Synthesis of N-substituted piperidine salts as potential muscarinic ligands for Alzheimer's applications

Article abstract of DOI:10.1002/jhet.1742

Several heterocyclic N-piperidine substituted salts were synthesized that were found to inhibit the specific binding of the antagonist [³H] quinuclidinyl benzilate in radioligand muscarinic binding assays ( ³H-QNB) in bioassays. One

Full text of DOI:10.1002/jhet.1742

Month 2013
Synthesis of N-Substituted Piperidine Salts as Potential Muscarinic
Ligands for Alzheimer’s Applications
a
b
b
a
*
John Boulos, Jan Jakubik, Alena Randakova, and Cristina Avila
aDepartment of Physical Sciences, Barry University, Miami Shores, Florida 33161, USA
bDepartment of Neurochemistry, Institute of Physiology of the Academy of Sciences of the Czech Republic, v.v.i., 142 20
Prague, Czech Republic
*
E-mail: jboulos@mail.barry.edu
Received March 27, 2012
DOI 10.1002/jhet.1742
Published online 00 Month 2013 in Wiley Online Library (wileyonlinelibrary.com).
Several heterocyclic N-piperidine substituted salts were synthesized that were found to inhibit the specific
binding of the antagonist [3H] quinuclidinyl benzilate in radioligand muscarinic binding assays (3H-QNB) in
bioassays. One of the heterocyclic salts, compound 7, met the significance criteria in these assays (>50%
inhibition) at 10 mM of the nonselective muscarinic antagonist (3H-QNB) in cells of the Wistar rat cerebral
cortex. Furthermore, this compound displayed 61% inhibition at 10 mM of the antagonist (3H-QNB) for the
M5 receptor (IC50 6.34 mM, Ki 3.93 mM, nH = 0.996) in human recombinant CHO cell lines. These data
obtained from Ricerca Biosciences suggested that compound 7 was selective for the M5 receptor. Another
study from the Czech Academy of Sciences demonstrated that compound 7 was 3 to 8 times more potent
at M2 than other subtypes of muscarinic receptors in competition with antagonist N-methylscopolamine
and selective for the M1 receptors over M3 and M5 in antagonizing accumulation of inositol phosphates
induced by muscarinic agonist carbachol.
J. Heterocyclic Chem, 00, 00 (2013).
INTRODUCTION
is a functionally selective M1/M4 receptor agonist with prom-
ising in vivo antidementia properties [4]. This drug was ini-
tially targeted for the treatment of AD, but clinical trials
revealed peripheral side effects that were not consistent with
M1 selectivity. More recent behavioral studies indicate, how-
ever, that the compound may decrease psychotic behaviors
in patients with AD suggesting that it might be useful in
the treatment of psychotic symptoms in patients with schizo-
phrenia. Xanomeline was shown to exhibit a novel mode of
interaction with the M1 receptor different from that used by
conventional agonists. There is evidence that the persistent at-
tachment of Xanomeline takes place away from the classical
binding site, whereas the active group interacts reversibly with
the primary receptor activation site. Other data have shown
that Xanomeline binds in a wash-resistant manner as an antag-
onist at the M5 receptor [5]. The ability of Xanomeline to
activate some subtypes of muscarinic receptors while antag-
onizing others in a wash-resistant manner represents a unique
and complex pharmacological profile. There is an obvious
need for more M1-selective receptor activators because of
unwanted side effects with compounds like Xanomeline.
