(±)8-Amino-5,6,7,8-tetrahydroisoquinolines as novel antinociceptive agents

Article abstract of DOI:10.1016/j.bmcl.2004.05.020

Several amine-substituted 8-amino-5,6,7,8-tetrahydroisoquinolines were examined as conformationally-constrained analogs of the nicotinic cholinergic (nACh) 3-(aminomethyl)pyridines. Although these ligands failed to bind at nACh receptors, the N-ethyl-N-methyl analog 3d was found to be at least equipotent with nicotine in rodent tests of antinociception. The mechanism of action of 3d is currently unknown.

Full text of DOI:10.1016/j.bmcl.2004.05.020

Bioorganic & Medicinal Chemistry Letters 14 (2004) 3651–3654
(%)8-Amino-5,6,7,8-tetrahydroisoquinolines as novel
antinociceptive agents
Małgorzata Dukat,a Mohamed Taroua,a Abdelaziz Dahdouh,a
Umamaheswar Siripurapu,a M. Imad Damaj,b Billy R. Martinb and Richard A. Glennona,*
aDepartment of Medicinal Chemistry, School of Pharmacy, Virginia Commonwealth University, Richmond, VA 23298, USA
bDepartment of Pharmacology and Toxicology, School of Medicine, Virginia Commonwealth University, Richmond, VA 23298, USA
Received 30 April 2004; revised 12 May 2004; accepted 12 May 2004
Available online
Abstract—Several amine-substituted 8-amino-5,6,7,8-tetrahydroisoquinolines were examined as conformationally-constrained
analogs of the nicotinic cholinergic (nACh) 3-(aminomethyl)pyridines. Although these ligands failed to bind at nACh receptors, the
N-ethyl-N-methyl analog 3d was found to be at least equipotent with nicotine in rodent tests of antinociception. The mechanism of
action of 3d is currently unknown.
Ó 2004 Elsevier Ltd. All rights reserved.
Effective pain management is one of today’s most seri-
ous unmet problems. Agents that bind at nicotinic
acetylcholinergic (nACh) receptors offer a relatively
novel and unexploited mechanistic approach to the
treatment of pain. In the course of our studies we have
investigated a variety of nACh receptor ligands includ-
ing conformationally-constrained ligands (reviewed1;2).
For example, N-substituted 3-aminomethylpyridines
such as 1 retain affinity for a4b2 nACh receptors and
possess antinociceptive properties.3 Although substitu-
tion at the pyridine 4-position has a tendency to reduce
nACh receptor affinity,3 compounds such as 2, which
might be viewed as conformationally-constrained ana-
logs of 1, bind at nACh receptors and display antino-
ciceptive character.4
Because the most pharmacologically relevant confor-
mation of nicotinic agents remains to be determined (but
see Ref. 5), we have explored other analogs of the
aminomethylpyridines where the side chain has been
constrained. In the present investigation, we prepared a
series of 8-aminotetrahydroisoquinolines (i.e., 3) as
conformationally-constrained analogs of 1 in order to
determine their nACh receptor affinities and examine
their antinociceptive properties. In general, aryl-unsub-
stituted analogs of 1 display low affinity (Ki > 500 nM)
for a4b2 nACh receptors when the amine is a primary or
secondary amine, or a tertiary amine where R1 ¼ Me
and R2 ¼ n-Pr; conversely, affinity is optimal
(Ki 6 30 nM) when R1 ¼ Me and R2 ¼ Et, or where the
amine is a pyrrolidine ring.1;3 Similar amine-substituted
derivatives of 3 were prepared for comparison.
R
N
Aminotetrahydroisoquinolines prepared for this inves-
tigation (Scheme 1) are shown in Table 1. Oxidation of
5,6,7,8-tetrahydroisoquinoline (4) with KMnO4 pro-
vided the known6 5,6-dihydro-7H-isoquinolin-8-one (5),
which was used as a common synthetic intermediate.
Compound 5 was reductively aminated using sodium
cyanoborohydride and the appropriate amine (method
A) to afford target 3. In method B, a methanolic solution
of 5 was allowed to react with the MeNH2, under cat-
alytic hydrogenation conditions to reduce the interme-
diate imine, to afford 3b; 3b was acylated with AcCl and
the corresponding amide 6 was reduced to 3d with
R1
R1
N
N
R2
R2
N
N
2
N
3
1
Keywords: Analgesia; Aminoisoquinolines; Tail flick; Hot plate.
* Corresponding author. Tel.: +1-804-828-8487; fax: +1-804-828-7404;
e-mail: glennon@hsc.vcu.edu
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.05.020

