Total synthesis of (-)-neovibsanin G and (-)-14-epi-neovibsanin G: Towards the synthesis of 15-O-methylneovibsanin F and 14-epi-15-O-methylneovibsanin F

Article abstract of DOI:10.1002/ejoc.201101796

The total synthesis of (-)-neovibsanin G and (-)-14-epi-neovibsanin G has been accomplished, establishing the absolute configurations of these caged diterpenes as 11S, and confirming Fukuyama's postulated biosynthesis of the neovibsanins. Synthetic studie

Full text of DOI:10.1002/ejoc.201101796

FULL PAPER
DOI: 10.1002/ejoc.201101796
Total Synthesis of (–)-Neovibsanin G and (–)-14-epi-Neovibsanin G:
Towards the Synthesis of 15-O-Methylneovibsanin F and
4-epi-15-O-Methylneovibsanin F
1
[a]
[a]
Jeffrey Y. W. Mak and Craig M. Williams*
Keywords: Total synthesis / Natural products / Terpenoids / Vibsanin / Cyclization
The total synthesis of (–)-neovibsanin G and (–)-14-epi-neo- based on the hypothesis that these methoxy substituted
vibsanin G has been accomplished, establishing the absolute caged neovibsanins were artifacts of isolation, formed by the
configurations of these caged diterpenes as 11S, and con- addition of methanol to the terminal alkenes of neovibsanin
firming Fukuyama’s postulated biosynthesis of the neovib-
sanins. Synthetic studies towards 15-O-methylneovibsanin F
G and 14-epi-neovibsanin G. The results strongly suggest,
however, that the methoxy caged neovibsanins are true natu-
ral products.
and 14-epi-15-O-methylneovibsanin
F were undertaken
methylneovibsanin H (7)[4] (path C) and the ketal bearing
Introduction
[
5]
neovibsanins A (4) and B (5) (path A), a synthetic study
towards the caged neovibsanins would complement our ef-
forts in this area. In this vein, we have recently completed
the total synthesis of two caged neovibsanins, (–)-neovib-
Vibsanins are rare diterpenes that exist exclusively in a
small number of Viburnum species.[ Based on their chemi-
cal structures, these natural products can be formulated into
three subtypes; eleven membered ring compounds, seven
membered ring compounds, and the rearranged type. This
last subtype, which is more commonly known as the neovib-
sanins, can be further divided into three subclasses; the bi-
cyclic, ketal and caged neovibsanins. These are postulated
to be derived biosynthetically through an acid-catalyzed re-
arrangement of the eleven membered vibsanin B (1)
1]
[6]
sanin G [(–)-9] and (–)-14-epi-neovibsanin G [(–)-10].
Herein, we disclose the full investigation that led to the total
synthesis of these diterpenes (i.e. 9 and 10), and report our
synthetic studies towards the methoxy caged neovibsanins
(11 and 12), leading to a better understanding to the pos-
sibility of their existence as artifacts of isolation.
Since the caged neovibsanins 9–12 were isolated after
soaking the dried leaves of Viburnum awabuki in MeOH at
[
2]
(
Scheme 1). After an acid-catalyzed rearrangement to give
enone 2 and subsequent intramolecular 1,4-addition, inter-
mediate 3 is formed. Two chemical outcomes are possible;
acid-catalyzed hemiketal formation (path A) gives rise to
the ketal neovibsanins, while acid-catalyzed dehydration
[2]
room temperature for thirty days, it seemed likely that
during this prolonged extraction period, residual plant acid
could catalyze the addition of MeOH to the terminal alk-
enes of 9 and 10 to give 11 and 12. Secondly, it was con-
ceded by Fukuyama that neovibsanin A (4) and B (5) might
(path B) forms carbocation 6. Likewise, 6 can undergo one
of two possible reactions. If the carbocation is captured by
water (path C), this presumably gives rise to the bicyclic
neovibsanins. If intramolecular capture of the carbocation
occurs (path D), however, cation 8 is formed, a species that
is postulated to be the precursor of the caged neovibsanins.
Our group has explored the total synthesis of many com-
plex natural products bearing caged bicyclic moieties, of
which many are vibsanins.[ Moreover, having recently also
investigated the biomimetic synthesis of the bicyclic 2-O-
[7]
be isolation artifacts of the true natural product, hemike-
tal 13, presumably via a cation of type 14 (Scheme 2), add-
ing further to the suspicion that 11 and 12 might also be
artifacts formed through a similar process.
Nevertheless, from a synthetic point of view, the alkenes
of 9 and 10 were seen as potential precursors of 11 and 12,
whereby the introduction of the methoxy group could be
effected by the capture of a tertiary carbocation by MeOH,
regardless of whether the C-15 oxygenated caged neovib-
sanins are artifacts or not. Thus, in order to access the four
caged neovibsanin targets, 9 and 10 were logical initial tar-
gets.
3]
[
a] School of Chemistry and Molecular Biosciences, University of
Queensland,
St Lucia 4072, Australia
Fax: +61-7-3365-4299
Securing advanced intermediate enone 18 was the key-
stone of the synthetic strategy towards 9 and 10 (Scheme 3).
The racemic form of this compound was a key intermediate
E-mail: c.williams3@uq.edu.au
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201101796.
Eur. J. Org. Chem. 2012, 2001–2012
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2001

J. Y. W. Mak, C. M. Williams
FULL PAPER
Scheme 1. Proposed biosynthesis of the neovibsanins from vibsanin B (1), and the hypothesis that the methoxy caged neovibsanins (11
and 12) are artifacts of isolation.
Scheme 2. Formation of possible artifacts of isolation, neovibsan-
ins A (4) and B (5).
in the synthesis of (Ϯ)-2-O-methylneovibsanin H (7),[4] act-
ing as a synthetic equivalent of biosynthetic intermediate 2 Scheme 3. Enantioselective NHC-Cu-catalyzed 1,4-addition syn-
(
Scheme 1). If 18 could be coerced into forming a carbo- thesis of 17, and the synthesis of key intermediate 18. (Inset 1:
ORTEP representation of X-ray crystallographic structure of 18.
cation of type-6 in the absence of a nucleophile, intramolec-
ular capture of the cation by the prenyl olefin would give
the core of neovibsanin G (9) and 14-epi-neovibsanin G
Alternative contributors to disorder are omitted for clarity. Ellip-
soids are shown at 30% probability. Inset 2: ORTEP representation
of X-ray crystallographic structure of 16. Ellipsoids are shown at
30% probability.).
(10).
2002
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 2001–2012

Total Synthesis of Neovibsanins
Results and Discussion
external nucleophiles was critical to the success of the cycli-
zation. Since it was known from the synthesis of 2-O-meth-
ylneovibsanin H (7) that carbocation (Ϯ)-20 could be
formed by treating (Ϯ)-18 with H SO in MeOH, these con-
Thus, the first objective was to obtain key intermediate 18
in an enantioenriched form in order to pursue an enantio-
selective synthesis of the caged neovibsane targets. This would
be best achieved by an asymmetric 1,4-addition of Grignard
reagent 15 to 3-methylcyclohexenone (14) (Scheme 3).
2
4
ditions were a sensible starting point in this investigation.
Clearly the use of MeOH as solvent, however, was not
applicable in this case. Nonetheless, classic carbocation
theory dictates that the stability of a carbocation is highly
dependent on the ability of the solvent to stabilize the
charge. Thus, a non-nucleophilic solvent with a high dielec-
tric constant and good physical and chemical compatibility
with H SO was required.
Mauduit and Alexakis had developed a chiral NHC-lig-
ated copper catalytic system[8] that allowed such a transfor-
mation to be realized. Thus, 16 was synthesized according
to published procedures from commercially available (S)-
[
9]
t-leucinol. Its structure and absolute configuration were
2
4
[
10]
confirmed by X-ray crystallography (Scheme 3, inset 2).
THF seemed to be the solvent with the most suitable
With 16 in hand, it was now possible to investigate the
enantioselective synthesis of ketone 17. Pleasingly, the 1,4-
addition of 15 to 3-methylcyclohexenone (14) in the pres-
physical properties. Thus, 19 was treated with H SO in an-
2
4
hydrous THF at room temperature overnight. When no re-
action was apparent, the reaction was heated to 40 °C for
ence of Cu(OTf) and 16 gave 17 in 70% yield, 91% ee.
2
3.5 hours. After aqueous work up, the crude product mix-
This reaction could be performed on up to 12 g scale with-
out any reduction in yield or selectivity.
Key intermediate 18 was then obtained following that
prescribed for the racemic series (Scheme 3).
ture was treated with diazomethane to give the desired cy-
clized products 21, 22 and 23 in a ratio of 1.0:1.5:6.8
(Scheme 4).
[4,11]
The abso-
Although these compounds were not separable by nor-
lute configuration of 18 was determined as 11R by X-ray
mal phase chromatography, clean fractions of each isomer
could be obtained following careful and repeated silver ni-
crystallography (Scheme 3, inset 1).[
12]
With key intermediate 18 in hand, two more objectives
remained in order to complete the core of 9 and 10, namely
the formation of the dihydrofuran ring system, and the con-
struction of the characteristic bicyclo[3.3.1]nonane ring sys-
tem. Towards the first objective, silyl deprotection of 18 was
achieved with catalytic HCl in THF. These conditions also
triggered an intramolecular 1,4-addition of the newly un-
veiled allylic alcohol onto the enone, furnishing 19 in 70%
yield (Scheme 4).
[13]
trate chromatography. The successful separation of these
isomers allowed for their full spectroscopic characteriza-
tion. The stereochemistry of 21 and 22 were assigned by
comparing the chemical shifts of the terminal alkene signals
of 21 (4.70 and 4.66 ppm) and 22 (4.90 and 4.83 ppm), with
neovibsanin G (9) (δ =4.70 ppm, two protons) and 14-epi-
neovibsanin G (10) (4.93 and 4.89 ppm). The stereochemis-
try of 21 and 22 were also independently confirmed by
1
NOESY 2D H NMR experiments. The key correlations
are shown in Figure 1.
Figure 1. Key NOESY correlations for determining the stereo-
chemistry of 21 and 22.
Having gained unprecedented access to the caged neovib-
sane core, attention was now focused on optimizing the
yield of the H SO cyclization reaction. The various condi-
2
4
tions attempted are summarized below in (Table 1).
Entry 2 of Table 1 was a pleasing result, allowing the
caged core to be accessed in reasonable yields from 19. As
a final optimization experiment, it was envisaged that the
silyl deprotection and dihydrofuran cyclization could easily
be achieved under the conditions of the bicyclo[3.3.1]-
nonane cyclization. Thus, when 18 was directly subjected to
the conditions of entry 2 of Table 1, the isomeric methyl
esters 21, 22 and 23 were obtained in a 1.0:2.0:5.7 ratio in
Scheme 4. Dihydrofuran and H
tions.
2 4
SO bicyclo[3.3.1]nonane cycliza-
Access to tricycle 19 allowed the crucial cyclization to 63% (74% yield based on recovered 19) (Scheme 5).
form the caged ring system to be investigated. It was clear Although the major product of the cyclization was the
that the formation of carbocation 20 in the absence of any tetrasubstituted alkene isomer 23, with terminal alkenes 21
Eur. J. Org. Chem. 2012, 2001–2012
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2003

