Preparation and characterization of N-(3-pyridinyl) spirocyclic diamines as ligands for nicotinic acetylcholine receptors

Article abstract of DOI:10.1016/j.bmcl.2009.01.099

Several N-pyridin-3-yl spirobicyclic diamines, designed as conformationally restricted analogs of tebanicline (ABT-594), were synthesized as novel ligands for nicotinic acetylcholine receptors (nAChR). The spirocyclic compounds exhibited weaker binding affinity, than other constrained analogs in accord with a pharmacophore model. Nevertheless, some (1a, 1b) possessed (partial) agonist potencies comparable to nicotine at the α4β2 subtype, but with greatly improved selectivity relative to the α3β4* nAChR.

Full text of DOI:10.1016/j.bmcl.2009.01.099

Bioorganic & Medicinal Chemistry Letters 19 (2009) 1682–1685
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Preparation and characterization of N-(3-pyridinyl) spirocyclic diamines
as ligands for nicotinic acetylcholine receptors
*
Kevin B. Sippy , David J. Anderson, William H. Bunnelle, Charles W. Hutchins, Michael R. Schrimpf
Neuroscience Research, Abbott Laboratories, Department R47W, 100 Abbott Park Rd., Building AP9A, Abbott Park, IL 60064-6117, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Several N-pyridin-3-yl spirobicyclic diamines, designed as conformationally restricted analogs of teban-
icline (ABT-594), were synthesized as novel ligands for nicotinic acetylcholine receptors (nAChR). The
spirocyclic compounds exhibited weaker binding affinity, than other constrained analogs in accord with
a pharmacophore model. Nevertheless, some (1a, 1b) possessed (partial) agonist potencies comparable to
Received 15 December 2008
Revised 28 January 2009
Accepted 29 January 2009
Available online 4 February 2009
nicotine at the
a
4b2 subtype, but with greatly improved selectivity relative to the
a3b4ꢀ nAChR.
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Nicotinic
nAChR
Acetylcholine
Receptors
a4b2
ABT-594
Diamines
The potential of nicotinic acetylcholine receptor (nAChR) ago-
nists as novel, broad-spectrum analgesics is now well recog-
nized,1–3 but development of effective drugs with suitable safety
and tolerability profiles has proved challenging. Multiple subtypes
of nAChRs are expressed throughout the central and peripheral
nervous systems, and it is likely that receptor subtypes that medi-
ate the analgesic response are different than those responsible for
many of the side effects. For example, knock-out and antisense
a
3b4ꢀ nAChR compared to the
a4b2 subtype) and accordingly
exhibits a very narrow separation between analgesic efficacy and
side effects in animal models.6 Tebanicline (ABT-594) was devel-
oped at Abbott Laboratories as a potent analgesic with greater
selectivity for the
a
4b2 subtype compared to epibatidine.8,9
Although tebanicline exhibited an improved therapeutic index that
enabled its advancement to human clinical trials, dose-limiting GI
side effects precluded full development of this compound.2 In an ef-
fort to identify analogs of tebanicline that discriminate between
nAChR subtypes, we have sought to rigidify the core structure of
tebanicline by replacing the ether linker with a cyclic secondary
amine (Fig. 1). Cyclization via the aza linker can be accomplished
in three distinct ways, to provide bridged (A) or fused (B) bicyclic
diamines, which we have described elsewhere,10,11 or spirobicyclic
diamines (C), which are the subject of this report. These spirocyclic
structures thus complete an interesting set of conformationally-
rigidified bicyclic diamine nAChR ligands.
Construction of the diamines was accomplished as outlined in
Scheme 1. Boc-protected methyl (S)-azetidine-2-carboxylate12
was allylated via the enolate. Ozonolysis and reductive amination
followed by thermal cyclization led to the lactam 8 which was re-
duced with LiAlH4 to the differentially-protected diamine 9.
Hydrogenolysis, Pd-mediated coupling of the liberated secondary
amine with 5-bromo-2-chloropyridine, and removal of the Boc-
protecting group provided 1a. Alternatively, coupling with 3-bro-
mopyridine led to the des-chloro analog 1b. In a similar fashion,
N-benzoyl-(S)-proline methyl ester12 was elaborated to benzyl-
protected 1,7-diazaspiro[4.4]nonane 11 as described by Culbertson
studies have implicated the
a4b2 subtype as critical for analgesic
efficacy,4,5 while activation of
a
3b4ꢀ nAChRs, abundantly ex-
pressed in the autonomic ganglia, has been linked to side effect lia-
bilities.6 Therefore, our medicinal chemistry efforts are directed
toward nAChR agonists that improved selectivity for the a4b2 sub-
type compared to the
a3b4ꢀ nAChR.
