Formal synthesis of (-)-aphanorphine using sequential photomediated radical reactions of dithiocarbamates

Article abstract of DOI:10.1002/anie.200701055

(Chemical Equation Presented) Time to change the light bulb: An alkyl dithiocarbamate, itself formed through a photoinitiated group-transfer cyclization of a carbamoyl radical, undergoes a second photomediated radical process initiated with a different light source. These two reactions, which proceed through the same cyclohexenyl radical intermediate, are key steps in a new asymmetric synthesis of the alkaloid aphanorphine. TEMPO = 2,2,6,6-tetramethyl-1-piperidinoxyl radical.

Full text of DOI:10.1002/anie.200701055

Angewandte
Chemie
DOI: 10.1002/anie.200701055
Alkaloid Synthesis
Formal Synthesis of (À)-Aphanorphine Using Sequential
Photomediated Radical Reactions of Dithiocarbamates**
Richard S. Grainger* and Emma J. Welsh
Aphanorphine (1), an alkaloid isolated from the freshwater
blue-green alga Aphanizomenon flos-aquae,[1] has attracted
considerable attention from the synthetic community owing
to its structural similarity to natural and non-natural analge-
sics such as morphine, eptazocine, and pentazocine
(Scheme 1).[2] Approaches to aphanorphine developed to
date have all relied on the formation of the B or C ring to
complete the C-norbenzomorphan skeleton, typically exploit-
ing the rigidity of the bridged tricyclic 3-benzazepine struc-
ture to set the second stereocenter from a preexisting
quaternary benzylic stereocenter at C1,[2a–d,i,k] or from an a-
branched amine at C4.[2n,o,r,t–v,x,y] In this communication we
present a complementary strategy for the synthesis of
aphanorphine which is characterized by the late-stage incor-
poration of the aromatic A ring, and formation of the
pyrrolidine C ring through a novel carbon–carbon bond-
forming reaction.
We have recently reported a new method for the
generation of carbamoyl (aminoacyl) radicals from dithiocar-
bamate precursors, and their subsequent intramolecular
addition—dithiocarbamate group-transfer reactions with
alkenes.[3] Application of this methodology to the synthesis
of the core 6-azabicyclo[3.2.1]octane ring system of aphanor-
phine was envisaged based upon a regioselective 5-exo-trig
cyclization of carbamoyl radical 2 followed by dithiocarba-
mate group transfer to give the functionalized bicyclic lactam
3 (Scheme 1).It was further envisaged that the dithiocarba-
mate group in 3 would provide a handle for phenol
annulation.Critical to the success of such an approach is
the ability of carbamoyl radicals generated from dithiocarba-
mate precursors to undergo potentially difficult cyclizations
onto unactivated alkenes.[3,4] Previous work by Quirante,
Scheme 1. Analgesics structurally related to aphanorphine and a retro-
synthesis of (À)-aphanorphine (1).
Bonjoch, et al.had shown that a-amino radicals undergo
analogous cyclizations onto alkenes carrying electron-with-
drawing groups at C9a (aphanorphine numbering);[5] how-
ever, unactivated alkenes did not undergo cyclization.The
effect of a further alkene substituent at C1, which may also
disfavor 5-exo-trig cyclization, was not evaluated.[6]
An asymmetric synthesis of the requisite secondary
cyclohexenylamine is outlined in Scheme 2 and relies on
Ellmanꢀs sulfinamide auxiliary to set the amino-substituted
stereocenter destined to be C4 of aphanorphine.[7,8] Conden-
sation of enantiomerically pure (R)-tert-butanesulfinamide
(5) with commercially available cis-4-heptenal gave the
expected (E)-sulfinimine 6 (Scheme 2).[9,10] Addition of 2-
methylallylmagnesium chloride gave rise to sulfinamide 7 in
excellent yield as a 83:17 mixture of diastereoisomers.[11] The
configuration of the major stereoisomer 7 was predicted to be
R on the basis of the Ellman model[7] and was ultimately
proven through a formal synthesis of (À)-aphanorphine.
