A synthetic study of atropurpuran: Construction of a pentacyclic framework by an intramolecular reverse-electron-demand diels-alder reaction

Article abstract of DOI:10.1002/anie.201103950

Masked talent: A tetracyclo[5.3.3.0^4,9.0^4,12] tridecane skeleton can be accessed by an intramolecular reverse-electron-demand Diels-Alder (REDDA) reaction of masked ortho-benzoquinone (MOB; see scheme). This reaction gives access to

Full text of DOI:10.1002/anie.201103950

DOI: 10.1002/anie.201103950
Natural Product Synthesis
A Synthetic Study of Atropurpuran: Construction of a Pentacyclic
Framework by an Intramolecular Reverse-Electron-Demand
Diels–Alder Reaction**
Takahiro Suzuki, Aya Sasaki, Naoki Egashira, and Susumu Kobayashi*
Aconitum is a genus of flowering plants that produce a variety
of poisons and compounds of medicinal importance. The
fascinating bioactivities of these compounds arise from C19
and C20 diterpene alkaloids, such as aconitine, hetisine, atisine,
and kobusine.[1] Isolation of non-alkaloidal diterpenes from
the genus Aconitum, however, has rarely been reported.[2] In
2009, Wang and co-workers reported the isolation of the
structurally unique non-alkaloidal diterpene atropurpuran
(1)[3] from Aconitum hemsleyanum var. atropurpureum
(Scheme 1). The structure of atropurpuran features an
C-19 (i.e., by forming a 1-pyrroline ring) and the hydroxyl-
ation of the C-1,10 double bond.
Despite the intriguing biosynthetic and structural proper-
ties of the tetracyclo[5.3.3.04,9.04,12]tridecane skeleton, there
are no previous reports that describe efforts to synthesize this
compound. Consequently, at the outset of our synthetic
investigation on 1, we attempted to establish a methodology
for the construction of the tetracyclic skeleton. Herein, we
report the first entry to a tetracyclo[5.3.3.04,9.04,12]tridecane
skeleton and the pentacyclic carbon framework of 1 through
an intramolecular Diels–Alder reaction of masked ortho-
benzoquinone (MOB).
Our strategy towards the synthesis of the tetracyclic
skeleton A was to utilize a reverse-electron-demand Diels–
Alder (REDDA) reaction of MOB (Scheme 2). The REDDA
reaction with MOB has recently emerged as a powerful tool
Scheme 1. Structure of atropurpuran and related compounds.
unprecedented cage-like skeleton that consists of five six-
membered rings (A, B, C, D, and E rings). Intriguingly, the B,
C, D, and E rings constitute the tetracyclo[5.3.3.04,9.04,12]-
tridecane skeleton, which includes two bicyclo[2.2.2]octane
units. This unusual cage-like structural motif is only found in a
few members of the diterpene alkaloids, namely arcutin (2),
arcutinine (3), and arcutinidine (4), which were isolated from
the roots of Aconitum arcuatum.[4] Compounds 2–4 are closely
related to atropurpuran and only differ in the substitution at
Scheme 2. Synthetic strategy toward the tetracyclic framework.
for the construction of highly functionalized complex mole-
cules.[5,6] We envisaged that the intramolecular REDDA
reaction of MOB C would directly provide the requisite
tetracyclic framework A via a transition state B to give an
anti-Bredt compound. The ketoacetal group at C-15,16[7] in A
could serve as a potential precursor for introducing requisite
functional groups in 1. Based on steric and electronic
considerations, the intramolecular REDDA reaction seems
to be highly dependent on the substituent at C-10 (X and Y).
Although the sp2-hybridized C-10 is suitable for the synthesis
of 1, we reasoned that the cycloadduct might be relatively
unstable because of a bridgehead double bond at C-9,11 with
a conjugated sp2-hybridized carbon center. Therefore, we
decided to prepare several keto- and alkoxy-substituted
precursors and investigate the feasibility of the REDDA
reaction of these precursors. MOB C is easily prepared from
tetralone D by diallylation and oxidative dearomatization
with hypervalent iodine.
