4-[(8-Alkyl-8-azabicyclo[3.2.1]octyl-3-yl)-3-arylanilino]-N,N-diethylbenz amides: High affinity, selective ligands for the delta opioid receptor illustrate factors important to antagonist activity

Article abstract of DOI:10.1016/S0960-894X(00)00209-2

The tropane derived compounds, 4-[(8-alkyl-8-azabicyclo[3.2.1]octyl-3-yl)-3-arylanilino]-N,N-diethylbenzamides (5a-d), were synthesized and found to have high affinity and selectivity for the δ receptor. Compounds 5a-d are structurally similar to the full agonist (-)-RTI-5989-54 (3); yet, efficacy studies for compounds in this series (5a-d) reveal greatly diminished agonist activity as well as antagonism not found in piperidine-based compounds like 3. (C) 2000 Elsevier Science Ltd. All rights reserved.

Full text of DOI:10.1016/S0960-894X(00)00209-2

Bioorganic & Medicinal Chemistry Letters 10 (2000) 1281±1284
4-[(8-Alkyl-8-azabicyclo[3.2.1]octyl-3-yl)-3-arylanilino]-
N,N-diethylbenzamides: High Anity, Selective Ligands for
the Delta Opioid Receptor Illustrate Factors Important to
Antagonist Activity
James B. Thomas, a Robert N. Atkinson, a Richard B. Rothman, b Jason P. Burgess, a
S. Wayne Mascarella, a Christina M. Dersch, b Heng Xu b and F. Ivy Carroll a,*
aChemistry and Life Sciences, Research Triangle Institute, Research Triangle Park, NC 27709, USA
bClinical Psychopharmacology Section, NIDA Addiction Research Section, PO Box 5180, Building C,
5500 Nathan Shock Dr., Baltimore, MD 21224, USA
Received 1 March 2000; accepted 31 March 2000
AbstractÐThe tropane derived compounds, 4-[(8-alkyl-8-azabicyclo[3.2.1]octyl-3-yl)-3-arylanilino]-N,N-diethylbenzamides (5a±d),
were synthesized and found to have high anity and selectivity for the d receptor. Compounds 5a±d are structurally similar to the
full agonist ( )-RTI-5989-54 (3); yet, ecacy studies for compounds in this series (5a±d) reveal greatly diminished agonist activity
as well as antagonism not found in piperidine-based compounds like 3. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
Relative to the piperazine analogues 1 or 2, the piper-
idine compound 3 is more conformationally ¯exible.
The reports of the antagonistic activity of the more
conformationally rigid 4 suggested that at least part of
this e€ect could result from a decrease in conforma-
tional ¯exibility of 4, and thus, antagonist activity might
be elicited from piperidine compounds similar to 3 if the
number of conformations available to 3 were restricted.
To test this hypothesis, the tropane (bridged piperidine)
derived compounds 5a±d were prepared and tested in
opioid binding and functional assays. In this communica-
tion we report that compounds of this nature demonstrate
high degrees of d selectivity and binding anity as well as
antagonist activity with potency modulated by the N-
substituent.
Since the discovery of the nonpeptide d opioid receptor
selective compounds, BW373U86 (1)1 and SNC-80 (2),2
a great variety of new opioid ligands3,4 have appeared
utilizing the diethylbenzamide substructure to in¯uence d
selectivity. This unique d address moiety has been incor-
porated into both classical and nonclassical opioid
ligands5 in a search for analgesics possessing a reduced
side-e€ect pro®le relative to the m opioid analgesic mor-
phine. Several groups, including our own,3,6,7 have stu-
died the e€ect on overall ligand activity produced by
transposition of the internal nitrogen atom in com-
pound 1 with the benzylic carbon as represented by
compound 3. These compounds were found to be d
receptor selective full agonists with the potency being
directly a€ected by the 3-methyl group as well as its cis
relative disposition to the 4-diaryl system. Recent
reports of antagonist activity in compound 4 by Su et
al.8 demonstrated that alteration of the placement of the
methyl groups in 1, and consequently, the conforma-
tional disposition of the piperazine ring could convert
agonists into antagonists.
Chemistry
Preparation of 5a±d (Scheme 1) began with reductive
amination of 3-tropanone (6) with 3-methoxyaniline
using titanium (IV) isopropoxide9 which gave 7 as a sin-
gle epimer (endo based on the observations of Abdel-
Magid et al.)10 in 38% yield. This intermediate was then
coupled to the butylated hydroxyanisole (BHA) ester of
4-¯uorobenzoic acid to give 8 in 83% yield.11 Removal of
the BHA group was accomplished by trans-esteri®cation
*Corresponding author. Tel.: +1-919±541-6679; fax: +1-919±541-
8868; e-mail: ®c@rti.org
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00209-2

