med. Chem. 13 (2005) 2547–2559
2549
OH
I
OTs
O
O
OH
O
R
COCH3
R
R
a
c
b
R
R
17
18
19
20
21
= Et
= CF3
= SCH3
= COCH3
= NO2
22
23
24
= Et
= SCH3
= COCH3
25
NH2
X
O
e
d
f
f
O
33
34
35
X = SCH3
= SOCH3
= SO2CH3
g
OMs
N
h
O
O
R
O
e
R
NH2
O
R
28
29
30
= Et
= SCH3
= CF3
R
R
= Et
= CF3
31
32
R
26
27
= CF3
= NO2
NH2
Y
OH
X
O
O
O
NHAc
NHAc
i
d
l
21
36
37
38
39
X = NH2
= NHAc
Y = OMs
= NBn2
40
j
k
Scheme 1. Reagents: (a) ethylene carbonate, K2CO3, DMF or toluene or xylene; (b) TsCl, Py. (c) NaI, acetone; (d) MsCl, TEA, DCM; (e) potassium
phthalimide, DMF; (f) hydrazine, 2-methoxyethanol; (g) 30% H2O2; (h) m-CPBA, DCM; (i) H2–Pd/C, MeOH; (j) CH3COCl, TEA, DCM;
(k) NHBn2, 2-propanol; (l) H2–Pd/C, 2.5/1 EtOH–EtOAc.
The final nucleophilic substitutions leading to the S iso-
mers of 1–16 are illustrated in Scheme 2. The R isomers
of 1–16 were synthesized by the same reactions as their
antipodes, but using the 2-substituted 1,4-benzodioxane
intermediates with inverted configuration. In particular,
compounds 1–6, 8–12, 14 and 15 were prepared by N-
alkylation of ortho substituted 2-phenoxyethylamines,
whereas compounds 7, 13 and 16 by aminolysis of ortho
substituted 2-phenoxyethyl halides or mesylates. In the
case of 7, the enantiomers of 2-N-benzylaminomethyl-
1,4-benzodioxane (41) were alkylated with 25 and the
resultant tertiary amines submitted to hydrogenolysis,
while the enantiomeric pairs of 13 and 16 were directly
prepared by reaction of (S)- and (R)-2-aminomethyl-
1,4-benzodioxane with 27 and 2-(2-chloroethoxy)ani-
sole, respectively.
3. Results and discussion
Table 1 reports the affinities, expressed as pKi values, at
the three cloned human a1-AR subtypes and 5-HT1A
2-phenoxyethylamine
(S)-1
OMs
31
I
(S)-3
(S)-5
(S)-6
(S)-8
(S)-9
(S)-10
(S)-11
(S)-12
(S)-7
2-(2-cyanophenoxy)ethylamine
2-(2-methylphenoxy)ethylamine
2-(2-t-butoxyphenoxy)ethylamine
O
(S)-2
32
O
(S)-4
O
O
2-(2-fluorophenoxy)ethylamine
2-(2-chlorophenoxy)ethylamine
40
(S)-14
33
34
35
HCl
(S)-15
NH2
NH Bn
2
27
(S)-13
(S)-16
O
O
O
2-(2-chloroethoxy)anisole
NHBn
H -Pd/C
2
25
O
(S)-41
Scheme 2. Synthesis of the S isomers of compounds 1–16.

2550
L. Fumagalli et al. / Bioorg. Med. Chem. 13 (2005) 2547–2559
Table 1. Experimental and predicted affinity constants, expressed as pKi (ꢀlogKiM), of the enantiomers of compounds 1–16 and of WB4101 for
cloned human a1-adrenoceptor subtypes and 5-HT1A receptor
Phenoxy o-substituent Compd
a–1a
a–1b
a–1d
5-HT1A
pKi exptl SE pKi pred pKi exptl SE pKi pred pKi exptl SE pKi pred pKi exptl SE pKi pred
H
(S)-1
(R)-1
(S)-2
7.34 ( 0.05)
7.07 ( 0.08)
7.51 ( 0.04)
7.16 ( 0.07)
7.58 ( 0.14)
7.40 ( 0.07)
6.22 ( 0.04)
6.22 ( 0.06)
7.06 ( 0.09)
7.22 ( 0.07)
7.16 ( 0.04)
7.05( 0.28)
7.26 ( 0.04)
7.63 ( 0.06)
7.47 ( 0.10)
7.15 ( 0.05)
7.62 ( 0.07)
7.27 ( 0.07)
7.72 ( 0.10)
7.58 ( 0.14)
7.17 ( 0.07)
n.d.
7.30
7.17
6.80
6.20
6.92
6.57
6.81
7.79
7.27
7.08
7.02 ( 0.03)
6.03 ( 0.12)
7.13 ( 0.05)
6.73 ( 0.03)
7.06 ( 0.07)
6.91 ( 0.05)
6.29 ( 0.08)
5.78 ( 0.05)
6.05( 0.08)
6.72 ( 0.06)
6.99 ( 0.08)
6.72 ( 0.08)
6.73 ( 0.05)
7.28 ( 0.10)
7.27 ( 0.04)
6.57 ( 0.06)
7.24 ( 0.29)
6.73 ( 0.09)
7.46 ( 0.26)
7.06 ( 0.07)
7.01
7.03
6.90
6.72
6.80
6.67
6.77
7.35
7.08
7.04
7.57 ( 0.08)
7.26 ( 0.12)
7.10 ( 0.05)
6.79 ( 0.08)
8.01 ( 0.17)
8.06 ( 0.30)
6.12 ( 0.04)
5.37 ( 0.09)
8.17 ( 0.22)
7.31 ( 0.19)
7.77 ( 0.18)
7.73 ( 0.13)
7.54 ( 0.19)
7.91 ( 0.27)
7.16 ( 0.05)
7.72 ( 0.24)
8.08 ( 0.32)
7.80 ( 0.23)
8.34 ( 0.17)
8.01 ( 0.17)
7.31
7.31
7.65
6.20
8.47
7.17
7.50
7.76
8.00
8.57 ( 0.07)
7.36 ( 0.09)
8.55 ( 0.04)
7.66 ( 0.08)
7.26 ( 0.04)
6.36 ( 0.11)
7.59 ( 0.03)
6.34 ( 0.11)
7.15( 0.06)
7.49 ( 0.04)
6.75 ( 0.05)
6.18 ( 0.14)
7.17 ( 0.08)
7.57 ( 0.07)
8.52 ( 0.02)
7.21 ( 0.07)
7.96 ( 0.04)
6.87 ( 0.05)
8.40
8.21
7.43
7.49
Me
(R)-2
(S)-3
Et
(R)-3
(S)-4
t-Bu
CN
(R)-4
(S)-5
6.92
(R)-5
(S)-6
CF3
7.34
7.21
8.81
7.88
(R)-6
(S)-7
COMe
F
(R)-7
(S)-8
(R)-8
(S)-9
Cl
(R)-9
(S)-10
(R)-10
(S)-11
(R)-11
(S)-12
(R)-12
(S)-13
(R)-13
(S)-14
(R)-14
(S)-15
(R)-15
(S)-16
(R)-16
SMe
SOMe
SO2Me
NO2
NHCOMe
NH2
OMe
7.757.62 ( 0.08)
7.35 ( 0.06)
7.23
7.65
6.756.90 ( 0.04)
n.d.
6.757.11 ( 0.11)
7.26
7.33 ( 0.04)
7.03
n.d.
6.69 ( 0.06)
n.d.
n.d.
6.96 ( 0.04)
n.d.
6.98 ( 0.15)
n.d.
6.62
6.94
7.44
7.52
7.91
8.51
7.00 ( 0.04)
n.d.
6.75
7.31
7.48
7.73
8.54
8.62
8.84
7.31 ( 0.27)
7.07 ( 0.11)
7.46 ( 0.19)
7.26 ( 0.25)
7.26 ( 0.11)
6.72 ( 0.13)
7.90 ( 0.09)
7.61 ( 0.09)
7.24 ( 0.12)
6.41 ( 0.14)
6.81 ( 0.14)
6.23 ( 0.13)
6.88 ( 0.10)
5.91 ( 0.17)
7.37 ( 0.14)
6.94 ( 0.04)
8.24 ( 0.04)
7.14 ( 0.06)
6.82
7.23
7.19
7.50
7.99
7.96 ( 0.13)
8.09 ( 0.29)
7.73 ( 0.03)
7.28 ( 0.16)
7.88 ( 0.17)
7.51 ( 0.16)
7.52 ( 0.04)
6.95( 0.07)
9.29 ( 0.11)
7.98 ( 0.08)
7.62
7.44 ( 0.06)
6.95( 0.06)
8.16 ( 0.43)
7.09 ( 0.03)
8.06 ( 0.08)
6.58 ( 0.13)
8.67 ( 0.02)
8.00
7.41
8.06
8.37 ( 0.05)
8.61 ( 0.04)
7.39 ( 0.03)
(S)-WB4101 9.39 ( 0.06)
(R)-WB4101 7.95( 0.04)
8.82
serotoninergic receptor of the enantiomeric pairs of
compounds 1–16 and, as comparison terms, of (S)-
and (R)-WB4101.
A little more detailed analysis of the binding data for
each of the four receptors reveals quite different affinity
decreases with respect to the 2,6-dimethoxy substituted
lead compound. In the case of the a1a-AR, a generalized
and considerable drop was observed from the sub-
nanomolar affinity of (S)-WB4101 to the values around
0.1 lM. More moderate decreases were instead regis-
tered for the a1b affinity, but analogously resulting in
nearly 0.1 lM Ki values due to the lower affinity of
the lead compound towards this AR subtype than for
the a1a-AR. For the third subtype, the a1d-AR, a large
minority of compounds, including S and R isomers,
showed Ki around 10 nM, thus approximating the
potencies of (S)- and (R)-WB4101. Finally, more than
one of the tested compounds displayed almost nanomo-
lar 5-HT1A affinity, sometimes surpassing the corre-
sponding homochiral enantiomers of WB4101.
Furthermore, in the case of (S)-1, (S)-2, (S)-4, (S)-8
and (S)-16, the very high or remarkable affinity for
the serotoninergic receptor was associated with a signif-
icant selectivity with respect to the three a1-AR
subtypes.
