Novel oxotremorine-related heterocyclic derivatives: Synthesis and in vitro pharmacology at the muscarinic receptor subtypes

Article abstract of DOI:10.1016/j.bmc.2007.09.003

A set of novel heterocyclic ligands (6-27) structurally related to Oxotremorine 2 was designed, synthesized and tested at muscarinic receptor subtypes (mAChRs). In the binding experiments at cloned human receptors (hm1-5), compounds 7 and 15 evidenced a remarkable affinity and selectivity for the hm2 subtype. The in vitro functional assays, performed on a selected group of derivatives at M₁, M₂, and M₃ tissue preparations, singled out the 3-butynyloxy-5-methylisoxazole trimethylammonium salt 7 as a potent unselective muscarinic agonist [pEC₅₀: 7.40 (M₁), 8.18 (M₂), and 8.14 (M₃)], whereas its 5-phenyl analogue 12 behaved as a muscarinic antagonist, slightly selective for the M₁ subtype [pK_B: 6.88 (M₁), 5.95 (M₂), 5.53 (M₃)]. Moreover, the functional data put in evidence that the presence of the piperidine ring may generate a functional selectivity, e.g., an M₁ antagonist/M₂ partial agonist/M₃ full agonist profile (compound 21), at variance with the corresponding quaternary ammonium salt (compound 22) which behaved as a muscarinic agonist at all M_1-3 receptors, with an appreciable selectivity for the cardiac M₂ receptors.

Full text of DOI:10.1016/j.bmc.2007.09.003

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 15 (2007) 7626–7637
Novel oxotremorine-related heterocyclic derivatives: Synthesis
and in vitro pharmacology at the muscarinic receptor subtypes
Clelia Dallanoce,a,* Marco De Amici,a Elisabetta Barocelli,b Simona Bertoni,b
Bryan L. Roth,c Paul Ernsbergerd and Carlo De Michelia
a
`
Istituto di Chimica Farmaceutica e Tossicologica ‘‘Pietro Pratesi’’, Universita degli Studi di Milano,
Via Mangiagalli 25, Milano 20133, Italy
Dipartimento di Scienze Farmacologiche, Biologiche e Chimiche Applicate, Universita degli Studi di Parma,
b
`
V.le GP Usberti 27/A, Parma 43100, Italy
cDepartment of Biochemistry Pharmacology, University of North Carolina Chapel Hill, Chapel Hill, NC 27599, USA
dDepartment of Nutrition, Case Western Reserve University, 10900 Euclid Avenue, Cleveland, OH 44106, USA
Received 3 May 2007; revised 29 August 2007; accepted 4 September 2007
Available online 11 September 2007
Abstract—A set of novel heterocyclic ligands (6–27) structurally related to Oxotremorine 2 was designed, synthesized and tested at
muscarinic receptor subtypes (mAChRs). In the binding experiments at cloned human receptors (hm1–5), compounds 7 and 15 evi-
denced a remarkable affinity and selectivity for the hm2 subtype. The in vitro functional assays, performed on a selected group of
derivatives at M1, M2, and M3 tissue preparations, singled out the 3-butynyloxy-5-methylisoxazole trimethylammonium salt 7 as a
potent unselective muscarinic agonist [pEC50: 7.40 (M1), 8.18 (M2), and 8.14 (M3)], whereas its 5-phenyl analogue 12 behaved as a
muscarinic antagonist, slightly selective for the M1 subtype [pKB: 6.88 (M1), 5.95 (M2), 5.53 (M3)]. Moreover, the functional data put
in evidence that the presence of the piperidine ring may generate a functional selectivity, e.g., an M1 antagonist/M2 partial agonist/
M3 full agonist profile (compound 21), at variance with the corresponding quaternary ammonium salt (compound 22) which
behaved as a muscarinic agonist at all M1–3 receptors, with an appreciable selectivity for the cardiac M2 receptors.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
the cerebral cortex, hippocampus, and striatum where
they are implicated in learning and memory. The M2 sub-
type is localized at presynaptic level in the brain and in
peripheral effector organs such as heart (M2) or smooth
muscles where it participates in a synergistic way with
the M3 subtype in the regulation of the contractility of
respiratory, gastrointestinal, and genitourinary tracts.
The M4 receptors, identified peripherally in the lung7
and centrally in the striatum, are believed to be involved
in the control of motor functions,8 cognitive processes,9
and antinociception.10,11 The M5 receptor mRNA has
been identified in various peripheral and cerebral blood
vessels and it has been demonstrated that dopaminergic
neurons innervating the striatum almost exclusively ex-
press the M5 receptor subtype;12 however, its physiolog-
ical role is far from being completely established.13
Recently, emerging evidence indicates that acetylcholine
acts also in a paracrine fashion through the activation
of the different muscarinic receptor subtypes located in
several non-innervated tissues such as keratinocytes, lym-
phocytes, placenta, and endothelial and ocular lens cells.
The muscarinic acetylcholine receptors (mAChRs) are
prototypical members of the superfamily of G protein-
linked receptors.1–3 The muscarinic actions of acetylcho-
line (ACh) are mediated by five molecularly distinct
mAChR subtypes (M1–M5)1–3 whose activation pro-
duces either an excitatory or an inhibitory modulation
of a number of central and peripheral physiological func-
tions.4 Different experimental approaches, including
immunohistochemical and mRNA hybridization tech-
niques, have shown that mAChRs are widely expressed
and critically regulate a number of physiological pro-
cesses.5,6 The M1 receptors are expressed at high density
in parasympathetic ganglia and forebrain areas including
Keywords: Muscarinic ligands; Oxotremorine-related compounds;
Muscarinic receptor subtypes; Cloned receptors; Binding assays;
Functional tests.
*
Corresponding author. Tel.: +39 02503 19327; fax: +39 02503
19326; e-mail: clelia.dallanoce@unimi.it
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.09.003

C. Dallanoce et al. / Bioorg. Med. Chem. 15 (2007) 7626–7637
7627
As autocrine mediator, acetylcholine seems to be in-
volved in the regulation of cell growth or proliferation
and in the release of smooth muscle relaxing agents.4
Derivatives 6–12 (Fig. 1) were readily prepared by
reacting known 3-hydroxy-5-methylisoxazole 28a25 or
3-hydroxy-5-phenylisoxazole 28b25 with 1,4-dichloro-2-
butyne to yield intermediates 29a, 29b, which were
sequentially treated with either dimethylamine or pyr-
rolidine to produce the corresponding free bases 30a,
30b and 31a, 31b, respectively (Scheme 1). Treatment
of the intermediates 30a and 31a with fumaric acid or
excess methyl iodide afforded fumarates 6 and 8 and
iodomethylates 7 and 9, respectively. The 5-phenyl-3-
oxyisoxazolyl oxalate 10 and fumarate 11, and methio-
dide 12 were likewise prepared from 30b and 31b
(Scheme 1). Similarly, the reaction of 3-hydroxyisoxaz-
ole 3226 with 4-chlorobut-2-yn-1-ol gave intermediate
33, which was reacted with N-hydroxyphthalimide
under the Mitsunobu protocol27–29 to yield derivative
34 (Scheme 2). Treatment of this intermediate with
hydrazine produced in good yield the oxime ether 35,
which was condensed with 3-quinuclidone and then
transformed into either fumarate 13 or iodomethylate 14.
During the past two decades, exploration of the role of
specific mAChR subtypes (M1–M5) in mediating the di-
verse physiological actions of ACh has been the subject
of detailed investigations. Such a knowledge is essential
for the development of novel therapeutic approaches
aimed at enhancing or inhibiting specific muscarinic
receptor-controlled responses. This is particularly true
for central mAChR subtypes, which regulate an extraor-
dinarily large number of cognitive, behavioral, sensory,
motor, and autonomic functions, whose imbalance is
implicated in the pathophysiology of several CNS dis-
eases, i.e. Alzheimer’s and Parkinson’s diseases, pain,
schizophrenia, depression, and epilepsy.14,15 To this
end, the discovery of new agonists4 has been a goal pur-
sued by numerous research groups in the effort to obtain
selective mAChR ligands clinically useful in the pallia-
tive treatment of the above-cited CNS neurodegenera-
tive diseases as well as in the therapy of peripheral
dysfunctions, e.g., intestinal hypomotility. However,
the task of assigning specific physiological functions to
the five mAChR subtypes has proved very challenging,
due to the lack of agonists provided with a high degree
of selectivity for the individual subtypes16 and to the co-
expression of multiple mAChRs in a variety of organs,
tissues, or cell types.5,6 These difficulties have been par-
tially circumvented by the generation of mutant mouse
strains deficient in each of the five mAChR subtypes.8
Isoxazolidin-3-one derivative 3720 was reacted with four
different amines (Scheme 3) to generate amines 38–41,
which were subsequently transformed into either oxa-
lates 15, 18, iodomethylates 16, 17, 19, and hydrocho-
ride 20. In a parallel way, mesylate 4220 (Scheme 4)
was converted into final salt derivatives 21–27.
3. Results and discussion
The new derivatives 6–27 were assayed for binding affin-
ity at human muscarinic receptor subtypes (hm1–5) in
transfected CHO cells labeled with [3H]quinuclidinyl
benzylate, and the results were compared with those of
well-recognized selective ligands and with the structur-
ally related derivative 3 (Table 1). The compounds
(10 lM) were preliminarily tested at hm2 and hm5, rep-
resentative of the two subgroups (M2, M4 and M1, M3,
M5, respectively) of muscarinic receptors. The deriva-
tives which displayed a percent inhibition of the radioli-
gand binding higher than 50% were then assayed at all
five receptor subtypes in order to evaluate their Ki values
(nM).
In the past, our research group investigated the pharma-
cological profile of a group of muscarinic ligands struc-
turally related to natural Muscarine 117–19 and to
Oxotremorine 220–23 (Fig. 1). The pharmacological pro-
file of representative derivatives, i.e. 3–5, was studied in
depth by taking into account both receptor occupancy
and in vivo tests. It emerged that the nonselective oxotre-
morine-like derivative 3 exhibited quite an interesting
analgesic activity.24 As an extension of our previous stud-
ies, we now report the synthesis and the results of the
pharmacological investigation of novel heterocyclic mus-
carinic ligands endowed with a butynyl side chain. On
one hand, we prepared the new 5-methyl/phenyl isoxaz-
ole derivatives 6–12 (Fig. 1) structurally related to 5, on
the other, we synthesized the group of derivatives 13–27
(Fig. 1), in which the terminal pyrrolidine function of
compounds 3–5 was replaced by more steric demanding
nonquaternized and quaternized moieties. To investigate
the effect of these structural modifications on the activity/
selectivity profiles at the five muscarinic receptor sub-
types, the target derivatives 6–27 were assayed at cloned
human muscarinic receptor subtypes (hm1–5), expressed
in Chinese Hamster Ovary (CHO) cells. Some of them
were also evaluated in functional tests at tissue prepara-
tions which contain M1À3 receptors.
The affinity, potency, and efficacy of compounds 6–9,
11–15, 21, and 22 were then evaluated by performing
the following three functional in vitro assays: rabbit
electrically stimulated vas deferens (M1),30 guinea pig
electrically driven left atrium (M2), and guinea pig ileum
(M3) (Table 2). Bethanechol and McN-A-343 were used
as the reference agonists at M2À3 and M1 subtypes,
respectively.
Inspection of the data reported in Table 1 puts in evi-
dence that in the set of novel derivatives 6–27, com-
pounds 7 and 15 displayed a remarkable affinity and
selectivity for the hm2 subtype; the Ki values obtained
for 7 and 15 (50 and 40 nM, respectively) at hm2 recep-
tors were lower than those of the reference isoxazolidi-
nyl derivative 3 (Ki = 70 nM). A qualitative analysis of
the structure–activity relationship evidences that
replacement of the pyrrolidine moiety in the isoxazoli-
2. Chemistry
The synthesis of target compounds was accomplished
along the reaction sequences depicted in Scheme 1–4.