Alzheimer’s disease (AD) is a neurodegenerative disorder
characterized by loss of memory, judgment, and language
functions and by the loss of specific neurons that project from
the basal forebrain to the hippocampus and cerebral cortex
[1]. Degeneration of these nerve cells results in a decrease
of neuronal markers such as the universal neurotransmitter
acetylcholine and the enzymes choline acetyltransferase
and esterase. The cholinergic hypothesis is based on the fact
that while basal forebrain neurons that express the M2
muscarinic receptor are at risk of degenerating, the cortical
neurons expressing the M1 subtype which synapse with them
are not altered [2]. In theory, one could design and synthesize
muscarinic selective M1 agonists that would activate the
cortical neurons. Such an approach has proven very difficult
to achieve because of lack of three-dimensional structural
information about the muscarinic binding site(s). To date,
three pharmacologically distinct muscarinic receptors and
five cloned muscarinic receptors (M1–M5) have been identi-
fied [3]. These receptors share high amino acid homology,
specifically at the transmembrane regions where agonist
binding is believed to occur. Therefore, it appears difficult
to design selective agonists that can differentiate between
these subtypes. Efficacious agonists would have to activate
specific neurons expressing the M1-subtype while having lit-
tle effect on neurons expressing other types of receptors. To
date, only a few compounds have shown sufficient promise
to proceed to clinical trials, and these have produced only
marginal stimulating responses. Xanomeline, compound 1,
© 2013 HeteroCorporation

J. Boulos, J. Jakubik, A. Randakova, and C. Avila
Vol 000
Our research group is currently working on the synthesis
of heterocyclic compounds with structural features similar
to Xanomeline that conform to Schulman’s [6] model of
the muscarinic pharmacophore. This model is based on de-
tailed conformational analyses of acetylcholine and other
known muscarinic agonists. In this model, the ammonium
head group of muscarinic agonists is shown to interact with
an aspartic acid residue, whereas a region of negative electro-
static potential interacts with a positive receptor residue or
forms a hydrogen bond at point Q (Fig. 1). Schulman’s
model is based on the following assumptions:
(1) The distance |PQ| varies by no more than 0.03nm over
a set of agonists. (2) The interaction dihedral angle “PNOQ”
and the distance |PQ| define the backbone of the drug, and the
range of accessible PNOQ values is 100ꢀ to 117ꢀ, with pos-
itive signs according to the Prelog convention. (3) The dis-
tance |PCt| defines the chain length requirements with full
agonists having |PCt| values around 0.85 nm.
have a piperidinyl group. In addition, compound 7 contains
a hydrophobic butoxy group.
REACTION SCHEMES AND STRATEGIES
The furfuryl ester N-substituted piperidinium salt 4
(Scheme 1) was first synthesized by reacting furfuryl
chloride with 2-bromoethanol to form 2 that was then
coupled with piperidine to yield the piperidine base 3. The
base was then treated with iodomethane to form the salt.
The p-butoxy benzoyl ester N-substituted piperidinium salt
7 (Scheme 2) was first synthesized by reacting 2-bromoetha-
nol with p-butoxybenzoylchloride to form 5 that was then
coupled with piperidine to yield the piperidine base 6. The
base was then treated with iodomethane to form the salt.
The thiophene N-substituted piperidinium salt 11
(Scheme 3) was synthesized by reducing 5-methyl
thiophenecarboxaldehyde with sodium borohydride to
form the alcohol 8 that was then converted to the
corresponding chloride 9. The chloride was then coupled
with piperidine to form the base 10 that was then
converted to the salt 11 by the addition of iodomethane.
Compounds 4, 7, and 11 were synthesized and all contain
the classical NCCO backbone required by Schulman’s
model for muscarinic binding and similar chemical moieties
present in Xanomeline. For example, all three compounds
EXPERIMENTAL
Reagents were purchased from Aldrich Chemical Company
(St. Louis, MO) unless otherwise noted, and all starting liquid
materials were distilled before use. NMR spectra were recorded
on a Varian 300 MHz spectrometer (Varian NMR Systems, Palo
Alto, CA) housed at Barry University. Mass spectra were
recorded on a Clarus 560 S GC/MS system (Perkin Elmer,
Shelton, CT). Elemental analyses were carried out by Galbraith
Laboratories (Knoxville, TN), and biological assays were
conducted at Ricerca Biosciences (Taiwan) and the Institute of
Physiology of the Czech Academy of Sciences as described
previously [7]. Melting points were recorded on a MEL-TEMP II
purchased from Laboratory Devices and are uncorrected.
Figure 1. Schulman’s model of the muscarinic pharmacophore.
Scheme 1
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

Month 2013
Synthesis of N-Substituted Piperidine Salts as
Potential Muscarinic Ligands for Alzheimer’s Applications
Scheme 2
Scheme 3
with a 10:1 mixture of CH2Cl2/MeOH to yield pure compound 3
(0.70 g, 35%). H NMR (300MHz, CD3COCD3) d7.4 (1H), 7.0
1
(1H), 6.4 (1H), 4.6 (2H), 4.2 (2H), 3.6 (4H), 2.4 (4H), 1.2 (2H).
MS: m/z 223 (M+), 111, 98 (base peak).
