Discovery of an opioid κ receptor selective pure antagonist from a library of N-substituted 4β-methyl-5-(3-hydroxyphenyl)morphans

Article abstract of DOI:10.1021/jm020084h

A library of compounds biased toward opioid receptor antagonist activity was prepared by incorporating N-phenylpropyl-4β-methyl-5-(3-hydroxyphenyl)morphans as the core scaffold using simultaneous solution phase synthetic methodology. From this library, N-

Full text of DOI:10.1021/jm020084h

3524
J . Med. Chem. 2002, 45, 3524-3530
Discover y of a n Op ioid K Recep tor Selective P u r e An ta gon ist fr om a Libr a r y of
N-Su bstitu ted 4â-Meth yl-5-(3-h yd r oxyp h en yl)m or p h a n s
J ames B. Thomas,† Robert N. Atkinson,† Nivedita Namdev,† Richard B. Rothman,‡ Kenneth M. Gigstad,†
Scott E. Fix,† S. Wayne Mascarella,† J ason P. Burgess,† N. Ariane Vinson,† Heng Xu,‡ Christina M. Dersch,‡
Buddy E. Cantrell,§ Dennis M. Zimmerman,§ and F. Ivy Carroll*,†
Chemistry and Life Sciences, Research Triangle Institute, Research Triangle Park, North Carolina 27709, Clinical
Psychopharmacology Section, NIDA Addiction Research Section, P.O. Box 5180, Building C, 4980 Eastern Avenue,
Baltimore, Maryland 21224, and Lilly Research Laboratories, Eli Lilly and Company, Lilly Corporate Center,
Indianapolis, Indiana 4628
Received February 20, 2002
A library of compounds biased toward opioid receptor antagonist activity was prepared by
incorporating N-phenylpropyl-4â-methyl-5-(3-hydroxyphenyl)morphans as the core scaffold
using simultaneous solution phase synthetic methodology. From this library, N-phenylpropyl-
4â-methyl-5-(3-hydroxyphenyl)-7R-[3-(1-piperidinyl)propanamido]morphan [(-)-3b] was identi-
fied as the first potent and selective κ opioid receptor antagonist from the 5-phenylmorphan
class of opioids.
Ch a r t 1
In tr od u ction
Since the discovery of three distinct opioid receptors
(µ, δ, and κ), researchers studying the underlying
mechanisms of opiate addiction have sought highly
potent and receptor subtype selective antagonists.1
While many agonist structures have been discovered for
the opioid receptor system, very few structures display-
ing potent pure antagonist activity have been identi-
fied.2 The 5-(3-hydroxyphenyl)morphans such as 1a
(Chart 1) are one class of compounds that has produced
both agonists and antagonists for the opioid receptors.3,4
These compounds are structurally similar to the trans-
3,4-dimethyl-(3-hydroxyphenyl)piperidine class of opioid
antagonists (2a ,b) with the exception that they are
conformationally rigid as compared to the phenylpip-
eridines due to the addition of a bridging ring. While
potent antagonists based on the 5-(3-hydroxyphenyl)-
morphan scaffold remained an elusive goal for many
years, we recently demonstrated that N-phenethyl-9â-
methyl-5-(3-hydroxyphenyl)morphan (1b) is a highly
potent opioid receptor antagonist that is pharmacologi-
cally very similar to N-phenethyl-trans-3,4-dimethyl-
(3-hydroxyphenyl)piperidine (2a ).5
In a previous paper, we reported the discovery of 4b
from the library of compounds based on general struc-
ture 4a (Chart 1).6 Subsequent structure-activity re-
lationships (SAR) based on this screening hit led to the
discovery of the highly potent and selective κ opioid
antagonist J DTic (4c).7 The discovery that 1b possessed
potent antagonist activity at the µ, δ, and κ opioid
receptor similar to that available in the 4-phenylpip-
eridine series suggested that libraries of compounds (3a )
where R3 contained an amino group located at different
distances relative to the rigid morphan structure might
act as a κ address moiety and thus bias the library for
the opioid κ receptor.6 As illustrated by general struc-
tures 3a and 4a , the libraries differ by the placement
of their diversity elements relative to the core antagonist
ligand. Thus, if the trans-3,4-dimethyl-(3-hydroxyphe-
nyl)piperidine substructure, common to both 3a and 4a ,
associates with a similar antagonist domain, then the
diversity elements affixed to these scaffolds can occupy
different regions of the opioid receptor. The use of this
dual scaffold strategy was undertaken to provide a more
* To whom correspondence should be addressed. Tel: 919 541-6679.
Fax: 919 541-8868. E-mail: fic@rti.org.
† Research Triangle Institute.
‡ NIDA Addiction Research Section.
§ Eli Lilly and Company.
10.1021/jm020084h CCC: $22.00 © 2002 American Chemical Society
Published on Web 07/04/2002

Opioid κ Receptor Selective Pure Antagonist
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 16 3525
Sch em e 1a
a
Reagents: (a) n-BuLi and then ClCH2C(OCH2OCH3)CH2 and then 1 N HCl, MeOH. (b) NaBH4, EtOH. (c) Benzoyl chloride, pyridine,
CH2Cl2, catalytic 4-N,N-dimethylaminopyridine. (d) 1-Chloroethylchloroformate, dichloroethane. (e) LiOH, H2O, CH3OH. (f) NaBH(OAc)3,
Ph(CH2)2CHO. (g) (COCl)2, DMSO, Et3N. (h) HONH2‚HCl, EtOH. (i) Na, i-PrOH, toluene. (j) 48% HBr, glacial HOAc, reflux. (k) BOP,
Et3N, C5H10N(CH2)2CO2H.
thorough exploration for complimentary binding sites
in the vicinity of the opioid antagonist binding domain
as compared to that provided by a single scaffold.
In this paper, we report that screening of a library
based on general structure 3a led to the discovery of
3b, the first 5-(3-hydroxyphenyl)morphan-based com-
pound to display selective antagonist activity for the κ
receptor over both the µ and δ opioid receptors (Chart
1). Additional studies with (-)- and (+)-3b showed that
the (-)-enantiomer [(-)-3b] was responsible for the
observed pharmacological activity.
with 1-chloroethyl chloroformate to remove the N-
methyl group by carbamate formation, hydrolysis of
both the newly formed carbamate and the ester groups
using lithium hydroxide in aqueous refluxing methanol,
and finally reductive alkylation with hydrocinnamal-
dehyde and sodium triacetoxyborohydride. Swern oxida-
tion of 8 followed by conversion of the carbonyl to the
oxime with hydroxylamine hydrochloride provided the
oxime intermediate 9 in 83% yield from 8. Reduction of
9 using sodium and 2-propanol in refluxing toluene gave
10 in 54% yield.10 Removal of the isopropyl group in 10
using 48% hydrobromic and glacial acetic acid then
provided the 7-amino-derivatized scaffold (11) in 84%
yield.