Considering the binding data, it is possible to observe
that the affinities of the S enantiomers are generally
higher than those of their antipodes, the only significant
exceptions being represented by the cyano and the acetyl
substituted analogues, 5 and 7, respectively, at the a1a
and a1b AR subtypes and the 5-HT1A receptor. Any-
how, the eudismic indexes are relatively small, exceeding
the unity in the only cases of the 5-HT1A affinity dis-
played by the enantiomeric pairs of 1, 4, 8 and 14.
Compared to the respective homochiral enantiomers of
WB4101, the ortho monosubstituted phenoxy deriva-
tives 1–16 are, for the most part, less potent. In particu-
lar, all the S isomers, excepting the monomethoxy
derivative (S)-16 at the 5-HT1A receptor, display lower
affinities than (S)-WB4101 and the gap is more pro-
nounced in the case of the three a1 subtypes. Otherwise,
a few exceptions can be counted among the R enantio-
mers, especially at the 5-HT1A receptor. Indeed, the
5-HT1A affinities of (R)-2, (R)-5, (R)-7 and, above all,
of (R)-16 exceed that of the WB4101 R enantiomer.
While such a birdꢁs-eye view of the values listed in Table 1
makes evident some significantly distinctive trends of the

L. Fumagalli et al. / Bioorg. Med. Chem. 13 (2005) 2547–2559
2551
four groups of binding data, a qualitative SAR analysis
of these latter appears problematic and little constructive
considering that the large variety in substituent proper-
ties does not always match with a very wide range of
affinity values. In fact, excluding the binding data of
(S)-4 and (R)-4, which frequently behave as modestly po-
tent outliers, the pKi range keeps almost unaltered for
both the R and S isomers only at the 5-HT1A receptor
(ꢁ2) and the a1b-AR (ꢁ1.4), whereas it sensibly decreases
at the other two ARs, even dropping to 0.9 in the case of
the a1a subtype. This notwithstanding, the mass of data
was judged suitable material for a QSAR investigation,
which could elicit latent correlations between the physi-
cochemical parameters of the aromatic substituents and
the biological data and sort out some determinants for
the selectivity modulation. Obviously, a broad range of
affinities facilitates the recognition of QSAR relation-
ships. Therefore, in order to expand the affinity scale,
the enantiomers of WB4101, in addition to those of 4,
were included in the series and the best linear correlations
were searched between each of the four affinities and the
physicochemical parameter values resulting from the
sum of the contributions of the two ortho substituents,
namely two methoxy groups in WB4101, a hydrogen
atom and another substituent in compounds 2–16 and
two hydrogen atoms in compound 1. Thus, the pKi range
was enlarged over two or three units also for the a1 affin-
ities, in particular, of the S isomers, whose QSAR inves-
tigation is reported here in detail.
the resonance constant of Swain and Lupton (R)27
was used, while the fragmental volume of aromatic sub-
stituents (V)28 and the length parameter of Verloop
(L)29 were employed as steric descriptors. The ability
of accepting H bonds was parameterized by the
HBb30 value derived from the fragmental system Sys-
tahl 2.0. The values of the input parameters are shown
in Table 3.
For the a1a-AR, Eq. 1 (Table 2) shows that the substitu-
ent volume exerts an adverse effect, while the negative
regression coefficient of the resonance parameter indi-
cates that the affinity values benefit from electron donor
substituents. Furthermore, the positive sign of HBb is
consistent with the stabilizing role of hydrogen bond
for the a1a-AR interaction, already postulated for the
reference ligand WB4101. The same parameters with
the same coefficient sign are found in the equation for
Table 3. Values of the physicochemical parametersa used in the QSAR
study
Compd
Subst
Rb
Vc
9.24
Ld
HBbe
(S)-1
(S)-2
H
Me
0
4.12
4.93
6.17
6.17
6.29
5.36
6.12
4.71
5.58
6.36
6.17
6.17
5.5
0
ꢀ0.13
ꢀ0.1
ꢀ0.13
0.19
22.7
0
(S)-3
(S)-4
Et
36.16
63.08
22
0
C(CH3)3
CN
CF3
0
(S)-5
(S)-6
0.23
0.09
0.51
0.03
0.01
0.2
0.62
0.40
0.28
0.19
0.2
31.12
37.16
11.41
20.32
37.7
A classical Hansch multivariate regression analysis
using the least-squares method was carried out to de-
rive QSAR equations for our data sets. For each of
the four receptors, the equation adopted as the final
model is given in Table 2 and the resultant predicted
pKi values for all the S isomers are listed in Table 1,
where they can be compared with the corresponding
experimental ones. The level of significance of the re-
sults was judged by the squared correlation coefficient
(r2), the squared cross-validated correlation coefficient
(q2), the standard error of the estimates (SE) and the
Fisher significance ratio (F). The statistical relevance
was also assured by a >5ratio of compounds ( n = 17)
to variables (3) and a reasonably extended pKi range.
The significance of each regression coefficient was ex-
pressed by its standard error. The substituent effects
were described by four physicochemical parameters,
suitably chosen out of 27. As electronic parameter,
(S)-7
(S)-8
(S)-9
Ac
F
Cl
ꢀ0.34
ꢀ0.15
ꢀ0.18
0.22
(S)-10
(S)-11
(S)-12
(S)-13
(S)-14
(S)-15
(S)-16
(S)-WB4101
SMe
SOMe
SO2Me
NO2
NHAc
NH2
OMe
2,6-OMe
42.50
47.04
25.08
46.03
19.53
27.76
46.28
0.1
0.16
ꢀ0.26
ꢀ0.18
ꢀ0.51
ꢀ1.02
7.150.67
4.84
6.04
7.96
0.21
0.3
0.6
a For each parameter, the reported value is the sum of the contribu-
tions of the two substituents in 2 and in 6. For compounds 2–16, one
of them is invariably a hydrogen atom. For compound 1, both sub-
stituents are a hydrogen atom.
b Data taken from Ref. 27.
c Data taken from Ref. 28.
d Data taken from Ref. 29.
e Data taken from Ref. 30.
Table 2. Equations for QSAR of the a1-ARs and the 5-HT1A receptor
Receptor
Equations
n
r2
q2
SE
F
a1a
(1) pKi ¼ 7:52ðÆ0:19Þ þ 0:94ðÆ0:22ÞHBb ꢀ 1:51ðÆ0:28ÞR ꢀ 0:024ðÆ0:007ÞV
17
0.78
0.74
0.34
15.50
a1b
(2) pKi ¼ 7:08ðÆ0:21Þ þ 0:37ðÆ0:11ÞHBb ꢀ 1:03ðÆ0:18ÞR ꢀ 0:0079ðÆ0:0073ÞV
(3) pKi ¼ 4:22ðÆ0:27Þ ꢀ 0:17ðÆ0:06ÞR þ 0:87ðÆ0:19ÞL ꢀ 0:054ðÆ0:012ÞV
(4) pKi ¼ 10:79ðÆ0:54Þ þ 0:79ð0:13ÞHBb ꢀ 0:58ðÆ0:11ÞL ꢀ 2:15ðÆ0:16ÞR
17
17
17
0.53
0.45
0.68
0.36
0.41
4.96
a1d
0.71
11.07
5-HT1A
0.850.81
0.27
24.31

2552
L. Fumagalli et al. / Bioorg. Med. Chem. 13 (2005) 2547–2559
a1b-AR affinity (Eq. 2, Table 2). This means that the
physicochemical properties favouring a1a and a1b affini-
ties are quite constant. Conversely, the statistical good-
ness of Eq. 2 is significantly minor than that for a1a-AR
and this may suggest that the role of the phenoxy moiety
is not so important for the a1b as for the a1a interaction.
For the a1d-AR, the best equation (Eq. 3, Table 2) lacks
the HBb parameter while including an additional steric
descriptor (L). The positive sign of this latter, combined
with the invariably negative one of V, would mean that
the phenoxy residue is inserted in a cavity, which is size-
limited in all three AR subtypes, but can accept, in the
case of the a1d-AR, substituents of rather elongated than
spherical shape.
with this moiety is subjected to tight steric constraints in
both a1 and 5-HT1A receptors. In fact, steric descriptors
are present in all the equations generally as the terms of
great weight and, excepting L in the a1d equation, their
effect is always negative. Contrariwise, the electron-rich-
ness is invariably a beneficial property and the ability to
accept H bonds increases the affinity for three receptors
(a1a,a1b and 5-HT1A) of the four considered. However,
such a similarity does not imply that the ortho monosub-
stitution at the phenoxy residue exerts univocal effects
on the affinity for the a1-ARs and the 5-HT1A receptor.
Indeed, the different incidence of the steric and the elec-
tronic parameters results in variegated affinity profiles
leading, in the case of (S)-1, (S)-2, (S)-4, (S)-8 and (S)-
16, to high 5-HT1A affinities combined with a specificity
for this receptor experimentally even more significant
than that predicted.
Finally, the last equation (Eq. 4, Table 2) shows that
electron-richness (see the negative coefficient of R) and
ability to accept H bonds (see the positive coefficient
of HBb) are beneficial properties also for the 5-HT1A
receptor affinity, which is however considerably lowered
by the substituent length. In particular, the negative
coefficient of L indicates that substituents of spherical
shape are better accepted, as in the case of (S)-1, (S)-2,
(S)-8, which exhibit almost nanomolar 5-HT1A affinities
and of (S)-4, whose remarkable affinity for the seroto-
ninergic receptor (26 nM) is 20–30-fold more higher
than for the a1-ARs. Furthermore, differently from the
QSARs for the a1 receptors, this significant 5-HT1A
affinity, in connection with such a bulky substituent,
cancels the negative effects of the V parameter making
it uninfluential in the present series of substituents as
demonstrated by unsuccessful attempts to introduce this
descriptor into Eq. 4.
