Synthesis and opioid receptor binding properties of a highly potent 4-hydroxy analogue of naltrexone

Article abstract of DOI:10.1016/j.bmcl.2005.02.032

Very high affinity for opioid receptors (e.g., K_i = 0.052 nM for μ) has been observed in the rationally designed naltrexone analogue 5. SAR and physical data supports the hypothesis that the 4-OH group of 5 stabilizes the 3-carboxamido group in the putative bioactive conformation.

Full text of DOI:10.1016/j.bmcl.2005.02.032

Bioorganic & Medicinal Chemistry Letters 15 (2005) 2107–2110
Synthesis and opioid receptor binding properties of a highly
potent 4-hydroxy analogue of naltrexone
Mark P. Wentland,a,* Qun Lu,a Rongliang Lou,a Yigong Bu,a Brian I. Knappb and
Jean M. Bidlackb
aDepartment of Chemistry and Chemical Biology, Rensselaer Polytechnic Institute, Troy, NY 12180, USA
bDepartment of Pharmacology and Physiology, School of Medicine and Dentistry, University of Rochester, Rochester, NY 14642, USA
Received 19January 2005; revised 9February 2005; accepted 11 February 2005
Available online 16 March 2005
Abstract—Very high affinity for opioid receptors (e.g., Ki = 0.052 nM for l) has been observed in the rationally designed naltrexone
analogue 5. SAR and physical data supports the hypothesis that the 4-OH group of 5 stabilizes the 3-carboxamido group in the
putative bioactive conformation.
Ó 2005 Elsevier Ltd. All rights reserved.
As part of our broad goal to identify long-acting opioid
CH
CH
2
2
receptor interactive agents useful in the treatment of co-
caine and heroin addiction in humans, we recently re-
ported the synthesis and opioid receptor binding data
of 3-desoxy-3-carboxamidonaltrexone 1.1 This novel
analogue of naltrexone (2) where its phenolic OH group,
typical of opioid receptor interactive agents, was re-
placed with a carboxamido group displayed high affinity
for the l opioid receptor. Relative to naltrexone, car-
boxamide 1 had approximately one order of magnitude
reduced affinity for l. This SAR was not in synch with
that noted for 2,6-methano-3-benzazocines [e.g., cyclaz-
ocine (4)] where comparable or enhanced affinity was
seen for the 8-carboxamides [e.g., 8-carboxamidocyc-
lazocine (8-CAC, 3)] relative to the corresponding phe-
nolic OH parent (Fig. 1).2
N
N
OH
CH
3
3
8
CH
3
4
O
X
O
X
1: X = CONH
2: X = OH
3: X = CONH
2
2
4: X = OH
Figure 1. Structures of lead compounds for this study.
proton NMR data for 1 and 3 in CDCl3. For 1, we ob-
served a strong intramolecular H-bond that bridges the
carboxamido group to the neighbouring ether oxygen of
the furan ring (NH protons appear as singlets at d 5.77
and 7.16 that do not coalesce upon warming to 45 °C).
These NMR data contrast that of 3 where the two car-
boxamide NHÕs appear as broad singlets (d 5.63 and
6.04) and coalesce into a very broad singlet (d 5.80) be-
tween 30 and 40 °C. 8-CAC (3) has no neighbouring fur-
an ring with which the carboxamide can
intramolecularly H-bond, which results in little or no
preference/barrier for/between conformations 3a/b in
the unbound state. This knowledge, coupled with the
divergent SAR noted above, led us to reason that
the l opioid receptor can bind 3b, the putative bioac-
tive conformation, with no conformational energy
penalty.
To explain this divergent SAR, we hypothesized that the
ether oxygen of the furan ring in the 4,5a-epoxymorph-
inan derivative 1 stabilizes a conformation (1a in Fig. 2)
that is different than the putative bioactive conforma-
tion 1b and that an energy penalty would be required
for the more stable (unbound state) conformer 1a to at-
tain the bioactive conformation resulting in reduced
binding affinity. Evidence supporting 1b as the bioactive
conformation was derived, in part, from comparative
Keywords: Opiate; SAR.
