Synthesis of 7-arylmorphinans. Probing the 'address' requirements for selectivity at opioid δ receptors

Article abstract of DOI:10.1021/jm9802214

Through arylation of 6-keto opiates with diaryliodonium iodide, a series of 7-aryl opiates (38) have been prepared in an effort to investigate the effect of conformational mobility of the δ 'address' moiety on opioid agonist and antagonist potencies. Eval

Full text of DOI:10.1021/jm9802214

J . Med. Chem. 1998, 41, 3091-3098
3091
Syn th esis of 7-Ar ylm or p h in a n s. P r obin g th e “Ad d r ess” Requ ir em en ts for
Selectivity a t Op ioid δ Recep tor s
Peng Gao, Dennis L. Larson, and Philip S. Portoghese*
Department of Medicinal Chemistry, College of Pharmacy, University of Minnesota, 308 Harvard Street S.E.,
Minneapolis, Minnesota 55455
Received April 10, 1998
Through arylation of 6-keto opiates with diaryliodonium iodide, a series of 7-aryl opiates (3-
8) have been prepared in an effort to investigate the effect of conformational mobility of the δ
“address” moiety on opioid agonist and antagonist potencies. Evaluation of the ligands in the
mouse vas deferens and guinea pig ileum preparations revealed that they were less potent
and less selective than the conformationally constrained ligands, naltrindole (1, NTI) and
7-(spiroindanyl)oxymorphone (2, SIOM), at δ opioid receptors. It is concluded that the
coplanarity of the address moiety with the C ring of the morphinan structure enhances δ
antagonist potency and selectivity.
In tr od u ction
At least three major types of opioid receptors (δ, µ, κ)
are involved in the modulation of a variety of opioid
effects. These receptors have been cloned, and recent
studies of mutant and chimeric opioid receptors have
provided fresh insight into the molecular basis for
recognition of opioid ligands.1
The availability of selective antagonists has contrib-
uted greatly to opioid receptor classification and to the
pharmacologic characterization of a variety of opioid
agonists. One approach employed to develop selective
nonpeptide opioid ligands has been the application of
the message-address concept.2 In this simple model,
the “address” is viewed as a recognition element in the
opioid ligand that confers affinity for a specific receptor
type, whereas the “message” component is envisaged to
be involved in promoting receptor activation.3 Although
the idea of discrete message and address elements has
led to the design of selective opioid ligands, it has been
useful only in a conceptual context because of the
possible overlap in the role of these elements.4
The nonpeptide δ-selective opioid receptor antagonist,
naltrindole5 (1, NTI), and agonist, spiroindanyloxymor-
phone6 (2, SIOM), represent examples where this ap-
proach was used successfully. Their selectivity has been
attributed to the presence of an address element in the
form of a benzene moiety attached through a scaffold
to the C ring of the morphinan nucleus. This was based
on a model that envisages the Phe4 phenyl group of
enkephalin as a key δ address element. The approach
also involved the inclusion of conformational constraint
using a rigid scaffold attached to an address in an effort
to enhance δ opioid receptor selectivity.
In the present study, we have attached a conforma-
tionally mobile address moiety to the C-7 position of
hydromorphone and naltrexone in order to determine
how the potency and selectivity of such ligands (3-8)
compare with those of NTI and SIOM. The results
suggest that the conformationally constrained δ address
in these ligands may be an important contributor to
potency and selectivity.
S0022-2623(98)00221-0 CCC: $15.00 © 1998 American Chemical Society
Published on Web 07/09/1998

3092 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 16
Gao et al.
Sch em e 1
Sch em e 2
Design Ra tion a le a n d Ch em istr y
The benzene moiety (the address) of NTI (1) and
SIOM (2) is conformationally constrained through a
pyrrolo or spirocyclopentano “scaffold” that is attached
to ring C of the morphinan nucleus. We have investi-
gated the influence of rotational freedom of the address
by linking a conformationally mobile phenyl group
directly to the C-7 position of hydromorphone and
naltrexone. In addition, the 6-keto group was reduced
in order to determine its role in selectivity and potency.
The 7-aryl group was incorporated into hydrocodone
(9a ), cyclopropylmethylnorhydrocodone (9b), and O-
benzylnaltrexone (17) via a phenylation procedure using
diaryliodonium iodide (Schemes 1 and 2).7 This method
provided the desired 7-phenyl intermediates (10-12, 18)
as major products together with minor amounts of
diarylated byproducts (19a , 19b). The stereochemistry
of the 7-phenyl group was confirmed to be 7R by the
positive NOE between the H-5 and H-7 â protons.
O-Dealkylation of 10a and 10b with BBr3 provided
target compounds 3a and 3b, respectively. The naltr-
exone derivative 3c was prepared as an analogue of 3b
using a similar procedure (Scheme 2). This involved
phenylation of O-benzylnaltrexone (17) with diphenyli-
odonium iodide to afford 18, followed by debenzylation
with HCl-HOAc to give 3c.
2.1 Hz indicated that the 6-hydroxy group is cis to the
7-phenyl group. The value of J 5,6 ) 5.8 Hz is consistent
with the coupling reported in structurally related 6-hy-
droxy opiates.8 It appears likely that the C ring is in a
chair conformation in view of the stabilizing effect of
an equatorial 7-phenyl group.
Alcohols 13a and 13b were esterified with acetic
anhydride for the preparation of acetate intermediates
16a and 16b , respectively. Subsequent treatment of
16a and 16b with boron tribromide afforded 8a and 8b,
respectively. Since H6-H7 coupling constant of the
alcohols were essentially the same upon acetylation (J
) 2.1 Hz), the configuration at C-6 and the conformation
of the C ring appeared to be unchanged. The 7,7-
diphenyl compound 4b was prepared through the boron
Compounds 13-15, which were obtained by the
reduction of ketones 10-12 with sodium borohydride,
were O-demethylated with boron tribromide to afford
5a , 6a , 6b, 7a , and 7b. Compound 5b was prepared
through an alternative route by reduction of 3b with
sodium borohydride. The H6-H7 coupling constant of

Synthesis of 7-Arylmorphinans
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 16 3093
Ta ble 1. Opioid Agonist Activity of 7-Phenyl Opiates in the
Ta ble 2. Opioid Antagonist Activity of 7-Phenyl Opiates in
MVD and GPI
the MVD and GPI
% max responsea or IC50 (nM)b
IC50 ratio
compd
MVD
GPI
compd
DADLEa (δ)
morphineb (µ)
EKb (κ)
3a
3b
3c
4a
4b
5a
5b
6a
6b
7a
7b
8a
8b
29.2 ( 2.8% (3)
31.4 ( 9.9% (3)
30.0 ( 5.1% (3)
5.4 ( 8.6% (3)
0.0 ( 0.0% (3)
178 ( 83.1 nM (3)
14.5 ( 6.8% (3)
15.2 ( 4.7% (3)
22.4 ( 10.4% (3)
56.1 ( 7.9% (4)
26.8 ( 5.6% (3)
34.3 ( 1.8% (3)
21.1 ( 8.6% (3)
53.9 ( 2.4% (3)
161 ( 62 nM (3)
-15.1 ( 16.1% (3)c
9.7 ( 14.0% (3)
475 ( 121 nM (10)
23.2 ( 2.0% (2)
58.9 ( 4.0 nM (3)
10.0 ( 5.4% (3)
39.2 ( 14.3% (3)
49.3 ( 9.1% (5)
28.3 ( 16.4 nM (3)
24.0 ( 7.9% (3)
118 ( 70.1 nM (6)
1 (NTI)c Ke ) 0.13 nMd
Ke ) 29 nM
Ke ) 46 nM
3a
3b
0.97 ( 0.2 (3)
17.6 ( 4.6 (3)
(Ke ) 5.9nM)
34.6 ( 8.5 (3)
(Ke ) 3 nM)
2.34 ( 0.60 (3)
63.2 ( 11.3 (6)
(Ke ) 1.6 nM)
e
3.03 ( 0.72 (5)
1.17 ( 0.29 (4)
9.07 ( 1.63 (4)
0.99 ( 0.05 (3)
4.17 ( 0.9 (10)
1.02 ( 0.42 (3)
6.38 ( 0.42 (3)
1.05 ( 0.17 (3) 0.75 ( 0.2 (3)
e
e
3c
17.3 ( 4.3 (6)
(Ke ) 6 nM)
3.9 ( 0.9 (7)
(Ke ) 34 nM)
4a
4b
1.69 ( 0.42 (3) 1.01 ( 0.31 (3)
e
e
5a
5b
6a
6b
7a
7b
8a
8b
1.85 ( 1.1 (2)
e
1.9 ( 0.5 (2)
e
1.20 ( 0.42 (3) 1.76 ( 0.40 (3)
1.38 ( 0.35 (3) 1.22 ( 0.72 (3)
0.34 ( 0.11 (3) 1.05 ( 0.09 (3)
a
Percent of maximal response ((SE) at 1 µM unless otherwise
e
e
b
noted. The IC50 is the concentration of the agonist required for
0.90 ( 0.30 (3) 1.08 ( 0.42 (3)
e
half-maximal response of the preparation. c Negative sign means
that the twitch amplitude was enhanced in the present of the test
compound.
