Synthetic studies on (+)-bucidarasin C: Two diastereoselective transannular reactions producing cis-decaline derivatives that show reversal selectivity

Article abstract of DOI:10.1016/j.tetlet.2012.01.058

This manuscript describes studies on the construction of the cis-decaline core of (+)-bucidarasin C. The transannular Diels-Alder (TADA) reaction of a macrocyclic lactone successfully afforded the desired cis-decaline derivative in a stereoselective manner. On the other hand, the stereoselective transannular cascade Michael (TACM) reaction of the parent macrocyclic lactone afforded the diastereomeric cis-decaline derivative as the major product.

Full text of DOI:10.1016/j.tetlet.2012.01.058

Tetrahedron Letters 53 (2012) 1518–1522
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Synthetic studies on (+)-bucidarasin C: two diastereoselective transannular
reactions producing cis-decaline derivatives that show reversal selectivity
⇑
Akinobu Nakahara, Misaki Kanbe, Masahisa Nakada
Department of Chemistry and Biochemistry, School of Advanced Science and Engineering, Waseda University, 3-4-1 Ohkubo, Shinjuku-ku, Tokyo 169-8555, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
This manuscript describes studies on the construction of the cis-decaline core of (+)-bucidarasin C. The
transannular Diels–Alder (TADA) reaction of a macrocyclic lactone successfully afforded the desired
cis-decaline derivative in a stereoselective manner. On the other hand, the stereoselective transannular
cascade Michael (TACM) reaction of the parent macrocyclic lactone afforded the diastereomeric cis-deca-
line derivative as the major product.
Received 22 December 2011
Revised 13 January 2012
Accepted 16 January 2012
Available online 24 January 2012
Ó 2012 Elsevier Ltd. All rights reserved.
Bucidarasins (Fig. 1) were isolated in 2001 from an extract of
Bucida buceras, which is a widely spreading timber and shade ever-
green tree found in tropical regions of northern South America.1
Although (+)-bucidarasin D is inactive, (+)-bucidarasins A–C show
significant in vitro cytotoxicity. The IC50 values of (+)-bucidarasins
Scheme 1 shows our retrosynthetic analysis of (+)-bucidarasins
C. We planned to construct the tetrahydrofuran moiety incorporat-
ing the acid- and base-sensitive bis-acetal at the late stage of the
total synthesis and set compound 1 as the late stage intermediate,
which was thought to be obtained from compound 2 via decarbox-
ylation and stereoselective reduction of the C2 ketone. The con-
struction of a stereogenic quaternary all-carbon center is one of
the most difficult problems in natural product synthesis. (+)-Bucid-
arasin C possesses two stereogenic quaternary all-carbon centers
at the C5 and C9 positions in the four contiguous stereogenic car-
bon centers, thus presenting a synthetic challenge.
We decided to employ the transannular Diels–Alder (TADA)
reaction of 3 to synthesize 2 because it could generate the requisite
cis-fused decaline core with four contiguous stereogenic centers
that include the C5 stereogenic quaternary all-carbon center.3
Compound 3 was expected to be prepared via the macrolactone
formation of 4, which would be prepared from 5. Compound 5
would be obtained by the oxidative ring-opening reaction of 6,
which was expected to be derived from 7. Finally, compound 7
was envisioned to be prepared from 8, a chiral building block
A–C range from 0.5 to 1.9 lM against various human tumor cell
lines such as KB (nasopharyngeal), A549 (lung), IA9 (ovarian), CAKI
(kidney), HCT-8 (ileocecal), MCF-7 (breast), HOS (bone), U87-MG
(glioblastoma), and SK-MEL-2 (melanoma). Moreover, their
potency of cytotoxicity is retained in the drug-resistant tumor cell
lines such as KB-VIN (vincristine resistant), KB-7d (etoposide resis-
tant), KB-CPT (camptothecin resistant), and IA9-PTX10 (paclitaxel
resistant), suggesting that they may have a novel mode of action
and could become clinical candidates. As (+)-bucidaracin D has
no cyctotoxicity, the bis-acetal moiety of (+)-bucidarasins A–C
plays a key role in producing the cytotoxicity; however, their mode
of action has not been reported so far.
Bucidarasins are a family of cis-clerodane diterpenoids, possess-
ing a unique tricyclic ring system incorporating a cis-dehydrodec-
alin core with a fused tetrahydrofuran. The cis-dehydrodecalin core
has two methyl groups trans to each other at the C8 and C9 posi-
tions, and an (E)-3-methyl-penta-2,4-dienyl group at the C9 posi-
tion. The C2 position of bucidarasins is oxygenated, and the C6
position is further oxygenated in (+)-bucidarasins A and B. (+)-
Bucidarasin C possesses four successive stereogenic centers on
the cis-dehydrodecalin core, an isobutyrate at the C2 position,
and two acetates as acetals on the tetrahydrofuran ring; hence,
(+)-bucidarasin C has seven stereogenic centers in total. The total
synthesis of (+)-bucidarasin C has not been reported so far.2 There-
fore, its potent bioactivity as well as unique structural features led
us to commence synthetic studies on (+)-bucidarasins C.
⇑
Corresponding author.
E-mail address: mnakada@waseda.jp (M. Nakada).
Figure 1. Structures of (+)-bucirarasin A–D.
0040-4039/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2012.01.058

