Synthesis and opioid receptor affinity of a series of 2,4-diaryl-substituted 3,7-diazabicylononanones

Article abstract of DOI:10.1021/jm0009484

3,7-Diazabicyclo[3.3.1]nonan-9-ones having aryl rings in positions 2 and 4 with systematically varied substituents were synthesized using a double Mannich procedure. Radioligand binding assays were performed to measure the affinity of the compounds to the μ-, δ-, and κ-opioid receptors. The affinity of all 2,4-diphenyl-substituted 3,7-diazabicyclo[3.3.1]nonan-9-ones to the μ- and δ-receptors was found to be low. In contrast, with exception of the nitro- and cyanophenyl-substituted compounds, most of the diazabicycles showed considerable affinity for the κ-receptor. In particular, the m-fluoro-, p-methoxy-, and m-hydroxy-substituted compounds have an affinity in the submicromolar range. Due to solubility problems in aqueous media, salts of HZ2 were synthesized. The methiodide shows high κ-affinity and may, thus, be a promising candidate for development of a peripheral κ-agonist, e.g. for use in the case of rheumatoid arthritis.

Full text of DOI:10.1021/jm0009484

3746
J . Med. Chem. 2000, 43, 3746-3751
Syn th esis a n d Op ioid Recep tor Affin ity of a Ser ies of 2,4-Dia r yl-Su bstitu ted
3,7-Dia za bicylon on a n on es
Tom Siener,† Antonella Cambareri,† Ulrich Kuhl,† Werner Englberger,‡ Michael Haurand,‡ Babette Ko¨gel,‡ and
Ulrike Holzgrabe*,†
Institute of Pharmacy and Food Chemistry, University of Wu¨rzburg, Am Hubland, 97074 Wu¨rzburg, Germany, and
Department of Pharmacology, Gru¨nenthal GmbH, Zieglerstrasse 6, 52078 Aachen, Germany
Received March 22, 2000
3,7-Diazabicyclo[3.3.1]nonan-9-ones having aryl rings in positions 2 and 4 with systematically
varied substituents were synthesized using a double Mannich procedure. Radioligand binding
assays were performed to measure the affinity of the compounds to the µ-, δ-, and κ-opioid
receptors. The affinity of all 2,4-diphenyl-substituted 3,7-diazabicyclo[3.3.1]nonan-9-ones to
the µ- and δ-receptors was found to be low. In contrast, with exception of the nitro- and
cyanophenyl-substituted compounds, most of the diazabicycles showed considerable affinity
for the κ-receptor. In particular, the m-fluoro-, p-methoxy-, and m-hydroxy-substituted
compounds have an affinity in the submicromolar range. Due to solubility problems in aqueous
media, salts of HZ2 were synthesized. The methiodide shows high κ-affinity and may, thus, be
a promising candidate for development of a peripheral κ-agonist, e.g. for use in the case of
rheumatoid arthritis.
Ch a r t 1. Structures of Asimadoline, U-69,593, and
TAN-67
In tr od u ction
The search for compounds which can effectively treat
strong pain, and not show side effects associated with
the use of morphine, remains a challenge in drug
development. Besides a wide variety of emerging targets
connected with analgesia,1-3 the three major subtypes
of the opioid receptors, µ, κ, and δ, are still an important
focus of efforts identifying subtype-selective ligands.4
κ-Agonists were originally believed to be free of depen-
dence, tolerance, and respiratory depression.5 However,
the first compounds in clinical trials for postsurgical
pain, spiradoline and enadoline, have been abandoned
due to dose-limiting dysphoria.6 To avoid the side effects
associated with the CNS, peripherically acting κ-ago-
nists recently were targeted for use in inflammatory
hyperalgesia, such as asimadoline (Chart 1) which has
potential utility in treating rheumatoid arthritis.1 In the
case of the arylacetamide derivative U-69,593 (Chart
1), a peripheral component could be also detected in a
rat model of unilateral neuropathy.7 In addition, it was
reported that κ-agonists can downregulate the expres-
sion of the human immunodeficiency virus (HIV-1) in
human microglial cells.8 With respect to δ-agonists,
preclinical results of the highly active TAN-67 (Chart
1) showed that ligands of the δ-receptor may be effective
and safe analgesics.9
assess the influence of substitution on the 2,4-aryl rings
on the affinity for the three opioid receptor subtypes.
