Discovery of a novel class of substituted pyrrolooctahydroisoquinolines as potent and selective δ opioid agonists, based on an extension of the message-address concept

Article abstract of DOI:10.1021/jm9608218

This paper describes the design and synthesis of compounds belonging to a novel class of substituted pyrrolooctahydroisoquinolines which are potent and selective δ opioid agonists. Molecular modeling studies performed on known, selective δ ligands such as (+)-3 and the potent δ agonist SNC 80 led to the identification of the carboxamido moiety of the latter as a putative nonaromatic δ address. Insertion of this moiety onto the octahyroisoquinoline opioid message resulted in (±)-5b, a potent and selective δ ligand. The active enantiomer, (-)-5b, displayed nanomolar affinity for the δ receptor (K(i) = 0.9 nM) with good μ/δ and κ/δ binding selectivity ratios (140 and 1480, respectively). In addition, (-)-5b behaved as a full δ agonist in the mouse vas deferens bioassay having an IC₅₀ = 25 nM and being antagonised in the presence of 30 nM naltrindole (NTI). These studies, based on the message-address concept, indicated that the nonaromatic (N,N-diethylamino)carbonyl moiety is a viable alternative to the classical benzene ring as a δ opioid address. Preliminary in vivo studies showed that (±)-5b produced a dose-related antinociception in the mouse abdominal constriction test after intracerebroventricular administration (ED₅₀ = 1.6 μg/mouse).

Full text of DOI:10.1021/jm9608218

3192
J . Med. Chem. 1997, 40, 3192-3198
Discover y of a Novel Cla ss of Su bstitu ted P yr r oloocta h yd r oisoqu in olin es a s
P oten t a n d Selective δ Op ioid Agon ists, Ba sed on a n Exten sion of th e
Messa ge-Ad d r ess Con cep t
Giulio Dondio,* Silvano Ronzoni, Drake S. Eggleston,† Marco Artico, Paola Petrillo, Giuseppe Petrone,
Luciano Visentin, Carlo Farina, Vittorio Vecchietti, and Geoffrey D. Clarke
SmithKline Beecham S.p.A., Via Zambeletti, 20021 Baranzate, Milan, Italy
Received December 4, 1996X
This paper describes the design and synthesis of compounds belonging to a novel class of
substituted pyrrolooctahydroisoquinolines which are potent and selective δ opioid agonists.
Molecular modeling studies performed on known, selective δ ligands such as (+)-3 and the
potent δ agonist SNC 80 led to the identification of the carboxamido moiety of the latter as a
putative nonaromatic δ address. Insertion of this moiety onto the octahydroisoquinoline opioid
message resulted in (()-5b, a potent and selective δ ligand. The active enantiomer, (-)-5b,
displayed nanomolar affinity for the δ receptor (Ki ) 0.9 nM) with good µ/ δ and κ/ δ binding
selectivity ratios (140 and 1480, respectively). In addition, (-)-5b behaved as a full δ agonist
in the mouse vas deferens bioassay having an IC50 ) 25 nM and being antagonised in the
presence of 30 nM naltrindole (NTI). These studies, based on the message-address concept,
indicated that the nonaromatic (N,N-diethylamino)carbonyl moiety is a viable alternative to
the classical benzene ring as a δ opioid address. Preliminary in vivo studies showed that (()-
5b produced a dose-related antinociception in the mouse abdominal constriction test after
intracerebroventricular administration (ED50 ) 1.6 µg/mouse).
Ch a r t 1
The therapeutically useful effects and adverse side
issues associated with morphine are primarily due to
an interaction with µ opioid receptors. Following the
great interest shown over the past 10 years in κ opioid
agonist-induced antinociception,1 attention has shifted
recently toward the potential of analgesics acting via δ
opioid receptors2 which might lack the negative proper-
ties associated with morphine. Considerable evidence
derived from animal models shows that existing δ opioid
agonists, predominantly peptides, produce antinocicep-
tion with relatively little effect on gastrointestinal
motility or respiratory depression and have little physi-
cal dependence liability in comparison with µ agonists.3
There is very little clinical evidence concerning their
therapeutic utility, although the δ selective peptide
[D-Ala2,D-Leu5]enkephalin (DADLE) has been shown to
cause effective pain relief when given intrathecally to
cancer patients.4 Widespread clinical proof of concept
studies await, therefore, the identification and develop-
ment of nonpeptidic drugs which act selectively as δ
opioid receptor agonists but which have more favorable
metabolism and pharmacokinetic properties than the
existing δ selective peptides.
The established rationale for the design of peptido-
mimetic drugs acting as selective δ receptor ligands has
been based on the message-address concept originally
proposed by Schwyzer5 and subsequently re-elaborated
by Portoghese.6 This concept attributes the role of the
opiate message to the Tyr1 residue of the tetrapeptidic
sequence of the endogenous peptides (Tyr1-Gly2-Gly3-
Phe4-...), whereas the δ address resides in the amino
acid sequence which starts with Phe4. In this context,
the residues Gly2-Gly3 represent a spacer maintaining
an appropriate distance between Tyr1 and Phe4.
Following this rationale the first nonpeptidic δ opioid
antagonist, naltrindole (NTI) (see Chart 1), was syn-
thesized6 as well as the closely related N-methyl ana-
logue, oxymorphindole (OMI), which has a partial δ
agonist profile in vitro.6 Recently, a novel class of
octahydroisoquinolines, formally derived from OMI
fragmentation,7 including the δ antagonist (+)-3 (SB
205588)7,8 and the δ agonists TAN679 and (-)-6 (SB
213698),8 has been described. In addition, two pipera-
zine derivatives, (()-BW373U8610 and one of its meth-
oxy analogues SNC 80,11 which show clear structural
differences compared to the previously known δ ligands,
have been identified as potent and selective δ agonists.
These piperazines do not apparently display the clas-
sical moieties responsible for interactions with the δ
receptor outlined in the message-address concept.
The present report describes molecular modeling
studies that have extended the message-address con-
† Current address: SmithKline Beecham Pharmaceuticals, 709
Swedeland Road, King of Prussia, PA 19406-0939.
X Abstract published in Advance ACS Abstracts, August 15, 1997.
S0022-2623(96)00821-7 CCC: $14.00 © 1997 American Chemical Society

Substituted Pyrrolooctahydroisoquinolines
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 20 3193
F igu r e 2. Stereoview of the superimposition between SNC
80 (gray) and 5b (black). Points used were the basic nitrogens,
the centroids of the oxygenated benzene rings, the centroids
of the second benzene ring of SNC 80 with that of pyrrole
nucleus of 5b and the two carbon atoms of the amidic moieties
(RMS ) 0.490).
F igu r e 1. Stereoview of the superimposition between SNC
80 (gray) and (+)-3 (black). Points used were the basic
nitrogens, the centroids of the oxygenated rings, and the
centroid of the second benzene ring of SNC 80 with that of
the pyrrole nucleus of (+)-3 (RMS ) 0.533).
cept to nonaromatic δ address moieties. In addition,
details of the synthesis of novel pyrrolooctahydroiso-
quinoline derivatives are given, together with prelimi-
nary pharmacological results which show them to be δ
selective agonists.
induced the amidic moiety to adopt a preferential
conformation outside the plane of the pyrrole ring,
ensuring a good fit with SNC 80. Thus, a model of
compound 5b was built, and after subsequent confor-
mational search, the resulting minimum energy con-
formation showed a good match with SNC 80 (Figure
2, RMS ) 0.490). Therefore, 5b 12 appeared a suitable
tool with which to test our hypothesis. Initially, the
racemic pyrrolooctahydroisoquinolines (()-5a -c were
synthesized and their opioid binding affinities were
evaluated. The agonist/antagonist properties of selected
compounds of interest were subsequently determined
in the mouse vas deferens (MVD), and their antinoci-
ceptive activity was evaluated in vivo.
