Total Synthesis of Actinophyllic Acid

Article abstract of DOI:10.1002/anie.201706312

Herein we report a total synthesis of the indolohydroazocine natural product actinophyllic acid. The target molecule was retrosynthetically deconvoluted to render a greatly simplified and symmetrical [4.4.1] bicyclic trienone, the desymmetrization of which was carefully examined under a variety of conditions, including oxidative, reductive, and transition-metal-catalyzed transformations. Ultimately, the successful synthetic strategy featured chemoselective catalytic dihydroxylation, desymmetrizing nitrile oxide dipolar cycloaddition, and palladium-catalyzed aminoarylation to sequentially modify the three olefins within the trienone, followed by a late-stage reductive cascade indolization and alkylation to complete the target molecule.

Full text of DOI:10.1002/anie.201706312

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201706312
German Edition: DOI: 10.1002/ange.201706312
Natural Products Synthesis
Total Synthesis of Actinophyllic Acid
Yu Yoshii, Hidetoshi Tokuyama, and David Y.-K. Chen*
Dedicated to Professor Yoshito Kishi on the occasion of his 80th birthday
Abstract: Herein we report a total synthesis of the indolohy-
droazocine natural product actinophyllic acid. The target
molecule was retrosynthetically deconvoluted to render
a greatly simplified and symmetrical [4.4.1] bicyclic trienone,
the desymmetrization of which was carefully examined under
a variety of conditions, including oxidative, reductive, and
transition-metal-catalyzed transformations. Ultimately, the
successful synthetic strategy featured chemoselective catalytic
dihydroxylation, desymmetrizing nitrile oxide dipolar cyclo-
addition, and palladium-catalyzed aminoarylation to sequen-
tially modify the three olefins within the trienone, followed by
a late-stage reductive cascade indolization and alkylation to
complete the target molecule.
P
lane symmetry is a ubiquitous feature in the chemical
sciences that has captivated our imaginations in terms of both
its generation and function. By definition, a compound
possessing plane symmetry harbors two identical structural
domains with mirrored chemical reactivity, and a selective
operation on one of its two identical structural domains
renders an asymmetric product with a new identity and
reactivity. This process, so-called desymmetrization, has
profoundly impacted organic synthesis, from the wide-
spread preparation of simple building blocks to highly
Scheme 1. a) Selected recent examples of desymmetrization in target-
oriented synthesis. b) Use of the [4.4.1] bicyclic trienone 2, with
a plane of symmetry, as a precursor to actinophyllic acid (1).
unique architecture adding to the scientific challenge (Sche-
[1]
[3]
elaborated molecular architectures. The success of desym-
metrization strategies relies heavily on the chemical method
of choice and the timing of its implementation in the overall
synthetic scheme. In the context of target-oriented total
synthesis, impressive feats have been realized through the
me 1b). Moreover, it would be possible to measure the
success of our proposed desymmetrization strategy against
the benchmark set by the elegant and efficient syntheses of
actinophyllic acid (1) by the Overman,
Kwon research groups. In accordance with the terminology
used by Hanessian, deconvolution of the cagelike structure of
actinophyllic acid (1) is by large a reflexive process. After
several generations of symmetry-driven retrosynthetic simpli-
fication, the [4.4.1] bicyclic trienone 2 was revealed as
a plausible synthetic precursor.
[
4a,b]
[4f,h]
Martin,
and
[
4i]
[
2]
exquisite insight of the practitioners (Scheme 1a). Although
the judicious application of desymmetrization-based synthetic
strategies can be highly advantageous, more importantly,
desymmetrization represents a generalized “thought process”
on the basis of which a synthetic plan could be logically
[
5]
[
1e]
[6]
conceived.
The simplicity of trienone 2 and its ready availability
The power of desymmetrization-based synthetic strategies
in target-oriented synthesis is best appreciated when the
target molecule does not possess any symmetry elements.