To enhance water solubility and to direct the compounds
to the peripheral receptors, HZ2 was methylated at the
nitrogen in position 7 and the product was used to form
various salts (Table 1).
Resu lts a n d Discu ssion
Recently, the 2,4-di-2-pyridine-substituted 3,7-di-
methyl-3,7-diaza-9-oxobicyclo[3.3.1]nonane-1,5-dicarboxylate,
HZ2, was reported to have selective κ-affinity10,11 com-
bined with strong antinociceptive activity comparable
to that of morphine and a potent action in inflammatory
and persistent pain.12 As a part of our efforts in opioid
ligand design, the purpose of the present study was to
Ch em istr y. To study structure-activity relationships
(SAR) in the series of 3,7-diazabicyclo[3.3.1]nonan-9-
ones, it was decided to systematically vary the substitu-
tion on the aryl rings in the 2- and 4-positions with
regard to lipophilicity and electronegativity. Using the
Craig plot,13 the following substitution pattern was
chosen: fluorine, chlorine, hydroxy, methoxy, nitro,
cyano, and methyl substituents on either position of the
phenyl ring (Table 1). In addition, the phenyl ring was
replaced with various pyridine, quinoline, and naph-
thalene rings.
* To whom correspondence should be addressed. Tel: 0931/8885460.
Fax: 0931/8885494. E-mail: holzgrab@pharmazie.uni-wuerzburg.de.
† University of Wu¨rzburg.
‡ Gru¨nenthal GmbH.
10.1021/jm0009484 CCC: $19.00 © 2000 American Chemical Society
Published on Web 09/13/2000

2,4-Diaryl-Substituted 3,7-Diazabicylononanones
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 20 3747
Ta ble 1. Structural Variations of the Diazabicyclo[3.3.1]nonan-9-ones
The 3,7-diazabicyclo[3.3.1]nonan-9-one skeleton could
be obtained by a double-step Mannich reaction. In
analogy to Petrenko-Kritschenko and Zoneff,14 the
condensation of 1 mol of dialkyl oxoglutarate, 1 mol of
methylamine, and 2 mol of the corresponding arylalde-
hyde led to the 2,6-diaryl-substituted γ-oxopiperidine-
3,5-dicarboxylates 1-9. With exception of the o- and
p-tolyl-, o-hydroxy-, and o-cyanophenyl-substituted pi-
peridinones, all other compounds could be obtained in
mostly good yields. Subsequently, the piperidin-4-ones
were converted with methylamine and formaldehyde in
methanol to give the 3,7-diaza-3-methylbicyclo[3.3.1]-
nonan-9-ones 11-19 (compare with ref 15). With the
exception of the o-methoxyphenyl-substituted compound
13c, which was not possible to synthesize, all bicyclics
were obtained in acceptable yields.
showed cis/trans isomerism with regard to the aryl
rings. After recrystallization from hot ethanol, all di-
azabicycles were converted to the cis configuration (for
details, see ref 16). Only in the case of the 2-quinolyl-
substituted compound 17a were the aryl substituents
found to have a trans arrangement to each other. In the
case of the p-nitrophenyl-substituted diazabicycles 5a ,e,
both the cis and trans isomers could be isolated.
Previous studies of representative diazabicyclic com-
pounds (15a ,b, 17a ,b, and 18a ,b) showed16 that the
compounds are configurationally stable under the course
of radioligand binding studies. In addition, the forma-
tion of rotational isomers was observed.16
To determine whether the diazabicycles were proto-
nated under physiological conditions, the pKa values
were representatively measured for HZ2, 11b, 12b, 13a ,
and 14b using a Sirius PCA 101 apparatus and a
Yasuda-Shedlovsky plot17 (Table 3). Due to solubility
NMR spectroscopic measurements of the first fraction
of crystals isolated from the reaction mixture often

3748 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 20
Siener et al.
