Towards the total synthesis of mycaperoxide B: Probing biosynthetic rationale

Article abstract of DOI:10.1002/ejoc.201101477

Studies towards the biomimetic synthesis of mycaperoxide B (1) are described. We have established the synthesis of four diastereoisomers of mycaperoxide B methyl ester (1a) by employing a Michael addition across an α,β-unsaturated ester precursor 2 as the

Full text of DOI:10.1002/ejoc.201101477

FULL PAPER
DOI: 10.1002/ejoc.201101477
Towards the Total Synthesis of Mycaperoxide B: Probing Biosynthetic
Rationale
Eduarda M. P. Silva,[a] Richard J. Pye,[b] Geoffrey D. Brown,[b] and
Laurence M. Harwood*[b]
Keywords: Natural products / Biomimetic synthesis / Michael addition / Peroxides / Oxygen heterocycles
Studies towards the biomimetic synthesis of mycaperoxide B
(1) are described. We have established the synthesis of four
diastereoisomers of mycaperoxide B methyl ester (1a) by em-
ploying a Michael addition across an α,β-unsaturated ester
precursor 2 as the key step. This result strongly suggests
stereocontrol in the addition of the hydroperoxide function-
ality to the E double bond and discloses the importance of
choosing the correct geometry of the α,β-unsaturated double
bond when attempting to synthesise mycaperoxide B. Four
diastereoisomeric tetrahydrofurans derived from an intramo-
lecular rearrangement of the 1,2-dioxolane enolate 12 were
also isolated and characterised.
in investigating the viability of a synthetic methodology
based on the second proposed biosynthetic pathway re-
Introduction
The mycaperoxide family[1–5] consists of eight structur- ported by Capon for the formation of this family of norses-
ally related marine norsesterterpenes with a common car- terterpene cyclic peroxides.[7] According to this hypothesis,
bon skeleton that consists of a tetramethyldecalin nucleus it was expected that the generation of a tertiary hydroperox-
possessing a side-chain at the C-1 position that bears a 1,2- ide and subsequent intramolecular Michael cyclisation ac-
dioxolane ring. Mycaperoxide B (1; Scheme 1) was isolated ross an α,β-unsaturated ester (Scheme 1) would allow the
in 1993 from marine sponges of the genus Mycale and preparation of the desired target.[7]
exhibits significant cytotoxicity against various cancer cell
The method that seemed to be most suitable for ob-
lines as well as antiviral activity.[1] Mycaperoxide B (1) con- taining the 1,2-dioxolane in a single step consisted of an
tains an interesting cyclic peroxide moiety in its structure intramolecular addition of a tertiary hydroperoxide group
and its absolute configuration was assigned to be to an α,β-unsaturated ester. The conjugate addition of the
2ЈS,3ЈS,6ЈS,1R,2R,9S,10S.[1] For these reasons, mycaperox- hydroperoxide to an α,β-unsaturated ester was first de-
ide B (1) represents a novel and synthetically challenging scribed by Bartlett and Chapuis in an approach towards the
target and development of an efficient facile synthetic route synthesis of polyether tetrahydrofurans.[8] More recently,
presents a rewarding and important objective. Following Massanet and co-workers described the preparation of 1,2-
previous work,[6] our group has been particularly interested dioxolane, 1,2-dioxane or 1,2-dioxepane rings by the intra-
Scheme 1. Retrosynthetic analysis of mycaperoxide B (1) inspired by the second biosynthetic pathway proposed by Capon.[7]
molecular Michael addition of a secondary hydroperoxide
[a] QOPNA, Department of Chemistry, University of Aveiro,
group onto an α,β-unsaturated ester as the key step.[9] In
3810-193 Aveiro, Portugal
Fax: +351-234-370-084
addition, 3-methoxy-6-[(methoxycarbonyl)methyl]-1,2-diox-
ane,[10a–10g] 3-methoxy-5-[(methoxycarbonyl)methyl]-1,2-di-
oxolane[10h] and spiro-peroxides,[10i] some of which have
anti-malarial activity, have been synthesised by the intramo-
lecular Michael addition of hydroperoxy ketals to α,β-un-
saturated esters. Posner and co-workers have described the
E-mail: eduarda.silva@ua.pt
[b] Department of Chemistry, University of Reading,
Whiteknights, Reading RG6 6AD, United Kingdom
Fax: +44-118-3786121
E-mail: l.m.harwood@reading.ac.uk
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201101477.
Eur. J. Org. Chem. 2012, 1209–1216
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1209

E. M. P. Silva, R. J. Pye, G. D. Brown, L. M. Harwood
FULL PAPER
intramolecular attack of a tertiary hydroperoxide group on amine.[14] A solution of the aldehyde 4 in dichloromethane
the α,β-unsaturated carbonyl of carvone.[11] In this case, the was added dropwise to a solution of the ylide 5 in dichloro-
spatial proximity of the two groups favoured the reaction. methane (Scheme 2). The reaction mixture was stirred over-
Finally, Woerpel and co-workers have recently synthesised night at room temperature under an inert atmosphere, after
1,2-dioxane systems by intramolecular palladium-catalysed which time TLC analysis of the mixture indicated complete
addition of hydroperoxides to olefins.[12] This method, how- consumption of the starting material along with the forma-
ever, appears to be specific to tertiary hydroperoxides be- tion of a new component. The crude material was purified
cause less substituted substrates did not cyclise.
