bis-Azaaromatic quaternary ammonium analogues: Ligands for α4β2* and α7* subtypes of neuronal nicotinic receptors

Article abstract of DOI:10.1016/S0960-894X(02)00687-X

A series of bis-nicotinium, bis-pyridinium, bis-picolinium, bis-quinolinium and bis-isoquinolinium compounds was evaluated for their binding affinity at nicotinic acetylcholine receptors (nAChRs) using rat brain membranes. N,N′-Decane-1,12-diyl-bis-nicoti

Full text of DOI:10.1016/S0960-894X(02)00687-X

Bioorganic & Medicinal Chemistry Letters 12 (2002) 3067–3071
bis-Azaaromatic Quaternary Ammonium Analogues: Ligands for
ꢀ4ꢁ2* and ꢀ7* Subtypes of Neuronal Nicotinic Receptors
Joshua T. Ayers, Linda P. Dwoskin, A. Gabriela Deaciuc, Vladimir P. Grinevich,
Jun Zhu and Peter A. Crooks*
Division of Pharmaceutical Sciences, College of Pharmacy, University of Kentucky, Lexington, KY 40536-0082, USA
Received 23 May 2002; accepted 5 August 2002
Abstract—A series of bis-nicotinium, bis-pyridinium, bis-picolinium, bis-quinolinium and bis-isoquinolinium compounds was
evaluated for their binding affinity at nicotinic acetylcholine receptors (nAChRs) using rat brain membranes. N,N0-Decane-1,12-
diyl-bis-nicotinium diiodide (bNDI) exhibited the highest affinity for [3H]nicotine binding sites (Ki=330 nM), but did not inhibit
[3H]methyllycaconitine binding (Ki >100 mM), indicative of an interaction with a4b2*, but not a7*receptor subtypes, respectively.
Also, bNDI inhibited (IC50=3.76 mM) nicotine-evoked 86Rb+ efflux from rat thalamic synaptosomes, indicating antagonist activity
at a4b2*nAChRs. N,N0-Dodecane-1,12-diyl-bis-quinolinium dibromide (bQDDB) exhibited highest affinity for [3H]methyllycaco-
nitine binding sites (Ki=1.61 mM), but did not inhibit [3H]nicotine binding (Ki>100 mM), demonstrating an interaction with a7*,
but not a4b2*nAChRs. Thus, variation of N-n-alkyl chain length together with structural modification of the azaaromatic qua-
ternary ammonium moiety afforded selective antagonists for the a4b2*nAChR subtype, as well as ligands with selectivity at a7*
nAChRs.
# 2002 Elsevier Science Ltd. All rights reserved.
S(ꢀ)-Nicotine (NIC; Fig. 1) produces many effects fol-
lowing stimulation of neuronal nicotinic acetylcholine
receptors (nAChRs), including stimulant effects,
analgesia, anxiolytic activity, and enhancement of
learning and memory. Nicotine and compounds acting
at nAChRs have been suggested as potential therapeutic
agents for several pathological conditions, including
Parkinson’s disease, Tourette’s syndrome and Alzhei-
mer’s disease.1ꢀ5 nAChRs are pentameric ligand-gated
ion channels, composed of two varieties of subunits, a
and b.6,7 To date, ten a (a1–a10) and four b (b1–b4)
subunits have been identified.8 nAChRs have the gen-
eral stoichiometry of 2a and 3b subunits that constitute
the ion channel.9 Expression of these nAChRs in Xeno-
pus oocytes has shown that association of a2, a3, and
a4 subunits with b2 and b4 subunits forms functional
heterologous receptor channels, whereas a7, a8, and
a9 subunits assemble to form homologous receptor
channels.10ꢀ12 Differences in subunit composition con-
tribute to nAChR pharmacology and functional diver-
sity.13,14 Research is currently underway to identify the
specific subunit composition of native nAChRs,8 the
putative nature of which is indicated by an asterisk
beside the subunit designation herein.