3652
M. Dukat et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3651–3654
LiAlH4. An alternative role of synthesis (method C)
involved NaBH4 reduction of 5 to its corresponding
alcohol 7, reaction of 7 with SOCl2 to afford the chloro
compound 8, and alkylation of 8 with N-methyl-N-
propylamine to afford 3f. Where a target compound was
prepared by more than one method, thin-layer chro-
matography and [1H]NMR were used to support
structural identity.
CH3
a
N
H
O
c
B
N
N
N
3b
5
4
d
A
b
f
C
CH3
CH3
R1
N
N
OH
In general, none of the compounds displayed appreciable
affinity for a4b2 nACh receptors with the exception of 3c
(Ki ¼ 330 nM) and 3d (Ki ¼ 930 nM) (Table 2). Com-
pound 3c binds at a4b2 nACh receptors with an affinity
comparable to that of aminomethylpyridine 1 where
R1 ¼ R2 ¼ Me (Ki ¼ 540 nM).3 However, 3d possessed
>30-fold lower affinity than its ring-open analog (i.e., 1
R1 ¼ Me, R2 ¼ Et; Ki ¼ 28 nM), and 3g binds with >300-
fold lower affinity than its corresponding aminomethyl-
pyridine.3 On this basis it seems unlikely that structures 3
represent pharmacologically-active conformations of 1
for interaction with a4b2 nACh receptors.
R2
N
O
N
N
3
6
7
g
e
CH3
CH3
CH3
CH3
N
Cl
N
h
N
N
N
8
3d
3f
Scheme 1. Reagents and conditions: (a) KMnO4, (b) R1R2NH,
NaCNBH4, (c) MeNH2, H2, (d) AcCl, (e) LiAlH4, (f) NaBH4, (g)
SOCl2, (h) HNR1R2.
Even though compounds 3 lacked appreciable affinity
for a4b2 nACh receptors, several were examined for
Table 1. Physicochemical properties of the aminotetrahydroisoquinolines
R1
N
R2
N
R1
R2
Methoda
Mp (°C)
%Yield
Recryst. solvent
Empirical formulab
3a
3b
H
H
H
Me
A
B
C
A
C
B
C
A
A
C
158–160
193–197
189–193
123–126
203–205
143–146
145–148
175–178
98–100
55
47
60
17
45
38
70
8
MeOH
C9H12N2Æ1.5C2H2O4
C10H14N2Æ1.8C2H2O4
2-PrOH/EtOH
2-PrOH/EtOH
2-PrOH/MeOH
MeOH/Et2O
2-PrOH/MeOH
2-PrOH/EtOH
MeOH/Et2O
MeOH/Et2O
2-PrOH/EtOH
3c
3d
Me
Me
Me
Et
C11H16N2Æ2C2H2O4
C11H16N2Æ2HCl
C12H18N2Æ2C2H4O4
3e
3f
H
Me
n-Pr
n-Pr
C12H18N2Æ1.5C2H2O4
C13H20N2ÆC2H2O4Æ0.5MeOHc
C13H18N2Æ1.6C2H2O4
15
73
3g
CH2CH2CH2CH2
135–138
a See Scheme 1.
b Compounds analyzed within 0.4% of theory for C, H, N. Compounds 3b–d were prepared by more than one route in an attempt to improve yields
and explore the general applicability of the reaction scheme.
c Compound 3f (C13H20N2) also analyzed correctly for C, H, N as its free base (bp 90 °C/0.03 mmHg).
Table 2. Pharmacological properties of the aminotetrahydroisoquinolines 3
Ki (nM) (SEM)
Tail-flick assay (ED50, mg/kg)
(subcutaneous)a;b
Tail-flick assay (ED50, lg/mouse)
(intrathecal)a;c
Hot-plate assay (ED50, mg/kg)
(subcutaneous)a;b
3a
3b
>10,000
>10,000
330 (36)
930 (120)
>10,000
>10,000
>10,000
1.2 (0.4)
1%
0%
NT
13 (10–17)
8%
10%
2%
3c
3d
2%
3.9 (1.4–11)
0.2 (0.1–0.3)
NT
0.6 (0.1–1.1)
NT
NT
12 (9–17)
NT
NT
3e
3f
3g
NT
11%
2%
1.5 (0.8–2.2)
6%
12 (8–16)
Nicotine
0.5 (0.3–0.7)
a Assays were performed as previously reported.8 ED50 values are followed in parenthesis by (S.E.M.). NT ¼ not tested.
b Where antinociception was not observed, %MPE is reported for a 10 mg/kg dose.
c Where antinociception was not observed, %MPE is reported for a 20–25 lg dose.