by flash column chromatography, eluting with petroleum
Recently, we demonstrated for a series of model alkene ether/diethyl ether (9:1), to furnish the desired product 6 in
1
substrates the introduction of the hydroperoxide functional 69% yield. The H NMR spectrum of compound 6 shows
group into the required position for a biosynthetically in- a new olefinic signal at δH = 6.78–6.74 ppm and a three-
spired synthesis of mycaperoxide B (1).[13] We also demon- proton singlet at δH = 3.73 ppm, which corresponds to the
strated that the introduction of a triethylsilylperoxy substit- methyl ester group. A singlet at δH = 1.84 ppm was assigned
uent into analogues of a potential synthetic precursor to to the resonance of the methyl group attached to the newly
the mycaperoxides is dependent on hydroxy protection and created double bond. Given that only one signal is associ-
1
this has led to the successful synthesis of triethylsilyl perox- ated with both these functional groups in both the H and
ide 3 (Scheme 1).[13]
13C NMR spectra of 6, it is credible to assume that the
Herein we describe the successful, although non-stereose- Wittig reaction had taken place with complete stereocon-
lective, synthesis of four diastereoisomers of mycaperoxide trol, as expected. The completion of the synthesis of the
B (1) by using the (E)-α,β-unsaturated methyl ester precur- proposed biomimetic precursor 7 (Scheme 2) was achieved
sor 2, as well as the formation of four diastereoisomeric in a yield of 81% by removal of the triethylsilyl group from
tetrahydrofuran compounds by intramolecular rearrange- the hydroperoxide 6 using a catalytic amount of pyridinium
1
ment of an enolate intermediate involved in the formation p-toluenesulfonate (PPTS). The H NMR spectrum of de-
of the dioxolane ring.
protected compound 7 shows two singlets at δH = 9.24 and
8.89 ppm in an approximately 1:1 ratio, which correspond
to the resonances of two hydroperoxide protons, which re-
veals that the product had been obtained as a mixture of
two diastereoisomers.
Results and Discussion
With the methodology for preparing the protected hydro-
With all the key features of the natural product in place
peroxide precursor 3[13] in hand, the next objective was to the most critical and important step in this synthesis had
assemble the remaining parts of the side-chain and, finally, now been reached: the construction of the cyclic peroxide
to prepare the 1,2-dioxolane ring. According to the retro- ring. The intramolecular cyclisation of the hydroperoxide
synthetic analysis outlined in Scheme 1, the remaining parts onto the conjugated methyl ester in 7 was first studied in
of the side-chain were to be assembled by a Wittig reaction model system 8 (Scheme 3), which was prepared by using a
between the aldehyde 4 and a stabilised ylide 5 that would methodology similar to that previously reported.[13] It was
be expected to set up the desired E stereochemistry for 2 initially decided to attempt the cyclisation reaction under
(Scheme 2).
basic conditions (Table 1, entries 1–8). The conditions that
The primary alcohol in the side-chain of 3 was selectively resulted in the optimum formation of the desired 1,2-diox-
oxidised to the aldehyde 4 in quantitative yield by using olane ring 10 (15%) along with the undesired epoxide 11
Dess–Martin periodinane (Scheme 2) with the crude mate- (40%) involved the use of 0.2 equiv. of triethylamine in
rial being used in the next step without further purification. methanol (Table 1, entry 3).
1
A triplet at δH = 9.70 ppm (J = 2.5 Hz) in the H NMR
It was reasoned that the best way to improve the low
spectrum and a carbonyl stretch at 1725 cm–1 in the IR yield of the desired product 10 would be to promote the
spectrum indicated successful formation of the desired rapid protonation of the intermediate enolate 9 and thereby
product. The ylide 5 (Scheme 2) required for the Wittig re- disfavour the formation of the unwanted epoxide 11.[15] The
action with 4 was synthesised according to the literature formation of an epoxide, by cleavage of the O–O bond, had
procedure from commercially available triphenylphosphane previously been reported in a publication concerning the
and methyl 2-bromopropionate in the presence of triethyl- structure elucidation of the marine natural product plakor-
Scheme 2. Preparation of the α,β-unsaturated methyl ester 7.
1210
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 1209–1216

Towards the Total Synthesis of Mycaperoxide B
Scheme 3. Preparation of 1,2-dioxolane 10 by using model system 8.
Table 1. Optimisation of the conditions for the cyclisation step with ance of two new double quartets at δH = 2.67 and 2.56 ppm,
compound 8.
which integrate to one proton and correspond to the reso-
nance of the proton adjacent to the methyl ester, concomi-
tant with the loss of the olefin signal and the two hydroper-
oxide signals of the starting material, confirmed the forma-
tion of 10. The H NMR spectrum of compound 11 shows
a new multiplet at δH = 3.93–3.86 ppm attributed to the
En-
try
Conditions
Yield [%]
10 11
65
14 48
15 40
8
1
1
2
3
4
5
6
7
8
KF, 18-crown-6, THF
–
–
–
–
–
–
Et3N (0.1 equiv.), MeOH
Et3N (0.2 equiv.), MeOH
Hünig’s base, MeOH
single proton of the epoxide ring. Two new singlets at δH
=
9
8
–
–
–
–
37
68
93
–
1.54 and 1.53 ppm, which integrate to three protons and
correspond to the resonance of the methyl group geminal
to the methyl ester, also confirmed the formation of epoxide
11. As the highest yield for the formation of the desired
1,2-dioxolane ring in the model compound was achieved by
using triethylamine in methanol (Table 1, entry 3), it was
decided to apply these same conditions to substrate 7
(Scheme 4). Thus, a methanolic solution of freshly distilled
triethylamine (0.05 m, 2 equiv.) was added dropwise over a
period of 2 h to a solution of the hydroperoxide 7 in meth-
anol and the reaction was monitored by TLC (petroleum
ether/diethyl ether, 2:1). After half the base had been added,
TLC analysis revealed the concurrent formation of two
products. The first spot had an Rf value higher than the
starting material and the second spot appeared almost on
the baseline. The solution was stirred under nitrogen for
48 h in total and the crude material was purified by flash
column chromatography by using a mixture of petroleum
ether/diethyl ether (2:1) as eluent.
NaOMe, MeOH
LHMDS, –78 °C, THF
LHMDS, TMSCl, –78 °C, THF
Et2NH (0.4 equiv.), MeOH/CF3CH2OH (1:1), 0 °C
MgBr2·OEt2, MeOH
BF3·OEt2, toluene
SnCl4, toluene
–
68
–
–
9
–
47
10
11
12
no reaction
decomposition
66
Yb(CF3SO3)3, THF, H2O
–
–
tin.[16] This interesting rearrangement occurred on treat-
ment of plakortin with sodium methoxide in methanol,
which led to the formation of the epoxide through ring con-
traction followed by the final formation of a tetra-
hydrofuran. By using trimethylsilyl chloride it was hoped to
capture and isolate the intermediate 9 as its silyl enol ether
and therefore avoid the formation of epoxide 11. Unfortu-
nately, no reaction was observed and only the starting mate-
rial was recovered (Table 1, entry 7). The use of Lewis acid
conditions resulted in no reaction or decomposition
(Table 1, entries 9–12).