A. Nakahara et al. / Tetrahedron Letters 53 (2012) 1518–1522
1519
Scheme 1. Retrosynthetic analysis of (+)-bucidarasin C.
Next, 5 was converted into seco-acid 4 to achieve macrocycliza-
tion (Scheme 4). The reaction of 5 with Ohira–Bestman’s reagent,8
subsequent reaction with LiAlH4, and then, Swern oxidation affor-
ded aldehyde 11. The Horner–Wadsworth–Emmons reaction of 11
afforded the corresponding enal, which was converted into 1,3-
dioxane 12. The terminal alkyne of 12 was formylated, and the
subsequent one-pot reduction of the aldehyde and alkyne with
Red-Al, followed by treatment with iodine, afforded the allylic
alcohol. The allylic alcohol was protected as t-butyldiphenylsilyl
(TBDPS) ether, followed by the palladium-mediated carbonylation
to afford methyl ester 13.
Scheme 2. Stereoselective synthesis of 7 from 8.
The diisobutylaluminum hydride (DIBAL-H) reduction of 13 and
the subsequent deprotection of the 1,3-dioxane, and the protection
of the allylic alcohol as methoxymethyl (MOM) ether gave 14. The
aldol reaction of 14 with methyl acetate and subsequent protection
of the resultant secondary hydroxyl as MOM ether, and deprotec-
tion of the TBDPS group with HF-pyridine, afforded a methyl ester
of 4.
prepared by us,4 by the Wittig reaction and stereoselective
hydrogenation.
The Wittig reaction of 8 afforded 9 (Scheme 2), but the epimer-
ization at the C9 position of 8 occurred after a prolonged reaction
time because the concomitant retro-aldol reaction of 8 occurred.5
Therefore, the Wittig reaction was stopped at an 88% conversion
to avoid epimerization. The hydroxyl-directed hydrogenation of 9
with the Crabtree’s catalyst ([Ir(cod)(Py)(PCy3)]PF6)6 stereoselec-
tively afforded the desired compound 7. Because the diol, which
was obtained by the hydrogenation of 7, was a known compound,
its structure was confirmed by a comparison of NMR data.7
Swern oxidation of 7 afforded 10 (Scheme 3), followed by
trimethylsilyl enol ether formation and Rubottom oxidation to give
6 quantitatively, which was treated with a lead tetraacetate in the
presence of methanol to afford methyl ester 5.
Hydrolysis of the methyl ester under commonly used condi-
tions was difficult to achieve, causing b-elimination to afford a,b-
unsaturated ester as the major product. However, this difficulty
was solved using the enzyme-mediated hydrolysis. Thus, the
methyl ester was successfully hydrolyzed with pig liver esterase
to afford 4 in KPB8 buffer using dimethyl sulfoxide (DMSO) as
the additive.9
With seco-acid 4 in hand, the macrolactonization of 4 was
examined. Gradual addition of a solution of 4 in CH2Cl2 using a syr-
inge pump to a solution of 2-methyl-6-nitrobenzoic anhydride
(MNBA)10 and 4-N,N-dimethylaminopyridine (DMAP) in CH2Cl2
under high dilution conditions at room temperature afforded mac-
rolactone 15. The major by-product was the dimer 16, which was
formed in an 11% yield.
The deprotection of MOM ethers of 15 under acidic conditions
(Scheme 5) gave diol 17, which was subjected to Dess–Martin oxi-
dation to afford keto-aldehyde 18. The Pinnick oxidation of 18, and
subsequent reaction with diazomethane afforded methyl ester 19.
Treatment of 19 with triisopropylsilyltrifluoromethane sulfonate
(TIPSOTf) and triethylamine afforded triisopropylsilyl (TIPS) ether
3, which was subjected to the TADA reaction. The TADA reaction
of 3 did not occur in the presence of Lewis acid at low temperature
and decomposition of the substrate occurred at room temperature.
Hence, the TADA reaction of 3 was carried out at the reflux temper-
ature of xylene to produce a mixture of three products, which was
then treated with tetrabutylammonium fluoride (TBAF) to afford
20a (56%, 2 steps), 20b (11%, 2 steps), and 20c (6%, 2 steps).
Analysis by nuclear Overhauser effect (NOE) experiments of the
products (Scheme 6) suggested that the structures of 20a and 20b
Scheme 3. Transformation of 7 to 5.