7628
C. Dallanoce et al. / Bioorg. Med. Chem. 15 (2007) 7626–7637
HO
H3C
O
O
N
N
+
NMe3
N
N
O
O
3
2
1
R1
R2
N
N
R3
O
O
N
N
R
O
O
O
4: Isoxazolinyl
5: Isoxazolyl
6: R=R1=R2=Me; R3=H
7: R=R1=R2=R3=Me
O
N
8: R=Me; R1-R2=(CH2)4; R3=H
9: R=Me; R1-R2=(CH2)4; R3=Me
10: R=Ph; R1=R2=Me; R3=H
11: R=Ph; R1-R2=(CH2)4; R3=H
12: R=Ph; R1-R2=(CH2)4; R3=Me
N
R
13: R=H
14: R=Me
N
O
R
N
R
N
O
O
N
O
O
15: R=H
21: R=H
22: R=Me
N
16: R=Me
O
R
N
R
N
O
O
N
17: R=Me
23: R=H
24: R=Me
N
O
R
N
R
N
O
N
O
O
O
18: R=H
25: R=H
26: R=Me
19: R=Me
N
O
O
R
N
R
O
N
O
N
20: R=H
27: R=H
N
Figure 1. Structure of model and target compounds.
din-3-one derivative 3 with the piperidine ring (15)
almost doubles the affinity at hm2 receptors. This results
in increased selectivity of 15 for hm2, as indicated by
affinity ratios, over hm1 (20), hm3 (13), hm4 (20), and
hm5 (14) subtypes. The corresponding affinity ratios
for the parent compound 3 were 13, 5.6, 3.2, and 7.3,
respectively. Worth noting, the corresponding methio-
dide 16 is devoid of any affinity for the five mAChR
subtypes.
ing dimethylamino derivative 6 which is a feeble ligand
for hm2 and hm4 (Ki values 3.8 and 1.1 lM, respec-
tively) and completely inactive at hm1, hm3, and hm5.
A set of derivatives (6–9, 11–15, 21, and 22) was selected
for evaluation in functional assays to assess their phar-
macological profile (Table 2). In this group of com-
pounds, we chose to insert also derivatives 8 and 14,
characterized by a negligible binding affinity, and 13,
whose binding data were unavailable. The most interest-
ing results refer to the comparison of the data of deriv-
ative 3 with those of its homologue 15 as well as the data
of derivative 6 with those of its methiodide 7. As shown
in Table 2, the replacement of the pyrrolidine moiety
Even higher values of hm2 selectivity characterize the
affinity profile of 5-methyl-3-oxyisoxazolyl trimethylam-
monium salt 7, e.g., hm1/hm2 > 200; hm3/hm2 80; hm4/
hm2 14, hm5/hm2 126 at variance with the correspond-

C. Dallanoce et al. / Bioorg. Med. Chem. 15 (2007) 7626–7637
7629
H3C
N
H3C
O
b
N
R
O
Cl
30a,b
HO
O
a
a: R = Me
b: R = Ph
N
N
R
R
O
O
N
28a,b
a: R = Me
b: R = Ph
29a,b
O
c
N
R
O
31a,b
d
e
f
d
6
8
9
30a
31a
31b
7
e
d
11
12
30b
10
e
Scheme 1. Reagents: (a) 1,4-dichloro-2-butyne, K2CO3, acetone; (b) NHMe2, DMF; (c) Pyrrolidine, DMSO; (d) C4H4O4, 2-propanol-MeOH;
(e) CH3I, MeOH-ether; (f) C2H2O4, ether.
O
HO
N
O
HO
O
a
b
O
O
N
N
O
O
O
N
O
N OH
34
32
33
O
H2N
O
N
O
N
O
O
c
d
34
N
N
O
O
35
36
e
f
13
14
36
Scheme 2. Reagents: (a) K2CO3, acetone, 4-chlorobut-2yn-1-ol; (b) PPh3, DEAD, THF; (c) H2NNH2ÆH2O, Et2O; (d) 3-quinuclidone, MeOH;
(e) C4H4O4, 2-propanol-MeOH; (f) CH3I, 2-propanol.
(derivative 3) with its piperidine homologue (derivative
15) brings about a sharp change in the pharmacological
profile. Indeed, compound 3 is a relatively potent,
nonselective muscarinic agonist (pEC50 = 7.31–7.98 at

7630
C. Dallanoce et al. / Bioorg. Med. Chem. 15 (2007) 7626–7637
e
f
15
16
O
N
a
N
O
O
38
N
O
b
f
17
N
O
Cl
39
40
41
N
O
e
f
N
N
O
18
19
37
c
N
O
O
O
d
g
20
N
Scheme 3. Reagents: (a) Piperidine, Cs2CO3, acetone; (b) Indoline, MeOH; (c) 1,2,3,4-Tetrahydroisoquinoline, MeOH; (d) Nortropane, Cs2CO3,
MeOH; (e) C2H2O4, 2-propanol-MeOH; (f) CH3I, MeOH-Et2O; (g) HCl-Et2O.
N
e
f
21
22
a
O
N
43
44
O
O
g
f
N
23
24
b
H3CO2SO
O
O
N
N
42
O
e
f
N
N
25
26
c
O
N
45
46
O
O
d
e
O
27
N
Scheme 4. Reagents: (a) Piperidine, MeOH; (b) Indoline, MeOH; (c) 1,2,3,4-Tetrahydroisoquinoline, MeOH; (d) Nortropane, Cs2CO3, MeOH;
(e) C2H2O4, 2-propanol-MeOH; (f) CH3I, MeOH-Et2O; (g) HCl-Et2O.
M1–M3 subtypes; a = 1), whereas 15 turns out to be a
functionally selective derivative characterized by a mod-
erate antagonist activity at M1 and M3 receptors
(pKB = 6.45 and 5.55, respectively; a = 0) and a partial
agonism, with exiguous intrinsic activity, at the M2 sub-
type (pKB = 5.94; a = 0.2) (Table 2).