Molecu la r Mod elin g
The aims of this study were as follows: (i) trying to
identify molecular similarities between the classical δ
ligands based on the message-address concept and the
novel piperazine derivatives; (ii) identifying which
moieties conferred agonist activity to SNC 80, and
finally (iii) combining the above findings to design novel
structures of putative δ selective agonists. As a starting
point for these studies, two representatives of those
classes of δ ligands for which crystallographically
derived molecular structures were known, i.e. (+)-3 (a
fragment of OMI possessing 4aS,11aR absolute config-
uration) and SNC 80,11 where chosen. The three-
dimensional coordinates of (+)-3 were used as input
geometries to build its 3D model while SNC 80 was built
according to its absolute stereochemistry.11 Energy
minimization using the MM2 force field and subsequent
extensive conformational searches were performed on
the above models to ensure that all conformers studied
were at their global minima.
Ch em istr y
Compound (()-1a is known and was prepared ac-
cording to the literature.13 Synthesis of compound (()-
1b was achieved according to the same method. Frac-
tional crystallization in absolute EtOH of the optically
active p-toluoyltartaric acid salts obtained from the
racemic ketone (()-1b and subsequent saponification
gave the two corresponding pure enantiomers (+)-1b
and (-)-1b in quantitative yield. Compound (()-1c was
obtained by de-ethylation of (()-1b with vinyl chloro-
formate14 and subsequent alkylation with (bromometh-
yl)cyclopropane.
The basic nitrogen bearing the allylic group and the
oxygenated benzene ring were identified as features that
could represent the opioid message in the piperazine
derivative SNC 80. These moieties were then super-
imposed to the corresponding fragments of (+)-3 (RMS
) 1.11). A better fit was obtained using also the
centroid of the second benzene ring of SNC 80 with that
of the pyrrole nucleus that served as a spacer in the
octahydroisoquinoline derivatives (Figure 1; RMS )
0.533). In this case the amidic moiety of SNC 80 lay
approximately in the same region of space as that
occupied by the classical, aromatic δ address. On the
basis of these overlaps, it was hypothesized that the
amidic moiety might be responsible for the δ opioid
selectivity shown by SNC 80 by playing the role of a
nonaromatic δ address. It was also possible that the
same moiety may confer agonist activity on SNC 80.
To confirm the validity of these hypotheses, the
putative nonaromatic δ address was attached to the
well-established opioid message represented by the
octahydroisoquinoline framework. Examination of the
superimposition shown in Figure 1 revealed that the
best position to insert a carboxamido moiety in the
octahydroisoquinoline nucleus was adjacent to the in-
dolic nitrogen of (+)-3. In addition, the presence of a
methyl group in position 3 of the target molecule
Compounds 2 (see Scheme 1) were obtained by Fis-
cher indole synthesis of the corresponding ketones 1b
with phenylhydrazine hydrochloride in refluxing MeOH
saturated with HCl.
Compounds 4a -c were obtained by Knorr synthesis
of the corresponding ketones 1a -c with N,N-diethyl-
2-phenylhydrazono-3-oxobutyramide15 in the presence
of zinc dust in acetic acid buffered with NaOAc.16
Compounds 2 and 4a -c were demethylated to the
corresponding phenols 3 and 5a -c with boron tribromide
in dry chloroform at room temperature.17
Resu lts a n d Discu ssion
The δ, µ, and κ opioid receptor binding affinities, along
with binding selectivity ratios for compounds 3 and 5a -
c, are shown in Table 1. For comparative purposes, the
opioid binding profiles of NTI, OMI, and SNC 80 are
also reported.
Compounds (()-5a -c exhibited high affinity for δ
receptors with Ki values ranging from 1.7 to 2.1 nM.
Furthermore, the N-ethyl derivative (()-5b was far less
potent at the µ and κ receptors, resulting in µ/ δ and
κ/ δ selectivity ratios of 210 and 690, respectively, while
(()-5a and (()-5c displayed low selectivity toward the
δ receptor. Thus, (()-5b was selected for further

3194 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 20
Dondio et al.
Sch em e 1a
mouse and was thus 5 times more potent than SNC 80
in this model [ED50 ) 9.5 (4.6-17.2) µg/mouse].
Con clu sion s
The present molecular modeling studies have dem-
onstrated how the message-address concept may be
extended to include nonaromatic moieties, such as N,N-
disubstituted carboxamido groups. This led to the
synthesis of compounds belonging to a novel class of
pyrrolooctahydroisoquinolines which behaved as potent
and selective δ opioid receptor ligands. The dialkyla-
midic moiety was also able to confer full δ agonist
activity when it replaced the aromatic δ address in the
framework of the δ antagonist (+)-3. Compound (-)-
5b therefore represents the prototype of a new class of
nonpeptidic, δ selective agonists which might serve as
tools to study further the pharmacology associated with
activation of the δ opioid receptors.
Preliminary studies with the racemate (()-5b have
already indicated that the compound is a potent and
efficaceous antinociceptive agent when injected icv in
the mouse.
Exp er im en ta l Section
Bin d in g Assa ys: Cell Cu ltu r e a n d P r ep a r a tion of th e
Cr u d e Mem br a n e F r a ction . NG108-15 neuroblastoma x
glioma hybrid cells (provided by European Collection of Animal
Cell Cultures, ECACC) were grown at 37 °C in 5% CO2-95%
humidified air atmosphere in Dulbecco’s MEM nutrient mix-
ture (without sodium pyruvate, using 4.5 g/L glucose) supple-
mented with 10% foetal calf serum, 2 mM glutamine, 2% HAT,
50 µg/mL streptomycin, and 50 units/mL penicillin. Cells at
confluence were harvested with 1 mM EDTA in Ca/Mg-free
phosphate-buffered saline with mechanical stirring and cen-
trifuged at 1000 rpm for 8 min. The pellets were stored at
-80 °C for a maximum of a month without any discernible
loss of binding activity. Prior to binding experiments, the cells
were suspended in ice cold 50 mM Tris-HCl, pH 7.4, buffer (3
× 107 cells/10 mL buffer) and homogenized by a PBI politron
(setting 5 for 15 s). The homogenate was centrifuged at 53000g
for 15 min at 4 °C. The resultant pellets were resuspended
in the same volume of buffer, incubated at 37 °C for 45 min,
and centrifuged at 53000g for 15 min. The pellets obtained
were finally resuspended in buffer, and 1.9 mL aliquots
(membranes from 3 × 105 cells) were used for the assay.
P r ep a r a tion of Mou se Br a in Mem br a n es. Whole brains
without cerebellum from male CD-1 mice (Charles River; 25-
30 g) were homogenized in 10 volumes (w/v) of ice cold 50 mM
Tris-HCl, pH 7.4, using a PBI tissue dispergerate (setting 5
for 15 s). The homogenate was centrifuged at 48000g for 10
min. The resulting pellets were resuspended in the same
volume of buffer, incubated at 37 °C for 45 min, and centri-
fuged at 48000g for 10 min. The pellets obtained were
resuspended in 100 volumes (original wet weight) of buffer and
used for the assay (1.9 mL sample).
Bin d in g Assa ys. The radiolabeled ligands employed in the
binding assays are as follow: [3H][D-Ala2, D-Leu5]enkephalin
([3H]DADLE; sp act. 32.3 µCi/nmol, New England Nuclear) has
been used to label δ binding sites in NG108-15 cell membranes
at the concentration of 1 nM, [3H][D-Ala2,MePhe4,Gly-ol5]-
enkephalin ([3H]DAMGO; sp act. 55.5 µCi/nmol, New England
Nuclear) at the concentration of 0.7 nM and [3H]U69593 (sp
act. 56.0 µCi/nmol, Amersham) at the concentration of 1.2 nM
have been used to label µ and κ binding sites, respectively, in
mouse brain membranes. The nonspecific binding was deter-
mined in presence of naloxone 10 µM (Salars, Como, Italy).