*
Corresponding author. Tel.: +1 518 276 2234; fax: +1 518 276
4887; e-mail: wentmp@rpi.edu
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.02.032

2108
M. P. Wentland et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2107–2110
CH2
CH2
CH3
CH2
OH
CH2
OH
N
N
N
N
CH3
8
8
3
H
H
CH3
CH3
3
H
H
O
O
N
O
N
O
O
O
N
H
O
O
N
H
1b
1a
3a
3b
H
H
CH2
OH
CH2
CH2
OH
CH2
N
N
N
N
OH
OH
3
3
3
H
H
3
H
H
O
O
N
O
H
N
OCH3
O
O
H
O
O
CH3
O
O
O
O
N
H
N
H
5b
6b
5a
6a
H
H
Figure 2. Conformational preferences of carboxamide groups.
In hopes of gaining additional insight into the SAR of
these carboxamido-containing opiates, we have designed
and synthesized a series of analogues of 1 as probes to
help further define its bioactive conformation. We rea-
soned that higher affinity would arise by rigidifying the
carboxamido group of 1 into a conformational state
similar to 1b. This could be accomplished by, for exam-
ple, incorporating the carboxamido group into a cyclic
structure with an appropriate 4-substituent via covalent
or non-covalent bonds. We now wish to report our re-
sults showing that such a rigidified naltrexone analogue
5 characterized by having a 4-OH group has very high
affinity for l and j opioid receptors that is 14- and 48-
fold, respectively, higher than 1. We also made and eval-
uated novel naltrexone analogues 6 (3-CONH2–4-
OCH3) and 8 (3,4-dihydroxy) as well as the N-cyclobu-
tylmethyl derivative 13 to aid in the interpretation of
these SAR data.
determine the effect of 4-OH substitution on naltrexone
itself, we made novel catechol 8 as the HBr salt in 73%
yield by treating known compound 73,4 with BBr3 in
chloroform (Scheme 2).
An additional 3-carboxamido-4-hydroxymorphinan
analogue, 13, was made to determine if changes (cyclo-
propylmethyl ! cyclobutylmethyl) at the basic nitrogen
would affect binding. Commercially available nalbu-
phine (9) was converted to its corresponding 3-triflate
13
ester 10 in 98% yield using PhN(Tf)2 in Et3N/CH2Cl2.
14
Treating 10 with Zn(CN)2/Pd(PPh)4 provided nitrile
11 in 83% yield, which was oxidized [(COCl)2, Et3N,
DMSO] to ketone 12 in 92% yield. Target 13 was made
CH
CH
2
2
N
N
4-Hydroxy-morphinans can be easily made via furan
ring cleavage of a wide variety of 4,5a-epoxymorph-
inan substrates using reductive,3–8 based-induced elimi-
nation,4,9 photolytic,10 and acid catalysis11 methods.
As shown in Scheme 1, we used one of these reductive
cleavage methods4 (Zn/HOAc/HCl) to prepare target 5
from 1 in 50% yield. The corresponding 4-methoxy tar-
get 6 was made in 72% yield by treating 5 with
(CH3)3SiCHN2, Et3N in methanol/acetonitrile.12 To
OH
OH
i
.
HBr
O
O
CH O
3
HO
OH
OH
7
8
Scheme 2. Reagents and conditions: (i) BBr3, CHCl3, 25 °C, 30 min.
CH2
N
CH2
CH2
N
N
OH
OH
OH
i
ii
O
O
O
O
H2N
H2N
OH
H2N
OCH3
O
O
O
1
5
6
Scheme 1. Reagents and conditions: (i) Zn, 37% HCl, HOAc, 125 °C, 15 min; (ii) (CH3)3SiCHN2, Et3N, MeOH, MeCN, 25 °C, 24 h.