J. Y. W. Mak, C. M. Williams
FULL PAPER
Table 1. Optimization of the H
2
SO
4
cyclization reaction.
and 22 obtained in relatively small quantities, this was ini-
tially believed to be of little concern, as 23 was seen as an
ideal synthetic precursor to the methoxy caged neovibsan-
ins (Scheme 1). Nonetheless, further optimization was
sought, in the pursuit of gaining complete control over the
regioselective outcome of the reaction. Accordingly, it
seemed prudent at this point to examine the mechanism of
the cyclization more closely in order to understand how the
regioselectivity could be improved (Scheme 5).
There were at least six chemical events that took place
in this acid mediated cascade. Firstly, acid-catalyzed silyl
deprotection occurred to give alcohol 24, which underwent
an intramolecular 1,4-addition to give an equilibrating mix-
ture of 24, 25 and 19, with the equilibrium heavily favoring
Entry Conditions
Result
1
2
3
5% H
then CH
5% H
then CH
0.3% H
5 h,
then CH
10% H SO
2
SO
4
in THF, room temp., 17 h,
2
in THF, 40 °C, 19 h,
2
1.4:7.4
products/RSM
60%, 5:1
products/RSM
mostly RSM,
2
N
2
SO
4
2
N
2
SO
4
(3 equiv.) in THF, 40 °C,
19. At this point, protonation of the lactone facilitates carb-
1
oxylate elimination, forming carbocation 20. Biosyntheti-
cally, it is possible that the cyclization of the olefin onto a
carbocation of this type is enzyme mediated. In vitro, the
stability of this carbocation was believed to be crucial in
order to achieve cyclization, as this carbocation must be
sufficiently long-lived in order for the olefin in the prenyl
side-chain to capture the cation without enzymatic confor-
mational pre-organization. As mentioned above, the solvent
was expected to be an important contributor to cation sta-
bility.
2
N
2
trace products
significant dec.
4
2
4
in THF, 40 °C, 2 h
As carbocation 20 was sufficiently long-lived, it was cap-
tured by the pendant olefin to give two diastereoisomeric
tertiary carbocations (26). A hybridization change at the
C-4 position triggered a C-5 epimerization to relieve the
increased steric strain between the two carbonyl groups to
give 27. Elimination, followed by methylation of the re-
sulting carboxylic acids 28 and 29 with diazomethane, gave
the observed products.
If the system was under complete kinetic control, and the
transition states of the competing elimination reactions
were equal in energy, note that the distribution of the alk-
enyl isomers should be 6:1 in favor of the terminal isomers
due to statistical reasons [as there were six methyl protons
(
H , 27) which could be eliminated compared to a single
m
tertiary proton (H )]. The use of Brønsted acid catalysis,
t
however, placed the entire reaction sequence under thermo-
dynamic control. Thus, under these conditions, the forma-
tion of the most stable olefin, tetrasubstituted isomer 23,
was expected to predominate; this was indeed observed ex-
perimentally.
Thus, based on these mechanistic considerations, in order
to regioselectively form the terminal alkenyl isomers, it
seemed clear that:
1. the reaction should be completely kinetically con-
trolled or the rate of equilibration of the final step should
be reduced, and/or,
2. the rate of the elimination reaction to give the terminal
alkenes should be increased relative to the competing unde-
sired elimination reaction.
Changing the solvent and the temperature of the H SO4
2
cyclization led to no improvement in yield or selectivity, and
thus, it was clear that a much more sophisticated approach
would be required in order to achieve regioselective cycliza-
Scheme 5. Formation of tricycles 21–23 directly from key interme-
diate 18, and mechanism of the H SO cyclization.
2
4
2004
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 2001–2012

Total Synthesis of Neovibsanins
tion. In order to address this issue, investigations began on Table 2. Lewis acid-catalyzed cyclization attempts.
developing non-equilibrating (i.e. kinetic) cyclization condi-
tions in order to selectively form the statistically more fa-
vorable (i.e. terminal alkenyl) products.
The first of these investigations focused on utilizing pal-
ladium catalysis to effect cyclization (Scheme 6). The for-
mation of π-allylpalladium species from allylic acetates and
their subsequent capture by soft carbon nucleophiles is well
[
14]
known. Thus, it was envisaged that treatment of 19 with
Pd would facilitate cyclization to give the targets’ core.
0
Entry[a] Conditions
Result
More importantly, the products were not expected to be a
Pd substrate, and hence, the reaction (should it proceed)
1
2
3
4
5
6
7
8
9
TMSOTf, (CH
TMSOTf, CH
TMSOTf, MeCN, –20 °C to room temp., Ar
TMSOTf, CCl , pyridine, room temp., Ar
, PhMe, room temp., Ar
2
Cl)
2
, room temp., Ar
decomposition
decomposition
decomposition
no reaction
0
2
2
Cl , –78 °C to room temp., Ar
would not be under thermodynamic control. Lactone 19
0
was, however, unreactive to Pd catalysis in all cyclization
4
+
–
Et
Me
BF
BF
EtAlCl
CH
EtAlCl
then CH
Et AlCl, THF, –78 to 45 °C, Ar
3
O BF
4
decomposition
attempts.
+
–
3
O BF
·OEt
·OEt
4 2 2
, CH Cl , –20 °C to room temp., Ar no reaction
, THF, –78 to 45 °C
, THF, –78 to reflux
3
2
no reaction
no reaction
33%, (21/22/23
1:1:1)
3
2
2
, THF, –78 to room temp., Ar, then
N
2 2
10
2
, Et
2
O, –78 °C to reflux, Ar,
trace products
2 2
N
11
2
decomposition
[a] Entries 1–8 were performed using racemic 19.
0
Scheme 6. Pd -catalyzed cyclization strategy via π-allylpalladium
species 30.
protected with catalytic HCl, and without further purifica-
tion, the crude mixture was treated with EtAlCl in THF
2
Next, the use of Lewis acids was investigated (Table 2). (entry 2). This resulted in an improvement of the combined
The Lewis acid TMSOTf appeared to be the ideal candidate yield of 34% from 18 over two steps compared to 33% from
for this reaction, as it could be used to dealkylate tert-butyl 19 over one step, (entry 10, Table 2), and an improvement
[15]
or benzyl esters via a S 1 mechanism.
In the initial of the selectivity to 2:1:1 (21:22:23).
attempts, employment of this reagent in (CH Cl)2 and Further investigations led to two more improvements
CH Cl resulted in decomposition of starting material. (Table 3). Firstly, it was clear that refluxing conditions re-
N
2
2
2
Changing to the much more polar solvent MeCN also led sulted in yield reductions; a temperature of 45 °C was the
to decomposition. No reaction was observed upon applica- optimum. Secondly, it was necessary to treat the crude mix-
[15]
tion of a modified Jung’s method
(entry 4). Meerwein ture of methyl esters with catalytic HCl in order to cyclize
had demonstrated that the ring opening of butyrolactones trace amounts of isomeric free allylic alcohols before
+
– [16]
could be achieved with Et O BF . On this system, how- chromatography. Under the optimized conditions for the
3
4
+
–
+
–
ever, both Me O BF and Et O BF were trialed without EtAlCl cyclization, the products 21, 22 and 23 could be
3
4
3
4
2
success (entries 5 and 6). Meanwhile, 19 was shown to be synthesized from 18 in 60% yield (73% based on recovered
completely inert to treatment with BF ·OEt (entries 7 & 19) in a ratio of 2.0:1.0:1.5.
3
2
8).
Notably, in the H SO cyclization, 22 was the major ter-
2 4
Pleasingly, when EtAlCl was explored (entry 9), the re- minal alkene product, while in the EtAlCl cyclization, 21
2
2
action yielded the desired products in equimolar ratio in a was the major product. Presumably, 21 was the kinetically
combined yield of 33%, a drastic improvement in regiose- favored terminal alkene isomer, but was also thermodynam-
lectivity over the H SO cyclization. The cyclization was re- ically less stable than 22 due to the proximity of the C-14
2
4
markably solvent and acid dependent. Repeating the experi- pendant group to the dihydrofuran methylene. Thus, when
ment in Et O resulted in only trace amounts of the products the cyclization was performed in H SO , 21 isomerized
2
2
4
being formed (entry 10), while changing the acid to the more quickly than 22 to tetrasubstituted 23, resulting in a
weaker Et AlCl resulted in gradual decomposition of the higher proportion of 22 to 21.
2
material (entry 11). The latter result was particular surpris-
Further optimization of the regioselectivity was sought.
ing; however, this could potentially be rationalized by the In Scheme 7, the general elimination reaction is shown
increased steric bulk of the Lewis acid hindering the sub- where E is a generalized electron deficient group. As dis-
strate’s ability to undergo the C-5 epimerization.
cussed above, if the elimination of E from 31 occurred much
A logical avenue for optimization was to perform the silyl more quickly than that of the tertiary proton, the terminal
deprotection, dihydrofuran cyclization and the bicy- alkenes would be formed selectively. For this to be achieved
clo[3.3.1]nonane cyclization in one pot. When 18 was through substrate modification, E would need to be less
treated with EtAlCl in THF, however, no reaction was ob- electronegative than hydrogen; to this end, organostannanes
2
served. Hence, in the next experiment, 18 was first silyl de- or organosilanes appeared to be suitable.
Eur. J. Org. Chem. 2012, 2001–2012
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2005

