Asymmetric synthesis of (+)-morphine. The phenanthrene route revisited

Full text of DOI:10.1021/jo9710994

5250
J . Org. Chem. 1997, 62, 5250-5251
Sch em e 1a
Asym m etr ic Syn th esis of (+)-Mor p h in e.
Th e P h en a n th r en e Rou te Revisited
J ames D. White,* Peter Hrnciar, and
Frank Stappenbeck
Department of Chemistry, Oregon State University,
Corvallis, Oregon 97331-4003
Received J une 19, 1997
Morphine (1) continues to occupy a position of unique
clinical importance in medicine, its wide prescription for
the treatment of severe pain and for relief of suffering
in the terminally ill placing it foremost among analgesic
agents.1 Synthesis of morphine engaged the interest of
organic chemists even before its complete structure was
revealed,2 and many optimistic schemes were devised in
the hope that a route could be found from a hydro-
phenanthrene precursor.3 Subsequent efforts brought
numerous syntheses of (()-morphine,4 but asymmetric
approaches have encountered difficulties that have made
this aspect of morphine synthesis a significant challenge.5
In fact, it was not until Overman’s publication in 1993
that a synthesis of natural (-)-morphine was achieved
by a route which did not involve resolution.6 We now
report an asymmetric synthesis of morphine having its
origins in those early studies7 that viewed the phenan-
threne nucleus as an appropriate platform upon which
to construct the morphine framework. Because our
interest is primarily in pharmacological properties of the
unnatural enantiomorph, particularly its binding to
a
(i) H2, [Rh(COD)Cl]2, (4R,5R)-(-)-MOD-DIOP, 100%, (94% ee);
(ii) Br2, HOAc, 93%; (iii) MsOH, P2O5, 75%; (iv) H2, Pd/C, NaHCO3,
100%; (v) LiOH, THF-H2O, 100%; (vi) KH, HCO2Me, DME, 0 °C,
85%; (vii) MVK, Et3N, CH2Cl2, 95%; (viii) NaOH, H2O, THF, 95%;
(ix) CH2N2, Et2O-CH2Cl2, 99%; (x) Br2, NaHCO3, CH2Cl2, 80%;
(xi) DBU, C6H6, 50 °C, 90%.
opioid receptors, the focus of our synthetic work has been
(+)-morphine (1).8 Our approach to 1 departs from all
previous schemes by invoking as the key step a carbenoid
C-H insertion to establish the C13-C15 bond.9 This
reaction is used to fashion a pentacyclic skeleton from
which the piperidine ring of 1 evolves at a final stage.
Asymmetric hydrogenation10 of the Stobbe condensa-
tion product 211 of isovanillin and dimethyl succinate
using a chiral rhodium catalyst prepared from the
bisphosphine 3 [(4R,5R)-MOD-DIOP]12 gave the succinate
derivative 4 in quantitative yield and 94% enantiomeric
excess (Scheme 1).13 The resulting (2S) configuration of
4 controls all subsequent stereochemical events through-
out the synthesis. Intramolecular Friedel-Crafts reac-
tion of 4 yielded exclusively the undesired tetralone from
cyclization para to the free phenol, but this was easily
corrected by prior bromination. This aryl blocking group
conveniently steers Friedel-Crafts acylation toward te-
tralone 5.14 Condensation of the carboxylic acid 5 with
methyl formate,15 followed by treatment of the resultant
R-formyl tetralone with methyl vinyl ketone (MVK),
(1) For a concise history of morphine and its development as an
analgesic, see Lednicer, D. Central Analgesics; Wiley: New York, 1982;
pp 137-213.
(2) (a) Gulland, J . M.; Robinson, R. Mem. Proc. Manch. Lit. Soc.
1925, 69, 79. (b) Robinson, R.; Sugasawa, S. J . Chem. Soc. 1933, 1079.
(3) These early endeavors were heavily influenced by degradative
experiments with morphine which yielded a phenanthrene system after
exhaustive dehydrogenation. For
a recent successful approach to
morphine along these lines, see Mulzer, J .; Durner, G.; Trauner, D.