Separation of the two diastereomers could not be achieved at
this stage, and so the mixture was carried through the
following steps.Following N-methylation of 7,[12] 1,7-diene 8
was subjected to ring-closing metathesis using the Grubbs
second-generation catalyst,[13] which furnished the trisubsti-
tuted alkene 9 in excellent yield.[14] Finally removal of the
sulfinyl auxiliary under acidic conditions gave the hydrochlo-
ride salt 10.[7] At this stage, a single recrystallization of 10 gave
enantiomerically pure material.[15]
[*] Dr. R. S. Grainger, E. J. Welsh
School of Chemistry
University of Birmingham
Edgbaston, Birmingham B152TT (UK)
Fax: (+44)121-414-4403
E-mail: r.s.grainger@bham.ac.uk
Homepage: http://www.chem.bham.ac.uk/staff/grainger.shtml
[**] We thank EPSRC, King’s College London, and University of
Birmingham for funding (DTA studentship to E.J.W.). R.S.G. thanks
EPSRC for the award of an Advanced Research Fellowship (2005–
2010). We additionally thank Dr. Andreas E. Goeta (Durham
University)for X-ray crystallography, Dr. Neil Spencer (University of
Birmingham)for NMR experiments, and Graham Burns (University
of Birmingham)for HPLC analysis.
Supporting information for this article (experimental procedures,
analytical data, and copies of 1H and 13C NMR spectra for all new
compounds, and structure of 3 from X-ray coordinates)is available
on the WWW under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2007, 46, 5377 –5380
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5377

Communications
product 3, isolated in 71% yield.The structure of 3 was
ultimately confirmed by X-ray crystallography.[17]
The formation of 3 is consistent with a 5-exo-trig cycliza-
tion of carbamoyl radical 2 to form a bridged azabicyclo-
[3.2.1]octane ring system (path a, Scheme 4), followed by a
Scheme 2. Asymmetric synthesis of secondary amine. Cy=cyclohexyl,
Mes=mesityl.
Scheme 4. Regio- and stereoselectivity in dithiocarbamate group--
transfer radical cyclization.
We have previously developed a two-step synthesis of
stereoselective group-transfer to the secondary radical 12 to
give the axial dithiocarbamate 3.We did not observe
formation of any products 14 containing the azabicyclo-
[2.2.2]octane ring system derived from a 6-endo-trig cycliza-
tion onto the less-substituted end of the double bond (path b).
The presence of a methyl group at the site of attack on the
double bond is clearly not sufficient to unduly influence this
particular radical cyclization.[6]
carbamoyl dithiocarbamates from secondary amines by first
treatment with triphosgene and pyridine to give the carba-
moyl chloride, followed by chloride displacement with sodium
diethyldithiocarbamate at room temperature in acetone.[3]
Although such a procedure could in principle be used in the
case of the free base derived from 10, we were wary of
volatility issues associated with such a low-molecular-weight
amine and instead sought to develop a method that avoided
the isolation of the free amine.Batey et al.have recently
reported carbamoylimidazolium salts as synthetic alternatives
to carbamoyl chorides in displacement reactions with amines,
alcohols, and thiols.[16] Treatment of ammonium salt 10 with
carbonyldiimidazole in the presence of potassium carbonate
gave the carbamoylimidazole 11 in essentially quantitative
yield (Scheme 3).Activation of 11 through N-methylation
gave a carbamoyl imidazolium salt.[16] Using sodium dieth-
yldithiocarbamate as a nucleophile, displacement was best
achieved in refluxing acetone to give the radical-cyclization
With formation of the bridged 6-azabicyclo[3.2.1]octane
ring system in hand, our attention turned to using the
dithiocarbamate group as a means to introduce the phenol
ring of the natural product.We were particularly attracted by
the idea of having a carbonyl group in place of the
dithiocarbamate in 3 (see below).Although a number of
sequences can be envisaged for such a transformation,[18] we
were intrigued by the possibility of initiating a new radical
process from dithiocarbamate 3.Although the trapping of
free radicals with both the 2,2,6,6-tetramethyl-1-piperidinoxyl
radical (TEMPO) and oxygen to form new carbon–oxygen
single bonds is well precedented,[19] use of dithiocarbamates
as precursors to alkyl radicals is relatively rare,[20] and we are
aware of only one example of a radical-chain process using
Bu3SnH having been reported.[21] Keen to avoid the use of
Bu3SnH due to toxicology and potential separation issues,[22]
we instead elected to further our investigations into the use of
light to initiate radical processes from dithiocarbamates.To
our delight, irradiation of a solution of dithiocarbamate 3 with
a medium-pressure mercury arc lamp in a quartz reaction
vessel in the presence of TEMPO resulted in the formal
replacement of a carbon–sulfur bond with a carbon–oxygen
bond and the clean formation of a single adduct 15
(Scheme 5).We believe this result is consistent with radical
12 being regenerated and stereoselectively trapped with
TEMPO[23] and represents a new reaction manifold for
dithiocarbamates.[24]
precursor 4 in 86% yield over two steps.Gratifyingly,
4
underwent clean photoinitiated group-transfer radical cycli-
zation under our standard conditions[3] to give a single
The TEMPO adduct 15 was directly oxidized to ketone 16
Scheme 3. Formation of carbamoyl dithiocarbamate and group-transfer
radical cyclization.
using mCPBA (Scheme 5).[25] Initial attempts to use ketone 16
5378
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 5377 –5380

Angewandte
Chemie
Keywords: cyclization · metathesis · oxidation · radical reactions ·
.
sulfinamides
[1] N.Gulavita, A.Hori, Y.Shimizu, P.Laszlo, J.Clardy,
Tetrahe-
dron Lett. 1988, 29, 4381 – 4384.
[2] Total and formal syntheses: a) S.Takano, K.Inomata, T.Sato, K.
Ogasawara, J. Chem. Soc. Chem. Commun. 1989, 1591 – 1592;
b) S.Takano, K.Inomata, T.Sato, M.Takahashi, K.Ogasawara,
J. Chem. Soc. Chem. Commun. 1990, 290 – 292; c) T.Honda, A.
Yamamoto, Y.Cui, M.Tsubuki, J. Chem. Soc. Perkin Trans. 1
1992, 531 – 532; d) AI..Meyers, W.Schmidt, B.Santiago,
Heterocycles 1995, 40, 525 – 529; e) A.N.Hulme, S.S.Henry,
A.I.Meyers, J. Org. Chem. 1995, 60, 1265 – 1270; f) A.Fadel, P.
Arzel, Tetrahedron: Asymmetry 1995, 6, 893 – 900; g) K.O.
Hallinan, T.Honda, Tetrahedron 1995, 51, 12211 – 12216;
h) M.Node, H.Imazato, R.Kurosaki, Y.Kawano, T.Inoue, K.
Nishide, K.Fuji, Heterocycles 1996, 42, 811 – 819; i) S.Shiotani,
H.Okada, K.Nakamata, T.Yamamoto, F.Sekino, Heterocycles
1996, 43, 1031 – 1047; j) A.Fadel, P.Arzel, Tetrahedron: Asym-
metry 1997, 8, 283 – 291; k) M.Shimizu, T.Kamikubo, K.
Ogasawara, Heterocycles 1997, 46, 21 – 26; l) A.Fadel, P.Arzel,
Tetrahedron: Asymmetry 1997, 8, 371 – 374; m) K.Tanaka, T.
Taniguchi, K.Ogasawara, Tetrahedron Lett. 2001, 42, 1049 –
1052; n) O.Tamura, T.Yanagimachi, T.Kobayashi, H.Ishibashi,
Org. Lett. 2001, 3, 2427 – 2429; o) J.R.Fuchs, R.L.Funk, Org.
Lett. 2001, 3, 3923 – 3925; p) AS. .ElAzab, T.Taniguchi, K.
Ogasawara, Heterocycles 2002, 56, 39 – 43; q) S.K. Taylor, M.
Ivanovic, L.J. Simons, M.M. Davis, Tetrahedron: Asymmetry
2003, 14, 743 – 747; r) O.Tamura, T.Yanagimachi, H.Ishibashi,
Tetrahedron: Asymmetry 2003, 14, 3033 – 3041; s) Y.Kita, J.
Futamura, Y.Ohba, Y.Sawama, J.K.Ganesh, H.Fujioka, J. Org.
Chem. 2003, 68, 5917 – 5924; t) H.Zhai, S.Luo, C.Ye, Y.Ma, J.
Org. Chem. 2003, 68, 8268 – 8271; u) H.Hu, H.Zhai, Synlett
Scheme 5. Photomediated dithiocarbamate–TEMPO exchange and
phenol annulation.
for phenol formation using modified Robinson annulation
conditions met with failure.[26] However, use of an alternative
procedure developed by Boger and Mullican proved more
successful.[27] Hence, reaction of 16 with dimethoxymethylene
malonate under basic conditions followed by acid-catalyzed
dehydration gave the electron-deficient pyrone 17 in 73%
overall yield.Inverse-electron-demand Diels–Alder reaction
with dimethoxyethylene ketal followed by in situ loss of CO2
and MeOH gave the tetrasubstituted aromatic compound 18.
Removal of the ester group was accomplished by first
hydrolysis to the corresponding carboxylic acid, followed by
copper-mediated decarboxylation.[27] The resulting lactam 19
had an optical rotation consistent with that reported in the
literature,[2v] and its preparation constitutes a formal total
synthesis of (À)-aphanorphine (1).Conversion of 19 to the
natural product has previously been accomplished in two
steps—reduction of the amide to the amine using LiAlH4 in
87–93% yield,[2o,v,w] followed by O-demethylation using BBr3
in yields ranging from 61–88%.[2a,b,i,k,o,x]
In conclusion, a formal total synthesis of (À)-aphanor-
phine has been achieved using a carbamoyl radical cyclization
to prepare the 6-azabicyclo[3.2.1]octane ring system of the
natural product, followed by a novel photomediated dithio-
carbamate–TEMPO exchange reaction to introduce oxygen
functionality and facilitate formation of the aromatic ring.
Lactam 19, a known intermediate in aphanorphine synthesis,
was prepared in 13 steps and 14% overall yield starting from
commercially available materials.In the course of this work a
new method for the synthesis of carbamoyl dithiocarbamates
has also been developed.
2003, 2129 – 2130; v) J.F.Bower, P.Szeto, T.Gallagher,
Chem.
Commun. 2005, 5793 – 5795; w) M.Katoh, H.Inoue, A.Suzuki,
T.Honda, Synlett 2005, 2820 – 2822; x) M.Li, P.Zhou, H.F.
Roth, Synthesis 2007, 55 – 60; y) Z.Ma, H.Zhai, Synlett 2007,
161 – 163; z) J.F.Bower, P.Szeto, T.Gallagher,
Org. Biomol.
Chem. 2007, 5, 143 – 150.
[3] R.S. Grainger, P. Innocenti, Angew. Chem. 2004, 116, 3527 –
3530; Angew. Chem. Int. Ed. 2004, 43, 3445 – 3448.
[4] This principle has been extensively developed by Zard for the
generation and reaction of a number of radicals using xanthate
precursors.For a recent review see: B.Quiclet-Sire, S.Z.Zard,
Chem. Eur. J. 2006, 12, 6002 – 6016.
[5] J.Quirante, X.Vila, C.Escolano, J.Bonjoch, J. Org. Chem. 2002,
67, 2323 – 2328.
[6] For competitive formation of [3.2.1] vs. [2.2.2] bicyclic ring
systems through radical cyclization see: a) W.F.Berkowitz, P.J.
Wilson, J. Org. Chem. 1991, 56, 3097 – 3102; b) A.Srikrishna, P.
Hemamalini, Tetrahedron 1992, 48, 9337 – 9354; c) M.Toyota,
M.Yokota, M.Ihara, Tetrahedron Lett. 1999, 40, 1551 – 1554;
d) M.Toyota, M.Yokota, M.Ihara, J. Am. Chem. Soc. 2001, 123,
1856 – 1861; e) E.W. Della, P.A. Smith, J. Chem. Soc. Perkin
Trans. 1 2001, 445 – 451.
[7] D.A.Cogan, G.Liu, J.Ellman,
8904.
Tetrahedron 1999, 55, 8883 –
[8] For a recent review on the use of sulfinimines in asymmetric
synthesis see: D.Morton, R.A.Stockman, Tetrahedron2006, 62,
8869 – 8905.
[9] G. Liu, D.A. Cogan, T.D. Owens, T.P. Tang, J.A. Ellman,
J.
Org. Chem. 1999, 64, 1278 – 1284.
[10] (R)-tert-Butanesulfinamide (5) is commercially available but can
be readily prepared on large scale: D.J.Weix, J.A.Ellman, Org.
Synth. 2005, 82, 157 – 165.
Received: March 9, 2007
Published online: June 6, 2007
Angew. Chem. Int. Ed. 2007, 46, 5377 –5380
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5379

Communications
[11] The d.r. value was determined by integration of the 1H NMR
[20] Dithiocarbamates have been used as mediators in free-radical
polymerization by the RAFT method.For a review see: a) G.