e
a
b
Tested in the MVD preparation. Tested in the GPI. c Data
from ref 5. Ke (nM) ) [antagonist]/(IC50 ratio - 1). e Not tested
due to agonist activity.
d
tribromide O-demethylation of 19b using a procedure
similar to that reported9 for the corresponding hydro-
codone analogue 4a .
Ta ble 3. Antagonism by 4b of the Antinociceptive Effect of
Opioid Agonists in Mice
ED50, nmol/mouse or µmol/kg
Biologica l Resu lts
agonist selectivity
control
treateda
ED50 ratiob
Sm ooth Mu scle P r ep a r a tion . Hydrochloride salts
of the target compounds were tested for opioid agonist
and antagonist activity in the electrically stimulated
mouse vas deferens10 (MVD) and the guinea pig ileal
longitudinal muscle11 (GPI) preparations. For deter-
mination of antagonist activity, the ligands (100 nM)
were incubated with the preparation for 15 min prior
to testing with the standard agonists, [D-Ala2,D-Leu5]-
enkephalin12 (DADLE), morphine, or ethylketazocine
(EK). These agonists are pharmacologically selective
for δ, µ, and κ receptors, respectively. DADLE was
employed in the MVD; morphine and EK were used in
the GPI. When the test compound was found to be
weakly active as an agonist, it was incubated with the
smooth muscle preparation at a single concentration (1
µM) to determine the maximal agonist response. When
a compound was found to be a full agonist, four to six
concentrations of the test compound were employed to
construct a concentration-response curve which was
compared to that of a standard agonist in the same
preparation. The opioid antagonist potency was ex-
pressed as an IC50 ratio, which represents the IC50 of
the standard agonist in the presence of the tested
compound (100 nM) divided by the control IC50 of the
standard agonist above.
In the GPI, compounds 3b, 4b, 5b, 7b, and 8b were
found to be full agonists, with IC50 values ranging from
28 to 475 nM, and in this regard, compound 7b was the
most active (Table 1). In the MVD, only 5a possessed
full agonist activity (IC50 ) 178 nM). The remaining
compounds displayed either partial agonist activity or,
in one case (3c), an enhanced electrically stimulated
twitch of the GPI.
Evaluation of antagonist activity (Table 2) showed
that the N-methyl compounds were inactive when tested
against DADLE, morphine, or EK. On the other hand,
several ligands in the N-cyclopropylmethyl series (3b,
3c, 4b) antagonized DADLE, with the most potent (4b)
having an IC50 ratio of 63. Interestingly, 4b exhibited
DPDPEc
DSLETc
δ1
δ2
µ
8.1 (5.9-11.2) 27.9 (18.1-41.9) 3.5 (2.0-5.9)
0.7 (0.5-0.9) 2.6 (2.0-3.2) 3.7 (2.6-5.0)
9.1 (0.4-12.6) 15.7 (9.7-23.3) 1.7 (1.0-3.1)
27.5 (4.8-63.2) 22.4 (1.9-44.8) 0.8 (0.1-2.9)
morphined
U50,488d
κ
a
b
Treated icv with 20 nmol of 4b. Treated ED50 divided by
control ED50.
c Administered icv; ED50 values expressed as nmol/
d
mouse. Administered sc; ED50 values expressed as µmol/kg.
full agonist activity in the GPI which probably is µ
receptor-mediated in view of its higher affinity for µ (Ke
) 0.11 ( 0.15 nM) relative to κ (Ki ) 7.58 ( 1.36 nM)
receptors. With the notable exception of 3c (morphine
IC50 ratio, 17), none of the compounds possessed sig-
nificant antagonism for morphine or EK in the GPI.
In Vivo Testin g. The most potent antagonist (4b)
in this series was tested using the tail flick procedure
in mice.13 Administration of 4b (20 nmol icv) with
standard agonists was timed so that the peak effect (45
min) coincided with the center of the observation period.
The ED50 values of the standard agonists, [D-Pen2,D-
Pen5]enkephalin (DPDPE; δ1),14 [D-Ser2,Leu5]enkephalin-
Thr5 (DSLET; δ2),15 morphine (µ), and (()-3,4-dichloro-
N-methyl-N-[2-(1-pyrrolidinyl)cyclohexyl]benzeneacet-
amide16 (U50,488; κ), were then determined and ex-
pressed as ED50 ratios (treated/control). As indicated
in Table 3, 4b antagonized DPDPE and DSLET equally
and to a greater extent than morphine. No antagonism
was noted for U50,488. Thus, although 4b is apparently
δ-selective in vivo, it does not distinguish between the
putative δ1 and the δ2 receptors.17,18
Discu ssion
With NTI (1), the aromatic δ address moiety is
conformationally constrained due to fusion with ring C
of the morphinan structure. This conformational rigid-
ity may be a contributing factor in the approximately
20-fold greater antagonist potency of NTI at δ receptors
relative to its analogue, 7-phenylnaltrexone (3c). Al-
though the equatorial 7-phenyl group of 3c (Figure 1)

3094 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 16
Gao et al.
F igu r e 1. Superposition of the low-energy conformers of NTI 1 and 7-phenylnaltrexone (3c).
F igu r e 2. Superposition of the low-energy conformers 3b and 4b.
is conformationally mobile, its rotamer population pre-
fers an orthogonal-like population with respect to the
plane of ring C of the morphinan structure due to steric
hindrance between the phenyl group ortho protons and
the vicinal carbonyl and methylene groups. It appears
likely that the lack of coplanarity between ring C and
the address moiety is the principal reason for the lower
δ antagonist potency of 3c. This is consistent with the
significantly lower δ antagonist potency of 7-spiroinda-
nyl and spirobenzocyclohexyl derivatives of naltrex-
one.19,20 In these cases the reduced potency was attrib-
uted to the fact that the plane of the address moiety is
perpendicular rather than coplanar to ring C as in NTI.
The δ antagonist selectivity of 7-phenylnaltrexone
(3c) is 2 orders of magnitude less than that of NTI due
to its lower potency at δ receptors and greater potency
at µ and κ receptors. The fact that 3c more potently
antagonizes µ and κ agonists compared to NTI may be
explained by the greater conformational mobility of its
address. In this regard, it has been reported that the
selectivity of NTI is due to increased affinity for δ
receptors conferred by its address and to decreased
affinity for non-δ receptors as a consequence of steric
interactions.21 Thus, the greater potency of 3c at µ and
κ receptors relative to that of NTI may be due to the
possibility that its 7-phenyl group can more easily be
accommodated because of its conformational mobility.