[*] Dr. T. Suzuki, A. Sasaki, N. Egashira, Prof. Dr. S. Kobayashi
Faculty of Pharmaceutical Sciences
Tokyo University of Science
2641 Yamazaki, Noda-shi, Chiba 278-8510 (Japan)
E-mail: kobayash@rs.noda.tus.ac.jp
Homepage: http://www.rs.noda.tus.ac.jp/kobalab/index.html
[**] This research was supported in part by a Grant-in-Aid for Young
Scientists (B, KAKENHI no. 22790022) from the Japan Society for
the Promotion of Science. We thank Prof. K. Miyamura and K. Ueji
(Department of Chemistry, Tokyo University of Science) for
assistance with X-ray analysis.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201103950.
Angew. Chem. Int. Ed. 2011, 50, 9177 –9179
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9177

Communications
The preparation of the REDDA precursor commenced
with the protection of ortho-eugenol 5 with a mesyl group
(Scheme 3). According to the reported procedure for 5,6-
Scheme 4. Preparation of alkoxy-substituted precursor. a) DIBAL,
CH2Cl2, ꢀ788C, 91%; b) Et3SiH, TFA, THF, 08C!RT, 70%; c) H2SO4,
MeOH, 89%; d) TESCl, imidazole, DMAP, CH2Cl2; e) 1n NaOH, THF,
73% (over 2 steps); f) PIDA, MeOH, 08C, respective yields: 15a
(66%), 15b (80%), 15c (quantitative), 15d (92%). DIBAL=diisobuty-
laluminum hydride, DMAP=N,N-4-dimethylaminopyridine, TES=
triethylsilyl, TFA=trifluoroacetic acid.
Table 1: REDDA reaction with alkoxy-substituted MOB.
Scheme 3. Attempts to prepare keto-substituted MOB. a) MsCl, TEA,
CH2Cl2, 08C, 93%; b) O3, CH2Cl2, ꢀ788C, then Me2S, 08C, 72%;
c) triethyl phosphonoacetate, NaH, THF, 08C, 94%; d) H2, Pd/C, 92%;
e) polyphosphoric acid, 80 8C, 85% (95% brsm); f) 1n NaOH, 1,4-
dioxane, reflux; g) BnBr, K2CO3, DMF, 95% (over 2 steps); h) allyl
bromide, NaH, THF/HMPA, 08C!RT, 80%; i) BCl3, CH2Cl2, ꢀ788C,
99%; j) PIDA, MeOH, 08C. Bn=benzyl, brsm=based on recovered
starting material, DMF=N,N-dimethylformamide, HMPA=hexame-
thylphosphoric triamide, Ms=methanesulfonyl, PIDA=(diacetoxy-
iodo)benzene, TEA=triethylamine, THF=tetrahydrofuran.
Entry
Substrate
X
T [8C]
Yield [%]
Ratio (16/17)[a]
[b]
1
2
3
4
5
6
15a
15b
15c
15d
15d
15d
H
OH
OMe
OTES
OTES
OTES
180
180
180
180
150
200
–
–
3:2
5:1
>20:1
>20:1
10:1
36
74
85
62
83
dimethoxy-1-tetralone,[8] mesylate 6 was converted to 8 in
59% overall yield. Because tetralone 8 was not suitable for
C,C-diallylation, the mesyl group was replaced with a benzyl
group in two steps.[9] Diallylation of the benzyl derivative 9
(allyl bromide, NaH) was successful and gave ketone 10 in
80% yield. Removal of the benzyl group with BCl3 afforded
phenol 11 in quantitative yield. Dearomatization of phenol 11
with PhI(OAc)2 in MeOH generated MOB 12, which turned
out to be very labile toward spontaneous dimerization.[10] The
dimer 13 was heated to 2208C in mesitylene with the
expectation of initiating a retro-Diels–Alder/intramolecular
Diels–Alder process,[11] but the desired product was not
observed.