1282
J. B. Thomas et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1281±1284
binding assays following previously reported proce-
dures6,7 (Table 1). Determination of agonism was
accomplished by measuring the stimulation of binding of
the GTP analogue [35S]GTP-g-S in guinea pig caudate
elicited by the test compounds (Table 2).6,7 The receptor
responsible for any observed agonist activity was deter-
mined by measuring stimulation in the presence of
selective antagonists. Measures of antagonism for the
three opioid receptors were obtained by monitoring the
test compounds ability to inhibit stimulation produced
by the selective agonists DAMGO (m), SNC-80 (d), and
U69,593 (k) (Table 3).6,7
Results and Discussion
The radioligand binding data for the compounds 5a±d
along with comparative data for SNC-80 (2) and the
piperidine derivative 3 are shown in Table 1. As is
apparent, the tropane analogues 5a±d all display excel-
lent anity for the d opioid receptor versus the m or k
receptors and possess slightly higher anity for d relative
to the piperidine 3 but less anity than the piperazine 2.
This translates into higher d versus m selectivity for all
the tropane derivatives except 5c relative to the piper-
idine derivative 3. Moreover, unlike the piperidine com-
pounds in general, the selectivity for d over m receptors
evident in the tropane analogues, 5a and 5b, rivals that
found in SNC-80 (2). Compound 5d demonstrates that
analogues possessing an aryl hydroxy substituent pro-
duce higher anity and lower selectivity relative to the
corresponding methyl ether derivatives like 5a. Similar
behavior was found for transitions between 1 and 2;
however, in this series, the methyl ether compounds are
generally more selective.2
Chart 1.
with re¯uxing sodium methoxide in toluene/N-methyl-
pyrrolidinone followed by saponi®cation of the methyl
ester. The zwitterionic intermediate was isolated as a
slurry of its HCl salt and converted without puri®cation
into the diethylamide using benzotriazol-1-yl-oxy-tris-
(dimethylamino) phosphonium hexa¯uorophosphate
(BOP a.k.a. Castro's reagent), diethylamine, and triethyl-
amine in a tetrahydrofuran (THF) slurry to give 9 in
78% yield. Conversion to the appropriately N-substituted
analogue was accomplished by treating 9 with phenyl
chloroformate followed by hydrolysis of the resulting
carbamate with potassium hydroxide in isopropyl alco-
hol. N-Alkylation with the appropriate alkyl bromide
then gave 5a±c. Compound 5d was prepared from 5a in
77% yield by treatment with boron tribromide.
In terms of d versus k selectivity, the tropane compounds
appear to possess values quite similar to those obtained
for the piperidine compound 3 with the exception of the
hydroxy substituted compound 5d. Indeed, comparison
of the selectivity observed for this compound to that of
its methoxy analogue 5a illustrates that a 5.5-fold loss in
d selectivity results from removal of the methyl in com-
pound 5a. Inspection of the binding data for the k
Biological
The binding anities of 5a±d and the reference com-
pounds SNC-80 (2) and ( )-RTI-5989-54 (3) for the m, d,
and k opioid receptors were determined using competitive
Scheme 1. (a) Ti(Oi-Pr)4, 3-methoxyaniline; (b) NaBH4, EtOH; (c) n-BuLi, THF, HMPA then 1-(2,6-di-tert-butyl-4-methoxy)-4-¯uorobenzoate; (d)
N-methylpyrrolidinone, NaOCH3, toluene; (e) EtOH, H2O; (f) Et2NH, BOP, Et3N (g) PhOCOCl; (h) KOH, i-PrOH, H2O; (i) R-Br, EtOH, K2CO3;
( j) BBr3.