Moad, E.Rizzardo, S.H.Thang, Aust. J. Chem. 2005, 58, 379 –
410; For trapping of alkyl radicals with dithiocarbamates see:
spectrum of the crude reaction mixture and confirmed by HPLC
analysis.
[12] S.Colonna, G.Germinario, A.Manfredi, C.J.M.Stirling,
J.
Chem. Soc. Perkin Trans. 1 1988, 1695 – 1698.
[13] M.Scholl, S.Ding, C.W.Lee, R.H.Grubbs,
953 – 956.
b) G.Bouhadir, N.Legrand, B.Quiclet-Sire, S.Z.Zard,
hedron Lett. 1999, 40, 277 – 280; c) S.H. Thang, Y.K. Chong,
R.T.A.Mayadunne, G.Moad, E. Rizzardo, Tetrahedron Lett.
Tetra-
Org. Lett. 1999, 1,
1999, 40, 2435 – 2438; d) D.Harakat, J.Pesch, S.Marinkovic, N.
Hoffmann, Org. Biomol. Chem. 2006, 4, 1202 – 1205.
[14] For a further example of ring-closing metathesis in the presence
of the sulfinamide group see: F.A.Davis, Y.Wu, Org. Lett. 2004,
6, 1269 – 1272.For a recent review on metathesis reactions in
total synthesis, see: KC. .Nicolaou, PG. .Bulger, D.Sarlah,
Angew. Chem. 2005, 117, 4564 – 4601; Angew. Chem. Int. Ed.
2005, 44, 4490 – 4527.
[15] The e.r. value was determined by chiral HPLC analysis of
corresponding carbamoylimidazole 11.
[16] J.A.Grzyb, M.Shen, C.Yoshina-Ishii, W.Chi, R.S.Brown,
R.A. Batey, Tetrahedron 2005, 61, 7153 – 7175.For earlier
[21] M.Yanagawa, O.Moriya, Y.Watanabe, Y.Ueno, T.Endo,
Chem. Soc. Jpn. 1988, 61, 2203 – 2204.
Bull.
[22] a) P.A.Baguley, J.C.Walton, Angew. Chem. 1998, 110, 3272 –
3283; Angew. Chem. Int. Ed. 1998, 37, 3072 – 3082; b) A.Studer,
S.Amrein, Synthesis 2002, 835 – 849.
[23] The relative stereochemistry in 15 has been assigned as shown on
the basis of nOe spectroscopy and the expectation that the
trapping of cyclohexenyl radical 12 by TEMPO will occur with
the same facial selectivity as that observed in the dithiocarba-
mate group-transfer reaction (Scheme 4).
precedent see: D.G. Oakenfull, K. Salvesen, W.P. Jenks,
J.
Am. Chem. Soc. 1971, 93, 188 – 194.
[17] CCDC-639680 contains the supplementary crystallographic data
for this paper.These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via www.ccdc.ca-
m.ac.uk/data_request/cif. The structure of 3 from X-ray coor-
dinates is also shown in the Supporting Information.
[18] This formal transformation has been achieved on allylic
dithiocarbamates by deprotonation with lithium diisopropyla-
mide, trapping with MeSSMe, and hydrolysis of the resulting
mixed thioacetal to an a,b-unsaturated aldehyde: T.Nakai, H.
Shiono, M.Okawara, Tetrahedron Lett. 1974, 15, 3625 – 3628.
Attempts to deprotonate adjacent to the dithiocarbamate group
in 3 followed by trapping with electrophiles have so far proved
unsuccessful.
[24] If the cyclization of 4 to 3 is run in the presence of TEMPO,
dithiocarbamate 3 is not obtained.Instead the initial carbamoyl
radical 2 is trapped with the TEMPO prior to cyclization.
[25] T.Inokuchi, H.Kawafuchi, Tetrahedron2004, 60, 11969 – 11975.
[26] a) D.L.Boger, M.D.Mullican, J. Org. Chem. 1980, 45, 5002 –
[19] C.Ollivier, P.Renaud in Radicals in Organic Synthesis, Vol. 2
5004.b) O.Barun, S.Nandi, K.Panda, H.Ila, H.Junjappa,
J.
(Eds: P.Renaud, MP..Sibi), Wiley-VCH, Weinheim,
pp.93 – 112.
2001,
Org. Chem. 2002, 67, 5398 – 5401.
[27] D.L.Boger, M.D.Mullican, J. Org. Chem. 1984, 49, 4033 – 4044.
5380
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 5377 –5380