C. Dallanoce et al. / Bioorg. Med. Chem. 15 (2007) 7626–7637
7631
Table 1. Binding affinity (Ki, nM) of compounds 6–27 at cloned
human muscarinic receptor subtypes (hm1–5) expressed in CHO cells
file. As a matter of fact, whereas the dimethylamino
derivative 6 behaves as a moderately potent agonist at
M2 and M3 receptors and as a partial agonist at M1
receptors, the corresponding methiodide 7 becomes a
potent full agonist at all three receptor subtypes with a
100-fold increase in both potency and affinity, e.g.,
pEC50 7.40 (M1), 8.18 (M2), and 8.14 (M3) for 7 versus
pEC50 5.46 (M1), 6.29 (M2), and 6.10 (M3) for 6.
Compound
hm1
hm2
hm3
hm4
hm5
Pirenzepine
Methoctramine
21.38
35.48
10.38
4-DAMP
3
0.68
392
905
70
222
510
6
>104
>104
3796
50
>104
3995
1095
682
>104
6310
>104
7514
>104
6885
1836
nta
7
8
>104
2041
>104
>104
>104
nta
The replacement of the N,N-dimethylamino or trimethy-
lammonium head with the pyrrolidine nucleus yields a
remarkable drop in the ability to bind and to activate
the muscarinic receptors as demonstrated by the pKB
or pEC50 values calculated for compounds 8 and 9 in
comparison with those of derivatives 6 and 7. The
further substitution of the methyl group with the more
lipophilic and larger phenyl group generates the 3-buty-
nyloxy-5-phenylisoxazole derivative 11, behaving as a
full M1 agonist (pEC50 = 4.35), a partial M2 agonist
(pEC50 = 4.66, a = 0.64), and an M3 antagonist
(pKB = 4.94, a = 0), and the corresponding methiodide
12, behaving as an antagonist at the muscarinic recep-
tors with a weak M1 selectivity (pKB = 6.88).
9
7680
7386
5932
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
6141
3376
4013
1448
5478
5986
>104
40
>104
544
819
537
792
>104
>104
>104
>104
>104
876
>104
>104
>104
>104
>104
5139
4116
>104
>104
>104
5664
>104
4550
7367
3919
4740
6041
1681
219
>104
4099
>104
474
>104
6958
5539
887
>104
2265
According to the change of the pharmacological profile
observed on passing from compound 3–15, in the
3-butynyloxy-D2-isoxazoline series replacement of the
pyrrolidine moiety with the piperidine ring caused
the conversion of a nonselective agonist [compound 4,
pEC50 = 6.76 (M1), 7.0 (M2), 7.41 (M3)] into a function-
ally selective M1 antagonist/M2 partial agonist/M3 full
>104
a nt, not tested.
As previously observed for structurally related ana-
logues,20 the transformation of the dimethylamine moi-
ety into the corresponding trimethylammonium salt
gives rise to a substantial change in the functional pro-
agonist (compound 21, M1: pKB = 5.91; M2: pEC50
=
5.96, a = 0.6; M3: pEC50 = 5.75, a = 1), although
characterized by a reduced potency. Conversely, its
Table 2. In vitro functional activity of compounds 6–9, 11–15, 21, and 22 at muscarinic receptor subtypes in rabbit vas deferens (M1), guinea pig left
atrium (M2) and guinea pig ileum (M3)
Compound
M1
Rabbit vas deferensa
M2
Guinea-pig left atrium
M3
Guinea-pig ileum
iac pKD
pKB
pEC50
iac pKD
pKB
pEC50
iac pKD
pKB
b
d
e
b
d
e
b
d
e
pEC50
Bethanechol
6.19 0.07 1.0 5.37 0.16
6.37 0.05 1.0 4.90 0.13
McN-A-343 6.40 0.04 1.0 7.20 0.28
3f
4f
6
7.31 0.11 1.0 6.52 0.20
6.76 0.16 1.0 6.44 0.18
7.73 0.10 1.0 6.73 0.29
7.00 0.07 1.0 6.40 0.11
6.29 0.20 1.0 3.56 0.22
8.18 0.14 1.0 7.03 0.12
0
7.98 0.12 1.0 6.91 0.28
7.41 0.03 1.0 6.38 0.39
6.10 0.07 1.1 4.40 0.11
8.14 0.12 1.0 6.17 0.13
5.46 0.19 0.6 5.45 0.24
7.40 0.09 0.9 6.97 0.06
7
8
0
5.66 0.27
4.81 0.07 4.82 0.12 0.4 4.81 0.37
6.26 0.11 0.9 5.78 0.47
9
0
6.01 0.01 6.60 0.06 1.0 6.72 0.32
4.66 0.06 0.6 ntg
0
11
12
13
14
15
21
22
4.35 0.03 1.0 ntg
ntg
6.88 0.15
ntg
5.95 0.25
0
4.94 0.15
5.53 0.12
0
0
0
5.37 0.27 0.9 3.82 0.16
5.85 0.49 5.51 0.06 0.7 5.97 0.10
<4
5.99 0.09 0.9 5.44 0.08
5.45 0.07 0.7 5.03 0.48
0
0
6.45 0.07
0.2
5.94 0.05
5.55 0.04
0
6.10 0.12 0.4 6.51 0.19
5.91 0.08 5.96 0.29 0.6 6.22 0.34
7.34 0.05 1.0 6.74 0.18
5.75 0.01 1.0 4.66 0.21
6.57 0.08 1.0 6.53 0.19
a See Ref. 30.
b pEC50 SE values are the negative logarithm of the agonist concentration that caused 50% of the maximum response attainable in that tissue.
c Intrinsic activity (a) was measured by the ratio between the maximum response of the compound and the maximum response of the reference
agonist.
d Apparent pKD SE values were calculated according to Furchgott33 and McKay.34
e Apparent pKB SE values were calculated according to Furchgott.33
f All the reported values were taken from the literature.20,21
g nt, not tested.

7632
C. Dallanoce et al. / Bioorg. Med. Chem. 15 (2007) 7626–7637
corresponding methiodide 22 becomes a full agonist of
higher potency at M2 and M3 subtypes (pEC50 = 7.34
and 6.57, respectively), and a partial agonist at M1
receptors (pEC50 = 6.10; a = 0.4). Methiodide 22 is the
only compound under study which shows an agonist
profile coupled with a degree of M2 selectivity.
4.1.2.2. 3-[(4-Chlorobut-2-yn-1-yl)oxy]-5-phenylisox-
azole (29b). Yellow oil, bp 165–170 ꢁC/0.5 mm Hg. Rf
0.56 (15% ethyl acetate/petroleum ether). 1H NMR:
4.20 (t, 2H, J = 1.8), 4.98 (t, 2H, J = 1.8), 6.19 (s, 1H),
7.43 (m, 3H), 7.71 (m, 2H). Anal. Calcd for
C13H10ClNO2: C, 63.04; H, 4.07; N, 5.66. Found: C,
63.40; H, 3.82; N, 5.38.
In summary, the data collected on the novel Oxotremo-
rine-like compounds indicate the 3-butynyloxy-5-
methyl-isoxazole moiety as a scaffold suitable to
generate potent muscarinic ligands. These results lay
the basis for further investigations aimed at finding
out the structural determinants needed to discriminate
the different muscarinic receptor subtypes.
A solution of 29a (0.49 g, 2.64 mmol) in N,N-dimethyl-
formamide (13 mL) and a tenfold excess of dimethyl-
amine was stirred at rt in a sealed metal container for
3 h. The crude reaction mixture was concentrated at
reduced pressure (50 ꢁC/0.5 mm Hg) and the residue, dis-
solved in 2 N HCl (5 mL), was treated with diethyl ether
(3 · 5 mL). The residual aqueous phase was made alkaline
by portionwise addition of solid K2CO3 and extracted
with ethyl acetate (3 · 5 mL). After the usual work-up,
the oily residue was purified by silica gel column chroma-
tography (eluant: 5% methanol/dichloromethane) to
afford the desired tertiary amine 30a (0.31 g, 60%).
4. Experimental
4.1. Chemistry
4.1.1. Materials and methods. 3-Hydroxy-5-methylisox-
azole 28a,25 3-hydroxy-5-phenylisoxazole 28b,25
3-hydroxyisoxazole 32,26 isoxazolidin-3-one 37,20 and
3-substituted-D2-isoxazoline 4220 were prepared accord-
ing to procedures described in the literature. Melting
4.1.2.3. N,N-Dimethyl-4-[(5-methylisoxazol-3-yl)oxy]-
but-2-yn-1-amine (30a). Colorless oil, bp 120–125 ꢁC/
0.5 mm Hg. Rf 0.30 (5% methanol/dichloromethane).
1H NMR: 2.28 (s, 6H), 2.32 (s, 3H), 3.31 (t, 2H,
J = 1.8), 4.88 (t, 2H, J = 1.8), 5.65 (s, 1H). Anal. Calcd
for C10H14N2O2: C, 61.84; H, 7.27; N, 14.42. Found:
C, 61.65; H, 7.48; N, 14.19.
points were determined on a model B 540 Buchi appara-
¨
tus and are uncorrected. Liquid compounds were char-
acterized by the oven temperature for Kugelrohr
distillations. 1H NMR spectra were recorded with a
Varian Mercury 300 (300 MHz) spectrometer in CDCl3
(unless otherwise specified) solutions at 20 ꢁC. Chemical
shifts (d) are expressed in ppm and coupling constants
(J) in Hertz. TLC analyses were performed on commer-
cial silica gel 60 F254 aluminum sheets: spots were fur-
ther evidenced by spraying with a dilute alkaline
potassium permanganate solution. Microanalyses (C,
H, Cl, and N) of new compounds agreed with the
theoretical value within 0.4%.
The procedure described above for the synthesis of 30a
was applied on a comparable amount of 29b to afford
derivative 30b in 52% yield.
4.1.2.4. N,N-Dimethyl-4-[(5-phenylisoxazol-3-yl)oxy]-
but-2-yn-1-amine (30b). Yellow oil, bp 170–175 ꢁC/
0.5 mm Hg. Rf 0.43 (5% methanol/dichloromethane);
1H NMR: 2.46 (s, 6H), 3.52 (t, 2H, J = 1.8), 4.86 (t,
2H, J = 1.8), 5.62 (s, 1H), 7.28 (m, 3H), 7.52 (m, 2H).
Anal. Calcd for C15H16N2O2: C, 70.29; H, 6.29; N,
10.93. Found: C, 69.95; H, 6.55; N, 10.78.
4.1.2. Synthesis of compounds 6–12. To a solution of
28a25 (3 g, 30.28 mmol) in acetone (40 mL) was added
potassium carbonate (8.3 g, 60.05 mmol). The suspen-
sion was heated at reflux for 1 h, then cooled at rt 1,4-Di-
chloro-2-butyne (17.59 mL, 0.180 mol) was then added
dropwise and, after heating at reflux for 24 h, the crude
reaction mixture was poured into water (120 mL) and
extracted with dichloromethane (3· 60 mL). The pooled
organic extracts were dried over anhydrous sodium
sulfate and concentrated at reduced pressure. The dark
brown residue was chromatographed on silica gel
(eluant: 10% ethyl acetate/petroleum ether) to give
derivative 29a (1.95 g, 35% yield).
A parallel protocol was applied to the reaction of 29a
(1 g, 5.39 mmol) with pyrrolidine (900 lL, 10.77 mmol)
in dimethylsulfoxide (5 mL) for 3 h. After the above-de-
scribed work-up, the oily residue was purified by silica
gel column chromatography (eluant: 2% methanol/
dichloromethane) to provide the desired tertiary amine
31a (0.56 g, 47%).
4.1.2.5. 5-Methyl-3-[(4-pyrrolidin-1-ylbut-2-yn-1-yl)oxy]-
isoxazole (31a). Pale yellow oil, bp 155–160 ꢁC/2 mm Hg.
Rf 0.50 (5% methanol/dichloromethane). 1H NMR: 1.58
(m, 4H), 2.10 (s, 3H), 2.38 (m, 4H), 3.21 (t, 2H, J = 1.9),
4.62 (t, 2H, J = 1.9), 5.41 (s, 1H). Anal. Calcd for
C12H16N2O2: C, 65.43; H, 7.32; N, 12.72. Found: C,
65.12; H, 7.09; N, 13.05.
4.1.2.1. 3-[(4-Chlorobut-2-yn-1-yl)oxy]-5-methylisoxaz-
ole (29a). Light yellow oil, bp 130–135 ꢁC/0.5 mm Hg. Rf
0.38 (10% ethyl acetate/petroleum ether). 1H NMR: 2.32
(s, 3H), 4.20 (t, 2H, J = 1.8), 4.90 (t, 2H, J = 1.8), 5.65 (s,
1H). Anal. Calcd for C8H8ClNO2: C, 51.77; H, 4.34; N,
7.55. Found: C, 52.02; H, 4.09; N, 7.31.
Intermediate 29b was similarly reacted with pyrrolidine
to give derivative 31b in 80% yield.
The procedure described above for the synthesis of 29a
was similarly applied to 28b25 to give derivative 29b in
30% yield.
4.1.2.6. 5-Phenyl-3-[(4-pyrrolidin-1-ylbut-2-yn-1-yl)oxy]-
isoxazole (31b). Mp 64–65 ꢁC (yellow prisms from diiso-
propyl ether). Rf 0.33 (5% methanol/dichloromethane).