Ch em istr y
The synthesis of (+)- and (-)-3b and (-)-3c-e utilized
the same synthetic methods developed for racemic 68
but started with either (+)- or (-)-5.9 To simplify
Scheme 1, only the synthesis of (-)-3b is illustrated.
As previously reported for racemic 5, treatment of
optically pure 5 with n-butyllithium followed by addition
to a solution of Okahara’s reagent followed by concomi-
tant deprotection and acid-catalyzed cyclization gave the
7-oxo derivative 6 as a single diastereomer and in this
case a single enantiomer. Conversion of 6 to 8 was
accomplished in 65% overall yield without purification
of intermediates by reducing the carbonyl group in 6
with sodium borohydride to give 7 followed by protection
of the hydroxyl group as the benzoate ester, treatment
Coupling of appropriately substituted carboxylic acid
derivatives to 11 provided the desired library of com-
pounds. Briefly, a mixture of the appropriate acid
derivative, the phenylmorphan scaffold 11, and the
coupling reagent, benzotriazol-1-yloxy-tris-(dimethy-
lamino)phosphonium hexafluorophosphate (BOP, Cas-
tro’s reagent), were combined in tetrahydrofuran (THF)/
triethylamine in parallel using an Argonaut Quest 210.
For (-)-3b, a 1-piperidinopropionic acid side chain was
used to give (-)-3b in 85% yield. For those acid
derivatives containing a Boc-protected guanidino group,
as in 3d ,e, a trifluoroacetic acid deprotection step was
added.

3526 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 16
Thomas et al.
Ta ble 1. Radioligand Binding Data for Test Compounds in µ,
δ, and κ Assays Using Opioid Receptor Obtained from Brain
Tissues (µ and δ, Rat; κ, Guinea Pig)
Ki (nM ( SD)
compd [3H]DAMGOa [3H]DADLEb [3H]U69,593c µ/κ
δ/κ
0.27
0.1
>790
61
122
2
132
79
(+)-11
(-)-11
(-)-3b
(+)-3b
(-)-3c
(-)-3d
(-)-3e
207 ( 24
3.3 ( 0.3
246 ( 22
893 ( 148
122 ( 5.6
4.3 ( 0.7
36 ( 5
0.23
0.03
34
22
5
0.2
14.7 ( 1.3
147 ( 9.8 >3400
775 ( 55
57 ( 4.4
5.2 ( 0.3
87 ( 6.3
2184 ( 93
1457 ( 113
12 ( 0.65
56.5 ( 3.4 23.3 ( 3
1744 ( 98
13.2 ( 1.7
7
nor-BNI, 65.0 ( 5.6
13
86 ( 7.2 1.09 ( 0.14 60
a
[3H]DAMGO [(D-Ala2, MePhe4, Glyol5)enkephalin]. Tritiated
ligand selective for µ opioid receptor. [3H]DADLE [(D-Ala2,
b
F igu r e 1. Observation of strong through-space interactions
between the 3R and 7â protons provided the relative stereo-
chemistry for the 7-amino group in intermediate 11 (R )
(CH2)3Ph).
D-Leu5)enkephalin]. Tritiated ligand selective for δ opioid receptor.
c [3H]U69,593 {[3H](5R,7R,8â)-(-)-N-methyl-N-[7-(1-pyrrolidinyl)-
1-oxaspiro[4.5]dec-8-yl]benzeneacetamide}. Tritiated ligand selec-
tive for κ opioid receptor.
Ta ble 2. Inhibition of Agonist-Stimulated [35S]GTP-γ-S
Binding by (-)-3b in Cloned Human µ, δ, and κ Opioid
Receptors
The relative stereochemical assignment for the 7-ami-
no group in the key intermediate 11 was based on the
observation of a strong through-space interaction in the
ROESY spectrum between the 3R11 and 7â protons
(Figure 1), which places the amino group on the endo
face of the ring system. The relative positions of the
protons critical to this assertion (3R and 7â) were
established through a series of steps beginning with a
full assignment of all protons and carbons using gradi-
ent-enhanced correlation spectroscopy (COSY), HSQC,
and HMBC spectra. From this point, the identity of the
H3R proton was determined by its COSY interaction
with one other proton (H4R) that was readily identified
by its COSY interaction with the 4â-methyl group. The
7â proton was assigned based on its chemical shift and
was differentiated from H1 by its large 1,2-diaxial
coupling constants with H6R and H8R. The protons on
C8 and C6 were in turn readily distinguishable since
both C8 protons showed a COSY correlation with H1
while H6R gave a three bond correlation with C4 in the
HMBC spectrum. The chemical shift of both C9 protons
was located by observation of a COSY correlation
between H9R and H1.
The relative stereochemical assignments were then
available from the ROESY spectrum, which indicated
an exo position for H3â due to its strong correlation with
the 4â-methyl group. The methyl group also showed a
strong interaction with H9â. Likewise, a mutual cor-
relation between H9R, H6R, and H8R indicates that
these protons occupy the endo face on the opposite side
of the azabicyclo ring system. Because H3â occupies an
endo position, H3R must occupy the endo face of the ring
system. Given this, the strong interaction observed
between H3R and H7â implies that the amino group
must reside on the endo face along with H6â and H8â.
Ke (nM ( SD)
compd
DAMGOa Cl-DPDPEb
U69,593c
µ/κ δ/κ
(-)-3b
nor-BNI, 13
41.7 ( 3.74 NDd
19 ( 1.52 4.6 ( 0.39
0.24 ( 0.01
175
0.04 ( 0.003 475 115
a
DAMGO [(D-Ala2, MePhe4, Glyol5)enkephalin]. Agonist selec-
tive for µ opioid receptor. Cl-DPDPE [Phe(p-Cl)4 cyclo[D-Pen2,
b
D-Pen5]enkephalin]. Agonist selective for δ opioid receptor. c U69,593
{[3H](5R,7R,8â)-(-)-N-methyl-N-[7-(1-pyrrolidinyl)-1-oxaspiro[4.5]-
dec-8-yl]benzeneacetamide}. Agonist selective for κ opioid receptor.
d
The Ke for the δ receptor was not determined because of the low
binding affinity at this site (see Table 1).
procedures (Table 1).12 The tissues used in the binding
assay included rat brain (µ and δ receptors) and guinea
pig brain (κ receptors). Measures of antagonism were
obtained by monitoring the test compounds’ ability to
inhibit stimulation of [35S]GTP-γ-S binding produced by
the selective agonists, (D-Ala2, MePhe4, Glyol5)enkephalin
(DAMGO, µ receptor), Phe(p-Cl)4 cyclo[D-Pen2, D-Pen5]-
enkephalin (Cl-DPDPE, δ receptor), and 5R,7R,8â-(-)-
N-methyl-N-[7-(1-pyrrolidinyl)-1-oxaspiro[4.5]dec-8-yl]-
benzeneacetamide (U69,593, κ receptor) in cloned human
opioid receptors (Table 2) transfected into Chinese
hamster ovary cells.13,14
Resu lts a n d Discu ssion
The initial library screening results showed that few
compounds possessed high affinity for the κ receptor.