2-(N-Piperidine ethyl) furfuryl ester N-methyl iodide (4).
Compound 3 (0.60 g, 0.0027 mol) and 1 mL of iodomethane were
added to a 25-mL boiling flask and stirred overnight. The mixture
was concentrated, and the crude salt was recrystallized from hexanol
to yield compound 4 (0.65 g, 67%), mp 165–167ꢀC. 1H NMR
(300MHz, D2O) d 7.7 (1H), 7.3 (1H), 6.6 (1H), 4.9 (2H), 3.8
(2H), 3.5–3.3 (4H), 3.1 (3H), 1.9–1.8 (4H), 1.65-1.55 (2H). Anal.
Calcd For C13H20NO3I: C 42.75%, H 5.52%, N 3.82%, I 34.75%.
Found: C 42.65%, H 5.38%, N 3.78%, I 36.19%.
2-Bromo ethyl p-butoxybenzoate (5).
To a 100-mL
round bottom flask were added 2-bromoethanol (0.625 g,
0.005 mol), triethylamine (0.505 g, 0.005 mol), and 5 mL of
dry anhydrous ether. The flask was cooled in an ice bath,
and p-butoxybenzoylchloride (1.06 g, 0.005 mol) was added
dropwise. The mixture was then refluxed for 30 min. The
solution was filtered, and 15 mL of 1 M HCl was added to
the filtrate. The mixture was extracted several times with
H2O. The ether layer was dried over anhydrous magnesium
sulfate, filtered, and concentrated under vacuum to yield
compound 5 (0.41 g, 27%). 1H NMR (300 MHz, CDCl3) d 8.0
(2H), 7.0 (2H), 4.6 (2H), 4.2 (2H), 3.6 (2H), 1.8 (2H),1.6 (2H),
1.0 (3H). MS: m/z 300 (M+), 121, 138 (base peak).
2-(2-Bromoethyl) furfuryl ester (2). To a 100-mL round
bottom flask were added 2-bromoethanol (6.25 g, 0.05 mol),
triethylamine (5.06 g, 0.05 mol), and 10 mL of anhydrous ether.
The flask was cooled in an ice bath, and 2-furfuryl chloride
(6.52 g, 0.05 mol) was added dropwise. The mixture was then
refluxed for 30 min. The solution was filtered, and 15 mL of
1 M HCl was added to the filtrate. The mixture was extracted
several times with H2O. The ether layer was then dried over
anhydrous magnesium sulfate, filtered, and concentrated under
2-(N-Piperidine ethyl) p-butoxybenzoate (6). Compound 5
(0.41 g, 0.00113 mol), 20 mL of acetonitrile, 1 g sodium carbonate,
and piperidine (0.116 g, 0.00113mol) were added to a 100-mL
boiling flask and left to stir overnight. The solution was filtered
and concentrated under vacuum to yield 0.39 g of crude
compound 6. The crude material was chromatographed over silica
with a 10:1 mixture of CH2Cl2/ MeOH to give compound 6
vacuum to yield compound
2
(3.35 g, 30%). 1H NMR
300 MHz, CDCl3) d 7.65 (1H), 7.3 (1H), 6.55 (1H), 4.7–4.6
(t, 2H), 3.7–3.6 (t, 2H). MS: m/z 219 (M+), 112, 95 (base peak).
1
(0.36 g, 88%). H NMR (300 MHz, CD3COCD3) d 8.0 (2H), 7.1
2-(N-Piperidine ethyl) furfuryl ester (3).
Compound 2
(2H), 4.5 (2H), 4.2 (2H), 2.8 (2H), 2.6 (4H), 1.8 (2H), 1.6 (4H),
1.5 (2H), 1.0 (5H). MS: m/z 305 (M+), 165, 98, 138 (base peak).
2-(N-Piperidine ethyl) p-butoxy benzoylester N-methyl
(2 g, 0.009 mol), 20 mL of acetonitrile, 1 g sodium carbonate,
and piperidine (0.77 g, 0.009 mol) were added to a 100-mL
boiling flask and left to stir overnight. The solution was
filtered and concentrated under vacuum to yield 1.05 g of crude
compound 3. The crude material was chromatographed over silica
iodide (7).