Ta ble 2. Opioid Receptor Binding Data of the Diazabicyclononanones 11-20: Inhibition Constants (Ki) or Percent Inhibition of the
Compounds at 3H-Radiolabeled Opioid Receptors in Rat Brain Membranesa
compd
Ar
-phenyl
p-F-phenyl
m-F-phenyl
o-F-phenyl
µ
δ1
δ2
κ
HZ3
11a
11b
11c
12a
12bc
12e
13a
13bc
14a
14bc
14c
14d
14e
14.5% (1)a
10.1% (1)
5% (1)
-9.2% (1)
14.2% (1)
35.6% (1)
10.5% (1)
26.5% (1)
-16.2% (1)
2.2% (1)
-0.3% (10)
-1.6% (1)
0.4% (1)
3.2% (1)
1.5% (1)
7.3% (1)
Ki ) 0.21 µM
Ki ) 0.14 µMb
Ki ) 0.026 µMb
47.7% (10)b
8.4% (10)b
p-OH-phenyl
m-OH-phenyl
m-OH, HCl
p-OCH3-phenyl
m-OCH3-phenyl
p-Cl-phenyl
m-Cl-phenyl
o-Cl-phenyl
3,4-diCl-phenyl
p-Cl-phenyl
p-NO2-phenyl
p-NO2-phenyl
m-NO2-phenyl
o-NO2-phenyl
p-NO2-phenyl
p-CN-phenyl
m-CN-phenyl
2-quinolyl
27% (1)
Ki ) 0.02 µMd
92.7% (10)b
Ki ) 0.018 µMb
52.2% (10)b,e
63% (10)b
6.9% (1)
3% (1)
-1.6% (1)
-2.1% (1)
-11.8% (1)
1.9% (1)
-5.3% (1)
1.8% (1)
-3.1% (1)
-4% (1)
-7.9% (1)
-1.1% (1)
6.1% (1)
3.2% (1)
5% (1)
8.8% (1)
2.7% (1)
6.1% (1)
-5% (1)
4.6% (1)
4.8% (1)
-0.8% (10)
42.3% (10)b
65.4% (10)b
38.9% (10)b
Ki ) 1.98 µMb
22.6% (10)d
27.8% (10)d
10% (10)b
15a cis
15a trans
15bc
15c
-23.5% (1)
14% (1)
2.9% (1)
3.4% (10)b
15e
16a
16b
17a
17b
18a
18b
19a
2.3% (1)
5.6% (1)
9.3% (1)
5.9% (10)d
17.2% (10)b
35.5% (10)b
Ki ) 0.19 µMb
25.8% (10)b
7.8% (10)b
36.6% (1)
-7.1% (1)
-14.4% (1)
-23.1% (1)
30.7% (1)
3.1% (1)
11.8% (1)
1.2% (1)
1.8% (1)
-2.1% (1)
1.1% (1)
-9% (1)
4-quinolyl
1-naphthyl
2-naphthyl
3-pyridyl
4-pyridyl
2-pyridyl
18.6% (10)b,e
54.8% (10)b
12.5% (10)b
Ki ) 0.015 µMd
61.7% (10)b
62.6% (10)b
65.0% (10)b
96.4% (10)d
Ki ) 0.82 µMd
19b
HZ2
20% (1)
5% (10)
-6.6% (1)
HZ2 oxalate
HZ2 perchlorate
HZ2 methoiodide
20c
2-pyridyl
2-pyridyl
2-pyridyl
m-Me-phenyl
37.6% (1)
32.4% (1)
29.2% (1)
-5.9% (1)
-16.1% (1)
-5.9% (1)
3.3% (1)
3.4% (1)
a
b
d
The respective test concentration is indicated in parentheses as µmol/L. Human κ-opioid site ([3H]diprenorphine). c Ref 18. Rat
κ-opioid site ([3H]CI 977). e Solubility problems.
Ta ble 3. pKa Values
κ-receptor subtype ranged from nanomolar to micromo-
lar concentrations. However, the affinity of some com-
pounds was found to be too low to perform a full dose-
effect curve. In these cases, the percentage of inhibition
of the radioligand binding was given. Thus, qualitative
SAR will be discussed in the following section.
compd
pKa1
pKa2
pKa3
pKa4
HZ2
11b
12b
13a
14b
10.99 ( 0.10
8.05 ( 0.06 3.47 ( 0.07 2.03 ( 0.01
13.08 ( 0.09 10.72 ( 0.15
10.29 ( 0.24
9.05 ( 0.23
10.63 ( 1.16
7.97 ( 0.01 9.12 ( 0.10 9.21 ( 0.26
5.51 ( 0.58
9.51 ( 0.58
Comparing the different types of substituents on the
aryl ring in the 2- and 4-positions, it can be stated that
nitro and cyano groups on the phenyl rings, as well as
naphthyl rings, are disadvantageous to the affinity for
the κ-receptor. Moreover, the cis and trans configura-
tions of 15a do not influence the affinity for the
κ-receptor. With respect to the halogenated compounds,
a fluorine atom creates a much higher affinity to the
κ-receptor than a chlorine atom. Even the 3,4-dichloro
substitution, which can be often found in potent aryl-
acetamide derivatives, such as U-50488, DuP 747, and
BRL 52580,5 does not enhance the affinity for the
κ-receptor. However, even though slightly different test
systems were used, it can be stated that the m-
fluorophenyl compound 11b has as high an affinity for
the κ-receptor as does the parent compound HZ2.