1520
A. Nakahara et al. / Tetrahedron Letters 53 (2012) 1518–1522
Scheme 4. Preparation of macrolactone 15.
Scheme 5. Preparation of 3 and the TADA reaction.
were as shown in Scheme 5. Compound 20c was found to be the
epimer of 20b because the hydrogenation of a mixture of 20b
and 20c (Scheme 6) afforded a separable mixture of 21b and 22,
and the structure of 22 was suggested to be as shown in Scheme
6 by NOE analysis. The major product 20a had the desired config-
with TS-3a. Consequently, the desired product 20a would be
formed as the major product. The TADA reaction of 3 was surmised
to afford products with the cis-fused c-lactone as shown in Scheme
7. Hence, 20c was thought to be kinetically formed by the proton-
ation of the enolate generated from the thermodynamically more
stable 20b during the reaction with basic TBAF.
uration, indicating that the TADA reaction of
3 successfully
constructed the cis-decaline core of (+)-busidaracins in a stereose-
lective manner.
On the other hand, compound 19, which is a parent compound
of 3, possessed a b-keto ester, an a,b-unsaturated ester, and an a,b-
In the TADA reaction of 3, the exo-addition would be favored
because the endo-addition is energetically unfavored due to the
severe strains arising from the mode of transannular reaction.
Indeed, only exo-adducts were formed by the TADA reaction of 3
(Scheme 5). Two transition states including a chair-structure,
TS-3a, and TS-3b are surmised to be energetically favored for the
TADA reaction of 3, but TS-3b would be unfavorable due to the
steric strains derived from 1,3-diaxial interactions when compared
unsaturated ketone at the positions suitable for the transannular
cascade Michael (TACM) reaction. Hence, the treatment of 19 with
an appropriate base was expected to produce 20a–c. Therefore, the
reactions of 19 under basic conditions were examined, and we
found that the reaction of 19 with lithium diisopropylamide
(LDA) under optimized conditions shown in Scheme 8 afforded
20a (9%), 20b (52%), and 20c (14%). Interestingly, the major prod-
uct was 20b and the diastereoselectivity was 9:66 (1:7.3) because

