Design, synthesis and binding at cloned muscarinic receptors of N-[5-(1'-substituted-acetoxymethyl)-3-oxadiazolyl] and N-[4-(1'-substituted-acetoxymethyl)-2-dioxolanyl] dialkyl amines

Article abstract of DOI:10.1016/S0968-0896(00)00092-4

Few muscarinic antagonists differentiate between the M₄ and M₂ muscarinic receptors. In a structure-activity study, aimed at discovering leads for the development of a M₄ muscarinic receptor-selective antagonist, we have s

Full text of DOI:10.1016/S0968-0896(00)00092-4

Bioorganic & Medicinal Chemistry 8 (2000) 1559±1566
Design, Synthesis and Binding at Cloned Muscarinic Receptors of
N-[5-(10-Substituted-Acetoxymethyl)-3-Oxadiazolyl] and
N-[4-(10-Substituted-Acetoxymethyl)-2-Dioxolanyl] Dialkyl Amines
Stefano Manfredini,a,* Ilaria Lampronti,a Silvia Vertuani,a Nicola Solaroli,a
Maurizio Recanatini,b David Bryanc and Michael McKinneyc,y
a
Á
Dipartimento di Scienze Farmaceutiche, Universita di Ferrara, Via Fossato di Mortara 17-19, I-44100 Ferrara, Italy
b
Á
Dipartimento Scienze Farmaceutiche, Universita di Bologna, Via Belmeloro 6, Bologna, Italy
cMayo Clinic Jacksonville, 4500 San Pablo Road, Jacksonville, FL 32224, USA
Received 11 January 2000; accepted 24 February 2000
AbstractÐFew muscarinic antagonists di€erentiate between the M4 and M2 muscarinic receptors. In a structure±activity study,
aimed at discovering leads for the development of a M4 muscarinic receptor-selective antagonist, we have synthesized and tested at
cloned muscarinic receptors the binding of a group of dioxolane- or oxadiazole-dialkyl amines, and compared them to our com-
pound 1, which contains the furan nucleus. Although none of these agents were particularly potent at M4 receptors (Kd values were
typically 30±70 nM), furan derivatives ( )1 and (+)1 were signi®cantly more potent at M4 receptors than at M2 receptors (ꢀ3- and
4-fold, respectively). The dioxolane derivatives 12b and 12c were more than 10-fold selective for the M4 versus the M2 receptors,
while the dioxolane derivative 12e was 15-fold more potent at M4 receptors than for M2 receptors. However, these agents bound to
M3 receptors with potencies like that for the M4 receptor, so they are not M4-selective. The M4/M2 relative selectivities of some of
our compounds are similar to the better hexahydrosiladifenidol derivatives, and may provide some important structural clues for
the development of potent and selective M4 antagonists. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
Most muscarinic antagonists are non-selective; a few are
marginally subtype-selective, but most of these are M1-
selective.5 Secoverine and tropicamide are two agents
reported to be 3±11-fold selective for the M4 subtype,
relative to the other subtypes.6 Recently, preparations of
cloned muscarinic receptors have become commercially
available; thus, materials for clearly distinguishing the
anities of antagonists at all ®ve subtypes are available
for rapid screening with binding assays.7,8 Most recently,
using cloned muscarinic receptor preparations for bind-
ing screening, a 38-fold M4/M2-selective benzoxazine
isoquinoline antagonist was discovered.9
The ®ve muscarinic receptors can be classi®ed into two
biochemical classes based upon structural similarity and
second messenger coupling.1 The M2 and M4 muscarinic
receptors are members of the subclass that couple to the
inhibition of adenylate cyclase and the activation of
potassium channels. These two receptors are dicult to
distinguish pharmacologically, in that they bind most
agonists and antagonists with similar anities.2 For
example, the potent antagonists AF-DX 384 and him-
bacine bind to M2 and M4 receptors with essentially the
same anities.3,4 Truly M4-selective agents would be
particularly useful studies of receptors in the central
nervous system, because most brain regions express
mixtures of several muscarinic receptor subtypes and
the M4 receptor is one of the more abundant ones, and
is the most abundant subtype in the striatum, a region
important in pharmacotherapy for Parkinson's and
Huntington's diseases.1
In our recent work, we developed muscarinic antago-
nists for di€erentiating between the ileal and bladder
receptors.10 A series of compounds with various sub-
stitutions on the furan ring were synthesized, of which
1 was the most selective in tissue screening assays. In
view of this unexpected tissue selectivity, both are
believed to express the M3 receptor; we started the pre-
sent study to determine the true receptor subtype selec-
tivity of these furan compounds in assays with cloned
muscarinic receptors. Additionally, we became inter-
ested in the possible M4-selectivity of our agents. In
*Corresponding author. Tel.: +39-532-291-292; fax: +39-532-291-296.
ySecond corresponding author.
0968-0896/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(00)00092-4

1560
S. Manfredini et al. / Bioorg. Med. Chem. 8 (2000) 1559±1566
Formulae 1 and general. a: R1=methyl; R2=phenyl; R3=cyclohexyl; d: R1=cyclohexyl; R2=phenyl; R3=cyclohexyl; e: R1=methyl; R2=cyclo-
hexyl; R3=cyclohexyl; h: R1=cyclohexyl; R2=cyclohexyl; R3=cyclohexyl; Het=furan; 1,2,4-oxadiazole; 1,3-dioxolane; R1, R2 R3=methyl,
cyclohexyl, phenyl.
order to explore the potential to develop M4-selective
antagonists, we prepared a new series of compounds
containing the dioxolane moiety.11 The dioxolane moi-
ety, originally synthesized as an analogue to muscar-
one,12 was subsequently shown to be suitable for
development of a potent muscarinic agonist, cis-dioxo-
lane.13,14 Addition of progressively larger lipophillic
groups to the dioxolane nucleus produces antagonist.14,15
Our dioxolane series is represented by compounds 12a±
g and several analogous agents were prepared with the
oxadiazole moiety (13a,d,e,h). This latter moiety has
been used in the ®eld of muscarinic receptors, but only
as an ester replacement in the modi®cation of arecoline
or azabicyclic muscarinic receptor ligands and not as a
sca€old to achieve subtype selectivity.16 20 The two
stereo-isomers of compound 1, compounds 12a±g, and
compounds 13a,d,e,h were screened in binding assays at
cloned M1±M4 receptors, all expressed in Chinese
hamster ovary (CHO) cell membranes.
appreciable yields (60±80%), only traces of compounds
5b,c could be recovered in these conditions (Scheme 1).