Unfortunately, the poor water solubility of this com-
pound prevented the intravenous application route in
the in vivo experiments.
problems in water, the determination was performed in
a mixture of water/dioxane and an extrapolation to 0%
dioxane had to be applied (compare with ref 18). Two
pKa values were found: the first one in a range of 10-
12 and the second between 8 and 10. Thus, at the pH
occurring under physiological conditions (7.4), a notable
amount of the diazabicycles will be once and mostly
twice positively charged. In case of a double protonation
the two additional hydrogens are in direct proximity in
a chair/chair conformation and, therefore, the diazabi-
cycle may adopt a chair/boat conformation with a boat
in the less substituted piperidin-4-one ring.11
P h a r m a cology a n d SARs. To evaluate the affinity
to either opioid receptor subtype, all compounds were
subjected to radioligand binding assays using [3H]-
naloxone, [3H]Cl-DPDPE, [3H]-D-Ala-deltorphine, and
[3H]CI 977 or [3H]diprenorphine for µ, δ1, δ2, and κ (rat
or human κ-opioid site) receptor subtypes, respectively.
The Ki values and percent inhibition of the radioligand
binding are displayed in Table 2. None of the diazabi-
cyclic compounds showed an affinity for the δ1- or δ2-
receptors. The ability to displace the µ-radioligand
naloxone was also limited and was always surpassed
by the affinity to the κ-receptor. The affinity for the
Among the hydroxy- and methoxyphenyl-substituted
series 12 and 13, compounds of high affinity were found.
It is worth pointing out that the OH group in the meta
position (12b) and the OCH3 group (13a ) in the para
position create the highest affinities. Taking addition-
ally the m-fluorophenyl compound into account, no rule

2,4-Diaryl-Substituted 3,7-Diazabicylononanones
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 20 3749
Ta ble 4. Binding Methods
[3H]naloxone (1 nM)
[3H]Cl-DPDPE (0.25 nM)
[3H]-D-Ala-deltorphine (1 nM)
[3H]CI 977 (1 nM)
rat brain area
brain without
cerebellum
brain without cerebellum,
pons, and medulla
oblongata
brain without cerebellum,
pons, and medulla
oblongata
brain without cerebellum,
pons, medulla oblongata,
and cortex
assay volume
nonspecific binding
preincubation
2 mL
2 mL
1 mL
1 mL
naloxone, 10-5
37 °C, 40 min
M
naloxone, 10-3
25 °C, 30 min
M
naloxone, 10-5
none
M
naloxone, 10-4
37 °C, 45 min
M
incubation buffer
50 mM Tris-HCl,
pH 7.4
50 mM Tris-HCl, pH 7.4,
containing 900 µg BSA,a
45 µM PMSF,b and
4.5 mM MgCl2 per assay
25 °C, 240 min
50 mM Tris-HCl, pH 7.4,
containing 900 µg BSAa
and 4.5 mM MgCl2
per assay
50 mM Tris-HCl, pH 7.4
incubation
25 °C, 30 min
37 °C, 90 min
25 °C, 70 min
filter presoak
incubation buffer
incubation buffer
incubation buffer
incubation buffer
a
b
BSA ) bovine serum albumin. PMSF ) phenylmethanesulfonyl fluoride.
concerning the locus of substitution can be given from
the data set. However, for each type of substitution, the
affinity is sensitive to the position of the substituent.
This is also true for the pyridine-substituted compounds
19 and HZ2. HZ2 shows the highest affinity. The shift
of the nitrogen in the aryl ring to the 3-position results
in a slight loss, and the shift to the 4-position provides
an almost complete loss of affinity.
Exp er im en ta l Section
Syn th esis. Melting points were determined with a Dr.
Tottoli melting point apparatus (Bu¨chi) and were not corrected.