A. Nakahara et al. / Tetrahedron Letters 53 (2012) 1518–1522
1521
Scheme 6. Conversion of 20b and 20c to 21b and 22, and NOE correlations in 20a, 20b, and 22.
Scheme 7. Plausible transition states of the TADA reaction of 3 (R = TIPS).
Scheme 8. The transannnular cascade Michael (TACM) reaction of 19.
20c is the epimer of 20b. On the other hand, the diastereoselectiv-
ity of the TADA reaction of 3 was 56:17 (3.3:1); that is, the diaste-
reoselectivity of the two reactions was reversed.
The reversal of the diastereoselectivity is surmised to arise from
the different reaction mechanism between the TADA reaction and
the TACM reaction. TADA reactions require s-cis conformation of
the diene for the concerted reaction with the dienophile. On the
other hand, the TACM reaction of 19 proceeded by step-wise C–C
bond formations. Thus, the first intramolecular Michael reaction
In summary, the cis-decaline derivative, which would be a key
intermediate for the total synthesis of (+)-bucidarasin C, was suc-
cessfully prepared by the stereoselective TADA reaction of a macto-
cyclic lactone. Moreover, the stereoselective TACM reaction of the
parent macrocyclic lactone was found to afford the diastereomeric
cis-decaline derivative as the major product. These two stereose-
lective transannular reactions that show reversal diastereoselectiv-
ity would be useful for the synthesis of natural products that
include a cis-decaline core.
of 19 generated the c-butyrolactone and the subsequent intramo-
lecular Michael reaction constructed the tricyclic carbon skeleton.
Therefore, the first C–C bond-forming reaction did not require s-cis
conformation of the dienolate that formed from 19. That is, the
dienolate could take the s-trans conformation to undergo the first
reaction. The difference described above suggests that the diastere-
oselectivity-determining transition states of the two reactions are
different, thereby leading to the observed reversal.
Acknowledgments
This work was financially supported in part by a Waseda Uni-
versity Grant for Special Research Projects, a Grant-in-Aid for Sci-
entific Research on Innovative Areas (No. 21106009) and the
Global COE program ‘‘Center for Practical Chemical Wisdom’’ by
MEXT.

1522
A. Nakahara et al. / Tetrahedron Letters 53 (2012) 1518–1522
5. Other olefination methods resulted in decomposition of 8.
References and notes
6. (a) Crabtree, R. H.; Morris, G. E. J. Organomet. Chem. 1977, 135, 395–403; (b)
Crabtree, R. H.; Davis, M. W. J. Org. Chem. 1986, 51, 2655–2661.
7. Herradon, B.; Seebach, D. Helv. Chim. Acta 1989, 72, 690–714.
8. (a) Ohira, S. Synth. Commun. 1989, 19, 561–564; (b) Müller, S.; Liepold, B.; Roth,
G. J.; Bestmann, H. J. Synlett 1996, 521–522.
1. Hayashi, K.-I.; Nakanishi, Y.; Bastow, K. F.; Cragg, G.; Nozaki, H.; Lee, K.-H.
Bioorg. Med. Chem. Lett. 2002, 12, 345–348.
2. Ida, R.; Nakada, M. Tetrahedron Lett. 2007, 48, 4855–4859.
3. (a) Suzuki, T.; Usui, K.; Miyake, Y.; Namikoshi, M.; Nakada, M. Org. Lett. 2004, 6,
553–556; (b) Hayashi, N.; Nakada, M. Tetrahedron Lett. 2009, 50, 232–235; (c)
Hayashi, N.; Suzuki, T.; Usui, K.; Nakada, M. Tetrahedron 2009, 65, 888–895.
4. (a) Iwamoto, M.; Miyano, M.; Utsugi, M.; Kawada, H.; Nakada, M. Tetrahedron
Lett. 2004, 45, 8647–8651; (b) Watanabe, H.; Iwamoto, M.; Nakada, M. J. Org.
Chem. 2005, 70, 4652–4658.
9. Our use of PLE in natural product synthesis, see: (a) Noguchi, N.; Nakada, M.
Org. Lett 2006, 8, 2039–2042; (b) Asakawa, K.; Noguchi, N.; Nakada, M.
Heterocycles 2008, 76, 183–190; (c) Asakawa, K.; Noguchi, N.; Takashima, S.;
Nakada, M. Tetrahedron: Asymmetry 2008, 19, 2304–2309.
10. Shiina, I.; Kubota, M.; Ibuka, R. Tetrahedron Lett. 2002, 43, 7535–7539.