Efficient entry to the hasubanan alkaloids: First enantioselective total syntheses of (-)-hasubanonine, (-)-runanine, (-)-delavayine, and (+)-periglaucine B

Article abstract of DOI:10.1002/anie.201102226

Maximized divergence: The hasubanan alkaloids given in the scheme have been synthesized in eight or nine steps from the aryl azide 1. The utility of 5-trimethylsilylcyclopentadiene as an easily removed, stabilizing stereocontrol element has been demonstrated.

Full text of DOI:10.1002/anie.201102226

DOI: 10.1002/anie.201102226
Natural Products Synthesis
Efficient Entry to the Hasubanan Alkaloids: First Enantioselective
Total Syntheses of (ꢀ)-Hasubanonine, (ꢀ)-Runanine, (ꢀ)-Delavayine,
and (+)-Periglaucine B**
Seth B. Herzon,* Nicholas A. Calandra, and Sandra M. King
In memory of David Y. Gin
We describe herein a simple and general logic strategy to
synthesize the hasubanan alkaloids,[1] a family of over 40
botanical natural products that share a common tetracyclic
propellane skeleton (see structure 1, Figure 1). The versatility
synthesis of (+)-cepharamine[11] and Castleꢀs synthesis of
(ꢀ)-acutumine.[12]
Our approach was designed to permit access to a maximal
number of targets. Retrosynthetically, deconstruction of the
hasubanan scaffold 7 by the pathway shown in Scheme 1A
Figure 1. Structures of the hasuban skeleton 1, the alkaloids 2–5, and
the starting material (6) used in this work.
of our approach is evinced by the first enantioselective
synthesis of (ꢀ)-hasubanonine (2),[2] the foremost member of
this family to be isolated, as well as by those of (ꢀ)-runanine
(3),[3] (ꢀ)-delavayine (4),[4] and (+)-periglaucine B (5),[5] in
eight or nine steps from 5-(2-azidoethyl)-1,2,3-trimethoxy-
benzene (6).[6] Hasubanonine (2)[7] and the related metabo-
lites metaphanine[8] and cepharamine[9] have been previously
prepared in racemic form, and numerous partial and formal
syntheses of other hasubanan alkaloids have also been
reported.[10] However, the only enantioselective total synthe-
ses of alkaloids bearing similarity to 2–5 are Schultzꢀs
Scheme 1. Synthetic strategy to access the hasubanan alkaloids.
A) Retrosynthetic analysis. B) Proposed synthesis of the intermediate
13.
affords the pronucleophile 8 and the azaquinone 9 as hypo-
thetical intermediates. We anticipated that 8 would serve as a
useful branching point for incorporation of the varying arene
substitution patterns found in the hasubanans. The azaqui-
none 9, which contains the trioxygenated cyclohexanone
fragment common to 2–5 and electrophilic sites (labeled a and
b) for attachment of 8, served as a nearly ideal synthetic
intermediate.
Bicyclic azaquinones such as 9 are unstable toward
isomerization to 5-hydroxyindoles (10)[13] by a pathway that
may comprise tautomerization and a 1,5-hydrogen atom shift,
as shown. Transient introduction of a quaternary center on the
azaquinone (at position b) was expected to mitigate this
pathway and convert 9 to a viable synthetic intermediate.
Thus, we considered implementing a Diels–Alder reaction
between the azidoquinone 11 and 5-trimethylsilylcyclopenta-
diene (12, Scheme 1B). Staudinger reduction would then
afford the tetracyclic imine 13. Following bond formation to
[*] Dr. S. B. Herzon, N. A. Calandra, S. M. King
Department of Chemistry, Yale University
225 Prospect Street, New Haven, CT, 06520-8107 (USA)
Fax: (+1)203-432-6144
E-mail: seth.herzon@yale.edu
Homepage: http://www.chem.yale.edu/herzongroup
[**] We thank Nathan Schley and Dr. Christopher Incarvito for X-ray
analysis of 29 and 30. Financial support from Yale University, the
Searle Scholars Program, Eli Lilly, Boehringer-Ingelheim, and the
Department of Defense (NDSEG fellowship to S.M.K) is gratefully
acknowledged.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201102226.