HN
H
O
N
N
H
Me
N
N
Cl
N
Cl
epibatidine
nicotine
tebanicline
The frog skin alkaloid epibatidine,7 has stimulated most of the re-
cent interest in nAChRs as a target for treatment of pain, but has
a poor subtype selectivity profile (four-fold more potent at the
* Corresponding author. Tel.: +1 847 937 4361.
E-mail address: kevin.sippy@abbott.com (K.B. Sippy).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.01.099

K. B. Sippy et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1682–1685
1683
O
O
O
a
b
HN
a
b
N
N
N
N
N
a
b
A
Py
Py
Py
O
O
O
O
Ph
Ph
Ph
c
Ph
15
16
17
X
HN
HN
Py
N
N
c
d
e, f
tebanicline X=O
rigidification form X=N
B
C
N
N
HN
18
HN
c
O
Ph
Ph
19
N
H
g, h
N
NH
Figure 1. Cyclization of a secondary amine to provide bridged, fused or spirocyclic
diamines.
HN
N
Boc
3a X = Cl (69%)
20
N
X
3b
X = H (74%)
Scheme 3. Reagents and conditions: (a) PPh3, 37% CH2O, AcOH (57%); (b)
N-(methoxymethyl)-N-(trimethylsilylmethyl)benzyl amine, CH2Cl2, TFA, 0 °C
(92%); (c) EtOH, H2, 20% Pd(OH)2, 50 °C (96%); (d) LiAlH4, THF (98%); (e) CH2Cl2,
Boc2O (63%); (f) EtOH, 10% Pd/C, H2 (100%); (g) Pd2(dba)3, BINAP (rac.), sodium tert-
butoxide, 5-bromo-2-X-pyridine, toluene; (h) EtOAc, 4M HCl/dioxane (69%).
H3C
CH3
b
a
c
CO2H
CO2CH3
N
H
N
N
CO2CH3
Boc
Boc
4
5
6
The 2,7-diazaspiro[4.4]nonanes were prepared via a 1,3-dipolar
cycloaddition of 2-methylenesuccinimide 16 with an azomethine
ylide14 as shown in Scheme 3 to generate the spiroamine 17. Imide
reduction, Boc protection, and debenzylation provided the mono-
protected diamine 20 which was carried through to the 3-pyridyl
derivatives 3a and 3b as before.
e, f
CHO
CO2CH3
7
N
N
d
N
Ph
N
Ph
N
Boc
Boc
9
O
8
Boc
Affinities of the compounds at the a4b2 nAChR were determined
h, i
g
from displacement of [3H]cytisine from rat brain membranes as
described previously.15 Agonist activities were evaluated from
NH
N
N
N
H
Boc
10
Ca2+ dynamics using FLIPR technology.10,16 For
a4b2, a cell line
N
X
X = Cl (56%)
1b X = H (71%)
1a
expressing the human recombinant receptor was employed, while
a human neuroblastoma cell line (IMR-32) was used to assess activ-
ity at the
a
3b4ꢀ subtype.
Scheme 1. Reagents and conditions: (a) (i) CH3OH, SOCl2, 0–90 °C; (ii) Boc2O, Et3N,
CH2Cl2, (95%); (b) LiHMDS, THF, ꢁ65 °C, 1-bromo-3-methylbut-2-ene (92%) (c) O3,
Ph3P, CH3OH, CH2Cl2, ꢁ65 °C (75%); (d) (i) PhCH2NH2, NaBH3CN, HOAc, CH3OH, 0 °C,
(ii) toluene, 110 °C (70%); (e) TFA, CH2Cl2 (f) (i) LiAlH4, THF, 0 °C-reflux (ii) Boc2O
(64%); (g) H2, Pd/C, EtOH (84%); (h) Pd2(dba)3, racemic-BINAP, sodium tert-
butoxide, 5-bromo-2-X-pyridine, toluene (i) TFA, CH2Cl2.