Indeed, only 3b gave greater than 70% inhibition of [3H]-
U69,593 binding at a concentration of 100 nM. Overall,
the results appeared to be lower than one would expect
based on our previous observations with the 5-(3-
hydroxyphenyl)morphan antagonists.5 As opioid recep-
tors are well-known to distinguish between enanti-
omers, we speculated that part of the lower affinity
might arise as a consequence of the addition of a
7-amido substituent to the phenylmorphan nucleus as
this would significantly increase the size of such deriva-
tives relative to unsubstituted phenylmorphans. Such
increases in size could in turn provide a clearer distinc-
tion between optical antipodes. These concerns were
validated by obtaining the binding data for the optical
isomers of 11 (Table 1), which revealed that (-)-11 with
Ki values of 3.3, 14.7, and 122 nM for the µ, δ, and κ,
Biologica l
The screening of compound libraries was performed
at a single concentration of test compound (100 nM) as
previously reported.6 Compounds showing >70% inhibi-
tion of radioligand binding and their close analogues
were resynthesized and submitted for determination of
their Ki value at each of the opioid receptors. The
binding affinities of (+)- and (-)-3b as well as (-)-3c-e
and the two optical isomers of 11 were determined using
competitive binding assays following previously reported

Opioid κ Receptor Selective Pure Antagonist
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 16 3527
Ch a r t 2
respectively, possesses far greater affinity at all opioid
receptors relative to (+)-11 with its Ki values of 207,
247, and 893 nM. These figures indicate a 63-, 17-, and
7-fold preference for the (-)-enantiomer by the µ, δ, and
κ receptors.
The addition of a 7R-amido-3-(1-piperidinyl)propionyl
side chain provided further evidence for a separation
of activities between enantiomers. Unlike (-)-11, how-
ever, (-)-3b possessed a high affinity for the opioid κ
receptor (Ki value of 4.3 vs 122 nM) and a lower affinity
for the µ receptor (Ki value of 147 vs 3.3 nM) and δ
receptor (Ki value of >3400 vs 147 nM). Taken together,
the comparison of the binding characteristics for com-
pounds (-)-3b and (-)-11 indicates that the 3-(1-
piperidinyl)propamido side chain reduces both µ and δ
affinity while concomitantly enhancing affinity at the
κ receptor. As compared with the standard κ antagonist
nor-binaltorphimine (nor-BNI, 13), compound (-)-3b
demonstrates roughly one-half of the µ vs κ selectivity
(34- vs 60-fold) but a decidedly improved κ vs δ selectiv-
ity (>790- vs 79-fold).
Several other library compounds that showed good
affinity in screening or were structurally similar to (-)-
3b were resynthesized in optically pure form using (-)-
11 for direct comparison with compound (-)-3b. Com-
pound (-)-3c, for example, differs from (-)-3b by an
additional methylene group in the chain connecting the
amino and carbonyl groups as well as possessing a
dimethylamino group on the pendant carbon rather
than a piperidine ring. In the binding experiments,
compound (-)-3c showed about 3-fold less affinity for
the κ receptor and about 3-fold higher affinity for the µ
receptor relative to (-)-3b. Neither compound showed
appreciable affinity for the δ receptor. Thus, the overall
effect of the structural changes in going from 3b to 3c
is elimination of the modest µ vs κ selectivity displayed
by (-)-3b.
Compound (-)-3d differs from (-)-3b by replacement
of the piperidine ring for a guanidine group while
maintaining the same carbon chain length. This struc-
tural difference was found to significantly alter not only
the receptor affinity but also the receptor preference of
the resulting ligand. As compared to (-)-3b, (-)-3d
favors the µ receptor over the κ receptor with a Ki value
of 5.2 vs the 147 nM value found for (-)-3b at the µ
receptor. This functional group change also provides a
dramatic increase in affinity for the δ opioid receptor
with a δ Ki value of 56.5 vs >3400 nM for (-)-3b. In
the κ receptor assay, (-)-3d shows diminished affinity
relative to (-)-3b (23.3 vs 4.3 nM). Altogether, the
replacement of piperidine with a guanidino group leads
to increases in affinity for both µ and δ opioid receptors
and a slight decrease in affinity for the κ receptor.
neither very potent nor selective for the κ receptor vs
the µ opioid receptor.
Compound (-)-3b was also assayed for its antagonist
activity by measuring its ability to inhibit agonist-
stimulated [35S]GTP-γ-S binding in cloned human opioid
receptors (Table 2). In this assay, (-)-3b retains the κ
receptor as its principle site of action, and its Ke value
showed a 18-fold increase relative to its Ki value in the
binding assay (0.24 vs 4.3 nM). In the µ receptor assay,
(-)-3b demonstrated a 5-fold decrease in its Ke value,
which together provided a significant increase in its µ
vs κ receptor selectivity (175-fold) relative to the binding
assay. The Ke value at the δ receptor was not measured
due to the very low affinity of (-)-3b for the δ opioid
receptor. Overall, the data from the functional assay
demonstrate that the novel 5-(3-hydroxyphenyl)mor-
phan-based antagonist (-)-3b is both potent and selec-
tive for the κ opioid receptor.
From a historical perspective, the pharmacological
behavior of compound (-)-3b parallels previous studies
that lead to the discovery of κ opioid selective antago-
nists. In a broad sense, the addition of pendant amino
groups to nonselective opioid receptor antagonist scaf-
folds to serve as κ selective address units has been
demonstrated by Portoghese as illustrated in Chart 2.
In this case, the nonselective antagonist naltrexone (12)
was modified to possess a pendant amine (denoted by
an asterisk in the chart) that served to promote affinity
and selectivity of the resulting ligand, nor-binaltor-
phimine (nor-BNI, 13), for the κ receptor.15-17 The
discovery of the structurally simplified κ antagonist 5′-
guanidino-naltrindole (GNTI, 14) followed the same
concept but in this case utilized a guanidine rather than
a tertiary amine as the κ address element.18
Compound (-)-3e like (-)-3d has a guanidino group
in its side chain rather than piperidine but differs from
(-)-3d in the length of its side chain. The addition of a
single methylene spacer provides a 2-fold increase in κ
affinity (13 vs 23 nM) but losses of receptor affinity in
both the µ (87 vs 5.2 nM) and δ receptor assays (1744
vs 56 nM) relative to (-)-3d . In this respect, (-)-3e
behaves more like the tertiary amine-bearing ligand (-)-
3c rather than the guanidino-bearing ligand (-)-3d . As
found earlier for (-)-3c, however, compound (-)-3e is
In a more closely related example, we recently dem-
onstrated that addition of a methylene amino group to
antagonist 4b gave a highly potent κ selective antago-
nist J DTic (4c). Taken together with our previous
observations of potent but nonselective opioid antagonist
activity in 1b, the discovery of κ selective antagonist
activity in (-)-3b appears to be consistent with the

3528 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 16
Thomas et al.