Compound 6 (0.36 g, 0.00118 mol) and 1 mL of
iodomethane were added to a 25-mL boiling flask and stirred
overnight. The mixture was concentrated to yield crude salt 7
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

J. Boulos, J. Jakubik, A. Randakova, and C. Avila
Vol 000
(0.50 g, 63%). The salt was recrystallized from n-butanol, mp
111–113ꢀC. 1H NMR (300 MHz, D2O) d 7.9 (2H), 7.0 (2H),
4.6 (2H), 4.1(2H), 3.8 (2H), 3.5–3.3 (4H), 3.1 (3H), 1.9–1.8
(4H), 1.7–1.5 (4H), 1.45–1.35 (2H), 0.9–0.8 (3H). Anal. Calcd
For C19H30NO3I: C 51.03%, H 6.71%, N 3.13%, I 28.38%.
Found: C 50.48%, H 6.59%, N 3.19%, I 30.29%.
5-Methyl-2-thiophene methanol (8). A solution containing
sodium methylate (3 g, 0.056 mol), sodium borohydride
(6 g, 0.16 mol), and 25 mL of methanol was slowly added to
added with stirring. The reaction mixture was allowed to stir
overnight, filtered, and concentrated on a rotary evaporator to
yield crude base 10 (9.12 g, 96%). To a solution containing
compound 10 (5.0 g, 0.0256 mol) in 20 mL of acetonitrile was
added methyl iodide (4 g, 0.028 mol). The solution was allowed
to stir overnight and concentrated. The crude salt 11 was
recrystallized from 1-butanol (8.16 g 52.85%), mp144–145ꢀC.
1H NMR (300 MHz, CDCl3) d 7.1 (d,1H), 6.8 (d,1H), 4.55
(s, 2H), 3.35–3.2 (m, 4H), 2.9 (s, 3H), 2.4 (s, 3H), 1.95–1.8
(m, 4H), 1.7–1.5 (m, 2H). Anal. Calcd For C12H20SNI: C
42.73%, H 5.98%, N 4.15%, I 37.62%, S 9.51%. Found: C
42.62%, H 5.77%, N 4.14%, I 38.68%, S 9.92%.
a
mixture containing 5-methyl thiophene carboxaldehyde
(26.39 g, 0.08 mol) and 50 ml of methanol with stirring and
cooling. After addition, the reaction mixture was then allowed to
warm to room temperature, poured over 200mL of crushed ice,
and acidified with 6 M HCl. The mixture was then extracted with
ether, and the combined ether layers was dried over anhydrous
magnesium sulfate, filtered, and concentrated to yield 8 (17.90 g,
76.63%), 1H NMR (300MHz, CDCl3) d 6.85 (d,1H), 6.65 (d,1H),
4.75 (s,2H), 2.5 (s,3H), 2.0 (bs, 1H).
RADIOLIGAND ASSAYS AND DISCUSSIONS
Compounds 4, 7, and 11 were assayed at Ricerca
Biosciences for muscarinic binding affinity on Wistar rat
cerebral cortex at 10 mM in 1% DMSO. Incubation time/
temperature: 60 min at 25ꢀC. Incubation buffer: 50 mM
phosphate buffer, pH 7.4. All compounds were tested in
the presence of 0.15 nM [3H] quinuclidinyl benzilate (a
muscarinic antagonist). These compounds were found to
significantly displace the antagonist from the muscarinic
receptor sites. Compounds 7, 11, and 4 displayed 81%,
61%, and 55% inhibition of the antagonist, respectively.
Although compounds 4 and 11 were able to bind signifi-
cantly, neither one had any significant binding selectivity
for any particular receptor. Compound 7, however, was
5-Methyl-2-thiophene methyl chloride (9).
Compound
8 (8.95 g, 0.0698 mol) was added to a mixture of carbon
tetrachloride (50 mL) and triphenyl phosphine (21.0 g, 0.0801 mol)
and refluxed for 1 h. The mixture was allowed to cool to room
temperature, and 160 mL of anhydrous pentane was added with
stirring. The solution was filtered, and filtrate was concentrated to
1
yield 9 (7.13 g, 68.56%). H NMR (300MHz, CDCl3) d 6.9 (d,
1H), 6.6 (d, 1H), 4.8 (s,2H), 2.5 (s,3H).
1-(5-Methyl-2-thiophene methyl)-1-methyl piperidinium
iodide (11).