J. Y. W. Mak, C. M. Williams
FULL PAPER
Table 3. Optimization of the EtAlCl
2
cyclization in THF.
able by-products. Although the failure to access allylsilane
34 was disappointing, the optimized EtAlCl2 cyclization
conditions nonetheless afforded the bicyclic core of neovib-
sanin G (9) and 14-epi-neovibsanin G (10) in good yield
and regioselectivity, meaning that the completion of their
respective total syntheses was still well within reach.
Entry
1
Conditions
EtAlCl (13.0 equiv., 0.16 m),
THF, –78 to 45 °C, Ar
i. cat. HCl (32%), THF, room temp., Ar; 34%
Result
2
no reaction
2
ii. EtAlCl
THF, –78 °C to reflux, Ar
iii. CH , Et O, room temp.
2
(13.0 equiv., 0.17 m),
(21/22/23
2.0:1.0:1.0)
2
N
2
2
3
i. cat. HCl (32%), THF, room temp., Ar; 62% (73%
ii. EtAlCl
THF, –78 to 45 °C, Ar
iii. CH , Et O, room temp.
2
(9.3 equiv., 0.16 m),
based on
recovered 19)
(21/22/23
2
N
2
2
iv. cat. HCl (32%), THF, room temp., Ar 1.5:1.0:1.2)
4
5
i. cat. HCl (32%),THF, room temp., Ar;
ii. EtAlCl (9.3 equiv., 0.16 m), (CH Cl)
78 to 45 °C, Ar
iii. CH , Et O, room temp.
unidentifiable
mixture
2
2
2
,
Scheme 8. Attempted conversion of 18 to the corresponding allyl-
silane 34.
–
2
N
2
2
i. cat. HCl (32%), THF, room temp., Ar; 28%
Two steps remained in the total synthesis. Subjection of
ii. EtAlCl
THF, –78 °C to reflux, Ar
iii. CH , Et O, room temp.
iv. cat. HCl (32%), THF, room temp., Ar
2
(2 equiv., 0.027 m),
(21/22/23
1.0:0.3:1.0)
methyl esters 21 and 22 to a reduction/oxidation sequence
[20]
with LiAlH and Dess–Martin periodinane buffered with
4
2
N
2
2
[
5b]
pyridine gave the corresponding aldehydes 35 and 36 in
9% and 77% yields, respectively (Scheme 9). Finally, cap-
5
6
i. cat. HCl (32%), THF, room temp., Ar; 60% (73%
ture of the sodium enolates of 35 and 36 with the appropri-
ate acid chloride completed the first syntheses of the caged
neovibsanins in (–)-neovibsanin G [(–)-9] and (–)-14-epi-
neovibsanin G [(–)-10].
ii. EtAlCl
THF, –78 to 45 °C, Ar
iii. CH , Et O, room temp.
2
(12.9 equiv., 0.30 m),
based on
recovered 19)
(21/22/23
2
N
2
2
iv. cat. HCl (32%), THF, room temp., Ar 2.0:1.0:1.5)
Scheme 7. Generalized elimination reaction to the terminal alkenes.
Interestingly, Fujisawa reported[17] that the ring opening
of γ-alkenyl-γ-butyrolactones with allylsilanes could be ef-
fected with complete regio- and stereoselectivity using
+
–
Me O BF . Based on these encouraging reports, an at-
3
4
tempt was made to convert the prenyl side chain of 18 to
the corresponding allylsilane 34, following an oxidative
cleavage/Wittig reaction sequence reported by Yamamoto
Scheme 9. Completion of the total synthesis of (–)-neovibsanin G
[(–)-9] and (–)-14-epi-neovibsanin G [(–)-10].
[
18]
(Scheme 8).
To this end, application of Jin’s olefin oxi-
[19]
dative cleavage protocol
with osmium tetroxide and so-
As the optical rotations of these compounds were oppo-
dium periodate on 18 smoothly gave the desired aldehyde site in sign to the naturally derived compounds, this estab-
2 chemoselectively in moderate yield (37%). Frustratingly, lished the absolute configuration of neovibsanin G (9) and
3
despite repeated attempts, treatment of 32 with ylid 33 fol- 14-epi-neovibsanin G (10) as 11S, the same configuration
lowing Yamamoto’s procedure[ returned only unidentifi- as neovibsanins A (4) and B (5). Thus, this strongly sup-
18]
2006
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 2001–2012

Total Synthesis of Neovibsanins
ports the biogenetic correlation of the entire neovibsanin
subtype to vibsanin B (1), which was also shown to possess
the 11S configuration.[
21]
Having synthesized neovibsanin G (9) and 14-epi-neovib-
sanin G (10), attention was now focused on the total syn-
thesis of 15-O-methylneovibsanin F (11) and 14-epi-15-O-
methylneovibsanin F (12). It was envisaged that alkene 23,
formed during the bicyclo[3.3.1]nonane cyclization, would
provide access to these natural products. Should this syn-
thetic plan fail, changing the substrate to the more reactive
terminal alkenes 21 and 22 was expected to suffice as an
alternative strategy.
The simplest method of introducing the methoxy func-
tionality appeared to be through an acid-catalyzed MeOH
addition reaction. Thus, 23 was treated with H SO4 in
2
MeOH at room temperature (Scheme 10). No reaction was
observed, however, even when the reaction was repeated
with the theoretically more reactive terminal alkene isomer
4
Scheme 11. NaBH reduction of ketones 21–23, and subsequent
22. Prolonged exposure (three days) of the tetrasubstituted acid-catalyzed MeOH addition attempts.
isomer 23 to the H SO /MeOH medium at room tempera-
2
4
ture led to a gradual formation of an unidentifiable mixture.
In a final attempt to install the methoxy functionality, an
oxymercuration/demercuration strategy was investigated.
Tetrasubstituted olefin 23 proved to be completely inert to
oxymercuration using Hg(OAc)2 or Hg(CF CO ) in
3
2 2
MeOH at room temperature. Heating of the reaction led to
[
24]
a gradual decomposition of the mercuric salts. The fail-
ure of 23 to undergo oxymercuration was presumably due
to the demanding steric requirements of the resulting orga-
nomercurial 42 (Scheme 12).
Scheme 10. Acid-catalyzed MeOH addition attempts.
As the dihydrofuran ring system of 21–23 were formed
via an acid-catalyzed 1,4-addition, its sensitivity to acid was
known. Thus, it was suspected that under these acidic con-
ditions, a competing retro-Michael process (leading to dihy-
Scheme 12. Attempted oxymercuration of 23 to give organomercu-
rial 42.
The substrate was then changed to the terminal alkenes
drofuran ring opening) could be interfering with the desired 21 and 22, as the C-14 steric bulk of the resulting organo-
transformation. Thus, ketone 23 was reduced with sodium mercurials would be substantially less than 42. Pleasingly,
borohydride in MeOH to quantitatively give a mixture of when 22 was treated with Hg(OAc) in MeOH, followed
2
two diastereoisomeric alcohols (39/40 68:32) (Scheme 11). by demercuration with NaBH at 0 °C, the desired MeOH
4
When 39 was subjected to the same conditions as 23 in adduct 38 was obtained in 16% (20% brsm) (Scheme 13).
Scheme 10, however, no reaction was observed, even after Attempts to optimize this reaction, including the addition
[
25]
three days. A mixture of terminal alkenes 21 and 22 were of base to prevent methoxy elimination, led to no further
also reduced to their corresponding alcohols (41), and sub- improvement on this result. Although 38 represented the
jected to the same acidic conditions. No reaction was ob- core of 14-epi-15-O-methylneovibsanin F (12), the yield of
served. Upon heating in a sealed tube, isomerization to the the reaction was too low to complete the total synthesis. In
corresponding tetrasubstituted alkenes 39 and 40 occurred. an attempt to access the core of 15-O-methylneovibsanin F
Turning to alternate conditions for the introduction of (11), a mixture of terminal alkenes 21 and 22 were subjected
the methoxy functionality, the hydroboration of 23 using to the same oxymercuration/demercuration conditions as
BH ·DMS and 9-BBN were attempted, but did not give the Scheme 13. Although oxymercuration proceeded to com-
3
[
22]
desired outcomes. A diastereoselective Shi epoxidation
/
pletion with ease for both alkenes, upon demercuration,
reductive ring opening[
proved to be fruitless.
23]
strategy was trialed but also only 38 was formed. The formation of 37 (Scheme 10) was
not observed.
Eur. J. Org. Chem. 2012, 2001–2012
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2007