ether. The ether was then washed (1 N NaOH, H2O), and the
solvent was removed under vacuum. The resulting residue of
product and water was dissolved in MeOH (30 mL/g), and
nitrogen was bubbled through the solution for 5 min. To this
was added concentrated HCl (2 mL/g), and the mixture was
allowed to stand at room temperature until the reaction was
complete as indicated by TLC [TLC condition: SiO2; elution
with 50% (80% CHCl3:18% CH3OH:2% NH4OH) in CHCl3;
detection: 5% phosphomolybdic acid in ethanol]. To this
mixture was added 50% NaOH to adjust the pH to ∼10, and
the methanol was removed under aspirator vacuum. The
aqueous residue was then extracted several times with 3:1
(methylene chloride:THF). The organic extracts were combined
and washed twice with water and once with brine, dried over
sodium sulfate, and evaporated to an oil. This material was
purified by flash chromatography on silica gel using 15-25%
(80% CHCl3:18% CH3OH:2% NH4OH) in CHCl3 to give 6 in
concepts of message and address substructures de-
scribed for previously discovered κ antagonists. In this
case, the N-phenylpropyl-4â-methyl-5-(3-hydroxyphen-
yl)morphan scaffold could be viewed as the opioid
“message” fragment while the 3-(1-piperidinyl) group
might be considered as the κ address moiety.19,20
However, one aspect of the results demonstrated here
is significantly different from the observations made by
Portoghese and is worthy of note. This relates to the
need for rigid scaffolds in the oxymorphone-based
antagonists such as 13 or 14 required to properly align
the pendant amine address element with an anionic
residue specific to the κ receptor. Such a requirement
has been the subject of earlier reports in oxymorphone-
based antagonist research.16,17 Clearly, such rigid scaf-
folds are absent in both the phenylmorphan-based
compound presented herein and the phenylpiperidine-
based compound 4c reported earlier. While the source
of these differences has not yet been ascertained, a
comparison of the structures of all of the known classes
of κ antagonist suggests that the need for a rigid scaffold
may be a characteristic particular to the oxymorphones.
1
70% yield from 5. H NMR: 7.24 (t, 1H, J ) 7.5 Hz), 6.77 (m,
3H), 4.55 (m, 1H), 3.49 (s, 1H), 2.91 (dd, 2H, J ) 17 and 16.5
Hz), 2.60 (m, 2H), 2.35 (m, 5H), 2.05 (m, 3H), 1.35 (m, 6H),
0.78 (d, 3H, J ) 6.8 Hz).
(-)-(1R ,4S ,5S )-5-[3-(1-Me t h yle t h oxy)p h e n yl]-2,4-d i-
m eth yl-2-a za bicyclo[3.3.1]n on a n -7-ol (7). To a solution of
6 (1 equiv) dissolved in absolute ethanol (7 mL/g of 6) was
added solid sodium borohydride (1 equiv) slowly over 10 min.
This mixture was allowed to stir at room temperature for 24
h after which time the ethanol was removed under aspirator
vacuum and the residue was carefully dissolved in 1.0 N HCl
(6 mL/g of 6). This solution was washed twice with ether (3
mL/g of 6 for each wash), and then, the aqueous solution was
made basic with 50% NaOH solution (pH 14), and the ether
layers were discarded. This solution was then saturated with
sodium chloride and extracted five times with 3:1 methylene
chloride:THF (3 mL/g of 6 for each extraction), and the
combined organic layers were dried over magnesium sulfate,
and the solvent was removed under aspirator vacuum to
provide 7 as a yellow oil in 95% yield from 6 as a mixture of
7-hydroxy diastereomers. This material was used in the next
Con clu sion s
This study shows that the addition of a 4â-methyl
group to N-substituted-5-(3-hydroxyphenyl)morphans
results in compounds that show pure opioid antagonist
activity. More important, the study shows that the
addition of a 7R-3-(1-piperidinyl)propanamido group to
the 4â-methyl-5-(3-hydroxyphenyl)morphan structure
provides a compound [(-)-3b], which possesses potent
and selective antagonist activity for the opioid κ recep-
tor. To our knowledge, this is the first demonstration
of κ selective antagonist activity in the 5-phenylmorphan
series.
1
step without purification. H NMR (CDCl3): δ 0.46-0.48 (d,
3H, J ) 6.90 Hz), 1.32-1.34 (d, 6H, J ) 6.02 Hz), 1.47-1.58
(m, 2H), 1.87-1.93 (dd, 2H, J ) 14.16, 5.09 Hz), 2.27-2.47
(m, 8H), 2.69-2.75 (dd, 1H, J ) 11.64, 5.26 Hz), 3.09 (br s,
1H), 4.02-4.05 (t, 1H, J ) 4.83 Hz), 4.47-4.59 (septet, 1H, J
) 6.07 Hz), 5.30 (br s, 1H), 6.68-6.81 (m, 3H), 7.16-7.62 (m,
1H).
Exp er im en ta l Section
Elemental analyses were obtained by Atlantic Microlabs,
Inc. and are within (0.4% of the calculated values. All optical
rotations were determined at the sodium D-line using a
Rudolph Research Autopol III polarimeter (1 dm cell). 1H
nuclear magnetic resonance (NMR) spectra were determined
on a Bruker Avance spectrometer operating at 300 MHz for
proton using tetramethylsilane as an internal standard. Silica
gel 60 (230-400 mesh) was used for all column chromatogra-
phy. Mass spectral data was obtained using a Finnegan LCQ
electrospray mass spectrometer in positive ion mode at atmo-
spheric pressure. All reactions were followed by thin-layer
chromatography (TLC) using Whatman silica gel 60 TLC
plates and were visualized by UV, charring using 5% phos-
phomolybdic acid in ethanol and/or exposure to iodine vapor.
All solvents were reagent grade. THF was dried over sodium
benzophenone ketyl and distilled prior to use. Methylene
chloride was distilled from calcium hydride prior to use.
(-)-(1R ,4S ,5S )-5-[3-(1-Me t h yle t h oxy)p h e n yl]-2,4-d i-
m eth yl-2-a za bicyclo[3.3.1]n on a n -7-on e (6). To a solution
of (S)-1,2,3,6-tetrahydro-1,3-dimethyl-4-[3-(1-methylethoxy)-
phenyl]pyridine (5) (1 equiv) dissolved in THF (20 mL/g) and
cooled to -10 °C was added n-butyllithium (1.6 M in hexanes)
slowly until a red color was maintained followed by an addition
of 1.1 equiv. This material was stirred for 1 h at -10 °C and
then cannulated quickly into a solution of Okahara’s reagent
(distilled to high purity) in THF (15 mL/g, 1.1 equiv) at -78
°C followed by stirring for 2 h. The temperature should be kept
below -30 °C during cannulation. This material was then
poured into 2 N HCl and extracted twice with ethyl ether. The
aqueous layer was allowed to stand for 15 min followed by
addition of 50% NaOH to pH 14 and extraction (3×) with ethyl
(-)-[(1R,4S,5S,7R)-5-[3-(1-Meth yleth oxy)ph en yl]-4-m eth -
yl-2-(3-p h en ylp r op yl)-2-a za bicyclo[3.3.1]n on a n -7-ol (8).