C. Dallanoce et al. / Bioorg. Med. Chem. 15 (2007) 7626–7637
7633
1H NMR: 1.80 (m, 4H), 2.62 (m, 4H), 3.48 (t, 2H,
J = 1.9), 4.98 (t, 2H, J = 1.9), 6.18 (s, 1H), 7.42 (m,
3H), 7.72 (m, 2H). Anal. Calcd for C17H18N2O2: C,
72.32; H, 6.43; N, 9.92. Found: C, 72.61; H, 6.12; N,
10.22.
(t, 2H, J = 1.8), 5.80 (s, 1H). Anal. Calcd for
C13H19IN2O2: C, 43.11; H, 5.29; N, 7.73. Found: C,
42.97; H, 5.48; N, 7.98.
4.1.2.13. 5-Phenyl-3-[(4-pyrrolidin-1-ylbut-2-yn-1-
yl)oxy]isoxazole methiodide (12). Mp 124–127 ꢁC (pale
yellow prisms from 90% absolute ethanol/diethyl ether).
1H NMR (D2O): 2.05 (m, 4H), 3.01 (s, 3H), 3.39 (m,
2H), 3.50 (m, 2H), 4.19 (t, 2H, J = 1.9), 4.92 (t, 2H,
J = 1.9), 6.42 (s, 1H), 7.41 (m, 3H), 7.69 (m, 2H). Anal.
Calcd for C18H21IN2O2: C, 50.96; H, 4.99; N, 6.60.
Found: C, 51.19; H, 5.18; N, 6.38.
To a solution of fumaric or oxalic acid (2.20 mmol) in
7 mL of 2-propanol/methanol (7:3) was added the
appropriate tertiary base (1.0 mmol). The corresponding
salt precipitated quantitatively on standing.
4.1.2.7. N,N-Dimethyl-4-[(5-methylisoxazol-3-yl)oxy]-
but-2-yn-1-amine fumarate 6 (30aÆC4H4O4). Mp 99–
101 ꢁC (colorless prisms from 90% 2-propanol/diethyl
4.1.3. Synthesis of compounds 13 and 14. To a solution of
3226 (5 g, 58.78 mmol) in acetone (50 mL) was added
potassium carbonate (16 g, 115.76 mmol). The suspen-
sion was heated at reflux for 1 h and then cooled at rt
4-chlorobut-2-yn-1-ol (13 g, 125 mmol)31 was hence
added and, after heating at reflux for 18 h, the reaction
mixture was filtered under vacuum. The filtrate was
poured into water (500 mL), extracted with dichloro-
methane (3· 200 mL), and the pooled organic extracts
were dried over anhydrous sodium sulfate and concen-
trated at reduced pressure. The dark brown residue
was chromatographed on silica gel (eluant: 30% ethyl
acetate/petroleum ether) to give derivative 33 (4.6 g,
51% yield).
1
ether). H NMR (D2O): 2.18 (s, 3H), 2.76 (s, 6H), 3.92
(bs, 2H), 4.80 (bs, 2H), 5.75 (s, 1H), 6.50 (s, 2H). Anal.
Calcd for C14H18N2O6: C, 54.19; H, 5.55; N, 9.03.
Found: C, 54.37; H, 5.10; N, 9.21.
4.1.2.8. 5-Methyl-3-[(4-pyrrolidin-1-ylbut-2-yn-1-yl)oxy]-
isoxazole fumarate 8 (31aÆC4H4O4). Mp 107–111 ꢁC
(colorless prisms from 2-propanol). 1H NMR (CD3OD):
2.05 (m, 4H), 2.34 (s, 3H), 3.35 (m, 4H), 4.12 (t, 2H,
J = 1.9), 4.92 (t, 2H, J = 1.9), 5.82 (s, 1H), 6.68 (s,
2H). Anal. Calcd for C16H20N2O6: C, 57.14; H, 5.99;
N, 8.33. Found: C, 57.04; H, 6.13; N, 8.15.
4.1.2.9. 5-Phenyl-3-[(4-pyrrolidin-1-ylbut-2-yn-1-yl)oxy]-
isoxazole fumarate 11 (31bÆ1/2 C4H4O4). Mp 115–116 ꢁC
(pale yellow prisms from 90% 2-propanol/diethyl ether).
1H NMR (CD3OD): 1.98 (m, 4H), 3.12 (m, 4H), 3.92 (t,
2H, J = 1.9), 5.01 (t, 2H, J = 1.9), 6.51 (s, 1H), 6.63 (s,
1H), 7.48 (m, 3H), 7.79 (m, 2H). Anal. Calcd for
C19H20N2O4: C, 67.05; H, 5.92; N, 8.23. Found: C,
67.29; H, 6.13; N, 8.15.
4.1.3.1. 4-(Isoxazol-3-yloxy)but-2-yn-1-ol (33). Yellow
oil, bp 165–170 ꢁC/0.5 mm Hg. Rf 0.38 (10% ethyl ace-
tate/petroleum ether). H NMR: 2.80 (bs, 1H), 4.30 (t,
2H, J = 1.8), 4.92 (t, 2H, J = 1.8), 6.00 (d, 1H,
J = 1.5), 8.13 (d, 1H, J = 1.5). Anal. Calcd for
C7H7NO3: C, 54.90; H, 4.61; N, 9.15. Found: C,
54.52; H, 4.95; N, 8.83.
1
4.1.2.10. N,N-Dimethyl-4-[(5-phenylisoxazol-3-yl)ox-
y]but-2-yn-1-amine oxalate 10 (30bÆC2H2O4). Mp 188–
189 ꢁC (colorless prisms from 90% methanol/diethyl
To a solution of 33 (3.45 g, 22.50 mmol), 2-hydroxy-iso-
indole-1,3-dione (3.67 g, 22.50 mmol) and triphenylphos-
phine (5.90 g, 22.5 mmol) in anhydrous tetrahydrofuran
(60 mL) was added dropwise a 40% toluene solution of
diethyl azodicarboxylate (4.12 mL, 22.50 mmol). The
mixture was stirred for 18 h at rt and then concentrated
in vacuo. The residue was purified by silica gel column
chromatography (eluant: 10% ethyl acetate/petroleum
ether) to give derivative 34 (3.41 g, 51% yield).
1
ether). H NMR (CD3OD): 2.91 (s, 6H), 4.12 (bs, 2H),
5.04 (bs, 2H), 6.51 (s, 1H), 7.48 (m, 3H), 7.78 (m, 2H).
Anal. Calcd for C17H18N2O6: C, 58.96; H, 5.24; N,
8.09. Found: C, 58.76; H, 5.16; N, 7.97.
The following procedure is representative. A solution of
30a (0.220 g, 1.13 mmol) in 2-propanol (5 mL) was trea-
ted with a fivefold excess of methyl iodide at rt The cor-
responding methiodide precipitated quantitatively upon
addition of anhydrous diethyl ether.
4.1.3.2. 2-{[4-(Isoxazol-3-yloxy)but-2-yn-1-yl]oxy}-1H-
isoindole-1,3(2H)-dione (34). Colorless needles, mp 65–
67 ꢁC (from n-hexane/ethyl acetate). Rf 0.31 (30% ethyl
acetate/petroleum ether). 1H NMR: 4.91 (t, 2H,
J = 1.7), 4.94 (t, 2H, J = 1.7), 5.95 (d, 1H, J = 2.0),
7.78 (m, 2H), 7.86 (m, 2H), 8.09 (d, 1H, J = 2.0). Anal.
Calcd for C15H10N2O5: C, 60.41; H, 3.38; N, 9.39.
Found: C, 60.66; H, 3.60; N, 9.12.
4.1.2.11. N,N-Dimethyl-4-[(5-methylisoxazol-3-yl)ox-
y]but-2-yn-1-amine methiodide (7). Mp 186–188ꢁC (col-
orless prisms from absolute ethanol). 1H NMR
(CD3OD): 2.32 (s, 3H), 3.21 (s, 9H), 4.40 (t, 2H,
J = 1.8), 5.00 (t, 2H, J = 1.8), 5.86 (s, 1H). Anal. Calcd
for C11H17IN2O2: C, 39.30; H, 5.10; N, 8.33. Found:
C, 39.55; H, 4.87; N, 8.11.
To a solution of 34 (1.08 g, 3.62 mmol) in diethyl ether
(5 mL) was added hydrazine monohydrate (0.18 g,
3.62 mmol). After vigorous stirring for 15 min at rt, fur-
ther diethyl ether (25 mL) was added and the reaction
mixture was stirred for additional 15 min. The white
precipitate was filtered off under vacuum and to the fil-
trate was added dropwise 5 mL of a 3.5 N HCl ethereal
solution. The amorphous precipitated hydrochloride of
4.1.2.12. 5-Methyl-3-[(4-pyrrolidin-1-ylbut-2-yn-1-
yl)oxy]isoxazole methiodide (9). Mp 72–74 ꢁC (pale yel-
low prisms from 90% absolute ethanol/diethyl ether). 1H
NMR (D2O): 2.09 (m, 4H), 2.00 (s, 3H), 3.01 (s, 3H),
3.40 (m, 2H), 3.50 (m, 2H), 4.19 (t, 2H, J = 1.8), 4.85

