Synthesis of the originally proposed structures of elatenyne and an enyne from Laurencia majuscula

Article abstract of DOI:10.1039/b814953d

A bidirectional synthesis of the originally proposed structures for the natural products elatenyne and a chloroenyne from Laurencia majuscula is described along with a reassessment of the structures of the halogenated enynes based upon a ¹³C NM

Full text of DOI:10.1039/b814953d

View Article Online / Journal Homepage / Table of Contents for this issue
www.rsc.org/obc
Volume 7 | Number 2 | 21 January 2009 | Pages 205–400
ISSN 1477-0520
FULL PAPER
2WT\XRP[ꢀBRXT]RT
Jonathan W. Burton et al.
Synthesis of the originally proposed
structures of elatenyne and an enyne
from Laurencia majuscula
In this issue...
1477-0520(2009)7:2;1-D

PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Synthesis of the originally proposed structures of elatenyne and an enyne from
Laurencia majuscula†‡
a
Helen M. Sheldrake,§a Craig Jamieson,b Sofia I. Pascu¶ꢀ and Jonathan W. Burton*a,c
Received 27th August 2008, Accepted 23rd October 2008
First published as an Advance Article on the web 20th November 2008
DOI: 10.1039/b814953d
A bidirectional synthesis of the originally proposed structures for the natural products elatenyne and a
chloroenyne from Laurencia majuscula is described along with a reassessment of the structures of the
halogenated enynes based upon a 13C NMR chemical shift/structure correlation.
recently the structure of a halogenated C15 natural product isolated
from L. majuscula was disclosed as the pyrano[3,2-b]pyran 4 again
on the basis of extensive NMR spectroscopic analysis and by
Introduction
Until the advent of modern spectroscopic techniques in the
later part of the 20th century, the structure determination of
natural products was a time-consuming process that involved
comparison with the structure 3 reported for elatenyne and with
(E)-dactomelyne [(E)-1].6
painstaking degradation and derivatisation of gram quantities of
a natural product to provide structure information followed by
total synthesis for structure confirmation. The development of
a myriad of spectroscopic techniques (most notably NMR) now
allows the structures of complex natural products to be determined
routinely. Nevertheless, unambiguous structure assignment by
NMR methods alone is not always straightforward especially
in closely related molecules and total synthesis frequently still
plays an important role in structure confirmation.1 In particular
through hetero-atom connectivity can still be a challenge to solve
by NMR methods. For example, the two natural products (Z)-
dactomelyne2 (Z)-1 and notoryne 23 are constitutional isomers
which both contain the same carbon and proton connectivity and
Elatenyne and the L. majuscula enyne were attractive targets
for total synthesis due to their unknown biological activity,
hence unambiguous structure assignment would be challenging
on the basis of NMR experiments alone.
densely functionalised pyrano[3,2-b]pyran cores and embed-
The natural products 1 and 2 belong to a much wider group
ded C2-symmetry.7,8 We have previously demonstrated that the
of C15 metabolites isolated from red algae and those marine
pyrano[3,2-b]pyran structures 3 and 4, originally assigned to the
natural products are incorrect, and proposed that the actual
structures of the natural products are related to notoryne in having
a core 2,2¢-bifuranyl.9,10 Herein we report the full details of the
total synthesis of the halogenated pyrano[3,2-b]pyrans 3 and 4.
At the outset of the project we were unaware that the structures
originally assigned to elatenyne and the chlorinated enyne from
L. majuscula were incorrect. During these synthetic studies we
uncovered a 13C NMR chemical shift correlation which allows
cis-fused pyrano[3,2-b]pyrans and 2,2¢-bifuranyls, such as (Z)-
dactomelyne [(Z)-1] and notoryne 2 to be readily distinguished
and which ultimately led us to reassign the structures of elatenyne
and the chloroenyne from L. majuscula.
organisms which feed on Laurencia species.4 In 1986 the bis-
brominated natural product elatenyne was isolated by Hall and
Reiss and was assigned a pyrano[3,2-b]pyran structure 3 on the
basis of extensive 1H and 13C NMR spectroscopic analysis.5 More
aDepartment of Chemistry, University of Cambridge, Lensfield Road,
Cambridge, UK CB2 1EW
bSchering-Plough Corporation, Newhouse, Lanarkshire, UK ML1 5SH
cDepartment of Chemistry, University of Oxford, Chemistry Research Labo-
ratory, Mansfield Road, Oxford, UK OX1 3TA. E-mail: jonathan.burton@
chem.ox.ac.uk; Tel: +44 1865 285 119
† Dedicated to Professor Andrew B. Holmes on the occasion of his 65th
birthday, with respect and admiration.