To a solution of 7 (1 equiv) in anhydrous methylene chloride
(35 mL/g of 15) at room temperature was added triethylamine
(1.1 equiv), a small amount of N,N-dimethylaminopyridine (0.5
equiv), and benzoyl chloride (2.0 equiv), and the resulting
mixture was stirred overnight under a nitrogen atmosphere.
Following this, the mixture was washed twice with 10% NaOH,
once with water and then dried over sodium sulfate, and the
solvent was removed under reduced pressure. The resulting
oil was not purified but carried directly to the next step. 1H
NMR (CDCl3): δ 8.18 (d), 8.02 (d), 7.56 (d), 7.46 (d), 7.20 (t),
6.75-6.69 (m), 5.32-5.28 (m), 4.56-4.52 (quintet), 3.65-3.50
(m), 3.30-3.15 (dd), 3.00 (s), 2.78-2.75 (q), 2.49 (s), 2.44 (s),
2.43-2.25 (m), 2.30 (s), 2.25-1.85 (m), 1.34-1.32 (d), 1.17-
1.12 (t), 0.76-0.74 (d).
To a solution of the intermediate oil (1 equiv) in anhydrous
1,2-dichloroethane (20 mL/g of intermediate) at reflux was
added 1-chloroethyl chloroformate (1.1 equiv) dropwise. The
resulting solution was heated under reflux for 2.5 h and then
cooled to room temperature. This mixture was then washed
with saturated bicarbonate solution and water, then the
organic layer was evaporated, and the resulting oil was
dissolved immediately in 1:1 methanol:water. To this was
added LiOH (1 g/g of intermediate) and then heated to reflux
until the reaction was complete as judged by TLC (∼2 h). After
it was cooled to room temperature, the methanol was removed
under aspirator vacuum and the remaining aqueous solution
was saturated with sodium chloride. This was then extracted

Opioid κ Receptor Selective Pure Antagonist
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 16 3529
with 3:1 CH2Cl2:THF until all of the product was removed,
and the combined extracts were washed once with water.
Removal of the solvent provided slightly impure material that
18% CH3OH:2% NH4OH) in CHCl3] to afford starting material
9 (18% recovered) and 10 (58%) as a yellow oil. 1H NMR
(CDCl3): δ 7.28-7.15 (m, 6H), 6.76-6.68 (m, 3H), 4.52 (sept.,
1H, J ) 6.1 Hz), 3.51 (m, 1H), 3.13 (m, 1H), 2.82 (m, 1H), 2.64
(m, 3H), 2.47 (m, 2H), 2.31 (m, 3H), 2.11 (m, 1H), 1.77 (m,
2H), 1.56 (m, 3H), 1.31 (d, 6H, J ) 6.0 Hz), 1.15 (m, 1H), 0.94
(m, 1H), 0.73 (d, 3H, J ) 6.9 Hz).
1
was not purified but carried directly to the next step. H NMR
(CDCl3): δ 0.48-0.65 (m, 3H), 1.32-1.34 (d, 6H), 1.54-1.72
(m, 2H), 1.97-2.12 (m, 4H), 2.58-2.64 (m, 1H), 3.45-3.59 (m,
3H), 3.91-4.02 (m, 1H), 4.47-4.68 (m, 1H), 6.68-6.78 (m, 3H),
7.17-7.35 (t, 1H).
(-)-3-[(1R,4S,5S,7R)-7-Am in o-4-m eth yl-2-(3-p h en ylp r o-
p yl)-2-a za bicyclo[3.3.1]n on -5-yl]p h en ol [(-)-11]. A solu-
tion of 10 (1 equiv) (2.53 g, 6.23 mmol) in glacial acetic acid (8
mL/g of 10) and 48% HBr (8 mL/g of 10) was heated to reflux
for 15 h. The reaction mixture was allowed to cool to room
temperature, added to ice (40 g/g of 10), and adjusted to pH
10 with 50% NaOH. The aqueous layer was extracted with
3:1 n-butanol:toluene (3 × 40 mL/g of 10), the organic layer
was collected and dried (Na2SO4), and the solvent was removed
under reduced pressure. The product was purified by flash
chromatography [50% (80% CHCl3:18% CH3OH:2% NH4OH)
To a solution of material from the previous step (1 equiv)
in anhydrous 1,2-dichloroethane (40 mL/g of intermediate) was
added hydrocinnamaldehyde (freshly opened, 1.2 equiv) and
NaBH(OAc)3 (1.2 equiv), and the resulting mixture was stirred
for 24 h. After this time, the resulting mixture was washed
with 1 N NaOH and the aqueous layer was back-extracted with
chloroform. The combined organic layers were dried over
sodium sulfate, and the solvent was removed at reduced
pressure to give crude 8 as a mixture of diastereomers. This
material was purified by flash chromatography on silica gel
to give 8 as a yellow oil in 65% yield from 7. 1H NMR (CDCl3):
δ 0.43-0.45 (d, 3H, J ) 6.82 Hz), 1.31-1.33 (d, 6H, J ) 6.03
Hz), 1.40-1.53 (m, 2H), 1.81-1.88 (m, 4-5 H), 2.27-2.70 (m,
10H), 3.14 (s, 1H), 4.06 (s, 1H), 4.51-4.55 (m, 1H), 6.08 (br s,
1H), 6.68-6.89 (m, 3H), 7.15-7.54 (m, 6H).
(-)-[(1R,4S,5S,7R)-5-[3-(1-Meth yleth oxy)ph en yl]-4-m eth -
yl-2-(3-phenylpropyl)-2-azabicyclo[3.3.1]nonan-7-one Oxime
(9). Dimethyl sulfoxide (6.6 equiv) in dry CH2Cl2 (3 mL/g of 8)
was added dropwise over 20 min to a solution of 2 M oxalyl
chloride (3 equiv) in CH2Cl2 at -78 °C. The reaction mixture
was allowed to warm to -20 °C. Maintaining a temperature
of -20 °C, 8 (1 equiv) in CH2Cl2 (4 mL/g of 8) was added
dropwise over 15 min to the reaction mixture. The reaction
was stirred for an additional 30 min and then quenched with
the careful addition of triethylamine (8 equiv). The reaction
mixture was allowed to warm to room temperature and
washed with saturated NaHCO3, the organic layer was col-
lected and dried (Na2SO4), and the solvent was removed under
reduced pressure. The crude product was purified by flash
chromatography [5-10% (80% CHCl3:18% CH3OH:2% NH4-
OH) in CH2Cl2] to afford the 7-oxo derivative of 8 (91%) as a
yellow oil. 1H NMR (CDCl3): δ 7.23 (m, 6H), 6.77 (m, 3H), 4.54
(sept., 1H, J ) 6.1 Hz), 3.47 (br, 1H), 2.83 (m, 2H), 2.68-2.52
(m, 5H), 2.43 (t, 2H, J ) 6.9 Hz), 2.09-1.95 (m, 3H), 1.74 (m,
3H), 1.33 (d, 6H, J ) 6.0 Hz), 0.79 (d, 3H, J ) 6.8 Hz).