Angew. Chem. Int. Ed. 2011, 50, 8863 –8866
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8863

Communications
the imine (position a), the unsaturation required for addition
13 might provide a handle for stereocontrol, although
enantioselective Diels–Alder reactions employing 12 were
unknown at the outset of our studies.
to position b may be regenerated by a retro-cycloaddition
reaction, rendering 13 functionally equivalent to 9. Concep-
tually related strategies have been previously applied using
cyclopentadiene or anthracene as blocking groups.[14] How-
ever, in the absence of two activating substituents, retro-
cycloaddition reactions of these adducts require conditions
(220–2508C, diphenyl ether, or 400–5008C, flash-vacuum
pyrolysis) that were expected to be incompatible with
functionalized advanced intermediates. By comparison,
The successful implementation of this strategy and total
syntheses of 2–5 are shown in Scheme 2. Imine 13 was
synthesized by a three-step sequence. First, 5-(2-azidoethyl)-
1,2,3-trimethoxybenzene (6)[6] was oxidized with hydrogen
peroxide in formic acid[16] to afford quinone 11 (48%). Regio-
and stereoselective Diels–Alder reaction with 5-trimethylsi-
lylcyclopentadiene (12),[17] mediated by the protonated form
of the Corey–Bakshi–Shibata oxazaborolidine 14,[18] pro-
duced the endo adduct 15 in 78% yield and 93% ee. The
selectivity of the addition reaction was anticipated based on
existing mechanistic models[18c] and was confirmed by X-ray
analysis of the related intermediate 29 (vide infra). Staudinger
reduction of 15 provided imine 13 in quantitative yield.
Imine 13 was transformed to (ꢀ)-hasubanonine (2) by a
short six-step sequence. First, 13 was activated toward
Diels–Alder
adducts
incorporating
the
7-
(trimethylsilyl)bicyclo[2.2.1]hept-2-ene substructure have
been shown to undergo significantly faster retro-cycloaddi-
tion reactions.[15] This rate enhancement has been attributed
to donation of electron density from the carbon–silicon
bonding orbital to the antibonding orbitals of the carbon–
carbon s bonds that are breaking in the reaction transition
state. We also envisioned that the cyclopentene fragment of
Scheme 2. Enantioselective total syntheses of (ꢀ)-hasubanonine (2), (ꢀ)-runanine (3), (ꢀ)-delavayine (4), and (+)-periglaucine B (5). Reagents
and conditions: 1. H2O2, HCO2H, 08C, 48%; 2. (S)-o-tol-CBS (14, 25 mol%), TfOH (20 mol%), 5-trimethylsilylcyclopentadiene (12), CH2Cl2,
ꢀ788C, 78%, 93% ee; 3. P(CH3)3, Et2O, 0!248C, 99%; 4a. CH3OTf, THF, ꢀ78!ꢀ60!ꢀ908C, then 17, 62%; 5a. PhCH3, 1358C, 86%;
6a. Crabtree’s catalyst, TFA, H2, CH2Cl2, 248C, 62%; 7a. TfOH, CH3CN, 0!248C, 75%; 8a. Bu3SnH, AIBN, PhCH3, 908C, 83%; 9a. [RhCl(PPh3)3],
H2 (1000 psi), TFA, PhCH3, 248C, 61%. 4b. CH3OTf, THF, ꢀ78!ꢀ60!ꢀ908C, then 21, 94%; 5b. PhCH3, 1358C, 85%; 6b. Crabtree’s catalyst,
TFA, H2, CH2Cl2, 248C, 81%; 7b. TfOH, CH2Cl2, ꢀ308C, 72%; 8b. [RhCl(PPh3)3], H2 (1000 psi), TFA, PhCH3, 248C, 65%. 4c. CH3OTf, THF, ꢀ78!
ꢀ60!ꢀ908C, then 24, 73%; 5c. PhCH3, 1358C, 87%; 6c. Crabtree’s catalyst, TFA, H2, CH2Cl2, 248C, 78%; 7c. TfOH, CH2Cl2, ꢀ40!ꢀ208C, 89%;
8c. [RhCl(PPh3)3], H2 (1000 psi), TFA, PhCH3, 248C, 73%. 9b. Co(acac)2, O2 (1 atm), isopropanol, 758C, then HCO2H, 758C, 55% (ratio of 28/
diastereomer 2.2:1). THF=tetrahydrofuran, TFA=trifluoroacetic acid, Tf=trifluoromethanesulfonyl, AIBN=azobisisobutyronitrile, acac=acetyla-
cetonate.
