Angewandte
Communications
Chemie
unexpected cyclization byproduct 18 through an intramolec-
structure of 25 was verified by X-ray crystallographic
analysis.[21]
Having successfully constructed the tetracyclic framework
ular arylation reaction[19] and a subsequent Wittig reaction,
while the methylenation with Peterson reagent, Tebbe
reagent, or Julia reagent did not give any of the desired
products. Much to our delight, the methylenation problem
was solved by employing Nysted reagent[20] with TiCl4 as
Lewis acid, rendering the desired product terminal alkene 19
in 31% yield and recovering 54% of starting material ketone
14. Subjection of intermediate 19 to heterogenous hydro-
genation conditions (10% Pd/C) resulted in the migration of
for dysideanone B, we proceeded to the final stage of the total
synthesis of dysideanone B. Thus, as depicted in Scheme 2,
methylenation of ketone 24 proceeded smoothly (87% yield)
with Wittig reagent to afford terminal alkene 11, whose
methyl protected hydroquinone moiety was oxidized using
AgO and HNO3, delivering quinone 26 in 86% yield and
setting the stage for the introduction of the ethoxy group. A
series of ethoxy group introduction reactions were performed.
Metal mediated reactions (e.g., EtOH with Co(OAc)2, Mn-
(OAc)3·2H2O, Ni(OAc)3·4H2O, or AgOAc)[22] led to either
low yields or decomposition of starting material. Finally, we
=
the exocyclic C C bond to the C7 and C8 position to give an
=
endocyclic C C bond byproduct, which could not be selec-
tively reduced without affecting the other olefin moiety. On
the other hand, homogenous hydrogenation of intermediate
19 with (Ph3P)3RhCl only gave 15% of desired product 20 and
76% of its C8-epimer 21 (d.r. = 1:5). Delightedly, the
diastereoselectivity reversed when the more advanced inter-
mediate ketoalkene 22 was subjected to the same reaction
conditions, affording the desired product 12 in 84% yield and
its C8-epimer 23 in 11% yield (d.r. = 8:1). It is noteworthy
that the bromide at C6’ and internal alkene between C1 and
C10 survived under this condition.
With alkene bromide 12 in hand, we focused our attention
on the investigation of cyclization reaction for the construc-
tion of the 6/6/6/6-tetracycle of dysideanone B, as shown in
Scheme 2. We envisioned that a C6’ radical, generated from
the parent aryl bromide, would attack the alkene from C1 to
give the more stable tertiary radical on C10, which then reacts
with a HC donor to form the 6/6/6/6-fused tetracyclic core of
dysideanone B. Thus, alkene bromide 12 was subjected to
radical reaction conditions (nBu3SnH and AIBN, Scheme 2).
The desired 6/6/6/6-fused tetracycles 24 was obtained in 61%
yield and a tetrasubstituted alkene byproduct 25 was isolated
in 23% yield, presumably because the tertiary radical under-
goes b-H elimination from C1 to give internal alkene 25. The
[23]
were pleased to find that Et3N in EtOH under O2 proved
effective for the introduction of ethoxy functionality, afford-
ing dysideanone B (1) in 53% yield and its regioisomer 27 in
36% yield. Driven by our suspicion that the ethoxy group
might come from the solvent EtOH, we opted to mimic the
isolation conditions. In this regard, quinone 26 was heated in
95% EtOH under air and dysideanone B (1) and its
regioisomer 27 were also isolated but in lower combined
yield (64%), revealing that the ethoxy group might come
from the solvent ethanol. The spectroscopic data of synthetic
dysideanone B (1) matched with those reported for the
natural product. The transformation of dysideanone B (1) to
its congener “dysideanone F” (9) was achieved in 92% yield
under acidic condition. The synthesis and characterization of
this hitherto unknown congener might facilitate its isolation
from natural sources in the future.
Having completed the first total synthesis of dysideano-
ne B (1), we turned our attention to the formation of the 6/6/5/
6-fused backbone of dysiherbol A, as depicted in Table 1.
However, the construction of the five-membered ring of
dysiherbol A (2) turned out to be a difficult task. Various
Heck reaction conditions[24] (different Pd catalysts such as
Pd2(dba)3, Pd(OAc)2, and Pd(dppf)Cl2, ligands such as SPhos,
P(o-tol)3, P(2-furyl)3, and BINAP, bases such as DBU, Et3N,
PMP, K2CO3, and NaHCO3) were explored on substrate 20,
but no desired product was observed (Table 1, entry 1). The
C8-epimer 21 was also investigated for the cyclization but did
not give any desired cyclization product (Table 1, entry 2). We
also examined the cyclization reaction on 14, whose ketone
functionality could be transformed to methyl group on C8 at
a late stage. Arylation byproduct 28 was obtained in 81%
yield with DBU as base (Table 1, entry 3), while addition
byproduct tertiary alcohol 29 was isolated in 86% yield with
Et3N as base (Table 1, entry 4) through a Grignard-type
nucleophilic addition of arylpalladium to ketone.[25] Molec-
ular modeling showed that the methoxy group on C5’ of the
benzene ring and the acetal group on C4 of the decalin ring
would be very close if the C10–C6’ single bond was generated.
We expected that the removal of the acetal group on C4 might
reduce the hindrance between these functionalities. To this
end, a range of Heck or reductive Heck reaction[26] conditions
were examined on ketoalkene bromide 12. However, the
desired cyclization product was also not observed. Indeed,
ring contraction byproduct 30 was obtained in 81% yield
when Pd2(dba)3 was used as catalyst with SPhos as ligand and
Scheme 2. Total synthesis of dysideanone B and “dysideanone F”.
AIBN=2,2’-azoisobutyronitrile. Thermal ellipsoids are shown at the
50% probability level.
Angew. Chem. Int. Ed. 2021, 60, 1 – 8
ꢀ 2021 Wiley-VCH GmbH
3
These are not the final page numbers!