Enantioselective total synthesis of (+)-homochelidonine by a Pd II-catalyzed asymmetric ring-opening reaction of a meso-azabicyclic alkene with an aryl boronic acid

Article abstract of DOI:10.1002/anie.200603945

(Chemical Equation Presented) An efficient and highly convergent enantioselective synthesis of (+)-homochelidonine has been achieved (see scheme; Cbz = benzyloxycarbonyl, MOM = methoxymethyl) and relied on a new and powerful desymmetrizing ring-opening reaction of a meso-azabicycle with an aryl boronic acid. The route should allow access to other hexahydrobenzo[c]phenanthridine alkaloids.

Full text of DOI:10.1002/anie.200603945

Angewandte
Chemie
DOI: 10.1002/anie.200603945
Natural Products Synthesis
Enantioselective Total Synthesis of (+)-Homochelidonine by a PdII-
Catalyzed Asymmetric Ring-Opening Reaction of a meso-Azabicyclic
Alkene with an Aryl Boronic Acid**
Helen A. McManus, Matthew J. Fleming, and Mark Lautens*
(+)-homochelidonine[1] (1) is a representative member of the
hexahydrobenzo[c]phenanthridine alkaloids with cis-fused
B/Crings, which occur in a number of plant species of the
Papaveraceae family. Other structurally similar naturally
The alkaloid is also cytotoxic, has been shown to inhibit
tubulin polymerization,[5] and is a major component of the
drug Ukrain, a semisynthetic antitumor preparation derived
from C. majus alkaloids.[6]
occurring
hexahydrobenzo[c]phenanthridine
alkaloids
The syntheses of racemic homochelidonine[7] and the
closely related chelidonine[8] have been reported in the
literature, but to the best of our knowledge, no enantiose-
lective total syntheses of these compounds exist to date.
Herein we disclose the first enantioselective synthesis of
(+)-homochelidonine which could be readily adapted to
prepare a variety of hexahydrobenzo[c]phenanthridine alka-
loids.
include (+)-chelidonine[2] (2), (À)-norchelidonine (3),
(+)-chelamidine (4), and (+)-chelamine (5, Scheme 1). Both
1 and 2 have been isolated from the roots of Chelidonium
majus L., and chelidonine (2) was isolated as early as 1839.[3]
Chelidonine has a wide range of pharmacological activities
and has been used successfully in experimental oncology.[4]
(+)-Homochelidonine has partially hydrogenated B and
Crings, fully aromatic A and D rings, a hydroxy group at the
C11 position, and three contiguous syn stereogenic centers.
Our proposed retrosynthetic plan for an enantioselective total
synthesis of (+)-homochelidonine is shown in Scheme 2. This
route exploits the racemic ring-opening chemistry of azabi-
cyclic alkenes with aryl boronic acids developed within our
research group, in which two adjacent syn stereocenters are
installed to form 1,2-dihydronaphthalenes.[9] It would there-
fore be necessary to develop an enantioselective metal-
catalyzed variant of this reaction, in which the key step would
be the addition of the trisubstituted aryl boronic acid 9 to
azabenzonorbornadiene 10 to yield cis-1-amino-2-aryl-dihy-
dronaphthalene intermediate 8. A suitably functionalized
ortho substituent on the boronic acid would allow for
cyclization onto the B ring. It was initially considered that
the syn-hydroxy group at the C11 position could be intro-
duced by functionalizing the double bond of dihydronaph-
thalene 7 to give the corresponding syn-epoxide 6, followed
by selective opening of the ring with hydride.
Scheme 1. Structures of some hexahydrobenzo[c]phenanthridine
alkaloids.
[*] Dr. H. A. McManus, Dr. M. J. Fleming, Prof. M. Lautens
Department of Chemistry
Davenport Chemical Laboratories
University of Toronto
80 St. George St., Toronto, ON M5S3H6 (Canada)
Fax: (+1)416-946-8185
E-mail: mlautens@chem.utoronto.ca
[**] We thank the NSERC, Merck Frosst Canada, and the University of
Toronto for support of our programs, JØrØmy Ruiz for initial studies
using the N-Cbz azabicycle, Alena Rudolph for initial studies into
the enantioselective reaction, and Dr. Christopher Dockendorff for
first bringing this natural product to our attention.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Scheme 2. Retrosynthetic analysis of (+)-1. Boc=tert-butylcarbonyl,
PG=protecting group.