The samples, in triplicate, containing NG108-15 or mouse
brain membranes, tritiated and unlabeled ligands (final
volume 2 mL) were incubated at 25 °C. The time necessary
to reach equilibrium conditions was 60 min for [3H]-DADLE
and 50 min for [3H]DAMGO and [3H]U69593. The incubation
was terminated by rapid filtration through Whatman GF/B
a
Reagents and conditions: (i) PhNHNH2‚HCl, MeOH/HCl,
reflux; (ii) MeCOC(NNHPh)CONEt2, Zn, AcOH, AcONa, reflux;
(iii) BBr3, CHCl3, room temperature.
evaluation in the MVD functional bioassay to assess its
agonist/antagonist properties. Activity of SNC 80 is also
reported for comparison (Table 2). Importantly, and as
predicted, insertion of a carboxamido group conferred
full δ agonist activity to compound (()-5b which had
an IC50 value of 34 nM in this in vitro assay. In the
presence of 30 nM NTI, there was a 10-fold rightward
shift of the concentration-response curve for (()-5b,
confirming the role of the δ receptors in the response
seen.
The binding of the indolooctahydroisoquinoline de-
rivatives (-)-3 and (+)-3 to the δ receptor was clearly
enantiospecific (eudismic ratio ) 900), and this prompted
the synthesis of the two corresponding enantiomers of
compound (()-5b. As shown in Table 1, compound (-)-
5b (SB 219825) was a very potent δ ligand with a Ki of
0.9 nM and µ/ δ and κ/ δ binding selectivity ratios of 140
and 1480, respectively. Enantioselectivity was also
evident in this class of pyrrolooctahydroisoquinolines
with the δ binding activity residing predominantly in
the (-)-enantiomer (eudismic ratio ) 3700). Compound
(-)-5b also behaved as a full δ agonist in the MVD
assay causing a complete, concentration-dependent
inhibition of electrically induced phasic contraction with
an IC50 ) 25 nM. There was a 18-fold shift of the
concentration-response curve in the presence of 30 nM
NTI (Table 2).
The antinociceptive activity of (()-5b has been de-
termined using the mouse abdominal constriction model
following intracerebroventricular (icv) administration.
For comparative purposes the antinociceptive potency
of SNC 80 was also determined. Both compounds
produced dose-related antinociception with the highest
doses employed causing complete protection. Com-
pound (()-5b exhibited an ED50 of 1.6 (1.0-2.6) µg/

Substituted Pyrrolooctahydroisoquinolines
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 20 3195
Ta ble 1. Binding Affinities to δ, µ and κ Receptors
binding affinities (Ki nM)a
selectivity ratios
compd
δ
µ
κ
µ/δ
κ/δ
(()-3
7.1 ( 0.8
1971 ( 190
2.2 ( 0.4
2.1 ( 0.1
1.9 ( 0.4
3535 ( 310
0.9 ( 0.2
1.7 ( 0.6
0.5 ( 0.1
0.8 ( 0.1
1.7 ( 0.5
2093 ( 365
1631 ( 130
2618 ( 190
93 (n ) 2)
407 ( 25
>5000
129 ( 30
14.1 ( 0.9
15 ( 2.6
66 ( 5.6
1300 ( 280
334 ( 13
290
1
1190
43
210
>1
140
8
50
0.1
(-)-3
167 ( 11
(+)-3
771 ( 160
500 (n ) 2)
1298 ( 120
6005 ( 530
1340 ( 490
63.1 ( 8.7
9.5 ( 2.8
350
240
690
2
1480
37
20
90
790
(()-5a
(()-5b
(+)-5b
(-)-5b
(()-5c
NTI
35
80
760
OMI
SNC 80
77 ( 6.3
1348 ( 330
a
Each value represents the mean ( SEM of independent experiments, each performed in triplicate (n ) 3) unless otherwise indicated
in parentheses.
Ta ble 2. Effect of (()-5b, (-)-5b, and SNC 80 in the MVD
Bioassay
Ch em istr y
Melting points were determined on a Bu¨chi 512 hot stage
apparatus and are uncorrected. Proton NMR spectra were
compd
IC50 (nM)a
IC50 (nM)a + 30 nM NTI
IC50 ratio
9.7
recorded on
a Bruker ARX-300 spectrometer at 303 K.
(()-5b
34
(23-50)
26
330
(192-569)
462
Chemical shifts were recorded in parts per million (δ units)
downfield from tetramethylsilane (TMS). Mass spectral data
were obtained on a Finnigan MAT TSQ-700 spectrometer. IR
spectra were recorded in KBr with a Perkin-Elmer 1420
spectrophotometer. Optical rotations were determined with
a Perkin-Elmer 241 polarimeter at 20 °C at the sodium D line.
Silica gel used for flash column chromatography was Kieselgel
60 (230-400 mesh) (E. Merck AG, Darmstadt, Germany).
Enantiomeric excesses were measured by chiral HPLC meth-
odology using Shimadzu LC9 equipment and Lichrocart (25
cm × 4.6 mm) Chiradex (5 µm); 0.1 M phosphate buffer (pH
4.0)/MeOH; 0.8 mL/min; concentration 100 γ/mL; UV detector
220 nm. Elemental analyses are indicated only by the symbols
of the elements; analytical results were within 0.4% of the
theoretical values unless otherwise indicated.
(-)-5b
17.8
79.2
(18-37)
(342-625)
SNC 80
8
634
(5-13)
(293-1373)
a
95% confidence limits are reported in parentheses.
filters using a Brandel cell harvester system. Filters used for
[3H]U69593 were presoaked in buffer containing polyethylen-
imine 0.05%. The radioactivity on the discs was measured by
liquid scintillation counting on a Camberra Packard 2500TR
beta counter. All the experiments were performed in triplicate.
Ma th em a tica l An a lysis of Bin d in g Da ta . Binding pa-
rameters deriving from competition experiments (IC50 values)
were calculated by nonlinear regression analysis using the
software package RS/1 (BBN Software Products, Corp.).18 Ki
values were calculated from IC50 using the Cheng-Prusoff
relationship.19
MVD Isola ted Tissu e Bioa ssa y. Vasa deferentia were
obtained from CD-1 mice weighing 25-35 g and were sus-
pended in a Mg2+-free, oxygenated (95% O2, 5% CO2) Krebs
buffer at 37 oC. For the δ agonist/antagonist studies, the
tissues were electrically stimulated with pulse trains having
the following parameters: train duration 50 ms, stimulus
duration 2 ms, frequency of stimuli 50 Hz, maximal voltage
60-70 V, train frequency 0.1 Hz. Concentration-response
curves for each compound were constructed cumulatively.
(()-t r a n s-2-E t h yl-4a -(3-m e t h oxyp h e n yl)-1,2,3,4,4a ,
5,6,7,8,8a-decah ydr oisoqu in olin -6-on e Hydr och lor ide [(()-
1b.HCl]. This product was obtained according to the method
described for its N-Me analogue (()-1a .13 The free base was
taken up in MeOH, and the resulting solution was brought to
acidic pH with HCl/Et2O. The solvent was removed in vacuo,
and the resulting solid was crystallized from acetone. The
precipitate was filtered, washed, and dried to yield (()-1b‚-
HCl: mp 243-244 °C; IR (KBr) 3460, 2480, 1715, 1465 cm-1
;
1H NMR (DMSO-d6) δ 11.20 (br s, 1H), 7.26 (dd, J ) 8.1, 8.1
Hz, 1H), 7.07 (dd, J ) 8.1, 1.0 Hz, 1H), 6.94 (dd, J ) 2.0, 1.0
Hz, 1H), 6.82 (dd, J ) 8.1, 2.0 Hz, 1H), 3.75 (s, 3H), 3.54 (br
d, J ) 11.2 Hz, 1H), 3.36-3.22 (m, 2H), 3.11-3.01 (m, 2H),
2.82-2.70 (m, 1H), 2.77 (d, J ) 13.5 Hz, 1H), 2.62-2.50 (m,
1H), 2.58 (d, J ) 13.5 Hz, 1H), 2.43-2.13 (m, 5H), 2.10-1.98
(m, 1H), 1.23 (t, J ) 8.1 Hz, 3H). Anal. (C18H25NO2‚HCl) C,
H, N, Cl.