7634
C. Dallanoce et al. / Bioorg. Med. Chem. 15 (2007) 7626–7637
35 was separated by decantation and used in the follow-
ing step without further purification.
(m, 2H), 1.60 (m, 4H), 2.46 (m, 4H), 2.77 (t, 2H,
J = 8.2), 3.26 (t, 2H, J = 1.7), 4.32 (t, 2H, J = 1.7),
4.36 (t, 2H, J = 8.2). Anal. Calcd for C12H18N2O2: C,
64.84; H, 8.16; N, 12.60. Found: C, 65.20; H, 7.87; N,
12.29.
A solution of the O-alkyl-hydroxylamine hydrochloride
35 (0.71 g, 3.47 mmol) and 3-quinuclidone (0.43 g,
3.47 mmol) in methanol (50 mL) was stirred at rt for
18 h. The crude reaction mixture was concentrated at re-
duced pressure and the residue was dissolved in 2 N HCl
(5 mL) and treated with diethyl ether (3 · 5 mL). The
residual aqueous phase was made alkaline by portion-
wise addition of solid K2CO3 and extracted with ethyl
acetate (3 · 5 mL). After the usual work-up, the oily res-
idue was purified by silica gel column chromatography
(eluant: 5% methanol/dichloromethane) to afford the de-
sired oxime ether 36 (0.52 g, 54%).
According to the procedure described for 38, derivatives
39, 40, and 41 were prepared by reacting 37 with indo-
line, 1,2,3,4-tetrahydroisoquinoline, and nortropane,32
respectively. Compound 39 was obtained heating at re-
flux the reaction mixture for 24 h whereas derivative
41 was formed in the presence of cesium carbonate
(equimolar with nortropane) at rt.
4.1.4.2. 2-[4-(2,3-Dihydro-1H-indol-1-yl)but-2-yn-1-
yl]isoxazolidin-3-one (39). Thick pale yellow oil (60%
1
yield). Rf 0.21 (20% ethyl acetate/petroleum ether). H
4.1.3.3. Quinuclidin-3-one O-[4-(isoxazol-3-yloxy)but-
2-yn-1-yl]oxime (36). Colorless viscous oil. Rf 0.42 (10%
methanol/dichloromethane). H NMR: 1.70 (m, 4 H),
NMR: 2.62 (t, 2H, J = 8.1), 2.97 (t, 2H, J = 7.9), 3.39
(t, 2H, J = 7.9), 3.92 (t, 2H, J = 1.6), 4.21 (t, 2H,
J = 8.1), 4.23 (t, 2H, J = 1.6), 6.58 (d, 1H, J = 7.1),
6.73 (t, 1H, J = 7.1), 7.09 (m, 2H). Anal. Calcd for
C15H16N2O2: C, 70.29; H, 6.29; N, 10.93. Found: C,
69.95; H, 6.53; N, 11.18.
1
2.52 (m, 1H), 2.67–2.94 (m, 4H), 3.52 (s, 2H), 4.58 (bs,
2H), 4.81 (bs, 2H), 5.89 (d, 1H, J = 1.9), 8.03 (d, 1H,
J = 1.9). Anal. Calcd for C14H17N3O3: C, 61.08; H,
6.22; N, 15.26. Found: C, 60.76; H, 6.57; N, 15.45.
Fumarate 13 and methiodide 14 were prepared accord-
ing to the above-described procedures.
4.1.4.3. 2-[4-(3,4-Dihydroisoquinolin-2(1H)-yl)but-2-yn-
1-yl]isoxazolidin-3-one (40). Thick yellow oil (51% yield).
Rf 0.14 (40% ethyl acetate/petroleum ether). H NMR:
1
4.1.3.4. Quinuclidin-3-one O-[4-(isoxazol-3-yloxy)but-
2-yn-1-yl]oxime fumarate 13 (36Æ3/2 C4H4O4). Mp 132–
137 ꢁC, dec (colorless prisms from 90% 2-propanol/
2.78 (t, 2H, J = 8.1), 2.82 (t, 2H, J = 6.1), 2.94 (t, 2H,
J = 6.1), 3.52 (t, 2H, J = 1.7), 3.78 (s, 2H), 4.32 (t, 2H,
J = 1.7), 4.37 (t, 2H, J = 8.1), 7.02 (m, 1H), 7.11 (m,
3H). Anal. Calcd for C16H18N2O2: C, 71.09; H, 6.71;
N, 10.36. Found: C, 71.30; H, 6.55; N, 10.08.
1
diisopropyl ether). H NMR (CD3OD): 2.04 (m, 2H),
2.16 (m, 2H), 2.85 (m, 1H), 3.30–3.45 (m, 4H), 4.15 (s,
2H), 4.63 (t, 2H, J = 1.9), 4.93 (t, 2H, J = 1.9), 6.13 (d,
1H, J = 1.8), 6.61 (s, 3H), 8.40 (d, 1H, J = 1.8). Anal.
Calcd for C20H23N3O9: C, 53.45; H, 5.16; N, 9.35.
Found: C, 53.63; H, 5.22; N, 9.58.
4.1.4.4. 2-[4-(8-Azabicyclo[3.2.1]oct-8-yl)but-2-yn-1-
yl]isoxazolidin-3-one (41). Viscous colorless oil (53%
yield). Rf 0.67 (20% methanol/dichloromethane). 1H
NMR: 1.35–1.50 (m, 4H), 1.64 (m, 2H), 1.85–2.05 (m,
4H), 2.74 (t, 2H, J = 8.1), 3.30 (t, 2H, J = 1.7), 3.38
(bs, 2H), 4.31 (t, 2H, J = 1.7), 4.38 (t, 2H, J = 8.1). Anal.
Calcd for C14H20N2O2: C, 67.71; H, 8.12; N, 11.28.
Found: C, 67.36; H, 8.41; N, 11.47.
4.1.3.5. Quinuclidin-3-one O-[4-(isoxazol-3-yloxy)but-
2-yn-1-yl]oxime methiodide (14). Mp 160–175 ꢁC, dec.
(colorless prisms from 90% 2-propanol/methanol). H
1
NMR (D2O): 1.97 (m, 2H), 2.16 (m, 2H), 2.80 (m,
1H), 3.00 (s, 3H), 3.38 (m, 2H), 3.52 (m, 2H), 4.26 (m,
2H), 4.61 (t, 2H, J = 1.6), 4.82 (t, 2H, J = 1.6), 6.09 (d,
1H, J = 1.8), 8.23 (d, 1H, J = 1.8). Anal. Calcd for
C15H20IN3O3: C, 43.18; H, 4.83; N, 10.07. Found: C,
43.29; H, 4.60; N, 9.85.
Oxalates 15 and 18 and methiodides 16 and 19 were pre-
pared according to the above-described procedures.
Iodomethylate 17 was obtained heating at reflux an ace-
tone solution of 39 for 6 h in the presence of a fivefold
excess of iodomethane. Derivatives 17 and 19 were puri-
fied by means of a silica gel column chromatography
and were obtained in 63% and 70% yield, respectively.
4.1.4. Synthesis of compounds 15–20. To a solution of
3720 (0.6 g, 3.46 mmol) in methanol (25 mL) was added
piperidine (684 lL, 6.92 mmol). After stirring at room
temperature for 36 h, the solvent was removed at re-
duced pressure, the crude reaction mixture was dissolved
in 2 N HCl (5 mL) and treated with diethyl ether (3 ·
5 mL). The residual aqueous phase was made alkaline
by portionwise addition of solid K2CO3 and extracted
with ethyl acetate (3 · 5 mL). After the usual work-up,
the oily residue was purified by silica gel column chro-
matography (eluant: 2% methanol/dichloromethane)
affording the desired tertiary amine 38 (0.54 g, 70%).
4.1.4.5. 2-(4-Piperidin-1-ylbut-2-yn-1-yl)isoxazolidin-3-
one 15 (38ÆC2H2O4). Mp 117–118 ꢁC (colorless prisms
from 90% 2-propanol/diethyl ether); H NMR (D2O):
1
1.29–1.42 (m, 1H), 1.50–1.72 (m, 3H), 1.79–1.90 (m,
2H), 2.79 (t, 2H, J = 8.2), 2.89 (dt, 2H, J = 2.6 and
12.1 and), 3.42–3.51 (m, 2H), 3.87 (t, 2H, J = 1.8),
4.33 (t, 2H, J = 1.8), 4.35 (t, 2H, J = 8.2). Anal. Calcd
for C14H20N2O6: C, 53.84; H, 6.45; N, 8.97. Found: C,
53.56; H, 6.34; N, 8.68.
4.1.4.1. 2-(4-Piperidin-1-ylbut-2-yn-1-yl)isoxazolidin-3-
one (38). Colorless oil, bp 155–160 ꢁC/0.5 mm Hg. Rf
0.64 (10% methanol/dichloromethane). H NMR: 1.42
4.1.4.6. 2-(4-Piperidin-1-ylbut-2-yn-1-yl)isoxazolidin-3-
one methiodide (16). Light yellow gummy solid. 1H
NMR (CD3OD): 1.60–1.80 (m, 2H), 1.83–1.98 (m, 4H),
1