We postulated that the failure of the REDDA reaction of
MOB 12 was a result of an extremely reactive conjugated
enedione structure and/or an unfavorable conformation for
the REDDA reaction because of the presence of the sp2-
hybridized C-10. Thus, several MOBs that include the sp3-
hybridized C-10 were prepared (Scheme 4). Reduction of
ketone 11 with DIBAL gave benzyl alcohol 14b. Treatment of
14b with triethylsilane or methanol under acidic conditions
afforded 14a or 14c, respectively. Silyl etherification of 14b
and selective deprotection with aqueous NaOH gave TES
ether 14d. Compounds 14a–d were subjected to oxidative
dearomatization with PhI(OAc)2 in MeOH to afford
REDDA precursors 15a–d. As expected, we observed no
dimerization of these MOBs during the oxidation reaction
and purification.[12]
[a] The ratio was determined by 1H NMR spectroscopy. [b] Decomposi-
tion occurred.
mesitylene in a sealed tube (1808C, 1 h; Table 1, entry 1)
resulted in decomposition of the starting material. When
hydroxy MOB 15b was heated (Table 1, entry 2), the dimer
was obtained as the major product along with a small amount
of 16b and 17b (36%, 3:2). However, the desired cycloadduct
16c was obtained in 74% yield by heating methyl ether 15c
(ca. 5:1 ratio of 16c to 17c; Table 1, entry 3). Finally, we found
that the REDDA reaction of silyl ether 15d (1808C; Table 1,
entry 4) afforded tetracyclic 16d in high yield and almost as a
single diastereomer. We reasoned that the steric repulsion
between the bulky TESO group and the syn allyl group might
force the anti allyl group to adopt a favorable conformation
for the desired intramolecular reaction.
The structure of tetracyclic 16d was determined by NMR
spectroscopy.[13] Removal of the TES group of 16d (TBAF,
THF) afforded crystalline 16b. Subsequent X-ray crystallo-
graphic analysis of 16b (Figure 1)[14] confirmed the anti-Bredt
tetracyclic skeleton.[15] This result established that we had
achieved the first artificial synthesis of the tetracy-
clo[5.3.3.04,9.04,12]tridecane skeleton.
We subsequently proceeded to the construction of the
pentacyclic skeleton of 1, despite the lack of functionality at
C-6 (Scheme 5). Addition of allylmagnesium bromide to
ketone 11 gave triene 18 in excellent yield. Oxidative
With precursors 15a–d in hand, the REDDA reaction was
carried out (Table 1). Heating of deoxy MOB 15a in
9178
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 9177 –9179

Bredt and cage-like complex molecules. Use of this method-
ology in synthetic efforts toward atropurpuran is currently
underway.
Received: June 10, 2011
Published online: August 24, 2011
Keywords: benzoquinones · Diels–Alder reaction · pentacycles ·
.
synthetic methods · terpenoids
Figure 1. ORTEP drawing of 16b (ellipsoids drawn at 50% probability).
[1] F.-P. Wang, Q.-H. Chen, X.-Y. Liu, Nat. Prod. Rep. 2010, 27, 529 –
570.
[2] To the best of our knowledge, only two examples have been
reported, see: a) S. W. Pelletier, A. M. Ateya, J. Finer-Moore,
N. V. Mody, L. C. Schramm, J. Nat. Prod. 1982, 45, 779 – 781;
b) C.-H. Yang, X.-C. Wang, Q.-F. Tang, W.-Y. Liu, J.-H. Liu,
Helv. Chim. Acta 2008, 91, 759 – 765.
[3] P. Tang, Q.-H. Chen, F.-P. Wang, Tetrahedron Lett. 2009, 50, 460 –
462.
[4] a) B. Tashkhodzhaev, S. A. Saidkhodzhaeva, I. A. Bessonova,
M. Y. Antipin, Chem. Nat. Compd. 2000, 36, 79 – 83; b) S. A.
Saidkhodzhaeva, I. A. Bessonova, N. D. Abdullaev, Chem. Nat.
Compd. 2001, 37, 466 – 469.
[5] For selected reviews, see: a) C.-C. Liao, R. K. Peddinti, Acc.