On the basis of these equations, it is difficult to rational-
ize the affinity shift of this series of ortho monosubsti-
tuted phenoxy derivatives towards the 5-HT1A
receptor and, to a minor extent, the a1d-AR. In this re-
gard, though prediction is neither the proper faculty nor
the main goal of QSARs, it is interesting to observe that
the equations for the a1a and a1b receptors are so struc-
tured as to let no ortho monosubstituted derivative com-
pete with the 2,6-dimethoxy substituted reference
compound (S)-WB4101. This would be possible if the
volumetric term had a more negative incidence. Such a
condition is realized, in connection with a much lower
influence of the R parameter, in the a1d equation, but
it is counteracted by the concomitant largely positive ef-
fect of the substituent length. Again, like in the two pre-
vious QSARs, the a1d affinity of the monosubstituted
derivatives, though generally higher than for the a1a
and a1b ARs, seems to be inevitably handicapped in
comparison with that of (S)-WB4101. Finally, the situa-
tion sensibly changes in the case of the 5-HT1A receptor
as a proof of its different nature. Negative values of R
exert the most beneficial effect and only the length of
the substituents becomes detrimental while their volume
is uninfluential. This results in many ortho monosubsti-
tuted derivatives, bearing an electron-donor and not too
long substituent, which show calculated and experimen-
tal 5-HT1A affinities comparable with that of (S)-
WB4101 and not far from those predicted for the corre-
sponding ortho disubstituted analogues. It is significant
that the exclusion of (S)-WB4101 from the QSAR anal-
ysis is uninfluential only in the case of the 5-HT1A
receptor.
The four equations were recalculated maintaining the
same physicochemical parameters, but excluding (S)-
WB4101, the only compound in the series without ortho
hydrogens at the phenoxy residue. The r2 values of the
new relations were 0.83 (5-HT1A receptor), 0.64 (a1a-
AR), 0.63 (a1d-AR) and 0.27 (a1b-AR). Such results
indicate that, apart from the QSAR for the a1b-AR,
these equations were still significant, retaining, in the
case of the 5-HT1A receptor, a very high correlation.
Furthermore, it is noteworthy that the equations recal-
culated for the a1a, a1d and 5-HT1A receptors without
(S)-WB4101 maintained the original coefficient signs,
excepting the R coefficient in the a1d-AR, which chan-
ged from a slightly negative value (ꢀ0.17) to +0.36. This
suggests that the inclusion of (S)-WB4101 in the present
analysis does not subvert the QSARs of the ortho mon-
osubstituted analogs, but somehow validates them. In
fact, as expected on the basis of its very high affinities
associated with the presence of two identical ortho sub-
stituents, (S)-WB4101 enhances the statistical signifi-
cance of such relationships and makes their nature
clearer.
In summary, the present results demonstrate the already
theorized importance of the 2,6-disubstitution, in partic-
ular with two methoxy groups, at the phenyl residue of
2-[(2-phenoxyethyl)aminomethyl]-1,4-benzodioxane for
the interaction with the a1-ARs and, especially, with
the a1a-AR subtype. Otherwise, the same results and
their QSAR analysis clearly indicate that such a feature
is not essential to the interaction with the 5-HT1A recep-
tor as proved by the high and specific 5-HT1A affinity of
the unsubstituted derivative (S)-1 as well as of several
WB4101 analogues suitably ortho monosubstituted at
the phenyl ring.
4. Conclusions
The present QSAR study on a wide series of (S)-
WB4101 S analogues differently ortho monosubstituted
at the phenoxy residue indicates that the site interacting

L. Fumagalli et al. / Bioorg. Med. Chem. 13 (2005) 2547–2559
2553
5. Experimental
(t, 2H), 6.89 (t, 1H), 6.99 (d, 1H), 7.39 (dt, 1H), 7.79 (dd,
1H). Anal. Calcd for C8H9NO4 (183.164).
5.1. Chemistry
5.1.6. 2-(2-Ethylphenoxy)-1-tosyloxyethane (22). Tosyl
chloride (22.8 g, 120 mmol) was added in small portions
to a stirred solution of 17 (20 g, 120 mmol) in pyridine
(70 mL) at 0 °C. The resulting mixture was stirred at
room temperature overnight, diluted with diethyl ether
(100 mL) and washed with 10% HCl. The aqueous layer
was separated and extracted with diethyl ether twice.
The organic phases were combined, dried and concen-
trated to give 25.5 g (66%) of 22 as a white solid: mp
79–80 °C; 1H NMR (CDCl3) d 1.1 (t, 3H), 2.45(s,
3H), 2.55 (q, 2H), 4.2 (t, 2H), 4.4 (t, 2H), 6.6–7.2 (m,
4H), 7.34 (d, 2H), 7.85(d, 2H). Anal. Calcd for
C17H20O4S (320.40).
Melting points were measured on Buchi melting point
apparatus and are uncorrected. H NMR spectra were
1
recorded operating at 60 MHz, 200 MHz or 300 MHz.
Chemical shifts are reported in parts per million relative
to residual solvent (CHCl3 or DMSO) as internal
standard. Signal multiplicity is designed according to
the following abbreviations: s = singlet, d = doublet,
dd = doublet of doublets, t = triplet, m = multiplet, br
s = broad singlet, br t = broad triplet. Optical rotations
were determined by a Perkin-Elmer 241 Polarimeter at
25 °C. Elemental analyses (CHN) of the new substances
are within 0.40% of theoretical values. Purifications were
performed by flash chromatography using silica gel
(particle size 40–63 lm, Merck).
5.1.7. 2-(2-Methylthiophenoxy)-1-tosyloxyethane (23).
Prepared from 19 as described for 22 and isolated after
crystallization from cyclohexane/ethyl acetate (1:1) as a
5.1.1. 2-(2-Ethylphenoxy)ethanol (17). A mixture of 2-
ethylphenol (20 g, 164 mmol), ethylene carbonate
(28.86 g, 328 mmol) and potassium carbonate (22.6 g,
164 mmol) in toluene (70 mL) was refluxed for 2 h. After
cooling at room temperature, water was added. The
aqueous layer was separated and extracted with toluene
twice. The organic phase was treated with 10% NaOH
(150 mL), washed with water twice (200 mL), dried
and concentrated to give the crude product, which was
purified by chromatography on silica gel. Elution with
cyclohexane–ethyl acetate (90:10) yielded 21 g (77%) of
1
white solid (64%): mp 71–73 °C; H NMR (CDCl3) d
2.37 (s, 3H), 2.42 (s, 3H), 4.0–4.6 (t, 4H), 6.0–7.2 (m,
4H), 7.32 (d, 2H), 7.85(d, 2H). Anal. Calcd for
C16H18O4S2 (338.44).
5.1.8. 2-(2-Acetylphenoxy)-1-tosyloxyethane (24). Pre-
pared from 20 as described for 22 and isolated after
crystallization from hexane/ethyl acetate (1:1) as a white
1
solid (43%): mp 87 °C; H NMR (CDCl3) d 2.4 (s, 3H),
2.50 (s, 3H), 4.60–4.80 (m, 4H), 6.80–8.00 (m, 8H). Anal.
Calcd for C17H18O5S (334.39).
1
17 as an oil. H NMR (CDCl3) d 1.28 (t, 3H), 2.45(br
s, 1H), 2.65(q, 2H), 4.3–4.37 (m, 4H), 6.7–7.4 (m,
4H). Anal. Calcd for C10H14O2 (166.22).
5.1.9. 2-(2-Acetylphenoxy)-1-iodoethane (25). A mixture
of 24 (4.7 g, 14 mmol) and sodium iodide (24.0 g,
140 mmol) in acetone (50 mL) was refluxed for 12 h.
The solvent was evaporated and the residue treated with
10% HCl and diethyl ether. The aqueous layer was sep-
arated and extracted with diethyl ether twice. The or-
ganic phases were combined, dried and concentrated
to give the crude product, which was purified by chro-
matography on silica gel. Elution with n-hexane/ethyl
acetate (90:10) afforded 3.5g (85%) of 25 as a solid:
5.1.2. 2-(2-Trifluoromethylphenoxy)ethanol (18). Pre-
pared from ortho-trifluoromethylphenol as described
for 17 but replacing toluene with DMF. After chroma-
tography on silica gel (cyclohexane–ethyl acetate
70:30) the product was isolated as a white solid (64%):
1
mp 75.8 °C; H NMR (CDCl3) d 2.20 (br s, 1H), 3.94
(m, 2H), 4.15(m, 2H), 6.97–7.05(m, 2H), 7.43–7.75
(m, 2H). Anal. Calcd for C9H9F3O2 (206.16).
1
mp 45–47 °C; H NMR (CDCl3) d 2.70 (s, 3H), 3.50
5.1.3. 2-(2-Methylthiophenoxy)ethanol (19). Prepared
from 2-methylthiophenol as described for 17. Crystalli-
zation from diethyl ether yielded 19 as a white solid
(t, 2H), 4.40 (t, 2H), 6.80–7.90 (m, 4H). Anal. Calcd
for C10H11IO2 (290.10).
1
(84%): mp 40–42 °C; H NMR (CDCl3) d 2.4 (s, 3H),
5.1.10. 2-(2-Trifluoromethylphenoxy)-1-mesyloxyethane
(26). Mesyl chloride (0.98 mL, 12.6 mmol) was added
dropwise to a stirred solution of 18 (2.47 g, 11.98 mmol)
and TEA (1.75mL) in dichloromethane at 0 °C. After
2 h at room temperature the reaction mixture was
washed with a saturated solution of NaHCO3, dried
and concentrated to give the crude product 3.35g
(98.5%) as a white solid which was used for the subse-
quent step without further purification: mp 51.5 °C;
1H NMR (CDCl3) d 3.07 (s, 3H), 4.35(m, 2H), 4.58
(m, 2H), 6.97–7.10 (m, 2H), 7.48–7.60 (m, 2H). Anal.
Calcd for C10H11F3O4S (284.25).
4.28 (t, 2H), 4.4–4.9 (br s, 3H), 6.7–7.5(m, 4H). Anal.
Calcd for C9H12O2S (184.25).
5.1.4. 2-(2-Acetylphenoxy)ethanol (20). Prepared from
2-hydroxyacetophenone as described for 17. After
chromatography on silica gel the product was iso-
1
lated as a yellow oil (48%): H NMR (CDCl3) d 1.60
(br s, 1H), 2.60 (s, 3H), 4.05(m, 2H), 4.25(m, 2H),
6.60–7.79 (m, 4H). Anal. Calcd for C10H12O3
(180.203).