Over the last decade, efforts have focused on the devel-
opment of nAChR agonists as therapeutic
agents.1ꢀ4,15,16 However, relatively little attention has
focused on the development of nAChR antagonists as
potential drug candidates, and there are very few
nAChR subtype selective antagonists available.17,18
bis-Quaternary ammonium salts, such as hexametho-
nium chloride (HEX) and decamethonium bromide
(DEC), are regarded as simplified analogues of d-tubo-
19,20
curarine (TBC)
(see Fig. 1), and have been utilized
to differentiate between subtypes of peripheral nicotinic
receptors.20ꢀ22 Thus, DEC inhibits neuromuscular
nicotinic receptors, while HEX primarily inhibits gang-
lionic nicotinic receptors.23,24 Recently, quaternary
ammonium N-n-alkyl analogues of NIC [i.e., N-n-
octylnicotinium iodide (NONI), N-n-decylnicotinium
iodide (NDNI) and cis-6-decyl-1-methyl-2,3,3a,4,5,9b-
hexahydro-1H-pyrrolo[2,3-f]quinolin-6-ium iodide (BCD;
Fig. 1)] have been reported as nAChR subtype selective
*Corresponding author. Fax: +1-859-257-7585; e-mail: pcrooks@
pop.uky.edu
Website: http://www.uky.edu/pharmacy/faculty/crooks
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00687-X

3068
J. T. Ayers et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3067–3071
Figure 1. Structures of S-(ꢀ)-nicotine, hexamethonium, deca-
methonium, d-tubocurarine, NONI, NDNI and BCD.
antagonists.25ꢀ30 These latter findings prompted us to
evaluate a series of bis-quaternary ammonium com-
pounds containing azaaromatic head groups for their
interaction with a4b2*and a7*nAChR subtypes.
Chemistry
Compounds 1–6 (Fig. 2) were prepared by first dissol-
ving NIC in glacial acetic acid, stirring the mixture for 5
min, and adding the appropriate diiodo- or dibromo-
alkane. The solution was stirred under reflux for 3 days,
the solvent was evaporated under reduced pressure,
and the resulting residue was treated with an aqueous
saturated solution of NaHCO3. The mixture was then
extracted with diethyl ether (3ꢁ50 mL), and then with
chloroform (3ꢁ50 mL). The aqueous layer was col-
lected, lyophilized and the resulting solid triturated with
chloroform. After filtration, the filtrate was dried over
anhydrous MgSO4. Filtration and removal of the sol-
vent afforded the desired bis-nicotinium salt. Com-
pounds 7–31 (Fig. 2) were prepared by reacting an
excess of the appropriate azaaromatic compound with a
variety of diiodo- or dibromoalkanes for 24 h at 65 ꢂC
in the absence of solvent. The resulting solid was col-
lected by filtration, dissolved in water and the aqueous
solution washed with diethyl ether (3ꢁ50 mL). The
Figure 2. Structures of bis-azaaromatic quaternary ammonium ana-
logues 1–31.
aqueous solution was then lyophilized to afford either a
solid or a viscous hygroscopic oil. All compounds were
1
characterized by H and 13C NMR spectroscopy, mass
spectroscopy and elemental analysis.31
Biological Assays
Compounds 1–31 were evaluated for their ability to
inhibit [3H]NIC and [3H]methyllycaconitine ([3H]MLA)

J. T. Ayers et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3067–3071
3069
binding to rat brain membranes, indicative of affinity
30,32
NIC exhibited a high affinity (Ki=1.4 nM) for the
a4b2*nAChR, and a relatively lower affinity ( Ki=0.45
mM) for the a7*nAChR subtype when compared to
classical antagonists, a-bungarotoxin (aBTX; Ki=7
nM) and MLA (Ki=2.3 nM). Although HEX and DEC
had 3-orders of magnitude lower affinity for the a4b2*
subtype compared to NIC, DEC had 6-fold higher affi-
nity (Ki=5.73 mM) for the a4b2*subtype compared to
HEX. Neither HEX nor DEC had affinity for the a7*
nAChR subtype. TBC exhibited Ki values of 4.79 and
9.72 mM at a7*and a4b2*nAChRs, respectively.
for a4b2*and a7*nAChRs, respectively.
