Angewandte
Communications
Chemie
[
20]
double-bond configuration. Our attempts to construct the
trisubstituted alkene in 35 directly by the Wadsworth–
Horner–Emmons reaction afforded an inseparable 1:1 Z/E
with MOMCl and Stille cross-coupling with the tin reagent 44
afforded 45. Final deprotection and base-promoted lactoni-
zation produced the dienophile 9, which is labile and was thus
used directly in the next step without purification.
[
21]
mixture. Oxidative elaboration of furan followed by acid-
mediated rearrangement delivered 36. We then attempted
to set up a radical cascade with 36 using different metal
catalysts and phenylsilane as hydrogen source. No cycliza-
tion product was detected with Co(acac) or Mn(dpm) as the
catalyst in the presence of oxygen; this likely reflects the
failure of radical initiation at the gem-disubstituted alkene.
To our delight, HAT-triggered radical cascade polycyclization
in the Fe-catalyzed conditions and subsequent silyl depro-
tection generated the tricyclic compound 37 with proper
stereochemistry in 45% isolated yield over two steps. It also
generated an inseparable mixture of another three diastereo-
mers in 19% combined yield over two steps. This cascade
simultaneously afforded two carbocycles and four contiguous
stereogenic centers. Then the secondary alcohol in 37 was
[
22]
Diels–Alder cycloaddition between 5 and 9 produced 8
and its diastereomer as endo-cycloadducts in 81% combined
yield in a 10:1 ratio. Although Diels–Alder cycloaddition has
been well applied in organic synthesis and natural product
[
23]
2
3
[
32]
synthesis, this cycloaddition proceeded surprisingly well at
room temperature. Such a spontaneous process is uncom-
[24]
[
33]
mon, suggesting that this reaction may participate in the
real biosynthesis of hispidanin A through a nonenzymatic
pathway, even though the biosynthetic hypothesis from the
[
25]
[
1]
Jiang group does not involve cycloaddition between 5 and 9.
Diastereoselective reduction of ketone and then acid-pro-
moted acetylation completed the total synthesis of hispida-
nin A (1), for which characterization data matched those of
[
26]
[1]
a natural sample.
[
27]
removed by a radical process, affording 38. Partial reduc-
tion of the lactone and subsequent elimination generated 39.
Oxidative cleavage of the double bond and base-promoted
elimination delivered the enone 40. Michael addition of
compound 11 and pyrolysis of the resulting intermediate
released one molecule of SO2 and generated 41. Finally,
nucleophilic addition of methyl lithium to the ketone,
followed by lactone formation, afforded the diene 5.
In conclusion, we have accomplished the asymmetric total
synthesis of the dimeric diterpenoid hispidanin A. The syn-
thesis involves 1) stereoselective construction of Z-alkene in
35 via Tanabeꢀs iron catalysis; 2) HAT-triggered and iron-
catalyzed radical polyene cyclization to generate the trans-
decalin architecture of the natural diterpenoid 5; 3) Yama-
motoꢀs cationic polyene cyclization to construct the basic
skeleton of the dienophile fragment; 4) site-selective bromi-
nation with TBCO and Stille cross-coupling to afford the
dienophile 9; and 5) a final step of bioinspired Diels–Alder
connection of the two diterpenoid monomers. The radical
polycyclization involving electron-deficient alkenes described
herein can likely be extended to total synthesis of other
diterpenoids functionalized at position C20.
We then began synthesizing the dienophile 9 from
[28]
compound 22 (Scheme 4), which was prepared in 90% ee
using Yamamotoꢀs enantioselective polyprenoid cycliza-
[
29]
[30]
tion.
Benzylic oxidation of 22
and basic hydrolysis
afforded 42. Site-selective bromination with 2,4,4,6-tetrabro-
mocyclohexa-2,5-dienone (TBCO) could be smoothly ach-
[31]
ieved to afford 43, once the reaction was timely quenched
to prevent formation of dibrominated product and other
byproducts. Other brominating reagents, such as NBS, Acknowledgements
bromine and TBAB, led to either complex mixture or an
unfavorable product brominated at the para-position of
We acknowledge financial support from the NSFC (21672153
and 21290180) and the Open Fund of State Key Laboratory of
Natural Medicines in China Pharmaceutical University
[
26b]
ketone on phenyl ring.
Subsequent protection of phenol
(SKLNMKF201601).
Conflict of interest
The authors declare no conflict of interest.
Keywords: diterpenoids · labdane · radical cascade ·
total synthesis · totarane
[
[
[
[
4] For recent examples, see: a) H.-Y. Jiang, W.-G. Wang, M. Zhou,
H.-Y. Wu, R. Zhan, X.-N. Li, X. Du, Y. Li, J.-X. Pu, H.-D. Sun,
Scheme 4. Synthesis of 9 and hispidanin A.
Angew. Chem. Int. Ed. 2017, 56, 1 – 5
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3
These are not the final page numbers!