M. P. Wentland et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2107–2110
2109
CH
CH
2
2
CH
2
N
N
N
OH
OH
OH
iii
iv
(11 → 12)
O
O
O
OH
H N
2
O
X
NC
OH
O
9: X = OH
12
13
i
10: X = OSO CF
2
3
ii
11: X = CN
Scheme 3. Reagents and conditions: (i) PhN(Tf)2, Et3N, CH2Cl2, 25 °C, 16 h; (ii) Zn(CN)2, Pd(PPh)4, DMF, 135 °C, 24 h; (iii) (COCl)2, Et3N,
DMSO, ꢀ78 °C, 20 min; (iv) Zn, 37% HCl, HOAc, 125 °C, 3 h.
in 71% yield from 12 via a reductive/hydrolytic process
using Zn in HCl/H2O/HOAc (Scheme 3).
ing the furan ring intact, 5 has 14-, 212- and 48-fold
higher affinity for l, d and j, respectively. These results
fit nicely with our pharmacophore hypothesis that the 4-
OH (as H-bond donor) stabilizes the Ôcarboxamide-
acceptorÕ putative bioactive conformation 5b. Addi-
tional physical and SAR data were generated to pre-
clude that (a) the 4-OH stabilizes the alternate
Ôcarboxamide-donorÕ conformation 5a and/or (b) the
newly introduced 4-OH group enhances binding via di-
rect interaction with receptor protein. Methyl ether 6,
an analogue close to 5 that is highly unlikely to adopt
Affinities of target compounds for human l, d and j opi-
oid receptors stably expressed in CHO cells were as-
sessed by generating Ki values using well-documented
receptor binding assays.15 These data are summarized
in Table 1. Ring-opened target compound 5 displays
very high affinity for l and j (Ki = 0.052 nM and
0.23 nM, respectively), and good affinity (Ki = 2.6 nM)
for the d receptor. Compared to 1, the original lead hav-
Table 1. Opioid receptor binding data for 3-carboxamido-4-hydroxymorphinans and related compounds
CH
CH
CH
CH
2
2
2
2
N
N
N
N
OH
OH
OH
OH
4
4
3
3
O
O
OH
O
OCH
O
OH
O
X
X
H N
2
H N
2
3
O
O
A
B
6
13
Compd
Ki (nM S.E.)a
[3H]Naltrindole (d)
X
[3H]DAMGO (l)
[3H]U69,593 (j)
1b,c
A
A
B
CONH2
OH
CONH2
0.71 0.058
0.11 0.006
0.052 0.004
550 40
60 3.2
2.6 0.26
11 0.36
0.19 0.005
0.23 0.018
2 (Naltrexone)c,d
5e
6e
11 0.69480 35
17 4.0
0.13 0.0083
23 2.6
8e
B
OH
130 6.6
4.2 0.36
2.2 0.16
0.27 0.013
13e
a Binding assays used to screen compounds are similar to those previously reported (see Ref. 15). Membrane protein from CHO cells that stably
expressed one type of the human opioid receptor were incubated with 12 different concentrations of the compound in the presence of either 1 nM
[3H]U69,593 (l), 0.25 nM [3H]DAMGO (d) or 0.2 nM [3H]naltrindole (j) in a final volume of 1 mL of 50 mM Tris–HCl, pH 7.5 at 25 °C.
Incubation times of 60 min were used for [3H]U69,593 and [3H]DAMGO. Because of a slower association of [3H]naltrindole with the receptor, a 3 h
incubation was used with this radioligand. Samples incubated with [3H]naltrindole also contained 10 mM MgCl2 and 0.5 mM phenylmethylsulfonyl
fluoride. Non-specific binding was measured by inclusion of 10 lM naloxone. The binding was terminated by filtering the samples through
Schleicher & Schuell No 32 glass fibre filters using a Brandel 48-well cell harvester. The filters were subsequently washed three times with 3 mL of
cold 50 mM Tris–HCl, pH 7.5, and were counted in 2 mL Ecoscint A scintillation fluid. For [3H]naltrindole and [3H]U69,593 binding, the filters
were soaked in 0.1% polyethylenimine for at least 60 min before use. IC50 values will be calculated by least squares fit to a logarithm-probit analysis.