In the present study we have found that 7-phenylhy-
dromorphone (3a ) was inactive as an antagonist or as
a full agonist in smooth muscle preparations. These
results are in harmony with a related study which
revealed that the 14-hydroxyl group is of critical im-
portance for the δ agonist potency of SIOM (2).22
Moreover, the finding that the cyclopropylmethyl ana-
logue 3b is less potent than its naltrexone analogue 3c
is consistent with the finding that the 14-hydroxyl group
contributes to the δ antagonist potency of NTI. It is
noteworthy that the 14-hydroxyl group also enhances
the opioid antagonist activity of naltrexone.23 This
similarity in the structure-activity relationship sug-
gests that there may be some common structural
features at the antagonist recognition site of µ and δ
receptors or that some common physicochemical factor
conferred by the 14-hydroxy group facilitates access to
the recognition sites.
The gem-diphenyl compound 4b was of interest
because of its unusual pharmacologic profile in smooth
muscle preparations. It was the most potent δ opioid
antagonist in the series when tested against DADLE
in the MVD (Ke ) 1.6 nM), whereas it functioned as a
weak, full agonist in the GPI. One possible explanation
for the greater antagonist potency relative to its
monophenyl analogue 3b is that the â-phenyl group of
4b induces its geminal partner to assume a conforma-
tion that is more coplanar with ring C. This is il-
lustrated in Figure 2 with superposition of the low-
energy conformers of 3b and 4b.
The remaining members of the series (5-8) exhibited
only feeble antagonist activity. However, several of the
cyclopropyl derivatives (5b, 7b, 8b) possessed moderate
agonist potencies in the GPI, and the N-methyl com-
pound 5a was a full agonist in the MVD. Of these
compounds, only 5a appeared to be a selective δ agonist,
as it was found to be inactive in the GPI. This was
surprising in view of the finding22 that the 14-hydroxyl
group is critical for the δ agonist activity of SIOM (2).
Although 3a is closely strucurally related to 5a , it is
not a full agonist in either of the smooth muscle
preparations. There is no obvious explanation as to why
substituting an hydroxyl for a ketone group should
make such a dramatic difference in activity, aside from
the possibility that the 6-hydroxy group may function
as a hydrogen bonding donor in altering the mode of
interaction with the δ receptor.
In conclusion, the attachment of a phenyl group to
the C-7 position of naltrexone to afford 3c changed
antagonist selectivity from µ to δ. This change presum-
ably is due to the aromatic group which functions as a

Synthesis of 7-Arylmorphinans
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 16 3095
-78 °C, followed by centrifugal chromatography using CHCl3-
EtOH-NH4OH (98:1:1), affording 11a (330 mg, 72%): mp
189-191 °C; 1H NMR (CDCl3) δ 7.38 (d, 2H, J ) 6.6, H-ArBr),
6.86 (d, 2H, J ) 6.6, H-ArBr), 6.68-6.78 (2d, 2H, J ) 8.1, H-1
and H-2), 4.87 (s, 1H, H-5), 3.93 (s, 3H, O-Me), 3.66-3.70 (m,
1H, H-7R), 3.38 (m, 1H, H-9), 3.07 (d, 1H, J ) 18.6, H-10),
1.52-1.66 (q, 1H, J ) 12.9, H-8R); MS (FAB) m/z 454 [M +
H]+.
δ address in a fashion similar to that of NTI (1). The
substantially lower δ selectivity of 3c is attributed to
the substantially greater conformational mobility and
lack of coplanarity of its 7-phenyl group when compared
to the rigidly held benzene moiety in NTI.
Some congeners in the series (i.e., 3b, 4b) appeared
to possess mixed agonist-antagonist activity in vitro. If
the agonist component is mediated through µ receptors,
they could be leads for the development of analgesics
devoid of abuse potential, in view of reports24-26 that δ
antagonists block the tolerance and physical dependence
of morphine without antagonizing antinociception.
N-(Cyclop r op ylm eth yl)-7r-(p-br om op h en yl)n or h yd r o-
cod on e (11b). Using the same procedure described above for
10b, compound 9b27 (1.10 g, 3.3 mmol) was reacted with
LiHMDSA [prepared from HMDSA (1.35 mL, 6.3 mmol) in
THF (10 mL), n-butyllithium (2.0 mL, 4.8 mmol, 2.4 M in
hexane)] and 4,4′-dibromophenyliodium iodide29 (2.0 g, 3.54
mmol) in DMF (100 mL). The crude product was purified by
column chromatography on silica gel (70 g), eluting with EtOH
(2%) in CHCl3 to afford 11b (0.51 g, 31%): 1H NMR (CDCl3) δ
7.37-7.38 (d, J ) 8.4, 2H, H-Ar), 6.85-6.88 (d, J ) 8.4, 2H,
H-Ar), 6.63-6.75 (2d, J ) 8.1, 2H, H-1and H-2), 4.85 (s, 1H,
H-5), 3.92 (s, 3H, O-Me), 3.36-3.67 (dd, J ) 3.6, 13.5, 1H, H-7),
3.58 (m, 1H, H-9), 2.95 (d, J ) 18.0, 1H, H-10), 0.90 (m, 1H,
H-19), 0.54-0.57 (m, 2H, H-20 and H-21), 0.15-0.19 (m, 2H,
H′-20 and H′-21); MS (FAB) m/z 494 [M + H]+, 492 [M - H]-.
Exp er im en ta l Section
Reagents were obtained from Aldrich Chemical Co. unless
otherwise noted. Hydrocodone bitartrate and naltrexone
hydrochloride were obtained from Mallinckrodt. All the reac-
tions were conducted under nitrogen (N2) unless specified.
Column chromatography was performed with silica gel (200-
400 mesh, Aldrich). Thin-layer chromatography was per-
formed on E. Merck silica gel 60 F-254 0.25 nm plates and
visualized with UV light or iodine vapor. Centrifugal prepara-
tive chromatography was performed on silica gel (EM Science
silica gel 60, PF254) layered on glass rotor plates and a
Chromatotron apparatus (Harrison Research, model 7924T,
Palo Alto, CA). Chromatographic solvent systems are reported
as volume/volume ratios. IR spectra were recorded on a
Nicolet 5DXC FT-IR instrument. NMR data were collected
on a GE Omega 300 MHz or Varian 500 MHz spectrometer at
room temperature (18-20 °C). The δ (ppm) scale was in
reference to the deuterated solvent, and coupling constants
are reported in hertz. Mass spectra were obtained on a
Finnigan 4000 or a VG707EHF spectrometer. Melting points
were determined in open capillary tubes with a Thomas-
Hoover melting point apparatus and are uncorrected. El-
emental analyses were performed by M-H-W Laboratory
(Phoenix, AZ) and are within (0.4% of theoretical values.
N -(Cyclop r op ylm e t h yl)-7r-p h e n yln or h yd r ocod on e
(10b). Hexamethyldisilazane (HMDSA) (0.9 mL, 2.88 mmol)
was mixed with THF (25 mL) at -78 °C under nitrogen. A
solution of n-butyllithium (1.2 mL, 2.88 mmol, 2.5 M in
hexane) was added over a 5 min period, and the resulting
solution was stirred for another 10 min. N-(Cyclopropylm-
ethyl)norhydrocodone (9b)27 (740 mg, 2.18 mmol) in THF (25
mL) was added, and stirring was continued at -78 °C for 30
min. The resulting solution was transferred to a mixture of
diphenyliodonium iodide28 (700 mg, 1.7 mmol) in 55 mL of
DMF at -45 °C. The final mixture was stirred at -45 °C for
2 h and then was allowed to warm to room temperature and
stirred overnight. The reaction was quenched with water (60
mL) and extracted with chloroform (3 × 50 mL). The extract
was dried (MgSO4), and the solvent was removed under
reduced pressure. The residue was taken up in a small
amount of CH2Cl2, applied to a silica gel rotor plate (2 mm
layer), and eluted with CHCl3-EtOH-NH4OH (98:1:1) to
afford 10b (366 mg, 40%): mp 275 °C dec (hydrochloride salt);
1H NMR (CDCl3) δ 7.20-7.24 (m, 3 H, H-Ar), 6.93-6.94 (m, 2
H, H-Ar), 6.71-6.93 (2d, 2H, J ) 8.1, H-1and H-2), 4.94 (s, 1
H, H-5), 4.19 (s, 3 H, OCH3), 3.46-3.94 (m, 2 H, H-7 and H-10),
3.27 (m, 1H, H-10), 2.58 (m, 1 H, H-9), 2.14 (m, 1 H, H-8),
2.00 (m, 1 H, H-14), 0.81 (m, 1 H, H-19), 0.65-0.67 (m, 2 H,
H-20 and H-21), 0.40-0.42 (m, 2 H, H′-20 and H′-21); 13C NMR
(CDCl3) δ 3.73, 3.94, 8.83, 20.60, 33.16, 34.88, 41.88, 45.26,
47.61, 54.83, 56.62, 56.83, 59.48, 91.96, 115.22, 119.63, 126.95,
128.07, 128.33, 136.69, 142.81, 145.25, 205.49; MS (FAB) m/z
416 [M + H]+.