1H spectra were recorded on a Varian XL 300 (1H NMR:
299.956, 75 MHz) spectrometer. The centers of the peaks of
CDCl3 and DMSO-d6 were used as internal references. Ab-
breviations for data quoted are d, doublet; t, triplet; and m,
multiplet. Coupling constants are given in hertz (Hz). TLC was
performed using Merck silica gel F254 plates, 0.2 mm; eluent
mixture: cyclohexane/ethyl acetate/methanol (10/4/1). Chemi-
cals were of analytical grade and were purchased from Aldrich,
Steinheim, or Merck, Darmstadt, FRG. Dry solvents were used
throughout. The analytical results of almost all compounds
are within (0.4% of the theoretical values. The deviation
within the theoretical and calculated values for some com-
pounds such as 12e, 16a ,b, and 17a ,b are caused by the
presence of different solvents (molar ratios of DMSO, ethanol,
or methanol in the crystals).
Gen er a l P r oced u r e for Syn th esis of 2,6-Dia r yl-Su b-
stitu ted 3,5-Dia lk yl-N-m eth yl-4-p ip er id on e-3,5-d ica r box-
yla tes 1-9 (m od ified a fter p r oced u r es in r ef 19). Exactly
0.04 mol of the corresponding arylaldehyde and 1.5 mL of a
40% aqueous solution of methylamine were dissolved in 20 mL
of methanol and cooled to 0 °C. Over a course of about 1 h,
0.02 mol of a dialkyl oxoglutarate was added dropwise to the
mixture at 0 °C. The solution was allowed to stand overnight
at 5 °C. The product was obtained by filtration of the
precipitate formed. When no precipitate appeared, the solvent
was removed in vacuo at 40-50 °C, and the remaining oil was
dissolved in ethanol or treated with diethyl ether. The crystals
obtained could be washed with a mixture of ethanol/diethyl
ether and recrystallized from ethanol.
The 2-quinolyl compound 17a corresponding to HZ2
shows a reduced but considerable affinity for the
κ-receptor, whereas the 4-quinolyl-substituted diazabi-
cycle 17b corresponding to 19b is almost inactive.
However, a writhing test of compounds 17 and 19
resulted in the exhibition of some analgesic activity.
Taken together, within this series of compounds, three
diazabicycles were found to have a considerable affinity
for the κ-receptor which is comparable to that of HZ2.
Despite their nanomolar affinity for the κ-opioid recep-
tor, none of the compounds tested, however, exhibits as
high an antinociceptive potency as HZ2.12 In addition,
due to relatively poor solubility, not all compounds could
be tested by the intravenous route in the tail-flick or
writhing test in vivo. To improve the solubility of the
compounds, oxalates and perchlorates of HZ2 were
synthesized. These salts showed almost the same af-
finity for the κ-receptor and exerted a strong analgesic
activity as did HZ2 (i.e. 100% writhing inhibition at a
dose of 10 mg/kg iv).
Gen er a l P r oced u r e for Syn th esis of 2,4-Dia r yl-Su b-
stitu ted 1,5-Dia lk yl-3,7-d im eth yl-3,7-d ia za -9-oxobicyclo-
[3.3.1]n on a n e-1,5-d ica r boxyla tes 11-19 (m od ified a fter
p r oced u r es in r ef 19). Exactly 0.0025 mol of the piperidin-
4-one was dissolved in 20 mL of methanol by refluxing. To this
solution, was added 0.6 mL of a 35% aqueous solution of
HCHO, followed by 0.4 mL of a 40% aqueous solution of
methylamine. The mixture was refluxed for 7-8 min, and
afterward it was allowed to stand for 3 h at ambient temper-
ature. Three different workup procedures were possible. First,
the obtained precipitate was filtered and washed with a
mixture of ethanol/diethyl ether. When no precipitate appeared
within 24 h, the solvent was removed in vacuo at 40-50 °C,
and the remaining oil was dissolved in a small amount of
ethanol and diethyl ether. Crystals usually appeared within
2 days. When no crystals were obtained during this time, the
solvent was removed again, and the remaining oil was treated
with diethyl ether. The raw product can be recrystallized from
ethanol. The perchlorates and oxalates of HZ2 were formed
according to known procedures.20
To direct HZ2 to the peripheral κ-opioid receptors
only, HZ2 was permanently charged by formation of a
methiodide which carries the additional methyl group
on the nitrogen in position 7. This methiodide analogue
retains complete inhibition in the radioligand binding,
indicating high affinity and selectivity toward the
κ-opioid site. In addition, it exerts a strong antinoci-
ceptive activity in the tail-flick assay (ED50 6.7 mg/kg
iv) and writhing test (ED50 1.05 mg/kg iv) performed in
mice. Furthermore pharmacological investigations are
in progress in order to prove whether the site of action
for the methiodide compound is predominantly periph-
eral.