Angew. Chem. Int. Ed. 2007, 46, 433 –436
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
433

Communications
The next task was to evaluate the palladium(II)-catalyzed
ring-opening reaction. Using our previously developed con-
ditions for the racemic system ([Pd(dppp)Cl2], Cs2CO3,
MeOH, 608C, 24 h),[9a] the reaction gave 18 as a single
diastereoisomer in 82% yield (Table 1, entry 1). The syn
relationship between the aryl and NHBoc groups was
confirmed by a 2D 1H NMR ROESY experiment. An
enantioselective version of this reaction was next investi-
gated. These reactions were carried out by first generating the
chiral palladium(II) catalyst in situ by stirring the catalyst
precursor bis(acetonitrile)dichloropalladium(II) with the
appropriate chiral ligand in MeOH at room temperature
prior to the addition of the reactants. A series of chiral ligands
were examined (Table 1, entries 2–7). These reactions were
carried out using 5 mol% catalyst, 5.5 mol% chiral ligand,
and at room temperature to maximize enantiodiscrimination.
The optimal ligand for the enantioselective ring opening of 10
with 17 was found to be (S)-tol-binap, which provided
dihydronaphthalene 18 in 90% yield and with 91% ee
(Table 1, entry 7).[12]
Scheme 3. Synthesis of N-Boc-azabicycle 10: a) BBr3, CH2Cl2, RT, 2 h,
quant.; b) CH2BrCl, Cs2CO3, DMF, 1108C, 3 h, 75%; c) N-Boc-pyrrole,
nBuLi, PhMe, À788C to RT, 17 h, 71%.
The first task was to prepare azabicycle 10 and suitably
functionalized aryl boronic acid 9. Azabicycle 10 was
prepared in three steps from 4,5-dibromovertrole (11,
Scheme 3). Treatment of 11 with boron tribromide gave 12
in quantitative yield, which was subsequently dialkylated with
bromochloromethane to give 13 in 75% yield. Slow addition
of nBuLi to dibromide 13 generated a benzyne intermediate,
which underwent an in situ Diels–Alder reaction with N-Boc-
pyrrole to furnish N-Boc-azabicycle 10 in 71% yield.
With azabicycle 10 in hand, we next set about preparing a
suitable boronic acid 9. A search of the literature revealed
that the synthesis of a 3,4-dimethoxyphenylboronic acid 17
with an ortho MOM-protected hydroxymethyl moiety had
been previously prepared by Nichols and co-workers.[10] The
use of this boronic acid in the ring-opening reaction with
azabicycle 10 would facilitate the formation of the B ring.
Boronic acid 17 was prepared in three steps following a
modified version of the reported procedure (Scheme 4).
2,3-Dimethoxybenzyl alcohol (14) was first regioselectively
brominated with NBS to give aryl bromide 15 in 89% yield.
The hydroxy group was then protected as the MOM ether in
71% yield by stirring 15 in dimethoxymethane at RT in the
presence of a catalytic amount of LiBr and p-toluenesulfonic
acid.[11] Aryl bromide 16 was converted into the correspond-
ing boronic acid 17 in 63% yield by lithium–halogen exchange
with nBuLi and subsequent quenching with triisopropyl
borate.
Table 1: Evaluation of ligands in the asymmetric ring-opening reaction.