Ana-
logue-induced inhibition of binding was expressed as a
percent of control and fitted by nonlinear, non-weigh-
ted least squares regression using a fixed slope sigmoi-
dal function. The log IC50 value represented the
logarithm of analogue concentration required to
decrease binding by 50%. IC50 values were corrected for
ligand concentration according to the Cheng–Prussoff
equation33 to yield inhibition constants (Ki), such that
Ki=IC50/(1+c/Kd), where c is the concentration of free
radioligand and Kd is the equilibrium dissociation con-
stant of the ligand. Protein concentration was deter-
mined using published methods.34 Results are reported
as Ki values (ꢃSEM) in Table 1. Previous results have
demonstrated a high correlation between inhibition of
[3H]NIC binding and inhibition of NIC-evoked 86Rb+
efflux, such that the latter assay has been utilized to
assess the functional interaction of ligands at the a4b2*
nAChR subtype.35 To ascertain if the most potent ana-
logues in the [3H]NIC binding assay act as antagonists
at a4b2*nAChRs, inhibition of NIC (3 mM)-evoked
86Rb+ efflux from rat thalamic synaptosomes was
determined.
In a homologous series of N-n-alkyl nicotinium ana-
logues, with alkyl groups ranging from C1–C12, the C10
analogue NDNI exhibited highest affinity (Ki=64 nM)
26,28
for the a4b2*subtype.
Furthermore, NDNI inhib-
ited NIC-evoked 86Rb+ efflux, indicating that it acts as
26,29
a pharmacological antagonist of a4b2*nAChRs.
Interestingly, BCD, a conformationally restricted ana-
logue of NDNI, had no affinity for the a4b2*subtype. 30
The C8 analogue, NONI, which has a shorter n-alkyl
chain length, exhibited a significantly lower affinity
(Ki=20 mM) at a4b2*.26,28 Neither NONI nor NDNI
exhibited high affinity for the a7*nAChR subtype.
Table 1. Ki Values for compounds 1–31 in the [3H]NIC and [3H]MLA binding assays using rat brain membranes
Compd
Ki [3H]NIC binding assay (mM)
Ki [3H]MLA binding assay (mM)
NIC
ꢀBTX
MLA
HEX
DEC
TBC
NONI
NDNI
1
2
3
4
5
6
7
8
9
10
11
12
13
14
0.0014ꢃ0.0001a
0.45ꢃ0.09
0.007ꢃ0.0003
0.0023ꢃ0.0008
>100
>100
1.56ꢃ0.01
21.7ꢃ3.80
5.73ꢃ0.76
9.72ꢃ2.23
19.7ꢃ1.50
0.064ꢃ0.004
20.0ꢃ2.84
1.54ꢃ0.17
5.03ꢃ0.80
0.33ꢃ0.01
0.43ꢃ0.06
1.95ꢃ0.19
>100
97.8ꢃ17.5
34.3ꢃ2.74
24.2ꢃ1.19
18.7ꢃ3.09
14.0ꢃ3.21
9.15ꢃ0.17
>100
>100
4.79ꢃ0.20
23.6ꢃ4.85
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
57.0ꢃ4.89
33.0ꢃ4.60
>100
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
>100
79.9ꢃ16.6
>100
>100
>100
>100
>100
>100
69.2ꢃ28.9
48.6ꢃ17.2
39.3ꢃ5.59
>100
13.3ꢃ2.12
2.39ꢃ0.69
3.58ꢃ0.45
7.81ꢃ1.90
2.58ꢃ0.46
1.61ꢃ0.21
16.0ꢃ1.92
11.7ꢃ5.63
5.95ꢃ1.01
7.25ꢃ3.02
5.46ꢃ0.78
2.21ꢃ0.25
22.5ꢃ1.30
>100
37.8ꢃ8.65
>100
17.9ꢃ3.59
65.5ꢃ14.5
7.89ꢃ0.92
>100
9.02ꢃ1.56
31
6.10ꢃ0.73
aData represent Ki value expressed mean ꢃ SEM.