Taken together, the results of the SAR study revealed
some interesting compounds. The quaternization of the
nitrogen in position 7 offers the perspective to develop
peripheral κ-agonists which might be used against pain
of rheumatic and osteoarthritic origin (compare with ref
5). Due to the high stability, the fluorine-substituted
compound 11b seems to be a promising candidate for
the development of a peripheral κ-agonist.20
Syn th esis of th e Meth iod id e of HZ2. Exactly 1 mmol of
HZ2 was dissolved in 10 mL of tetrahydrofuran, and 16 mmol
methyliodide was added. The solution was allowed to stand
for 24 h at ambient temperature and was protected from light.
The so obtained crystals were separated, washed with tet-

3750 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 20
Siener et al.
rahydrofuran, and recrystallized from a mixture of methanol/
diethyl ether: yield 79%, mp 115 °C dec.
surface of the tail was measured. The heat source was adjusted
to produce a baseline tail-flick latency of 3-5 s prior to any of
the experiments and was left at a constant setting thereafter.
p Ka Va lu es. The pKa values were measured potentiometri-
cally in water using a Sirius PCA101 apparatus. Exactly
0.002-0.005 g of the substances was dissolved in 8-12 mL of
aqueous dioxane (50%) and diluted to 20.0 mL with a 0.15 M
KCl solution. The titration was performed starting from pH
11. Using the Yasuda-Shedlovsky plot,17 the pKa values could
be calculated and extrapolated to 0% dioxane.
P h a r m a cology. All animal testing was performed in ac-
cordance with the recommendations and policies of the Inter-
national Association for the Study of Pain and the national
law on the care of animals in experiments (German Animal
Welfare Law). All study protocols were approved by the local
government committee for animal research, which is also an
ethics committee.
Ack n ow led gm en t. Thanks are due to the DFG for
financial support (Ho-1368/3-2), Dr. C. Nachtsheim,
Bonn, FRG, for providing HZ2, Prof. Dr. M. Wiese,
University of Halle, for the opportunity to measure the
pKa values using the Sirius PCA101 apparatus, and
Mrs. I. Knoblauch for carefully measuring the NMR
spectra.
Su p p or tin g In for m a tion Ava ila ble: Analytical data
(molecular formula, yield, and melting point), elemental
analyses, and spectroscopic data (IR, 1H) of the intermediates
1-9 and final products 11-19 are available free of charge via
the Internet at http://pubs.acs.org.
Recep tor Bin d in g Stu d ies. The binding studies with
membranes from rat brains for the µ-, δ1/δ2-, and κ-opioid
receptor sites were done as essentially described by Brandt et
al.11 In brief, Wistar SPF rats, weighing about 200 g, were
killed by decapitation. Membrane suspensions were prepared
essentially as described by Wood et al.21 Protein concentration
was determined by the method of Lowry et al.,22 using bovine
serum albumin as the standard. Incubation was initiated by
addition of the membrane suspension. Details of the binding
studies for the rat membrane can be found in Table 3 (compare
with ref 12). All incubations were run in triplicate and
terminated by rapid filtration under mild vacuum (Brandel
cell harvester type M-24R) and three washes with 5 mL of ice-
cold buffer using FP-100 Whatman GF/B filter mats. The
radioactivity of the samples was counted, after a stabilization
and extraction period of at least 15 h, by use of Ready Protein
scintillation fluid (Beckman). The radioligands for µ-, δ1/δ2-,
and κ-opioid receptor sites (rat brain membranes) sites were
[3H]naloxone, [3H]Cl-DPDPE/[3H]-D-Ala-deltorphine, and [3H]-
CI 977, respectively. The receptor binding toward the human
κ-opioid receptor was done with membranes from CHO-K1
cells transfected with the human κ-opioid receptor (Receptor
Biology, Inc., Beltsville, MD). The membranes were thawed
rapidly, diluted 90-fold with 50 mM Tris-HCl pH 7.4 and
resuspended by homogenization. Threefold incubations were
carried out in 50 mM Tris-HCl pH 7.4 at 20 °C for 1.5 h in a
total volume of 250 µL, containing 200 µL of a membrane
suspension with 45 µg of protein and 2 nM [3H]diprenorphine
as the radioactive ligand. Nonspecific binding was defined in
the presence of 100 µM naloxone. Incubations were started
by the addition of the membrane suspension and terminated
by rapid filtration with a Brandel 96 sample harvester using
GF/B-UniFilterplates (Packard) presoaked with 50 µL of 50
mM Tris-HCl (pH 7.4/well), followed by washing three times
with 250 µL of 50 mM Tris-HCl (pH 7.4). After the filter plates
had dried at 50 °C for 1 h, 35 µL of Ultima Gold MV scintillant
(Packard) was added per well, and the radioactivity was
determined using a Wallac 1450 MicroBeta scintillation
counter.