Linear regression analysis and IC50 concentrations were
evaluated according to Tallarida and Murray.20
In Vivo An tin ocicep tive Stu d ies. Male Swiss mice
(Charles River; 20-35 g) were used throughout these studies.
The mouse phenyl-p-benzoquinone-induced (PPQ) abdominal
constriction test (MAC) was performed according to the
procedure described by Siegmund et al.21 modified by Milne
and Twomey.22 The icv injection was performed according to
the method of Domino et al.23 by insertion of a disposable 30
(-)-t r a n s-2-E t h yl-4a -(3-m e t h oxyp h e n yl)-1,2,3,4,4a ,
5,6,7,8,8a-decah ydr oisoqu in olin -6-on e Hydr och lor ide [(-)-
1b.HCl]. A solution of 5.97 g (20.77 mmol) of (()-1b in 80
mL of EtOH was added to a solution of 8.02 g (20.77 mmol) of
(+)-di-O,O′-p-toluoyl-D-tartaric acid in 80 mL of EtOH. After
a gentle warming, the resulting solution was filtered and the
less soluble diastereomeric salt crystallized from the filtrate
on standing. The salt was recrystallized from EtOH, up to a
constant optical activity, to give 5.62 g of (+)-di-O,O′-p-
1
gauge /2 in. needle mated to a 50 µL luer syringe (Hamilton),
through the soft bone 1.5 mm to the right bregma on the
coronal suture. The needle was inserted through a stainless-
steel tube that acted as a stopper (needle protrusion, 3.5 mm).
Drugs were injected 5 min before the intraperitoneal (ip)
administration of an aqueous solution of PPQ (2 mg/kg at 37
°C, in a final volume of 10 mL/kg). The treated mice were
placed in a compartemented perpex box maintained at room
temperature and were observed for a period of 8 min. During
this period the number of abdominal constriction responses
for each animal was recorded.
toluoyl-D-tartrate: mp 161-163 °C; [R]20 ) +57.42 (c ) 2,
D
MeOH). Anal. (C38H43NO10) C, H, N.
The tartrate salt was transformed into the free base by
dissolving it in 5% NaOH solution, extracting with CH2Cl2,
and evaporating the solvent, yielding 2.3 g (77%) of (-)-1b as
an oil. [R]20 ) -83.85 (c ) 2, CHCl3). The corresponding
D
hydrochloride salt was formed and crystallized as described
above for (()-1b.HCl. IR and 1H NMR matched those of the
racemate (()-1b.HCl.
Da ta Eva lu a tion . The degree of graded antinociceptive
protection afforded by the drug was determined according to
the method described by Locke et al.24
ED50 values, and their 95% confidence intervals (in paren-
theses), were determined using the method of Finney.25
(+)-t r a n s-2-E t h yl-4a -(3-m e t h oxyp h e n yl)-1,2,3,4,4a ,
5,6,7,8,8a-decah ydr oisoqu in olin -6-on e Hydr och lor ide [(+)-
1b.HCl]. The mother liquors obtained from the first crystal-

3196 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 20
Dondio et al.
1
lization of the preceding description were evaporated to
dryness. The residue was treated with 5% NaOH solution,
extracted with CH2Cl2, and evaporated to afford 2.75 g (9.6
mmol) of the enriched free base which was dissolved in 45 mL
of EtOH. A solution of 3.78 g (9.6 mmol) of (-)-di-O,O′-p-
toluoyl-L-tartaric acid, dissolved in 45 mL of EtOH, was added
to the hot solution of the free base and the diastereomeric salt
crystallized on standing. The salt was recrystallized until
constant optical activity to give 5.82 g of (-)-di-O,O′-p-
274-277 °C; [R]20 ) +147.0 (c ) 2, MeOH); IR and H NMR
D
matched those of the racemate (()-2.HCl. Anal. (C24H28N2O.-
HCl) C, H, N, Cl.
(-)-tr a n s-2-E t h yl-4a -(3-m et h oxyp h en yl)-1,2,3,4,4a ,5,
11,11a -octa h yd r oin d olo[2,3-g]isoqu in olin e Hyd r och lo-
r id e [(-)-2‚HCl]. This compound was prepared from (+)-1b
using the same procedure reported for (()-2: yield 47%; mp
1
273-276 °C; [R]20 ) -143.1 (c ) 2, MeOH); IR and H NMR
D
matched those of the racemate (()-2. Anal. (C24H28N2O.HCl)
toluoyl-L-tartrate: mp 162-163 °C; [R]20 ) -55.36 (c ) 2,
C, H, N, Cl.
D
MeOH). Anal. (C38H43NO10) C, H, N.
N,N-Dieth yl-2-p h en ylh yd r a zon o-3-oxobu tyr a m id e. A
solution of 15.7 g (0.1 mol) of N,N-diethyl-3-oxobutyramide,
12 g (0.14 mol) of AcONa in 20 mL of water, and 75 mL of
EtOH was cooled to 10 °C, and 0.1 mol of a freshly prepared
solution of phenyldiazonium chloride were added dropwise.
The precipitated solid was filtered and dried in vacuo yielding
22.6 g (87%) of product: mp 63-65 °C; IR (KBr) 2970, 1720,
The tartrate salt was transformed into the free base by
dissolving it in 5% NaOH solution, extracting with CH2Cl2,
and evaporating the solvent, yielding 2.4 g (80%) of (+)-1b as
an oil, [R]20 ) +82.20 (c ) 2, CHCl3). The corresponding
D
hydrochloride salt was formed and crystallized as described
for (()-1b.HCl. IR and 1H NMR matched those of the
racemate (()-1b.HCl.
1620, 1605, 1560, 1245 cm-1
; 1H NMR (CDCl3) δ 9.30 (s, 1H),
7.41-7.22 (m, 5H), 3.60 (q, J ) 6.4 Hz, 2H), 3.22 (q, J ) 6.4
Hz, 2H), 2.51 (s, 3H), 1.35 (t, J ) 6.4 Hz, 3H), 1.20 (t, J ) 6.4
Hz, 3H), MS (TSP) m/ z 262.1 (MH+).
(()-tr a n s-2-(Cyclopr opylm eth yl)-4a-(3-m eth oxyph en yl)-
1,2,3,4,4a ,5,6,7,8,8a -d eca h yd r oisoqu in olin -6-on e Hyd r o-
ch lor id e [(()-1c.HCl]. A solution of 1.2 g (4.2 mmol) of (()-
1b and 1.8 g (12.6 mmol) of Proton-Sponge (Aldrich) in 34 mL
of 1,2-dichloroethane was treated with 1.4 mL (16.8 mmol) of
vinyl chloroformate at 0 °C under a nitrogen atmosphere. The
reaction mixture was stirred for 15 min and then refluxed for
3 h. The solvent was removed in vacuo, and the residue was
taken up in water and extracted with Et2O. The organic layer
was washed with 3% HCl and then was dried over Na2SO4,
and the solvent was removed in vacuo. The residue was then
dissolved in EtOH, treated with an excess of HCl/EtOH, and
refluxed for 1 h. The solvent was removed in vacuo, obtaining
0.88 g (3.08 mmol) of the de-ethylated intermediate which was
dissolved in 15 mL of DMF; 0.44 g (3.23 mmol) of (bromometh-
yl)cyclopropane were added together with 0.64 g (4.6 mmol)
of K2CO3 and a catalytical amount of KI. The reaction mixture
was stirred at 60 °C for 2 h, then the solvent was removed in
vacuo, and the crude product was purified by flash chroma-
tography (AcOEt/MeOH/concentrated NH4OH, 90:10:0.8). The
resulting solid was dissolved in acetone and treated with an
excess of HCl/Et2O. The precipitate was filtered, washed, and
dried to yield 280 mg (30%) of (()-1c·HCl: mp 78 °C dec; IR
(()-tr a n s-2-[(Diet h yla m in o)ca r b on yl]-6-et h yl-8a -(3-
m eth oxyp h en yl)-3-m eth yl-4,4a ,5,6,7,8,8a ,9-octa h yd r o-1H-
p yr r olo[2,3-g]isoqu in olin e Hyd r och lor id e [(()-4b·HCl].