C. Dallanoce et al. / Bioorg. Med. Chem. 15 (2007) 7626–7637
7635
2.81 (t, 2H, J = 8.1), 3.17 (s, 3H), 3.30 (t, 2H, J = 1.8)
3.48 (m, 2H), 4.33-4.49 (m, 6H). Anal. Calcd for
C13H21IN2O2: C, 42.87; H, 5.81; N, 7.69. Found: C,
43.22; H, 5.49; N, 7.91.
4.1.5.2. 1-[4-(4,5-Dihydroisoxazol-3-yloxy)but-2-yn-1-
yl]indoline (44). Thick yellow oil (81% yield). Rf 0.57
1
(30% ethyl acetate/petroleum ether); H NMR: 2.95 (t,
2H, J = 9.5), 2.98 (t, 2H, J = 8.1), 3.42 (t, 2H, J = 8.1),
3.99 (t, 2H, J = 1.8), 4.40 (t, 2H, J = 9.5), 4.72 (t, 2H,
J = 1.8), 6.58 (d, 1H, J = 7.7), 6.73 (t, 1H, J = 7.3),
7.10 (m, 2H). Anal. Calcd for C15H16N2O2: C, 70.29;
H, 6.29; N, 10.93. Found: C, 69.92; H, 6.51; N, 11.23.
4.1.4.7. 2-[4-(2,3-Dihydro-1H-indol-1-yl)but-2-yn-1-
yl]isoxazolidin-3-one methiodide (17). Light yellow gum-
my solid. Rf 0.22 (10% methanol/dichloromethane); H
1
NMR (CD3OD): 2.74 (t, 2H, J = 7.5), 3.50 (m, 2H),
3.62 (s, 3H), 4.22 (m, 1H), 4.31 (t, 2H, J = 7.5), 4.35
(t, 2H, J = 1.6), 4.45 (m, 1H), 4.81 (m, 2H), 7.58 (m,
3H), 7.78 (d, 1H, J = 7.6). Anal. Calcd for
C16H19IN2O2: C, 48.26; H, 4.81; N, 7.03. Found: C,
47.90; H, 5.10; N, 7.35.
4.1.5.3. 2-[4-(4,5-Dihydroisoxazol-3-yloxy)but-2-yn-1-
yl]-1,2,3,4-tetrahydroisoquinoline (45). Thick colorless oil
(65% yield). Rf 0.63 (40% ethyl acetate/petroleum ether);
1H NMR: 2.84 (t, 2H, J = 5.9), 2.94 (t, 2H, J = 5.9), 2.99
(t, 2H, J = 9.5), 3.56 (t, 2H, J = 1.8), 3.77 (s, 2H), 4.41 (t,
2H, J = 9.5), 4.80 (t, 2H, J = 1.8), 7.03-7.20 (m, 4H).
Anal. Calcd for C16H18N2O2: C, 71.09; H, 6.71; N,
10.36. Found: C, 71.42; H, 6.32; N, 10.59.
4.1.4.8. 2-[4-(3,4-Dihydroisoquinolin-2(1H)-yl)but-2-yn-
1-yl]isoxazolidin-3-one oxalate 18 (40ÆC2H2O4). Mp 157–
162 ꢁC (colorless prisms from 10% methanol/2-propa-
1
nol). H NMR (CD3OD): 2.79 (t, 2H, J = 8.1), 3.19 (t,
4.1.5.4. 8-[4-(4,5-Dihydroisoxazol-3-yloxy)but-2-yn-1-
yl]-8-azabicyclo[3.2.1]octane (46). Thick colorless oil
(57% yield). Rf 0.71 (20% methanol/dichloromethane);
1H NMR: 1.38–1.56 (m, 4H), 1.63 (m, 2H), 1.81–1.99
(m, 4H), 2.99 (t, 2H, J = 9.5), 3.27 (t, 2H, J = 1.8),
3.40 (bs, 2H), 4.41 (t, 2H, J = 9.5), 4.78 (t, 2H,
J = 1.8). Anal. Calcd for C14H20N2O2: C, 67.71; H,
8.12; N, 11.28. Found: C, 67.78; H, 8.39; N, 11.48.
2H, J = 6.3), 3.54 (t, 2H, J = 6.3), 4.15 (s, 2H), 4.38 (t,
2H, J = 8.1), 4.40 (m, 4H), 7.15-7.35 (m, 4H). Anal.
Calcd for C18H20N2O6: C, 59.99; H, 5.59; N, 7.77.
Found: C, 59.71; H, 5.52; N, 7.61.
4.1.4.9. 2-[4-(3,4-Dihydroisoquinolin-2(1H)-yl)but-2-yn-
1-yl]isoxazolidin-3-one methiodide (19). Pale yellow waxy
compound. Rf 0.41 (10% methanol/dichloromethane).
1H NMR (CD3OD): 2.83 (t, 2H, J = 8.1), 3.35 (s, 3H),
3.95 (m, 4H), 4.41 (t, 2H, J = 8.1), 4.49 (m, 2H), 4.52
(m, 2H), 4.79 (m, 2H), 7.21–7.40 (m, 4H). Anal. Calcd
for C17H21IN2O2: C, 49.53; H, 5.13; N, 6.80. Found:
C, 49.84; H, 5.47; N, 6.53.
Hydrochloride 23, oxalates 21, 25, and 27, methiodides
22, 24, and 26 were prepared according to the above-de-
scribed protocols. Derivatives 24 and 26 were purified
by means of a silica gel column chromatography.
4.1.5.5. 1-[4-(4,5-Dihydroisoxazol-3-yloxy)but-2-yn-1-
yl]piperidine oxalate 21 (43ÆC2H2O4). Mp 131–133ꢁC
(colorless prisms from 2-propanol); 1H NMR (CD3OD):
1.62-1.73 (m, 2H,), 1.80–1.92 (m, 4H), 3.01 (t, 2H,
J = 9.8), 3.25-3.37 (m, 4H), 4.06 (t, 2H, J = 1.8), 4.39
(t, 2H, J = 9.8), 4.87 (t, 2H, J = 1.8). Anal. Calcd for
C14H20N2O6: C, 53.84; H, 6.45; N, 8.97. Found: C,
53.83; H, 6.44; N, 8.80.
To a solution of 41 (0.13 g, 0.52 mmol) in diethyl ether
(3 mL) was added 3 N HCl in diethyl ether (1 mL).
The corresponding salt precipitated as a very hygro-
scopic colorless powder. After washing several times
with anhydrous diethyl ether, the hydrochloride was
dried under vacuum.
4.1.4.10. 2-[4-(8-Azabicyclo[3.2.1]oct-8-yl)but-2-yn-1-
yl]isoxazolidin-3-one hydrochloride (20). Thick yellow
oil. 1H NMR (CD3OD): 1.82 (m, 4H), 1.92–2.12
(m, 4H), 2.25 (m, 2H), 2.67 (t, 2H, J = 7.9), 3.99
(bs, 2H), 4.09 (bs, 2H), 4.12 (bs, 2H), 4.21 (t, 2H,
J = 7.9).
4.1.5.6. 1-[4-(4,5-Dihydroisoxazol-3-yloxy)but-2-yn-1-
yl]piperidine methiodide (22). Colorless gummy solid. 1H
NMR (CD3OD): 1.59–1.79 (m, 2H), 1.93 (m, 4H), 3.03
(t, 2H, J = 9.9), 3.17 (s, 3H), 3.43–3.54 (m, 4H), 4.40 (t,
2H, J = 9.9), 4.42 (t, 2H, J = 1.8), 4.90 (t, 2H, J = 1.8).
Anal. Calcd for C13H21IN2O2: C, 42.87; H, 5.81; N,
7.69. Found: C, 43.21; H, 5.59; N, 7.52.
4.1.5. Synthesis of compounds 21–27. Derivatives 43, 44,
45, and 46 were prepared, according to the procedure
described for 38, by reacting mesylate 42 for 12 h at rt
with two equivalents of piperidine, indoline, 1,2,3,4-tet-
rahydroisoquinoline, and nortropane, respectively. The
tertiary base 46 was obtained in the presence of cesium
carbonate (equimolar with 42).
4.1.5.7. 1-[4-(4,5-Dihydroisoxazol-3-yloxy)but-2-yn-1-
yl]indoline hydrochloride (23). Mp 118–119ꢁC (colorless
prisms from 2-propanol/diethyl ether). 1H NMR
(CD3OD): 2.94 (t, 2H, J = 9.9), 3.30 (t, 2H, J = 7.7),
3.92 (t, 2H, J = 7.7), 4.36 (t, 2H, J = 9.9), 4.49 (t, 2H,
J = 1.8), 4.79 (t, 2H, J = 1.8), 7.26–7.43 (m, 4H). Anal.
Calcd for C15H17ClN2O2: C, 61.54; H, 5.85; Cl, 12.11;
N, 9.57. Found: C, 61.40; H, 5.97; Cl, 12.32; N, 9.38.
4.1.5.1. 1-[4-(4,5-Dihydroisoxazol-3-yloxy)but-2-yn-1-
yl]piperidine (43). Colorless liquid, bp 155–160 ꢁC/5 mm
Hg (80% yield). Rf 0.64 (10% methanol/dichlorometh-
1
ane). H NMR: 1.43 (m, 2H), 1.60 (m, 4H), 2.47 (m,
4.1.5.8. 1-[4-(4,5-Dihydroisoxazol-3-yloxy)but-2-yn-1-
yl]indoline methiodide (24). Pale yellow waxy compound
(78% yield). Rf 0.54 (10% methanol/dichloromethane).
1H NMR (CD3OD): 2.96 (t, 2H, J = 9.5), 3.45–3.54
(m, 2H), 3.61 (s, 3H), 4.21 (m, 1H), 4.39 (t, 2H,
4H), 2.98 (t, 2H, J = 9.5), 3.30 (t, 2H, J = 1.8), 4.41 (t,
2H, J = 9.5), 4.79 (t, 2H, J = 1.8). Anal. Calcd for
C12H18N2O2: C, 64.84; H, 8.16; N, 12.60. Found: C,
64.58; H, 8.28; N, 12.31.