A 4.0 g of anhydrous sodium carbonate was
added to a solution of compound 9 (7.13 g, 0.0487 mol) in
40 mL of acetonitrile. Piperidine (4.15 g, 0.0487 mol) was then
Figure 2. Carbachol-induced accumulation of inositol phosphate. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

Month 2013
Synthesis of N-Substituted Piperidine Salts as
Potential Muscarinic Ligands for Alzheimer’s Applications
found to be more selective for the M5 receptor. This
compound displayed 61% inhibition at 10 mM of the antag-
onist (3H-QNB) for the M5 receptor (IC50 6.34 mM, Ki
3.93 mM, nH = 0.996) in human recombinant CHO cell
lines. Incubation time/temperature: 2 h at 25ꢀC; buffer:
50 mM Tris–HCl, pH 7.4 10 mM MgCl2, 1 nM EDTA;
Ligand: 0.8 nM [3H] N-methylscopolamine; KD 1.3 nM,
BMAX 5.8 pmol/mg protein. Percent inhibitions at 10 mM
for the M1, M2, M3, and M4 were 47, 36, 23, and 16,
respectively. A significant antagonistic response (70% at
20 mM) was also noted with compound 7 in the muscarinic
M1 functional assay (muscarinic M1, GTPgS binding) with
an IC50 of 8.9 mM in human CHO-K1 cell lines. Another
study was carried on compound 7 at the Institute of
Physiology of the Czech Academy of Sciences on similar
cell lines and set ups. KI values were measured: M1
(2.88 Æ 0.13mM, nH 1.00); M2 (0.91 Æ 0.02 mM, nH 0.802);
M3 (3.16 Æ 0.06 mM, nH 0.837); M4 (7.54 Æ 0.31 mM,
nH 1.05); M5 (2.88 Æ 0.29 mM, nH 1.189) (mean Æ SD
from three independent experiments performed in quadrupli-
cates). In these assays, compound 7 was found to have
highest potency at the M2 receptor and lowest at the M4.
Competition of 100 mM on carbachol-induced accumu-
lation of inositol phosphates confirmed greater potency of
compound 7 at M1 and M3 than at M5 receptors (Fig. 2).
Effects of compound 7 to inhibit carbachol-induced inhibition
of cAMP synthesis were insignificant. More analogs of
compound 7 are being synthesized to increase both receptor
functional selectivity and affinity. Compound 7 is currently
being evaluated for its bitopic antagonist properties at all
M1–M5 receptors.
REFERENCES AND NOTES
[1] Coyle, J. T.; Price, D. L.; Delong, M. R. Science 1983, 219, 1184.
[2] (a) Mash, D. C.; Flynn, D. D.; Potler, L. T. Science 1985, 228,
1115; (b) Meyer, E. M.; Arendash, G. W.; Judkins, J. H.; Ying, L.;
Kem, W. R. J. Neurochem. 1987, 49, 1758; (c) Narang, P. K.; Cutler,
N. R. Prog. Neuro–Psychopharmacol. Biol. Psychiatry 1986, 10, 519.
[3] (a) Buckley, N. J.; Bonner, T. I.; Buckley, C. M.; Brann, M. R.
Mol. Pharmacol. 1989, 35, 469; (b) Bonner, T. I.; Buckley, N. J.;
Young, A. C.; Brann, M. R. Science 1987, 237, 527; (c) Kubo, T.;
Fukuda, K.; Mikami, A.; Maeda, A.; Takahashi, H.; Mishima, M.;
Haga, T.; Haga, K.; Ichiyama, A.; Kangawa, K.; Kojima, M.; Matsuo, H.;
Kirosi, T.; Numa, S. Nature 1986, 323, 411.
[4] (a) Bymaster, F. P.; Whitesitt, C. A., et al. Drug Development
Res. 1997, 40, 158; (b). Fang, L.; Jumpertz, S.; Trankle, C. J. Med. Chem.
2010, 53, 2094.
[5] Grant, M. K.; El-Fakahany, E. JPET 2005, 1, 153, 313.
[6] Schulman, J. M.; Sabio, M. L.; Disch, R. L. J. Med. Chem.
1983, 26, 817.
[7] (a) Jakubik, J.; El-Fakahany, E. E.; Dolezal, V. Mol. Pharma-
col. 2006, 70, 656; (b) Michal, P.; Rudajev, V.; El-Fakahany, E. E.;
Dolezal, V. Eur. J. Pharmacol. 2009, 606, 50.
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet