Investigation of phenolic bioisosterism in opiates: 3-sulfonamido analogues of naltrexone and oxymorphone

Article abstract of DOI:10.1021/ol005561+

formula presented The phenolic hydroxy group of opiate-derived ligands is of known importance for biological activity. On the basis of its putative role as a hydrogen-bonding donor in the interaction with opioid receptors, it was replaced with a sulfonamide group because of their similar pK_a values. The first thebaine-derived 3-amino (8a, 8b) and subsequent sulfonamide analogues (10a, 10b) were synthesized from naltrexone (1a) and oxymorphone (1b) in a linear nine-step synthesis. The sulfonamides were tested in vitro and found inactive.

Full text of DOI:10.1021/ol005561+

ORGANIC
LETTERS
2000
Vol. 2, No. 6
819-821
Investigation of Phenolic Bioisosterism
in Opiates: 3-Sulfonamido Analogues of
Naltrexone and Oxymorphone
Christopher R. McCurdy, Robert M. Jones, and Philip S. Portoghese*
Department of Medicinal Chemistry, College of Pharmacy, UniVersity of Minnesota,
Minneapolis, Minnesota 55455
porto001@maroon.tc.umn.edu
Received January 18, 2000
ABSTRACT
The phenolic hydroxy group of opiate-derived ligands is of known importance for biological activity. On the basis of its putative role as a
hydrogen-bonding donor in the interaction with opioid receptors, it was replaced with a sulfonamide group because of their similar pKa
values. The first thebaine-derived 3-amino (8a, 8b) and subsequent sulfonamide analogues (10a, 10b) were synthesized from naltrexone (1a)
and oxymorphone (1b) in a linear nine-step synthesis. The sulfonamides were tested in vitro and found inactive.
The opioid antagonist, naltrexone (1a), and its receptor
agonist, oxymorphone (1b), are thebaine-derived structures
that share a common opiate-type template. It is well-
established that the phenolic hydroxy group of these and
related opiates is important for recognition by opioid
receptors.1 Indeed, substitution at this position greatly reduces
or abolishes activity. Given the putative role of the phenolic
hydroxy proton as a hydrogen-bonding donor in enhancing
binding, we have investigated the replacement with a group
whose pKa is similar. There is good precedent for such an
approach, as replacement of a phenolic hydroxy with a
sulfonamide group in catecholamines has been reported to
afford ligands which retain the intensity of action and
biological profile for adrenergic receptors.2 This was at-
tributed to the similar pKa value of the sulfonamide and
phenolic hydroxy groups, whose protons were presumed to
function as donors in hydrogen bonding to the adrenergic
receptor. Since adrenergic receptors, like opioid receptors,
interact with G protein-coupled receptors, it seemed reason-
able that a smiliar modification of thebaine-derived opiates
might afford retention of biological activity. Here we describe
the first reported synthesis and biological evaluation of
3-methylsulfonamide-substituted opiates in an effort to obtain
ligands that retain affinity for opioid receptors. Such
modification was of interest because it would permit the
introduction of functionality without eliminating an acidic
proton to that region of the opiate.
Direct conversion of a phenol to an aniline usually employs
high temperatures that are not suitable for the stability of
the opiate structure. The recent reports3 of palladium-
catalyzed amination have opened an efficient route for
conversion of phenols to anilines. Indeed, such an approach
was very recently utilized in the synthesis of 8-amino-
substituted benzomorphan analogues.4 Although the 8-phe-
nolic hydroxy group of benzomorphans is in an equivalent
position to the 3-phenolic hydroxy group of the thebaine-
(3) Wolfe, J. P.; Wagaw, S.; Marcoux, J. F.; Buchwald, S. L. Acc. Chem.
Res. 1998, 31, 805-181.
(4) Wentland, M. P.; Xu, G.; Cioffi, C. L.; Ye, Y.; Duan, W.; Cohen,
D. J.; Colasurdo, A. M.; Bidlack, J. M. Bioorg. Med. Chem. Lett. 2000, 10,
183-187.
(1) Fries, D. S. In Principles of Medicinal Chemistry; Foye, W. O.,
Lemke, T. L., Williams, D. A., Eds.; Lippincott, Williams and Wilkins:
Philadelphia, PA, 1995; pp 247-269.
(2) Larsen, A. A.; Lish, P. M. Nature 1964, 203, 1283-1284.
10.1021/ol005561+ CCC: $19.00 © 2000 American Chemical Society
Published on Web 02/15/2000