7r-(m -Nitr op h en yl)h yd r ocod on e (12a ). Following the
procedure described for 10b, compound 9a (320 mg, 1.06 mmol)
was treated with LiHMDSA [from HMDSA (0.42 mL, 2.0
mmol), n-butyllithium (0.60 mL, 1.5 mmol, 2.5 M in hexane)]
and 3,3′-dinitrophenyliodonium iodide28,29 (800 mg, 1.6 mmol)
in DMF (60 mL), followed by centrifugal chromatography using
3% ethanol in chloroform, to afford 205 mg (45%) of 12a : mp
1
221-225 °C; H NMR (CDCl3) δ 8.09-8.12 (dd, 1H, J ) 2.4,
8.1, H-ArNO2), 7.86-7.87 (m, 1H, H-ArNO2), 7.43-7.48 (dd,
1H, J ) 4.5, 8.1, H-ArNO2), 7.32-7.36 (dd, 1H, J ) 1.5, 6.6,
H-ArNO2), 6.72-6.80 (2d, 2H, J ) 6, H-1and H-2), 4.93 (s, 1H,
H-5), 3.92 (s, 3H, O-Me), 3.83-3.9 (m, 1H, H-7R), 3.39 (m, 1H,
H-9), 3.11 (d, 1H, J ) 18.9, H-10), 2.05-2.12 (m, 1H), 1.9-
1.94 (m, 1H), 1.62-1.75 (m, H-8); MS (FAB) m/z 421 [M + H]+.
N-(Cyclop r op ylm eth yl)-7r-(m -n itr op h en yl)n or h yd r o-
cod on e (12b). Using the procedure of 10b, compound 9b27
(0.98 g, 2.88 mmol) was reacted with LiHMDSA [prepared
from HMDSA (1.34 mL, 6.3 mmol), n-butyllithium (1.95 mL,
4.7 mmol, 2.4 M in hexane) in THF], and the mixture was
added to 3,3′-dinitrophenyliodium iodide28,29 (1.80 g, 3.6 mmol)
in DMF (100 mL). The product was purified through a silica
gel column (70 g), eluting with EtOH (3%) in CHCl3 to afford
12b (220 mg, 17%): 1H NMR (CDCl3) δ 8.07-8.11 (m, 1H,
H-ArNO2), 7.87-7.88 (m, 1H, H-ArNO2), 7.41-7.47 (m, 1H,
H-ArNO2), 7.32-7.34 (m, 1H, H-ArNO2), 6.66-6.76 (2d, J )
8.4, 2H, H-1 and H-2), 4.89 (s, 1H, H-5), 3.92 (s, 3H, O-Me),
3.78-3.84 (dd, J ) 4.2, 13.2, 1H, H-7), 3.55-3.56 (m, 1H, H-9),
2.95-3.01 (d, 1H, H-10), 0.85 (m, 1H, H-19), 0.51-0.57 (m,
2H, H-20 and H-21), 0.13-0.17 (m, 2H, H′-20 and H′-21).
17-(Cyclop r op ylm eth yl)-3-h yd r oxy-4,5r-ep oxy-7r-p h e-
n ylm or p h in a n -6-on e (3b). To a solution of N-(cyclopropy-
lmethyl)-7R-phenylnorhydrocodone (10b) (0.64 g, 1.54 mmol)
in CH2Cl2 (10 mL) at 0 °C was added boron tribromide (6.0
mL, 1 M in CH2CH2) over 10 min. The mixture was stirred
at 0 °C for 1 h, methanol (12 mL) was added, and the mixture
was stirred for another hour. The reaction mixture then was
adjusted with saturated aqueous NaHCO3 to pH 8 and
extracted with chloroform (4 × 30 mL). The chloroform extract
was washed with water and dried (MgSO4), and the solvent
was removed under reduced pressure to give the crude base.
This was taken up in a small amount of chloroform and applied
to a silica gel rotor plate (2 mm layer) and eluted with 2%
ethanol in chloroform to afford pure 3b base (360 mg, 58%):
1H NMR (CDCl3) δ 7.12-7.23 (m, 3H, H-Ph), 6.94-6.99 (m,
2H, H-Ph), 6.52-6.64 (2d, 2H, J ) 7.8, H-1 and H-2), 3.94 (s,
1H, H-5), 3.90-3.96 (dd, 1H, J ) 3.6, 14.2, H-7), 3.44-3.50
(m, 1H, H-9), 0.81-0.88 (m, 1H, H-19), 0.42-0.50 (m, 2H, H-20
and H-21), 0.08-0.12 (m, 2H, H′-20 and H′-21); MS (FAB) m/z
402 [M + H]+. Treatment of the base in methanol with
7r-(p-Br om oph en yl)h ydr ocodon e (11a). Using the same
procedure as described for 10b above, hydrocodone (9a ) (300
mg, 1.0 mmol) was reacted with LiHMDSA [prepared from
hexamethyldisilazane (0.42 mL, 2.0 mmol) and n-butyllithium
(0.60 mL, 1.5 mmol, 2.5 M in hexane)] and 4,4′-dibromophe-
nyliodonium iodide29 (960 mg, 1.7 mmol) in 40 mL of DMF at

3096 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 16
Gao et al.
ethereal HCl gave 3b‚HCl, which was crystallized from
17-(Cyclop r op ylm eth yl)-3-m eth oxy-4,5r-ep oxy-7r-(m -
n itr op h en yl)m or p h in a n -6r-ol (15b). A mixture of com-
pound 12b (220 mg, 0.476 mmol) and NaBH4 (390 mg, 10.1
mmol) in EtOH (40 mL) afforded 15b (220 mg, 99%): 1H NMR
(CDCl3) δ 8.01-8.07 (m, 2H, H-Ar), 7.38-7.54 (m, 2H, H-Ar),
6.65-6.74 (2d, 2H, J ) 8.4, H-1 and H-2), 4.72 (d, J ) 6.0,
1H, H-5), 4.09-4.19 (m, 1H, H-6), 3.84 (s, 3H, O-Me), 3.42 (m,
1H, H-9), 2.96 (d, 1H, J ) 18.3, H-10R), 0.83 (m, 1H, H-19),
0.50-0.52 (m, 2H, H-20 and H-21), 0.10-0.15 (m, 2H, H′-20
and H′-21); MS (FAB) m/z 463 [M + H]+.
MeOH-Et2O, mp 201-205 °C. Anal. (C23H25NO3‚HCl) C, H,
N.