7636
C. Dallanoce et al. / Bioorg. Med. Chem. 15 (2007) 7626–7637
J = 9.5), 4.46 (m, 1H), 4.89 (bs, 4H), 7.57 (m, 3H), 7.78
(d, 1H, J = 8.1). Anal. Calcd for C16H19IN2O2:
C,48.26; H, 4.81; N, 7.03. Found: C, 48.56; H, 4.55;
N, 6.78.
CO2 inhalation. Isolated preparations were set up fol-
lowing the techniques previously described.21
4.2.1. Electrically stimulated rabbit vas deferens. Accord-
ing to Eltze,36 the prostatic portion of each vas deferens
was mounted in a 10 mL organ bath, containing a mod-
ified Krebs solution (mM composition: NaCl 134, KCl
3.4, CaCl2 2.8, KH2PO4 1.3, NaHCO3 16, MgSO4 0.6,
and glucose 7.7) kept at 31 ꢁC and bubbled with 95%
O2-5% CO2. Yohimbine (1.0 lM) was present through-
out the experiment to prevent prejunctional a2-adreno-
ceptos stimulation. For isometric recordings, the
tissues were left to equilibrate for 45 min under a passive
load of 0.75 g before electrical field stimulation through
platinum electrodes was applied by square-wave pulses
(0.5 ms, 0.05 Hz, supramaximal intensity 450 mA;
LACE Elettronica Mod. ES3, Ospedaletto PI, Italy).
4.1.5.9. 2-[4-(4,5-Dihydroisoxazol-3-yloxy)but-2-yn-1-yl]-
1,2,3,4-tetrahydroisoquinoline oxalate 25 (45ÆC2H2O4). Mp
157–159 ꢁC (colorless prisms from 2-propanol). 1H
NMR (CD3OD): 3.02 (t, 2H, J = 9.5), 3.18 (t, 2H,
J = 6.6), 3.56 (t, 2H, J = 6.6), 4.20 (t, 2H, J = 1.8),
4.38 (t, 2H, J = 9.5), 4.42 (s, 2H), 4.89 (t, 2H,
J = 1.8), 7.25 (m, 4H). Anal. Calcd for C18H20N2O6:
C, 59.99; H, 5.59; N, 7.77. Found: C, 59.80; H, 5.76;
N, 7.52.
4.1.5.10. 2-[4-(4,5-Dihydroisoxazol-3-yloxy)but-2-yn-1-
yl]-1,2,3,4-tetrahydroisoquinoline methiodide (26). Pale
yellow waxy compound (85% yield). 1H NMR
(CD3OD): 3.04 (t, 2H, J = 9.5), 3.27 (s, 3H), 3.33 (m,
2H), 3.78–3.94 (m, 2H), 4.40 (t, 2H, J = 9.5), 4.49 (t,
2H, J = 1.8), 4.70 (d, 1H, J = 16.0), 4.73 (d, 1H,
J = 16.0), 4.93 (t, 2H, J = 1.8), 7.24 (d, 1H, J = 7.3),
7.35 (m, 3H). Anal. Calcd for C17H21IN2O2: C, 45.93;
H, 5.13; N, 6.80. Found: C, 45.61; H, 5.40; N, 6.54.
4.2.2. Electrically stimulated guinea pig left atrium. The
left atria were mounted in 20 mL organ baths under
0.5 g tension at 33 ꢁC, immersed in a modified Krebs–
Henseleit solution (mM composition: NaCl 118.9, KCl
4.6, CaCl2 2.5, KH2PO4 1.2, NaHCO3 25, MgSO4.7H2O
1.2, and glucose 11.1), gassed with a 95%O2–5%CO2
mixture. After a period of stabilization of 45 min, tissues
were electrically stimulated through platinum electrodes
by square-wave submaximal pulses (2 Hz, 5 ms, 5 V)
and inotropic activity was recorded isometrically.
4.1.5.11. 8-[4-(4,5-Dihydroisoxazol-3-yloxy)but-2-yn-
1-yl]-8-azabicyclo[3.2.1]octane oxalate 27 (46ÆC2H2O4).
Colorless amorphous solid. 1H NMR: 1.67 (m, 4H), 1.98
(m, 2H), 2.18 (m, 2H), 2.35 (m, 2H), 3.01 (t, 2H,
J = 9.5), 3.83 (bs, 2H), 4.10 (bs, 2H), 4.43 (t, 2H,
J = 9.5), 4.78 (bs, 2H). Anal. Calcd for C16H22N2O6:
C, 56.80; H, 6.55; N, 8.28. Found: C, 56.59; H, 6.81;
N, 8.47.
4.2.3. Guinea pig ileum. Ileal segments 2–3 cm long were
set up under 1.0 g tension at 37 ꢁC in 10 mL organ baths
filled with Krebs-Henseleit solution (see above), bubbled
with carbogen. Tissues were allowed to equilibrate for
45 min and afterwards contractile responses were iso-
metrically recorded.
4.2. Pharmacology
Stable cell lines expressing the hm1–hm5 muscarinic
receptors were obtained via the resources of the NIMH
PDSP and were tested for receptor expression by radio-
ligand binding to L-quinuclidinyl [phenyl-4-3H]benzylate
([3H]QNB). Saturation experiments were performed
using 5–2500 pM [3H]QNB in 50 mM Tris–HCl buffer,
pH 7.4, to determine the KD of [3H]QNB for each recep-
tor subtype. Competition binding experiments were
performed using 180 pM [3H]QNB and 10–50 lg of
membrane protein. All binding experiments were per-
formed in the presence of 1% DMSO. Non-specific bind-
ing was defined by 0.5 lM atropine. After equilibrium
was reached (120 min incubation at 25 ꢁC), bound and
free radioactivity were separated by filtration using
Whatman GF-C filters. Filters were covered with 6 mL
of counting cocktail (Eco-Scint, Research Products
International) and radioactivity was measured in a scin-
tillation counter (Beckman) at 40% efficiency. The IC50
values were obtained by nonlinear curve fitting to a lo-
gistic equation (Prism, GraphPad Software, San Diego,
CA). Ki values were derived from IC50 values using
Cheng–Prusoff equation.35
4.2.4. Protocols. Agonist concentration–response curves
were constructed in each tissue by cumulative applica-
tion of concentrations of the test compounds.37 The
agonist potency was expressed as pEC50 (Àlog EC50)
calculated by linear regression analysis using the least
square method. Intrinsic activity (ia) was calculated as
a fraction of the maximal response to the reference full
agonist, Bethanechol or McN-A-343. Concentration-re-
sponse curves of the agonists were reconstructed in the
presence of atropine 0.1 lM and hexamethonium
100 lM. When the compounds were tested as antago-
nists, a dose–response curve to the full agonists Bethane-
chol or McN-A-343 was repeated after 30 min
incubation with the test compounds (10 nM–100 lM).
Apparent affinity (pKD) of drugs behaving as full ago-
nists was estimated by means of the receptor inactiva-
tion technique33 using phenoxybenzamine (1–10 lM
for 20–30 min) as the alkylating agent. For partial ago-
nists pKD was estimated by the method of McKay34 by
comparing their concentration-response curves with
those of the reference full agonist. The potency of the
compounds acting as antagonists was expressed as
pKB value, the calculated molar concentration of the test
compounds that causes a twofold increase in the EC50
values of the muscarinic agonists used in the functional
tests calculated according to Furchgott’s method.33 All
data are expressed as means of 6–8 experiments.
Male guinea pigs (250–350 g) and New Zealand white
rabbits (3.0–3.5 kg) (Morini, S. Polo, Italy) were used.
The tissues for in vitro experiments were removed from
animals fasted 24 h before the experiments and killed by