20
in CH2Cl2] to afford (-)-11 (84%) as a yellow oil; [R]D -40.8°
(c 1.04, CHCl3). 1H NMR (CDCl3): δ 7.27-7.07 (m, 6H), 6.65-
6.58 (m, 3H), 4.33 (br, 2H), 3.54 (br, 1H), 2.79 (m, 1H), 2.66-
2.53 (m, 3H), 2.46 (t, 2H, J ) 7.0 Hz), 2.31 (m, 3H), 2.04 (br,
1H), 1.77 (t, 2H, J ) 7.2), 1.53 (m, 1H), 1.14 (m, 1H), 0.98 (m,
1H), 0.70 (d, 3H, J ) 6.9 Hz). 13C NMR (CDCl3): δ 157.5, 151.8,
142.8, 129.7, 128.7, 126.1, 116.7, 113.9, 113.1, 56.3, 54.7, 53.9,
52.1, 47.2, 40.8, 38.1, 33.8, 33.0, 29.4, 19.1. Anal. (C24H32N2O)
C, H, N.
(+)-3-[(1R,4S,5S,7R)-7-Am in o-4-m eth yl-2-(3-p h en ylp r o-
p yl)-2-a za bicyclo[3.3.1]n on -5-yl]p h en ol [(+)-11]. This com-
pound was prepared according to the same procedure as
described above for (-)-3-[(1R,4S,5S,7R)-7-amino-4-methyl-2-
(3-phenylpropyl)-2-azabicyclo[3.3.1]non-5-yl]phenol [(-)-11] with
the exception that the synthesis begins with (R)-1,2,3,6-
tetrahydro-1,3-dimethyl-4-[3-(1-methylethoxy)phenyl]pyri-
dine rather than (S)-1,2,3,6-tetrahydro-1,3-dimethyl-4-[3-(1-
methylethoxy)phenyl]pyridine. The product was obtained from
20
the phenol deprotection step in 83% yield as a yellow oil; [R]D
+39° (c 1.0, CHCl3). The NMR spectra were identical to those
obtained for (-)-11. Anal. (C24H32N2O) C, H, N.
(-)-N-[(1R,4S,5S,7R)-5-(3-Hyd r oxy)p h en yl-4-m eth yl-2-
(3-p h en ylp r op yl)-2-a za bicyclo[3.3.1]n on -7-yl]-3-(1-p ip e-
r id in yl)p r op a n a m id e [(-)-3b]. To a solution of 11 (1 equiv)
was added BOP reagent (1.1 equiv), 1-piperidinepropionic acid
(2 equiv), and triethylamine (5 equiv) in dry THF (250 mL/g
of 11). The reaction mixture was stirred under N2 at room
temperature for 4 h. The mixture was diluted with Et2O (20
mL) and washed with saturated NaHCO3, followed by water.
The organic layers were collected and dried (Na2SO4), and the
solvent was removed under reduced pressure. The product was
purified by flash chromatography [33% (80% CHCl3:18% CH3-
OH:2% NH4OH) in CHCl3] to afford (-)-3b (85%) as an off-
The ketone from the previous step (1 equiv) and hydroxy-
lamine hydrochloride (5 equiv) in EtOH (absolute, 17 mL/g of
ketone) were heated under reflux for 3 h. The reaction mixture
was allowed to cool to room temperature, and the ethanol was
removed under reduced pressure. The oil thus obtained was
dissolved in 2 M NaOH (17 mL/g of ketone), and the product
was extracted with 3:1 CH2Cl2:THF (4 × 10 mL/g of ketone).
The organic layers were collected and dried (Na2SO4), and the
solvent was removed under reduced pressure. The product
obtained was purified by flash chromatography [5-10% (80%
CHCl3:18% CH3OH:2% NH4OH) in CH2Cl2] to afford 9 (90%)
1
white foam. H NMR (CDCl3): δ 8.70 (br, 1H), 7.27-7.13 (m,
6H), 6.90-6.67 (m, 3H), 4.64 (m, 1H), 3.22 (br, 1H), 3.05 (m,
1H), 2.80-2.02 (m, 14H), 1.82-1.34 (m, 10H), 1.31-0.97 (m,
20
1
4H), 0.72 (d, 3H, J ) 6.9 Hz); [R]D -50.9° (c 1.42, CH3OH).
as a yellow oil. H NMR (CDCl3): δ 10.09 (br, 1H), 7.26-7.13
Anal. (C32H45N3O2) C, H, N.
(m, 6H), 6.88-6.72 (m, 3H), 4.54 (m, 1H), 3.63 (d, 1H, J ) 17
Hz), 3.29 (br, 1H), 2.94-2.85 (m, 2H), 2.69-2.41 (m, 5H), 2.29
(d, 1H, J ) 15.9 Hz), 2.04-1.65 (m, 6H), 1.33 (d, 6H, J ) 6.0
Hz), 0.76 (d, 3H, J ) 6.9 Hz).
(+)-N-[(1S,4R,5R,7S)-5-(3-Hyd r oxy)p h en yl-4-m eth yl-2-
(3-p h en ylp r op yl)-2-a za bicyclo[3.3.1]n on -7-yl]-3-(1-p ip e-
r id in yl)p r op a n a m id e [(+)-3b]. This compound was prepared
in 88% yield according to the method described for (-)-3b
starting from (+)-3-[(1S,4R,5R,7S)-7-amino-4-methyl-2-(3-phen-
ylpropyl)-2-azabicyclo[3.3.1]non-5-yl]phenol [(+)-11]. The method
of purification and 1H NMR (CDCl3) were identical to those
(-)-[(1R,4S,5S,7R)-5-[3-(1-Meth yleth oxy)ph en yl]-4-m eth -
y l-2-(3-p h e n y lp r o p y l)-2-a z a b i c y c lo [3.3.1]n o n a n -7-
a m in e (10). Compound 9 (1 equiv) (5.51 g, 13.1 mmol) in a
minimum of dry 2-propanol was added dropwise over 1 h to a
refluxing mixture of dry toluene (35 mL/g of 9) and sodium
(150 equiv). After complete addition of oxime, two portions of
2-propanol (23 mL/g of 9) were added dropwise over 30 min.
The reaction mixture was heated to reflux until all of the
sodium was consumed. The reaction mixture was allowed to
cool to 50 °C and then quenched by careful addition of water
(135 mL/g of 9). The toluene layer was separated, and the
aqueous layer was extracted with CHCl3 (4 × 90 mL/g of 9).
The organic layers were combined and dried (Na2SO4), and
the solvent was removed under reduced pressure. The product
was purified by flash chromatography [25-50% (80% CHCl3:
20
found for the antipode; [R]D +49.8° (c 0.275, CH3OH). Anal.