J. Y. W. Mak, C. M. Williams
FULL PAPER
ing succumbed to biomimetic total syntheses, this confirms
the original biosynthesis postulated by Fukuyama.[
2]
Based on the hypothesis that the methoxy caged neovib-
sanins (11 and 12, Scheme 1) were isolation artifacts formed
from 9 and 10 during prolonged (30 d) MeOH extraction
of the dried leaves of V. awabuki, attempts were made to
access the core of these natural products from the corre-
sponding alkenyl precursors. Only the core of 14-epi-15-O-
methylneovibsanin F (12) could be obtained, however, and
the results of the synthetic study strongly suggested that the
Scheme 13. Oxymercuration/demercuration synthesis of the 14-epi-
15-O-methylneovibsanin F core.
If the methoxy caged neovibsanins (11 and 12) were in- methoxy caged neovibsanins are true natural products.
deed artifacts of isolation, prolonged exposure of 9 and 10
to MeOH (Scheme 1) would offer insight. Thus, in order to
conclusively determine whether the methoxy caged neovib-
sanins were true natural products or artifacts, a solution of
synthetic (–)-14-epi-neovibsanin G [(–)-10] in MeOH was
allowed to stand for 30 days at room temperature in an
atmosphere of air, emulating Fukuyama’s[ original isola-
tion procedure (Scheme 14). A parallel experiment em-
ploying the same conditions was performed, except that sil-
Experimental Section
General Experimental Methods: Anhydrous diethyl ether and tetra-
hydrofuran were freshly distilled from sodium/benzophenone under
argon. Anhydrous methanol was prepared by distillation from cal-
cium hydride under argon. Argon was dried by passing through a
drying tube containing 4 Å molecular sieves, silica gel beads and
Drierite. Column chromatography was undertaken with distilled
2]
ica gel was added as a potential mild acid source. The first solvents. Petroleum spirit refers to the hydrocarbon fraction with
experiment returned only starting material, while in the lat- boiling points between 40–60 °C. Melting points were determined
ter experiment, a product tentatively assigned with the on a DigiMelt Stanford Research Systems melting point apparatus.
Enantiomeric excess determinations were performed by the Univer-
sity of Queensland Enantioselective Chromatography Facility on a
structure 43 was formed. Notably, in both experiments, no
addition of MeOH to any site on the substrate was ob-
served. Thus, these experiments, together with the results
from the synthetic studies above, strongly suggested that the
methoxy caged neovibsanins are true natural products and
not artifacts of isolation.
Daicel Chiralpak IC column (4.6ϫ250 mm, DAICEL Chemical
1
13
IND, LDT). The H and C NMR spectra were acquired on
Bruker AV300 (300 MHz; 75 MHz), AV400 (400 MHz; 100 MHz)
1
13
or AV500 (500 MHz; 125 MHz) instruments. The H and
C
NMR spectra data of the natural products were acquired on a
Bruker 900 MHz NMR spectrometer equipped with a cryoprobe.
Chemical shifts were reported in parts per million (ppm) on a δ
1
scale relative to the solvent peak ( H, CDCl
3
δ 7.24 ppm, C
6 6
D δ
1
3
7.15 ppm; C CDCl
3
δ 77.0 ppm, C
6
D
6
δ 128.0 ppm). High resolu-
tion electrospray ionization (HRMS) accurate mass measurements
were recorded in positive mode on a Bruker MicrOTOF-Q (quad-
rupole-Time of Flight) instrument with a Bruker ESI source. Accu-
rate mass measurements were carried out with external calibration
using sodium formate as a reference calibrant. Optical rotations
were performed on a JASCO P-2000 polarimeter.
Ketone 17: Following a similar procedure to Kehrli et al.,[8c] a flask
was charged with magnesium shavings (4.07 g, 170 mmol), and was
dried by heating with a heat gun under vacuum. After charging the
flask with argon, anhydrous diethyl ether was added (10 mL) as
well as
a
portion of 5-bromo-2-methylpent-2-ene (4.0 mL,
0 mmol). Once the reaction was initiated, a solution of remaining
-bromo-2-methylpent-2-ene (15.0 mL, 112 mmol) in anhydrous di-
3
5
ethyl ether (70 mL) was added via a pressure equalising dropping
funnel to maintain a gentle reflux. On completion, the reaction
was refluxed for a further 30 min, then allowed to cool to room
temperature. Into a separate heat gun dried flask was added 16
(1.89 g, 4.36 mmol) and copper(II) triflate (1.18 g, 3.27 mmol). The
flask was then placed under vacuum for 1.5 h. After charging the
flask with argon, anhydrous diethyl ether was added (140 mL), then
cooled to –30 °C. The solution of Grignard reagent 15 was added
dropwise to the vigorously stirred copper suspension by syringe.
Upon addition of Grignard reagent 15, the reaction initially turns
greyish-blue, then bright yellow, and then dark blue. A solution of
Scheme 14. Testing the natural product status of the methoxy caged
neovibsanins by subjecting synthetic 14-epi-neovibsanin G [(–)-10]
to conditions that emulate the isolation procedure.
Conclusions
This synthetic study involving the caged neovibsanins has
led to the total synthesis of (–)-neovibsanin G [(–)-9] and
14 (12.0 g, 109 mmol) in anhydrous diethyl ether (480 mL) was
(
–)-14-epi-neovibsanin G [(–)-10]. With members from the added dropwise to the vigorously stirred reaction at –30 °C via can-
[
5,26]
[4]
[6]
ketal,
bicyclic and now the caged neovibsanins hav- nula, over 1.5 h, maintaining the reaction at the same temperature.
2008
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 2001–2012