After several attempts, in order to overcome the above
described synthetic diculties, we envisaged route (b) as
a possible alternative pathway. Instead of reacting the
appropriate amine on the functionalized dioxolane ring,
as for route (a), chloroacetaldehyde dimethyl acetal was
directly converted into the corresponding amino-deri-
vatives (6b,c), and then transacetalized with iso-
propylidenglycerine-1-benzyl ether in the presence of
TSOH, to give compounds 7b,c in 57±59% overall
yields. Next deprotection by hydrogenolysis, as for 4,
gave the expected 5b,c in 47±68% yields.
3,5-Substituted-1,2,4-oxadiazoles (10a,d) were obtained,
adapting the procedure described by Street et al.,23 via
cyclization of the amidoxime precursor 9 with the appro-
priate N-substituted-amino acid ethyl ester (Scheme 2).24
Compound 9 was in turn obtained by the reaction of
benzyloxyacetonitrile (8)25 with hydroxylamine.
Due to the presence of the oxadiazole ring, it was
impossible to use the standard catalytic hydrogenation,
as for 3 and 7b,c, for the next removal of the benzyl
protective group. An alternative procedure, based on
the use of BCl3 at 73 ꢁC, allowed us to obtain the
intermediate 11a,d in good yields and short reaction
times. Finally, the amino-alcohols 5a-d and 11a,d were
condensed on the appropriate substituted acid chlor-
ides, prepared in situ from the corresponding carboxylic
acid and oxalyl chloride, in benzene (Scheme 3).
Results
Chemistry
The synthetic procedure for derivatives 12a±h and
13a,d,e,h are outlined in Schemes 1±3. In the case of the
dioxolanes 12a±h two di€erent routes have been envi-
saged. Route (a) was based on our previous studies, the
methodology now being improved by the use of
.
BF3 Et2O as the catalyst instead of toluensulfonic acid
(TSOH). Indeed, taking advantage of our other previous
®ndings on the transacetalization of THP-derivatives21,22
we have evaluated some Lewis acids (i.e. trimethylsilyl-
tri¯ate, SnCl4 and BF32O) as possible catalysts for the
dioxolane ring formation. Among these catalysts,
BF3 Et2O gave the better results both in terms of yields
and stereospeci®city: while cis:trans mixtures (4:1 ratios)
.
were obtained with TSOH, BF3 Et2O gave only the cis
Biology
All compounds were then tested for their capability to
displace [3H]QNB binding on cloned M1±M4 human
muscarinic receptors. Anities, determined by compe-
tition experiments, are reported in Table 1.
.
compound without traces of the other isomer. More-
over, the reaction could be accomplished at lower
temperature, about 35 ꢁC, without any traces of
decomposition products. Cis:trans ratios were inferred
Discussion
Compound 1, which was selected as a lead at the
beginning of this study, showed modest potency
(69 nM) and signi®cant selectivity (4.46-fold; p<0.05)
for the M4 subtype (Table 1). Compound ( )1 was also
signi®cantly selective (3.54-fold; p<0.05) for the M4
receptors versus the M1 receptors. Most M1-selective
antagonists do not di€erentiate the M1 and M4 recep-
tors,1 so compound ( )1 seems to be an exception to
this generality. Compound ( )1 also displayed a tendency
1
by H NMR analysis, according to our previous stu-
dies.11 In order to deprotect the hydroxy function at
position 4, removal of the benzyl group was performed
in reductive conditions, with Pd/C as the catalyst, to
give compound 4 in almost quantitative yield. Next
substitution of the chlorine atom at position 2 was
initially attempted in sealed vial, as described previously
by us.11 Whereas compounds 5a,d were obtained in

S. Manfredini et al. / Bioorg. Med. Chem. 8 (2000) 1559±1566
1561
Scheme 1. a: ꢁR1=methyl; b: R1=isopropyl; c: R1=butyl; d R1=cyclohexyꢁl. Reagents and conditions: (i) isopropylidenglycerine-1-benzylether,
.
BF3 Et2O, 40 C; (ii) H2, Pd/C; (iii) MeNHR1; (iv) MeNHR1, KI, K2CO3, 60 C; (v) 2,2-dimethyl-4-benzyl-dioxolane, TSOH; (vi) H2, Pd/C.
to be selective for the M4 over the M3 receptors, but with
the small number of assays performed did not quite reach
the p=0.05 level of signi®cance. Compound (+)1 was
also signi®cantly M4-selective (4.28-fold; p<0.05) relative
to the M2 subtype. However, this compound was sig-
ni®cantly (p<0.05) more potent at the M3 receptors
(Kd=65 nM) than was compound ( )1. Compounds
( )1 and (+)1 were about equipotent at the M1 recep-
tors. The substitution of the furan ring, in the lead
compound 1, with the dioxolane (12a±h) and oxadiazole
one (13a,d,e,h), induced an overall lipophilicity varia-
tion of 0.6 and 1.2 log P units in the case of furan±
dioxolane and of dioxolane±oxadiazole, respectively
(CLOGP values were calculated with the C-QSAR pro-
gram).26 As a result of this modi®cation, a consistent
decrement of activity at muscarinic receptors was
observed, in particular for the oxadiazole derivatives.