Actinophyllic acid (1) readily fulfills this criterion, with its
were particularly attractive; however, its three olefinic
residues had to be modified sequentially to fulfill the
objective of desymmetrization. In this context, upon ketal
protection of trienone 2 (ethylene glycol, p-TsOH·H O), we
2
soon discovered that under carefully controlled reaction
conditions, the isolated olefin could be selectively dihydroxy-
[
*] Y. Yoshii, Prof. Dr. D. Y.-K. Chen
lated (OsO , NMO, 78% yield over two steps) without
4
Department of Chemistry, Seoul National University
Gwanak-1 Gwanak-ro, Gwanak-gu, Seoul 151-742 (South Korea)
E-mail: davidchen@snu.ac.kr
[7]
affecting the conjugated diene (Scheme 2). This latter
process was followed by oxidative cleavage of the dihydroxy-
lated carbocycle within 3 (Pb(OAc) ) and subsequent recon-
4
Y. Yoshii, Prof. Dr. H. Tokuyama
Graduate School of Pharmaceutical Sciences, Tohoku University
Aoba, Aramaki, Aoba-ku, Sendai 980-8578 (Japan)
stitution of the cyclic framework through double reductive
amination of the intermediate dialdehyde with PMBNH in
2
6
5% overall yield. As a precautionary maneuver, particularly
Supporting information for this article can be found under:
https://doi.org/10.1002/anie.201706312.
in view of the likelihood of employing oxidative transforma-
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
Scheme 2. Total synthesis of actinophyllic acid hydrochloride (1·HCl). Reagents and conditions: a) p-TsOH·H O (1.0 equiv), ethylene glycol, 238C,
2
3
2
h; b) OsO (0.06 mol%), NMO (1.5 equiv), acetone, 238C, 72 h, 78% for two steps; c) Pb(OAc) (1.5 equiv), NaHCO (2.8 equiv), benzene,
4
4
3
38C, 1 h; then MS4ꢀ (1.3 wt equiv), NaBH(OAc) (2.4 equiv), PMBNH (1.7 equiv), 1,2-dichloroethane, 238C, 1.5 h; then NaHCO (2.8 equiv),
3
2
3
Boc O (1.8 equiv), 1,2-dichloroethane, 238C, 1 h, 65%; d) TeocCl (1.3 equiv), CH Cl , 238C, 1 h, 89%; e) 6 (2.2 equiv), KHCO (3.3 equiv), ethyl
2
2
2
3
acetate, 238C, 48 h, 68%; f) LiOMe (1.0m in MeOH, 20 equiv), MeOH, 508C, 72 h; g) DDQ (1.0 equiv), toluene, reflux, 16 h, 36% for two steps;
h) TFA/CH Cl (1:10), 238C, 12 h; then 9 (3.0 equiv), Cs CO (5.0 equiv), (2-furyl) P (1.6 equiv), Pd(OAc) (0.8 equiv), THF, reflux, 4 h, 65%; i) HCl
2
2
2
3
3
2
(
(
4.0m aqueous)/MeOH (1:1), reflux, 48 h, 76%; j) PdCl (4.5 equiv), MeOH/H O/AcOH (8:2:1), H (1 atm), 238C, 6 h; then TFA; k) LDA
2
2
2
4.3 equiv), THF, À788C, 0.5 h; then CH O (ca. 0.5m in THF, 14 equiv), À788C, 5 min; then TFA (0.9m in THF, 14 equiv), 238C; then HCl (4.0m
2
aqueous), 708C, 13 h, 15% overall yield from 11. Boc O=di-tert-butyl dicarbonate; DDQ=2,3-dichloro-5,6-dicyano-1,4-benzoquinone; LDA=
2
lithium diisopropylamide; NMO=N-methylmorpholine N-oxide; PMB=para-methoxybenzyl; Teoc=2-(trimethylsilyl)ethylcarbonyl; TFA=trifluoro-
acetic acid; p-TsOH·H O=para-toluenesulfonic acid monohydrate.
2
tions or p-philic Lewis acidic reagents on the conjugated
diene, p-methoxybenzyl amine 4 was readily converted into
Teoc carbamate 5 (TeocCl) in 89% yield.