Entry
Ligand[a]
Yield [%][b]
ee [%][c]
1[d]
2
3
4
5
dppp
(S)-binap
82
48
66
25
61
92
90
–
87
40
14
77
37
91
(S,S)-chiraphos
(R)-monophos
(R)-segphos
(R,R)-Me-duphos
(S)-tol-binap
6
7
[a] dppp=propane-l,3-diylbis(diphenylphosphane), binap=2,2’-bis(di-
phenylphosphanyl)-1,1’-binaphthyl, chiraphos=2,3-bis(diphenylphos-
phino)butane, monophos=(3,5-dioxa-4-phosphacyclohepta[2,1-a;3,4-
a’]dinaphthalen-4-yl)dimethylamine,
segphos=5,5’-bisdiphenylphos-
phino[4,4’]bi[benzo[1,3]dioxolyl], duphos=2,5-diethylphospholano]ben-
zene, tol=tolyl. [b] Yields of isolated products. [c] ee values determined
by HPLC analysis (Chiralpak AD, hexane/2-propanol 90:10, flow rate
1.0 mLminÀ1): tr: 15.24 and 27.47 min, 208C. [d] Reaction carried out at
608C using 1 mol% catalyst.
Once we had established high enantioselectivity and
reactivity for the palladium-catalyzed ring-opening reaction
between 10 and 17, we turned our attention to the cyclization
of the dihydronaphthalene product 18 and the formation of
the B ring of homochelidonine. Selective removal of the
MOM group proved to be problematic. A variety of
conditions were attempted: TMSCl,[13] LiBF4,[14] B-chloro-
catecholborane,[15] HCl/THF,[16] CBr4/iPrOH,[17] PPTS/
Scheme 4. Synthesis of boronic acid 17: a) NBS, THF, RT, 30 min,
89%; b) CH2(OMe)2, LiBr, TsOH, RT, 15 h, 72%; c) nBuLi, THF,
À788C 45 min; then B(OiPr)3, À788C to RT, 18 h; then aqueous
NH4Cl, 63%. NBS=N-bromosuccinimide, Ts=toluene-4-sulfonyl,
MOM=methoxymethyl.
434
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 433 –436

Angewandte
Chemie
tBuOH,[18] 50% aqueous AcOH,[13] and p-TsOH/MeOH[19]
.
However, we only isolated the expected product 19 in low
yield (< 25%), contaminated with the aromatic compounds
20 and 21, presumably formed by concomitant removal of the
Boc group and then elimination of ammonia under the acidic
conditions (Scheme 5).
Scheme 6. Synthesis of dihydronaphthalene 23: a) TMSI, NEt3, CH2Cl2,
reflux, 15 min; then CbzCl, RT, 3 h, 80%; b) [Pd(MeCN2)Cl2] (5 mol%),
(S)-binap (5.5 mol%), 17, Cs2CO3, MeOH, RT, 6 h, 89%, 90% ee
(80%, 99% ee after one recrystallization). Cbz=benzyloxycarbonyl,
TMS=trimethylsilyl.
installed on the correct side of the Cring, it was now
necessary to carry out a regioselective hydride-mediated
reduction of the epoxide, followed by removal of the Cbz
group and methylation of the amine. This was carried out in a
one-pot reaction in which epoxide 27 was heated with LiAlH4
in 1,4-dioxane to give (+)-homochelidonine in 87% yield.
The spectroscopic properties of the synthetic material
were in agreement with those of the natural product.[7c,21] The
optical rotation ([a]2D5 = + 120 (c = 1.0 in EtOH)) confirmed
the absolute stereochemistry.[1g] HPLCanalysis of this com-
pound on a chiral stationary phase gave an ee value of 99%,
thus indicating that the enantiopurity of dihydronaphthalene
23 was maintained throughout the final sequence.
In summary, we have developed a new and general
strategy for the synthesis of the hexahydrobenzo[c]phenan-
thridine alkaloids with a novel and highly enantioselective
palladium(II)-catalyzed ring-opening reaction of a meso-
azabicycle with an aryl boronic acid as the key step. We
have demonstrated the power of this methodology for the first
Scheme 5. Attempted cleavage of the MOM group of 18.