However, compounds 12b, 12c and 12e were more than
10-fold selective for the M4 versus the M2 receptors. Of
these, 12b and 12c also distinguished between the M4
and M1 receptors. Compound 12e bound to the M1 and
M4 receptors with similar potency, but was notable for
having the highest (15-fold; p<0.2) selectivity for the
M4 versus the M2 receptors. Compound 12 was remi-
niscent of drugs like 4-DAMP, which bind with similar
potencies at M1, M3, and M4 receptors, but with low
anity to the M2 subtype.

1562
S. Manfredini et al. / Bioorg. Med. Chem. 8 (2000) 1559±1566
Scheme 2. a: R1=methyl; R2=phenyl; R3=cyclohexyl; d: R1=cyclohexyl; R2=phenyl; R3=cyclohexyl. Reagents and conditions: (i) 60 ꢁC,
NH2OH; (ii) NaH, a: N,N-dimethylamino- or d: N-cyclohexyl-N-methyl-glycine ethyl ester; (iii) BCl3.
Scheme. 3. a: R1=methyl; R2=phenyl; R3=cyclohexyl; b: R1=isopropyl; R2=phenyl; R3=cyclohexyl; c: R1=butyl; R2=phenyl; R3=cyclohexyl;
e: R1=methyl; R2=cyclohexyl; R3=cyclohexyl; f: R1=isopropyl; R2=cyclohexyl; R3=cyclohexyl; g: R1=butyl; R2=cyclohexyl; R3=cyclohexyl;
h: R1=cyclohexyl; R2=cyclohexyl; R3=cyclohexyl. Reagents and conditions: i: R2R3CHCOOH, (COCl)2.
Table 1. Anities (Kd) for muscarinic receptors
Conclusion
Compound
M1 Kd
(nM)
M2 Kd
(nM)
M3 Kd
(nM)
M4 Kd
(nM)
Compounds 13a and 13d were two or more orders of
magnitude less potent at the muscarinic receptors than the
other drugs listed in Table 1, indicating that substitution
of the oxadiazole ring for the dioxolane structure is detri-
mental to muscarinic activity.
(
)1
244Æ6a
308Æ18a
199Æ17.2a
65Æ10.5a
34Æ5a
28
69Æ11.9a
50Æ6a
36Æ7a
30
(+)1
12a
12b
12c
12d
12e
12f
12g
13a
13d
161Æ3a
80Æ8a
189
214Æ29a
192Æ24a
327
202
316
392
302
27
32
33
39
73Æ8a
796Æ55a
26Æ3a
53Æ10a
Since this substitution does not a€ect the size of the
molecule and the distance between the polar head and
the lipophilic tail, the anity may be reduced due to the
increased hydrophilicity and/or di€erent electronic
properties. In our opinion, having maintained the same
substitution pattern (R1 to R3) of furan and dioxolane
634
588
>10,000
98
29
65
49
1280
>10,000
>10,000
>10,000
>10,000
3430
>10,000
6710
6810
aData are averages of three to ®ve experiments with each receptor.

S. Manfredini et al. / Bioorg. Med. Chem. 8 (2000) 1559±1566
1563
series, these results con®rm the importance of the over-
all lipophilicity in the anity pro®le of compound 1.
338 mmol). The reaction mixture was heated at re¯ux
conditions for 2 h, cooled and washed with a saturated
solution of NaHCO3 (20 mLÂ2). This mixture was then
washed with a saturated solution of NaCl and water.
The organic layer was dried (Na2SO4), evaporated under
vacuum, and the residue was puri®ed by column chro-
matography on silica gel (eluent hexane:Et2O 4:1) to give
The dioxolane derivatives 12b and 12c were more than
10-fold selective for the M4 versus the M2 receptors,
while the dioxolane derivative 12e was 15-fold more
potent at M4 receptors than for M2 receptors. However,
these agents bound to M3 receptors with potencies like
that for the M4 receptor, so they are not M4-selective.
1
the compound 3 (yield 98%). H NMR (CDCl3) d 3.52
(d, 2H, J=3.6 Hz), 3.86 (m, 2H), 4.15 (m, 1H), 4.31 (m,
2H), 4.69 (m, 2H), 5.16 (t, 1H, J=3.6 Hz, H-2 cis), 5.37
(t, 1H, J=3.6 Hz, H-2 traces of trans), 7.20±7.50 (m, 5H).
The potencies observed in the present assays, for the iso-
lated enantiomers of 1, are higher than those observed for
the racemate in isolated tissues;10 however, it is not un-
usual for receptors in tissues to bind less eciently than
receptors in homogenates, due to di€usional and other
mechanical constraints.27 Further studies will address this
question also with respect to compounds 12a±h.
Synthesis of 2-chloromethyl-4-hydroxymethyl-1,3-dioxo-
lane (4). The compound 3 (6.9 g, 28 mmol) was dissolved
in methyl alcohol, hydrogenated at 40 psi over 10% Pd/
C for 18 h. The catalyst was removed by ®ltration on a
Celite pad, and the solvent was then evaporated under
vacuum. The residue oil was puri®ed by column chro-
matography on silica gel (eluent hexane:Et2O 4:1) to
However, it is noteworthy that in the current research
we have been able to demonstrate the true enantio-
selectivity of 1 in its binding to the M4 muscarinic
receptors; this illustrates the advantages of screening
with cloned receptors.