Scheme 1a
a Legend: (a) ethylene glycol, p-TSA, 65 °C, 1 mmHg; (b) N-phenyltrifluoromethylsulfonimide, TEA, CH2Cl2, 0 °C; (c) Pd(OAc)2, CO,
DPPF, MeOH, DMSO, 65 °C, CO atm; (d) H2NNH2-H2O, 50 °C; (e) THF/H2O, 2 N HCl, t-BuNO2, 0 °C; (f) (1) benzene, TEA, reflux,
(2) PhCH2OH; (g) Pd/C, H2, MeOH; (h) CH3SO2Cl, TEA, CH2Cl2, 0 °C; (i) HCl(aq), EtOH, reflux, 8 h.
derived opiates, under several conditions, including those
employed in the synthesis of the 8-aminobenzomorphans,
the palladium-catalyzed amination of 3a and 3b was unsuc-
cessful. Since the 8-position of the benzomorphan structure
does not contain an ortho epoxy group, it is possible that
this may be a reason for the failure of this reaction with the
thebaine-derived opiates. Consequently, we embarked on an
alternate route involving carbon-carbon palladium cross-
coupling chemistry in order to generate the 3-carbomethoxy
opiates, which could subsequently be converted to amines
via the Curtius rearrangement.
The desired target compounds (10a, 10b) were synthesized
as outlined in Scheme 1. The 6-keto group of 1a and 1b
was protected by forming the ethylene glycol derived
dioxolanes (2a, 2b). Compounds 2a and 2b were converted
to the corresponding triflates 3a (96%) and 3b (79%) using
equivalent amounts of N-phenyltrifluoromethanesulfonimide
and triethylamine. To a solution of triflate (3a or 3b)
containing DMSO-methanol (1.5:1) was added palladium
acetate (0.1 equiv), diphenylphosphoferrocene (0.21 equiv),
and triethylamine (2.2 equiv). Carbon monoxide gas was
bubbled into this solution for 10 min, and the entire reaction
mixture was sealed under a carbon monoxide atmosphere
and heated to 65 °C for 3 h to afford the carboxymethyl
esters5 (4a, 78%; 4b, 81%). Hydrolysis of 4a and 4b with
lithium hydroxide in a mixture of THF and water afforded
the free acids (not shown). One-pot Curtius rearrangements
were attempted utilizing sodium azide or DPPA in either
tert-butyl alcohol or benzyl alcohol with no success, per-
sumably due to decomposition of the azide prior to reaction
with the ester. Therefore, it was determined that the azide
moiety would need to be synthesized on the opiate scaffold
itself. The esters (4a and 4b) were converted to the
hydrazides (5a, 60%; 5b, 55%) in neat hydrazine monohy-
drate. Treatment of a solution of 5a or 5b in a mixture of
tetrahydrofuran and 2 N HCl with tert-butyl nitrite yielded
the acyl azides 6a (99%) and 6b (92%). The azides were
dissolved in benzene and in the presence of triethylamine
were heated to reflux to form the isocyanate Curtius
rearrangement6 products which, without isolation, were
converted to the benzyloxycarbamates (7a, 38%; 7b, 24%)
by refluxing with benzyl alcohol. Catalytic hydrogenolysis
of the carbamates afforded the corresponding anilines (8a,
98%; 8b, 98%). Subsequent conversion of 8a or 8b to the
methyl sulfonamides (9a, 50%; 9b, 49%) followed by
deprotection of the 6-dioxolane afforded the desired sul-
fonamides 10a (68%) and 10b (70%).
The opiate sulfonamides (10a, 10b) were tested on the
electrically stimulated guinea pig ileum7 and mouse vas
deferens8 preparations at a concentration of 1 µM and were
(5) Hamel, P.; Riendeau, D.; Brideau, C.; Chan, C.-C.; Desmarais, S.;
Delorme, D.; Dube, D.; Ducharme, Y.; Ethier, D.; Grimm, E.; Falgueyret,
J.-P.; Guay, J.; Jones, T. R.; Kwong, E.; McAuliffe, M.; McFarlane, C. S.;
Piechuta, H.; Roumi, M.; Tagari, P.; Young, R. N.; Girard, Y. J. Med. Chem.
1997, 40, 2866-2875.
(6) Smith, P. A. S. Org. React. 1946, 3, 337.
(7) Rang, H. P. J. Pharmacol. Chemother. 1965, 22, 356-365.
(8) Henderson, G.; Hughes, J.; Kosterlitz, H. W. Br. J. Pharmacol. 1972,
46, 746-766.
820
Org. Lett., Vol. 2, No. 6, 2000

found to be devoid of agonist or antagonist activity.
Similarly, they exhibited no significant binding (Ki > 1 µM)
to cloned µ, κ, or δ opioid receptors transiently expressed
in HEK293 cells. The absence of significant binding or
function may be due to the bulky nature of the sulfone
moiety.
In conclusion, this is the first reported synthesis of
3-sulfonamide-substituted derivatives of thebaine-derived
opiates. In addition, this report provides a nontrivial synthetic
route to 3-amino opiates via the Curtius rearrangement. These
compounds were synthesized to determine whether a sul-
fonamide group could replace the phenolic 3-hydroxy group
of an opiate in view of their similar pKa values. Unlike
catecholamines, where activity was retained when one of the
phenolic groups was replaced by a sulfonamide group, similar
modification in the opiates abolishes activity. The reason for
this is unclear, but in contrast to the catecholamines, it
appears likely that the steric bulk of the sulfone moiety makes
the sulfonamide group an unsuitable substitution in opiate
ligands.
Acknowledgment. This work is supported by NIDA/NIH
research grants T32DA07234 and DA01533. We thank
Michael Powers and Mary Lunzer for capable technical
assistance.
Supporting Information Available: Experimental pro-
cedures and spectral data of selected compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL005561+
Org. Lett., Vol. 2, No. 6, 2000
821