Although pyridinyl ethers related to tebanicline are exception-
ally potent ligands at nAChRs, simple replacement of oxygen by
nitrogen causes a sharp decrement in binding potency (Fig. 2).
Thus, 21 displaces [3H]cytisine with sevenfold greater potency
than nicotine,15 while the aza-analog 22 is 11-fold weaker than
nicotine.17 N-Methylation of the linking nitrogen (23) is even more
detrimental to binding affinity, resulting in this instance in a com-
pound that is 700 times less potent than nicotine. Nevertheless, we
have demonstrated that structures where the N-alkyl substituent
is tethered to the azacyclic ring exhibit subnanomolar binding
affinities for both bridged and fused bicyclic motifs.10,11 From the
data in Table 1, it is evident that the spirocyclic diamines exhibit
binding potencies in the double-digit nanomolar range, ranking
them as somewhat less potent than nicotine but substantially
more potent than unconstrained analogs like 23. As observed for
the pyridinyl ethers,9,18 the azetidines 1 are slightly more potent
than the homologous pyrrolidines 2 and 3. The 2,7-diazaspi-
ro[4.4]nonanes 3, where the distance between the secondary
amine and the pyridine has been extended, show weaker binding
affinity.
et al.,13 and after swapping the benzyl for Boc-protecting group as
outlined in Scheme 2, palladium coupling and subsequent depro-
tection provided the 7-(6-chloro-pyridin-3-yl)-1,7-diaza-spiro[4.4]-
nonane analog 2a and the des-chloro analog 2b.
a
b
N
N
N
NH
N
N
Boc
CF3
CF3
Ph
c
Ph
O
O
11
12
13
d, e
N
NH
N
NH
Boc
N
X
2a X = Cl (77%)
2b
21 X = O
22 X = NH
23 X = NCH3
X
N
X = H (73%)
14
H
Scheme 2. Reagents and conditions: (a) TFAA, Et3N, THF, 0 °C (87%); (b) Boc2O,
EtOAc, 20% Pd(OH)2, H2 (75%); (c) K2CO3, CH3OH (83%); (d) Pd2(dba)3, BINAP (rac.),
sodium tert-butoxide, 5-bromo-2-X-pyridine, toluene (e) TFA, CH2Cl2.
N
Figure 2. Pyridinyl ether and aza related analogs of tebanicline.

1684
K. B. Sippy et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1682–1685
Tønder and Olessen21,22 that employs two receptor interaction
points (a and b) located 2.9 Å along the trajectory of the NH bond
of the basic amine and the lone pair of the pyridine ring, respec-
tively, as well as the centroid (c) of the heterocycle. According to
Table 1
In vitro pharmacological data for the spirocyclic diamines
Compound
Rat
a
4b2
Human
a
M)
4b2
Human
EC50 (l
a
M)
3b4
b
b
Kia (nM)
EC50
(l
this model, the optimum geometry for high affinity at the
a4b2
Nicotine
Epibatidine
0.94
0.047
0.048
0.15
1.2
21
11
48
33
4.7 (100%)
0.12 (154%)
0.020 (115%)
nd
1.6 (122%)
2.5 (48%)
25 (44%)
>100 (6%)
>100 (20%)
>100 (9%)
>100 (4%)
9.4 (100%)
0.027 (132%)
0.19 (169%)
6.4 (108%)
6.1 (90%)
nAChR is achieved when distances a–b and a–c fall in the ranges
7.3–8.0 Å and 6.5–7.4 Å, respectively, with \bac within 30.4–
35.8ꢀ. To a first approximation, structures A, B, and C (see Fig. 1)
represent different rotamers about the exocyclic CAN bond exter-
nal to the pyridine substituted ring. This has the effect of changing
orientation of the NAH bonds relative to the pyridine ring, and
consequently, the ability of the respective compounds to fit the
parameters of the pharmacophore model. Structures A, B, and C
were generated and the geometry optimized using CFF force
field.23 Rotation about the N-pyridine bond was then allowed to
produce the best fit within the parameters of the model.22
Tebanicline
A19
B20
1a
6.9 (23%)
1b
2a
2b
3a
>100 (4%)
>100 (8%)
>100 (15%)
>100 (2%)
>100 (10%)
240
74
3b
a
Values are means of at least three experiments.