17-(Cyclop r op ylm eth yl)-3,6r-d ih yd r oxy-4,5r-ep oxy-7r-
p h en ylm or p h in a n (5b). A mixture of 3b (260 mg, 0.648
mmol), NaBH4 (450 mg, 11.7 mmol), and EtOH (45 mL) was
stirred for 22 h at room temperature. Acetone (20 mL) was
added to quench the reaction, and the resulting mixture was
stirred for an additional 20 min and filtered. The filtrate was
concentrated and mixed with chloroform (2 mL), and the
mixture was applied to a silica gel rotor plate (2 mm) and
eluted with EtOH (3.8%) in CHCl3 to afford 5b (167 mg,
64%): 1H NMR (CDCl3) δ 7.10-7.26 (m, 5H, H-Ph), 6.53-6.65
(2d, 2H, J ) 8.4, H-1and H-2), 4.57 (d, J ) 5.7, 1H, H-5), 4.05-
4.07 (dd, 1H, J ) 1.5, 5.7, H-6), 3.46-3.47 (m, 1H, H-9), 2.91
(d, 1H, J ) 18.3, H-10â), 2.60-2.65 (m, 1H, H-7), 2.31-2.55
(m, 3H, H-10R, H-14, H-17), 1.50-1.74 (m, 1H, H-8R), 1.39-
1.44 (m, 1H, H-8â), 1.82-1.94 (m, 1H, H-19), 0.48-0.53 (m,
The following compounds (4b, 5a , 6a -8b) were prepared
using compounds described herein as precursors and the same
procedure as described for 3b above, except for differences
noted:
17-(Cyclop r op ylm e t h yl)-3-h yd r oxy-4,5r-e p oxy-7,7-
d ip h en ylm or p h in a n -6-on e (4b). Compound 19b (210 mg,
0.42 mmol) treated with boron tribromide (1.8 mmol) by this
procedure furnished 4b, 29 mg (14%) after centrifugal chro-
matography using EtOAc-hexanes (1:3): 1H NMR (CDCl3) δ
7.08-7.46 (m, 8H, H-Ph), 6.63-6.75 (2d, J ) 7.8 Hz, 2H, H-1
and H-2), 6.54-6.57 (m, 2H, H-Ph), 4.96 (s, 1H, H-5), 3.63 (m,
1H, H-9), 2.99 (d, J ) 18.3 Hz, 1H, H-10); MS (FAB) m/z 478
[M + H]+. The base in CHCl3 was treated with ethereal HCl,
and after removal of the solvent, the residual solid was
crystallized from i-PrOH-Et2O to afford the hydrochloride
salt, mp 226 °C. Anal. (C32H31NO3‚HCl‚H2O) C, H, N.
2H, H-20 and H-21), 0.12-0.13 (m, 2H, H′-20 and H′-21); 13
C
NMR (CD3COCD3) δ 4.80 (s), 4.78 (s), 9.83 (t), 21.50 (s), 23.99
(s), 37.59 (s), 42.65 (q), 44.11 (t), 46.14 (t), 46.73 (s), 57.99 (t),
60.66 (s), 72.61 (t), 91.54 (t), 118.38 (t), 120.12 (t), 125.82 (q),
127.38 (t), 128.95 (t), 128.99 (t), 131.06 (q), 138.60 (q), 143.47
(q), 145.86 (q). Compound 5b (63 mg, 0.156 mmol) was
dissolved in chloroform (2 mL) and treated with ethereal HCl
to furnish a solid, which was crystallized from MeOH-Et2O
to afford 5b‚HCl (60 mg, 87%): mp 240 °C dec; MS (FAB) m/z
404 [M + H]+. Anal. (C26H29NO3‚HCl) C, H, N, Cl.
17-Met h yl-4,5r-ep oxy-7r-p h en ylm or p h in a n -3,6r-d iol
(5a ). Obtained from 13a (378 mg, 1.0 mmol) in 50% yield,
using 9% EtOH in chloroform as the chromatography eluant:
1H NMR (CD3COCD3) δ 7.01-7.20 (m, 5H, H-Ph), 6.52-6.70
(2d, 2H, J ) 8.1, H-1 and H-2), 4.37 (d, J ) 6.0, 1H, H-5),
3.39-3.46 (dd, 1H, J ) 6.0, 11.5, H-6), 3.05-3.08 (m, 1H, H-9),
2.96 (d, 1H, J ) 18.3, H-10), 2.31 (s, 3H, NMe), 2.24-2.39 (m,
2H, H-14, H-10), 1.20-1.32 (m, 1H, H-8R); 13C NMR (CD3-
COCD3) δ 18.61 (s), 20.64 (s), 24.21 (s), 37.44 (q), 41.87 (p),
42.89 (t), 44.40 (t), 45.89 (t), 47.82 (s), 57.43 (t), 60.29 (t), 71.94
(t), 91.26 (t), 117.47 (t), 118.91 (t), 125.38 (q), 126.36 (t), 128.27
(t), 131.12 (t), 138.93 (q), 144.95 (q), 146.40 (q); MS (FAB) m/z
364 [M + H]+. The base 5a (158 mg, 0.419 mmol) was
dissolved in chloroform (5 mL) and treated with Et2O-HCl (8
mL) to give a solid, which was crystallized from MeOH-Et2O
Compounds 13a -15b were prepared using the same pro-
cedure as described for 5b above, except for differences noted:
17-Meth yl-3-m eth oxy-4,5r-epoxy-7r-ph en ylm or ph in an -
6r-ol (13a ). The reaction between 7R-phenylhydrocodone7
(10a ) (120 mg, 0.32 mmol) and NaBH4 (200 mg, 5.2 mmol) in
1
EtOH (8 mL) afforded 13a (106 mg, 87%): mp 84-86 °C; H
NMR (CDCl3) δ 7.08-7.28 (m, 5H, H-Ph), 6.65-6.74 (2d, J )
8.4, 2H, H-1 and H-2), 4.72 (d, J ) 6.0, 1H, H-5), 4.18-4.20
(m, 1H, H-6), 3.85 (s, 3H, O-Me), 3.12-3.15 (m, 1H, H-9), 3.06
(d, J ) 18.3, 1H, H-10â), 2.69-2.73 (m, 1H, H-7), 2.44-2.50
(m, 1H, H-10R), 2.43 (m, 3H, N-Me), 2.37-2.43 (m, 1H, H-14),
1.43-1.48 (m, 1H, H-8); 13C NMR (CDCl3) δ 20.92 (s), 23.79
(s), 37.55 (s), 43.11 (q), 43.54 (p), 44.94 (t), 46.05 (t), 48.09 (s),
56.89 (p), 60.46 (t), 72.33 (t), 91.40 (t), 113.69 (t), 119.99 (t),
126.84 (q), 128.74 (t), 128.82 (t), 128.83 (t), 131.25 (q), 142.32
(q), 143.66 (q), 146.88 (q).
to afford 5a ‚HCl (165 mg, 95%), mp > 240 °C. Anal. (C23H25
NO3‚HCl) C, H, N, Cl.
-
17-Meth yl-4,5r-ep oxy-7r-(p-br om op h en yl)m or p h in a n -
3,6r-d iol (6a ). Obtained from 14a (445 mg, 0.96 mmol) in
34% yield (145 mg) and using 10% EtOH in chloroform as the
chromatography eluant: 1H NMR (CDCl3) δ 7.25-7.28 (d, J
) 8.4, 2H, H-Ar), 6.91-6.94 (d, J ) 8.4, 2H, H-Ar), 6.49-6.60
(2d, J ) 8.7, 2H, H-1 and H-2), 4.51 (d, J ) 5.4, 1H, H-5),
3.91-3.93 (m, 1H, H-6), 3.17 (m, 1H, H-9), 2.93 (d, J ) 18.3,
1H, H-10), 2.39 (s, 3H, N-Me); MS (FAB) m/z 442 [M + H]+,
m/z 440.0 [M - H]-. Compound 6a (144 mg, 0.325 mmol) was
dissolved in CHCl3 (2 mL) and converted to the hydrochloride
salt, which was recrystallized from 2-propanol twice to afford
6a ‚HCl (94 mg, 60%): mp 255 °C dec; MS (FAB) m/z 442 [M
+ H]+. Anal. (C23H24NO3Br‚HCl‚0.5H2O) C, H, N.