5.1.5. 2-(2-Nitrophenoxy)ethanol (21). Prepared as de-
scribed for 17 but replacing toluene with DMF. The
resulting crude product, isolated as a yellow oil (57%),
was used for the subsequent step without further purifi-
cation: 1H NMR (CDCl3) 3.42 (s, 1H), 3.84 (t, 2H), 4.09
5.1.11. 2-(2-Nitrophenoxy)-1-mesyloxyethane (27). Pre-
pared from 21 as described for 26 and isolated as an
orange oil (98%): 1H NMR (CDCl3) d 3.04 (s, 3H),
4.32 (m, 2H), 4.52 (m, 2H), 6.98–7.08 (m, 2H), 7.5 (t,

2554
L. Fumagalli et al. / Bioorg. Med. Chem. 13 (2005) 2547–2559
1H), 7.74 (d, 1H). Anal. Calcd for C9H11NO6S
(261.16).
resulting mixture was heated at 90 °C for 4 h. After cool-
ing at room temperature, water was removed and the
residue was dissolved with ethanol, which was then con-
centrated. The same treatment was done for four times.
At the end 34 hydrochloride was isolated as white solid
10 g (67%): H NMR (CDCl3) d 2.60 (s, 3H), 4.05(m,
2H), 4.25(m, 2H), 6.60–7.79 (m, 4H). Anal. Calcd for
C9H13O2SN (199.26).
5.1.12. N-[2-(2-Ethylphenoxy)ethyl]phthalimido (28). So-
dium hydride (1.46 g, 50.8 mmol) was suspended in
DMF (35mL) under N 2 atmosphere. A solution of
phthalimide (7.46 g, 50.8 mmol) in DMF (35 mL) was
added dropwise and the resulting mixture was heated
at 70 °C for 1 h. Then 22 (19.0 g, 59.3 mmol) was di-
luted in DMF (25mL) and added stepwise to the stir-
ring solution. The resulting mixture was refluxed for
3 h and after cooling at room temperature was
quenched into frozen water (150 mL). The resulting
precipitate was isolated yielding 15.60 g (90%) of 28
as a white solid: mp 61–63 °C; 1H NMR (CDCl3) d
1.1 (t, 3H), 2.24 (d, 2H), 4.18 (s, 4H), 6.6–7.4 (m,
4H), 7.5–8.0 (m, 4H). Anal. Calcd for C18H17NO3
(295.34).
1
5.1.19. 2-(2-Methylsulfonylphenoxy)ethylamine (35). 3-
Chloroperbenzoic acid (12 g, 28 mmol) was added to a
stirring solution of 34 (5g, 27.5mmol) in dichlorometh-
ane (40 mL) at 0 °C. The reaction mixture was stirred at
room temperature for 24 h. At the end of the reaction,
the mixture was washed with a saturated solution of
NaHCO3. The organic phase was separated, dried and
concentrated to give 35 (3.75g, 63%) as an oil: 1H
NMR (CDCl3) d 2.10 (s, 2H), 3.25(s, 3H), 4.25(t,
2H), 7.00–8.15(m, 4H). Anal. Calcd for C H13O3SN
9
5.1.13. N-[2-(2-Methylthiophenoxy)ethyl]phthalimido
(29). Prepared from 23 as described for 28: mp 117–
120 °C: H NMR (CDCl3) d 2.3 (s, 3H), 4.22 (m, 4H),
6.7–7.4 (m, 4H), 7.5–8.15 (m, 4H). Anal. Calcd for
C17H15NO3S (313.37).
(215.26).
1
5.1.20. 2-(2-Aminophenoxy)ethanol (36). A solution of 21
(11 g, 60 mmol) in methanol (220 mL) was added with
10% Pd/C (1 g) and vigorously shaken under hydrogen
at room temperature. The catalyst was removed by fil-
tration and the filtrate concentrated to give 36 (9.02 g)
as a red solid: mp 90.9 °C; H NMR (CDCl3) d 2.6–
3.3 (br s, 3H), 3.97 (t, 2H), 4.12 (t, 2H), 6.6–6.9 (m,
4H). Anal. Calcd for C8H11NO2 (153.18).
5.1.14. N-[2-(2-Trifluoromethylphenoxy)ethyl]phthalim-
ido (30). Prepared from 26 as described for 28: mp
1
1
126 °C; H NMR (CDCl3) d 4.17 (t, 2H), 4.32 (t, 2H),
6.95–7.00 (m, 2H), 7.46 (m, 2H), 7.71 (m, 2H), 7.85
(m, 2H). Anal. Calcd for C17H12F3O3N (335.28).
5.1.21. 2-(2-Acetlyaminophenoxy)ethanol (37). Acetyl
chloride (7.82 mL, 0.110 mmol) was added stepwise to
a cool solution of 36 (7.99 g, 52 mmol) and TEA
(15.33 mL, 0.110 mmol) in dichloromethane (48 mL).
The reaction mixture was stirred at room temperature
for 2 h, treated with water (50 mL) dried and concen-
trated. The residue obtained was dissolved in methanol
(40 mL) and 10% NaOH (21.5mL). After stirring at
room temperature for 30 min the solvent was removed.
The residue was treated with 10% HCl (30 mL) and
dichloromethane (50 mL). The organic phase was sepa-
rated, dried and concentrated to give 37 (10.83 g, 98%)
5.1.15. 2-(2-Ethylphenoxy)ethylamine (31). Hydrazine
hydrate (2.7 mL) was added dropwise to a stirred mix-
ture of 28 (15g, 50.8 mmol) in 2-methoxyethanol at
125 °C. After refluxing for 2 h 10% HCl was added until
pH = 3. After cooling, the reaction mixture was concen-
trated, the residue treated with 10% NaOH (100 mL)
and washed with diethyl ether. The organic phases were
combined, dried and concentrated. Distillation under
vacuum (bp 75 °C at 0.5mbar) of the residue gave
1
5.2 g (62%) of 31 as a colourless oil: H NMR (CDCl3)
d 1.04 (t, 3H), 1.30 (br s, 2H), 2.52 (q, 2H) 2.91 (t, 2H),
3.81 (t, 2H), 6.5–7.4 (m, 4H). Anal. Calcd for C10H15NO
(165.23).
1
as an oil: H NMR (CDCl3) d 2.14 (s, 3H), 3.97 (t,
2H), 4.12 (t, 2H), 6.86–7.02 (m, 3H), 8.2 (br s, 1H),
8.24 (m, 1H). Anal. Calcd for C10H13NO3 (195.22).
5.1.16. 2-(2-Trifluoromethylphenoxy)ethylamine (32).
Prepared from 30 as described for 31 and isolated, after
chromatography on silica gel (dichloromethane/isopro-
5.1.22.
2-(2-Acetylaminophenoxy)-1-mesyloxyethane
(38). Prepared from 37 as described for 26 and isolated
after crystallization from diisopropyl ether/2-propanol
(120 mL:100 mL) as a white solid (53%): mp 87.9 °C 1H
NMR (CDCl3) d 1.2 (s, 3H), 3.1 (s, 3H), 4.23 (t, 2H),
4.65(t, 2H), 6.8 (m, 1H), 6.98 (m, 2H), 8.29 (br s, 1H),
8.39 (m, 1H). Anal. Calcd for C11H15NO5S (273.39).
1
panol 95:5), as a yellow oil (88.7%): H NMR (CDCl3)
d 1.62 (br s, 2H), 3.11 (t, 2H), 4.07 (t, 2H), 6.96–7.05
(m, 2H), 7.41–7.58 (m, 2H). Anal. Calcd for
C9H10F3ON (205.18).
5.1.17. 2-(2-Methylthiophenoxy)ethylamine (33). Pre-
pared from 29 as described for 31 and isolated after dis-
tillation under vacuum (bp 108 °C at 0.8 mbar) as an oil
(42%): 1H NMR (CDCl3) d 1.7–2.1 (m, 1H), 2.3–2.6 (m,
1H), 2.4 (s, 3H), 3.12 (t, 2H), 4.12 (t, 2H), 6.7–7.5(m,
4H). Anal. Calcd for C9H13NOS (183.27).
5.1.23. 2-(2-Acetylaminophenoxy)-1-dibenzylaminoethane
(39). A solution of 38 (8.6 g, 32 mmol) and dibenzyl-
amine (12.21 mL, 64 mmol) in isopropanol (50 mL)
was refluxed overnight. At the end, the solvent was re-
moved and the residue dissolved in dichloromethane
(50 mL) and treated with 10% NaOH (50 mL). The
organic phase was separated, washed with water
(50 mL), dried and concentrated to give the crude prod-
uct, which was purified by chromatography on silica gel.
5.1.18. 2-(2-Methylsulfinylphenoxy)ethylamine (34). 30%
H2O2 (10.2 mL) was added to a solution of 33 hydro-
chloride (14 g, 63.70 mmol) in water (40 mL). The

L. Fumagalli et al. / Bioorg. Med. Chem. 13 (2005) 2547–2559
2555
25
anol) (½aꢂ ¼ þ35:22 (c 1, CHCl3) for the free amine);
Elution with cyclohexane/ethyl acetate (1:1) afforded 39
as an oil. After crystallization from cyclohexane (50 mL)
39 was isolated (7.63 g, 64%) as a white solid: mp
98.1 °C; H NMR (CDCl3) d 2.09 (s, 3H), 2.94 (t, 2H),
3.73 (s, 4H), 4.1 (t, 2H), 6.77 (m, 1H), 6.96 (m, 2H),
7.23–7.42 (m, 10H), 7.79 (br s, 1H), 8.36 (m, 1H). Anal.
Calcd for C24H26N2O2 (374.48).
D
1H NMR identical to that of (S)-1.
1
5.1.28. (S)-2-[((2-(2-Methylphenoxy)ethyl)amino)methyl]-
1,4-benzodioxane hydrochloride [(S)-2]. A solution of
2-(2-methylphenoxy)ethylamine (1.4 g, 9.25mmol)
and (2R)-mesyloxymethyl-1,4-benzodioxane (1.2 g,
4.91 mmol) in 2-methyl-1-propanol (4 mL) was refluxed
for 48 h. After removing the solvent, 10% Na2CO3 and
dichloromethane were added. The aqueous layer was
separated and extracted with dichloromethane again.
The organic phases were combined, dried and concen-
trated to give a residue, which was purified by chroma-
tography on silica gel. Elution with cyclohexane/ethyl
acetate (1:1) afforded 0.44 g (30.13%) of (2S)-2-[((2-(2-
5.1.24. 2-(Acetylaminophenoxy)ethylamine (40). A solu-
tion of 39 (7.63 g, 20 mmol) in ethyl acetate (100 mL)
and ethanol (250 mL) was added with 10% Pd/C
(0.5g) and vigorously shaken under hydrogen at room
temperature. The catalyst was removed by filtration,
the filtrate concentrated and dissolved with dichloro-
methane (100 mL). The solvent was then removed to
1
methylphenoxy)ethyl)amino)methyl]-1,4-benzodioxane:
give 40 (3.88 g) as a white solid: mp 62.4 °C: H NMR
25
D
½aꢂ ¼ ꢀ32:2 (c 1, CHCl3);1H NMR (CDCl3) d 1.40 (br
(CDCl3) d 1.49 (s, 2H), 2.18 (s, 3H), 3.09 (t, 2H), 4.04
(t, 2H), 6.88–7.04 (m, 3H), 8.32 (m, 1H), 8.44 (br s,
1H): C10H14N2O2 (194.23).
s, 1H), 2.15(s, 3H), 3.00 (m, 2H), 3.15(m, 2H), 4.00–
4.10 (m, 3H), 4.10–4.25(m, 2H), 6.80 (m, 6H), 7.15
(m, 2H). The secondary amine was dissolved in ethyl
acetate/chloroform (5:1) (5 mL) and 4 N HCl/EtOH
was added. Ethyl acetate (10 mL) was added and evap-
orated for three times. After crystallization with ethyl
5.1.25. (2S)-2-Benzylaminomethyl-1,4-benzodioxane [(S)-
41]. A solution of (2R)-2-iodomethyl-1,4-benzodioxane
(1.5g, 5.43 mmol) and benzylamine (1.18 mL,
10.86 mmol) in isopropanol was refluxed for 18 h. After
cooling at room temperature, Na2CO3 (700 mg) and sil-
ica gel (10 g) was added. The solvent was removed and
the residue was purified by chromatography on silica
gel. Elution with hexane/ethyl acetate (85:15) afforded
acetate/ethanol (10:1), (S)-2 was isolated as a white so-
25
lid: mp 167.7 °C; ½aꢂ ¼ ꢀ61:34 (c 1, methanol); 1H
NMR (DMSO-d6) d D2.18 (s, 3H), 3.27–3.46 (m, 4H),
4.08 (dd, 1H), 4.31 (t, 2H), 4.40 (dd, 1H), 4.72 (m,
1H), 6.80–6.95(m, 6H), 7.14 (m, 2H), 9.74 (br s, 2H).
Anal. Calcd for C18H22ClNO3 (335.82).
850 mg (81.2%) of (S)-41 as
a
colourless oil:
25
D
1
½aꢂ ¼ ꢀ41:4 (c 1, CHCl3); H NMR (CDCl3) d 2.90
5.1.29.
(R)-2-[((2-(2-Methylphenoxy)ethyl)amino)-
(d, 2H), 3.85 (s, 2H), 3.95–4.50 (m, 3H), 6.90 (s, 4H),
7.35(s, 5H). Anal. Calcd for C 16H17NO2 (255.31).
methyl]-1,4-benzodioxane hydrochloride [(R)-2]. Prepared
from 2-(2-methylphenoxy)ethylamine and (2S)-mesyl-
oxymethyl-1,4-benzodioxane as described for (S)-2: mp
25
25
5.1.26. (S)-2-[((2-(Phenoxy)ethyl)amino)methyl]-1,4-ben-
zodioxane hydrochloride [(S)-1]. A solution of (2R)-2-
iodomethyl-1,4-benzodioxane (1.7 g, 6.16 mmol) and
2-phenoxyethylamine (1.69 g, 12.32 mmol) in EtOH
(5mL) was refluxed for 48 h. After removing the solvent,
10% NaOH and diethyl ether were added. The aqueous
layer was separated and extracted with diethyl ether
again. The organic phases were combined, dried and
concentrated to give a residue, which was purified by
chromatography on silica gel. Elution with cyclo-
hexane/ethyl acetate–isopropanol–NH4OH (8:6:2:0.03)
afforded 1.37 g (78%) of (2S)-2-[((2-(Phenoxy)ethyl)-
167.3 °C; ½aꢂ ¼ þ60:1 (c 1, ethanol) (½aꢂ ¼ þ32:9 (c
D
D
1, CHCl3) for the free amine); 1H NMR identical to that
of (S)-2.
5.1.30. (S)-2-[((2-(2-Ethylphenoxy)ethyl)amino)methyl]-
1,4-benzodioxane hydrochloride [(S)-3]. Prepared from
(2R)-2-iodomethyl-1,4-benzodioxane (1.5g, 5.43 mmol)
and 31 (1.8 g, 10.86 mmol) as described for (S)-1. (2S)-
2-[((2-(2-Ethylphenoxy)ethyl)amino)methyl]-1,4-benzo-
25
D
dioxane was isolated as an oil: ½aꢂ ¼ ꢀ25:3 (c 1,
1
CHCl3); H NMR (CDCl3) d 1.2 (t, 3H), 1.9 (s, 1H),
2.68 (q, 2H), 2.8–3.25 (m, 4H), 3.65–4.55 (m, 5H), 6.4–
7.4 (m, 8H). (S)-3 was isolated as a white solid (17%
amino)methyl]-1,4-benzodioxane
as
an
oil:
25
D
1
½aꢂ ¼ ꢀ33:44 (c 1, CHCl3); H NMR (CDCl3) d 1.7
based on the starting amount of (2R)-2-iodomethyl-1,4-
25
D
(s, 1H), 2.7–3.35(m, 4H), 3.8–4.6 (m, 5H), 6.7–7.1 (m,
9H). The secondary amine was dissolved in EtOH
(10 mL) and 4 N HCl/EtOH (5mL) was slowly added.
The resulting precipitate was dried yielding 0.995g
(64%, based on the starting amount of (2R)-2-iodo-
benzodioxane): mp 141 °C; ½aꢂ ¼ ꢀ55:22 (c 1, metha-
1
nol) H NMR (DMSO-d6) d 0.92 (t, 3H), 2.47 (q, 2H),
2.95–3.53 (m, 4H), 3.6–4.3 (m, 5H), 4.62 (s, 2H), 6.5–
7.2 (m, 8H). Anal. Calcd for C19H24ClNO3 (385.86).
methyl-1,4-benzodioxane) of (S)-1 as a white solid: mp
5.1.31. (R)-2-[((2-(2-Ethylphenoxy)ethyl)amino)methyl]-
1,4-benzodioxane hydrochloride [(R)-3]. Prepared from
(2S)-2-iodomethyl-1,4-benzodioxane and 31 as de-
25
D
233 °C; ½aꢂ ¼ ꢀ67:0 (c 1, methanol); 1H NMR
(DMSO-d6) d 2.9–3.8 (m, 4H), 3.8–5.2 (m, 5H), 6.5–7.7
(m, 9H), 9.82 (br s, 2H). Anal. Calcd for C17H20ClNO3
(285.34).
25
scribed for (S)-3: mp 141–142 °C; ½aꢂ ¼ þ55:19 (c 1,
D
25
D
amine); H NMR identical to that of (S)-3.
methanol) (½aꢂ ¼ þ25:25( c 1, CHCl3) for the free
1
5.1.27. (R)-2-[((2-(Phenoxy)ethyl)amino)methyl]-1,4-ben-
zodioxane hydrochloride [(R)-1]. Prepared from (2S)-2-
iodomethyl-1,4-benzodioxane and 2-phenoxyethylamine
5.1.32.
methyl]-1,4-benzodioxane hydrochloride [(S)-4]. Prepared
from (2R)-mesyloxymethyl-1,4-benzodioxane (1 g,
(S)-2-[((2-(2-t-Butoxyphenoxy)ethyl)amino)-
25
as described (S)-1: mp 230 °C; ½aꢂ ¼ þ65:74 (c 1, meth-
D

2556
L. Fumagalli et al. / Bioorg. Med. Chem. 13 (2005) 2547–2559
1
CHCl3) for the free amine); H NMR identical to that
of (S)-6.
8.17 mmol)
(1.58 g, 8.17 mmol) as described for (S)-2. The free
and
2-(2-t-butoxyphenoxy)ethylamine
25
amine was isolated as an oil: ½aꢂ ¼ ꢀ24:4 (c 1, CHCl3);
D
1H NMR (CDCl3) d 1.40 (s, 9H), 1.95(br s, 1H), 3.00
(m, 2H), 3.15(m, 2H), 3.9–54.19 (m, 3H), 4.20–4.40
(m, 2H), 6.75–6.95(m, 6H), 7.10–7.35(m, 2H). ( S)-4
was isolated as a white solid (37% based on the starting
5.1.38. (S)-2-[((2-(2-Acetylphenoxy)ethyl)amino)methyl]-
1,4-benzodioxane hydrochloride [(S)-7]. A solution of
(S)-41 (800 mg, 3.13 mmol), 25 (910 mg, 3.13 mmol)
and K2CO3 (500 mg) in toluene (15 mL) was refluxed
for 48 h. The solvent was removed to give the residue,
which was purified by chromatography on silica gel.
Elution with hexane/ethyl acetate (90:10) afforded
560 mg of tertiary amine, which was dissolved in a sus-
pension of 10% Pd/C in ethanol. At the end the catalyst
was filtrate and the solvent removed. The residue was
dissolved in ethyl acetate (30 mL) and treated with 5%
NaOH (30 mL). The aqueous layer was separated and
extracted with ethyl acetate again. The organic phases
were combined, washed with water, dried and concen-
trated to give the residue, which was purified by chroma-
tography on silica gel. Elution with dichloromethane/
methanol (97:3) afforded (S)-2-[((2-(2-acetylphen-
amount of (2R)-mesyloxymethyl-1,4-benzodioxane): mp
25
D
156.8 °C; ½aꢂ ¼ ꢀ63:7 (c 1, methanol) 1H NMR
(DMSO-d6) d 1.32 (s, 9H), 3.30–3.50 (m, 4H), 4.08
(dd, 1H), 4.34–4.40 (m, 3H), 4.73 (m, 1H), 6.88 (m,
5H), 7.00 (d, 1H), 7.19 (m, 2H), 9.80 (br s, 2H). Anal.
Calcd for C21H28ClNO3 (377.90).
5.1.33.
(R)-2-[((2-(2-t-Butoxyphenoxy)ethyl)amino)-
methyl]-1,4-benzodioxane hydrochloride [(R)-4]. Prepared
from (2S)-mesyloxymethyl-1,4-benzodioxane and 2-(2-t-
butoxyphenoxy)ethylamine as described for (S)-4:
25
D
25
D
mp 156.8 °C; ½aꢂ ¼ þ55:71 (c 1, methanol) (½aꢂ
¼
þ23:45( c 1, CHCl3) for the free amine); 1H NMR iden-
tical to that of (S)-4.
oxy)ethyl)amino)methyl]-1,4-benzodioxane
(340 mg,
43%) as a yellow oil. The secondary amine was dissolved
in ethanol (5mL) and 4 N HCl/EtOH (5mL) was
added. The resulting precipitate was filtrated and dried
5.1.34. (S)-2-[((2-(2-Cyanophenoxy)ethyl)amino)methyl]-
1,4-benzodioxane hydrochloride [(S)-5]. Prepared from
(2R)-2-iodomethyl-1,4-benzodioxane
mmol) and 2-(2-cyanophenoxy)ethylamine (1.08 g,
(1.47 g,
5.32
to give (S)-7 (320 mg, 26% based on the starting amount
25
D
of (S)-41): mp 118 °C; ½aꢂ ¼ ꢀ53:23 (c 1, methanol); 1H
6.65mmol) as described for ( S)-1. After chromatogra-
phy on silica gel (dichloromethane/methanol 97:3) the
NMR (CDCl3) d 2.58 (s, 3H), 3.30 (m, 2H), 3.40–3.52
(m, 2H), 4.16 (dd, 1H), 4.31–4.40 (m, 1H), 4.50–4.54
(m, 1H), 6.84–6.90 (m, 5H), 7.10 (t, 1H), 7.45 (t, 1H),
7.70 (dd, 1H), 9.7 (br s, 1H), 10.7 (br s, 1H). Anal. Calcd
for C19H23N2O4Cl (378.84).
25
D
(c 1, CHCl3). (S)-5 was obtained as a white solid:
secondary amine was isolated as an oil: ½aꢂ ¼ ꢀ26:4
25
D
mp 163–166 °C; ½aꢂ ¼ ꢀ60:29 (c 1, methanol) 1H
NMR (CDCl3) d 2.10 (br s, 1H), 2.90–3.35(m, 4H),
3.90–4.60 (m, 5H), 6.80–7.20 (m, 6H), 7.35–7.80 (m,
2H). Anal. Calcd for C18H19N2ClO3 (346.81).
5.1.39. (R)-2-[((2-(2-Acetylphenoxy)ethyl)amino)methyl]-
1,4-benzodioxane hydrochloride [(R)-7]. Prepared from
(R)-41 and 25 as described for (S)-7: mp 97.8 °C;
25
1
5.1.35. (R)-2-[((2-(2-Cyanophenoxy)ethyl)amino)methyl]-
1,4-benzodioxane hydrochloride [(R)-5]. Prepared from
(2S)-2-iodomethyl-1,4-benzodioxane and 2-(2-cyano-
phenoxy)ethylamine as described for (S)-5: mp 165.7
½aꢂ ¼ þ60:12 (c 1, methanol); H NMR (CDCl3) iden-
D
tical to that of (S)-7.
5.1.40. (S)-2-[((2-(2-Fluorophenoxy)ethyl)amino)methyl]-
1,4-benzodioxane hydrochloride [(S)-8]. Prepared from
(2R)-2-iodomethyl-1,4-benzodioxane (1 g, 3.62 mmol)
and 2-(2-fluorophenoxy)ethylamine (1.2 g, 7.24 mmol)
as described for (S)-1 but replacing ethanol with isopro-
25
25
°C; ½aꢂ ¼ þ59:2 (c 1, methanol) (½aꢂ ¼ þ25:05( c 1,
D
CHCl3) for the free amine); H NMR identical to that
D
1
of (S)-5.
5.1.36. (S)-2-[((2-(2-Trifluoromethylphenoxy)ethyl)amino)-
methyl]-1,4-benzodioxane hydrochloride [(S)-6]. Prepared
from (2R)-2-iodomethyl-1,4-benzodioxane (1.02 g,
37 mmol) and 32 (0.95g, 4.63 mmol) as described for
(S)-1. After chromatography on silica gel (dichloro-
panol. The secondary amine was isolated as an oil:
½aꢂ ¼ ꢀ31:72 (c 1, CHCl3). (S)-8 was obtained as a
25
D
white solid (0.51 g, 42% based on the starting amount of
(2R)-2-iodomethyl-1,4-benzodioxane): mp 148–154 °C;
25
½aꢂ ¼ ꢀ62:79 (c 1, CHCl3); 1H NMR (DMSO-d6) d
D
methane/methanol 98:2) the free amine was isolated as
3.26–3.31 (m, 1H), 3.41–3.46 (m, 3H), 4.03 (dd, 1H),
4.35–4.41 (m, 3H), 4.65–4.69 (m, 1H), 6.83–6.9 (m,
4H), 6.92–7.0 (m, 1H), 7.14–7.27 (m, 3H), 9.54 (br s,
2H): Anal. Calcd for C17H19O3FNCl (339.78).
25
D
as a white solid (370 mg, 26% based on the starting
an oil: ½aꢂ ¼ ꢀ28:3 (c 1, CHCl3). (S)-6 was isolated
amount of (2R)-2-iodomethyl-1,4-benzodioxane: mp
25
D
150–152 °C; ½aꢂ ¼ ꢀ48:1 (c 1, methanol) 1H NMR
(DMSO-d6) d 3.60–3.33 (m, 4H), 4.10 (dd, 1H), 4.43
(d, 1H), 4.56 (m, 2H), 4.75 (m, 1H), 6.93 (m, 4H), 7.20
(t, 1H), 7.37 (d, 1H), 7.70 (m, 2H), 9.90 (m, 2H). Anal.
Calcd for C18H19ClF3NO3 (389.79).
5.1.41. (R)-2-[((2-(2-Fluorophenoxy)ethyl)amino)methyl]-
1,4-benzodioxane hydrochloride [(R)-8]. Prepared from
(2S)-2-iodomethyl-1,4-benzodioxane and 2-(2-fluoro-
phenoxy)ethylamine as described for (S)-8: mp
25
25
213.4 °C; ½aꢂ ¼ þ58:06 (c 1, CHCl3) (½aꢂ ¼ þ31:22
D
D
1
5.1.37.
(R)-2-[((2-(2-Trifluoromethylphenoxy)ethyl)-
(c 1, CHCl3) for the free amine); H NMR identical to
that of (S)-8.
amino)methyl]-1,4-benzodioxane hydrochloride [(R)-6].
Prepared from (2S)-2-iodomethyl-1,4-benzodioxane
and 32 as described for (S)-6: mp 150.7 °C;
5.1.42. (S)-2-[((2-(2-Chlorophenoxy)ethyl)amino)methyl]-
1,4-benzodioxane hydrochloride [(S)-9]. Prepared from
25
25
½aꢂ ¼ þ48:08 (c 1, methanol) (½aꢂ ¼ þ27:19 (c 1,
D
D

L. Fumagalli et al. / Bioorg. Med. Chem. 13 (2005) 2547–2559
2557
(2R)-2-iodomethyl-1,4-benzodioxane (1 g, 3.62 mmol)
and 2-(2-chlorophenoxy)ethylamine (2.48 g, 14.48
mmol) as described for (S)-1 but replacing ethanol
with 2-methylpropanol. After purification by chroma-
tography on silica gel (dichloromethane/methanol
1H), 6.83–6.91 (m, 4H), 7.17–7.16 (m, 2H), 7.50–7.55
(m, 1H), 7.64–7.67 (m, 1H). Anal. Calcd for
C18H22O4NCl (383.87).
5.1.47. (R)-2-[((2-(2-Methanesulfinylphenoxy)ethyl)amino)-
methyl]-1,4-benzodioxane hydrochloride [(R)-11]. Pre-
pared from (2S)-2-iodomethyl-1,4-benzodioxane and
98:2) the secondary amine was isolated as a yellow oil
25
D
(880 mg, 76.5%): ½aꢂ ¼ ꢀ26:5 6 c( 1, CHCl3); 1H
25
NMR (CDCl3) d 1.95(br s, 1H), 2.95(m, 2H), 3.12
(m, 2H), 4.05(dd, 1H), 4.13–4.20 (m, 2H), 4.25–4.40
(m, 2H), 6.83–6.95(m, 6H), 7.21 (t, 1H), 7.36 (d,
1H). (S)-9 was isolated as a white solid (460 mg,
34 as described for (S)-11: mp 136 °C; ½aꢂ ¼ þ47:22
D
25
D
amine); H NMR identical to that of (S)-11.
(c 1, methanol) (½aꢂ ¼ þ28:6 (c 1, CHCl3) for the free
1
38.3% based on the starting amount of (2R)-2-iodo-
5.1.48.
(S)-2-[((2-(2-Methanesulfonylphenoxy)ethyl)-
25
methyl-1,4-benzodioxane): mp 164.6 °C; ½aꢂ ¼ ꢀ53:33
amino)methyl]-1,4-benzodioxane hydrochloride [(S)-12].
Prepared from (2R)-2-iodomethyl-1,4-benzodioxane
(520 mg, 1.88 mmol) and 35 (750 mg, 3.76 mmol) as de-
scribed for (S)-1 but replacing ethanol with isopropanol.
The secondary amine was isolated (460 mg, 67%) as an
D
(c 1, methanol); 1H NMR (DMSO-d6) d 3.42–3.57
(m, 4H), 4.13 (dd, 1H), 4.42–4.48 (m, 3H), 4.77 (m,
1H), 6.88–7.09 (m, 5H), 7.25 (m, 1H), 7.38 (t, 1H),
7.49 (d, 1H), 9.78 (br s, 2H). Anal. Calcd for
C17H19Cl2NO3 (356.23).
25
1
oil: ½aꢂ ¼ ꢀ15:91 (c 1, CHCl3); H NMR (CDCl3) d
D
2.15(s, 1H), 2.90–3.40 (m, 4H), 3.25(s, 3H), 4.00–4.60
(m, 5H), 6.80–8.10 (m, 4H). (S)-12 was isolated as a
white solid (250 mg, 39% based on the starting amount
of (2R)-2-iodomethyl-1,4-benzodioxane): mp 94–97 °C;
5.1.43. (R)-2-[((2-(2-Chlorophenoxy)ethyl)amino)methyl]-
1,4-benzodioxane hydrochloride [(R)-9]. Prepared from
(2S)-2-iodomethyl-1,4-benzodioxane and 2-(2-chloro-
phenoxy)ethylamine as described for (S)-9: mp
25
1
½aꢂ ¼ ꢀ44:09 (c 1, methanol); H NMR (DMSO-d6) d
D
25
D
25
D
164.6 °C; ½aꢂ ¼ þ53:28 (c 1, methanol) (½aꢂ
¼
3.29–3.31 (m, 6H), 3.40–3.53 (m, 2H), 4.0–4.06 (m,
1H), 4.33–4.37 (m, 1H), 4.53–4.63 (m, 2H), 6.8–7.00
(m, 4H), 7.19–7.24 (m, 1H), 7.37–7.36 (m, 1H), 7.72–
7.80 (m, 1H), 7.80–7.81 (m, 1H). Anal. Calcd for
C18H22ClNO5S (399.47).
þ29:22 (c 1, CHCl3) for the free amine); 1H NMR iden-
tical to that of (S)-9.
5.1.44.
methyl]-1,4-benzodioxane hydrochloride [(S)-10]. Pre-
pared from (2R)-2-iodomethyl-1,4-benzodioxane
(S)-2-[((2-(2-Methylthiophenoxy)ethyl)amino)-
5.1.49. (R)-2-[((2-(2-Methanesulfonylphenoxy)ethyl)-
amino)methyl]-1,4-benzodioxane hydrochloride [(R)-12].
Prepared from (2S)-2-iodomethyl-1,4-benzodioxane and
(1.11 g, 4.02 mmol) and 33 (1.47 g, 8.05mmol) as de-
scribed for (S)-1. The secondary amine was isolated as
25
D
25
D
1
an oil (442 mg, 34%): ½aꢂ ¼ ꢀ23:79 (c 1, CHCl3); H
35 as described for (S)-12: mp 93–96 °C; ½aꢂ ¼ þ45:84
25
(c 1, methanol) (½aꢂ ¼ þ17:43 (c 1, CHCl3) for the free
NMR (CDCl3) d 2.04 (s, 1H), 2.4 (s, 3H), 2.8–3.3 (m,
4H), 3.8–4.55 (m, 5H), 6.6–7.5 (m, 8H). (S)-10 was iso-
lated as a white solid (387 mg, 29% based on the starting
D
amine); H NMR identical to that of (S)-12.
1
amount of (2R)-2-iodomethyl-1,4-benzodioxane): mp
5.1.50. (S)-2-[((2-(2-Nitrophenoxy)ethyl)amino)methyl]-
1,4-benzodioxane hydrochloride [(S)-13]. A solution of
(2S)-aminomethyl-1,4-benzodioxane (1.5g, 9.08 mmol)
and 27 (1.4 g, 5.45 mmol) in 2-methylpropanol was re-
fluxed for 24 h. At the end, the solvent was removed
and the residue dissolved in dichloromethane (20 mL)
and treated with NaHCO3 (20 mL). The aqueous layer
was separated and extracted with dichloromethane
again. The organic phases were combined, washed with
water, dried and concentrated to give a residue, which
was purified by chromatography on silica gel. Elution
with dichloromethane/methanol (98:2) afforded 1.06 g
(59%) of (S)-2-[((2-(2-nitrophenoxy)ethyl)amino)methyl]-
25
D
167-169 °C; ½aꢂ ¼ ꢀ50:47 (c 1, methanol); 1H NMR
(DMSO-d6) d 2.4 (s, 3H), 3.1–3.8 (m, 5H), 6.7–7.45
(m, 8H), 9.85(m, 1H). Anal. Calcd for C 18H22ClNO3S
(367.88).
5.1.45. (S)-2-[((2-(2-Methylthiophenoxy)ethyl)amino)-
methyl]-1,4-benzodioxane hydrochloride [(R)-10]. Pre-
pared from (2S)-2-iodomethyl-1,4-benzodioxane and
33 as described for (S)-10: mp 167–170 °C; ½aꢂ
25
D
¼
25
þ49:3 (c 1, methanol) (½aꢂ ¼ þ23:7 (c 1, CHCl3) for
D
the free amine); H NMR identical to that of (S)-10.
1
5.1.46.
(S)-2-[((2-(2-Methanesulfinylphenoxy)ethyl)-
1,4-benzodioxane as
a
white solid: mp 75.6 °C;
25
D
1
amino)methyl]-1,4-benzodioxane hydrochloride [(S)-11].
Prepared from (2R)-2-iodomethyl-1,4-benzodioxane
(1.04 g, 3.76 mmol) and 34 (1.5g, 7.52 mmol) as de-
scribed for (S)-1 but replacing ethanol with isopropanol
½aꢂ ¼ ꢀ51:3 (c 1, methanol); H NMR (CDCl3) d 1.95
(br s, 1H), 2.9–3.04 (m, 2H), 3.11 (t, 2H), 3.99–4.09 (dd,
1H), 4.2–4.33 (m, 4H), 6.8–6.99 (m, 4H), 7.03–7.09 (m,
2H), 7.52 (dt, 1H), 7.86 (dd, 1H). The secondary amine
was dissolved in ethanol (2 mL) and 4.6 N HCl/EtOH
(2 mL) was slowly added. The resulting precipitate was
dried to give (S)-13 (700 mg, 36% based on the starting
(5mL). The secondary amine was isolated as an oil
25
D
(1.2 g, 92%): ½aꢂ ¼ ꢀ27:46 (c 1, CHCl3); 1H NMR
(CDCl3) d 2.20 (s, 1H), 2.85(s, 3H), 2.9–3.4 (m, 4H),
4.0–4.6 (m, 5H), 6.9–8.0 (m, 8H). (S)-11 was isolated as
a white solid (510 mg, 35% based on the starting amount
amount of 27) as
a
white solid: mp 155.1 °C;
25
D
1
½aꢂ ¼ ꢀ55:74 (c 1, methanol); H NMR (DMSO-d6) d
of (2R)-2-iodomethyl-1,4-benzodioxane): mp 134-
3.38–3.52 (m, 4H), 4.06–4.15 (dd, 1H), 4.45 (d, 1H),
4.60 (m, 2H), 4.73 (m, 1H), 6.93 (m, 4H), 7.23 (t, 1H),
7.09 (d, 1H), 7.45(t, 1H), 7.96 (d, 1H), 9.81 (br s, 2H).
Anal. Calcd for C17H19ClN2O5Æ1/2H2O (365.26).
25
136 °C; ½aꢂ ¼ ꢀ48:06 (c 1, methanol); 1H NMR
D
(DMSO-d6) d 2.73 (s, 3H), 3.26–3.33 (m, 2H), 3.40–3.46
(m, 3H), 4.0–4.1 (m, 1H), 4.36–4.40 (m, 3H), 4.41(m,

2558
L. Fumagalli et al. / Bioorg. Med. Chem. 13 (2005) 2547–2559
5.1.51. (R)-2-[((2-(2-Nitrophenoxy)ethyl)amino)methyl]-
1,4-benzodioxane hydrochloride [(R)-13]. Prepared from
(2R)-aminomethyl-1,4-benzodioxane and 27 as de-
as described by Testa et al.31) were resuspended in Tris
HCl, 50 mM, pH = 7.7 containing 10 lM pargyline
and 0.1% ascorbic acid and incubated for 30 min at
25 °C with 0.5nM [ 3H]-Prazosin (NEN, 80.5Ci/mmol)
in the absence or presence of different concentrations
of the tested compounds. Prazosin 1 M was used to
determine non-specific binding.
25
scribed for (S)-13: mp 154.8 °C; ½aꢂ ¼ þ57:2 (c 1, meth-
D
25
D
free amine); H NMR identical to that of (S)-13.
anol) (mp 75.7 °C; ½aꢂ ¼ þ53:22 (c 1, methanol) for the
1
5.1.52. (S)-2-[((2-(2-Acetylaminophenoxy)ethyl)amino)-
methyl]-1,4-benzodioxane hydrochloride [(S)-14]. Pre-
pared from (2R)-mesyloxymethyl-1,4-benzodioxane
(1.25g, 5.15mmol) and 40 (2 g, 10.2 mmol) as described
for (S)-2. After purification by chromatography on silica
gel (dichloromethane/MeOH–TEA 97.5/2.5/0.5) (S)-2-
[((2-(2-acetylaminophenoxy)ethyl)amino)methyl]-1,4-benzo-
dioxane was isolated (1.14 g, 65%) as a yellow oil:
Binding studies at 5-HT1A receptors were carried out
using crude membrane preparations from rat hippocam-
pus, which were resuspended in Tris HCl 50 mM
(pH = 7.7, 10 lM pargyline and 4 mM CaCl2) and incu-
bated for 30 min at 25 °C with 1 nM [3H]-8-OH-DPAT,
in the absence or presence of different concentrations of
the tested compounds. 5-HT1A 1 lM was used to deter-
mine non-specific binding.
25
1
½aꢂ ¼ ꢀ43:9 (c 1, methanol); H NMR (CDCl3) d 2.13
D
(br s, 1H), 2.18 (s, 3H), 2.97 (t, 2H), 3.08 (t, 2H), 4.01
(dd, 1H), 4.13 (t, 2H), 4.27 (dd, 1H), 4.33 (m, 1H), 6.74–
7.02 (m, 7H), 8.2 (s, 1H), 8.33 (dd, 1H). (S)-14 was iso-
lated as a white solid (650 mg, 33% based on the starting
Incubations were stopped by rapid filtration, through
GF/B fiber filters, which were then washed, dried and
counted in
spectrometer.
a LK1214 rack b-liquid scintillation
amount of (2R)-mesyloxymethyl-1,4-benzodioxane): mp
25
D
122.7 °C; ½aꢂ ¼ ꢀ56:84 (c 1, water); 1H NMR (DMSO-
d6) d 2.24 (s, 3H), 3.41–3.53 (m, 4H), 4.13 (dd, 1H), 4.35
(t, 2H), 4.41 (dd, 1H), 4.9 (m, 1H), 6.94–6.99 (m, 5H),
7.07 (m, 2H), 8.05(d, 1H), 9.67 (s, 1H), 9.8 (br s, 1H).