‡ Electronic supplementary information (ESI) available: Experimental
procedures for the preparation of a number of compounds including 8, 16,
26, 29, 31–34, 37–46, 55, 58, 59, 68, 76–78, and 1H NMR and 13C NMR
spectra of compounds 3, 4, 9, 14, 15, 18, 19, 38, 39, 53, 56, 57, 60–66, 70–75
and 81–86. CCDC reference number 698760. For ESI and crystallographic
data in CIF or other electronic format see DOI: 10.1039/b814953d
Retrosynthetic analysis
We aimed to utilise a two-directional synthesis of the two targets
3 and 4.11 Thus, we envisaged that both halogenated pyrano[3,2-
b]pyrans would be synthesised from the C2-symmetric tetrol 5
(Fig. 1). Having had previous experience with the intramolecular
hydrosilation of exo-cyclic enol ethers,12,13 we postulated that the
tetrol 5 would be available by intramolecular hydrosilation of
the appropriately functionalised bis-exo-cyclic enol ether derived
§ Current Address: Institute of Cancer Therapeutics, School of Life
Sciences, University of Bradford, West Yorkshire, UK BD7 1DP.
¶ Author to whom correspondence regarding the X-ray crystallography
should be addressed.
ꢀ Current Address: Department of Chemistry, University of Bath, Bath,
UK BA2 7AY.
238 | Org. Biomol. Chem., 2009, 7, 238–252
This journal is
The Royal Society of Chemistry 2009
©

Fig. 1 Retrosynthetic analysis of the pyrano[3,2-b]pyrans 3 and 4.
from 6; an unstudied reaction with six-membered exo-cyclic enol
anomeric acetates 14 as a mixture of 3 diastereomers. Treatment
of the anomeric acetates 14 with methanolic hydrochloric acid
at room temperature gave solely the three diastereomeric 2,2¢-
bifuranyls 15. Use of methanolic hydrochloric acid at reflux
provided a mixture of the two diastereomeric pyrano[3,2-b]pyrans
9 along with the 2,2¢-bifuranyls 15. The pyrano[3,2-b]pyrans 9 are
present as a ca. 1.3 : 1 mixture of diastereomers (ca. 83% of the
mixture at equilibrium) along with the 2,2¢-bifuranyls 15 (ca. 17%
of the mixture at equilibrium).21–23 The five isomeric acetals could
be separated by careful chromatography. The structures of the
pyrano[3,2-b]pyrans 9 were tentatively assigned as follows. Due
to reduced torsional strain it was expected that the pyrano[3,2-
b]pyrans 9 would be thermodynamically more stable than the 2,2¢-
bifuranyls 15 and hence the major products in the equilibrium
mixture should be the pyrano[3,2-b]pyrans 9. Furthermore, the
pyrano[3,2-b]pyrans 9 were readily oxidised to the bis-d-lactone 8
(vide infra) which was spectroscopically distinct from the known
bis-g-lactone 13.18
ethers.14 Alternatively the known hydroboration of a-oxygenated
exo-cyclic enol ethers could also be used to convert 6 into the
desired tetrol 5.15 The diol 6 would be prepared from the dihydroxy
bis-d-lactone 7 which would, in turn, be prepared by oxidation
of the bis-d-lactone 8. To the best of our knowledge the bis-d-
lactone 8 has not been reported in the literature. The preparation
of this bis-lactone would not be possible by cyclisation of an open
chain intermediate diacid or diester (11) as under both kinetic and
thermodynamic conditions this would give rise to the correspond-
ing known bis-g-lactone 13 (Scheme 1).16 We therefore aimed to
synthesise the bis-d-lactone 8 by oxidation of the corresponding
bis-methyl acetal 9.17 The bis-acetal 9 would be prepared from
the corresponding dialdehyde 10 under equilibrating conditions
with acid catalysis. Ultimately the dialdehyde would be syn-
thesised from the commercially available tartrate-derived aceto-
nide 12.