Next, we carefully analyzed the options to further
elaborate intermediates 17, 20, and 21 towards actinophyllic
acid (1), particularly in connection with the introduction of
the requisite indole moiety. In this analysis, we soon realized
that although several plausible synthetic pathways could be
With the bicyclic diene carbamate 5 in hand, the stage was
set for the key desymmetrization event (Scheme 3). In view of
the rich repertoire of alkene- or conjugated-diene-specific
transformations, while keeping in mind the structure of the
target molecule, the methods initially considered for the
proposed desymmetrization were largely focused on trans-
formations that would enable the installation of the indole
moiety required for actinophyllic acid (1). In this context,
several arylation protocols were examined but soon aban-
doned primarily owing to reactivity and selectivity issues
(
Scheme 3a). Oxidative or redox-neutral transformations
that involved either oxygenation or amination also proved
largely ineffective. We were pleased to find, however,
synthetically useful hydrogenation conditions that afforded
a nearly equal mixture of alkenes 17 (the first desymmetrized
intermediate obtained) and 18 in 80% combined yield
(
Scheme 3b).
Although the precise mechanism behind the formation of
[8]
the migratory hydrogenation product 18 is unclear, alkene
1
8 underwent catalytic dihydroxylation (OsO , [K Fe(CN) ],
4 3 6
5
8% yield), and the resulting diol 19 could be desymmetrized
by silylation (TBSCl, 84% yield). Presumably, the steric
pressure within TBS ether 20 prevented further silylation.
Alcohol 20 could be oxidized (IBX) to provide ketone 21 as
a synthetically more versatile intermediate. The success of the
selective silylation has a broader implication in view of the
recent advances in enantioselective desymmetrization silyla-
Scheme 3. a) Early exploration of the desymmetrization of diene 5.
b) Synthesis of unsymmetrical alkene 17, bis(TBS ether) 20, and
ketone 21. See the Supporting Information for details. IBX=2-iod-
oxybenzoic acid; TBS=tert-butyldimethylsilyl.
[9]
tion reactions, thus suggesting potential access to optically
active intermediates.
2
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
devised on the basis of intermediates 17, 20, and 21, the
accepted mechanism postulated for the aminoarylation
[
11]
overall number of synthetic steps was likely to far exceed
reaction.
We initially speculated that the originally
[
4a,b]
[4f,h]
those required by the Overman,
Martin,
and Kwon
formed aminoarylation product might have undergone epi-
merization, or that the arylation reaction might take place
without the assistance of the amino group. However, deuter-
ation and control experiments failed to support these spec-
ulations (Scheme 5a,b). Interestingly, cleavage of the Teoc
[4i]
groups. Therefore, an alternative desymmetrization solu-
tion had to be identified to ensure comparable overall
efficiency.
Gratifyingly, our continued efforts in desymmetrization
ultimately led to the discovery of a cycloaddition process that
proved highly effective on diene 5 (Scheme 2). In this
instance, the bromonitrile oxide generated in situ from
dibromooxime 6 underwent regioselective [2+3] cycloaddi-
tion with diene 5 to afford isoxazoline 7 in 68% yield as
[
10]
a single stereoisomer. The use of KHCO was previously
3
[
10b]
adopted by the Baran group,
was routinely performed on a gram scale with excellent
reproducibility. Bromoisoxazoline further underwent
and the optimized procedure
7
methoxy substitution (MeOLi) and oxidation (DDQ) to
afford isoxazole 8, a masked b-ketoester with a concealed
carboxylate group and the desired C2 (actinophyllic acid
numbering) oxidation state for actinophyllic acid (1).