(C32H45N3O2) C, H, N.
(-)-N-[(1S,4R,5R,7S)-5-(3-H yd r oxy)p h en yl-4-m et h yl-
2-(3-p h en ylp r op yl)-2-a za b icyclo[3.3.1]n on -7-yl]-4-(N,N-
d im eth yla m in o)bu ta n a m id e [(-)-3c]. This compound was
prepared in 89% yield according to the method described for
(-)-3b starting from (+)-11 and 4-(dimethylamino)butyric acid
hydrochloride. The product was obtained as an off-white foam
following purification by flash chromatography [33% (80%
1
CHCl3:18% CH3OH:2% NH4OH) in CHCl3]. H NMR (CDCl3):
δ 7.28-7.09 (m, 6H), 6.89 (d, 1H, J ) 7.0 Hz), 6.64-6.59 (m,

3530 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 16
Thomas et al.
(2) Aldrich, J . V. Analgesics. In Burger’s Medicinal Chemistry and
Drug Discovery; Wolff, M. E., Ed.; J ohn Wiley & Sons: New
York, 1996; Vol. 3.
3H), 6.31 (m, 1H), 4.65 (br, 1H), 3.16 (br, 1H), 3.02 (d, 1H, J
) 7.8 Hz), 2.69-2.14 (m, 16H), 1.91 (m, 1H), 1.86-1.75 (m,
4H), 1.58 (m 1H), 1.36-0.83 (m, 3H), 0.71 (d, 3H, J ) 6.5 Hz);
(3) Rogers, M. E.; May, E. L. Improved synthesis and additional
pharmacology of the potent analgetic (-)-5-m-hydroxyphenyl-
2-methylmorphan. J . Med. Chem. 1974, 17, 1328-1330.
(4) Awaya, H.; May, E. L.; Aceto, M. D.; Merz, H.; Rogers, M. E.;
Harris, L. S. Racemic and optically active 2,9-dimethyl-5-(m-
hydroxyphenyl)morphans and pharmacological comparison with
the 9-demethyl homologues. J . Med. Chem. 1984, 27, 536-539.
(5) Thomas, J . B.; Zheng, X.; Mascarella, S. W.; Rothman, R. B.;
Dersch, C. M.; Partilla, J . S.; Flippen-Anderson, J . L.; George,
C. F.; Cantrell, B. E.; Zimmerman, D. M.; Carroll, F. I. N-
Substituted 9â-methyl-5-(3-hydroxyphenyl)morphans are opioid
receptor pure antagonists. J . Med. Chem. 1998, 41, 4143-4149.
(6) Thomas, J . B.; Fall, M. J .; Cooper, J . B.; Rothman, R. B.;
Mascarella, S. W.; Xu, H.; Partilla, J . S.; Dersch, C. M.;
McCullough, K. B.; Cantrell, B. E.; Zimmerman, D. M.; Carroll,
F. I. Identification of an opioid κ receptor subtype-selective
N-substituent for (+)-(3R,4R)-dimethyl-4-(3-hydroxyphenyl)pi-
peridine. J . Med. Chem. 1998, 41, 5188-5197.
(7) Thomas, J . B.; Atkinson, R. N.; Rothman, R. B.; Fix, S. E.;
Mascarella, S. W.; Vinson, N. A.; Xu, H.; Dersch, C. M.; Lu, Y.;
Cantrell, B. E.; Zimmerman, D. M.; Carroll, F. I. Identification
of the first trans-(3R,4R)-dimethyl-4-(3-hydroxyphenyl)piperi-
dine derivative to possess highly potent and selective opioid κ
receptor antagonist activity. J . Med. Chem. 2001, 44, 2687-
2690.
(8) Thomas, J . B.; Gigstad, K. M.; Fix, S. E.; Burgess, J . P.; Cooper,
J . B.; Mascarella, S. W.; Cantrell, B. E.; Zimmerman, D. M.;
Carroll, F. I. A stereoselective synthetic approach to N-alkyl-
4â-methyl-5-phenylmorphans. Tetrahedron Lett. 1999, 40, 403-
406.
(9) Werner, J . A.; Cerbone, L. R.; Frank, S. A.; Ward, J . A.; Labib,
P.; Tharp-Taylor, R. W.; Ryan, C. W. Synthesis of trans-3,4-
dimethyl-4-(3-hydroxyphenyl)piperidine opioid antagonists: Ap-
plication of the cis-thermal elimination of carbonates to alkaloid
synthesis. J . Org. Chem. 1996, 61, 587-597.
(10) Bertha, C. M.; Flippen-Anderson, J . L.; Rothman, R. B.; Porreca,
F.; Davis, P.; Xu, H.; Becketts, K.; Cha, X.-Y.; Rice, K. C. Probes
for narcotic receptor-mediated phenomena. 20. Alteration of
opioid receptor subtype selectivity of the 5-(3-hydroxyphenyl)-
morphans by application of the message-address concept: Prepa-
ration of δ-opioid receptor ligands. J . Med. Chem. 1995, 38,
1523-1537.
20
[R]D -50.6° (c 0.62, CH3OH). Anal. (C30H43N3O2) C, H, N.
(-)-N-[(1R,4S,5S,7R)-5-(3-H yd r oxy)p h en yl-4-m et h yl-
2-(3-p h en ylp r op yl)-2-a za bicyclo[3.3.1]n on -7-yl]-3-gu a n i-
d in op r op a n a m id e [(-)-3d ]. A solution on N,N′-di-Boc-N′′-
trifylguanidine (4.39 g, 11.2 mmol) in dry CH2Cl2 was added
to a solution of â-alanine (1.00 g, 11.2 mmol) and triethylamine
(2.34 mL, 16.8 mmol) in dry CH2Cl2 (10 mL). The reaction
mixture was stirred at room temperature for 4 h. The organic
layer was washed with 1 M NaHSO4, followed by water, the
organic layer was collected and dried (Na2SO4), and solvent
was removed under reduced pressure yielding crude product,
which was purified by flash chromatography (1:1:0.004 EtOAc:
pentane:AcOH) to afford the Boc-protected guanidino acid (3.58
1
g, 97%) as a white solid. H NMR (CDCl3): δ 10.42 (br, 1H),
8.77 (br, 1H), 3.68 (t, 2H, J ) 6.1 Hz), 2.69 (t, 2H, J ) 6.1 Hz),
1.49 (s, 18H). 13C NMR (CDCl3): δ 175.9, 162.8, 156.3, 152.9,
83.4, 79.8, 36.2, 34.5, 28.2, 28.0. The acid thus prepared was
coupled to (-)-11 in the manner described for (-)-3b. The
product from this reaction was not purified but dissolved in
CH2Cl2 and treated with an equal volume of trifluoroacetic acid
at -30 °C. This was stirred for 30 min and allowed to warm
to room temperature whereupon the volatiles were removed
on a rotary evaporator. The resulting resin was dissolved in
methanol and slurried with silica gel, and the solvent was
removed under reduced pressure. Purification by flash chro-
matography using the resulting silica gel on a silica gel column
and 33% (80% CHCl3:18% CH3OH:2% NH4OH) in CHCl3 as
1
the eluent gave (-)-3d as an off-white foam (95%). H NMR
(CDCl3): δ 8.77 (t, 1H, J ) 5.7 Hz), 7.28-7.09 (m, 6H), 6.67
(m, 3H), 6.10 (br, 1H), 4.65 (br, 1H), 3.63 (m, 2H), 3.17 (br,
1H), 3.01 (d, 1H, J ) 7.8 Hz), 2.69-2.54 (m, 7H), 2.40 (m, 3H),
2.10 (m, 1H), 1.76 (m, 2H), 1.61-1.48 (m, 2H), 1.47 (s, 9H),
20
1.41 (s, 9H), 1.17-0.85 (m, 3H), 0.71 (d, 3H, J ) 6.9 Hz); [R]D
-26.5° (c 0.17, CH3OH). Anal. (C28H39N5O2) C, H, N.