3070
J. T. Ayers et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3067–3071
Based on the interesting pharmacological profile of the
N-n-alkylnicotinium analogues and the well-established
pharmacology of the bis-trimethylammonium alkanes,
including HEX and DEC, a series of bis-azaaromatic
quaternary ammonium analogues were synthesized and
evaluated for interaction with a4b2*and a7*nAChR
subtypes in order to identify ligands which are both
potent and selective at these nAChRs. The bis-nicoti-
nium compounds (1–6) were thought to be a particu-
larly interesting series of target compounds, since they
are related structurally to NONI and NDNI.
quinolinium analogue, bQDDB (25) exhibited a Ki
value of 1.61 mM for the a7*nAChR and no affinity for
the a4b2*nAChR subtype. Importantly, replacing the
bis-nicotinium headgroups with quinolinium moieties
afforded a compound with no affinity for a4b2*
nAChR, but had similar affinity to NIC at the a7*
nAChR subtype. Thus, two interesting bis-azaaromatic
quaternary ammonium compounds have emerged from
this structure activity study, bNDI and bQDDB, which
exhibit good affinity and selectivity for a4b2*and a7*
nAChR subtypes, respectively. Subtype-selective
nAChR ligands, such as bNDI and bQDDB, constitute
useful agents for both basic and clinical research aimed
at determining the role of nAChRs in physiological
function.
The bis-nicotinium analogues 1–6 exhibited Ki values
for the a4b2*subtype ranging from 0.33–20 mM, the
most potent being the C10 analogue, bNDI (Table 1).
None of the bis-nicotinium analogues had affinity for
the a7*nAChR subtype. bNDI had a similar potency to
NDNI at a4b2*nAChRs. Moreover, bNDI and bNUB
inhibited NIC-evoked 86Rb+ efflux from rat thalamic
synaptosomes (IC50=3.76ꢃ1.1 mM and 5.22ꢃ1.4 mM,
respectively), without producing intrinsic activity, indi-
cating that these bis-nicotinium analogues are func-
tional antagonists at the a4b2*nAChR subtype.
Acknowledgements
This work was supported by grants from the National
Institute on Drug Abuse (DA10934, DA00399).
References and Notes
1. Lloyd, G. K.; Williams, M. J. Pharmacol. Exp. Ther. 2000,
292, 461.
2. Holladay, M. W.; Dart, M. J.; Lynch, J. K. J. Med. Chem.
1997, 40, 4169.
3. Lin, N.; Meyer, M. D. Exp. Opin. Ther. Patents 1998, 8,
991.
4. Schmitt, J. D. Current Med. Chem. 2000, 7, 749.
5. Court, J. A.; Piggott, M. A.; Lloyd, S.; Cookson, N.; Bal-
lard, C. G.; McKeith, I. G.; Perry, R. H.; Perry, E. K. Neu-
roscience 2000, 98, 79.
6. Le Novere, N.; Changeux, J. P. J. Mol. Evol. 1995, 40, 155.
7. Ortells, M.; Lunt, G. G. Trends Neurosci. 1995, 18, 121.
8. Lukas, R. J.; Changeux, J. P.; Novere, N. L.; Albuquerque,
E. X.; Balfour, D. J.; Berg, D. K.; Bertrand, D.; Chiappinelli,
V. A.; Clarke, P. B. S.; Collins, A. C.; Dani, J. A.; Grady,
S. R.; Kellar, K. J.; Lindstrom, J. M.; Marks, M. J.; Quik,
M.; Taylor, P. W.; Wonnacott, S. Pharmacol. Rev. 1999, 51,
397.