C. Dallanoce et al. / Bioorg. Med. Chem. 15 (2007) 7626–7637
7637
19. Angeli, P.; Cantalamessa, F.; Cavagna, R.; Conti, P.; De
Amici, M.; De Micheli, C.; Gamba, A.; Marucci, G.
J. Med. Chem. 1997, 40, 1099.
20. Dallanoce, C.; Conti, P.; De Amici, M.; De Micheli, C.;
Barocelli, E.; Chiavarini, M.; Ballabeni, V.; Bertoni, S.;
Impicciatore, M. Bioorg. Med. Chem. 1999, 7, 1539, and
references cited therein.
Acknowledgments
This research has been supported by grants from the
University of Milano. The authors wish to thank Dr.
Giuseppe Domenichini (University of Parma) for his
skillful technical assistance.
21. Barocelli, E.; Ballabeni, V.; Bertoni, S.; Dallanoce, C.; De
Amici, M.; De Micheli, C.; Impicciatore, M. Life Sci.
2000, 67, 717.
References and notes
22. De Amici, M.; Conti, P.; Fasoli, E.; Barocelli, E.;
Ballabeni, V.; Bertoni, S.; Impicciatore, M.; Roth, B.
L.; Ernsberger, P.; De Micheli, C. Farmaco 2003, 58,
739.
23. Disingrini, T.; Muth, M.; Dallanoce, C.; Barocelli, E.;
Bertoni, S.; Kellershohn, K.; Mohr, K.; De Amici, M.;
Holzgrabe, U. J. Med. Chem. 2006, 49, 366.
24. Barocelli, E.; Ballabeni, V.; Bertoni, S.; De Amici, M.;
Impicciatore, M. Life Sci. 2001, 68, 1775.
1. Caulfield, M. P. Pharmacol. Ther. 1993, 58, 319.
2. Wess, J. Crit. Rev. Neurobiol. 1996, 10, 69.
3. Caulfield, M. P.; Birdsall, N. J. M. Pharmacol. Rev. 1998,
50, 279.
4. Eglen, R. M. Auton. Autacoid. Pharmacol. 2006, 26, 219.
5. Levey, A. I. Life Sci. 1993, 52, 441.
6. Vilaro, M. T.; Mengod, G.; Palacios, J. M. Prog. Brain
Res. 1993, 98, 95.
7. Chelala, J. L.; Kilani, A.; Miller, M. J.; Martin, R. J.;
Ernsberger, P. Pharmacol. Toxicol. 1998, 83, 200.
8. Wess, J. Annu. Rev. Pharmacol. Toxicol. 2004, 44, 423.
9. Mulugeta, E.; Karlsson, E.; Islam, A.; Kalaria, R.;
Mangat, H.; Winblad, B.; Adem, A. Brain Res. 2003,
960, 259.
10. Sheardown, M. J.; Shannon, H. E.; Swedberg, M. D.;
Suzdak, P. D.; Bymaster, F. P.; Olesen, P. H.; Mitch, C.
H.; Ward, J. S.; Sauerberg, P. J. Pharmacol. Exp. Ther.
1997, 281, 868.
11. Mulugeta, E.; El-Bakri, N.; Karlsson, E.; Elhassan, A.;
Adem, A. Brain Res. 2003, 982, 284.
12. Yamada, M.; Basile, A. S.; Fedorova, I.; Zhang, W.;
Duttaroy, A.; Cui, Y.; Lamping, K. G.; Faraci, F. M.;
Deng, C.-X.; Wess, J. Life Sci. 2003, 74, 345, and
references cited therein.
25. Sorensen, U. S.; Falch, E.; Krogsgaard-Larsen, P. J. Org.
Chem. 2000, 65, 1003.
26. Iwai, I.; Nakamura, N. Chem. Pharm. Bull. 1966, 14,
1277.
27. Mitsunobu, O. Synthesis 1981, 1, 1.
28. Grochowski, E.; Jurczak, J. Synthesis 1976, 682.
29. Xu, R.; Sim, M.-K.; Go, M.-L. J. Med. Chem. 1998, 41,
3220.
30. This functional test is commonly used to characterize
ligands for M1 receptors. However, Melchiorre et al.
suggested that inhibition of the neurogenic contractions in
this tissue preparation could be mediated by M4 receptors:
Budriesi, R.; Cacciaguerra, S.; Di Toro, R.; Bolognesi, M.
L.; Chiarini, A.; Minarini, A.; Rosini, M.; Spampinato, S.;
Tumiatti, V.; Melchiorre, C. Br. J. Pharmacol. 2001, 132,
1099.
´
31. Gomez, C.; Huerta, F. F.; Pastor, I. M.; Yus, M.
Tetrahedron 1997, 53, 17201.
13. Eglen, R. M.; Nahorski, S. R. Br. J. Pharmacol. 2000, 130,
13.
14. Power, A. E. Curr. Med. Chem. 2004, 11, 987.
15. Sur, C.; Kinney, G. G. Curr. Neuropharmacol. 2005, 3, 63.
16. Felder, C. C.; Bymaster, F. P.; Ward, J.; De-Lapp, N.
J. Med. Chem. 2000, 43, 4333, and references cited therein.
17. De Amici, M.; De Micheli, C.; Grana, E.; Rodi, R.; Zonta,
F.; Santagostino-Barbone, M. G. Eur. J. Med. Chem.
1989, 24, 171.
32. Daly, J. W.; Gupta, T. H.; Padgett, W. L.; Pei, X.-F.
J. Med. Chem. 2000, 43, 2514.
33. Furchgott, R. F.; Bursztyn, P. Ann. N. Y. Acad. Sci. 1967,
144, 882.
34. McKay, D. Br. J. Pharmacol. 1966, 26, 9.
35. Cheng, Y.-C.; Prusoff, W. H. Biochem. Pharmacol. 1973,
22, 3099.
36. Eltze, M. Eur. J. Pharmacol. 1988, 151, 205.
37. Van Rossum, J. M. Arch. Int. Pharmacodyn. Ther. 1963,
143, 299.
18. De Amici, M.; Dallanoce, C.; De Micheli, C.; Grana, E.;
Barbieri, A.; Ladinsky, H.; Schiavi, G. B.; Zonta, F.
J. Med. Chem. 1992, 35, 1915.