At this juncture, following the selective functionalization
of two of the three olefins in triene 2, the remaining alkene in
8
was poised to undergo an arylation reaction to render
a substrate for late-stage indolization. Moreover, the pro-
posed arylation could be made more appealing by synchro-
nizing it with the formation of the adjacent pyrrolidine, a dual
objective well-suited for the palladium-catalyzed aminoary-
[
11]
lation reaction developed by Wolfe and co-workers. Teoc
carbamate 8 was first treated with TFA to unmask its
secondary amine while leaving the acid-labile dioxolane
intact, and the resulting amine·TFA salt was directly sub-
jected to the aminoarylation conditions. The yield of amino-
arylation product 10 varied depending on the reaction
conditions (Scheme 4; see also the Supporting Information).
The choice of aryl halide, phosphine ligand, base, and solvent
were all examined. Although an increase in the catalyst
loading had a positive effect on the reaction yield, we opted
for a lower catalyst loading for economic reasons and
routinely recycled the crude reaction mixture.
Scheme 5. Probing the stereochemical course of the aminoarylation
reaction (see the Supporting Information for details).
carbamate, followed by workup with aqueous NaHCO3,
afforded cyclized pyrrolidine 22 (Scheme 5c), a compound
that failed to undergo arylation under our standard reaction
conditions (Scheme 5d). At the present, our working model
for the observed stereochemical outcome is consistent with an
anti-aminoarylation pathway instead of the originally antici-
pated syn-aminoarylation (Scheme 5e), in which the electro-
philic palladium center together with the conformational
constraint that prohibited proper PdÀN and C=C alignment
The stereochemical assignment of aminoarylation product
1
0 was established on the basis of extensive 2D NMR studies,
and although inconsequential for our synthesis, the observed
stereostructure was not in agreement with the generally
would appear to favor anti-aminoarylation under reversible
[12]
palladation conditions. Furthermore, coordination of the
palladium center with a dioxolane oxygen atom could also
[
13]
contribute to the observed stereochemical outcome.
With nitroarene 10 in readiness for the planned indoliza-
tion, the requirement to reduce its nitro group together with
the rupture of the isoxazole moiety motivated us to execute
both reductive events as a single operation, followed by
spontaneous indolization (Scheme 2). This proposed cascade
transformation proved nontrivial in practice. In particular, the
isoxazole moiety was resilient towards a variety of reducing
conditions examined. Ultimately, upon removal of the
Scheme 4. Summary of the optimization of the aminoarylation. All
reactions were performed under an Ar atmosphere (see the Supporting
Information for details).
dioxolane moiety in 10 (HCl, MeOH), PdCl -catalyzed
2
hydrogenation successfully afforded the targeted indole 12.
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
The final conversion of indole 12 into actinophyllic acid (1)
[2] For selected recent examples of desymmetrization-based target-
oriented total synthesis, see: a) J. T. Malinowski, R. J. Sharpe,
J. S. Johnson, Science 2013, 340, 180; b) R. J. Sharpe, J. T.
Malinowski, J. S. Johnson, J. Am. Chem. Soc. 2013, 135, 17990;
c) M. Nagatomo, M. Koshimizu, K. Masuda, T. Tabuchi, D.
Urabe, M. Inoue, J. Am. Chem. Soc. 2014, 136, 5916.
was realized through slight modification of the original
[
4a,b]
procedure reported by Overman and co-workers,
and
after acidic hydrolysis (HCl) provided synthetic actinophyllic
[
14]
acid hydrochloride (1·HCl) with spectroscopic properties in
[
4]
perfect agreement with those reported previously.
[
3] For isolation and structure elucidation of actinophyllic acid, see:
a) A. R. Carroll, E. Hyde, J. Smith, R. J. Quinn, G. Guymer, P. I.
Forster, J. Org. Chem. 2005, 70, 1096; b) T. Taniguchi, C. L.
Martin, K. Monde, K. Nakanishi, N. Berova, L. E. Overman, J.
Nat. Prod. 2009, 72, 430.