Under a nitrogen atmosphere, 1.4 g (4.9 mmol) of (()-1b and
1.54 g (5.8 mmol) of N,N-diethyl-2-phenylhydrazono-3-oxobu-
tyramide were dissolved in a mixture of 5 mL of glacial AcOH
and 0.48 g (5.8 mmol) of AcONa. The solution was heated to
60 °C, and then 1.47 g (22.5 mmol) of zinc dust was added
portionwise. The resulting mixture was refluxed for 2 h then
cooled to room temperature. The precipitate was removed by
decantation and washed with 5 mL of glacial AcOH. The
combined acidic solutions were diluted with iced water (50
mL), the pH was adjusted to 8 with 20% NaOH, and then the
solution was extracted with AcOEt. The organic layer was
dried (Na2SO4), and the solvent was evaporated, affording a
residue that was purified by flash chromatography (AcOEt/
MeOH/concentrated NH4OH, 90:10:1). The resulting solid was
dissolved in acetone and the solution treated with an excess
of Et2O/HCl. The solvent was evaporated and the solid
triturated with Et2O, yielding 1.5 g (67%) of (()-4b·HCl: mp
272-274 °C dec; IR (KBr) 3410, 3200, 2920, 2500, 1600, 1580
(KBr) 3400, 2940, 1715, 1600 cm-1 1H NMR (DMSO-d6) δ
;
10.00 (br s, 1H), 7.27 (dd, J ) 7.9, 7.9 Hz, 1H), 7.06 (dd, J )
7.9, 1.0 Hz, 1H), 6.93 (dd, J ) 2.0, 1.0 Hz, 1H), 6.81 (dd, J )
7.9, 2.0 Hz, 1H), 3.74 (s, 3H), 3.62 (br d, J ) 11.6 Hz, 1H),
3.46-3.35 (m, 2H), 3.05-2.93 (m, 2H), 2.78 (d, J ) 14.3 Hz,
1H), 2.65-2.40 (m, 2H), 2.59 (d, J ) 14.3 Hz, 1H), 2.33-2.00
(m, 6H), 1.12-1.02 (m, 1H), 0.64-0.57 (m, 2H), 0.39-0.32 (m,
cm-1 1H NMR (DMSO-d6) δ 10.84 (br s, 1H), 10.39 (s, 1H),
;
7.19 (dd, J ) 8.2, 8.2 Hz, 1H), 7.03 (dd, J ) 8.2, 1.0 Hz, 1H),
6.91 (dd, J ) 2.0, 1.0 Hz, 1H), 6.76 (dd, J ) 8.2, 2.0 Hz, 1H),
3.69 (s, 3H), 3.50 (br d, J ) 11.0 Hz, 1H), 3.42-3,22 (m, 6H),
3.12 (m, 1H), 3.03 (m, 2H), 2.95 (d, J ) 15.0 Hz, 1H), 2.78-
2.50 (m, 5H), 2.18 (ddd, J ) 11.2, 11.2, 2.0 Hz, 1H), 1.98 (s,
3H), 1.22 (t, J ) 6.4 Hz, 3H), 1.02 (t, J ) 6.4 Hz, 6H), MS
.
2H), MS (EI) m/ z 314.2 (MH+). Anal. (C20H27NO2 HCl) H,
.
N, Cl; C: calcd, 68.65; found, 68.07.
(TSP) m/ z 424.2 (MH+). Anal. (C26H37N3O2 HCl) C, H, N, Cl.
(()-tr a n s-2-E t h yl-4a -(3-m et h oxyp h en yl)-1,2,3,4,4a ,5,
11,11a -octa h yd r oin d olo[2,3-g]isoqu in olin e Hyd r och lo-
r id e [(()-2·HCl]. A solution of 470 mg (1.64 mmol) of (()-1b
and 357 mg (2.47 mmol) of phenylhydrazine hydrochloride in
33 mL of MeOH saturated with HCl was refluxed under a
nitrogen atmosphere for 3 h and then cooled at room temper-
ature. The reaction mixture was evaporated to dryness, the
residue was dissolved in AcOEt and treated with an excess of
1 N NaOH, and the aqueous phase was counterextracted with
AcOEt. The combined extracts were dried (Na2SO4) and
evaporated. The solid residue was purified by flash chroma-
tography (CH2Cl2/MeOH/concentrated NH4OH, 94:5:0.5), yield-
ing 457 mg of (()-2, which were dissolved in 10 mL of acetone
and treated with an excess of HCl/Et2O. The precipitate was
(-)-tr a n s-2-[(Diet h yla m in o)ca r b on yl]-6-et h yl-8a -(3-
m eth oxyp h en yl)-3-m eth yl-4,4a ,5,6,7,8,8a ,9-octa h yd r o-1H-
p yr r olo[2,3-g]isoqu in olin e Hyd r och lor id e [(-)-4b‚HCl].
This compound was prepared from (-)-1b using the same
procedure reported for (()-4b: yield 85%; mp 273-276 °C dec;
1
[R]20
D ) -20.32 (c ) 1, MeOH); IR and H NMR matched those
.
of the racemate (()-4b.HCl. Anal. (C26H37N3O2 HCl) C, H,
N, Cl.
(+)-tr a n s-2-[(Diet h yla m in o)ca r b on yl]-6-et h yl-8a -(3-
m eth oxyp h en yl)-3-m eth yl-4,4a ,5,6,7,8,8a ,9-octa h yd r o-1H-
p yr r olo[2,3-g]isoqu in olin e Hyd r och lor id e [(+)-4b·HCl].
This compound was prepared from (+)-1b using the same
procedure reported for (()-4b: yield 74%; mp 273-275 °C dec;
1
[R]20D ) +20.65 (c ) 1, MeOH); IR and H NMR matched those
.
of the racemate (()-4b.HCl. Anal. (C26H37N3O2 HCl) C, H,
filtered, washed, and dried to yield 400 mg (61%) of (()-
2·HCl: mp 273-275 °C; IR (KBr) 3400, 3200, 1605, 1460 cm-
;
1
N, Cl.
1H NMR (free base, CDCl3) δ 7.67 (br s, 1H), 7.44 (m, 1H),
7.21 (m, 1H), 7.11-7.00 (m, 5H), 6.62 (br d, J ) 8.1 Hz, 1H),
3.69 (s, 3H), 3.10 (d, J ) 15.7 Hz, 1H), 3.10 (dd, J ) 12.5, 2.1
Hz, 1H), 3.01-2.85 (m, 4H), 2.68-2.57 (m, 2H), 2.50 (q, J )
6.7 Hz, 2H), 2.41 (m, 1H), 2.10-1.98 (m, 2H), 1.14 (t, J ) 6.7
Hz, 3H). Anal. (C24H28N2O.HCl) C, H, N, Cl.