17-Meth yl-3-m eth oxy-4,5r-ep oxy-7r-(p-br om op h en yl)-
m or p h in a n -6r-ol (14a ). A mixture of 11a (440 mg, 0.97
mmol) and NaBH4 (750 mg, 19.5 mmol) in EtOH (35 mL)
afforded crude 14a (445 mg, 99%), which was used directly
for the next reaction without further purification: 1H NMR
(CDCl3) δ 7.35-7.38 (d, J ) 9.0, 2H, H-Ar), 7.04-7.07 (d, J )
9.0, 2H, H-Ar), 6.67-6.75 (2d, J ) 7.8, 2H, H-1 and H-2), 4.71
(d, J ) 5.7, 1H, H-5), 4.13-4.16 (m, 1H, H-6), 3.85 (s, 3H,
O-Me), 3.20 (m, 1H, H-9), 3.07 (d, J ) 18.3, 1H, H-10), 2.47 (s,
3H, N-Me); MS (FAB) m/z 456 [M + H]+.
17-(Cyclop r op ylm et h yl)-3-m et h oxy-4,5r-ep oxy-7r-(p -
br om op h en yl)m or p h in a n -6r-ol (14b). Compound 11b (480
mg, 0.97 mmol) and NaBH4 (750 mg, 19.5 mmol) in EtOH (35
mL) gave 14b (485 mg, 99%): 1H NMR (CDCl3) δ 7.34-7.37
(d, J ) 8.4, 2H, H-Ar), 7.04-7.07 (d, J ) 8.4, 2H, H-Ar), 6.64-
6.74 (2d, J ) 8.7, 2H, H-1 and H-2), 4.71 (d, J ) 5.4, 1H, H-5),
4.14-4.16 (m, 1H, H-6), 3.85 (s, 3H, O-Me), 3.45 (m, 1H, H-9),
2.96 (d, J ) 18.3, 1H, H-10), 0.85 (m, 1H, H-19), 0.52-55 (m,
2H, H-20 and H-21), 0.14-0.17 (m, 2H, H′-20 and H′-21); MS
(FAB) m/z 496 [M + H]+.
17-Meth yl-3-m eth oxy-4,5r-ep oxy-7r-(m -n itr op h en yl)-
m or p h in a n -6r-ol (15a ). Compound 12a (123 mg, 0.29 mmol)
treated with NaBH4 (250 mg, 6.5 mmol) in EtOH (15 mL)
furnished 15a (124 mg, 99%): 1H NMR (CDCl3) δ 8.06-8.07
(m, 2H, H-Ar), 7.50-7.54 (m, 1H, H-Ar), 7.40-7.46 (m, 1H,
H-Ar), 6.70-6.80 (2d, 2H, J ) 9.1, H-1 and H-2), 4.74 (d, J )
5.4, 1H, H-5), 4.20 (m, 1H, H-6), 3.86 (s, 3H, O-Me), 3.30 (m,
1H, H-9), 3.09 (d, 1H, J ) 19.7, H-10R), 2.53(s, 3H, N-Me);
MS (FAB) m/z 422 [M + H]+, m/z 420 [M - H]-.
17-(Cyclop r op ylm et h yl)-4,5r-ep oxy-7r-(p -b r om op h e-
n yl)m or p h in a n -3,6r-d iol (6b). Obtained from 14b (480 mg,
0.96 mmol) in 52% yield (240 mg) and employing 10%
methanol in chloroform as the chromatography eluant: 1H
NMR (CDCl3) δ 7.31-7.34 (d, J ) 8.4, 2H, H-Ar), 6.97-7.00
(d, J ) 8.4, 2H, H-Ar), 6.55-6.70 (2d, J ) 8.1, 2H, H-1 and
H-2), 4.63 (d, J ) 5.4, 1H, H-5), 4.03-4.05 (m, 1H, H-6), 3.65
(m, 1H, H-9), 1.02 (m, 1H, H-19), 0.58-0.60 (m, 2H, H-20 and
H-21), 0.23-0.29 (m, 2H, H′-20 and H′-21); MS (FAB) m/z 482
[M + H]+, m/z 480.1 [M - H]-. Compound 6b (240 mg, 0.50
mmol) was dissolved in CHCl3-MeOH (10:1) and treated with
excess ethereal HCl. The solution was left overnight, and the
precipitate was recrystallized from i-PrOH-Et2O twice to
afford 6b‚HCl (96 mg, 38%), mp 220 °C dec. Anal. (C26H28
NO3Br‚HCl) C, H, N.
-
17-Meth yl-4,5r-ep oxy-7r-(m -n itr op h en yl)m or p h in a n -
3,6r-d iol (7a ). Prepared from 15a (120 mg, 0.28 mmol) in
78% (98 mg,) and using 10% methanol in chloroform as the

Synthesis of 7-Arylmorphinans
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 16 3097
chromatography eluant: 1H NMR (CDCl3) δ 7.99-8.01 (m, 2H,
H-Ar), 7.33-7.46 (m, 2H, H-Ar), 6.52-6.61 (2d, J ) 8.4, 2H,
H-1 and H-2), 4.61 (d, J ) 5.4, 1H, H-5), 4.05-4.08 (m, 1H,
H-6), 3.16-3.17 (m, 1H, H-9), 2.97 (d, 1H, J ) 18.3, H-10â),
2.70-2.74 (m, 1H, H-7), 2.40 (s, 3H, N-Me), 2.32-2.44 (m, 2H,
H-14, H-10R), 1.69-1.83 (m, 1H, H-8â), 1.31-1.36 (m, 1H,
H-8R); 13C NMR (CD3COCD3) δ 20.94 (s), 23.91 (s), 36.94 (s),
41.75 (p), 43.16 (t), 43.63 (q), 45.71 (t), 47.99 (s), 60.35 (t), 71.39
(t), 90.93 (t), 118.47 (t), 120.17 (t), 122.14 (t), 123.92 (t), 129.44
(t), 130.17 (q), 135.28 (t), 138.69 (q), 145.74 (q), 145.87 (q),
148.69 (q); MS (FAB) m/z 409 [M + H]+, 407 [M - H]-.
Compound 7a was dissolved in CHCl3 (2 mL) and treated with
ethereal HCl to furnish a light yellow solid that was recrystal-
lized from MeOH-Et2O to afford 7a ‚HCl, mp 228 °C dec.
Anal. (C23H24N2O5‚HCl‚H2O) C, H, N, Cl.
17-(Cyclopr opylm eth yl)-4,5r-epoxy-7r-(m-n itr oph en yl)-
m or p h in a n -3,6r-d iol (7b). Prepared from 15b (220 mg,
0.476 mmol) in 61% (130 mg) yield, using 5% EtOH in CHCl3
as chromatographic eluant: 1H NMR (CDCl3) δ 7.99-8.01 (m,
2H, H-Ar), 7.33-7.46 (m, 2H, H-Ar), 6.52-6.61 (2d, J ) 8.4,
2H, H-1 and H-2), 4.61 (d, J ) 5.4, 1H, H-5), 4.05-4.08 (m,
1H, H-6), 3.16-3.17 (m, 1H, H-9), 2.97 (d, 1H, J ) 18.3, H-10â),
2.70-2.74 (m, 1H, H-7), 2.40 (s, 3H, N-Me), 2.32-2.44 (m, 2H,
H-14, H-10R), 1.69-1.83 (m, 1H, H-8â), 1.31-1.36 (m, 1H,
H-8R); 13C NMR (CD3COCD3) δ 4.79 (s), 21.43 (s), 23.87 (s),
42.12 (q), 45.58 (t), 46.76 (s), 57.96 (t), 59.14 (s), 60.04 (s), 71.30
(t), 90.91 (t), 118.59 (t), 118.67 (t), 120.08 (t), 122.13 (t), 123.94
(t), 129.44 (q), 130.18 (t), 135.32 (q), 138.65 (q), 145.73 (q),
145.96 (q), 148.66 (q); MS (FAB) m/z 449 [M + H]+. The
hydrochloride was prepared in CHCl3-MeOH (10:2) and
crystallized from i-PrOH-Et2O; mp 235 °C dec. Anal.
(C26H28N2O4‚HCl‚1.7H2O) C, H, N, Cl.