Average of six replicates; maximal response normalized to 100
b
l
M nicotine is
given in parenthesis. nd, not determined.
As shown in Table 2, the bridged and fused ring diamine struc-
tures A and B, respectively, both achieve a good fit to the pharma-
cophore model, with each of the three calculated parameters
within the optimal range of the model. This is consistent with the
potent (nanomolar) binding affinity for these series (Table 1). In
contrast, spirocyclic structure C cannot achieve a conformation that
satisfies the criteria of the model. The best-fit conformation for C
c
N
N
H
2.9 Å
2.9 Å
approaches the optimal ranges for a–b and \bac, but the a–c dis-
b
0
a
tance is too short by 1.1 ÅA. Moreover, this conformer involves sub-
stantial twisting of the pyridinyl-N bond away from the minimal
energy conformation, an energy cost that will further disfavor fit
to the pharmacophore model. Thus, the reduced binding affinity
of C relative to A and B is in accord with the pharmacophore model.
In summary, the spirocyclic diamine represents a new core for
construction of nAChR ligands. The lower affinity of the spirocyclic
structures relative to fused- and bridged-bicyclic counterparts is
consistent with recently developed pharmacophore models. On
the other hand, the spirocyclic amines may exhibit enhanced selec-
Figure 3. The definition of a novel nicotinic pharmacophore model22, where a is the
site point corresponding to the protonated nitrogen atom, b is the site point
corresponding to the electronegative atom capable of forming a hydrogen bond, c is
the center of a heteroaromatic ring or a C@O bond. \bac is the angle measured
between the interatomic distance vectors a–b and a–c.
Of these spirocyclic diamines, only the azetidines 1a,b display
any agonist activity. Both exhibit potencies comparable to nicotine
at the a4b2 nAChR that is considered to be crucial to the analgesic
response, but they are only partial agonists (40–50% of the full nic-
tivity for the a4b2 subtype over other nAChRs.
otine response) at this receptor. Interestingly, although 1b has
somewhat higher affinity than 1a for the
a4b2 nAChR, it is about
References and notes
10-fold weaker as an agonist. This might reflect a species difference
(rat receptor for binding, human for functional activity), or may
indicate that the structural requirements for the binding confor-
mation of the receptor (considered to be the desensitized state)
are changed in the activated (channel-opening) conformation. On
the other hand, neither has much activity at the
with 1a producing a maximal response of only 23% efficacy (rela-
tive to nicotine), and 1b a barely detectable response up to the
maximum concentration of 100 lM.
1. Jain, K. K. Curr. Opin. Inv. Drugs 2004, 5, 76.
2. Meyer, M. D. Drug Dev. Res. 2006, 67, 355.
3. Bunnelle, W. H.; Decker, M. W. Exp. Opin. Ther. Pat. 2003, 13, 1003.
4. Bitner, R. S.; Nikkel, A. L.; Curzon, P.; Donnelly-Roberts, D. L.; Puttfarcken, P. S.;
Namovic, M.; Jacobs, I. C.; Meyer, M. D.; Decker, M. W. Brain Res. 2000, 871, 66.
5. Marubio, L. M.; del Mar Arroyo-Jimenez, M.; Cordero-Erausquin, M.; Lena, C.; Le
Novere, N.; de Kerchove d’Exaerde, A.; Huchet, M.; Damaj, M. I.; Changeux, J.-P.
Nature 1999, 398, 8085.
6. Sullivan, J. P.; Bannon, A. W. CNS Drug Rev. 1996, 2, 21.
7. Spande, T. F.; Garraffo, H. M.; Edwards, M. W.; Yeh, H. J. C.; Pannell, L.; Daly, J.
W. J. Am. Chem. Soc. 1992, 114, 3475.
8. Bannon, A. W.; Decker, M. W.; Holladay, M. W.; Curzon, P.; Donnelly-Roberts,
D. L.; Puttfarcken, P. S.; Bitner, R. S.; Diaz, A.; Dickenson, A. H.; Porsolt, R. D.;
Williams, M.; Arneric, S. P. Science 1998, 279, 77.