J. B. Thomas et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1281±1284
Table 1. Radioligand binding results at the m, d, and k opioid receptorsa
1283
Ki (nM Æ S.D.)
d [3H]DADLEc
Compound
m [3H]DAMGOb
k[3H]U69,593d
m/d
k/d
2, SNC-80
3, ( )-RTI5989-54
5a
5b
5c
5d
1614 Æ 131
2623 Æ 307
2547 Æ 344
>3400
1.57 Æ 0.19
5.85 Æ 0.31
3.28 Æ 0.31
3.7 Æ 0.16
3.51 Æ 0.32
1.26 Æ 0.08
3535 Æ 1841
1448 Æ 196
1076 Æ 84
1166 Æ 99
788 Æ 45
1028
448
776
>918
315
544
2251
247
328
355
224
58
1105 Æ 114
686 Æ 94
73 Æ 7.7
aMu and d binding was performed using rat brain membranes, and k assays were performed using guinea pig brain membranes.
b[3H]DAMGO [(D-Ala2,MePhe4,Gly-ol5)enkephalin]. Tritiated ligand selective for m opioid receptor.
c[3H]DADLE [(D-Ala2,D-Leu5)enkephalin]. Tritiated ligand selective for d opioid receptor.
d[3H]U69,593 {[3H](5a,7a,8b)-( )-N-methyl-N-[7-(1-pyrrolidinyl)-1-oxaspiro[4,5]dec-8-yl]benzeneacetamide}. Tritiated ligand selective for k opioid
receptor.
Table 2. % Stimulation of [35S]GTP-g-S binding by DAMGO, SNC-80, U69,593, and compounds 5a±d (10 mM) in guinea pig caudate membranes
Unblocked condition%
stimulation (mean Æ S.D.)
m Condition%
d Condition%
k Condition%
stimulation (mean Æ S.D.)a
stimulation (mean Æ S.D.)b
stimulation (mean Æ S.D.)c
DAMGO
SNC-80 (2)
U69,593
5a
5b
5c
5d
115 Æ 10
102 Æ 7
148 Æ 7
46 Æ 3
37 Æ 8
29 Æ 5
79 Æ 9
90 Æ 10
8.6 Æ 1.2
105 Æ 2
30 Æ 3
6 Æ 5
1.9 Æ 1.7
4 Æ 3
34 Æ 5
103 Æ 25
0
41 Æ 4
0
0
0
0
0
33 Æ 2
28 Æ 13
44 Æ 2
40 Æ 3
47 Æ 18
aThe m-selective condition used 20 nM naltrindole (NTI, antagonist selective for d opioid receptor) and 6 nM nor-binaltorphimine (nor-BNI,
antagonist selective for k opioid receptor).
bThe d-selective condition used 6000 nM CTAP (antagonist selective for m opioid receptor) and 6 nM nor-BNI.
cThe k-selective condition used 6000 nM CTAP and 20 nM NTI.
Table 3. % Inhibition of DAMGO, SNC-80, and U69,593 (10 mM)-stimulated [35S]GTP-g-S binding by compounds 5a±d (10 mM) in guinea pig
caudate membranes
Compound
% Inhibition of DAMGO-stimulated
35S]GTP-g-S binding
% Inhibition of SNC-80-stimulated
35S]GTP-g-S binding
% Inhibition of U69,593-stimulated
35S]GTP-g-S binding
[
[
[
5a
5b
5c
5d
0
25 Æ 1.1
49 Æ 0.4
69 Æ 1
0
0
25 Æ 3
0
0
23 Æ 0.6
0
34 Æ 6
receptor reveals that this change in selectivity results
primarily from a 15-fold increase in k anity for 5d
versus 5a. As mentioned previously, this is not unex-
pected since the methyl ethers in both the piperidine and
piperazine series, like 3 and 2, respectively, show greater
selectivity as the ether compared with the phenol.
contrast to the piperidine-based compound 3 which
demonstrated full agonist activity in similar assays with
an observed emax equivalent to that found for 2.7 Thus,
compared with similarly functionalized piperidine-based
ligands, the more rigid tropane analogues demonstrate
dramatically lower ability to stimulate GTP-g-S binding.
Inspection of Table 2 reveals that the tropane analogues
5a±c all exhibit selective d opioid stimulation of GTP
binding but that this stimulation diminishes as the size of
the N-substituent increased. Thus, the N-allyl analogue 5a
has 41% stimulation, whereas the N-prenyl analogue 5c
shows only 28% stimulation. Surprisingly, compound 5d,
which is highly selective for the d receptor in the binding
assay, produces stimulation for all opioid receptors in
the GTP-g-S functional assay (Table 2). Thus, changing
a methoxy substituent (5a) to a hydroxy substituent (5d)
dramatically impacts the behavior of 5d such that sti-
mulation is now observed for all opioid receptors. The
partial agonist activity for compounds 5a±d stands in
In Table 3, the data listed show the antagonist activity
of the tropane analogues 5a±d. Across the series of
methoxy substituted compounds 5a±c, the antagonist
activity is observed to increase with the size of the N-
substituent. As noted above, it was across the same series
(5a±c) that stimulation of GTP binding was found to
decrease. Thus, taken together, the trend in the data
clearly suggests that further manipulation of the N-sub-
stituent could lead to more potent d antagonists with
diminished agonist activity. The aryl hydroxy substituted
compound 5d demonstrates an even greater degree of
antagonist activity than the similarly substituted meth-
oxy derivative 5a. This observation is in line with