An alternative approach used N-Cbz-protected azabicycle
22, since we surmised that the resulting product, dihydro-
naphthalene 23, would be stable to the acidic conditions
required to remove the MOM group. N-Boc-azabicycle 10
was converted into N-Cbz-azabicycle 22 in 80% yield in a
one-pot reaction by using TMSI for removal of the Boc
group,[20] then addition of CbzCl to protect the resulting
secondary amine. The asymmetric ring-opening reaction with
boronic acid 17 gave dihydronaphthalene 23 in 89% yield and
90% ee (Scheme 6). One recrystallization from Et2O gave
dihydronaphthalene 23 in 80% yield and 99% ee. This
reaction has been carried out on a multigram scale without
any loss of enantiodiscrimination.[21]
It was now possible to selectively remove the MOM group
by stirring dihydronaphthalene 23 in concentrated HCl and
THF/iPrOH to give the benzyl alcohol 24 in 75% yield
(Scheme 7). Bromination of the benzyl
alcohol group with carbon tetrabromide
and triphenylphosphine, followed by a
cyclization reaction with sodium hydride
on the crude reaction mixture, resulted in
the formation of the B ring, thus providing
dihydronaphthalene 25 in 90% yield. Our
attention now turned to the introduction of
the syn-hydroxy group at C11. A single
bromohydrin isomer was obtained in 75%
by reaction of 25 with NBS in wet THF. The
regio- and stereochemistry can be rational-
ized by the intermediate bromonium ion
being formed on the least hindered face of
the alkene, followed by attack of water at
the benzylic position. Reaction of bromo-
hydrin 26 with sodium tert-butoxide in THF
yielded the syn-epoxide 27 in quantitative
yield. With the oxygen functionality now
Scheme 7. Completion of the synthesis of (+)-1: a) HCl, iPrOH/THF, RT, 8 h, 75%; b) CBr4, PPh3,
CH2Cl2, 08C, 1 h; then NaH, DMF, 08C, 3 h, 90%; c) NBS, THF/H2O, RT, 90 min, 75%; d) KOtBu,
THF, À788C, 30 min, quant.; e) LiAlH4, 1,4-dioxane, reflux, 12 h, 87%.
Angew. Chem. Int. Ed. 2007, 46, 433 –436
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
435

Communications
[7] a) I. Ninomiya, O. Yamamoto, T. Naito, Heterocycles 1977, 7,
137; b) I. Ninomiya, O. Yamamoto, T. Naito, J. Chem. Soc.
Perkin Trans. 1 1983, 2165; c) I. Ninomiya, O. Yamamoto, T.
Naito, J. Chem. Soc. Perkin Trans. 1 1983, 2171; d) M. Hanaoka,
S. Yoshida, C. Mukai, Tetrahedron Lett. 1985, 26, 5163; e) M.
Yoshida, T. Watanabe, T. Ishikawa, Tetrahedron Lett. 2002, 43,
6751.
[8] a) W. Oppolzer; K. Keller, J. Am. Chem. Soc. 1971, 93, 3836;
b) W. Oppolzer, Heterocycles 1980, 14, 1615; c) M. Cushman, T.-
C. Choong, J. T. Valko, M. P. Kolek, Tetrahedron Lett. 1980, 21,
3845; d) M. Cushman, T.-C. Choong, J. T. Valko, M. P. Kolek, J.
Org. Chem. 1980, 45, 5067.
time in natural product synthesis and thus have completed the
first enantioselective total synthesis of (+)-homochelidonine
in 14 steps. The longest linear sequence is 11 steps and the
overall yield is 15% from 4,5-dibromovertrole. Studies to find
general conditions for the desymmetrization of meso-azabi-
cycles with boronic acids and the application of this method-
ology to the synthesis of other hexahydrobenzo[c]phenan-
thridine alkaloids is currently under investigation.
Received: September 26, 2006
Published online: December 5, 2006
[9] a) M. Lautens, C. Dockendorff, Org. Lett. 2003, 5, 3695; for a
review of ring-opening reactions prior to 2003, see: b) M.
Lautens, K. Fagnou, S. Hiebert, Acc. Chem. Res. 2003, 36, 48; for
ring-opening reactions after 2003, see: c) M. Lautens, S. Hiebert,
J. Am. Chem. Soc. 2004, 126, 1437; d) P. Leong, M. Lautens, J.
Org. Chem. 2004, 69, 2194; e) Y.-h. Cho, V. Zunic, H. Senboku,
M. Olsen, M. Lautens, J. Am. Chem. Soc. 2006, 128, 6837; f) Y.-h.
Cho, A. Fayol, M. Lautens, Tetrahedron: Asymmetry 2006, 17,
416.