Refer en ces
(1) Williams, M.; Kowaluk, E. A.; Arneric, S. P. Emerging Molecular
Approaches to Pain Therapy. J . Med. Chem. 1999, 42, 1481-
1500.
(2) Yaksh, T. L. Spinal systems and pain processing: development
of novel analgesic drugs with mechanistically defined models.
Trends Pharmacol. Sci. 1999, 20, 329-337.
(3) Eglen, R. M.; Hunter, J . C.; Dray, A. Ions in the fire: recent
ion-channel and approaches to pain therapy. Trends Pharmacol.
Sci. 1999, 20, 337-342.
(4) Holzgrabe, U.; Nachtsheim, C.; Siener, T.; Drosihn, S.; Brandt,
W. Opioid-Agonisten und -Antagonisten, Opioid-Rezeptoren.
Pharmazie 1997, 52, 4-22.
(5) Barber, A.; Gottschlich, R. Novel development with selective,
nonpeptidic kappa-opioid receptor agonists. Exp. Opin. Invest.
Drugs 1997, 7, 1351-1368.
(6) Kowaluk, E. A.; Arneric, S. P. Novel Molecular Approaches to
Analgesia. Annu. Rep. Med. Chem. 1998, 33, 11-20.
(7) Catheline, G.; Guilbaud, G.; Kayser, V. Peripheral component
in the enhanced antinociceptive effect of systemic U-69,593, a
κ-opioid receptor agonist in mononeuropathic rats. Eur. J .
Pharmacol. 1998, 357, 171-178.
(8) Chao, C. C.; Gekker, G.; Hu, S.; Sheng, W. S.; Shark, K. B.; Bu,
D. F.; Archer, S.; Bidlack, J . M.; Peterson, P. K. κ Opioid
receptors in human microglia downregulate human immunode-
fiency virus 1 expression. Proc. Natl. Acad. Sci. U.S.A. 1996,
93, 8051-8056.
(9) Dondio, G.; Ronzoni, S.; Petrillo, P. Non-peptide δ opioid agonists
and antagonists. Exp. Opin. Invest. Drug. 1997, 6, 1075-1098.
(10) Borsodi, A.; Benyhe, S.; Holzgrabe, U.; Marki, A.; Nachtsheim,
C. Structurally Novel Group of Ligands Selective for Kappa-
Opioid Receptors. Reg. Pept. 1994, 54, 27-28.
(11) Brandt, W.; Drosihn, S.; Haurand, M.; Holzgrabe, U.; Nacht-
sheim, C. Search for the Pharmacophore in Kappa-agonistic
Diazabicyclo[3.3.1]nonan-9-one-1,5-diesters and Arylacetamides.
Arch. Pharm. Pharm. Med. Chem. 1996, 329, 311-323.
(12) Ko¨gel, B.; Christoph, T.; Friderichs, E.; Hennies, H.-H.; Mat-
thiesen, T.; Schneider, J .; Holzgrabe, U. HZ2 a Selective κ-Opioid
Agonist. CNS Drug Rev. 1998, 4, 54-70.
(13) Craig, P. N. Interdependence between Physical Parameters and
Selection of Substituent Groups for Correlation Studies. J . Med.
Chem. 1971, 14, 680-684.
(14) Petrenko-Kritschenko, P.; Zoneff, N. U¨ ber die Condensation von
Acetondicarbonsa¨ureestern mit Benzaldehyd unter Anwendung
von Ammoniak. Ber. Dtsch. Chem. Ges. 1906, 39, 1358-1361.