1284
J. B. Thomas et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1281±1284
expectations since typical opioid antagonists display
greater activity with hydroxyl groups relative to methoxy
groups.12 The result from the present study suggests that
further modi®cation of N-substituents in the hydroxy
series will lead to antagonists of greater activity than
found in the methoxy series. For comparison, compound
3 does not inhibit SNC-80 stimulated GTP binding at a
concentration of 10 mM.6,7 Overall, the data available
from this study reveal that reorganization of ring alkyl
substituents and limiting conformational ¯exibility can
minimize agonist activity and promote antagonist
activity. Additional studies will have to be performed in
order to determine what direct e€ect, if any, that the
bridging ring may induce beyond increasing system rigid-
ity. Especially critical for future studies will be determi-
nation of favored system conformations for the tropanes
relative to the piperidines.
Acknowledgements
This research was supported by the National Institute
on Drug Abuse, Grant DA09045.
References
1. Chang, K. J.; Rigdon, G. C.; Howard, J. L.; McNutt, R. W.
J. Pharm. Exp. Ther. 1993, 267, 852.
2. Calderon, S. N.; Rothman, R. B.; Porreca, F.; Flippen-
Anderson, J. L.; McNutt, R. W.; Xu, H.; Smith, L. E.; Bilsky,
E. J.; Davis, P.; Rice, K. C. J. Med. Chem. 1994, 37, 2125.
3. Dondio, G.; Ronzoni, S.; Petrillo, P. Exp. Opin. Ther. Pat.
1999, 9, 353.
4. Zhang, X.; Rice, K. C.; Calderon, S. N.; Kayakiri, H.;
Smith, L.; Coop, A.; Jacobson, A. E.; Rothman, R. B.;
Davis, P.; Dersch, C. M.; Porreca, F. J. Med. Chem. 1999,
42, 5455.
5. Dondio, G.; Ronzoni, S.; Eggleston, D. S.; Artico, M.;
Petrillo, P.; Petrone, G.; Visentin, L.; Farina, C.; Vecchietti,
V.; Clarke, G. D. J. Med. Chem. 1997, 40, 3192.
6. Thomas, J. B.; Herault, X. M.; Rothman, R. B.; Burgess, J.
P.; Mascarella, S. W.; Xu, H.; Horel, R. B.; Dersch, C. M.;
Carroll, F. I. Bioorg. Med. Chem. Lett. 1999, 9, 3053.
7. Thomas, J. B.; Atkinson, R. N.; Herault, X. M.; Rothman,
R. B.; Mascarella, S. W.; Dersch, C. M.; Xu, H.; Horel, R. B.;
Carroll, F. I. Bioorg. Med. Chem. Lett. 1999, 9, 3347.
8. Su, Y.-F.; McNutt, R. W.; Chang, K.-J. J. Pharmacol. Exp.
Ther. 1998, 287, 1.
Conclusion
We have demonstrated that 4-[(8-alkyl-8-azabicyclo
[3.2.1]octyl-3-yl)-3-arylanilino]-N,N-diethylbenzamides
5a±d display high anity binding and selectivity for the
d opioid receptor. Furthermore, the data obtained from
this series shows that antagonist activity can be realized
in the nitrogen-transposed compounds like 3, if the 3-
methyl group is removed and the system made more
rigid by the addition of a bridging ring between the 2
and 6 positions. Accordingly, the hypothesis that the
antagonism observed for 4 could result from limitations to
the conformational ¯exibility found in compound 1 is
supported. Further studies aimed at better understanding
the conformational preferences of compounds 5a±d are
currently underway and will be reported in due course.
9. Mattson, R. J.; Pham, K. M.; Leuck, D. J.; Cowen, K. A. J.
Org. Chem. 1990, 55, 2552.
10. Abdel-Magid, A. F.; Carson, K. G.; Harris, B. D.; Mar-
yano€, C. A.; Shah, R. D. J. Org. Chem. 1996, 61, 3849.
11. Hattori, T.; Satoh, T.; Miyano, S. Synthesis 1996, 514
12. Aldrich, J. V. In Burger's Medicinal Chemistry and Drug
Discovery; Wol€, M. E., Ed.; John Wiley & Sons: New York,
1996; Vol. 3.