8864 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 8863 –8866

addition to the carbon–nitrogen p bond by treatment with
methyl triflate at ꢀ608C. The temperature profile of this step
was critical; lower temperatures resulted in incomplete
methylation, whereas further warming promoted rapid
retro-cycloaddition of the iminium salt 16. The iminium salt
16 was then cooled to ꢀ908C and treated with the acetylide
17, resulting in the formation of the 1,2-addition product 18 in
62% yield, as a single detectable diastereomer (1H NMR
analysis). The relative stereochemistry of the addition prod-
uct 18 was assigned by elaboration to (ꢀ)-hasubanonine (2)
and by analogy to that of the related crystalline product 29
(vide infra). The retro-cycloaddition reaction of 18 was
achieved by heating in toluene at 1358C (86% yield).
Chemoselective hydrogenation using Crabtreeꢀs catalyst
(19)[19] furnished the cis alkene 20 (62%). A three-step
sequence comprising acid-mediated cyclization (75%),
debromination (83%), and hydrogenation (61%) then pro-
vided synthetic (ꢀ)-hasubanonine (2).
Scheme 3. Retro-cycloaddition of the 2-(trimethylsilyl)acetylene addi-
tion products 29 and 30. 1. PhCH3, 1358C, 99%; 2. Ph2O, 2208C, 15%.
We next prepared the alkaloids (ꢀ)-runanine (3) and (ꢀ)-
delavayine (4). To access (ꢀ)-runanine (3), the acetylide 21
was added to the iminium ion 16, to afford the 1,2-addition
product 22 (94%). To access (ꢀ)-delavayine (4), the acetylide
24 was employed in the addition step, affording the 1,2-
addition product 25 (73%). Two four-step sequences were
used to convert 22 and 25 to (ꢀ)-runanine (3) and (ꢀ)-
delavayine (4), respectively.
As the alkene 27, which is the penultimate precursor to
(ꢀ)-delavayine (4), might be converted to (+)-periglaucine B
(5) directly by a formal olefin hydration/conjugate addition
sequence, we surveyed a number of conditions to effect this
transformation. We found that the desired hydration product
28 could be formed by heating a mixture of 27 and cobalt
bis(acetylacetonate) in isopropanol under an atmosphere of
dioxygen.[20] The diastereoselectivity in the hydration step was
2.2:1 in favor of 28. Addition of excess formic acid directly to
the reaction mixture promoted cyclization of 28, providing
(+)-perglaucine B (5) in 55% yield.[21]
inductive effect of the trimethylsilyl substituent is manifested
primarily in the transition state for the retro-cycloaddition
reaction, potentially in the form of an asynchronous, polar-
ized transition structure.
In summary, we have completed the first enantioselective
total syntheses of (ꢀ)-hasubanonine (2), (ꢀ)-runanine (3),
(ꢀ)-delavayine (4), and (+)-periglaucine B (5). Our route to
each target proceeds in eight or nine steps from the aryl azide
6 (the latter was obtained in three steps from commercial
reagents, without purification of intermediates). Our
approach has also demonstrated the utility of 5-trimethyl-
silylcyclopentadiene 12 as an easily removable, stabilizing,
stereocontrol element in the synthesis of complex molecules.
The logic developed in these studies is likely to find
application in the synthesis of other members of this large
family of alkaloids.
Received: March 31, 2011
Published online: June 3, 2011
In the context of the work reported herein, 5-trimethyl-
silylcyclopentadiene (12) has served the dual purpose of
stabilizing the azaquinone 9 and providing a handle for setting
the absolute stereochemistry. The facility with which the
retro-cycloaddition reaction occurs, even in the presence of
only a single activating group, suggests this novel controlling
group may find application in other settings. To illustrate
clearly the rate enhancement provided by the trimethylsilyl
substituent, we prepared the homologous 2-(trimethylsilyl)-
acetylene addition products 29 and 30 (Scheme 3, see the
Supporting Information for details). Thermolysis of 29
(toluene, 1358C, 3 h) afforded the retro-cycloaddition prod-
uct 31 in quantitative yield. By comparison, higher temper-
ature (2208C) was required to promote retro-cycloaddition of
the unsubstituted adduct 30, leading to extensive decompo-
sition and low yield of 31 (15%).[22] Relevant to the experi-
ments above, both 29 and 30 were highly crystalline, allowing
unambiguous determination of relative (and in the case of 29,
absolute) stereochemistry by X-ray analysis.[23] The crystallo-
graphic data reveal that the carbon–carbon s bonds that are
broken in these transformations (shown in red) are of nearly
the same length in 29 and 30 (1.56–1.57 ꢁ). Thus, the
Keywords: cycloaddition · hasubanan alkaloids ·
natural products
.