Total Synthesis of Neovibsanins
For smaller scale reactions, the addition of enone 14 solution
1
[
6 6
Methyl Ester 21: H NMR (C D , 500 MHz): δ = 5.36 (br. m, 1
should be performed via a syringe pump over 30 min.] The reaction
mixture was stirred for another 15 min until starting material was
consumed. The reaction mixture was then poured onto saturated
ammonium chloride solution (350 mL) and stirred overnight. The
reaction was partitioned, and the aqueous phase was extracted with
diethyl ether (3ϫ200 mL). The combined organic phases were suc-
cessively washed with saturated sodium hydrogen carbonate solu-
tion (350 mL) and brine (350 mL) before drying over magnesium
sulfate and evaporating to dryness in vacuo. The crude oily product
was distilled in vacuo. The fraction between 90 and 100 °C at
H), 4.70 (q, J = 1.5 Hz, 1 H), 4.66 (s, 1 H), 4.66–4.62 (m, 1 H),
4.31 (dt, J = 12.0, 2.6 Hz, 1 H), 3.37 (s, 3 H), 2.67 (dd, J = 14.9,
3.5 Hz, 1 H), 2.49 (br. m, 1 H), 2.46 (dd, J = 15.0, 8.3 Hz, 1 H),
2.32 (br. s, 1 H), 2.13 (m, 2 H), 1.90 (s, 3 H), 1.68–1.35 (m, 5 H),
1.52 (d, J = 0.5 Hz, 3 H), 1.10–1.05 (m, 1 H), 0.94 (dd, J = 12.6,
1
3
6 6
2.3 Hz, 1 H), 0.81 (s, 3 H) ppm. C NMR (C D , 125 MHz): δ =
205.4, 173.3, 148.0, 138.4, 134.7, 110.0, 83.1, 76.6, 51.4, 49.1, 45.1,
42.7, 39.1, 38.0, 35.6, 32.7, 32.3, 30.9, 28.0, 23.6, 22.5 ppm. HRMS
+
+
30 4
(ESI): Calculated for C21H O Na [M + Na ] 369.2036, found
2
4
369.2029. [α]
D
= –29.3 (c = 0.39, EtOAc).
0
.5 Torr was collected to give 17 as a colorless oil (14.7 g, 70%).
1
Methyl Ester 22: H NMR (C
6
D
6
, 500 MHz): δ = 5.37 (br. m, 1
Spectral data matched the data reported for the racemic com-
pound. After oxidation
the corresponding enone, the enantiomeric excess of the reaction
was determined by chiral HPLC (2% 2-propanol in hexane) as
H), 4.90 (q, J = 1.5 Hz, 1 H), 4.83 (s, 1 H), 4.44 (ddd, J = 12.1,
.5, 2.9 Hz, 1 H), 4.32 (complex d, J = 12.1 Hz, 1 H), 3.37 (s, 3
H), 2.65 (dd, J = 14.9, 3.4 Hz, 1 H), 2.43 (dd, J = 15.0, 8.9 Hz, 1
H), 2.41 (m, 1 H), 2.19 (m, 1 H), 2.18 (dd, J = 16.0, 7.0 Hz, 1 H),
.08 (dd, J = 16.0, 6.0 Hz, 1 H), 1.93 (s, 3 H), 1.80 (br. s, 1 H),
.60–1.52 (m, 1 H), 1.55 (s, 3 H), 1.49–1.39 (m, 2 H), 1.26 (dtd, J
12, 4.5, 2.0 Hz, 1 H), 1.21 (dd, J = 13, 2.5 Hz, 1 H), 1.08 (m, 1
[
3i]
[11b]
of a small amount of the product to
5
1
91%. H NMR (CDCl
3
, 400 MHz): δ = 5.05 (m, 1 H), 2.25 (t, J =
2
1
6
.8 Hz, 2 H), 2.18 (d, J = 13.6 Hz, 1 H), 2.09 (d, J = 13.2 Hz, 1
H), 1.94–1.81 (m, 4 H), 1.65 (s, 3 H), 1.57 (s, 3 H), 1.63–1.50 (m,
2
Et O).
=
24
H), 1.26 (m, 2 H), 0.91 (s, 3 H) ppm. [α]
D
= –13.2 (c = 13.3,
1
3
H), 0.78 (s, 3 H) ppm. C NMR (C
6
D
6
, 125 MHz): δ = 205.2,
73.4, 146.2, 137.6, 137.4, 111.1, 83.3, 74.7, 51.4, 48.4, 40.2, 39.7,
7.7, 35.5, 32.8, 30.9, 30.6, 29.9, 28.2, 22.4, 22.3 ppm. HRMS
2
1
3
Methyl Esters 21–23: Part A: To a stirred solution of 18 (60.6 mg,
.136 mmol) in anhydrous tetrahydrofuran (7 mL) was added hy-
+
+
(
3
30 4
ESI): Calculated for C21H O Na [M + Na ] 369.2036, found
0
2
5
69.2038. [α]
D
= +0.31 (c = 0.76, EtOAc).
drochloric acid (32%, 70 μL) dropwise at room temperature under
argon. After 1 h, the reaction was quenched by addition of satu-
rated sodium hydrogen carbonate solution (7.5 mL), then extracted
with diethyl ether (3ϫ12.5 mL). The combined organic layers were
washed with brine (12.5 mL), dried with magnesium sulfate, and
then concentrated in vacuo to give 19 as a colorless oil. No further
1
Methyl Ester 23: H NMR (C D , 500 MHz): δ = 5.39 (br. m, 1
6
6
H), 4.45–4.36 (m, 2 H), 3.39 (s, 3 H), 3.04 (br. s, 1 H), 2.62 (dd, J
= 14.9, 3.3 Hz, 1 H), 2.48 (t, J = 6.2 Hz, 1 H), 2.40 (dd, J = 14.9,
9.0 Hz, 1 H), 2.24 (dd, J = 15.9, 6.9 Hz, 1 H), 2.22–2.16 (m, 1 H),
2.12 (dd, J = 15.9, 6.2 Hz, 1 H), 2.08–2.00 (m, 1 H), 1.89 (s, 3 H),
purification was performed. Part B: A mixture of 19 above in anhy- 1.55 (s, 3 H), 1.51 (m, 1 H), 1.48 (d, J = 1.2 Hz, 3 H), 1.44 (m, 1
drous tetrahydrofuran (4.9 mL) was cooled to –78 °C under argon,
to which ethylaluminium dichloride (1.8 m in toluene, 980 μL,
H), 1.22 (m, 1 H), 1.05 (dd, J = 12.5, 2.3 Hz, 1 H), 0.79 (s, 3 H)
ppm. C NMR (C D , 125 MHz): δ = 205.3, 173.4, 136.8, 136.7,
6 6
1
3
1
.76 mmol) was added dropwise. The reaction mixture was then
129.7, 122.7, 83.1, 74.9, 51.5, 48.3, 41.7, 39.6, 35.75, 35.66, 32.8,
32.5, 30.7, 27.7, 23.7, 20.6, 19.9 ppm. HRMS (ESI): Calculated for
warmed rapidly to room temperature, then heated to 45 °C for 4 h,
during which time the reaction mixture became reddish-brown in
color. On cooling to room temperature, the reaction mixture was
diluted with ethyl acetate (7.5 mL) before carefully adding water
+
+
24
C H O Na [M + Na ] 369.2036, found 369.2042. [α]D = –148.1
2
1
30
4
(c = 1.3, EtOAc).
1
Lactone 19: H NMR (CDCl
3
, 500 MHz): δ = 5.82 (m, 1 H), 5.00
(7.5 mL) while vigorously stirring. The reaction mixture was ex-
(
m, 1 H), 4.44 (dp, J = 11.5, 1.5 Hz, 1 H), 4.28 (d, J = 11.0 Hz, 1
tracted with ethyl acetate (3ϫ12.5 mL). The combined organic
phases were washed with brine (12.5 mL), dried with magnesium
sulfate, then concentrated in vacuo to give a reddish-brown semi-
solid. No further purification was performed. Part C: Using
smooth glassware and a smooth stir bar, the crude material was
dissolved in diethyl ether (7 mL) at room temperature and a solu-
tion of diazomethane in diethyl ether was added dropwise, until
TLC analysis (silica, 10:1 diethyl ether/dichloromethane) showed
complete consumption of the carboxylic acids, which appear as
baseline spots. After degassing the reaction with a stream of nitro-
gen, the solvent was removed in vacuo to give a reddish-brown oil.
No further purification was performed. The crude material was
subjected to the procedure outlined in Part A again. This was nec-
essary to encourage any primary alcohols isomeric to the product
to cyclize to form the dihydrofuran ring. The crude oil was purified
by column chromatography (silica gel, 1:3 ethyl acetate/petroleum
spirit) to give 19 (6 mg, 13%) and a mixture of 21, 22 and 23
H), 4.14 (dd, J = 8.0, 4.5 Hz, 1 H), 2.91, (dd, J = 17.0, 7.5 Hz, 1
H), 2.59–2.47 (m, 4 H), 2.18 (s, 3 H), 2.03 (m, 1 H), 1.94 (dd, J =
1
7.5, 6.0 Hz, 1 H), 1.90–1.79 (m, 2 H), 1.63 (d, J = 1.0 Hz, 3 H),
1
3
1
.54 (s, 3 H), 1.29–1.21 (m, 2 H), 0.92 (s, 3 H) ppm. C NMR
, 125 MHz): δ = 206.3, 175.1, 136.1, 131.9, 123.7, 123.4,
8.2, 80.3, 68.4, 43.0, 40.8, 38.8, 37.0, 32.0, 31.01, 30.99, 25.6, 23.0,
(CDCl
3
8
2
5
1
2.3, 17.5 ppm. H NMR (C
6 6
D , 500 MHz): δ = 5.18 (m, 1 H),
.06 (br. t, J = 7.0 Hz, 1 H), 4.41 (br. d, J = 11.2 Hz, 1 H), 4.15
(
=
(
dd, J = 7.4, 5.0 Hz, 1 H), 3.97 (d, J = 11.2 Hz, 1 H), 2.75 (dd, J
16.5, 7.4 Hz, 1 H), 2.29 (dd, J = 16.5, 4.8 Hz, 1 H), 2.16–2.09
m, 3 H), 1.81 (m, 1 H), 1.74–1.67 (m, 1 H), 1.70 (s, 3 H), 1.65 (s,
3
1
3
1
3
C
H), 1.56 (s, 3 H), 1.47 (dq, J = 17.2, 3.4 Hz, 1 H), 1.34 (dd, J =
7.2, 5.7 Hz, 1 H), 1.15–1.08 (m, 1 H), 1.02–0.96 (m, 1 H), 0.54 (s,
1
3
6 6
H) ppm. C NMR (C D , 125 MHz): δ = 205.3, 174.1, 136.8,
31.5, 124.6, 123.1, 87.4, 81.2, 68.4, 43.0, 40.6, 39.2, 36.5, 32.9,
1.0, 30.6, 25.8, 22.6, 22.4, 17.6 ppm. HRMS (ESI): Calculated for
+
+
23
20
28 4 D
H O Na [M + Na ] 355.1880, found 355.1875. [α] = –29.3
(28 mg, 60%, 73% brsm, 2.0:1.0:1.5). The isomeric methyl esters
(
1
c = 0.49, EtOAc). IR (ATR): ν˜ = 2973, 2928, 2869, 1777, 1732,
717, 1380, 1162, 1116, 833, 570 cm .
could be separated completely through repeated and careful ar-
gentation chromatography in the dark (1:100 crude product to sil-
ica ratio, 10% silver nitrate on silica gel, 1:3 ethyl acetate/petroleum
spirit with 1% triethylamine). The silver nitrate impregnated silica
gel was prepared by forming a silica gel slurry with a solution of
silver nitrate in acetonitrile (10% w/w silver nitrate to silica gel),
and removing the acetonitrile in vacuo in the dark.
–
1
Aldehyde 32: To a solution of 18 (43.6 mg, 0.0976 mmol) in distilled
dioxane (0.75 mL) and water (1.0 mL) were added 2,6-lutidine
(22.5 μL, 0.194 mmol), osmium tetroxide (2.5% in tert-butanol,
20.1 μL, 0.00192 mmol) and sodium periodate (25.0 mg,
0.117 mmol) successively. After 30.5 h, water (2 mL) and dichloro-
Eur. J. Org. Chem. 2012, 2001–2012
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2009