(+)-tr a n s-2-E t h yl-4a -(3-m et h oxyp h en yl)-1,2,3,4,4a ,5,
11,11a -octa h yd r oin d olo[2,3-g]isoqu in olin e Hyd r och lo-
r id e [(+)-2·HCl]. This compound was prepared from (-)-1b
using the same procedure reported for (()-2: yield 43%; mp
(()-tr a n s-2-[(Dieth yla m in o)ca r bon yl]-3,6-d im eth yl-8a -
(3-m eth oxyph en yl)-4,4a,5,6,7,8,8a,9-octah ydr o-1H-pyr r olo-
[2,3-g]isoqu in olin e Hyd r och lor ide [(()-4a·HCl]. This com-
pound was prepared from (()-1a 13 using the same procedure
reported for (()-4b: yield 15%; mp 250 °C dec; IR (KBr) 3410,
3200, 2915, 2510, 1605, 1580 cm-1; 1H NMR (DMSO-d6) δ 10.47
(br s, 1H), 10.43 (s, 1H), 7.06 (dd, J ) 7.9, 7.9 Hz, 1H), 6.87-
6.81 (m, 2H), 6.60 (dd, J ) 7.9, 2.0 Hz, 1H), 3.71 (s, 3H), 3.45
(br d, J ) 11.2 Hz, 1H), 3.43-3,12 (m, 5H), 2.94 (d, J ) 15.9
Hz, 1H), 2.74 (br s, 3H), 2.67-2.42 (m, 6H), 2.42 (d, J ) 13.7

Substituted Pyrrolooctahydroisoquinolines
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 20 3197
1.01 (t, J ) 6.4 Hz, 6H), MS (TSP) m/ z 396.4 (MH+). Anal.
Hz, 1H) 2.04 (ddd, J ) 13.7, 13.7, 1.0 Hz, 1H), 1.91 (s, 3H),
1.00 (t, J ) 6.4 Hz, 6H), MS (TSP) m/ z 410.7 (MH+). Anal.
.
(C24H33N3O2 HCl) C, H, N, Cl.
.
(C25H35N3O2 HCl) C, H, N, Cl.
(()-tr a n s-2-[(Dieth yla m in o)ca r bon yl]-6-eth yl-8a -(3-h y-
d r oxyp h en yl)-3-m eth yl-4,4a ,5,6,7,8,8a -octa h yd r o-1H-p yr -
r olo[2,3-g]isoqu in olin e [(()-5b]. This compound was pre-
pared from (()-4b using the same procedure reported for (()-
3: yield 20%, crystallization solvent EtOH; mp 238-240 °C;
IR (KBr) 3200, 2980, 2940, 1600 cm-1; 1H NMR (DMSO-d6) δ
10.22 (s, 1H), 9.03 (s, 1H), 6.97 (dd, J ) 7.9, 7.9 Hz, 1H), 6.86
(d, J ) 2.0 Hz, 1H), 6.83 (br d, J ) 7.9 Hz, 1H), 6.47 (dd, J )
7.9, 2.0 Hz, 1H), 3.45-3.23 (m, 4H), 2.90 (d, J ) 15.9 Hz, 1H),
2.83 (dd, J ) 10.5, 2.0 Hz, 1H), 2.63-2.42 (m, 6H), 2.35-2.20
(m, 3H), 2.12 (br d, J ) 9.6 Hz, 1H), 1.88 (s, 3H), 1.80-1.79
(m, 2H), 1.02 (t, J ) 6.4 Hz, 6H), 0.88 (t, J ) 6.4 Hz, 3H), MS
(TSP) m/ z 410.2 (MH+). Anal. (C25H35N3O2‚0.5H2O) C, H, N.
(-)-(4aS,8aR)-tr a n s-2-[(Dieth ylam in o)car bon yl]-6-eth yl-
8a -(3-h yd r oxyp h en yl)-3-m eth yl-4,4a ,5,6,7,8,8a ,9-octa h y-
d r o-1H-p yr r olo[2,3-g]isoq u in olin e [(-)-5b]. This com-
pound was prepared from (-)-4b using the same procedure
reported for (()-3: yield 17%, crystallization solvent EtOH;
(()-tr a n s-6-(Cyclop r op ylm eth yl)-2-[(d ieth yla m in o)ca r -
bon yl]-8a -(3-m eth oxyp h en yl)-3-m eth yl-4,4a ,5,6,7,8,8a ,9-
octa h yd r o-1H-p yr r olo[2,3-g]isoqu in olin e Hyd r och lor id e
[(()-4c·HCl]. This compound was prepared from (()-1c using
the same procedure reported for (()-4b: yield 62%; mp 190-
195 °C. IR (KBr) 3400, 3200, 2915, 2580, 1600 cm-1; 1H NMR
(DMSO-d6) δ 10.58 (br s, 1H), 10.40 (s, 1H), 7.20 (dd, J ) 7.9,
7.9 Hz, 1H), 7.04 (dd, J ) 7.9, 1.0 Hz, 1H), 6.91 (dd, J ) 2.0,
1.0 Hz, 1H), 6.76 (dd, J ) 7.9, 2.0 Hz, 1H), 3.69 (s, 3H), 3.50
(br d, J ) 11.2 Hz, 1H), 3.50-3,22 (m, 5H), 3.01-2.93 (m, 4H),
2.78-2.50 (m, 5H), 2.18 (m, 1H), 1.94 (s, 3H), 1.05-0.98 (m,
2H), 1.02 (t, J ) 6.4 Hz, 6H), 0.62-0.58 (m, 2H), 0.42-0.38
.
(m, 2H), MS (EI) m/ z 450.5 (MH+). Anal. (C28H39N3O2 HCl)
C, H, N, Cl.
(()-tr a n s-2-E t h yl-4a -(3-h yd r oxyp h en yl)-1,2,3,4,4a ,5,
11,11a -octa h yd r oin d olo[2,3-g]isoqu in olin e Hyd r och lo-
r id e [(()-3·HCl]. To a stirred solution of 0.55 mL (5.8 mmol)
of boron tribromide in 17 mL of dry CHCl3 was added dropwise,
a solution of 383 mg (0.96 mmol) of (()-2.HCl in 5 mL of CHCl3
under a nitrogen atmosphere and at room temperature. After
30 min the solution was poured onto 17 g of ice containing 2
mL of concentrated NH4OH and stirred for 30 min. The
precipitate was collected by filtration; the filtrate was extracted
with CH2Cl2, dried (Na2SO4), evaporated, and combined with
the precipitate. The crude product was purified by flash
chromatography (CH2Cl2/MeOH/concentrated NH4OH, 79:15:
1), and the resulting solid was dissolved in 5 mL of MeOH
and treated with an excess of HCl/Et2O. The precipitate was
filtered, washed, and dried to yield 120 mg (33%) of (()-
3·HCl: mp >300 °C; IR (KBr) 3450, 3260, 3200, 1600, 1450
mp 239-241 °C; [R]20 ) -57.94 (c ) 1, MeOH); ee >99.5%
D
(HPLC; 0.1 M phosphate buffer (pH 4.0)/MeOH ) 85:15); IR
and NMR matched those of the racemate (()-5b; MS (TSP)
m/ z 410.2 (MH+). Anal. (C25H35N3O2‚0.5H2O) C, H, N.
(+)-(4aR,8aS)-tr a n s-2-[(Dieth ylam in o)car bon yl]-6-eth yl-
8a -(3-h yd r oxyp h en yl)-3-m eth yl-4,4a ,5,6,7,8,8a ,9-octa h y-
d r o-1H-p yr r olo[2,3-g]isoq u in olin e [(+)-5b]. This com-
pound was prepared from (+)-4b using the same procedure
reported for (()-3: yield 16%, crystallization solvent EtOH;
mp 239-240 °C; [R]20 ) +57.49 (c ) 1, MeOH); ee >99.5%
D
(HPLC; 0.1 M phosphate buffer (pH 4.0)/MeOH ) 85:15); IR
and NMR matched those of the racemate (()-5b; MS (TSP)
m/ z 410.2 (MH+). Anal. (C25H35N3O2‚0.5H2O) C, H, N.