[1] a) M. Matsui in The Alkaloids, Vol. 33, Academic Press, New
York, 1988, p. 307; b) A. J. Pihko, A. M. P. Koskinen, Tetrahe-
dron 2005, 61, 8769.
[2] M. Tomita, T. Ibuka, Y. Inubushi, Y. Watanabe, M. Matsui,
Tetrahedron Lett. 1964, 5, 2937.
[3] M. Zhi-Da, L. Ge, X. Guang-Xi, M. Iinuma, T. Tanaka, M.
Mizuno, Phytochemistry 1985, 24, 3084.
[4] A. Patra, Phytochemistry 1987, 26, 2391.
[5] M.-H. Yan, P. Cheng, Z.-Y. Jiang, Y.-B. Ma, X.-M. Zhang, F.-X.
Zhang, L.-M. Yang, Y.-T. Zheng, J.-J. Chen, J. Nat. Prod. 2008,
71, 760.
[6] J. A. Soderquist, I. Kock, M. E. Estrella, Org. Process Res. Dev.
2006, 10, 1076.
[7] a) T. Ibuka, K. Tanaka, Y. Inubushi, Tetrahedron Lett. 1970, 11,
4811; b) T. Ibuka, K. Tanaka, Y. Inubushi, Chem. Pharm. Bull.
1974, 22, 782.
[8] a) T. Ibuka, K. Tanaka, Y. Inubushi, Tetrahedron Lett. 1972, 13,
1393; b) T. Ibuka, K. Tanaka, Y. Inubushi, Chem. Pharm. Bull.
1974, 22, 907.
Angew. Chem. Int. Ed. 2011, 50, 8863 –8866
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8865

Communications
[9] a) Y. Inubushi, M. Kitano, T. Ibuka, Chem. Pharm. Bull. 1971,
review, see: b) A. J. H. Klunder, J. Zhu, B. Zwanenburg, Chem.
Rev. 1999, 99, 1163.
[15] P. Magnus, P. M. Cairns, J. Moursounidis, J. Am. Chem. Soc.
1987, 109, 2469.
19, 1820; b) T. Kametani, H. Nemoto, T. Kobari, K. Shishido, K.
Fukumoto, Chem. Ind. 1972, 13, 538.
[10] a) M. Tomita, T. Ibuka, M. Kitano, Tetrahedron Lett. 1966, 7,
6233; b) T. Ibuka, M. Kitano, Chem. Pharm. Bull. 1967, 15, 1944;
c) M. Tomita, M. Kitano, T. Ibuka, Tetrahedron Lett. 1968, 9,
3391; d) D. A. Evans, Tetrahedron Lett. 1969, 10, 1573; e) D. A.
Evans, C. A. Bryan, G. M. Wahl, J. Org. Chem. 1970, 35, 4122;
f) S. L. Keely, Jr., A. J. Martinez, F. C. Tahk, Tetrahedron 1970,
26, 4729; g) D. A. Evans, C. A. Bryan, C. L. Sims, J. Am. Chem.
Soc. 1972, 94, 2891; h) T. Kametani, T. Kobari, K. Fukumoto, J.
Chem. Soc. Chem. Commun. 1972, 288; i) B. Belleau, H. Wong, I.
Monkovic, Y. G. Perron, J. Chem. Soc. Chem. Commun. 1974,
603; j) T. Kametani, T. Kobari, K. Shishido, K. Fukumoto,
Tetrahedron 1974, 30, 1059; k) I. Monkovic, H. Wong, B. Belleau,
I. J. Pachter, Y. G. Perron, Can. J. Chem. 1975, 53, 2515; l) I.
Monkovic, H. Wong, Can. J. Chem. 1976, 54, 883; m) S. Shiotani,
T. Kometani, Tetrahedron Lett. 1976, 17, 767; n) H. Bruderer, D.
Knopp, J. J. Daly, Helv. Chim. Acta 1977, 60, 1935; o) M. A.