Anal. Calcd for C19H23N2O4Cl (378.86).
At least three different experiments, in triplicate, were
carried out for each compound and usually each com-
pound was tested simultaneously on the different a1-
AR subtypes. Prazosin or 5-HT1A were always tested
in parallel, as reference drugs. The percentage inhibi-
tory effects obtained in the different experiments were
pooled together and the inhibition curves were analyzed
using the ꢀone-site competitionꢁ equation built into
GraphPad Prism 4.0 (GraphPAD Software, San Diego,
CA). This analysis gives the IC50 (i.e., the drug concen-
tration inhibiting specific binding by 50%), calculated
with the relative standard error. Ki values were then
calculated by IC50 using the Cheng and Prusoff equa-
tion in which the Kd of [3H]-Prazosin for a1a, a1b, a1d
AR-subtypes were 0.4, 0.4 and 0.7 nM, respectively,
whereas the Kd of [3H]-8-OH-DPAT for 5-HT1A recep-
tors was 1.2 nM.
5.1.53. (R)-2-[((2-(2-Acetylaminophenoxy)ethyl)amino)-
methyl]-1,4-benzodioxane hydrochloride [(R)-14]. Pre-
pared from (2S)-mesyloxymethyl-1,4-benzodioxane and
25
40 as described for (S)-14: mp 117.1 °C; ½aꢂ ¼ þ48:76
D
25
D
amine); H NMR identical to that of (S)-14.
(c 1, water) (½aꢂ ¼ þ41:61 (c 1, water) for the free
1
5.1.54. (S)-2-[((2-(2-Aminophenoxy)ethyl)amino)methyl]-
1,4-benzodioxane hydrochloride [(S)-15]. A solution of
(S)-14 (0.85g, 2.48 mmol) in 6 N HCl (5mL) was heated
for 1 h. At the end water (10 mL) was added and then
concentrated to give the crude product, which was crys-
tallized from ethanol (16 mL) affording (S)-15 as a white
25
D
solid: mp 217 °C; ½aꢂ ¼ ꢀ56:65( c 1, methanol); 1H
Acknowledgements
NMR (DMSO-d6) d 3.30–3.50 (m, 4H), 4.13 (dd, 1H),
4.46–4.51 (m, 3H), 4.92 (m, 1H), 6.93–6.98 (m, 4H),
7.10 (t, 1H), 7.25(d, 1H), 7.43 (m, 2H), 10.0 (m, 5H).
Anal. Calcd for C17H22N2O3Cl2 (373.28).
Financial support provided by the Italian Ministero
`
dellꢁIstruzione, dellꢁUniversita e della Ricerca Scientifica
is gratefully acknowledged.
5.1.55. (R)-2-[((2-(2-Aminophenoxy)ethyl)amino)methyl]-
1,4-benzodioxane hydrochloride [(R)-15]. Prepared from
References and notes
25
D
(R)-14 as described for (S)-15: mp 217 °C; ½aꢂ
¼
1. (a) Ford, A. P. D. W.; Williams, T. J.; Blue, D. R.; Clarke,
D. E. Trends Pharmacol. Sci. 1994, 15, 167; (b) Hieble, J.
P.; Bylund, D. B.; Clarke, A. E.; Eikenberg, D. C.; Langer,
S. Z.; Lefkowitz, R. J.; Minneman, K. P.; Ruffolo, R. R.,
Jr. Pharmacol. Rev. 1995, 47, 266; (c) Hieble, J. P.;
Ruffolo, R. R., Jr. Prog. Drug Res. 1996, 47, 81.
1
þ51:63 (c 1, methanol); H NMR identical to that of
(S)-15.
5.2. Biology
2. Faure, C.; Pimoule, C.; Arbilla, S.; Langer, S. Z.; Graham,
D. Eur. J. Pharmacol. Mol. Pharmacol. 1994, 15, 167.
3. Malloy, B. J.; Price, D. T.; Price, R. R.; Bienstock, a. M.;
Dole, M. K.; Funk, B. L.; Rudner, X. L.; Richardson, C. D.;
Donatucci, C. F.; Schwinn, D. A. J. Urol. 1998, 160, 937.
4. Fenton, H.; Green, P. N.; Shapero, M.; Wilson, C. Nature
1965, 206, 725.
5.2.1. Binding assays. The pharmacological profile of
both the S and R enantiomers of compounds 1–16 was
assessed by measuring their affinities for a1a, a1b, a1d
AR-subtypes and 5-HT1A serotoninergic receptor with
in vitro binding studies.
Briefly, membranes derived from Chinese Hamster
Ovary (CHO) cells expressing a1-AR subtypes (prepared
5. Melchiorre, C.; Angeli, P.; Bolognesi, M.; L.Ferri, V.;
`
Chiarini, A.; Giardina, D.; Gulini, U.; Leonardi, A.;

L. Fumagalli et al. / Bioorg. Med. Chem. 13 (2005) 2547–2559
2559
Marucci, G.; Minarini, A.; Pigini, M.; Quaglia, W.;
Rosini, M.; Tumiatti, V. Pharm. Acta Helv. 2000, 74, 181.
6. Bolchi, C.; Catalano, P.; Fumagalli, L.; Gobbi, M.;
Pallavicini, M.; Pedretti, A.; Villa, L.; Vistoli, G.; Valoti,
E. Bioorg. Med. Chem. 2004, 12, 4937.
E.; Leonardi, A.; Melchiorre, C. J. Med. Chem. 2002, 45,
32.
18. Ferri, V.; Pallavicini, M.; Piccini, D.; Valoti, E.; Villa, L.
Farmaco 1988, 12, 1153.
19. Villa, L.; Valoti, E.; Villa, A. M.; Pallavicini, M.; Ferri, V.;
Iuliano, E.; Brunello, N. Farmaco 1994, 49, 587.
20. Valoti, E.; Pallavicini, M.; Villa, L.; Pezzetta, D. J. Org.
Chem. 2001, 66, 1018.
21. Kapur, H.; Mottram, P.; Green, P. N. J. Pharm.
Pharmacol. Commun. 1978, 30, 259.
`
7. Melchiorre, C.; Giardina, D.; Gallucci, P.; Brasili, L.
J. Pharm. Pharmacol. 1982, 34, 644.
`
8. Melchiorre, C.; Brasili, L.; Giardina, D.; Pigini, M.;
Strappaghetti, G. J. Med. Chem. 1984, 27, 1535.
9. Castagnino, E.; Strappaghetti, G.; Corsano, S.; Gallucci,
`
P.; Brasili, L.; Giardina, D. Farmaco (Ed. Sci.) 1984, 39,
569.
10. Giardina, D.; Angeli, P.; Brasili, L.; Gulini, U.; Melchi-
22. Dearden, J. J.; Cronin, M. T. D.; Higgins, C.; Mottram,
D. R.; Kapur, H. J. Pharm. Pharmacol. Commun. 1998, 4,
89.
23. Piascik, M. T.; Perez, D. J. Pharmacol. Exp. Ther. 2001,
298, 403.
24. Waugh, D. J.; Gaivin, R. J.; Zuscik, M. J.; Gonzalez-
Cabrera, P.; Ross, S. A.; Yun, J.; Perez, D. M. J. Biol.
Chem. 2001, 276, 25366.
`
orre, C.; Strappaghetti, G. Eur. J. Med. Chem. 1984, 19,
411.
11. Pigini, M.; Brasili, L.; Giannella, M.; Giardina, D.; Gulini,
`
U.; Quaglia, W.; Melchiorre, C. J. Med. Chem. 1988, 31,
2300.
12. Quaglia, W.; Pigini, M.; Giannella, M.; Melchiorre, C.
J. Med. Chem. 1990, 33, 2946.
25. Pedretti, A.; Silva, E. S.; Villa, L.; Vistoli, G. Biochem.
Biophys. Res. Commun. 2004, 319, 493.
13. Quaglia, W.; Pigini, M.; Tayebati, S. K.; Piergentili, A.;
Giannella, M.; Marucci, G.; Melchiorre, C. J. Med. Chem.
1993, 36, 1520.
14. Quaglia, W.; Pigini, M.; Tayebati, S. K.; Piergentili, A.;
Giannella, M.; Leonardi, A.; Taddei, C.; Melchiorre, C.
J. Med. Chem. 1996, 39, 2253.
26. Bolchi, C.; Fumagalli, L.; Moroni, B.; Pallavicini,
M.; Valoti, E. Tetrahedron: Asymmetry 2003, 14, 3779.
27. Swain, C. G.; Lupton, E. C. J. Am. Soc. Chem. 1968, 90,
4328.
28. Testa, B.; Seiler, P. Arzneim. Forsch Drug Res. 1981, 31,
1053.
15. Quaglia, W.; Pigini, M.; Piergentili, A.; Giannella, M.;
Marucci, G.; Poggesi, E.; Leonardi, A.; Melchiorre, C.
J. Med. Chem. 1999, 42, 2961.
16. Bolognesi, M. L.; Budriesi, R.; Cavalli, A.; Chiarini, A.;
Gotti, R.; Leonardi, A.; Minarini, A.; Poggesi, E.;
Recanatini, M.; Rosini, M.; Tumiatti, W.; Melchiorre,
C. J. Med. Chem. 1999, 42, 4214.
29. Verloop, A. The Sterimol Approach in Drug Design;
Marcel Dekker: New York, 1987.
30. Rey, S.; Caron, G.; Ermondi, G.; Gailllard, P.; Pagliara,
A.; Carrupt, P. A.; Testa, B. J. Mol. Graphics Modell.
2001, 19, 521.
31. Testa, R.; Taddei, C.; Poggesi, E.; Destefani, C.;
Cotecchia, S.; Hieble, J. P.; Sulpizio, A. C.; Naselsky,
D.; Bergsma, D.; Ellis, S.; Swif, A.; Ganguly, S.; Ruffolo,
R. R.; Leonardi, A. Pharmacol. Commun. 1995, 6, 79–86.
17. Quaglia, W.; Pigini, M.; Piergentili, A.; Giannella, M.;
Gentili, F.; Marucci, G.; Carrieri, A.; Carotti, A.; Poggesi,