Having established a procedure for formation of the cis-fused
pyrano[3,2-b]pyran 9 we required a method for oxidation of
the methyl acetals directly to the corresponding bis-d-lactone
8. Following Grieco’s procedure,17 exposure of a mixture of the
acetals 9 to mCPBA and BF3·OEt2 followed by aqueous workup
gave the bis-peroxyester 16 which on treatment with a solid
supported guanidine base24 gave the desired bis-d-lactone 8 as a
white crystalline solid in good yield (Scheme 2). The IR spectrum
of the bis-d-lactone was indicative of a 6-ring lactone (nmax
1740 cm-1) and was significantly different from the IR spectrum of
the bis-g-lactone 13 (nmax 1771 cm-1) thus confirming its structure.
Disappointingly, it proved impossible to covert the bis-d-lactone 8
Results and discussion
First generation route to the bis-exo-cyclic enol ether 6
Given the imbedded C2-symmetry within the halogenated
pyrano[3,2-b]pyrans 3 and 4 we restricted our synthetic plans
to those which would result in a bidirectional synthesis of the
target molecules starting from tartaric acid. The tartrate acetonide
12 was readily converted into the known bis-lactone 1318 (nmax
1771 cm-1) by an efficient 4 step procedure19 (Scheme 1). Following
the method of Rychnovsky,20 the bis-lactone 13 was reduced with
DIBAL and the intermediate alkoxides were acylated to give the
Scheme 1 Formation of the acetals 9 and 15. Reagents and conditions: (a) DIBAL, toluene, -78 ◦C, then (carbethoxymethylene)triphenylphosphorane,
MeOH, -78 ◦C → RT, 83%; (b) H2, Pd/C, EtOH, 91%; (c) TFA, water, 100%; (d) DIBAL, CH2Cl2, -78 ◦C, then Ac2O, pyridine, DMAP, CH2Cl2,
-78 → -20 ◦C, 83%; (e) MeOH, HCl, 90%.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 238–252 | 239
©

Scheme 3 Synthesis of the bis-enol ether 19. Reagents and conditions:
(a) NaI, TMSCl, MeCN, then hexamethyldisilazane.
Scheme 2 Synthesis of the bis-d-lactone 8. Reagents and conditions:
˚
(a) mCPBA, BF·OEt2,
4
A
sieves, CH2Cl2; (b) 1,5,7-triazabicy-
clo[4.4.0]dec-5-ene on polystyrene.
enol ether was confirmed by X-ray crystallographic analysis of a
later intermediate, the bis-epoxide 18.
into the corresponding hydroxylated bis-lactone 7, or to form the
enolate of 8 without decomposition.
We subsequently discovered that subjection of any of the acetals
9, 14 or 15 to the above reaction conditions gave the desired enol
ether 19 as the sole product although the yield from the anomeric
acetates 14 was considerably lower than from the acetals 9 or 15.
We postulate that the novel rearrangement of the 2,2¢-bifuranyl
acetals 15 to give the pyrano[3,2-b]pyran enol ether 19 proceeds
as follows (Fig. 3). Silylation of the most sterically accessible
oxygen lone pair in 15 occurs first which leads to the oxocarbenium
ion 20. The oxocarbenium ion 20 is trapped by the lone pair of
the oxygen atom on the adjacent THF ring to give the tricyclic
oxonium ion 21, which fragments to give a second oxocarbenium
ion 22. This oxocarbenium ion is then captured by another oxygen
atom lone pair to form the second tricyclic oxonium ion 23
which fragments to give the pyrano[3,2-b]-pyran system 24. The
resulting oxocarbenium ion can then be readily converted into
the bis-anomeric iodide (24→26) which, on addition of base,
gives the bis-enol ether 19. We have briefly studied this reaction
Second generation route to the bis-exo-cyclic enol ether 6
Having been unable to oxidise the enolate derived from the bis-d-
lactone 8 we decided to approach the a,a¢-dihydroxy bis-d-lactone
7 by installation of the desired hydroxy groups prior to bis-lactone
formation.