(15) Mannich, C.; Mohs, P. U¨ ber Derivate eines aus zwei Piperidin-
ringen kondensierten bicyclischen Systems. Ber. Dtsch. Chem.
Ges. 1930, 63, 608-612.
(16) Siener, T.; Holzgrabe, U.; Drosihn, S.; Brandt, W. Conforma-
tional and configurational behaviour of κ-agonistic 3,7-diaza-
bicyclo[3.3.1]nonan-9-ones - synthesis, nuclear magnetic reso-
nance studies and semiempirical PM3 calculations. J . Chem.
Soc., Perkin Trans. 2 1999, 1827-1834.
(17) Yasuda, M. Dissociation Constants of Some Carboxylic Acids in
Mixed Aqueous Solvents. Bull. Chem. Soc. J pn. 1959, 32, 429-
432.
(18) Avdeef, A.; Barret, D. A.; Shaw, P. N.; Knaggs, R. D.; Davis, S.
S. Octanol-, Chloroform-, and Propylene Glycol Dipelargonat-
Water Partitioning of Morphine-6-glucuronide and Other Re-
lated Opiates. J . Med. Chem. 1996, 39, 4377-4381.
(19) Holzgrabe, U.; Erciyas, E. Synthese und Stereochemie potentiell
stark analgetischer 2,4-m-diarylsubstituierter 3,7-Diazabicyclo-
[3.3.1]nonan-9-on-1,5-diester. Arch. Pharm. (Weinheim) 1992,
325, 657-663.
IC50 values were calculated using the computer software
Figure P (version 6.0c; Biosoft, Cambridge, U.K.), and Ki values
were obtained using the Cheng-Prusoff equation.23
Nocicep tive Testin g Meth od s. The experiments were
carried out in male NMRI mice, weighing 23-26 g (tail-flick)
and 31-37 g (writhing test). The animals were kept under
standard laboratory conditions (group housing, approximately
22 °C, 12-h light-dark cycle), with free access to standard
laboratory food and tap water. Both were withdrawn during
the tests. Writhing was induced by ip injection of 0.35 mL of
a 0.02% solution of phenylquinone (PQ) according to the
method described by Hendershot and Forsaith.24 The charac-
teristic writhing response was observed and counted from 5
to 20 min after PQ administration. The tail-flick test (TF) was
carried out by a modification of the method described by
D’Amour and Smith.25 A TF analgesy meter (tail-flick type 55/
12/10.fl; Labtec, Dr. Hess, Germany) was used to measure TF
latency. The time it took a mouse to withdraw its tail from a
radiant heat source (bulb, 12 V/55 W) focused on the dorsal

2,4-Diaryl-Substituted 3,7-Diazabicylononanones
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 20 3751
(20) Kuhl, U.; Cambareri, A.; Sauber, C.; So¨rgel, F.; Hartmann, R.;
Euler, H.; Kirfel, A.; Holzgrabe, U. Synthesis, X-ray analysis and
Spectroscopic Characterization of the Hemiaminal Cyclisation
Product from 2-Dipyridine Substituted 3,7-Diazabicyclo[3.3.1]-
nonanone 1,5-diester. J . Chem. Soc., Perkin Trans. 2 1999,
2083-2088.
(23) Cheng, Y. C.; Prusoff, W. H. Relationship between the inhibition
constant (Ki) and the concentration of inhibition which causes
50% inhibition (I50) of an enzymatic reaction. Biochem. Phar-
macol. 1973, 22, 3099-3108.
(24) Hendershot, L. C.; Forsaith, J . Antagonism of the frequency of
phenylcholine-induced writhing in the mouse by weak analgesics
and nonanalgesics. J . Pharmacol. Exp. Ther. 1959, 125, 237-
240.
(21) Wood, P. L.; Charleson, S. E.; Lane, D.; Hudgin, R. L. Multiple
opiate receptors: Differential binding of µ, κ and δ agonists.
Neuropharmacology 1981, 20, 1215-1220.
(25) D’Amour, F. E.; Smith, D. L. A method for determining loss of
pain sensations. J . Pharmacol. Exp. Ther. 1941, 72, 74-78.
(22) Lowry, O. W.; Rosebrough, N. J .; Farr, A. L.; Randall, R. J .
Protein measurement with the folin phenol reagent. J . Biol.
Chem. 1951, 193, 265-275.
J M0009484