In conclusion, a desymmetrization-based total synthesis of
actinophyllic acid (1) has been realized. Insightful structural
deconvolution of the target molecule uncovered a symmet-
rical [4.4.1] bicycle 2 as a possible readily accessible starting
material. Sequential implementation of judiciously selected
and carefully optimized catalytic dihydroxylation, nitrile
oxide dipolar cycloaddition, and intramolecular aminoaryla-
tion effectively differentiated the three olefins within 2,
followed by reductive indolization and late-stage alkylation to
complete the total synthesis. The expedient synthetic pathway
showcases “strategy-based” design (desymmetrization) in
contrast to the “reaction-based” approach of the Overman
[
4] For synthetic studies towards actinophyllic acid and its total
synthesis, see: a) C. L. Martin, L. E. Overman, J. M. Rohde, J.
Am. Chem. Soc. 2008, 130, 7568; b) C. L. Martin, L. E. Overman,
J. M. Rohde, J. Am. Chem. Soc. 2010, 132, 4894; c) R. G.
Vaswani, J. J. Day, J. L. Wood, Org. Lett. 2009, 11, 4532; d) H.
Zaimoku, T. Taniguchi, H. Ishibashi, Org. Lett. 2012, 14, 1656;
e) I. Z. Galicia, L. A. Maldonado, Tetrahedron Lett. 2013, 54,
2180; f) B. A. Granger, I. T. Jewett, J. D. Butler, B. Hua, C. E.
Knezevic, E. I. Parkinson, P. J. Hergenrother, S. F. Martin, J. Am.
Chem. Soc. 2013, 135, 12984; g) D. Mortimer, M. Whiting, J. P. A.
Harrity, S. Jones, I. Coldham, Tetrahedron Lett. 2014, 55, 1255;
h) B. A. Granger, I. T. Jewett, J. D. Butler, S. F. Martin, Tetrahe-
dron 2014, 70, 4094; i) L. Cai, K. Zhang, O. Kwon, J. Am. Chem.
Soc. 2016, 138, 3298; j) F. Xue, T. Xiao, K.-F. Zhang, L.-P. He, Y.
Qin, X.-Y. Liu, D. Zhang, Tetrahedron 2017, 73, 2109.
(
aza-Cope Mannich), Martin (cationic cascade), and Kwon
groups (phosphine-catalyzed [3+2]). Further demonstration
and variation of desymmetrization-based target-oriented
total synthesis are currently in progress in our laboratory.
[
5] S. Hanessian, S. Giroux, B. L. Merner in Design and Strategy in
Organic Synthesis—From the Chiron Approach to Catalysis,
Wiley-VCH, Weinheim, 2013.
Acknowledgements
[
6] a) S. Itꢀ, H. Ohtani, S. Narita, H. Honma, Tetrahedron Lett. 1972,
This research was supported by the Seoul National University
Foreign Faculty Fund, the Seoul National University New
Faculty Resettlement Fund, a National Research Foundation
of Korea (NRF) grant funded by the Korean government
13, 2223; b) for a review of [6+4] cycloaddition reactions, see:
J. H. Rigby in Organic Reactions, Vol. 49 (Ed.: L. A. Paquette),
Wiley, New York, 1997, p. 331.
[
[
[
7] Protection (or derivatization) of the ketone moiety in 2 proved
crucial for the success of this process. For the selective
dihydroxylation of related systems, see: a) J. H. Rigby, V.
Saint Claire, S. V. Cuisiat, M. J. Heeg, J. Org. Chem. 1996, 61,
7992; b) J. H. Rigby, J. Hu, M. J. Heeg, Tetrahedron Lett. 1998,
39, 2265.
(
MSIP; Nos. 2012R1A2A2A01002895, 2013R1A1A2057837,
and 2014–011165, Center for New Directions in Organic
Synthesis), Novartis, Campus Asia Program, JSPS
KAKENHI (Grant Nos.JP16H01127, JP16H00999), and
a Grant-in aid for Scientific Research (A) (26253001). We
thank Professor John Wolfe (University of Michigan) for
a discussion on the palladium-catalyzed aminoarylation
reaction and Professor Seiji Mori (Ibaraki University) for
preliminary theoretical computation of the aminoarylation
reaction of substrate 8.
8] For related examples of migratory 1,4-hydrogenation, see:
a) C. J. Flann, L. E. Overman, J. Am. Chem. Soc. 1987, 109,
6115; b) K. S. Feldman, M.-J. Wu, D. P. Rotella, J. Am. Chem.