Thus, the a,a¢-dihydroxy bis-d-lactone 7 would be synthesised
from the corresponding anomeric peracid ester 17 which, in turn,
would be available by opening of the bis-epoxide 18 with mCPBA
(Fig. 2). The bis-epoxide 18 would be prepared by epoxidation
of endo-cyclic bis-enol ether 19. We envisaged that the bis-enol
ether would be formed by the elimination of two equivalents of
methanol from the methoxy acetals 9. Exposure of the pyrano[3,2-
b]pyran 9a to a large excess of iodotrimethylsilane followed by
the addition of hexamethyldisilazane gave the bis-enol ether 19 in
quantitative yield (Scheme 3)25,26 which was used in the subsequent
reactions without further purification. The structure of the bis-
1
1
by H NMR in d3-MeCN. H NMR analysis of a solution of
the pyrano[3,2-b]pyran 9a in d3-MeCN immediately after the
addition of iodotrimethylsilane shows the presence of a species
we assigned to the bis-anomeric iodide 26. Addition of HMDS to
the above solution immediately results in the exclusive formation
of the bis-enol ether 19 by 1H NMR analysis. Exposure of one of
the C2-symmetric 2,2¢-bifuranyl acetals 15 to iodotrimethylsilane
1
followed by H NMR analysis indicated the presence of the bis-
anomeric iodide 26 and a second species which we assigned to
the corresponding 2,2¢-bifuranyl bis-anomeric iodide 29 (ca. 1 :
1 mixture of 26 and 29).27 Over many minutes the 2,2¢-bifuranyl
bis-anomeric iodides 29 were converted into the corresponding
pyrano[3,2-b]pyran bis-anomeric iodides 26 presumably by way
of the anomeric iodide 27 or equivalent intermediate. Exposure
of the anomeric acetates 13 to the same reaction conditions
followed by 1H NMR gave the 2,2¢-bifuranyl anomeric iodides 29
which slowly converted into the corresponding pyrano[3,2-b]pyran
Fig. 2 Retrosynthesis of the a,a¢-dihydroxy bis-d-lactone 7.
Fig. 3 Proposed mechanism for the formation of the bis-enol ether 19.
240 | Org. Biomol. Chem., 2009, 7, 238–252 This journal is The Royal Society of Chemistry 2009
©

bis-anomeric iodides 26 over a number of hours. In a separate
experiment, treatment of a solution of the bis-anomeric acetates
13 in toluene with TMSI followed by the addition of HMDS
gave the known 2,2¢-bifuranyl bis-enol ether 30.28 These results are
in accord with the proposed mechanism. The 2,2¢-bifuranyl bis-
methyl acetals 15 are rapidly converted into a mixture of the bis-
anomeric iodides 26 and 29. The 2,2¢-bifuranyl anomeric iodides
then rearrange to the pyrano[3,2-b]pyran anomeric iodides 26 as
shown in Fig. 3. This rearrangement involves a number of charged
intermediates and therefore proceeds readily in acetonitrile. With
the bis-anomeric acetates 13 the conversion of the 2,2¢-bifuranyl
anomeric iodides 29 into corresponding pyrano[3,2-b]pyran 26
is slower than with the bis-methyl acetals 15. This is probably
due to a methoxy group being a better electron donor than an
acetoxy group and hence intermediates such as 22 are formed more
rapidly when R = Me than when R = Ac. Hence in toluene the
rearrangement of the bis-anomeric acetates 13 is far slower and as
a result the 2,2¢-bifuranyl bis-enol ether 30 is the ultimate product.
The driving force for the rearrangement of the acetals 15 to give
the pyrano[3,2-b]pyran 19 may arise from the release of torsional
strain in moving from a 2,2-bifuranyl to a pyrano[3,2-b]pyran.