Schwartz, R. A. Wallace, Tetrahedron Lett. 1979, 20, 3257; p) D.
Trauner, S. Porth, T. Opatz, J. W. Bats, G. Giester, J. Mulzer,
Synthesis 1998, 653; q) S. B. Jones, L. He, S. L. Castle, Org. Lett.
2006, 8, 3757; r) T. X. Nguyen, Y. Kobayashi, J. Org. Chem. 2008,
73, 5536; s) D. K. Nielsen, L. L. Nielsen, S. B. Jones, L. Toll,
M. C. Asplund, S. L. Castle, J. Org. Chem. 2009, 74, 1187.
[11] A. G. Schultz, A. Wang, J. Am. Chem. Soc. 1998, 120, 8259.
[12] a) F. Li, S. S. Tartakoff, S. L. Castle, J. Am. Chem. Soc. 2009, 131,
6674; b) F. Li, S. S. Tartakoff, S. L. Castle, J. Org. Chem. 2009, 74,
9082.
[16] H. Orita, M. Shimizu, T. Hayakawa, K. Takehira, Bull. Chem.
Soc. Jpn. 1989, 62, 1652.
[17] A. J. Ashe, J. Am. Chem. Soc. 1970, 92, 1233.
[18] a) E. J. Corey, T. Shibata, T. W. Lee, J. Am. Chem. Soc. 2002, 124,
3808; b) D. H. Ryu, T. W. Lee, E. J. Corey, J. Am. Chem. Soc.
2002, 124, 9992; c) D. H. Ryu, G. Zhou, E. J. Corey, J. Am. Chem.
Soc. 2004, 126, 4800. For a review, see: d) E. J. Corey, Angew.
Chem. 2002, 114, 1724; Angew. Chem. Int. Ed. 2002, 41, 1650.
[19] a) R. H. Crabtree, H. Felkin, G. E. Morris, J. Organomet. Chem.
1977, 141, 205; b) J. W. Suggs, S. D. Cox, R. H. Crabtree, J. M.
Quirk, Tetrahedron Lett. 1981, 22, 303.
[20] S. Inoki, K. Kato, T. Takai, S. Isayama, T. Yamada, T.
Mukaiyama, Chem. Lett. 1989, 515.
[21] NMR spectroscopic data for synthetic 2–5 were in complete
accord with those reported for the natural isolates. However,
inspection of these data revealed that many of the resonances of
natural 2–5 were incorrectly assigned. These were clarified by
2D NMR and NOESY experiments. Complete spectroscopic
data and revised assignments are presented in the Supporting
Information.
[22] Grieco has reported efficient Lewis acid catalyzed retro-cyclo-
addition reactions of cyclopentadiene Diels – Alder adducts.
P. A. Grieco, N. Abood, J. Org. Chem. 1989, 54, 6008. However,
30 did not undergo retro-cycloaddition in the presence of several
Lewis and protic acid additives.
[23] Intermediate 29 was prepared in enantiomerically enriched form
using the catalyst 14. The absolute stereochemistry depicted is
supported by the observed Flack [0.15(0.14)] and Hooft [0.00-
(0.12)] parameters. Intermediate 30 was prepared as a racemate.
CCDC 819339 (29) and 819340 (30) contain the supplementary
crystallographic data for this paper. These data can be obtained
free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
[13] a) R. I. T. Cromartie, J. Harley-Mason, J. Chem. Soc. 1952, 2525;
b) J. Harley-Mason, A. H. Jackson, J. Chem. Soc. 1954, 1165;
c) Y. Ozaki, Z.-S. Quan, K. Watabe, S.-W. Kim, Heterocycles
1999, 51, 727; d) E. J. L. Stoffman, D. L. J. Clive, Org. Biomol.
Chem. 2009, 7, 4862.
[14] For a seminal early report, see: a) G. Stork, G. L. Nelson, F.
Rouessac, O. Gringore, J. Am. Chem. Soc. 1971, 93, 3091. For a
8866 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 8863 –8866