9. Cooper, E.; Courtier, S.; Ballivet, M. Nature 1991, 350,
235.
10. Papke, R. L.; Boulter, J.; Patrick, J. H.; Heinemann, S.
Neuron 1989, 3, 589.
The simple bis-pyridinium analogues (7–13) generally
had a 10-fold lower affinity for the a4b2*subtype com-
pared to the bis-nicotinium analogues (1–6). As the
carbon chain in the bis-pyridinium analogues increased
from C5 to C12, the affinity for the a4b2*subtype also
increased by an order of magnitude. Thus, the C12 ana-
logue 13 was the most potent (Ki=9.15 mM) in the series
at the a4b2*subtype. None of the compounds in the
bis-pyridinium series exhibited high affinity for the a7*
nAChR subtype. Interestingly, the structurally related
bis-picolinium analogues (14–19) generally had low
affinity for both a4b2*and a7*subtypes.
The bis-quinoliniums analogues (20–25) generally
exhibited low affinity for the a4b2*subtype. The struc-
ture activity relationship for the bis-isoquinoliniums
analogues (26–31) resulted in a range of affinity values
for the a4b2*subtype ( Ki=6.1–65.5 mM; except bIQDI,
Ki>100 mM). As a group, the bis-quinolinium and bis-
isoquinolinium analogues exhibited Ki values ranging
from 1.61 - 16.0 mM for the a7*nAChR subtype. Thus,
the two most potent analogues in these two groups were
the C12 analogues 25, bQDDB, and 31, bIQDDB; and
moreover, bQDDB demonstrates excellent selectivity
for the a7*nAChR subtype as well.
11. Luetje, C. W.; Patrick, J. J. Neurosci. 1991, 11, 837.
12. Chavez-Noriega, L. E.; Crona, J. H.; Washburn, M. S.;
Urrutia, A.; Elliot, K. J.; Johnson, E. C. J. Pharmacol. Exp.
Ther. 1997, 280, 346.
13. Cachelin, A. B.; Rust, G. Pflugers Arch. Eur. J. Physiol.
1995, 429, 449.
14. Harvey, S. C.; Luetje, C. W. J. Neurosci. 1996, 16, 3798.
15. Tønder, J. E.; Olesen, P. H. Curr. Med. Chem. 2001, 8,
651.
16. Glennon, R. A.; Dukat, M. In Neuronal Nicotinic Recep-
tors; Arneric, S. P.; Brioni, J. D., Eds.; Wiley-Liss: New York,
1999; p 271.
17. Dwoskin, L. P.; Xu, R.; Ayers, J. T.; Crooks, P. A. Exp.
Opin. Ther. Patents 2000, 10, 1561.
18. Dwoskin, L. P.; Crooks, P. A. J. Pharmacol. Exp. Ther.
2001, 298, 395.
19. Everett, A. J.; Lowe, L. A.; Wilkinson, S. Chem. Commun.
Chem. Soc. London 1970, 1020.
20. Koelle, G. B. In The Pharmacological Basis of Ther-
apeutics, 5th Ed.; Goodman, L. S.; Gilman, A. Eds.; MacMil-
lan Pub. Co.: New York, 1975; p 565.
From the structure activity trends in compounds 1–31,
it is apparent that replacing the N-trimethylammonium
moieties such as HEX and DEC with azaaromatic moi-
eties can afford molecules with affinity for a4b2*and
a7*nAChR subtypes. Of particular interest is the find-
ing that two compounds in this study exhibit high
selectivity at these nAChR subtypes. In this respect, the
bis-nicotinium analogue, bNDI (4) exhibited a Ki value
of 330 nM at the a4b2*nAChR subtype, and function-
ally inhibited NIC-evoked 86Rb+ efflux (IC50=3.76
mM), while demonstrating no affinity for the a7*
nAChR subtype, indicating that this analogue is a
selective a4b2*nAChRs antagonist. In contrast, the bis-

J. T. Ayers et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3067–3071
3071
21. Khromov-Borisov, N.; Michelson, M. Pharmacol. Rev.
1966, 18, 1051.
22. Michelson, M.; Zaimal, E. Acetylcholine, An Approach to
the Molecular Mechanism of Action; Pergammon: Oxford,
1973; p 243.
23. Paton, W.; Zaimis, E. J. Nature (London) 1948, 162, 810.
24. Barlow, R. B.; Ing, H. R. Brit. J. Pharmacol. 1948, 3,
298.
25. Crooks, P. A.; Ravard, A.; Wilkins, L. H.; Teng, L. H.;
Buxton, S. T.; Dwoskin, L. P. Drug Dev. Res. 1995, 36, 91.
26. Dwoskin, L. P.; Wilkins, L. H.; Pauly, J. R.; Crooks, P. A.
Ann. N.Y. Acad. Sci. 1999, 868, 617.
27. Wilkins, L. H.; Haubner, A.; Ayers, J. T.; Crooks, P. A.;
Dwoskin, L. P. J. Pharmacol. Exp. Ther. 2002, 301, 1088.
28. Wilkins, L. H.; Grinevich, V. P.; Ayers, J. T.; Crooks, P.
A.; Dwoskin, L. P.; J. Pharmacol. Exp. Ther. 2002, submitted
for publication.
31. 4: 1H NMR (300 MHz, CDCl3) d 9.55 (1H, s, C2-H), 9.44
(1H, d, C6-H), 8.53 (1H, d, C4-H), 8.09 (1H, m, C5-H), 4.98
(2H, t, C00 1-CH2), 3.74 (1H, t, pyrrolidine CH2), 3.30 (1H, t,
pyrrolidine CH2, 2.50 (2H, m, C00 2-CH2), 2.29 (3H, s, pyrro-
lidine N-CH3), 1.65.20 (5H, m, pyrrolidine CH2CH2), 1.30.57
(6H, m, C00 3–5-CH2); 13C NMR (75 MHz, CDCl3) d 146.6,
143.9ꢁ2, 143.4, 128.4, 66.9, 61.9, 57.0, 40.8, 36.1, 32.1, 28.5,
28.2, 25.8, 23.5, C30H48I2N4. Yellow oil, 27% yield. 25: 1H
NMR (300 MHz, DMSO-d6) d 9.58 (1H, d, C2-H), 9.29 (1H,
d, C3-H), 8.61 (1H, d, C8-H), 8.50 (1H, d, C4-H), 8.29 (1H, t,
C7-H), 8.18 (1H, t, C5-H), 8.04 (1H, d, C6-H), 5.04 (2H, t,
C’1-CH2), 1.95 (2H, m, C02-CH2), 1.2.4 (8H, m, C03-6-CH2);
13C NMR (75 MHz, DMSO-d6) d 149.5, 147.4, 137.4, 135.6,
130.7, 129.9, 129.7, 122.1, 118.9, 57.3, 29.5,28.9, 28.8, 28.5,
25.7, C30H38Br2N2. glassy solid, 92% yield.
32. Marks, M. J.; Stitzel, J. A.; Romm, E.; Wehner, J. M.;
Collins, A. C. Mol. Pharmacol. 1986, 30, 427.
29. Wilkins, L. H.; Miller, D. K.; Ayers, J. T.; Crooks, P. A.;
Dwoskin, L. P.; J. Pharmacol. Exp. Ther., 2002, submitted for
publication.
33. Cheng, Y.; Prussoff, W. H. Biochem. Pharmacol. 1973, 22,
3099.
34. Bradford, M. M. Ann. Biochem. 1976, 72, 248.
30. Xu, R.; Dwoskin, L. P.; Grinevich, V. P.; Sumithran, S. P.;
Crooks, P. A. Drug Dev. Res. 2002, 55, 173.
35. Marks, M. J.; Bullock, A. E.; Collins, A. C. J. Pharmacol.
Exp. Ther. 1995, 274, 833.