1
give the compound 4 (40%). H NMR (CDCl3) d 2.50±
4.15 (br, 1H), 3.22 (d, 2H, J=3.9 Hz), 3.51 (m, 2H),
3.70±3.90 (m, 2H), 4.12 (m, 1H), 5.03 (t, 1H, J=3.9 Hz,
H-2 cis).
In conclusion, we believe these data of interest for
structure±activity relationship (SAR) studies in the ®eld
of muscarinic agents and, in view of the demand for
subtype-speci®c ligands, particularly in the case of M4
receptors. Indeed, the reinvestigation of compound 1
resulted in the discovery of a new M4-selective ligand,
that might be useful as a lead compound in the design of
more potent and selective ligands.
Synthesis of 2-dimethylaminomethyl- and 2-(N-methyl-
N-cyclohexylaminomethyl)-4-hydroxymethyl-1,3-dioxo-
lanes (5a,d). The compound 4 (0.7 g, 4.5 mmol) was dis-
solved in ethyl alcohol (10 mL) and mixed with the
appropriate amine (15 mmol) in a sealed vial for 12 h at
120 ꢁC. The solvent was removed on a rotary eva-
porator and the residue was puri®ed by column chro-
matography on silica gel (eluent CH2Cl2:MeOH 8:2) to
give the compounds 5a,d as oils.
Experimental
Chemistry
5a: yield 82%; 1H NMR (CDCl3) d 2.43 (s, 6H), 2.79 (d,
2H, J=2.6 Hz), 3.44 (br, 1H), 3.51 (m, 2H), 3.65±4.20
(m, 2H), 4.25 (m, 1H), 5.10 (t, 1H, J=2.6 Hz, H-2 cis).
Material and methods. Reaction courses and product
mixtures were routinely monitored by thin-layer chro-
matography (TLC) on silica gel precoated F254 Merck
plates, with detection under 254 nm UV lamp and/or by
spraying with a diluted potassium permanganate solu-
tion. Nuclear magnetic resonance (1H NMR) spectra
were determined for solution in CDCl3±DMSO-d6 on a
Bruker AC-200 spectrometer, and peak positions are
given in parts per million (d) down®eld from tetra-
methylsilane as internal standard, whereas coupling
constants (J) are in hertz. Melting points were obtained
in open capillary tubes and are uncorrected. Column
chromatographies were performed with Merck 60±200
mesh silica gel. Ambient temperature was 22±25 ꢁC. All
drying operations were performed over anhydrous
magnesium or sodium sulfate. Microanalyses, unless
indicated, were in agreement with calculated values
within Æ0.4%.
5d: yield 67%; 1H NMR (CDCl3) d 1.29±1.35 (m, 10H),
2.43 (s, 3H), 2.61 (m, 1H), 2.79 (d, 2H, J=2.6 Hz), 3.44
(br, 1H), 3.51 (m, 2H), 3.65±4.20 (m, 2H), 4.25 (m, 1H),
5.10 (t, 1H, J=2.6 Hz, H-2 cis).
Route b.
Synthesis of (2,2-dimethoxy-ethyl)-isopropyl-methyl-amine
and (2,2-dimethoxy-ethyl)-butyl-methyl-amine (6b,c). A
mixture of chloroacetaldehydedimethylacetal (2) (0.7 g,
4.6 mmol), the appropriate N-methylamine-derivative
(4.6 mmol), KI (catalytic amount) and K2CO3 (1.27 g,
9.2 mmol) in acetonitrile was heated under argon atmo-
sphere and at re¯ux conditions for 18h. The solvent was
then removed under vacuum and the resulting oil was
puri®ed by column chromatography on silica gel (eluent
CH2Cl2:MeOH 9.5:0.5) to give the compounds 6b and 6c.
Synthesis of 2,4-substitued-dioxolanes. Route a.
6b: yield 35%; 1H NMR (CDCl3) d 1.05 (m, 6H), 2.27 (s,
3H), 2.65 (m, 2H), 2.97 (m, 1H), 3.24 (s, 6H), 4.43 (m, 1H).
Synthesis of 2-chloromethyl-4-benzyloxymethyl-1,3-diox-
olane (3). To a stirred and re¯uxed solution of iso-
propylidenglycerine-1-benzylether (11 mL, 50mmol) and
chloroacetaldehydedimethylacetal (2) (12 mL, 100 mmol)
in Et2O (800 mL) was added BF3 in Et2O (48 g, 43 mL,
1
6c: yield 43%; H NMR (CDCl3) d 0.95 (t, 3H, J=7
Hz), 1.33±1.39 (m, 4H), 2.36 (m, 2H), 2.66 (m, 2H), 3.24
(s, 6H), 4.43 (m, 1H).

1564
S. Manfredini et al. / Bioorg. Med. Chem. 8 (2000) 1559±1566
Synthesis of 2-(N,N-methylisopropylaminomethyl)- and 2
-(N,N-methylbutylaminomethyl)-4-benzyloxymethyl-1,3-
dioxolanes (7b,c). To a stirred and re¯uxed solution of
6b,c (100mmol) and isopropylidenglycerine-1-benzylether
(11 mL, 50 mmol) in benzene (200 mL) were added water
(9 mL) and TSOH (1 g, 5 mmol). The reaction mixture
was heated at re¯ux conditions for 24 h, cooled and
washed with a saturated solution of NaHCO3/H2O
(2Â100). The organic phase was then washed with a
saturated solution of NaCl/H2O and water. The organic
layer was ®nally dried (MgSO4) and evaporated in
vacuo. The residue was puri®ed by column chromato-
graphy on silica gel (eluent hexane:Et2O 4:1).
the pink color of the solution. When the addition was
complete, the mixture was heated at re¯ux conditions
for an additional 2 h, then ®ltered and 2-benzyloxy-
acetonitrile (8) (2 g, 14 mmol) was added to the above
prepared hydroxylamine free base solution. The mix-
ture was heated at re¯ux conditions for 48 h. The
reaction mixture was then cooled to room temperature,
dried over Na2SO4 and evaporated to dryness. The
residue was dissolved in Et2O (20 mL) and crystallized
from hexane to give 9.
Yield 63%; mp 73±74 ꢁC; 1H NMR (DMSO-d6) d 3.86 (s,
2H), 4.44 (s, 2H), 5.56 (br, 2H), 7.20±7.40 (m, 5H), 8.80±
9.50 (br, 1H).
1
7b, cis isomer: yield 45%; H NMR (CDCl3) d 1.16 (d,
3H, J=6.6 Hz), 1.19 (d, 3H, J=6.6 Hz), 2.65 (s, 3H),
2.96 (d, 2H, J=2.6 Hz), 3.29 (m, 1H), 3.48±4.35 (m, 5H),
4.63 (m, 2H), 5.18 (t, 1H, J=2.6 Hz, H-2 cis), 7.20±7.50
(m, 5H).
Synthesis of 3-benzyloxymethyl-5-dimethylaminomethyl-
and 3-benzyloxymethyl-5-methylcyclohexylaminomethyl-1,
2,4-oxadiazoles (10a,d). Compound 9 (0.36 g, 2 mmol)
was dissolved in THF (anhydrous, 10mL). NaH 80%
(0.05 g, 2.1 mmol) was added to the solution, and the
mixture was heated at re¯ux for 1 h followed by the addi-
tion of the appropriate amino acid ethyl ester (4mmol).
Re¯ux was continued for additional 90 min. The mix-
ture was then cooled and ®ltered through a Celite pad,
and the residue was puri®ed by column chromato-
graphy on silica gel (eluent Et2O:CH2Cl2 1:4) to give
10a,d as oils.
1
Trans isomer: yield 12%; H NMR (CDCl3) d 5.32 (t,
1H, J=2.6 Hz, H-2 trans).
1
7c, cis isomer: yield 52%; H NMR (CDCl3) d 0.91 (t,
3H, J=7.2 Hz), 1.28 (m, 2H), 1.47 (m, 2H), 2.43 (s, 3H),
2.52 (m, 2H), 2.70 (d, 2H, J=2.2 Hz), 3.45±4.15 (m,
4H), 4.25 (m, 1H), 4.63 (m, 2H), 5.06 (t, 1H, J=2.2 Hz,
H-2 cis), 7.20±7.50 (m, 5H).
1
10a: yield 60%; H NMR (CDCl3) d 2.36 (s, 6H), 3.79
(s, 2H), 4.63 (s, 2H), 4.65 (s, 2H), 7.20±7.45 (m, 5H).
1
Trans isomer: yield 7%; H NMR (CDCl3) d 5.16 (t,
1H, J=2.2 Hz, H-2 trans).
10d: yield 62%; 1H NMR (CDCl3) d 1.10±1.90 (m, 10H),
2.42 (s, 3H), 2.46 (m, 1H), 3.96 (s, 2H), 4.67 (s, 2H), 4.49
(s, 2H), 7.25±7.45 (m, 5H).
Synthesis of 2-(N-methyl-N-isopropylaminomethyl)- and
2-(N-methyl-N-butylaminomethyl)-4-hydroxymethyl-1,3-
dioxolanes (5b,c). The compounds 7b,c (28 mmol) were
dissolved in methyl alcohol, hydrogenated at 40 psi over
10% Pd/C for 18 h. The catalyst was removed by ®ltra-
tion on a Celite pad, and the solvent was then evapo-
rated under vacuum. The obtained oil was puri®ed by
column chromatography on silica gel (eluent CH2Cl2:
MeOH 4:1).
Synthesis of 3-hydroxymethyl-5-dimethylaminomethyl-
and of 3-hydroxymethyl-5-methylcycloesylaminomethyl-
1,2,4-oxadiazoles (11a,d). Compounds 10a,d were dis-
solved in CH2Cl2 (anhydrous, 30mL) under argon atmo-
sphere. The solution was cooled to 73ꢁC and boron
trichloride solution, 1 M in dichloromethane (4 mL,
4 mmol), was slowly added. The reaction was then treated,
after 10 min, by very slow addition of a mixture of
CH2Cl2 and MeOH (1:1). After 10 min, trimethylamine
(2.6 mL) was added and the reaction mixture warmed to
room temperature, dried over Na2SO4 and concentrated
in vacuo. The crude oil obtained was puri®ed by column
chromatography on silica gel (eluent CH2Cl2:MeOH
5:0.25) to give 11a,d as oils.
5b: yield 68%; 1H NMR (CDCl3) d 1.16 (d, 3H,
J=6.6 Hz), 1.19 (d, 3H, J=6.6 Hz), 2.65 (s, 3H), 2.96 (d,
2H, J=2.6 Hz), 3.29 (m, 1H), 3.48±4.35 (m, 5H), 5.18 (t,
1H, J=2.6 Hz, H-2 cis), 5.80±7.50 (br, 1H).
5c: yield 47%; 1H NMR (CDCl3) d 0.91 (t, 3H,
J=7.2 Hz), 1.28 (m, 2H), 1.47 (m, 2H), 2.43 (s, 3H), 2.52
(m, 2H), 2.70 (d, 2H, J=2.2 Hz), 3.45±4.15 (m, 4H), 4.25
(m, 1H), 4.97 (s. br. 1H), 5.06 (t, 1H, J=2.2 Hz, H-2 cis).
1
11a: yield 62%; H NMR (CDCl3) d 2.35 (s, 6H), 3.78
(s, 2H), 3.80±4.20 (br, 1H), 4.76 (s, 2H).
Synthesis of 3,5-substituted-1,2,4-oxadiazoles.
11d: yield 77%; 1H NMR (CDCl3) d 0.90±2.00 (m,
10H), 2.34 (s, 3H), 2.39 (m, 1H), 3.60±4.60 (br, 1H),
3.88 (s, 2H), 4.75 (s, 2H).
Synthesis of 2-benzyloxyacetamidoxime (9). A sodium
butoxide solution was prepared by dissolving sodium
(0.34 g, 16 mmol) in butanol (10 mL) under argon
atmosphere and heating at re¯ux conditions for 1 h. The
above obtained sodium butoxide solution (kept
warmed) was added to a well-stirred n-butanol solution
(10 mL) of hydroxylamine hydrochloride (1.2 g,
15 mmol) and phenolphthaleine (catalytic amount). The
rate of the addition should be slow enough to maintain
General procedure for the preparation of N-[4-(10-substi-
tuted-acetoxymethyl)-2-dioxolanyl] and N-[5-(10-substitu-
ted-acetoxymethyl)-3-oxadiazolyl] dialkyl amines 12a±h
and 13a,d,e,h. Phenylcyclohexyl acetic acid or dicyclo-
hexyl acetic acid (0.87 mmol) were dissolved in benzene
(anhydrous, 5 mL) under argon atmosphere; to the

S. Manfredini et al. / Bioorg. Med. Chem. 8 (2000) 1559±1566
1565
solution was then added oxalyl chloride (110 mL,
1.26 mmol). After 1 h under stirring at room tempera-
ture, the solvent was evaporated and the residue was
dissolved in anhydrous benzene (3 mL). This solution
was added dropwise, under argon, to a solution of the
appropriate amino-alcohol 5a±d or 11a,d (0.76 mmol) in
benzene (anhydrous, 10 mL) containing triethylamine
(0.4 mL) and DMAP (catalytic amount). The mixture
was stirred at room temperature for 24 h, then washed
with a saturated solution of NaHCO3, the aqueous layer
was dried (MgSO4), concentrated, and the residue was
puri®ed by column chromatography on silica gel (eluent
CH2Cl2: MeOH 5:0.25) to give the expected compound.
(CDCl3) d 0.83 (t, 3H, J=10.8 Hz), 0.90±1.45 (m, 14H),
1.45±1.85 (m, 12H), 2.01 (t, 1H, J=7.4 Hz), 2.24 (s, 3H),
2.32 (m, 3H), 2.52 (d, 2H, J=4.5 Hz), 3.65±3.80 (m,
2H), 3.95±4.10 (m, 2H), 4.19 (m, 1H), 4.94 (t, 1H,
J=4.5 Hz, H-2 cis). Anal. C, H, N (C24H43NO4).
4-(10-Dicyclohexyl-acetoxymethyl)-2-(N-methyl-N-cyclo-
1
hexyl-aminomethyl)-1,3-dioxolane (12h). Yield 76%; H
NMR (CDCl3) d 0.75±1.40 (m, 18H), 1.40±1.90 (m, 14H),
2.02 (t, 1H, J=7.2 Hz), 2.25 (m, 1H), 2.29 (s, 3H), 2.59 (d,
2H, J=4.5 Hz), 3.50±4.05 (m, 4H), 4.20 (m, 1H), 4.91 (t,
1H, J=4.5 Hz, H-2 cis). Anal. C, H, N (C26H45 NO4).
3-(10-Cyclohexyl-10-phenylacetoxymethyl)-5-(N,N-dimethyl-
aminomethyl)-1,2,4-oxadiazole (13a). Yield 55%; 1H
NMR (CDCl3) d 0.90±2.10 (m, 10 H), 2.28 (s, 6H), 2.38
(m, 1H), 3.26 (d, 1H, J=10.6 Hz), 3.80 (s, 2H), 5.03 (d,
1H, J=13.7 Hz), 5.21 (d, 1H, J=13.7 Hz), 7.10±7.35 (m,
5H). Anal. C, H, N (C20H27N3O3).
4-(10-Cyclohexyl-10-phenyl-acetoxymethyl)-2-(N,N-dime-
1
thyl-aminomethyl)-1,3-dioxolane (12a). Yield 60%; H
NMR (CDCl3) d 1.00±1.40 (m, 6H), 1.45±1.80 (m, 5H),
2.27 (s, 6H), 2.47 (m, 2H), 3.23 (d, 1H, J=10.6 Hz), 3.65±
3.90 (m, 2H), 4.00±4.30 (m, 3H), 4.95 (t, 1H, J=2.60 Hz,
H-2 cis), 7.20±7.40 (m, 5H). Anal. C, H, N (C21H31NO4).
3-(10 -Cyclohexyl-10 -phenyl-acetoxymethyl)-5-(N-cyclo-
hexyl-N-methyl-aminomethyl)-1,2,4-oxadiazole (13d).
4-(10-Cyclohexyl-10-phenyl-acetoxymethyl)-2-(N-isopropyl-
N-methyl-aminomethyl)-1,3-dioxolane (12b). Yield 51%;
1H NMR (CDCl3) d 0.91 (d, 6H, J=6.6 Hz), 1.00±1.40
(m, 6H), 1.40±2.05 (m, 5H), 2.21 (s, 3H), 2.47 (m, 2H),
2.79 (m, 1H), 3.18 (d, 1H, J=10.6 Hz), 3.30±4.15 (m,
5H), 4.86 (t, 1H, J=4.5 Hz, H-2 cis), 7.05±7.30 (m, 5H).
Anal. C, H, N (C23H35NO4).
1
Yield 97%; H NMR (CDCl3) d 0.90±1.50 (m, 11H),
1.50±2.15 (m, 10 H), 2.37 (s, 3H), 2.38 (m, 1H), 3.36 (d,
1H, J=10.6 Hz), 3.92 (s, 2H), 5.08 (d, 1H, J=13.6 Hz),
5.26 (d, 1H, J=13.6 Hz), 7.20±7.40 (m, 5H). Anal. C, H,
N (C25H35N3O3).
3-(10-Dicyclohexyl-acetoxymethyl)-5-(N,N-dimethyl-ami-
4-(10-Cyclohexyl-10-phenyl-acetoxymethyl)-2-(N-butyl-N-
methyl-aminomethyl)-1,3-dioxolane (12c). Yield 72%; 1H
NMR (CDCl3) d 0.88 (t, 3H, J=7.2 Hz), 1.00±1.80 (m,
15H), 2.23 (s, 3H), 2.34 (m, 2H), 2.49 (d, 2H, J=4.4Hz),
3.17 (d, 1H, J=10.8 Hz), 3.50±3.80 (m, 2H), 3.80±4.25 (m,
3H), 4.90 (t, 1H, J=4.4 Hz, H-2 cis), 7.10±7.40 (m, 5H).
Anal. C, H, N (C24H37NO4).
nomethyl)-1,2,4-oxadiazole (13e). Yield 77%; H NMR
1
(CDCl3) d 0.80±1.50 (m, 10H), 1.50±1.80 (m, 12H), 2.15
(t, 1H, J=7.2 Hz), 2.36 (s, 6H), 3.80 (s, 2H), 5.19 (s,
2H). Anal. C, H, N (C20H33N3O3).
3-(10 -Dicyclohexyl-acetoxymethyl)-5-(N-cyclohexyl-N-
methyl-aminomethyl)-1,2,4-oxadiazole (13h). Yield 85%;
1H NMR (CDCl3) d 0.80±1.50 (m, 18H), 1.50±1.90 (m,
14H), 2.10 (t, 1H, J=7.2 Hz), 2.32 (s, 3H), 2.37 (m, 1H),
3.89 (s, 2H), 5.13 (s, 2H). Anal. C, H, N (C25H41N3O3).
4-(10-Cyclohexyl-10-phenyl-acetoxymethyl)-2-(N-methyl-
N-cyclohexyl-aminomethyl)-1,3-dioxolane (12d). Yield
1
75%; H NMR (CDCl3) d 0.90±1.35 (m, 13H), 1.35±
1.75 (m, 8H), 2.20 (m, 1H), 2.26 (s, 3H), 2.53 (d, 2H,
J=4.6 Hz), 3.17 (d, 1H, J=10.6 Hz), 3.55±3.85 (m, 2H),
3.85±4.20 (m, 3H), 4.85 (t, 1H, J=4.6 Hz, H-2 cis),
7.10±7.40 (m, 5H). Anal. C, H, N (C26H39NO4).
Biology
Materials
4- (10 - Dicyclohexyl - acetoxymethyl) - 2 - (N,N - dimethyl -
aminomethyl)-1,3-dioxolane (12e). Yield 85%; H NMR
(CDCl3) d 0.90±1.35 (m, 10H) 1.35±1.80 (m, 12H), 2.02
(t, 1H, J=7.4 Hz), 2.29 (s, 6H), 2.56 (d, 2H, J=4.4 Hz),
3.65±3.80 (m, 2H), 3.95±4.10 (m, 2H), 4.21 (m, 1H), 5.01
(t, 1H, J=4.4 Hz, H-2 cis). Anal. C, H, N (C21H37
NO4).
[3H]QNB (speci®c activity 43 Ci/mmol) was purchased
from DuPont/NEN (Boston, MA). Atropine was from
Research Biochemicals, Inc. (Natick, MA).
1
Membrane preparation
Human muscarinic receptors (hm1±hm4) transfected
into CHO cells were purchased from Receptor Biology,
Inc. The membranes were stored and prepared according
to RBI recommendations. Each assay tube contained
approximately 16±31 mg of membranes.
4-(10 -Dicyclohexyl-acetoxymethyl)-2-(N-isopropyl-N-
methyl-aminomethyl)-1,3-dioxolane (12f). Yield 60%;
1H NMR (CDCl3) d 0.80±1.35 (m, 10H), 0.92 (d, 6H,
J=6.6 Hz), 1.35±1.80 (m, 12H), 2.02 (t, 1H, J=7.4 Hz),
2.24 (s, 3H), 2.50 (d, 2H, J=4.6 Hz), 2.79 (m, 1H), 3.70±
3.90 (m, 2H), 4.15 (m, 2H), 4.20 (m, 2H), 4.90 (t, 1H,
4.6 Hz, H-2 cis). Anal. C, H, N (C23H41NO4).
Radioligand binding assays
Competiton assays between [3H]QNB (0.2 nM) and the
unlabeled muscarinic antagonists (1 nM±10 mM) were
performed in a total volume of 1 mL of 50 mM sodium±
potassium phosphate bu€er. At the end of the incubation
4-(10-Dicyclohexyl-acetoxymethyl)-2-(N-butyl-N-methyl-
1
aminomethyl)-1,3-dioxolane (12g). Yield 60%; H NMR

1566
S. Manfredini et al. / Bioorg. Med. Chem. 8 (2000) 1559±1566
period, each suspension was aspirated onto Whatman
GF/B glass ®ber ®lters in a Brendel cell harvester. The
®lters were washed three times with cold 50mM
sodium±potassium phosphate bu€er; radioactivity was
solubilized and and counted in Redi-Safe cocktail (Beck-
man). Atropine (100 mM) was present in some tubes to
determine the level of nonspeci®c binding.
10. Feriani, A.; Gaviraghi, G.; Toson, G.; Mor, M.; Barbieri,
A.; Grana, E.; Boselli, C.; Guarneri, M.; Simoni, D.; Manfred-
ini, S. J. Med. Chem. 1994, 37, 4278.
11. Manfredini, S.; Simoni, D.; Guarneri, M.; Ferroni, R.;
Zaidi, J.; Grana, E.; Boselli, C. Drug Des. Dis 1995, 13, 1.
12. Fourneau, J. P.; Bovet, D.; Bovet, F.; Montezin, G. Bull.
Soc. Chim. Biol. 1994, 26, 135.
13. Belleau, B.; Puranen, J. J. Med. Chem. 1963, 6, 325.
14. Chang, K. J.; Deth, R. C.; Triggle, D. J. J. Med. Chem.
1972, 15, 243.
15. Aronstam, R. S.; Triggle, D. J.; Eldefrawi, M. E. Mol.
Pharmacol. 1979, 15, 227.
16. Suzuki, T.; Uesaka, H.; Hamajima, H.; Ikami, T. Chem.
Pharm. Bull. 1999, 47, 876.
17. Ward, J. S.; Calligaro, D. O.; Bymaster, F. P.; Shannon,
H. E.; Mitch, C. H.; Whitesitt, C.; Brunsting, D.; Sheardown,
M. J.; Olesen, P. H.; Swedberg, M. D.; Jeppesen, L.; Sauer-
berg, P. J. Med. Chem. 1998, 41, 379.
Data analysis
Concentration±response data were ®tted with a four
parameter logistic model, ALLFIT.28 The initial para-
meters were constrained using values assuming compe-
tition between [3H]QNB and muscarinic antagonists. Kd
was calculated by correction of IC50 using the Cheng
and Pruso€ formula29 and Kd for [3H]QNB of 50 pM.
18. Showell, G. A.; Baker, R.; Davis, J.; Hargreaves, R.;
Freedman, B.; Hoogsteen, K.; Shailendra, P.; Snow, R. J. J.
Med. Chem. 1992, 35, 911.
Acknowledgements
19. Sauerberg, P.; Kindtler, J. W.; Nielsen, L.; Sheardown, M.
J.; Honore, T. J. Med. Chem. 1991, 34, 687.
20. Oerlek, B. S.; Blaney, F. E.; Brown, F.; Clark, M. S. G.;
Hadley, M. S.; Hatcher, J.; Riley, G. J.; Rosenberg, H. E.;
Wadsworth, H. J.; Wyman, P. J. Med. Chem. 1991, 34,
2726.
This work was supported by University of Ferrara
(grants 1997 and 1998), NIH grants AG09973 and
AG12653, and the Mayo Foundation.
References
21. Manfredini, S.; Baraldi, P.; Bazzanini, R.; Guarneri, M.;
Simoni, D. Tetrahedron Lett. 1994, 5709.
1. McKinney, M. Prog. Brain Res. 1993, 98, 333.
2. McKinney, M.; Miller, J. H.; Nickelson, L.; Aksoy, S. Mol.
Pharmacol. 1991, 40, 1014.
3. Miller, J. H.; Gibson, V. A.; McKinney, M. J. Pharmacol.
Exp. Ther. 1991, 259, 601.
4. Miller, J. H.; Aagaard, P. J.; Gibson, V. A.; McKinney, M.
J. Pharmacol. Exp. Ther. 1992, 263, 663.
5. Eglen, R. M.; Choppin, A.; Dillon, M. P.; Hegde, S. Curr.
Opin. Chem. Biol. 1999, 3, 426.
6. van Zwieten, P. A.; Doods, H. N. Cardiovascular Drugs &
Therapy 1995, 9, 159.
7. Dorje, F.; Wess, J.; Lambrecht, G.; Tacke, R.; Mutschler,
E.; Brann, M. R. J. Pharmacol. Exp. Ther. 1991, 256, 727.
8. Bukely, N. J.; Bonner, T. I.; Buckley, C. M.; Brann, M. R.
Mol. Pharmacol. 1989, 35, 469.
9. Augelli-Szafran, C. E.; Jaen, J. C.; Moreland, D. W.; Nel-
son, C. B.; Penvose-Yi, J. R.; Schwarz, R. D. Biorg. Med.
Chem. Lett. 1998, 8, 1991.
22. Manfredini, S.; Baraldi, P.; Bazzanini, R.; Guarneri, M.;
Simoni, D. J. Chem. Soc. Chem. Commun. 1994, 583.
23. Street, L.; Baker, R.; Book, T.; Kneen, C.; MacLeod, A.;
Merchant, K.; Showell, G.; Saunders, J.; Herbert, H.; Freed-
man, S.; Harley, E. J. Med. Chem. 1990, 33, 2690.
24. Manfredini, S.; Bazzanini, R.; Baraldi, P.; Simoni, D.;
Vertuani, S.; Pani, A.; Pinna, E.; Scintu, F.; Lichino, D.; La
Colla, P. Bioorg. Med. Chem. Lett. 1996, 6, 1279.
25. Quaterman, J.; Stevens, T. J. Chem. Soc. 1955, 3292.
26. C-QSAR Ver. 1.87, BioByte, Claremont, CA, 1994.
27. McKinney, M. In Current Protocols in Pharmacology;
Enna, S. and Williams, M., Eds; John Wiley & Sons, Inc.:
New York, 1998; unit 1.3, pp 1±33.
28. DeLean, A.; Munson, P.; Rodbard, D. Am. J. Physiol.
1978, 235, E97.
29. Cheng, Y.; Pruso€, W. Biochem. Pharmacol. 1973, 22,
3099.