Angew. Chem., Int. Ed. Engl. 1996, 35, 2830.
(4) For a recent review, see Hudlicky, T.; Butora, G.; Fearnley, S.
P.; Gum, A. G.; Stabile, M. R. Studies in Natural Products Chemistry;
Rahman, A-u., Ed.; Elsevier: Amsterdam, 1996; pp 43-154.
(5) For example, see Schwartz, M. A.; Pham, P. T. K. J . Org. Chem.
1988, 53, 2318.
(6) Hong, C. Y.; Kado, N.; Overman, L. E. J . Am. Chem. Soc. 1993,
115, 11028.
(7) (a) Pschorr, A. T. Ber. 1896, 29, 296. (b) Fieser, L. F.; Holmes,
H. L. J . Am. Chem. Soc. 1936, 58, 2319. (c) Robinson, R.; Gosh, R. J .
J . Chem. Soc. 1944, 506.
(8) For a recent approach to the unnatural enantiomer of morphine,
see Butora, G.; Hudlicky, T.; Fearnley, S. P.; Gunn, A. G.; Stabile, M.
R.; Abboud, K. Tetrahedron Lett. 1996, 37, 8155.
(9) Of the 16 completed morphine syntheses only one, that of
Ginsburg, attacks the morphine problem from this direction (a)
Ginsburg, D.; Pappo, R. J . Chem. Soc. 1953, 1524. (b) Ginsburg, D.;
Elad, D. J . Chem. Soc. 1954, 3052.
(11) J ohnson, W. S.; Daub, G. Org. React. 1951, 6, 1.
(12) Morimoto, T.; Chiba, M.; Achiwa, K. Tetrahedron Lett. 1989,
30, 735.
(13) Enantiomeric excess of 4 was determined by chiral HPLC using
a Chiralpak AD column and 9:1 hexane-2-propanol (containing 5%
trifluoroacetic acid) as eluent.
(14) For a previous application of this tactic, see White, J . D.;
Caravatti, G.; Kline, T. B.; Edstrom, E.; Rice, K. C.; Brossi, A.
Tetrahedron 1983, 39, 2393.
(15) Banwell, M. G.; Lambert, J . N.; Corbett, M; Greenwood, R. J .;
Gulbis, J . M.; Mackay, M. F. J . Chem. Soc., Perkin Trans. 1 1992, 1415.
Saponification of the methyl ester of 5 is necessary in order to avoid
racemization in this step.
(10) Harada, K. In Asymmetric Synthesis; Morrison, J . D., Ed.;
Academic Press: Orlando, 1985; Vol. 5, p 345.
S0022-3263(97)01099-2 CCC: $14.00 © 1997 American Chemical Society

Communications
J . Org. Chem., Vol. 62, No. 16, 1997 5251
Sch em e 2a
Sch em e 3a
a
(i) NaH, MeI, C6H6, ∆, 95%; (ii) HBr, MeCN, 96%; (iii) Dess-
Martin periodinane, CHCl3, 99%; (iv) PhSeCl, MsOH, CH2Cl2; (v)
NaIO4, 65% from 16; (vi) LiAlH4, THF, ∆, 90%; (vii) ref 17.
benoid C-H insertion17 which creates the pentacyclic
nucleus of 12. In the event, decomposition of 11 afforded
principally 12 accompanied by two minor products, one
of which arose from intramolecular hydrogen transfer
and the other from fragmentation of the diazo ketone.18
Attempts to effect expansion of the bridged cyclopen-
tanone of 12 through Beckmann rearrangement to a
δ-lactam were initially unsuccessful. Although oxime 13
was produced (as a 1.2:1 mixture of syn and anti isomers)
in good yield, it failed to yield any trace of a δ-lactam
under acidic conditions.19 Fortunately, the brosylate 14
derived from 13 was more compliant, affording 15 and
the regioisomeric δ-lactam in the ratio 11:1, respectively,
upon exposure to acetic acid. N-Methylation of 15,
followed by removal of the MOM protecting group and
oxidation of the resultant alcohol, gave keto lactam 16
(Scheme 3). Introduction of the conjugated olefin into
16 was effected by R-phenylselenylation of the ketone
under acidic conditions20 and subsequent oxidation of the
selenide with periodate. As expected, treatment of 17
with lithium aluminum hydride reduced the enone from
the R face21 and also reduced the lactam to furnish (+)-
codeine (18). The latter was identical with the natural
substance in all respects, except for its optical rotation.