(()-tr a n s-6-(Cyclop r op ylm eth yl)-2-[(d ieth yla m in o)ca r -
b on yl]-8a -(3-h yd r oxyp h en yl)-3-m et h yl-4,4a ,5,6,7,8,8a ,9-
octa h yd r o-1H-p yr r olo[2,3-g]isoqu in olin e Hyd r och lor id e
[(()-5c·HCl]. This compound was prepared from (()-4c using
the same procedure reported for (()-3: yield 35%; mp 270-
cm-1 1H NMR (DMSO-d6) δ 10.60 (s, 1H), 10.30 (br s, 1H),
;
9.25 (s, 1H), 7.33 (d, J ) 7.4 Hz, 1H), 7.18 (d, J ) 7.4 Hz, 1H),
7.03 (dd, J ) 7.4, 7.4 Hz, 1H), 6.97 (dd, J ) 7.8, 2.0 Hz, 1H),
6.95-6.89 (m, 3H), 6.53 (dd, J ) 7.8, 2.0 Hz, 1H), 3.66 (br d,
J ) 11.4 Hz, 1H), 3.42 (br d, J ) 11.4 Hz, 1H), 3.29 (ddd, J )
11.4, 11.4, 9.5 Hz, 1H), 3.20-3.02 (m, 3H), 3.00-2.80 (m, 3H),
2.72-2.60 (m, 1H), 2.60-2.48 (m, 2H), 2.11 (dd, J ) 13.8, 13.8
Hz, 1H), 1.21 (t, J ) 6.4 Hz, 3H). Anal. (C23H26N2O.HCl) H,
N, Cl, C: calcd, 72.14; found, 71.69.
1
272 °C dec; IR (KBr) 3010, 2700, 1595, 1580 cm-1; H NMR
(DMSO-d6) δ 10.50 (br s, 1H), 10.40 (s, 1H), 9.30 (s, 1H), 7.06
(dd, J ) 7.9, 7.9 Hz, 1H), 6.90-6.85 (m, 2H), 6.58 (dd, J ) 7.9,
2.0 Hz, 1H), 3.61 (br d, J ) 11.2 Hz, 1H), 3.45 (br d, J ) 11.2
Hz, 1H), 3.40-3,12 (m, 4H), 3.01-2.92 (m, 3H), 2.67-2.42 (m,
6H), 2.14 (ddd, J ) 12.0,12.0, 2.0 Hz, 1H), 1.88 (s, 3H), 1.08-
1.00 (m, 2H), 1.01 (t, J ) 6.4 Hz, 6H), 0.62-0.58 (m, 2H), 0.40-
0.36 (m, 2H), MS (EI) m/ z 435.3 (M+). Anal. (C27H37N3O2‚HCl)
C, H, N, Cl.
Com p u ter Mod elin g Stu d ies. Models of compound 5b
and SNC 80 were constructed with standard bond lengths and
angles from the fragment database in MacroModel V5.0
(Columbia University, New York, NY 10027) using a Silicon
Graphics workstation (Indigo II). The structure of compound
(+)-3 was built by inverting the known X-ray coordinates of
the relative counterpart, (-)-3. All the compounds have been
modeled as free bases and minimized by the MacroModel/
BatchMin V5.0 program using the MM2 force field. To
perform an extensive conformational search, a MonteCarlo/
Energy minimization26 was carried out (Ei - Emin e 40 kJ /
mol). Representative minimum energy conformations of each
compound were used to perform superimposition studies. For
compound (-)-5b, X-ray data agreed favorably with the low-
energy conformers used throughout our studies.27
(+)-(4a S,11a R)-tr a n s-2-E t h yl-4a -(3-h yd r oxyp h en yl)-
1,2,3,4,4a,5,11,11a-octah ydr oin dolo[2,3-g]isoqu in olin e Hy-
d r och lor id e [(+)-3·HCl]. This compound was prepared from
(+)-2‚HCl using the same procedure reported for (()-3: yield
63%; mp > 300 °C; [R]20 ) +141.1 (c ) 1, MeOH); ee >99.5%
D
(HPLC; 0.1 M phosphate buffer (pH 4.0)/MeOH ) 60:40); IR
1
and H NMR matched those of the racemate (()-3.HCl. Anal.
(C23H26N2O.HCl) H, N, Cl; C: calcd, 72.14; found, 71.72
(-)-(4a R,11a S)-tr a n s-2-E t h yl-4a -(3-h yd r oxyp h en yl)-
1,2,3,4,4a,5,11,11a-octah ydr oin dolo[2,3-g]isoqu in olin e Hy-
d r och lor id e [(-)-3‚HCl]. This compound was prepared from
(-)-2 using the same procedure reported for (()-3: yield 63%;
mp >300 °C; [R]20D ) -141.5 (c ) 1, MeOH); ee >99.5% (HPLC;
0.1 M phosphate buffer (pH 4.0)/MeOH ) 60:40); IR and 1H
NMR matched those of the racemate (()-3.HCl. Anal.
(C23H26N2O.HCl) H, N, Cl, C: calcd, 72.14; found, 71,62.
A sample of the hydrobromide salt was prepared for X-ray
analysis. The free base was dissolved in MeOH, and then the
resulting solution was brought to acidic pH with 48% HBr.
The solvent was removed and the resulting solid crystallized
from MeOH, mp >300 °C. Anal. (C23H26N2O.HBr·MeOH) C,
H, N, Br.
(()-tr a n s-2-[(Dieth yla m in o)ca r bon yl]-3,6-d im eth yl-8a -
(3-h ydr oxyph en yl)-4,4a,5,6,7,8,8a,9-octah ydr o-1H-pyr r olo-
[2,3-g]isoqu in olin e Hyd r och lor id e [(()-5a·HCl]. This com-
pound was prepared from (()-4a using the same procedure
reported for (()-3: yield 12%; mp 250 °C dec; IR (KBr) 3450,
3120, 2970, 1600, 1580 cm-1; 1H NMR (DMSO-d6) δ 10.45 (br
s, 1H), 10.40 (s, 1H), 9.30 (s, 1H), 7.06 (dd, J ) 7.9, 7.9 Hz,
1H), 6.88-6.82 (m, 2H), 6.59 (dd, J ) 7.9, 2.0 Hz, 1H), 3.46
(br d, J ) 11.2 Hz, 1H), 3.43-3,12 (m, 5H), 2,94 (d, J ) 15.9
Hz, 1H), 2.74 (br s, 3H), 2.67-2.42 (m, 6H), 2.41(d, J ) 13.7
Hz, 1H) 2.04 (ddd, J ) 13.7, 13.7, 1.0 Hz, 1H), 1.90 (s, 3H),
Ack n ow led gm en t. We thank A. Cerri and R. Mena
for providing NMR and mass spectroscopic data and L.
Antonini and M. Digiuni for their biological assistance.
Su p p or tin g In for m a tion Ava ila ble: Data for single-
crystal X-ray structure analysis of compounds (-)-3 and (-)-
5b (14 pages). Ordering information is given on any current
masthead page.
Refer en ces
(1) Millan, M. J . κ-Opioid Receptor and Analgesia. Trends Phar-
macol. Sci. 1990, 11, 70-76.

3198 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 20
Dondio et al.
(2) Rapaka, R. S.; Porreca F. Development of Delta Opioid Peptides
as Nonaddicting Analgesics. Pharm. Res. 1991, 8, 1-8.
(3) (a) Cheng, P. Y.; Wu, D.; Decena, J .; Soong, Y.; McCabe, S.; Szeto,
H. H. Opioid-induced Stimulation of Fetal Respiratory Activity
by [D-Ala2]-deltorphin I. Eur. J . Pharmacol. 1993, 230, 85-88.
(b) Cowan, A.; Zhu, X. Z.; Mosberg, H. I.; Omnaas, J . R.; Porreca,
F. Direct Dependence Studies in Rats with Agents Selective for
Different Types of Opioid Receptor. J . Pharmacol. Exp. Ther.
1988, 246, 950-955. (c) Porreca, F.; Mosberg, H. I.; Hurst, R.;
Hruby, V. J .; Burks, T. F. A Comparison of the Analgesic and
Gastrointestinal Transit Effects of [D-Pen2, L-Cys5]enkephalin
after Intracerebroventricular and Intrathecal Administration to
Mice. Life Sci. 1983, 33, 457-460. (d) Galligan, J . J .; Mosberg,
H. I.; Hurst, R.; Hruby, V. J .; Burks, T. F. Cerebral Delta Opioid
Receptors Mediate Analgesia but not the Intestinal Motility
Effects of Intracerebroventricularly Administered Opioids. J .