Ki values of unlabelled compounds were calculated from the equation Ki = (IC50)/1 + S where S = (concentration of radioligand)/(Kd of radioli-
gand)—see Ref. 20. Data are the mean SEM from at least three experiments performed in triplicate.
b See Ref. 1.
c In our original work (Ref. 1), binding affinities for compounds 1 and 2 were measured for opioid receptors from guinea pig brain membranes.
Relative affinities for 1 and 2 using human or guinea pig receptors were very similar.
d Obtained from commercial sources.
e Proton NMR, IR and MS were consistent with the assigned structures of all new compounds. C, H and N elemental analyses were obtained for all
new targets and most intermediates and were within 0.4% of theoretical values.

2110
M. P. Wentland et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2107–2110
the Ôcarboxamide-acceptorÕ conformation 6b due to
incompatibility of the oxygenÕs lone pairs, has substan-
tially lower (212-, 185- and 100-fold) affinity than 5 for
l, d and j, respectively.16 Proton NMR data in CDCl3
for 5 and 6 lend additional support of our pharmaco-
phore hypothesis. Chemical shift difference (ca.
0.5 ppm) and the broad line shape for the two NHs of
5 is very different than seen with 1 but very close to that
observed for 3. Also, the proton of the phenolic hydro-
xyl of 5 is highly deshielded (d 13.3) and appears as a
sharp singlet, properties characteristic of participation
in an intramolecular H-bond as donor (i.e., Ôcarboxam-
ide-acceptorÕ conformation 5b).17 Results from NMR
dilution experiments with 5 also confirms the relative
stability of conformer 5b. At five concentrations (100,
50, 10, 5 and 1 mM) of 5 in CDCl3, no change was noted
in the chemical shift or line shape of the phenolic H
whereas the amide HÕs (broad singlets at d 6.2 and 6.7
at 100 mM) coalesced into a broad singlet at d 5.8 at
1.0 mM; both observations indicate that conformer 5b
(Ôcarboxamide-acceptorÕ) is highly stabilized relative to
5a.18 The NHÕs of the poorly active methyl ether 6 have
magnetic environments very similar to those of 1 signify-
ing the higher stability of conformer 6a (Ôcarboxamide-
donorÕ) relative to 6b. Like naltrexone, compounds 1
and 5 were pure antagonists at the l receptor, as mea-
sured by its inhibition of DAMGO-stimulated
[35S]GTPcS binding (unpublished results).
evant to the actual biological system, research aimed at
the design, synthesis and evaluation of novel carboxa-
mido-containing opiates where the CONH2 (or isosteric
group) is rigidified in the putative bioactive conforma-
tion via covalent attachments is underway in our
laboratories.
Acknowledgements
We gratefully acknowledge the contributions of Dr.
Xufeng Sun and mass spectroscopist Dr. Dmitri Zago-
revski (both of Rensselaer) and the technical assistance
provided by Dana J. Cohen of the University of Roches-
ter. Funding of this research was from NIDA (DA12180
and KO5-DA00360).
References and notes
1. Wentland, M. P.; Lou, R.; Dehnhardt, C. M.; Duan, W.;
Cohen, D. J.; Bidlack, J. M. Bioorg. Med. Chem. Lett.
2001, 11, 1717.
2. Wentland, M. P.; Lou, R.; Ye, Y.; Cohen, D. J.;
Richardson, G. P.; Bidlack, J. M. Bioorg. Med. Chem.
Lett. 2001, 11, 623.
3. Brine, G. A.; Boldt, K. G.; Coleman, M. L.; Bradley, D.
J.; Carroll, F. I. J. Org. Chem. 1978, 43, 1555.
4. Coop, A.; Rothman, R. B.; Dersch, C.; Partilla, J.;
Porreca, F.; Davis, P.; Jacobson, A. E.; Rice, K. C. J.
Med. Chem. 1999, 42, 1673.
5. Schmidhammer, H.; Krassnig, R.; Greiner, E.; Traynor, J.
R. Heterocycles 1998, 49, 489.
6. Freund, M.; Speyer, E.; Guttmann, E. Berichte 1920, 53B,
2250.
7. Abu-Khurmah, M.; Oelschlager, H.; Volke, J. Pharmazie
1998, 53, 524.