a
(i) NaBH4, i-PrOH-CH2Cl2, 99%; (ii) H2, Pd/C, MeOH, 78%;
(iii) CH2(OMe)2, P2O5, CHCl3 80%; (iv) LiOH, THF-H2O, 99%; (v)
(COCl)2, C6H6; (vi) CH2N2, 63%; (vii) Rh2(OAc)4, CH2Cl2, 50%; (viii)
H2NOH‚HCl, NaOAc, MeOH, 90%; (ix) p-BrC6H4SO2Cl, Et3N; (x)
AcOH, rt, 62% from 13.
produced the tricyclic lactol 6. Exposure of 6 to base
furnished the product 7 of Robinson annulation, whose
relative configuration was established by X-ray crystal-
lographic analysis of the corresponding ethyl ester.
Careful methylation of the acid and subsequent bromi-
nation of the enone gave the dibromophenanthrenone
derivative 8, and exposure of this R-bromo enone to 1,8-
diazabicyclo[5.4.0]undec-7-ene (DBU) afforded the tetra-
cycle 9.16
Hydrogenation of the benzofuran nucleus of 9 unex-
pectedly reduced the ketone and left no oxygen substitu-
ent at C6. Hence, 9 was first treated with sodium
borohydride to yield exclusively the 6R alcohol in which
the C-O bond is aligned nearly coplanar with the furan
(Scheme 2). This orientation not only ensures proper
placement of the crucial 13R hydrogen during the cata-
lytic hydrogenation which follows, but also denies the
reduced product 10 opportunity for hydrogenolysis.
Straightforward transformation of 10 to diazo ketone 11
set the stage for the pivotal, rhodium(II)-catalyzed car-
Final
de-
methylation of 18 to (+)-morphine (1) followed the
protocol previously published by Rice.22
In summary, a new asymmetric route to (+)-morphine
has been demonstrated that takes isovanillin in 28 steps
and ca. 3% overall yield to the unnatural enantiomer.
The synthesis, which can be easily adapted to natural
(-)-morphine by hydrogenation of 2 using catalyst pre-
pared from the bisphosphine enantiomeric with 3, affords
good opportunities for the preparation of analogues en
route.
(16) Cyclization probably occurs via the isomeric â,γ-unsaturated
ketone.
(17) Taber, D. F. In Comprehensive Organic Synthesis; Trost, B. M.,
Fleming, I., Eds.; Pergamon, Oxford, U.K., 1991; Vol. 3, p 1045.
(18) The two byproducts (ca. 15% each) have been identified as i
and ii, respectively:
Ack n ow led gm en t. We are indebted to Professor
J ohn H. Block, College of Pharmacy, Oregon State
University, for a sample of (-)-codeine and to Professor
P. Shing Ho, Department of Biochemistry and Biophys-
ics, Oregon State University, for assistance with the
X-ray structure determination of a derivative of 7.
Financial support was provided by the National Science
Foundation.
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures and characterization data for 2, 4-13, and 15-18 (13
pages).
(19) Gawley, R. E. Org. React. 1988, 35, 1. The principal reaction
pathway is believed to involve Beckmann fragmentation.
(20) Iijima, I.; Rice, K. C. Heterocycles 1977, 6, 1157.
(21) Weller, D. D.; Rapoport, H. J . Med. Chem. 1976, 19, 1171.
(22) Rice, K. C. J . Med. Chem. 1977, 20, 164.
J O9710994