17-Met h yl-3-m et h oxy-4,5r-ep oxy-6r-O-a cet yl-7r-p h e-
n ylm or p h in a n (16a ). Compound 13a (400 mg, 1.05 mmol)
was mixed with acetic anhydride (5 mL) and glacial acetic acid
(10 mL) under N2. The resulting solution was stirred at 110
°C (oil bath) for 24 h. After removal of solvents under reduced
pressure, the residue was taken up in a small volume of
chloroform and washed with saturated aqueous NaHCO3. The
organic layer was removed under reduced pressure to afford
16a (380 mg, 99%): 1H NMR (CDCl3, 200 MHz) δ 7.01-7.22
(m, 5H, H-Ph), 6.69-6.77 (2d, J ) 8.4, 2H, H-1 and H-2),
5.46-5.50 (m, 1H, H-6), 4.79 (d, J ) 2.1, 1H, H-5), 3.83 (s,
3H, O-Me), 3.30 (m, 1H, H-9), 3.12 (d, J ) 19, 1H, H-10), 2.76
(s, 3H, N-Me).
17-(Cyclop r op ylm et h yl)-3-m et h oxy-4,5r-ep oxy-6r-O-
a cetyl-7r-p h en ylm or p h in a n (16b). By the same procedure
employed for preparing 16a , a solution of 13b (600 mg, 1.44
mmol) stirred at 110 °C with acetic anhydride (8 mL) and
AcOH (16 mL) furnished crude 16b (570 mg, 99%): 1H NMR
(CDCl3) δ 7.04-7.22 (m, 5H, H-Ph), 6.61-6.70 (2d, J ) 8.1,
2H, H-1 and H-2), 5.47-5.50 (dd, J ) 2.1, 5.8, 1H, H-6), 4.77
(d, J ) 5.8, 1H, H-5), 3.83 (s, 3H, O-Me), 3.50 (m, 1H, H-9),
2.96 (d, J ) 18.3, 1H, H-10), 0.90 (m, 1H, H-19), 0.51-0.54
(m, 2H, H-20 and H-21), 0.14-0.15 (m, 2H, H′-20 and H′-21);
MS (FAB) m/z 460 [M + H]+.
EtOH in CHCl3: 1H NMR (CDCl3) δ 7.04-7.22 (m, 5H, H-Ph),
6.61-6.70 (2d, J ) 8.1, 2H, H-1 and H-2), 5.47-5.50 (dd, J )
2.1, 5.8, 1H, H-6), 4.77 (d, J ) 5.8, 1H, H-5), 3.50 (m, 1H, H-9),
2.96 (d, J ) 18.3, 1H, H-10), 0.90 (m, 1H, H-19), 0.51-0.54
(m, 2H, H-20 and H-21), 0.14-0.15 (m, 2H, H′-20 and H′-21);
MS (FAB) m/z 386.2 [M + H]+. Treatment of the base with
ethereal HCl gave the hydrochloride salt, which was recrystal-
lized twice from i-PrOH-Et2O (138 mg, 61%): mp 210 °C dec;
1H NMR (CD3OD) δ 7.08-7.23 (m, 5H, H-Ph), 6.71 (s, 2H, H-1
and H-2), 5.45-5.48 (dd, J ) 1.8, 6.2, 1H, H-6), 4.92 (d, J )
6.2, 1H, H-5), 3.85 (m, 1H, H-9), 1.52 (s, 3H, H-Ac); MS (FAB)
m/z 446 [M + H]+. Anal. (C28H31NO4‚HCl) C, H, N, Cl.
3-O-Ben zyl-7r-p h en yln a ltr exon e (18). The same proce-
dure employed for 10b was used to carry out the reaction
between 3-O-benzylnaltrexone6 (17) (0.84 g, 1.95 mmol),
hexamethyldisilazane (1.8 mL, 8.56 mmol), n-butyllithium (2.4
M, 2.0 mL, 4.8 mmol), and diphenyliodonium iodide (1.4 g, 3.4
mmol) in DMF (80 mL). The crude product was purified by
centrifugal chromatography on silica gel, eluting with ethyl
acetate-hexane (2:8) to afford compound 18 (85 mg, 8.5%):
1H NMR (CDCl3, 200 MHz) δ 7.23-7.47 (m, 10H, H-Ph), 6.53-
6.73 (2d, 2H, J ) 8, H-1 and H-2), 5.16-5.32 (dd, 2H, J ) 12
Hz, CH2-Ph), 4.70 (s, 1H, H-5), 2.95-3.48 (m, H-10 and H-7),
0.88 (m, 1H, H-19), 0.51-0.60 (m, 2H, H-20 and H-21), 0.14-
0.16 (m, 2H, H′-20 and H′-21).
7r-P h en yln a ltr exon e (3c). A solution of 18 (200 mg, 0.39
mmol) and concentrated HCl (4 mL) in glacial acetic acid (4
mL) was stirred at 85 °C for 1 h. The solvent was evaporated
with the aid of ethanol, and the residue was taken up in ethyl
acetate (25 mL) and washed with aqueous KHCO3 (5%) and
then water. After the ethyl acetate extract was dried over
MgSO4, it was evaporated to dryness. The residue (∼160 mg)
was applied to a silica gel (1 mm) rotor plate and eluted with
ethyl acetate-hexane (2:8) to afford compound 3c (120 mg,
73%): 1H NMR (CDCl3, 200 MHz) δ 7.20-7.24 (m, 3H, H-Ph),
6.97-7.01 (m, 2H, H-Ph), 6.52-6.70 (2d, 2H, J ) 9.4, H-1 and
H-2), 4.91 (s, 1H, H-5), 4.40-4.48 (dd, 1H, J ) 5.4, 12.6, H-7),
3.26 (d, J ) 5.8, H-9), 3.13 (d, 2H, J ) 18.4, H-10), 0.895 (m,
1H, H-19), 0.51-0.55 (m, 2H, H-20 and H-21), 0.17-0.19 (m,
2H, H′-20 and H′-21); 13C NMR (CDCl3) δ 3.31, 9.22, 22.26,
28.09, 28.15, 30.63, 39.07, 43.56, 50.07, 51.43, 58.87, 62.11,
90.86, 119.21, 126.54, 127.85, 128.71, 222.00; MS (FAB) m/z
418 [M + H]+. Compound 3c (100 mg, 0.22 mmol) was treated
with ethereal HCl to afford 3c‚HCl (50 mg) after recrystalli-
zation three times from i-PrOH-Et2O. Anal. (C26H27NO4‚
HCl‚2.5H2O) C, H, N.
Ack n ow led gm en t. This work is supported by the
National Institution on Drug Abuse. We thank Michael
Powers, Veronika Phillips, and Idalia Sanchez for
capable technical assistance.
Refer en ces
(1) Dhawan, B. N.; Cesselin, F.; Raghubir, R.; Reisine, T.; Bradley,
P. B.; Portoghese, P. S.; Hamon, M. International Union of
Pharmacology. XII. Classification of Opioid Receptors. Pharma-
col. Rev. 1996, 48, 567-592.
17-Met h yl-4,5r-ep oxy-7r-p h en ylm or p h in a n -3,6r-d iol
6-Aceta te (8a ). Obtained from 16a (380 mg, 1.05 mmol) in
52% (189 mg) yield, using 10% EtOH in CHCl3 as eluant in
the chromatography step: 1H NMR (CDCl3) δ 7.00-7.25 (m,
5H, H-Ph), 6.59-6.71 (2d, J ) 8.2, 2H, H-1 and H-2), 5.39-
5.42 (m, 1H, H-6), 4.73 (d, J ) 6.2, 1H, H-5), 3.26 (m, 1H, H-9),
3.06 (d, J ) 18.6, 1H, H-10), 2.44 (s, 3H, N-Me), 1.51 (s, 3H,
H-Ac); MS (FAB) m/z 406 [M + H]+. The base (180 mg, 0.54
mmol) was dissolved in CHCl3 (2 mL) and MeOH (0.2 mL) and
treated with ethereal HCl to afford the crude hydrochloride,
which was recrystallized twice from MeOH-Et2O (144 mg,
(2) Portoghese, P. S. Bivalent Ligands and the Message-Address
Concept in the Design of Selective Opioid Receptor Antagonists.