Chem. Res. 2002, 35, 856 – 866; b) D. Magdziak, S. J. Meek,
T. R. R. Pettus, Chem. Rev. 2004, 104, 1383 – 1429; c) C.-C. Liao,
Pure Appl. Chem. 2005, 77, 1221 – 1234; d) S. P. Roche, J. A.
Porco Jr., Angew. Chem. 2011, 123, 4154 – 4179; Angew. Chem.
Int. Ed. 2011, 50, 4068 – 4093.
[6] For an example of our study, see: T. Suzuki, S. Kobayashi, Org.
Lett. 2010, 12, 2920 – 2923.
Scheme 5. Construction of the pentacyclic skeleton of atropurpuran.
a) AllylMgBr, THF, 08C, 96%; b) PIDA, MeOH, 08C, 89%; c) mesity-
lene, 1808C, 1 h, 79%; d) Grubbs’ II catalyst, CH2Cl2, 99%; e) Grubbs’
II catalyst, CH2Cl2, 56%; f) PIDA, MeOH, 08C, 78%; g) mesitylene,
1808C, 1 h, 75%; e) H2, Pd(OH)2/C, THF, 78%; f) Tf2O, pyridine,
CH2Cl2, 85% (99% brsm). Tf=trifluoromethanesulfonyl.
[7] The atropurpuran numbering system is used in this manuscript.
[8] S. Lahiri, C. Ramarao, B. Venkateswara Rao, A. V. Rama Rao,
M. S. Chorghade, Org. Process Res. Dev. 1999, 3, 71 – 72.
[9] Mesyl protection is quite important for the effective conversion
to tetralone 9. The overall yields from benzyl ether or o-eugenol
itself were 31% and 20%, respectively.
1
[10] Formation of dimer 13 was confirmed by LRMS and H NMR
spectroscopy. However, the stereochemistry of dimer 13 was not
confirmed.
[11] S. K. Chittimalla, H.-Y. Shiao, C.-C. Liao, Org. Biomol. Chem.
2006, 4, 2267 – 2277.
[12] These MOBs gradually dimerized under storage in a freezer
(ꢀ258C). Hence, MOBs were used for the REDDA reaction
directly after column chromatography.
dearomatization of 18 with PIDA followed by the REDDA
reaction of the resulting MOB afforded tetracyclic adduct 19
as a single diastereomer. Treatment of 19 with Grubbsꢀ
second-generation catalyst provided the desired pentacyclic
skeleton 21 in excellent yield. Alternatively, triene 18 was
treated with Grubbsꢀ second-generation catalyst to give cis-
hydrophenanthrene 20 as a major product.[16] Oxidation of 20
with PIDA and the REDDA reaction of the tricyclic MOB
was achieved to construct the same pentacyclic compound 21.
Finally, hydrogenation of 21 (H2, Pd(OH)2/C) and subsequent
dehydration with Tf2O and pyridine afforded 22 to complete
the construction of the pentacyclic framework of 1.
In summary, we have achieved the construction of the
pentacyclic framework of atropurpuran by an intramolecular
REDDA reaction. Characteristic features of the present study
are: 1) the first synthesis of a tetracyclo[5.3.3.04,9.04,12]-
tridecane skeleton and 2) a concise approach to the atropur-
puran skeleton. These results demonstrate the power of the
REDDA reaction by using MOB for the construction of anti-
[13] See the Supporting Information.
[14] CCDC 827527 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.
cam.ac.uk/data_request/cif.
ꢀ
ꢀ
ꢀ
[15] A torsion angle of 146.38 for the C10 C9 C11 C12 bonds is
consistent with our previous observation on an E-olefin that
contains an eight-membered ring, see: R. Matsui, K. Seto, K.
Fujita, T. Suzuki, A. Nakazaki, S. Kobayashi, Angew. Chem.
2010, 122, 10266 – 10271; Angew. Chem. Int. Ed. 2010, 49, 10068 –
10073.
[16] A spirocyclic compound was obtained as a minor product
(29%).
Angew. Chem. Int. Ed. 2011, 50, 9177 –9179
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
9179