(-)-N-[(1R,4S,5S,7R)-5-(3-Hyd r oxy)p h en yl-4-m eth yl-2-
(3-p h en ylp r op yl)-2-a za bicyclo[3.3.1]n on -7-yl]-4-gu a n id i-
n ob u t a n a m id e [(-)-3e]. A solution on N,N′-di-Boc-N′′-
trifylguanidine (3.80 g, 9.70 mmol) in dry CH2Cl2 was added
to a solution of 4-aminobutyric acid (1.00 g, 9.70 mmol) and
triethylamine (2.03 mL, 14.55 mmol) in dry CH2Cl2 (10 mL).
The reaction mixture was stirred at room temperature for 4
h. The organic layer was washed with 1 M NaHSO4, followed
by water. The organic layer was collected and dried (Na2SO4),
and solvent was removed under reduced pressure yielding
crude product that was purified by flash chromatography (1:
1:0.004 EtOAc:pentane:AcOH) to afford the Boc-protected
guanidino acid (2.92 g, 87%) as a white solid. 1H NMR
(CDCl3): δ 8.39 (br, 1H), 3.49 (m, 2H,), 2.45 (t, 2H, J ) 7.1
Hz), 1.92 (m, 2H), 1.50 (s, 9H), 1.49 (s, 9H). 13C NMR (CDCl3):
δ 177.4, 163.4, 157.1, 153.5, 83.8, 80.1, 40.1, 32.0, 28.5, 28.4,
25.5. The acid thus prepared was coupled to (-)-11 in the
manner described for (-)-3d to give (-)-3e as an off-white foam
(100%) following purification by flash chromatography using
33% (80% CHCl3:18% CH3OH:2% NH4OH) in CHCl3. 1H NMR
(CDCl3): δ 8.46 (t, 1H, J ) 5.6 Hz), 7.33-7.02 (m, 6H), 6.69
(d, 2H, J ) 6.6 Hz), 6.63 (d, 1H, J ) 6.7 Hz), 5.72 (br, 1H),
4.61 (br, 1H), 3.36 (m, 2H), 3.20 (br, 1H), 3.01 (d, 1H, J ) 8.3
Hz), 2.83-2.55 (m, 5H), 2.33 (m, 3H), 2.11 (m, 3H), 1.80 (m,
4H), 1.48 (s, 9H), 1.43 (s, 9H), 1.32-0.83 (m, 5H), 0.70 (d, 3H,
J ) 6.8 Hz); [R]D20 -24.7° (c 0.45, CH3OH). Anal. (C29H41N5O2)
C, H, N.
(11) The configuration of the 3- or 4-position substituents is desig-
nated as illustrated in Figure 1, i.e., substituents on the same
side of the piperidine ring as the propano bridge are designated
R/endo.
(12) Thomas, J . B.; Mascarella, S. W.; Rothman, R. B.; Partilla, J .
S.; Xu, H.; McCullough, K. B.; Dersch, C. M.; Cantrell, B. E.;
Zimmerman, D. M.; Carroll, F. I. Investigation of the N-
substituent conformation governing potency and µ receptor
subtype-selectivity in (+)-(3R,4R)-dimethyl-4-(3-hydroxyphenyl)-
piperidine opioid antagonists. J . Med. Chem. 1998, 41, 1980-
1990.
(13) Toll, L.; Berzetei-Gurske, I. P.; Polgar, W. E.; Brandt, S. R.;
Adapa, I. D.; Rodriguez, L.; White, A.; Kennedy, J . M.; Craymer,
K.; Farrington, L.; Auh, J . S. Standard binding and functional
assays related to (NIDA) Medications Development Division
testing for potential cocaine and narcotic treatment programs.
NIDA Res. Monogr. Ser. 1998, 178, 440-466.
(14) Traynor, J . R.; Nahorski, S. R. Modulation by µ-opioid agonists
of guanosine-5′-O-(3-[35S]thio)triphosphate binding to mem-
branes from human neuroblastoma SH-SY5Y cells. Mol. Phar-
macol. 1995, 47, 848-854.
(15) Portoghese, P. S.; Lipkowski, A. W.; Takemori, A. E. Binaltor-
phimine and nor-binaltorphimine, potent and selective κ-opioid
receptor antagonists. Life Sci. 1987, 40, 1287-1292.
(16) Portoghese, P. S.; Nagase, H.; Takemori, A. E. Only one
pharmacophore is required for the κ opioid antagonist selectivity
of norbinaltorphimine. J . Med. Chem. 1988, 31, 1344-1347.
(17) Olmsted, S. L.; Takemori, A. E.; Portoghese, P. S. A remarkable
change of opioid receptor selectivity on the attachment of a
peptidomimetic κ address element to the δ antagonist, natrin-
dole: 5′-[N2-alkylamidino)methyl]naltrindole derivatives as a
novel class of κ opioid receptor antagonists. J . Med. Chem. 1993,
36, 179-180.
Ack n ow led gm en t. This research was supported by
the National Institute on Drug Abuse, Grant DA 09045.
We thank NIDA (OTDP, Division of Treatment Re-
search and Development) for obtaining the functional
assay data provided herein.
(18) J ones, R. M.; Hjorth, S. A.; Schwartz, T. W.; Portoghese, P. S.
Mutational evidence for a common κ antagonist binding pocket
in the wild-type κ and mutant µ[K303E] opioid receptors. J . Med.
Chem. 1998, 41, 4911-4914.
(19) Schwyzer, R. ACTH: A short introductory review. Ann. N. Y.
Acad. Sci. 1977, 247, 3-26.
(20) Portoghese, P. S. An approach to the design of receptor-type-
Refer en ces
selective non-peptide antagonists of peptidergic receptors:
opioid antagonists. J . Med. Chem. 1991, 34, 1757-1762.
δ
(1) Zimmerman, D. M.; Leander, J . D. Selective opioid receptor
agonists and antagonists: Research tools and potential thera-
peutic agents. J . Med. Chem. 1990, 33, 895-902.
J M020084H