M. Dukat et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3651–3654
3653
antinociceptive action in the mouse tail-flick assay (Ta-
ble 2). Interestingly, compound 3d was found to be twice
as potent as (ꢀ)-nicotine. The antinociceptive actions of
3d following subcutaneous administration were evident
for 120 min, relative to 30 min for nicotine (data not
shown). In addition, the antinociceptive effects, unlike
those of nicotine, were not antagonized by the non-
competitive nACh antagonist mecamylamine (1 mg/kg)
nor the competitive antagonist dihydro-b-erythroidine
(1 mg/kg) (data not shown). Similar results were ob-
tained in the mouse hot-plate assay. Administered via
the intrathecal route, 3d was equipotent with (ꢀ)-nico-
tine in the tail-flick assay. In contrast, compounds
3a and 3b were inactive both in the tail-flick and
hot-plate assays, whereas 3b was nearly equipotent
with 3d and nicotine when administered via the
intrathecal route. Compound 3c was inactive in the
tail-flick assay and nearly 20 times less potent than 3d in
the hot-plate assay. In summary, several of the
compounds displayed antinociceptive action, but only
3d was active under all three assay conditions––and
was at least equipotent with (ꢀ)-nicotine in these
assays.
N
CH3
N
H
N
H3CO
N
9
10
In summary, with respect to interaction at a4b2 nACh
receptors, it is quite doubtful that the aminotetrahy-
droisoquinolines 3, represent pharmacologically impor-
tant conformations of 1 due to their low affinity.
Nevertheless, compound 3d in particular displayed an-
tinociceptive actions both in the tail-flick and hot-plate
assays and, as such, might represent an interesting
structural and mechanistic template for further devel-
opment of novel analgesic agents.
Acknowledgements
This work was supported, in part, by funding from DA
05274 and the British Technology Group (BTG). The
authors are also grateful to Drs. Ken Kellar and
Yingxian Xiao (Georgetown University) for providing
us with a radioligand binding profile for 3d at the cloned
nACh receptor subtypes.
With its low affinity for a4b2 nACh receptors, and the
inability to antagonize its effects with nACh antagonists,
it is difficult to reconcile the actions of 3d with a nico-
tinic receptor mechanism unless there is involvement of
a nACh receptor subtype other than a4b2 receptors.
Compound 3d was evaluated at several such subtypes
and found to show little affinity: a2b2 (Ki ¼
10; 200 ꢁ 1700 nM), a2b4 (Ki ¼ 49; 900 ꢁ 4800 nM),
a3b2 (Ki ¼ 4410 ꢁ 730 nM), a3b4 (Ki ¼ 46; 900 ꢁ
3200 nM), a4b2 (Ki ¼ 5860 ꢁ 880 nM), a4b4 (Ki ¼
40; 000 ꢁ 1600 nM), a6b4 (Ki ¼ 49; 400 ꢁ 4100 nM), a7
(Ki ¼ 10; 500 ꢁ 200 nM). Of course, the possibility exists
that 3d might still be acting via a subtype that was not
specifically examined. However, because a4b2 receptors
represent a major population of nACh receptors in
mammalian brain, it seems unlikely that 3d produces its
antinociceptive effects via such a mechanism. Com-
pound 3d was also examined at 75 other neurotrans-
mitter receptors (CEREP) and displayed remarkably
low affinity for each; for example: Ki > 10; 000 nM for
hA1, hA2, and hA3 adenosine receptors, a1-, a2-, hb1-,
hb2-adrenergic receptors, central and peripheral BZ
receptors, hD1–D5 dopamine receptors, H1 and H2 his-