9. Holladay, M. W.; Wasicak, J. T.; Lin, N.-H.; He, Y.; Ryther, K. B.; Bannon, A. W.;
Buckley, M. J.; Kim, D. J. B.; Decker, M. W.; Anderson, D. J.; Campbell, J. E.;
Kuntzweiler, T.; Donnelly-Roberts, D. L.; Piattoni-Kaplan, M.; Briggs, C. A.;
Williams, M.; Arneric, S. P. J. Med. Chem. 1998, 41, 407.
a
3b4ꢀ receptor,
Both 1a and 1b were tested in the mouse hot plate model for
analgesia. While 1a does exhibit an analgesic response at the high-
est dose tested (62
lmol/kg, ip), hypolocomotion and decreased
body temperature were observed at the same dose. The des-chloro
analog 1b showed no behavioral effects at 62
with the lower agonist activity relative to 1a.
lmol/kg, consistent
10. Bunnelle, W. H.; Daanen, J. F.; Ryther, K. B.; Schrimpf, M. R.; Dart, M. J.; Gelain,
A.; Meyer, M. D.; Frost, J. M.; Anderson, D. J.; Buckley, M. J.; Curzon, P.; Cao, Y.-
J.; Puttfarcken, P. S.; Searle, X.; Ji, J.; Putman, C. B.; Surowy, C. S.; Toma, L.;
Barlocco, D. J. Med. Chem. 2007, 50, 3627.
11. Ji, J.; Schrimpf, M. R.; Sippy, K. B.; Bunnelle, W. H.; Li, T.; Anderson, D. J.;
Faltynek, C.; Surowy, C. S.; Dyhring, T.; Ahring, P. K.; Meyer, M. D. J. Med. Chem.
2007, 50, 5493.
Pharmacophore models for the
a4b2 nAChR place importance
on the spacing and orientation between a basic (or quaternized)
amine and a hydrogen bond acceptor (heterocycle or carbonyl
group). For example, Figure 3 depicts the model developed by
12. The chiral amino acids were commercially available, but the stereogenic center
is racemized during the synthesis.
13. Culbertson, T. P.; Sanchez, J. P.; Gambino, L.; Sesnie, J. S. J. Med. Chem. 1990, 32,
2270.
Table 2
Calculated pharmacophore modeling values for bridged, fused and spiro diamines
14. Padwa, A.; Dent, W. Org. Synth. 1989, 67, 133.
Structure
a–b (Å)
a–c (Å)
\bac (°)
15. Anderson, D. J.; Arneric, S. P. Eur. J. Pharmacol. 1994, 253, 261.
16. Kuntzweiler, T. A.; Arneric, S. P.; Donnelly-Roberts, D. L. DrugDev. Res. 1998, 44, 14.
17. Balboni, G.; Marastoni, M.; Merighi, S.; Borea, P. A.; Tomatis, R. Eur. J. Med.
Chem. 2000, 35, 979.
18. Abreo, M. A.; Lin, N.-H.; Garvey, D. S.; Gunn, D. E.; Hettinger, A.-M.; Wasicak,
J. T.; Pavlik, P. A.; Martin, Y. C.; Donnelly-Roberts, D. L.; Anderson, D. J.;
Optimal range22
A (bridged)
B (fused)
7.3–8.0
7.4
7.6
6.5–7.4
6.5
6.5
30.4–35.8
34.97
34.09
C (spiro)
8.1
5.4
28.2

K. B. Sippy et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1682–1685
1685
Sullivan, J. P.; Williams, M.; Arneric, S. P.; Holladay, M. W. J. Med. Chem. 1996,
39, 817.
21. Tønder, J. E.; Olesen, P. H. Curr. Med. Chem. 2001, 8, 651.
22. Tønder, J. E.; Olesen, P. H.; Hansen, J. B.; Begtrup, M.; Pettersson, I. J. Comput.
Aided Mol. Des. 2001, 15, 247.
23. Ewig, C. S.; Berry, R.; Dinur, U.; Hill, J. R.; Hwang, M. J.; Li, H.; Liang, C.; Maple, J.;
Peng, Z.; Stockfisch, T. P.; Thacher, T. S.; Yan, L.; Ni, X.; Hagler, A. T. J. Comput.
Chem. 2001, 22, 1782.
19. Data for bridged compound
diazabicyclo[2.2.1]heptane taken from Ref. 10.
20. Data for fused compound is that of (1S,5S)-3-(3-pyridyl)-3,6-
diazabicyclo[3.2.0]heptane taken from Ref. 11.
A
is that of (1S,4S)-2-(3-pyridyl)-2,5-
B