Pharmacol. Exp. Ther. 1984, 229, 641-648. (e) Sheldon, R. J .;
Riviere, P. J . M.; Malarchik, M. E.; Mosberg, H. I.; Burks, T. F.;
Porreca, F. Opioid Regulation of Mucosal Ion Transport in the
Mouse Isolated J ejunum. J . Pharmacol. Exp. Ther. 1990, 253,
144-151.
(13) (a) Cantrell, B. E.; Paschal, J . W.; Zimmerman, D. M. An
Efficient Synthesis of the 4a-Aryl-6-oxodecahydroisoquinolines.
J . Org. Chem. 1989, 54, 1442-1445. (b) Zimmerman, D. M.;
Cantrell, B. E.; Reel, J . K.; Hemrick-Luecke, S. K.; Fuller, R.
W. Characterization of the Neurotoxic Potential of m-Methoxy-
MPTP and Use of its N-Ethyl Analogue as a Means of Avoiding
Exposure to a Possible Parkinsonism-Causing Agent. J . Med.
Chem. 1986, 29, 1517-1520.
(14) Olofson, R. A.; Schnur, R. C.; Bunes, L.; Pepe, J . P. Selective
N-Dealkylation of Tertiary Amines with Vinyl Chloroformate:
an Improved Synthesis of Naloxone. Tetrahedron Lett. 1977,
1567-1570.
(15) Parmerter, S. M. The Coupling of Diazonium Salts with Aliphatic
Carbon Atoms. In Organic Reactions; Adams, R., Blatt, A. H.,
Boekelheide, V., Cope, A. C., Curtin, D. Y., McGrew; F. C.,
Niemann, C., Eds; J ohn Wiley & Sons, Inc.: New York, 1959;
Vol. 10, pp 3-142.
(16) Shvedov, V. I.; Altukhova, L. B.; Grinev, A. N. Monoarylhydra-
zones of Di- and Tricarbonyl Compounds in the Knorr Synthesis
of Pyrroles. Khimiya Beterot. Soed. 1972, 3, 342-344.
(17) Bhatt, M. V.; Kulkarni, S. U. Cleavage of Ethers. Synthesis 1983,
249-282.
(18) Baron, B. M.; Siegel, B. W.; Slone, A. L.; Harrison, B. L.;
Palfreyman, M. G.; Hurt, S. D. [3H]-5,7-Dichlorokynurenic Acid,
a Novel Radioligand Labels NMDA Receptor-associated Glycine
Binding Sites. Eur. J . Pharmacol. 1991, 206, 149-154.
(19) Cheng, Y.-C.; Prusoff, W. H. Relationship Between the Inhibition
Constant (KI) and the Concentration of Inhibitor which Causes
50 per cent Inhibition (I50) of an Enzymatic Reaction. Biochem.
Pharmacol. 1973, 22, 3099-3108.
(20) Tallarida R. J .; Murray R. B. Manual of Pharmacologic Calcula-
tions with Computer Programmes; Springer-Verlag: New York,
1987.
(21) Siegmund, E.; Cadmus, R.; Lu, G. A Method for Evaluating Both
Non-Narcotic and Narcotic Analgesics. Proc. Soc. Exp. Biol. Med.
1957, 95, 729-731.
(4) Moulin. D. E.; Max, M. B.; Kaiko, R. F.; Inturrisi, C. E.; Maggard,
J .; Yaksh, T. L.; Foley, K. M. The Analgesic Efficacy of Intrath-
ecal [D-Ala2, D-Leu5]enkephalin in Cancer Patients with Chronic
Pain. Pain 1985, 23, 213-221.
(5) Schwyzer, R. ACTH: a Short Introductory Review. Ann. N. Y.
Acad. Sci. 1977, 297, 3-26.
(6) Portoghese, P. S.; Sultana, M.; Takemori, A. E. Design of
Peptidomimetic δ Opioid Receptor Antagonists Using the Mes-
sage-Address Concept. J . Med. Chem. 1990, 33, 1714-1720.
(7) Dondio, G.; Clarke, G. D.; Giardina, G.; Petrillo, P.; Rapalli, L.;
Ronzoni, S.; Vecchietti, V. Potent and Selective Non-peptidic
Delta Opioid Ligands Based on The Novel Heterocycle-Con-
densed Octahydroisoquinoline Structure. Regul. Pept. 1994, 21,
43-44.
(8) Dondio, G.; Clarke, G. D.; Giardina, G.; Petrillo, P.; Petrone, G.;
Ronzoni, S.; Visentin, L.; Vecchietti, V. The Role of the “Spacer”
in the Octahydroisoquinoline Series: Discovery of SB 213698,
(22) Milne, G. M.; Twomey, T. M. The Analgetic Properties of
Piroxicam in Animals and Correlation with Experimentally
Determined Plasma Levels. Agents Actions 1980, 10, 31-37.
(23) Domino, E. F.; Gole, D.; Koek, W. Metaphit, a Proposed Phen-
cyclidine Receptor Acylator: Disruption of Mouse Motor Behav-
ior and Absence of PCP Antagonist Activity. J . Pharmacol. Exp.
Ther. 1987, 243, 95-100.
a
Non-peptidic, Potent and Selective Delta Opioid Agonist.
Analgesia 1995, 1 (4-6), 394-399.
(9) (a) Nagase, H.; Wakita, H.; Kawai, K.; Endoh, H.; Matsura, H.;
Tanaka, C.; Takezawa, Y. Syntheses of Non-peptidic Delta
Opioid Agonists and their Structure Activity Relationships. J pn.
J . Pharmacol. 1994, 64 (Suppl. I), 35. (b) Knapp, R. J .; Lands-
man, R.; Waite, S.; Malatynska, E.; Varga, E.; Haq, W.; Hruby,
V. J .; Roeske, W. R.; Nagase, H.; Yamamura, H. I. Properties of
TAN-67, a Nonpeptidic δ-Opioid Receptor Agonist, at Cloned
Human δ- and µ-Opioid Receptors. Eur. J . Pharmacol. 1995, 291,
129-134.
(24) Locke, K. W.; Dunn, R. W.; Hubbard, J . W.; Vanselouse, C. L.;
Cornfeltd, M.; Fielding, S.; Strupczewski, J . T. HP 818:
A
Centrally Acting Analgesic with Neuroleptic Properties. Drug
Dev. Res. 1990, 19, 239-256.
(25) Finney, D. J . Probit Analysis, 3rd ed.; Cambridge University
Press: Cambridge, UK, 1971.
(10) Chang, K. J .; Rigdon, G. C.; Howard, J . L.; McNutt, R. W. A
Novel, Potent and Selective Non-peptide Delta-receptor Agonist
BW 373U86. J . Pharmacol. Exp. Ther. 1993, 267, 852-857.
(11) 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. Probes for Narcotic Receptor Mediated Phenom-
ena. 19. Synthesis of (+)-4-[(RR)-R((2S,5R)-4-allyl-2,5-dimethyl-
1-piperazinyl)-3-methoxybenzyl]-N,N-diethylbenzamide (SNC 80):
a Highly Selective, Nonpeptide δ Opioid Receptor Agonist. J .
Med. Chem. 1994, 37, 2125-2128.
(26) Chang, G.; Guida, W. C.; Still, W. C. An Internal Coordinate
Monte Carlo Method for Searching Conformational Space. J . Am.
Chem. Soc. 1989, 111, 4379-4386.
(27) The crystallographic data were not suitable for verifying the
absolute configuration of (-)-5b since the anomalous scattering
signal was quite weak. However, its absolute configuration may
be assigned as 4aS,8aR because the starting material for its
synthesis was the same enantiomerically pure ketone (-)-1b
used for the synthesis of (+)-3.
(12) Dondio, G.; Ronzoni, S. International Patent Application WO95/
04734 (16.02.95); Chem. Abstr. 1995, 122, 314563j.
J M9608218