Soc. 1989, 111, 6457; c) M. Di Filippo, I. Izzo, S. Raimondi, F. D.
Riccardis, G. Sodano, Tetrahedron Lett. 2001, 42, 1575; d) F. Li,
N. C. Warshakoon, M. J. Miller, J. Org. Chem. 2004, 69, 8836.
9] For selected examples of desymmetrization through enantiose-
lective silylation, see: a) Y. Zhao, J. Rodrigo, A. H. Hoveyda,
M. L. Snapper, Nature 2006, 443, 67; b) Y. Zhao, A. Mitra, A. H.
Hoveyda, M. L. Snapper, Angew. Chem. Int. Ed. 2007, 46, 8471;
Angew. Chem. 2007, 119, 8623; c) J. M. Rodrigo, Y. Zhao, A. H.
Hoveyda, M. L. Snapper, Org. Lett. 2011, 13, 3778; d) X. Sun,
A. D. Worthy, K. L. Tan, Angew. Chem. Int. Ed. 2011, 50, 8167;
Angew. Chem. 2011, 123, 8317.
Conflict of interest
The authors declare no conflict of interest.
Keywords: alkaloids · cycloaddition · desymmetrization ·
natural products · total synthesis
[
10] For a pioneering example of the use of bromooxime 6 in dipolar
cycloaddition, see: a) A. A. Hagedorn, B. J. Miller, J. O. Nagy,
Tetrahedron Lett. 1980, 21, 229; for the use of KHCO3 in
a closely related dipolar cycloaddition reaction, see: b) T. J.
Maimone, J. Shi, S. Ashida, P. S. Baran, J. Am. Chem. Soc. 2009,
1
31, 17066.
[
1] For selected reviews, see: a) S. R. Magnuson, Tetrahedron 1995,
[
11] For a comprehensive study by the Wolfe group, see: a) B. M.
Bertrand, J. D. Neukom, J. P. Wolfe, J. Org. Chem. 2008, 73, 8851;
for a review, see: b) D. M. Shultz, J. P. Wolfe, Synthesis 2012, 351.
51, 2167; b) M. C. Willis, J. Chem. Soc. Perkin Trans. 1 1999,
1765; c) M. Wang, M. Feng, B. Tang, X. Jiang, Tetrahedron Lett.
2014, 55, 7147; d) J. Merad, M. Candy, J.-M. Pons, C. Bressy,
Synthesis 2017, 1938; for discussions on the strategic application
of desymmetrization in target-oriented synthesis, see: e) E. J.
Corey, X. M. Cheng in The Logic of Chemical Synthesis, Wiley-
VCH, Weinheim, 1989.
[12] Personal communication, Professor John P. Wolfe, University of
Michigan.
[13] Preliminary computational studies (Professor Seiji Mori, Ibaraki
University, Japan) suggested a dioxolane-coordinated species
4
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
along the reaction pathway. A more detailed theoretical study is
smoothly in the final transformations leading to actinophyllic
currently in progress.
acid hydrochloride.
[
14] The low yield of actinophyllic acid hydrochloride may be caused
by the isolation and purification of this highly polar compound.
LCMS analysis of the reaction progress and NMR analysis of the
crude reaction mixture indicated that all reactions proceeded
Manuscript received: June 21, 2017
Version of record online: && &&, &&&&
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Natural Products Synthesis
Y. Yoshii, H. Tokuyama,
D. Y.-K. Chen*
&&&& — &&&&
Total Synthesis of Actinophyllic Acid
Find a mirror to break: The indolohydro-
azocine natural product actinophyllic acid
was synthesized by an expedient route
based on the desymmetrization of a tri-
enone precursor. The developed synthe-
sis features chemoselective catalytic
dihydroxylation of the trienone, desym-
metrizing dipolar cycloaddition with a ni-
trile oxide, and palladium-catalyzed ami-
noarylation as key steps.
6
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!