Having developed an efficient synthesis of the bis-enol ether 19
from a mixture of the acetals 9 and 15, we aimed directly to form
the corresponding anomeric peracid esters 17 by treatment of the
bis-enol ether 19 with excess mCPBA. In practice, this strategy
was not effective. However, epoxidation of the bis-enol ether 19
with mCPBA in CH2Cl2 and methanol29 delivered the bis-methyl
acetals 31 as an inseparable mixture of anomers (Scheme 4).30
The free hydroxy groups in the acetals 31 were protected as
benzyl ethers (32) to avoid water solubility issues which we had
encountered with the lactone 8. Disappointingly, the oxidation
of the bis-methyl acetals 32 under Grieco’s conditions17 failed
to deliver any of the desired bis-d-lactone. Similarly attempted
oxidation of the anomeric acetates 3331 under the same condi-
tions, following precedent from the work of Hoppe,32 was also
unsuccessful resulting in substrate decomposition. We therefore
sought to isolate the pure bis-epoxide 18 before exploring further
routes towards 7.
The bis-epoxide 18 was readily synthesised by treatment of
the bis-enol ether 19 with dimethyldioxirane in acetone,33,34 and
was isolated as a white crystalline solid which was characterised
crystallographically.9 Disappointingly, attempted formation of the
a-hydroxy anomeric peroxyesters 17 by opening of the bis-epoxide
18 with mCPBA was not successful. We attempted to convert
the epoxide into the desired bis-d-lactone 7 from the correspond-
ing anomeric sulfides by Pummerer rearrangement35 which was
also unsuccessful (see ESI for substrate preparation‡). We also
attempted to open the bis-epoxide 18 with iodomethyllithium36 or
dimethylsulfonium methylide37 which would have given us direct
access to the exo-cyclic bis-enol ether 6, but again these reactions
were unsuccessful.
Third generation route to the bis-exo-cyclic enol ether 6
Due to the failure of the epoxide-opening reactions and subse-
quent synthetic manipulations, we turned our attention to the
direct functionalisation of enol ethers. We planned to convert the
bis-endo cyclic enol ether 19 into the hydroxymethyl substituted
bis-enol ether 35 which could be converted into the desired
intermediate 6 by Evans–Mislow rearrangement of the corre-
sponding sulfoxides 36, or undergo intramolecular hydrosilation
or hydroboration itself (Fig. 4).38
In the event, this plan was unsuccessful; attempted metallation
of the enol ether 19 with tBuLi resulted in decomposition of
the substrate. Ley has shown that anomeric sulfones undergo
lithiation at the anomeric position and react with a wide variety
of electrophiles.39,40 Furthermore, in many cases, spontaneous
elimination of benzenesulfinic acid occurs to give a functionalised
endo-cyclic enol ether (such as 35).40 We therefore investigated
this methodology for the synthesis of 35. Exposure of the bis-
endo-cyclic enol ether 19 to freshly prepared benzenesulfinic
acid39–41 gave the desired anomeric bis-sulfone 37a in poor yield
Scheme 4 Elaboration of the enol ether 19. Reagents and conditions: (a) mCPBA, MeOH, CH2Cl2, 74%; (b) NaH, BnBr, DMF, 39–54%; (c) PhI(OAc)2,
BF3·OEt2, CH2Cl2 then Et3N, 33 15%, 34 19%; (d) DMDO, acetone, NaHCO3, 98%.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 238–252 | 241
©

Fig. 4 Retrosynthesis of the bis-enol ether 6.
˚
Scheme 5 Synthesis of anomeric sulfones. Reagents and conditions: (a) PhSO2H, 4 A molecular sieves, CH2Cl2, 4–10%; (b) PhSO2H, CaCl2, CH2Cl2, 38
27%, 39 10%; (c) PhSH, BF3·OEt2, CH2Cl2, 87%; (d) mCPBA, NaHCO3, EtOAc, 100%.
(Scheme 5).30 Attempted preparation of the bis-sulfones 37 from
the corresponding bis-methyl acetals 9 using benzenesulfinic acid
and calcium chloride39,40 gave the 2,2¢-bifuranyl anomeric sulfones
39 (mixture of 3 diastereomers) and the pyrano[3,2-b]pyran 38.