J. Y. W. Mak, C. M. Williams
2.4 Hz, 1 H), 0.65 (s, 3 H) ppm. 13C NMR (C
, 125 MHz): δ =
FULL PAPER
methane (4 mL) were added to quench the reaction. The layers
were partitioned, and the aqueous layer was extracted with dichlo-
romethane (2ϫ4 mL). The combined organic phases were washed
with brine, dried with anhydrous sodium sulfate, before removing
the solvent in vacuo. The crude product was purified by column
6 6
D
205.5, 200.4, 147.9, 138.1, 134.8, 110.0, 82.9, 76.7, 49.0, 45.1, 44.7,
42.5, 38.3, 36.9, 32.2, 30.9, 30.2, 28.7, 23.5, 22.5 ppm. HRMS
+
+
(ESI): Calculated for C20
H
28
O
3
Na [M + Na ] 339.1931, found
2
4
339.1925. [α]
D
= –35.9 (c = 0.16, EtOAc).
chromatography to give the product 32 as a colorless oil (15.0 mg,
1
(–)-14-epi-Neovibsanin G [(–)-10]: To a solution of 36 (2.0 mg,
.0063 mmol) in anhydrous tetrahydrofuran (1 mL) was added
3
6
1
1
2
7%). H NMR (CDCl
3
, 500 MHz): δ = 9.77 (t, J = 1.3 Hz, 1 H),
0
.62 (d, J = 15.8 Hz, 1 H), 6.36 (d, J = 15.8 Hz, 1 H), 6.05 (br. m,
H), 4.14 (dddd, J = 14.7, 2.1, 2.1, 2.1 Hz, 1 H), 4.03 (dddd, J =
4.7, 1.9, 1.9, 1.9 Hz, 1 H), 2.56 (dd, J = 17.8, 8.8 Hz, 1 H), 2.45–
.40 (m, 3 H), 2.30 (dd, J = 8.2, 6.6 Hz, 1 H), 2.24 (s, 3 H), 2.08
dropwise sodium hexamethyldisilazide solution (0.25 m in tetra-
hydrofuran, 25 μL, 0.0063 mmol) at –78 °C under argon. The reac-
tion was warmed to –50 °C and held at this temperature for 5 min,
then immediately cooled to –78 °C, before a solution of 3,3-dimeth-
ylacryloyl chloride in anhydrous tetrahydrofuran (5% v/v, 14 μL,
(
0
m, 1 H), 1.99 (m, 1 H), 1.78–1.72 (m, 1 H), 1.60–1.54 (m, 1 H),
1
3
.86 (s, 9 H), 0.86 (s, 3 H), 0.012 (s, 3 H), 0.010 (s, 3 H) ppm.
, 100 MHz): δ = 201.0, 196.9, 174.7, 145.7, 133.6,
28.7, 124.2, 84.5, 61.3, 48.8, 38.3, 34.3, 33.7, 31.5, 30.5, 28.7, 25.9,
1.6, 18.3, –5.4, –5.5 ppm. HRMS (ESI): Calculated for
C
0.0087 mmol) was added to the reaction dropwise under argon. The
NMR (CDCl
1
2
C
3
reaction mixture was quenched by the addition of saturated sodium
hydrogen carbonate solution (1 mL). After warming to room tem-
perature, and adding water to dissolve any solids, the reaction mix-
ture was extracted with diethyl ether (3ϫ2 mL). The combined or-
ganic layers were washed with brine (2 mL), dried with magnesium
sulfate, before evaporating in vacuo. Purification of the crude mix-
ture by column chromatography (silica gel, 1:12 ethyl acetate/petro-
leum spirit with 1% triethylamine) gave [(–)-10] (0.24 mg, 9%, 13%
brsm), which was further purified by chiral HPLC to give a color-
less oil. Spectra data of [(–)-10] matched that of natural 14-epi-
+
+
24
23 36 5 D
H O SiNa [M + Na ] 443.2224, found 443.2220. [α] = +17.0
(c = 0.47, EtOAc).
Aldehyde 36: Following a procedure similar to that of Chen,[5b]
methyl ester 22 (7.4 mg, 0.021 mmol) was dissolved in anhydrous
tetrahydrofuran (1.0 mL) under argon. Lithium aluminium hydride
(5.2 mg, 0.14 mmol) was then added in one portion at room tem-
perature and stirred under argon for 10 min. The reaction was
quenched with sodium sulfate decahydrate and stirred until the re-
action became completely white, before filtering through celite and
eluting with ethyl acetate. After removing the solvent in vacuo, the
residue was taken up in dichloromethane/pyridine (1:1, 0.5 mL) un-
der argon, to which was added Dess–Martin periodinane (38 mg,
[2]
1
neovibsanin G.
6 6
H NMR (C D , 900 MHz): δ = 7.39 (d, J =
12.5 Hz, 1 H), 5.64 (m, 1 H), 5.30 (m, 1 H), 5.22 (dd, J = 12.5,
10.5 Hz, 1 H), 4.93 (br. s, 1 H), 4.89 (br. s, 1 H), 4.44 (ddd, J =
12.0, 5.5, 3.0 Hz, 1 H), 4.32 (br. d, J = 12.4 Hz, 1 H), 2.45 (dd, J
=
14.3, 3.8 Hz, 1 H), 2.37 (dd, J = 14.0, 8.2 Hz, 1 H), 2.24–2.22
0
.090 mmol) in one portion. After 10 min, TLC showed that the
reaction was complete. The reaction was diluted with diethyl ether
1 mL), and the resulting white suspension was poured onto 10%
(
m, 2 H), 2.03 (d, J = 0.9 Hz, 3 H), 1.96 (s, 3 H), 1.81 (br. s, 1 H),
1
5
.68–1.64 (m, 1 H), 1.60 (m, 1 H), 1.57 (s, 3 H), 1.46 (td, J = 13.5,
.5 Hz, 1 H), 1.36 (s, 3 H), 1.29 (dd, J = 13.0, 2.1 Hz, 1 H), 1.22
(
sodium thiosulfate solution (1 mL) and saturated sodium hydrogen
carbonate solution (1 mL), and stirred until the white precipitate
was dissolved. The layers were partitioned, and the aqueous layer
was extracted with diethyl ether (2ϫ2 mL). The combined organic
layers were successively washed with saturated sodium hydrogen
carbonate solution (2 mL) and brine (2 mL), before drying over
magnesium sulfate and removing the solvent in vacuo. Column
chromatography (silica gel, 1:3 ethyl acetate/petroleum spirit) gave
(
br. d, J = 13.9 Hz, 1 H), 1.18 (br. d, J = 12.6 Hz, 1 H), 0.77 (s, 3
13
H) ppm. C NMR (C
1
4
HRMS (ESI): Calculated for C25H O Na [M + Na ] 421.2349,
34 4
found 421.2359. [α]
6
D
6
, 225 MHz): δ = 205.3, 163.3, 159.7,
46.1, 137.5, 136.5, 136.4, 115.2, 114.0, 111.3, 83.6, 74.9, 48.4, 42.7,
0.1, 37.6, 33.0, 30.9, 30.41, 30.37, 29.7, 27.0, 22.6, 22.1, 20.2 ppm.
+
+
24
D
= –153.1 (c = 0.036, EtOH).
(
–)-Neovibsanin G [(–)-9]: Treating a mixture of 35 and 36 (1:1 ratio,
1
9.7 mg, 0.031 mmol, 0.015 mmol of 35) with an analogous pro-
cedure to 36 gave [(–)-9] as an oil (0.5 mg, 8%, 12% brsm), which
was further purified by chiral HPLC to give a colorless oil. Spectra
the aldehyde product 36 (4.9 mg, 73%) as an amorphous solid. H
NMR (C
1
=
6
D
6
, 500 MHz): δ = 9.34 (t, J = 1.5 Hz, 1 H), 5.25 (br. m,
H), 4.91 (sext, J = 1.5 Hz, 1 H), 4.84 (br. m, 1 H), 4.43 (ddd, J
12.2, 5.5, 3.0 Hz, 1 H), 4.29 (ddd, J = 12.2, 1.8, 0.9 Hz, 1 H),
data of [(–)-9] matched[ that of natural neovibsanin G.
27]
[2] 1
H
NMR (C
1
6 6
D , 900 MHz): δ = 7.40 (d, J = 12.4 Hz, 1 H), 5.63 (m,
H), 5.24 (m, 1 H), 5.17 (dd, J = 12.4, 10.6 Hz, 1 H), 4.70 (s, 2
2.43 (dd, J = 15.2, 4.5 Hz, 1 H), 2.39 (dd, J = 15.0, 7.8 Hz, 1 H),
2.29 (br. m, 1 H), 2.19 (br. m, 1 H), 1.98 (ddd, J = 17.8, 7.3, 1.7 Hz,
1 H), 1.89 (ddd, J = 17.8, 4.2, 1.4 Hz, 1 H) 1.89 (s, 3 H), 1.80 (br.
H), 4.62 (ddd, J = 11.9, 5.5, 3.1 Hz, 1 H), 4.30 (m, 1 H), 2.56 (dd,
J = 13.8, 3.9 Hz, 1 H), 2.41 (dd, J = 14.0, 7.0 Hz, 1 H), 2.37 (m, 1
H), 2.34 (br. s, 1 H), 2.03 (d, J = 0.90 Hz, 3 H), 1.97 (s, 3 H), 1.70
m, 1 H), 1.57–1.52 (m, 2 H), 1.56 (d, J = 0.5 Hz, 3 H), 1.42–1.35
1
3
(
m, 1 H), 1.22–1.17 (m, 2 H), 0.99 (m, 1 H), 0.62 (s, 3 H) ppm.
NMR (C , 125 MHz): δ = 205.2, 200.3, 146.0, 137.4, 137.3,
11.2, 83.0, 74.7, 48.4, 44.7, 40.2, 37.6, 37.1, 32.6, 30.8, 30.6, 30.3,
C
(
m, 2 H), 1.53 (s, 3 H), 1.51 (m, 1 H), 1.46 (dt, J = 12.5, 3.2 Hz, 1
6
D
6
H), 1.40 (m, 1 H), 1.35 (d, J = 1.0 Hz, 3 H), 1.12 (m, 1 H), 1.02
(dd, J = 12.5, 2.2 Hz, 1 H), 0.79 (s, 3 H) ppm. C NMR (C D ,
6 6
225 MHz): δ = 205.8, 163.3, 159.6, 147.9, 137.2, 136.3, 134.9, 115.2,
1
2
C
1
2
1
3
+
8.8, 22.5, 22.2 ppm. HRMS (ESI): Calculated for C20
H
28
O
3
Na
+
24
[
M + Na ] 339.1931, found 339.1937. [α]
D
= +1.35 (c = 0.49,
14.3, 110.1, 83.9, 77.0, 48.5, 45.0, 42.5, 42.3, 38.3, 32.9, 32.6, 30.9,
EtOAc).
9.5, 27.0, 23.5, 22.5, 20.2 ppm. HRMS (ESI): Calculated for
Aldehyde 35: Treating 21 (3.0 mg, 0.0087 mmol) with an analogous
+
+
22
25 34 4 D
H O Na [M + Na ] 421.2349, found 421.2356. [α] = –152.6
1
procedure to its epimer 22 gave 35 as an oil (1.6 mg, 59%).
NMR (C , 500 MHz): δ = 9.35 (t, J = 1.7 Hz, 1 H), 5.21 (br. m,
H), 4.70 (q, J = 1.4 Hz, 1 H), 4.65 (s, 1 H), 4.61 (ddd, J = 12.1, Alcohols 39 and 40: To a solution of ketone 23 (21 mg, 0.061 mmol)
H
(
c = 0.02, EtOH).
6
D
6
1
5
.4, 3.3 Hz, 1 H), 4.28 (dddd, J = 12.1, 3.1, 2.2, 0.7 Hz, 1 H), 2.50
in anhydrous methanol (1.5 mL) at 0 °C was added sodium borohy-
dride (8 mg, 0.21 mmol) in one portion under argon, then warmed
to room temperature. Upon consumption of the starting material
as indicated by TLC, the reaction was cooled to 0 °C, and saturated
(
(
(
dd, J = 15.0, 4.4 Hz, 1 H), 2.39 (dd, J = 14.7, 7.2 Hz, 1 H), 2.35
br. m, 1 H), 2.29 (br. s, 1 H), 1.96 (dd, J = 7.3, 1.8 Hz, 1 H), 1.94
dd, J = 4.0, 1.6 Hz, 1 H), 1.87 (s, 3 H), 1.65 (br. d, J = 11.3 Hz,
1
1
H), 1.61–1.53 (m, 1 H), 1.52 (s, 3 H), 1.49–1.45 (m, 1 H), 1.39– ammonium chloride solution (1 mL) was added dropwise. The or-
.27 (m, 2 H), 1.03 (td, J = 13.6, 5.2 Hz, 1 H), 0.91 (dd, J = 12.6, ganic solvent was evaporated in vacuo, and the residue was ex-
2010
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 2001–2012

Total Synthesis of Neovibsanins
tracted with diethyl ether (3ϫ2 mL). The combined organic phases
were washed with saturated brine (2 mL), dried with magnesium
sulfate, before removing the solvent in vacuo. The crude mixture
was purified by column chromatography (silica gel, 1:3 ethyl acet-
ate/petroleum spirit) to give quantitatively the diastereoisomeric
alcohols 39 and 40 as clear oils in a ratio of 68:32.
and authentic neovibsanin G and 14-epi-neovibsanin G. CCDC-
841524 (for 18) and -857590 (for 16) contain 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.
1
Alcohol 39: H NMR (C
6
D
6
, 500 MHz): δ = 4.97 (m, 1 H), 4.29–
Acknowledgments
4.28 (m, 2 H), 4.19 (m, 1 H), 4.07 (br. s, 1 H), 3.30 (s, 3 H), 3.02
(
br. s, 1 H), 2.52 (t, J = 6.0 Hz, 1 H), 2.21 (dd, J = 16.0, 7.3 Hz, 1
The authors thank the Australian Research Council (DP
DP0666855) for financial support. Prof. Paul V. Bernhardt and Dr.
Patricia Y. Hayes (School of Chemistry and Molecular Biosciences,
University of Queensland) performed the X-ray crystallographic
data collection and chiral HPLC purification (Enantioselective
chromatography facility) respectively. Dr. Greg Pierens (Centre for
Advanced Imaging, University of Queensland) assisted in the
NMR spectroscopic data collection of synthetic neovibsanin G and
14-epi-neovibsanin G, while Prof. Yoshiyasu Fukuyama (Faculty of
Pharmaceutical Sciences, Tokushima Bunri University) provided
the NMR spectra of natural neovibsanin G and 14-epi-neovibsanin
G. Dr. Marc Mauduit (Ecole Nationale Supérieure de Chimie de
Rennes) and Prof. Alexandre Alexakis (Département de Chimie
Organique, Université de Genéve) provided invaluable advice con-
cerning the asymmetric 1,4-addition. Their contributions to this
work are gratefully acknowledged.
H), 2.15 (dt, J = 15.0, 5.2 Hz, 1 H), 2.04 (dd, J = 16.0, 5.7 Hz, 1
H), 1.98 (m, 1 H), 1.88 (dt, J = 13.9, 2.2 Hz, 1 H), 1.56–1.50 (m,
1
6
1
H), 1.54 (s, 3 H), 1.49–1.42 (m, 2 H), 1.47 (s, 3 H), 1.28 (d, J =
.2 Hz, 3 H), 1.18 (ddd, J = 17.7, 11.8, 5.9 Hz, 1 H), 1.04 (dd, J =
2.6, 2.1 Hz, 1 H), 0.78 (s, 3 H) ppm. 13C NMR (C
6 6
D , 125 MHz):
δ = 173.3, 137.2, 135.7, 129.6, 122.7, 87.5, 74.9, 68.1, 51.3, 42.9,
4
1.7, 39.3, 35.74, 35.67, 32.7, 32.5, 27.6, 23.9, 23.6, 20.6, 19.9 ppm.
+
+
HRMS (ESI): Calculated for C21
H
32
O
4
Na [M + Na ] 371.2193,
found 371.2201. [α]2
5
D
= –209.3 (c = 1.08, EtOAc).
, 500 MHz): δ = 5.19 (m, 1 H), 4.40–
.39 (m, 2 H), 4.20 (m, 1 H), 3.35 (s, 3 H), 3.06 (br. s, 1 H), 2.62
br. s, 1 H), 2.56 (m, 1 H), 2.24 (dd, J = 15.5, 7.5 Hz, 1 H), 2.23–
1
6 6
Alcohol 40: H NMR (C D
4
(
2
1
2
(
1
.18 (m, 1 H), 2.10 (dd, J = 16.0, 6.0 Hz, 1 H), 2.11–2.05 (m, 1 H),
.90 (ddd, J = 14.5, 9.0, 3.0 Hz, 1 H), 1.62 (ddd, J = 14.0, 9.0,
.5 Hz, 1 H), 1.54 (d, J = 0.5 Hz, 3 H), 1.52–1.45 (m, 2 H), 1.49
d, J = 1.5 Hz, 3 H), 1.25–1.18 (m, 1 H), 1.20 (d, J = 6.5 Hz, 3 H),
.07 (dd, J = 12.5, 2.0 Hz, 1 H), 0.80 (s, 3 H) ppm. 13C NMR
, 125 MHz): δ = 173.5, 137.2, 136.1, 129.8, 122.6, 83.9, 74.8,
5.3, 51.4, 42.4, 41.8, 39.6, 35.9, 35.7, 32.9, 32.6, 27.7, 24.1, 23.7,
(C
6
D
6
[1] Y. Fukuyama, M. Kubo, T. Esumi, K. Harada, H. Hioki, Het-
erocycles 2010, 81, 1571–1602.
6
2
+
[2] Y. Fukuyama, M. Kubo, H. Minami, H. Yuasa, A. Matsuo, T.
Fujii, M. Morisaki, K. Harada, Chem. Pharm. Bull. 2005, 53,
0.6, 19.9 ppm. HRMS (ESI): Calculated for C21
H
32
O
4
Na [M +
+
26
Na ] 371.2193, found 371.2196. [α]
D
= –185.3 (c = 0.57, EtOAc).
7
2–80.
[3] a) J. Y. W. Mak, C. M. Williams, Nat. Prod. Rep. 2012, DOI:
0.1039/C2NP00067A; b) S. Chow, C. Kreß, N. Albæk, C.
Isopropyl Methyl Ether 38: To a solution of 22 (14 mg, 0.040 mmol)
in anhydrous methanol (4.4 mL) was added mercury(II) acetate
1
Jessen, C. M. Williams, Org. Lett. 2011, 13, 5286–5289; c) J. M.
Faber, C. M. Williams, Chem. Commun. 2011, 47, 2258–2260;
d) B. D. Schwartz, J. R. Denton, P. V. Bernhardt, H. M. L.
Davies, C. M. Williams, Synthesis 2009, 17, 2840–2846; e) B. D.
Schwartz, J. R. Denton, H. M. L. Davies, C. M. Williams, Aust.
J. Chem. 2009, 62, 980–982; f) B. D. Schwartz, J. R. Denton,
Y. Lian, H. M. L. Davies, C. M. Williams, J. Am. Chem. Soc.
2009, 131, 8329–8332; g) B. D. Schwartz, C. M. Williams, P. V.
Bernhardt, Beilstein J. Org. Chem. 2008, 4(34), DOI: 10.3762/
bjoc.4.34; h) M. J. Gallen, C. M. Williams, Eur. J. Org. Chem.
(
17 mg, 0.053 mmol) in one portion at room temperature under
argon. After 5 min, TLC showed consumption of starting material
and the appearance of a baseline spot corresponding to the
organomercurate intermediate. The reaction mixture was cooled to
0
0
After 10 min, TLC showed consumption of the organomercurate
by the disappearance of the baseline spot. Reaction mixture was
passed through a short silica gel plug (pre-treated with a 1% trieth-
ylamine solution in diethyl ether), and eluted with diethyl ether.
After removing the solvent in vacuo, the crude reaction mixture
was taken up in diethyl ether and passed through a second silica
gel plug (pre-treated with a 1% triethylamine solution in diethyl
ether), and eluted with diethyl ether. After removing the solvent in
vacuo, the crude mixture was purified by column chromatography
°C in an ice/brine bath, then sodium borohydride (1.5 mg,
.040 mmol) was added in one portion to give a grey suspension.
2
008, 27, 4697–4705; i) M. J. Gallen, C. M. Williams, Org. Lett.
2
008, 10, 713–715; j) B. D. Schwartz, D. P. Tilly, R. Heim, S.
Wiedemann, C. M. Williams, P. V. Bernhardt, Eur. J. Org.
Chem. 2006, 14, 3181–3192; k) C. M. Williams, L. N. Mander,
P. V. Bernhardt, A. C. Willis, Tetrahedron 2005, 61, 3759–3769.
4] A. P. J. Chen, C. M. Williams, Org. Lett. 2008, 10, 3441–3443.
5] a) A. P. J. Chen, C. C. Müller, H. M. Cooper, C. M. Williams,
Tetrahedron 2010, 66, 6842–6850; b) A. P. J. Chen, C. C.
Müller, H. M. Cooper, C. M. Williams, Org. Lett. 2009, 11,
[
[
(silica gel, 1:3 to 1:1 ethyl acetate/petroleum spirit with 1% triethyl-
1
amine) to give 38 as a colorless oil (2.5 mg, 16%, 20% brsm). H
NMR (C , 500 MHz): δ = 5.31 (m, 1 H), 4.70 (ddd, J = 12.3,
.7, 2.1 Hz, 1 H), 4.45 (dd, 1 H), 3.38 (s, 3 H), 2.91 (s, 3 H), 2.67
dd, J = 14.7, 3.6 Hz, 1 H), 2.41 (dd, J = 15.0, 8.7 Hz, 1 H), 2.32–
3
758–3761.
6] J. Y. W. Mak, C. M. Williams, Chem. Commun. 2012, 48, 287–
89.
6
D
6
[
[
[
5
(
2
1
2
7] M. Kubo, Y. Kishimoto, K. Harada, H. Hioki, Y. Fukuyama,
Bioorg. Med. Chem. Lett. 2010, 20, 2566–2571.
.19 (m, 3 H), 2.08 (dd, J = 15.0, 7.0 Hz, 1 H), 1.95 (s, 3 H), 1.37–
.31 (m, 3 H), 1.24–1.18 (m, 2 H), 1.17–1.13 (m, 1 H), 1.03 (dd, J
8] a) S. Kehrli, D. Martin, D. Rix, M. Mauduit, A. Alexakis,
Chem. Eur. J. 2010, 16, 9890–9904; b) J. Wencel, M. Mauduit,
H. Henon, S. Kehrli, A. Alexakis, Aldrichim. Acta 2009, 42,
43–50; c) D. Martin, S. Kehrli, M. d’Augustin, H. Clavier, M.
Mauduit, A. Alexakis, J. Am. Chem. Soc. 2006, 128, 8416–
8417.
=
13.0, 3.5 Hz, 1 H), 0.86 (s, 3 H), 0.85 (s, 3 H), 0.82 (s, 3 H) ppm.
C NMR (C D , 75 MHz): δ = 205.3, 173.6, 141.3, 134.6, 83.0,
6 6
1
3
7
2
C
(
6.8, 74.9, 51.4, 48.8, 48.3, 45.5, 41.8, 35.9, 34.4, 32.0, 30.6, 30.2,
9.2, 28.3, 26.6, 22.3, 20.9 ppm. HRMS (ESI): Calculated for
+
+
23
22 34 5 D
H O Na [M + Na ] 401.2298, found 401.2289. [α] = –33.9
[
9] H. Clavier, L. Coutable, L. Toupet, J.-C. Guillemin, M. Maud-
c = 0.071, EtOAc).
uit, J. Organomet. Chem. 2005, 690, 5237–5254.
[10] CCDC-857590 contain the supplementary crystallographic
data for this compound. These data can be obtained free of
Supporting Information (see footnote on the first page of this arti-
cle): 1H and C spectra. Spectral comparison between synthetic
13
Eur. J. Org. Chem. 2012, 2001–2012
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2011

J. Y. W. Mak, C. M. Williams
FULL PAPER
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
11] a) B. D. Schwartz, A. Porzelle, K. S. Jack, J. M. Faber, I. R.
[20] D. B. Dess, J. C. Martin, J. Org. Chem. 1983, 48, 4155–4156.
[21] Y. Fukuyama, H. Minami, S. Takaoka, M. Kodama, K. Ka-
wazu, H. Nemoto, Tetrahedron Lett. 1997, 38, 1435–1438.
[22] M.-X. Zhao, Y. Shi, J. Org. Chem. 2006, 71, 5377–5379.
[
Gentle, C. M. Williams, Adv. Synth. Catal. 2009, 351, 1148–
1
154; b) R. Heim, S. Wiedemann, C. M. Williams, P. V. Bern- [23] a) D. Tanner, T. Groth, Tetrahedron 1997, 53, 16139–16146; b)
hardt, Org. Lett. 2005, 7, 1327–1329.
P. Ma, V. S. Martin, S. Masamune, K. B. Sharpless, S. M. Viti,
J. Org. Chem. 1982, 47, 1378–1380.
[24] J. Chatt, Chem. Rev. 1951, 48, 7–43.
[25] F. G. Bordwell, M. L. Douglass, J. Am. Chem. Soc. 1966, 88,
993–999.
[
12] CCDC-841524 contains the supplementary crystallographic
data for this compound. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
[
[
13] C. M. Williams, L. N. Mander, Tetrahedron 2001, 57, 425–447.
14] J. Tsuji, in Palladium Reagents and Catalysts: Innovations in
Organic Synthesis, Wiley, Chichester, 1995, pp. 297–328.
15] M. F. Jung, M. A. Lyster, J. Am. Chem. Soc. 1977, 99, 968–
[26] H. Imagawa, H. Saijo, T. Kurisaki, H, Yamamoto, M. Kubo,
Y. Fukuyama, M. Nishizawa, Org. Lett. 2009, 11, 1253–1255.
1
13
[27] The H and C spectra of (–)-9 and (–)-10 matched the spectra
of natural neovibsanin G and 14-epi-neovibsanin G provided
by Prof. Y. Fukuyama in every way. There are, however, a
number of typographical discrepancies between the spectra of
natural neovibsanin G and the its tabulated NMR spectro-
[
[
[
969.
16] H. Meerwein, P. Borner, O. Fuchs, H. J. Sasse, H. Schrodt, J.
Spille, Chem. Ber. 1956, 89, 2060–2079.
[2]
17] a) M. Kawashima, T. Fujisawa, Bull. Chem. Soc. Jpn. 1988, 61,
scopic data in the original article (ref. ). The chemical shifts
of neovibsanin G at the C-10 position should read 2.37 and
4
051–4055; b) T. Fujisawa, M. Kawashima, S. Ando, Tetrahe-
1
13
dron Lett. 1984, 25, 3213–3216.
42.5 ppm ( H NMR and C NMR respectively), while at C-18,
1
[
[
18] Y. Yamamoto, T. Furuta, J. Org. Chem. 1990, 55, 3971–3972.
19] W. Yu, Y. Mei, Y. Kang, Z. Hua, Z. Jin, Org. Lett. 2004, 6,
the H NMR chemical shifts should read 4.62 and 4.30 ppm.
Received: December 13, 2011
Published Online: February 16, 2012
3217–3219.
2012
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 2001–2012