tamine receptors, hm1–m5 muscarinic cholinergic
References and notes
1. Glennon, R. A.; Dukat, M.; Liao, L. Curr. Top. Med.
Chem. 2004, 4, 631–644.
2. Glennon, R. A. Prog. Med. Chem. 2004, 42, 55–123.
3. Glennon, R. A.; Fiedler, W.; Dukat, M.; Damaj, I.; Martin,
B. Eur. J. Med. Chem. 1996, 31, 875–888.
4. Cheng, Y.-X.; Dukat, M.; Dowd, M.; Fiedler, W.; Martin,
B.; Damaj, M. I.; Glennon, R. A. Eur. J. Med. Chem. 1999,
34, 177–190.
5. Ullrich, T.; Krich, S.; Binder, D.; Mereiter, K.; Anderson,
D. J.; Meyer, M. D.; Pyerin, M. J. J. Med. Chem. 2002, 45,
4047–4054.
6. Glassco, W.; Suchocki, J.; George, C.; Martin, B. R.; May,
E. L. J. Med. Chem. 1993, 36, 3381–3385.
7. Vernier, J.-M.; Holsenback, H.; Cosford, N. D. P.; Whit-
ten, J. P.; Menzaghi, F.; Reid, R.; Rao, T. S.; Sacaan, A. I.;
Lloyd, K.; Suto, C. M.; Chavez-Noriega, L. E.; Washburn,
M. S.; Urrutia, A.; McDonald, I. A. Bioorg. Med. Chem.
Lett. 1998, 8, 2173–2178.
receptors, l-, d-, and j-opioid receptors, h5-HT1A
,
5-HT1B, 5-HT2B, 5-HT2C, 5-HT3, 5-HT5A, 5-HT6, and 5-
HT7 serotonin receptors, nor did it bind at the norepi-
nephrine or dopamine transporter, or to Kþ, Naþ,
Caþþ, or Clꢀ ion channels.
8. Assays were performed as previously detailed in: Dukat,
M.; Damaj, I. M.; Young, R.; Vann, R.; Collins, A. C.;
Marks, M. J.; Martin, B. R.; Glennon, R. A. Eur. J.
Pharmacol. 2002, 435, 171–180. In brief, the binding assay
was conducted using rat brain (minus cerebellum) homo-
genates and [3H](ꢀ)-nicotine as radioligand. IC50 values
were determined from a plot of the log concentration versus
percent displacement and converted to Ki values (at least in
triplicate). The tail-flick assay employed male ICR mice
(20–25 g; Harlan Laboratories, Indianapolis, IN) that were
housed in an AALAC approved facility in groups of six
and had free access to food and water. The study was
approved by the Institutional Animal Care and Use Com-
mittee of VCU. The antinociceptive response was calcu-
lated as percent maximum possible effect (%MPE) where
Compounds 96 and 107 are other examples of previously
reported conformationally-constrained nicotine analogs.
Both lack significant affinity for a4b2 nACh receptors
yet possess antinociceptive character. Whereas the an-
tinociceptive action of 10 was attenuated by mecamyl-
amine,7 that of 9,6 like that of 3d, was not. The
aminotetrahydroisoquinoline moiety common to 3d and
9 and might be responsible for their antinociceptive
properties.

3654
M. Dukat et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3651–3654
%MPE ¼ [(test-control)/(10-control)] ꢂ 100. Groups of 6–
cases was 5 lL. In the hot-plate test, mice were placed into a
10 cm wide glass cylinder on a hot plate (Thermojust
Apparatus) maintained at 55 °C, and the antinociceptive
response was calculated as %MPE. Groups of 6–12 mice
were used for each dose, and mice were tested 5 min after
subcutaneous injections.
12 mice were used for each dose, and mice were tested 5 min
after either subcutaneous or intrathecal injections. Intra-
thecal injections were performed free-hand between the L5
and L6 lumbar space in unanesthetized mice according to
the method of Hylden and Wilcox9 using a 30-gauge needle
attached to a glass microsyringe. The injection volume in all
9. Hyldon, J. L.; Wilcox, G. L. Eur. J. Pharmacol. 1980, 67, 313.