Keywords: alkaloids · asymmetric catalysis · polycycles ·
ring-opening reactions · total synthesis
.
[1] a) E. Schmidt, F. Selle, Arch. Pharm. 1890, 228, 441; b) J. Slavik,
Cesk. Farm. 1955, 4, 15; c) J. Slavik, L. Slavíkovµ, J. Brabenec,
Collect. Czech. Chem. Commun. 1965, 30, 3697; d) R. H. F.
Manske, Can. J. Res. Sect. B 1943, 21, 140; e) N. Takao, M.
Kamiguchi, K. Iwasa, N. Morita, Arch. Pharm. 1984, 317, 223;
f) J. Slavik, E. Taborska, H. Bochorakova, Collect. Czech. Chem.
Commun. 1994, 89, 429; g) M. Ne›as, J. Dostµl, I. Kejnovskµ, M.
Vorlí›kovµ, J. Slavík, J. Mol. Struct. 2005, 734, 1.
[2] a) F. von Bruchhausen, H. W. Bersch, Chem. Ber. 1930, 63, 2520;
b) E. Späth, F. Kuffner, Chem. Ber. 1931, 64, 370; c) H. W.
Bersch, Arch. Pharm. 1958, 291, 491; d) E. Seoane, An. R. Soc.
Esp. Fis. Quim. Ser. B 1965, 61, 755; e) C.-Y. Chen, D. B.
MacLean, Can. J. Chem. 1967, 45, 3001; f) S. Naruto, S. Arakawa,
H. Kaneko, Tetrahedron Lett. 1968, 10, 1705; g) N. Takao, N.
Bessho, M. Kamigauchi, K. Iwasa, K. Tomita, T. Fujiwara, S.
Fugii, Tetrahedron Lett. 1979, 20, 495.
[10] A. M. Qandil, D. Ghosh, D. E. Nichols, J. Org. Chem. 1999, 64,
1407.
[11] J.-L. Gras, Y.-Y. K. W. Chang, A. Guerin, Synthesis 1985, 74.
[12] Further optimization by varying the reaction conditions with
(S)-tol-binap as the ligand was attempted. However no improve-
ment in the ee value was observed; see the Supporting Informa-
tion.
[13] C. Hermann, G. C. G. Pais, A. Geyer, S. M. Kꢀhnert, M. E.
Maier, Tetrahedron 2000, 56, 8461.
[14] R. E. Ireland, M. D. Varney, J. Org. Chem. 1986, 51, 635.
[15] B. Cao, H. Park, M. M. Joulliꢁ, J. Am. Chem. Soc. 2002, 124, 520.
[16] D. L. Boger, S. R. Brunette, R. M. Garbaccio, J. Org. Chem.
2001, 66, 5163.
[3] J. M. Probst, Ann. Pharm. 1839, 29, 113.
[17] A. S.-Y. Lee, Y.-J. Hu, S.-F. Chu, Tetrahedron 2001, 57, 2121.
[18] H. Monti, G. Lꢁandri, M. Klos-Ringuet, C. Corriol, Synth.
Commun. 1983, 13, 1021; PPTS = pyridinium p-toluenesulfo-
nate.
[4] G. Grynkiewicz, E. Chojecka-Koryn, M. Gadzikowska, A.
Chodkowska, E. Jagiello-Wójtowicz, Eur. J. Med. Chem. 2001,
36, 951.
[5] A. Panzer, A. M. Joubert, P. C. Bianchi, E. Hamel, J. C. Seegers,
Eur. J. Cell Biol. 2001, 80, 111.
[6] J. W. Nowicky, G. Manolakis, D. Meijer, V. Vatanasapt, W. J.
Brzosko, Drugs Exp. Clin. Res. 1992, 18, 1.
[19] S. K. Panday, N. Langlois, Synth. Commun. 1997, 27, 1373.
[20] R. S. Lott, V. S. Chauhan, C. H. Stammer, J. Chem. Soc. Chem.
Commun. 1979, 495.
[21] See the Supporting Information.
436
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 433 –436