Trends Pharmacol. Sci. 1989, 10, 230-235.
(3) Schwyzer, R. ACTH: A Short Introductory Review. Ann. N.Y.
Acad. Sci. 1977, 297, 3-26.
(4) Portoghese, P. S. The Role of Concepts in Structure-Activity
Relationship Studies of Opioid Ligands. J . Med. Chem. 1992,
35, 1928-1937.
(5) 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.
(6) Portoghese, P. S.; Moe, S. T.; Takemori, A. E. A Selective δ1
Opioid Receptor Agonist Derived from Oxymorphone. Evidence
for Separate Recognition Sites for δ1 Opioid Receptor Agonists
and Antagonists. J . Med. Chem. 1993, 36, 2572-2574.
(7) Gao, P.; Portoghese, P. S. Monophenylation of Morphinan-6-ones
with Diphenyliodonium Iodide. J . Org. Chem. 1995, 60, 2276-
2278.
1
70%): mp 275 °C dec; H NMR (CDCl3) δ 7.08-7.23 (m, 5H,
H-Ph), 6.71 (s, 2H, H-1 and H-2), 5.45-5.47 (dd, J ) 1.8, 6.2,
1H, H-6), 4.92 (d, J ) 6.2, 1H, H-5), 3.85 (m, 1H, H-9), 2.91 (s,
3H, N-Me). Anal. (C25H27NO4‚HCl) C, H, N, Cl.
17-(Cyclop r op ylm eth yl)-4,5r-ep oxy-7r-p h en ylm or p h i-
n a n -3,6r-d iol 6-Aceta te (8b). Prepared from 16b (570 mg,
1.43 mmol) in 54% yield following chromatography using 10%
(8) Sayre, L. M.; Portoghese, P. S. Stereospecific Synthesis of 6R-
and 6â-Amino Derivatives of Naltrexone and Oxymorphone. J .
Org. Chem. 1980, 45, 3366-3368.

3098 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 16
Gao et al.
(9) Gao, P.; Portoghese, P. S. Boron Tribromide-Catalyzed Rear-
rangement of 7,7-Diphenylhydromorphone to 6,7-Diphenylmor-
phine: A Novel Conversion of Ketones to Allylic Alcohols. J . Org.
Chem. 1996, 61, 2466-2469.
(20) Fang, X.; Larson, D. L.; Portoghese, P. S. 7-Spirobenzocyclohexyl
Derivatives of Naltrexone, Oxymorphone, and Hydromorphone
as Selective Opioid Receptor Ligands. J . Med. Chem. 1997, 40,
3064-3070.
(21) Farouz-Grant, F.; Portoghese, P. S. Pyrrolomorphinans as δ
Opioid Receptor Antagonists. The Role of Steric Hindrance in
Conferring Selectivity. J . Med. Chem. 1997, 40, 1977-1981.
(22) Kshirsagar, T. A.; Fang, X.; Portoghese, P. S. 14-Desoxy Ana-
logues of Naltrindole and 7-Spiroindanyloxymorphone. The Role
of the 14-Hydroxy Group at δ Opioid Receptors. J . Med. Chem.
(submitted).
(23) Casy, A. F.; Parfitt, R. T. Opioid Analgesics; Plenum Press: New
York, 1986; pp 54-68.
(24) Abdelhammid, E. E.; Sultana, M.; Portoghese, P. S.; Takemori,
A. E. Selective Blockage of Delta Opioid Receptors Prevents the
Development of Morphine Tolerance and Dependence in Mice.
J . Pharmacol. Exp. Ther. 1991, 258, 299.
(25) Fundytus, M. E.; Schiller, P. W.; Shipiro, M.; Weltrowska, G.;
Coderre, T. J . Attenuation of Morphine Tolerance and Depen-
dence with Highly Selective Delta Opioid Receptor Antagonist
TIPP. Eur. J . Pharmacol. 1995, 286, 105-108.
(26) Suzuki, T.; Tsuji, M.; Mori, T.; Misawa, M.; Nagase, H. Effect of
Naltrindole on the Development of Physical Dependence on
Morphine in Mice: A Behavioral and Biochemical Study. Life
Sci. 1995, 57, PL247-PL252.
(10) Lord, J . A. H.; Waterfield, A. A.; Hughes, J .; Kosterlitz, H. W.
Endogenous Opioid Peptides: Multiple Agonists and Receptors.
Nature 1977, 267, 495-499.
(11) Rang, H. B. Stimulant Actions of Volatile Anaesthetics on
Smooth Muscle. Br. J . Pharmacol. 1964, 22, 356-365.
(12) Fournie-Zaluski, M.-C.; Gacel, G.; Maigret, B.; Premilat, S.;
Roques, B. P. Structural Requirements for Specific Recognition
of Mu or Delta Opiate Receptors. Mol. Pharmacol. 1981, 20,
484-491.
(13) Tulunay, F. C.; Takemori, A. E. The Increased Efficacy of
Narcotic Antagonists Induced by Various Narcotic Analgesics.
J . Pharmacol. Exp. Ther. 1974, 190, 395-400.
(14) Mosberg, H. I.; Hurst, R.; Hurby, V. I.; Gee, K.; Yamamura, H.
I.; Galligan, J . J .; Burks, T. F. Bis-penicillamine Enkephalins
Show Pronounced Delta Receptor Sensitivity. Proc. Natl. Acad.
Sci. U.S.A. 1983, 80, 5871-5874.
(15) Handa, B. K.; Lane, A. C.; Lord, J . A. H.; Morgan, B. A.; Rance,
M. J .; Smith, C. F. C. Analogs of â-LPH61-64 Possessing
Selective Agonist Activity at Mu-Opiate Receptors. Eur. J .
Pharmacol. 1981, 70, 531-540.
(16) von Voigtlander, P. F.; Lahti, R. A.; Ludens, J . H. U-50488: A
Selective and Structurally Novel Non-mu (Kappa) Opioid Ago-
nist. J . Pharmacol. Exp. Ther. 1983, 224, 7-12.
(17) Sofuoglu, M.; Portoghese, P. S.; Takemori, A. E. Differential
Antagonism of Delta Opioid Agonist by Naltrindole and its
Benzofuran Analogue (NTB) in Mice: Evidence for Delta Opioid
Receptor Types. J . Pharmacol. Exp. Ther. 1991, 275, 676-680.
(18) J iang, Q.; Takemori, A. E.; Sultana, M.; Portoghese, P. S.; Bowen,
W. D.; Mosberg, H. I.; Porreca, F. Differential Antagonism of
Opioid Delta Antinociception by [DAla2,Leu5Cys6]Enkephalin
and Naltrindole 5′-Isothiocyanate: Evidence for Delta Receptor
Subtypes. J . Pharmacol. Exp. Ther. 1991, 257, 1069-1075.
(19) Portoghese, P. S.; Sultana, M.; Moe, S. T.; Takemori, A. E.
Synthesis of Naltrexone-Derived δ-Opioid Antagonists. Role of
Conformation of the δ Address Moiety. J . Med. Chem. 1994, 37,
579-585.
(27) Gates, M.; Montzka, T. A. Some Potent Morphine Antagonists
Possessing High Analgesic Activity. J . Med. Chem. 1964, 7, 127-
131.
(28) Beringer, M.; Dexler, M.; Gindler, M.; Lumpkin, C. C. Diaryli-
odonium Salts. I. Synthesis. J . Am. Chem. Soc. 1953, 75, 2705-
2708.
(29) Beringer, M.; Talk, R. A.; Karniol, M.; Lillien, I.; Masullo, G.;
Mausner, M.; Sommer, E. Diaryliodonium Salts. IX. The Syn-
thesis of Substituted Diphenyliodonium Salts. J . Am. Chem. Soc.
1959, 81, 342-351.
J M9802214