Ultimately we found that the anomeric sulfones 37 could be
prepared in good yield from the bis-methyl acetals 9 by way of
the corresponding anomeric sulfides 40. Thus, exposure of the
bis-methylacetals 9 to thiophenol in the presence of a Lewis acid
delivered the anomeric sulfides 40 as an inseparable 2 : 1 mixture of
diastereomers in good yield.30 The anomeric sulfides were readily
oxidised to the corresponding inseparable mixture of anomeric
Fig. 5 Retrosynthetic analysis of the bis-enol ether 6.
sulfones 37.
Yet again we were frustrated by our inability to convert the
anomeric sulfones 37 into the bis-enol ether 35. Attempted
lithiation of the sulfones with BuLi or LDA followed by addition
of trioxane failed to give the desired product. Use of D2O as the
electrophile did not result in any deuterium incorporation. In all
of the attempted lithiations, only varying levels of decomposition
of the starting material were observed.
Fourth generation route to the bis-exo-cyclic enol ether 6
Our final approach to the hydrosilation substrate is illustrated in
Fig. 5. Thus, we proposed to synthesise the desired bis-exo-cyclic
enol ether 6 by rearrangement of the bis-epoxide 41. The bis-
epoxide 41 would be made from the bis-endo-cyclic enol ether 42
which we proposed to synthesise by elimination of two equivalents
of methanol from the bis-methyl-acetal 43, analogous to the
preparation of the bis-enol-ether 19.
Scheme 6 Synthesis of the bis-acetal 43. Reagents and conditions:
(a) DIBAL, toluene, -78 ◦C, then (acetylmethylene)triphenylphosphorane,
MeOH, -78 ◦C → RT, 80%; (b) H2, Pd/C, EtOH, 99%; (c) BF3·OEt2,
CH2Cl2, MeOH, water, 43, 50%.
After screening a wide range of protic and Lewis acids, we found
that stirring the diketone 44 with BF3·OEt2 in methanol gave
the desired pyrano[3,2-b]pyran 43 identical with the previously
prepared racemic sample.22 Also isolated from the reaction mixture
The diketone 44 required for the synthesis of the acetal 43
was readily prepared from the tartrate acetonide 12 (Scheme 6).
242 | Org. Biomol. Chem., 2009, 7, 238–252
This journal is
The Royal Society of Chemistry 2009
©

was a single diastereomer of the 2,2¢-bifuranyl 45 which rapidly
decomposed; the dioxabicyclic[2.2.1]heptane 46 was formed in
varying amounts when other acid catalysts were used. The next
step in the proposed synthesis of the bis-enol ether 42 required the
elimination of two equivalents of methanol from the bis-methyl
acetal 43.
Disappointingly, the conditions used for the formation of the
bis-enol ether 19 from the bis-acetal 9 gave the desired enol
ether in reasonable yield (50%) but contaminated with a number
of inseparable impurities. A range of reaction conditions were
screened; however, many of these resulted in formation of signif-
icant quantities of the bicyclic ketone 46. Ultimately we found
that exposure of the bis-methyl acetal 43 to bromotrimethylsilane
followed by addition of DBU gave the desired bis-enol ether 42
in 55% yield (Scheme 7). The bis-epoxide 41 was readily formed
from the bis-enol ether 42 on exposure to dimethyl-dioxirane in
dichloromethane.42
for the synthesis of complex natural products,11 it has so far
proved unsuccessful in our case. This may be in part due to the
bowl shaped conformation of the cis-fused pyrano[3,2-b]pyran
intermediates which can result in the reaction on one side of the
molecule having considerable influence on the reactivity of the
opposite side of the molecule. Indeed, it might well have proved
possible to synthesise the bis-enol ether 6 if a two-directional
approach had not been used.
Structure determination
Our failure to synthesise the bis-enol ether 6 was most disap-
pointing; however, this failure had resulted in the synthesis of a
large number of 2,2¢-bifuranyls and cis-fused pyrano[3,2-b]pyrans.
Careful analysis of all of these compounds revealed that the 13C
NMR chemical shifts of the central oxygen-bearing carbons fell
into two distinct groups: for the pyrano[3,2-b]pyrans, the 13C NMR
chemical shifts of the central oxygen-bearing carbons C-8a and C-
4a resonate at less then d = 76 ppm, whereas the corresponding
carbon atoms in the 2,2¢-bifuranyls (C-2 and C-2¢) resonate at
greater than d = 76 ppm. We were alerted to this chemical shift
pattern by the vastly different chemical shifts of the central oxygen
bearing carbons in the various anomeric sulfones (Fig. 6). In
this paper we report the synthesis of a large number of cis-fused
pyrano[3,2-b]pyrans and 2,2¢-bifuranyls and >98% of these fit this
pattern.