8. Sawa, Y. K.; Tsuji, N.; Okabe, K.; Miyamoto, T.
Tetrahedron 1965, 21, 1121.
9. Coop, A.; Rice, K. C. Heterocycles 1999, 50, 39 .
10. Schultz, A. G.; Green, N. J.; Archer, S.; Tham, F. S. J.
Am. Chem. Soc. 1991, 113, 6280.
11. Grundt, P.; Martinez-Bermaejo, F.; Lewis, J. W.; Hus-
bands, S. M. J. Med. Chem. 2003, 46, 3174.
12. Aoyama, T.; Terasawa, S.; Sudo, K.; Shioiri, T. Chem.
Pharm. Bull. 1984, 32, 3759.
13. Hendrickson, J. B.; Bergeron, R. Tetrahedron Lett. 1973,
14, 4607.
14. Kubota, H.; Rice, K. C. Tetrahedron Lett. 1998, 39,
2907.
We also made catechol derivative 8, the ring-opened ver-
sion of naltrexone, to directly assess the contribution of
the 4-OH to binding affinity in the 3-OH subseries and
indirectly gauge if the 4-OH of 5 contributes to high
affinity binding via intramolecular effects (as we hypoth-
esized) or intermolecular interaction(s) with receptor.
Our hypothesis was corroborated when we observed
that compared to naltrexone (2), there is a substantial
reduction (155-fold) in binding affinity for l upon cleav-
age of the furan ring to give 8. This result greatly con-
trasts the corresponding 3-CONH2 pair, 1 and 5,
where the ring-cleaved analogue is 14-fold more potent.
Compound 8 had a 2- and 12-fold lower affinity than 2
for d and j receptors, respectively. Lastly, the cyclobu-
tylmethyl analogue 13, which has the same 3-CONH2,
4-OH motif and proton NMR pattern as 5, has similarly
high affinity (within 2-fold) for all three receptors.
Our results show that 4-OH substitution on morphinans
significantly enhances binding affinity when the 3-substi-
tuent is CONH2 but greatly decreases affinity with a 3-
OH group, the prototypic substituent of opiates. There
are hundreds of 4-hydroxy-morphinans in the literature
with some, for example thebainone,19 dating as far back
as the 1920s. To our knowledge, however, none have a
3-CONH2 or similar group and none have reached the
therapeutic prominence of the corresponding 4,5a-
epoxymorphinans. We have also presented SAR and
proton NMR data showing the benefits of the 4-OH
group with a 3-CONH2 substituent is possibly a conse-
quence of stabilizing an intramolecular H-bonded bioac-
tive conformation where carboxamide is acceptor and 4-
OH is donor. As these NMR studies were conducted in
CDCl3 rather than in an aqueous environment more rel-
15. Neumeyer, J. L.; Zhang, A.; Xiong, W.; Gu, X.; Hilbert, J.
E.; Knapp, B. I.; Negus, S. S.; Mello, N. K.; Bidlack, J. M.
J. Med. Chem. 2003, 46, 5162.
16. The corresponding ring-opened methyl ether (i.e., 3-OH,
4-OMe) of naltrexone (a.k.a. 3-hydroxycyprodime) has
moderate affinity for l (Ki = 6.15 nM), however, this study
did not do a direct comparison with naltrexone: Schmid-
hammer, H.; Jennewein, H. K.; Krassnig, R.; Traynor, J.
R.; Patel, D.; Bell, K.; Froschauer, G.; Mattersberger, K.;
Jachs-Ewinger, C.; Jura, P.; Fraser, G. L.; Kalinin, V. N.
J. Med. Chem. 1995, 38, 3071.
17. Sun, S.; Lees, A. J. Chem. Commun. 2000, 17, 1687.
18. Schopfer, U.; Stahl, M.; Brandl, T.; Hoffmann, R. W.
Angew. Chem., Int. Ed. 1997, 36, 1745.
19. Gulland, J. M.; Robinson, R. J. Chem. Soc. 1923, 123, 998.
20. Cheng, Y. C.; Prusoff, W. H. Biochem. Pharmacol. 1973,
22, 3099.