The least polar fraction was isolated as a colourless oil
in a yield of 10% and NMR spectroscopic analysis (see
The preparation of both the desired product 10 and the Table S1, Supporting Information) suggested it was the de-
hydroxy epoxide 11 was confirmed by analysis of the NMR sired dioxolane product 12. The second fraction to be iso-
1
spectroscopic data. The H NMR spectrum of compound lated consisted of recovered starting material (21%) and the
10 shows a new multiplet at δH = 4.11–4.07 ppm attributed most polar fraction, which was obtained in 43% yield, was
to the single proton of the peroxide ring. Also, the appear- confirmed to be the tetrahydrofuran byproduct 13 by NMR
Scheme 4. Preparation of the six-membered peroxide ring 12.
Eur. J. Org. Chem. 2012, 1209–1216
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1211

E. M. P. Silva, R. J. Pye, G. D. Brown, L. M. Harwood
FULL PAPER
analysis (see Table S2, Supporting Information; Scheme 4). was a significant discrepancy between our observations and
1
The H NMR spectrum of the first fraction (see Table S1, the chemical shifts quoted for C-5Ј and C-7Ј. In our sample
Supporting Information) shows the appearance of two new these signals were extremely broad and poorly resolved in
double double doublets at δH = 4.104 (J = 7, 7, 7 Hz) and the 13C NMR spectrum, which made assignment difficult;
4.141 ppm (J = 7, 7, 7 Hz), which combined integrate as indeed some of the assignments for C-5Ј and C-7Ј reported
one proton, along with two new double quartets at δH
=
in Table S1 (see the Supporting Information) were more
2.658 (J = 7, 7 Hz) and 2.620 ppm (J = 7, 7 Hz), which also easily made on the basis of 2D experiments.
integrate to one proton, concomitant with the loss of the
The formation of four of the possible isomers starting
olefin signal and the two hydroperoxide signals from the from the R diastereoisomer at C-6Ј of 7 is illustrated in
starting material. Examination of the 1H–1H COSY, HSQC Scheme 5. The hydroperoxide can perform a Michael ad-
and HMBC spectra showed that the two new double double dition to either the Re or Si face of the double bond conju-
doublets at δH = 4.104 and 4.141 ppm correspond to the gated to the ester and the resulting enolate could sub-
proton at C-3Ј on the peroxide ring and that the two new sequently be protonated on either the Si or Re side, thus
double quartets at δH = 2.658 and 2.620 ppm correspond allowing the formation of up to four products from this
to the proton adjacent to the methyl ester group at C-2Ј. single diastereoisomer, and a total of eight if both C-6Ј epi-
From these results it was evident that cyclic peroxide forma- mers of 7 are considered. As intramolecular attack of the
tion had occurred and the colourless oil was tentatively as- hydroperoxide on both the Re and Si faces of the double
signed the gross structure 12 of the mycaperoxide skeleton. bond appears to be equally probable, we propose that the
However, although the 1H NMR spectroscopic data for the observed stereoselectivity results from an overall stereose-
least polar material were very similar to those previously lective trans addition across the trans double bond (i.e.,
reported for mycaperoxide B (1), they did not match ex- either Si addition of H+ to the enolate from the Si Michael
actly. Examination of the 13C NMR spectrum of 12 was addition or Re addition of H+ to the enolate from the Re
extremely revealing. Many of the signals were “tripled” (see Michael addition). The four diastereoisomers that would re-
Table S1, Supporting Information), but C-3Ј and C-8Ј both sult from the operation of such a selective mechanism on
appeared as four distinct resonances, which indicates the both the 6ЈR and 6ЈS substrates are depicted in Figure 1
presence of four inseparable diastereoisomers. Given that and possess either a 1,4-trans (12a,b) or a 1,4-cis (12c,d)
there are three stereogenic centres at C-2Ј, C-3Ј and C-6Ј, relationship across the 1,2-dioxolane ring.
and therefore eight possible diastereoisomers, the observed
Capon and MacLeod developed an empirical rule de-
formation of only four such isomers implies some form of signed to assign the relative stereochemistry about the C-
stereoselectivity during the cyclisation reaction. As for the 2Ј, C-3Ј and C-6Ј atoms of the mycaperoxides.[17] It was
1H NMR spectrum, although the chemical shifts of the ma- proposed that the relative configuration between C-2Ј and
jority of the carbon signals compared very favourably with C-3Ј in cyclic peroxy systems such as mycaperoxide B (1)
both mycaperoxide B (1) and its methyl ester (1a),[1] there could be assigned on the basis of the 1H NMR shift for the
Scheme 5. Possible formation of up to four diastereoisomers in the Michael addition reaction when using the 6ЈR diastereoisomer.
1212
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 1209–1216

Towards the Total Synthesis of Mycaperoxide B
between the 3Ј-H (δH = 4.141 ppm) and the 6Ј-Me (inter-
changeably δH = 1.114/1.122 ppm), which was absent for
the corresponding protons in 12a/12b. The observation of
such a weak correlation between the rather distant 1- and
4-positions of this cis-dioxolane ring may be related to its
conformational mobility, as revealed by line-broadening in
the 13C NMR spectrum (particularly severe for C-5Ј and C-
7Ј). Such line-broadening was also observed by Tanaka et
al. when characterising mycaperoxide A, the structure of
which was later confirmed by X-ray crystallography.[1]
These same cis and trans relationships were observed
very clearly for the tetrahydrofuran ring in the byproducts
13a–13d. As shown in Scheme 3, these compounds arise
from an alternative reaction of the enolate produced by ad-
dition of the hydroperoxide group to the alkene in 7.
HRMS revealed a molecular formula of C25H44O5 for the
most polar fraction. The 1H NMR spectrum of this fraction
shows a series of double doublets at δH = 4.048 (J = 9,
5.5 Hz), δH = 4.062 (J = 9, 5.5 Hz), δH = 4.105 (J = 7.5,
7.5 Hz) and δH = 4.104 ppm (J = 7.5, 7.5 Hz), which corre-
late to tetrahydrofuran methine protons (Figure 2, see
Table S2, Supporting Information). Singlets at δH = 3.45,
Figure 1. NMR spectroscopic data for the cyclic peroxide ring of
the four diastereoisomers isolated and 13C NMR spectroscopic
data for mycaperoxide B methyl ester (1a).
C-2Ј secondary methyl group. Examination of the 1H NMR 3.33, 3.093 and 3.086 ppm were shown to be the protons
shifts for the C-2Ј secondary methyl group in several dioxol- associated with the tertiary alcohol 2Ј-OH adjacent to the
ane-containing natural products revealed that, in those ester group as they underwent exchange with D2O. The
cases in which C-2Ј and C-3Ј are in an erythro configuration presence of an alcohol was also confirmed by the presence
(R,R or S,S), the 1H NMR shift for the C-2Ј methyl is more of a strong absorption band at 3435 cm–1 in the IR spec-
upfield (δ = 1.13–1.14 ppm) than in those examples in trum. The appearance of four distinct signals for this group
which they are threo (R,S or S,R; δ = 1.24–1.26 ppm).[17] indicated once again that four diastereoisomers were pres-
The authors state that this observation is independent of ent. The presence of four diastereoisomers was also con-
the stereochemistry of C-6Ј, but all the examples considered firmed by examination of the 13C NMR spectrum, which
possess an axial C-3Ј oxymethine proton. Consequently, shows four distinct carbon signals associated with almost
this technique cannot be extended to assign the stereochem- every carbon. The structures of the four diastereoisomers
istry about C-2Ј in compounds in which the C-3Ј proton is and the NMR spectroscopic data associated with each are
equatorial.
represented in Figure 2.
The stereochemistry of the 1,2-dioxolane ring bearing
NOE studies suggest that for compounds 13a and 13b,
substituents has been deduced for several different natural 3Ј-H and 6Ј-CH3 are in a cis relationship to each other, and
products containing this functionality by applying this em- that for compounds 13c and 13d, 3Ј-H and 6Ј-CH3 are in a
pirical rule.[1,4,17,18] In our case, the chemical shift of the C- trans relationship. However, the NMR spectroscopic data
2Ј methyl group in compounds 12a–d is between δH = 1.23 assigned to structures 13a and 13b are interchangeable, as
and 1.25 ppm (Figure 1), which suggests that the stereo- are the data assigned to structures 13c and 13d. There is no
chemistry of the C2Ј–C3Ј fragment is threo and lends sup- means, by NMR studies, of deciding which cis-tetra-
port to our proposal of a stereoselective trans addition ac- hydrofuran is derived from the 6ЈS diastereoisomer and
ross the double bond.
which is derived from the 6ЈR hydroperoxide starting mate-
The relative stereochemistry at C-6Ј in 12a–12d (Fig- rial. Similarly, it is impossible to assign the relative stereo-
ure 1) is consistent with the chemical shifts observed in the chemistry of the trans-tetrahydrofurans. To conclude, the
13C NMR spectra for the C-6Ј methyl group of these dia- relative stereochemistries of the cyclised materials 12a–d are
stereoisomers. Six-membered cyclic peroxides are reported proposed to be 2ЈR,3ЈS,6ЈR, 2ЈS,3ЈR,6ЈR, 2ЈS,3ЈR,6ЈS and
to exist predominantly in chair conformations and equato- 2ЈR,3ЈS,6ЈS, respectively. Unfortunately, none of these four
rial methyl groups (δC = 23.5–24.0 ppm) are known to reso- diastereoisomers have data that exactly match the NMR
nate significantly downfield relative to axial methyl groups spectra of mycaperoxide B methyl ester (1a) with a
(δC = 20.5–20.9 ppm).[17] The chemical shifts of the 6Ј- 2ЈS,3ЈS,6ЈS stereochemistry. It is suggested that to synthe-
methyl group in 12a and 12b (δC = 20.66 ppm) suggest that sise mycaperoxide B by using this methodology it will be
this group should be axial and for 12c and 12d (inter- necessary to start with the Z alkene of 7 to furnish the
changeably δC = 23.53/23.43 ppm) this methyl group should correct stereochemistry about the C-3Ј position instead of
be equatorial. The cis relationship between the substituents the E alkene, which was used in our studies. Assuming that
at the 6Ј- and 3Ј-positions of 12c/12d was then confirmed by the Michael addition step remains stereospecifically trans,
the observation of a weak nuclear Overhauser enhancement as in Scheme 5 (i.e., Si addition of the hydroperoxide is fol-
Eur. J. Org. Chem. 2012, 1209–1216
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1213

E. M. P. Silva, R. J. Pye, G. D. Brown, L. M. Harwood
FULL PAPER
structure of mycaperoxide B (1) by a pathway following the
proposed biosynthesis of Capon.[7] All the steps leading to
the formation of compound 12 progressed efficiently until
the Michael addition cyclisation step. The formation of the
cyclic peroxide occurred in a low yield and examination of
the NMR spectroscopic data revealed the formation of four
diastereoisomers of mycaperoxide B (1). Nonetheless, the
successful formation of the cyclic peroxide 12 and the tetra-
hydrofuran 13 provide evidence to support Capon’s second
biosynthetic proposal concerning the formation of the char-
acteristic cyclic peroxide common to the mycaperoxide fam-
ily. Following these results, future work will involve the
preparation of a (Z)-α,β-unsaturated methyl ester precursor
analogue of 2 and the preparation of the 1,2-dioxolane ring
from this substrate.
Experimental Section
General Methods: 1H NMR spectra were recorded with either a
Bruker AV500 (500 MHz), a Bruker AMX400 (400 MHz) or a
Bruker DPX250 (250 MHz) spectrometer. Spectra were referenced
to the residual solvent peak or TMS as the internal standard. 13C
NMR spectra were recorded with the same spectrometers listed
above at either 125, 100 or 62.5 MHz. 2D NMR spectra, such as
HSQC, HMBC, 1H–1H COSY and NOESY, were all recorded at
high resolution in order to make full assignments of 12a–12d and
13a–13d. HRMS were recorded under conditions of CI with am-
monia as the ionising source or ESI. Finnigan MAT 95 and Bruker
micro TOF mass spectrometers were used, respectively. IR spectra
were recorded as a thin film between two sodium chloride plates
with a Perkin–Elmer 1720-X FT-IR spectrometer. Flash column
chromatography was performed according to the procedure devel-
oped by Still et al.[19] by using Merck 60 silica gel (particle size
0.040–0.063 nm, 230–400 mesh ASTM) with head pressure pro-
duced by means of hand bellows. TLC was performed with Merck
aluminium plates coated with 0.25 mm silica 60 containing F254.
Reagents obtained from Acros, Aldrich, Avocado, Fluka and Lanc-
aster fine chemicals suppliers were either used directly as supplied
or following purification according to procedures described by Per-
rin and Armarego.[20]
6Ј-(9,10-trans-1α-Hydroxy-2α,5,5,9β-tetramethyldecahydronaphth-
alen-1β-yl)-4Ј-methyl-4Ј-(triethylsilylperoxy)hexanal (4): Dess–Mar-
tin periodinane (0.61 g, 15 wt.-%, 0.22 mmol) was added dropwise
over 5 min to a stirred solution of 1-[6Ј-hydroxy-3Ј-methyl-3Ј-(tri-
ethylsilylperoxy)hexyl]-9,10-trans-2α,5,5,9β-tetramethyldecahydro-
naphthalen-1α-ol (3; 85 mg, 0.18 mmol) in dichloromethane (8 mL)
under nitrogen. The resulting solution was stirred at room tempera-
ture for a further 2 h. The reaction mixture was diluted with diethyl
ether (15 mL), washed with sodium hydroxide (1 m, 15 mL) and
extracted with diethyl ether (3ϫ30 mL). The combined organic ex-
tracts were washed with brine (15 mL), dried with Na2SO4 and
concentrated in vacuo to furnish the aldehyde 4 as a colourless oil
(84 mg, quant.) which was used without further purification. Rf =
Figure 2. NMR spectroscopic data of the four diastereoisomers of
13.
lowed by Si addition of H+ to the resulting enolate), then
it should be possible to produce the natural diastereoisomer
of mycaperoxide B starting from the 6ЈS diastereoisomer of
the Z alkene of 7.
0.33 (20% diethyl ether/petroleum ether). IR (thin film): ν
=
˜
max
3539, 2948, 2934, 2867, 1725, 1460, 1372, 1238, 1017, 975, 802,
1
741 cm–1. H NMR (250 MHz, CDCl3): δH = 9.70 (t, J = 2.5 Hz,
1 H, 1Ј-H), 2.43–2.37 (br. t, J = 7.5 Hz, 2 H, 2Ј-H), 1.85–1.77 (m,
2 H, 3Ј-H), 1.75–1.01 (m, 16 H, 2-, 3-, 4-, 6-, 7-, 8-, 10-H and 5Ј-,
6Ј-H), 1.07 (s, 3 H, 4Ј-CH3), 0.93–0.86 (m, 12 H, SiCH2CH3 and
9-CH3), 0.80–0.76 (m, 9 H, 2-CH3 and 5-CH3), 0.59 (q, J = 7.5 Hz,
Conclusions
The key objective of this project has been accomplished,
namely the first synthesis of a compound having the gross 6 H, SiCH2CH3) ppm. 13C NMR (63 MHz, CDCl3): δC = 203.1
1214
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 1209–1216

Towards the Total Synthesis of Mycaperoxide B
(C-1Ј), 84.0 and 83.9 (COOTES), 77.4 and 77.3 (C-1), 46.7 and
18.6 (C-7), 16.9 and 16.8 (2-CH3), 16.3 and 16.0 (9-CH3), 12.3 (2Ј-
46.6 (C-10), 43.82 and 43.80 (C-6), 42.2 (C-2Ј), 37.0 and 36.8 (C- CH3) ppm. HRMS (ESI): calcd. for C25H44O5NaSi [M + Na]+·
2), 34.2 (5α-CH3), 33.7 (C-5), 33.3 and 33.2 (C-5Ј), 32.5 and 32.4
(C-8), 31.8 (C-6Ј), 30.7 (C-3Ј), 28.0 and 27.9 (C-3), 22.4 (5β-CH3),
22.1 (C-4), 22.0 (4Ј-CH3), 19.1 and 19.0 (C-7), 16.9 and 16.8 (2-
CH3), 16.7 and 16.6 (9-CH3), 7.3 and 7.2 (SiCH2CH3), 3.9
(SiCH2CH3) ppm. HRMS (ESI): calcd. for C27H52O4NaSi [M +
Na]+· 491.3533; found 491.3539.
447.3082; found 447.3081.
Methyl 2-{6-[2-(tert-Butyldiphenylsilanyloxy)ethyl]-6-methyl-1,2-di-
oxan-3-yl}propanoate (10): Freshly distilled triethylamine (0.32 mL,
32.0 μmol, 0.1 m in methanol) was added dropwise to a solution
of methyl (2E)-8-(tert-butyldiphenylsilanyloxy)-6-hydroperoxy-2,6-
dimethylocta-2-enoate (8; 152 mg, 0.32 mmol) in methanol (5 mL)
under nitrogen. The resulting solution was stirred at room temp.
for a further 24 h. The solvents were subsequently concentrated in
vacuo and the crude material was purified by flash chromatography
on silica. Elution with petroleum ether/diethyl ether (9:1) gave a
first fraction, pure compound 10, obtained as a colourless oil
(23 mg, 15%) as a mixture of inseparable diastereoisomers. Further
elution gave a second fraction, pure compound 11,obtained as a
colourless oil (73 mg, 40%) as a mixture of inseparable diastereo-
isomers.
Methyl (2ЈE)-8Ј-(9,10-trans-1α-Hydroxy-2α,5,5,9β-tetramethylde-
cahydronaphthalen-1β-yl)-2Ј,6Ј-dimethyl-6Ј-(triethylsilylperoxy)oct-
2Ј-enoate (6): A solution of 6Ј-(9,10-trans-1α-hydroxy-2α,5,5,9β-tet-
ramethyldecahydronaphthalen-1β-yl)-4Ј-methyl-4Ј-(triethylsilylper-
oxy)hexanal (4; 50 mg, 0.11 mmol) in dichloromethane (2 mL) was
added dropwise over 15 min to a stirred solution of methyl 2-(tri-
phenylphosphanyl)propionate (5; 44 mg, 0.13 mmol) in dichloro-
methane (2 mL) under nitrogen. The resulting solution was left to
stir overnight at room temperature. The solvents were removed in
vacuo and the crude residue was purified by flash column
chromatography on silica, eluting with petroleum ether/diethyl Methyl 2-{6-[2-(tert-Butyldiphenylsilanyloxy)ethyl]-6-methyl-1,2-di-
ether (9:1), to give the title compound as a colourless oil (39 mg,
oxan-3-yl}propanoate (10): Rf = 0.63 (20% diethyl ether/petroleum
69%). Rf = 0.89 (20% diethyl ether/petroleum ether). IR (thin film):
ether). IR (thin film): νmax = 1711, 1586 cm–1. 1H NMR (400 MHz,
˜
ν
˜
max = 3561, 2951, 2873, 1714, 1461, 1280, 1018, 805, 741 cm–1. 1H CDCl3): δH = 7.74–7.65 (m, 4 H, Ph), 7.47–7.35 (m, 6 H, Ph), 4.11–
NMR (400 MHz, CDCl3): δH = 6.78–6.74 (m, 1 H, 3Ј-H), 3.73 (s,
4.07 (m, 1 H, 3-H), 3.76 (m, 2 H, 8-H), 3.68 (s, 3 H, COOMe),
3 H, COOCH3), 2.21–2.15 (m, 2 H, 4Ј-H), 1.841 and 1.838 (s, 3 H, 2.67 (dq, 0.5 H, 2-H), 2.56 (dq, 0.5 H, 2-H), 2.13–1.87 (m, 1 H,
2Ј-CH3), 1.76–1.11 (m, 18 H, 2-, 3-, 4-, 6-, 7-, 8-, 10-H and 5Ј-, CH2), 1.83–1.73 (m, 1 H, CH2), 1.71–1.50 (m, 2 H, CH2), 1.66 (s,
7Ј-, 8Ј-H), 1.164 and 1.158 (s, 3 H, 6Ј-CH3), 0.98 (t, J = 8.0 Hz, 9 1.5 H, CH3), 1.59 (s, 1.5 H, CH3), 1.43–1.20 (m, 2 H, CH2), 1.12
H, SiCH2CH3), 0.94 and 0.93 (s, 3 H, 9-CH3), 0.866 (d, J = 6.5 Hz, (s, 3 H, 2-CH3), 1.05 (s, 9 H, tBu) ppm. 13C NMR (100 MHz,
3 H, 2-CH3), 0.867 (s, 3 H, 5α-CH3), 0.83 (s, 3 H, 5β-CH3), 0.67 CDCl3): δC = 174.7 and 174.6 (C=O), 136.0 (Ph), 135.9 (Ph), 134.1
(q, J = 7.4 Hz, 6 H, SiCH2CH3) ppm. 13C NMR (100 MHz,
CDCl3): δC = 168.7 (C=O), 142.7 (C-3Ј), 127.4 (C-2Ј), 84.0
(COOTES), 76.9 (C-1), 51.7 (COOCH3), 46.2 (C-10), 43.43 and
43.37 (C-9), 41.8 and 41.7 (C-6), 36.6 and 36.4 (C-2), 35.5 and 35.4
(C-5Ј), 33.8 (5α-CH3), 33.3 (C-5), 32.7 and 32.6 (C-8), 32.1 and
32.0 (C-3), 31.4 (C-7Ј), 27.6 and 27.5 (C-8Ј), 23.34 and 23.27 (C-
4Ј), 22.0 (5β-CH3), 21.9 and 21.8 (6Ј-CH3), 21.7 (C-4), 18.7 and
18.6 (C-7), 16.5 and 16.4 (2-CH3), 16.3 and 16.2 (9-CH3), 12.3 (2Ј-
CH3), 6.8 (SiCH2CH3), 3.9 (SiCH2CH3) ppm. HRMS (ESI): calcd.
for C31H58O5NaSi [M + Na]+· 561.3939; found 561.3946.
(Ph), 130.2 (Ph), 128.1 (Ph), 81.7 and 81.6 (C-3), 80.1 (C-6), 79.8
(q), 60.5 and 59.9 (C8), 52.3 (COOCH3), 43.3 and 43.2 (C-2), 37.9
(C-7), 32.9 and 32.3 (C-5), 27.2 (tBu), 24.0 and 23.6 (C-4), 22.7 (6-
CH3), 19.5 (q, tBu), 14.0 and 13.9 (2-CH3) ppm. HRMS (CI):
calcd. for C27H39O5Si [M + H]+ 471.2567; found 471.2574.
Methyl 8-(tert-Butyldiphenylsilanyloxy)-6-hydroxy-2,3-epoxy-2,6-di-
methylocta-6-enoate (11): Rf = 0.17 (20% diethyl ether/petroleum
ether). IR (thin film): ν
= 3550, 1712, 1591 cm–1 1H NMR
.
˜
max
(400 MHz, CDCl3): δH = 7.69–7.66 (m, 4 H, Ph), 7.46–7.41 (m, 6
H, Ph), 3.93–3.86 (m, 1 H, 3-H), 3.87 (t, J = 12 Hz, 2 H, 8-H),
3.83 (s, 3 H, COOMe), 1.85–1.51 (m, 6 H, 3 CH2), 1.54 (s, 1.5 H,
2-CH3), 1.53 (s, 1.5 H, 2-CH3), 1.25 (s, 1.5 H, CH3), 1.23 (s, 1.5 H,
Methyl (2ЈE)-6Ј-Hydroperoxy-8Ј-(9,10-trans-1α-hydroxy-2α,5,5,9β-
tetramethyldecahydronaphthalen-1β-yl)-2Ј,6Ј-dimethyloct-2Ј-enoate
(7): A solution of methyl (2ЈE)-8Ј-(9,10-trans-1α-hydroxy-
2α,5,5,9β-tetramethyldecahydronaphthalen-1β-yl)-2Ј,6Ј-dimethyl-
6Ј-(triethylsilylperoxy)oct-2Ј-enoate (6; 42 mg, 0.078 mmol) and
pyridinium p-toluenesulfonate (2.1 mg, 0.0083 mmol) in absolute
ethanol (2 mL) was left to stir at room temperature for 6 h. The
solvents were removed in vacuo and the crude residue was purified
by flash column chromatography on silica, eluting with petroleum
ether/diethyl ether (2:1), to give the title compound as a colourless
oil (27 mg, 81%). Rf = 0.21 (20% diethyl ether/petroleum ether).
CH3), 1.04 (s, 9 H, tBu) ppm. 13C NMR (100 MHz, CDCl3): δC
=
172.1 (C=O), 135.6 (Ph), 132.6 (Ph), 130.0 (Ph), 127.9 (Ph), 72.2
(C-6), 62.7 and 62.6 (C-3), 61.7 (C-8), 57.8 (C-2), 52.6 (COOCH3),
41.6 (CH2) 41.3 (CH2), 38.5 (CH2), 38.4 (CH2), 26.8 (tBu), 26.6
and 26.3 (6-CH3), 22.8 (CH2), 19.0 (q), 13.5 (2-CH3) ppm. HRMS
(ESI): calcd. for C27H39O5Si [M + H]+· 471.2567; found 471.2548.
Methyl 2-{6-[2-(9,10-trans-1-Hydroxy-2α,5,5,9β-tetramethyldecahy-
dronaphthalen-1-yl)ethyl]-6-methyl-1,2-dioxan-3-yl}propanoate (12):
A solution of freshly distilled triethylamine (3.95 mL, 0.05 m in
methanol) was added dropwise to a solution of methyl (2ЈE)-6Ј-
IR (thin film): ν
= 3535, 3390, 2934, 2868, 1699, 1460, 1438,
˜
max
1
1283, 734 cm–1. H NMR (400 MHz, CDCl3): δH = 9.24 (s, 0.5 H,
OOH), 8.89 (s, 0.5 H, OOH), 6.81–6.74 (m, 1 H, 3Ј-H), 3.733 and hydroperoxy-8Ј-(9,10-trans-1α-hydroxy-2α,5,5,9β-tetramethyl-
3.729 (s, 3 H, COOCH3), 2.37–2.13 (m, 2 H, 4Ј-H), 1.89–1.11 (m, decahydronaphthalen-1β-yl)-2Ј,6Ј-dimethyloct-2Ј-enoate (7; 42 mg,
21 H, 2-, 3-, 4-, 6-, 7-, 8-, 10-H, 5Ј-, 7Ј-, 8’-H and 2Ј-CH3), 1.19
and 1.09 (s, 3 H, 6Ј-CH3), 0.95 (s, 3 H, 9-CH3), 0.90 (d, J = 6.7 Hz,
0.10 mmol) in methanol (1 mL) under nitrogen. The resulting solu-
tion was stirred at room temperature for 2 d. The reaction mixture
1.5 H, 2-CH3), 0.88 (s, 3 H, 5α-CH3), 0.87 (d, J = 6.4 Hz, 1.5 H, was subsequently concentrated in vacuo and the crude material
2-CH3), 0.83 (s, 3 H, 5β-CH3) ppm. 13C NMR (100 MHz, CDCl3): purified by flash column chromatography on silica. Elution with
δC = 168.8 and 168.7 (C=O), 142.7 and 142.4 (C-3Ј), 127.6 and
127.5 (C-2Ј), 84.3 and 84.1 (C-6Ј), 77.2 (C-1), 51.7 (COOCH3), 47.3
and 47.0 (C-10), 43.7 and 43.6 (C-9), 41.6 (C-6), 37.2 and 36.8 (C-
2), 36.0 and 35.1 (C-5Ј), 33.8 (5α-CH3), 33.5 (C-5), 32.4 and 31.9
petroleum ether/diethyl ether (2:1) gave a first fraction, pure com-
pound 12, obtained as a colourless oil (4.1 mg, 10%) as a mixture
of inseparable diastereoisomers. Further elution allowed the recov-
ery of the starting material (8.7 mg, 21%) and a second fraction,
(C-8), 31.7 and 31.6 (C-3), 31.4 (C-7Ј), 26.1 and 25.9 (C-8Ј), 23.1 pure compound 13, obtained as a colourless oil (17.9 mg, 43%) as
and 23.0 (C-4Ј), 22.0 (5β-CH3), 21.5 and 21.4 (C-4), 20.5 (6Ј-CH3), a mixture of inseparable diastereoisomers.
Eur. J. Org. Chem. 2012, 1209–1216
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1215

E. M. P. Silva, R. J. Pye, G. D. Brown, L. M. Harwood
FULL PAPER
wanishi, S. Itagaki, T. Horii, M. Kobayashi, Bioorg. Med.
Chem. Lett. 2004, 14, 3513–3516; d) M. Kawanishi, N. Kotoku,
S. Itagaki, T. Horii, M. Kobayashi, Bioorg. Med. Chem. Lett.
2004, 14, 5297–5307; e) C. Xu, J. M. Raible, P. H. Dussault,
Org. Lett. 2005, 7, 2509–2511; f) H.-H. Liu, H.-X. Jin, Q.
Zhang, Y.-K. Wu, H.-S. Kim, Y. Wataya, Chin. J. Chem. 2005,
23, 1469–1473; g) H.-H. Liu, Q. Zhang, H.-X. Jin, X. Shen,
Y.-K. Wu, Chin. J. Chem. 2006, 24, 1180–1189; h) H.-H. Liu,
H.-X. Jin, Y.-K. Wu, Chin. J. Chem. 2004, 22, 1029–1033; i)
H.-X. Jin, Q. Zhang, H.-S. Kim, Y. Wataya, H.-H. Liu, Y. Wu,
Tetrahedron 2006, 62, 7699–7711.
Methyl 2-{6-[2-(9,10-trans-1-Hydroxy-2α,5,5,9β-tetramethyldecahy-
dronaphthalen-1-yl)ethyl]-6-methyl-1,2-dioxan-3-yl}propanoate (12):
R = 0.83 (30% diethyl ether/petroleum ether). IR (thin film): ν
˜
f
max
= 2932, 2867, 1735, 1458, 1375, 1264, 1199, 1162, 909 cm–1. HRMS
(ESI): calcd. for C25H44O5Na [M + Na]+· 447.3081; found
447.3076. For the NMR spectroscopic data, see Table S1, Support-
ing Information.
Methyl 2-Hydroxy-2-{5-[2-(9,10-trans-1-hydroxy-2α,5,5,9β-tetra-
methyldecahydronaphthalen-1-yl)ethyl]-5-methyltetrahydrofuran-2-
yl}propanoate (13): Rf = 0.12 (30% diethyl ether/petroleum ether).
[11] P. M. O’Neill, N. L. Searle, K. J. Raynes, J. L. Maggs, S. A.
Ward, R. C. Storr, B. K. Park, G. H. Posner, Tetrahedron Lett.
1998, 39, 6065–6068.
[12] J. R. Harris, S. R. Waetzig, K. A. Woerpel, Org. Lett. 2009, 11,
3290–3293.
IR (thin film): ν
= 3435, 2932, 2867, 1730, 1460, 1372, 1261,
˜
max
1179, 1060, 977 cm–1. HRMS (ESI): calcd. for C25H44O5Na [M +
Na]+· 447.3081; found 447.3076. For the NMR spectroscopic data,
see Table S2, Supporting Information.
[13] E. M. P. Silva, R. J. Pye, C. Cardin, L. M. Harwood, Synlett
2010, 509–513.
[14] I. Smonou, S. Khan, C. S. Foote, Y. Elemes, I. M. Mavridis, A.
Pantidou, M. Orfanopoulos, J. Am. Chem. Soc. 1995, 117,
7081–7087.
Supporting Information (see footnote on the first page of this arti-
cle): NMR spectroscopic data for compounds 12a–d and 13a–d.
Acknowledgments
[15] The epoxidation of electron-deficient olefins by using hydrogen
peroxide in a strong alkaline solution is generally known as the
Weitz–Scheffer epoxidation. The most important modification
of the asymmetric Weitz–Scheffer epoxidation has been devel-
oped by using optically active secondary hydroperoxide as the
oxidation reagent in the presence of either KOH or an organic
base. This reaction has important applications for the epoxid-
ation of electron-poor olefins (E. Weitz, A. Scheffer, Ber. Dtsch.
Chem. Ges. 1921, 54, 2327–2344). For recent reviews of asym-
metric epoxidation, see: a) H. Adolfsson, Y. Shi in Modern
Oxidation Methods (Ed.: J.-E. Bäckvall), Wiley-VCH,
Weinheim, Germany, 2004, pp. 21; b) K. Matsumoto, T. Kat-
suki, in: Catalytic Asymmetric Synthesis, 3rd ed. (Ed.: I.
Ojima), Wiley-VCH, New York, 2010, p. 839; c) B. S. Lane, K.
Burgess, Chem. Rev. 2003, 103, 2457–2474; d) D. Díez, M. G.
Núñez, A. B. Antón, P. García, R. F. Moro, N. M. Garrido,
I. S. Marcos, P. Basabe, J. G. Urones, Curr. Org. Synth. 2008,
5, 186–216; e) K. Matsumoto, Y. Sawada, T. Katsuki, Pure
Appl. Chem. 2008, 80, 1071–1077; f) O. A. Wong, Y. Shi, Chem.
Rev. 2008, 108, 3958–3987.
E. M. P. S. is grateful to the Portuguese Foundation for Science and
Technology (grant number SFRH/BPD/66961/2009) for a post-
doctoral grant. We are also grateful to Dr. John Mckendrick, Uni-
versity of Reading, for useful discussions.
[1] J. Tanaka, T. Higa, K. Suwanborirux, U. Kokpol, G. Bernardi-
nelli, C. W. Jefford, J. Org. Chem. 1993, 58, 2999–3002.
[2] R. J. Capon, S. J. Rochfort, S. P. B. Ovenden, J. Nat. Prod.
1997, 60, 1261–1264.
[3] R. J. Capon, S. J. Rochfort, S. P. B. Ovenden, R. P. Metzger, J.
Nat. Prod. 1998, 61, 525–528.
[4] P. Phuwapraisirisan, S. Matsunaga, N. Fusetani, N. Chaitana-
wisuti, S. Kritsanapuntu, P. Menasveta, J. Nat. Prod. 2003, 66,
289–291.
[5] R. J. Capon, J. Nat. Prod. 1991, 54, 190–195.
[6] a) L. M. Harwood, J. Robertson, S. Swallow, Synlett 1999,
1359–1362; b) L. M. Harwood, J. Robertson, S. Swallow, Syn-
lett 1999, 185–188; c) L. M. Harwood, A. C. Brickwood, V.
Morrison, J. Robertson, S. Swallow, J. Heterocycl. Chem. 1999,
36, 1391–1408.
[16] M. D. Higgs, D. J. Faulkner, J. Org. Chem. 1978, 43, 3454–
3457.
[17] R. J. Capon, J. K. MacLeod, Tetrahedron 1985, 41, 3391–3404.
[18] a) R. J. Capon, J. K. MacLeod, A. C. Willis, J. Org. Chem.
1987, 52, 339–342; b) S. P. B. Ovenden, R. J. Capon, Aust. J.
Chem. 1998, 51, 573–579.
[7] a) S. P. B. Ovenden, R. J. Capon, J. Nat. Prod. 1999, 62, 214–
218; b) R. J. Capon, Eur. J. Org. Chem. 2001, 633–645.
[8] P. A. Bartlett, C. Chapuis, J. Org. Chem. 1986, 51, 2799–2806.
[9] F. M. Guerra, E. Zubía, M. J. Ortega, F. J. Moreno-Dorado,
G. M. Massanet, Tetrahedron 2010, 66, 157–163.
[19] W. C. Still, M. Kahn, A. Mitra, J. Org. Chem. 1978, 43, 2923–
2925.
[10] a) N. Murakami, M. Kawanishi, S. Itagaki, T. Horii, M. Ko-
bayashi, Tetrahedron Lett. 2001, 42, 7281–7285; b) N. Murak-
ami, M. Kawanishi, S. Itagaki, T. Horii, M. Kobayashi, Bioorg.
Med. Chem. Lett. 2002, 12, 69–72; c) N. Murakami, M. Ka-
[20] D. D. Perrin, W. L. F. Armarego, in: Purification of Laboratory
Chemicals, 3rd ed., Pergamon Press, Oxford, 1988.
Received: October 11, 2011
Published Online: December 27, 2011
1216
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 1209–1216