Scheme
7
Synthesis of the epoxide 41. Reagents and conditions:
(a) TMSBr, DBU, 55%; (b) DMDO, CH2Cl2, 100%.
There is considerable precedent for the rearrangement of
epoxides to give allylic alcohols, including those with an exo-
cyclic olefin; however, we again were thwarted in our attempts
to synthesise the bis-enol ether 6 from the bis-epoxide 41.
Exposure to the epoxide to a wide range of reagents and
conditions [Al(OiPr)3 in toluene;43 Al2O3,44 TMSBr/DBU;45,46
LDA;47 Li/H2NCH2CH2NH2;47 MeMgNCyiPr;48 PhSeH, oxida-
tive workup;49 KOtBu] did not give any of the desired product.
We had invested considerable synthetic effort in trying to
make the bis-exo-cyclic enol ether 6 precursor for the proposed
hydroboration or intramolecular hydrosilation reaction to give the
tetrol 5. All of the routes which we investigated towards 6 involved
two-directional synthesis. While this can be a very efficient strategy
In order to be able to put forward such a chemical shift corre-
lation, it is imperative that the structures of all of the pyrano[3,2-
b]pyrans and 2,2¢-bifuranyls have been assigned correctly. Hoff-
mann has used 1H NMR to investigate the conformation of
the cis-fused pyrano[3,2-b]pyran skeleton 4750 and the related
tetraoxadecalin 48 (TOD) system51 which has also been extensively
studied by Fuchs.52–56 cis-Fused pyrano[3,2-b]-pyrans and TODs
are conformationally flexible and may exist in the O-proximal or
1
O-distal conformations (Fig. 7). The H NMR vicinal coupling
constants J8ax,8a = J4ax,4a are characteristically large in the O-distal
conformer with the corresponding coupling constants being small
in the O-proximal conformer (J8eq,8a = J4eq,4a); in the analysis of
equilibrating mixtures of conformers in solution, reference values
Fig. 6 13C NMR chemical shifts for 2,2¢-bifuranyls and pyrano[3,2-b]pyrans.
The Royal Society of Chemistry 2009 Org. Biomol. Chem., 2009, 7, 238–252 | 243
This journal is
©

Fig. 7 Conformations of the cis-fused pyrano[3,2-b]pyran 47 and the
tetraoxadecalin 48.
Fig. 9 X-Ray crystal structure of the anomeric sulfone 38 showing two
molecules in the unit cell (50% probability ellipsoids).
of 1.2 Hz (O-proximal) and 10.6 Hz (O-distal) have been used
for these coupling constants in TODs.51,57 In the unfunctionalised
pyrano[3,2-b]pyran 47, H-4a and H-8a appear in the H NMR
in the O-proximal chair–chair conformation and the anomeric
substituents axial as required for maximum anomeric stabilisation.
The 2,2¢-bifuranyls were characterised by the 13C NMR chemi-
cal shift of C-2 and C-2¢ being >d = 76 ppm. The 2,2¢-bifuranyls
showed a larger 1H–1H coupling constant between the H-2 and H-
2¢ and one of their vicinal neighbours. Furthermore, these protons
1
spectrum as a narrow triplet (J 2.9 Hz) due to small axial–axial
and axial–equatorial couplings to H-4ax and H-4eq,50 a feature that
was characteristic of the pyrano[3,2-b]pyrans synthesised in this
work.50,58 Furthermore, in TODs which exist in the O-proximal
conformation 48-O-prox, J4a,8a is ~1.6 Hz,53 but in the O-distal
conformation, J4a,8a is ~6 Hz.51 The 13C shift of C-4a and C-8a
in a wide range of TODs has been reported to fall in the range
69–70 ppm.52,53
1
appeared as a well defined multiplet in the H NMR spectrum.
Additionally, the H-2/2¢ multiplets were typically more complex
and wider than the corresponding mutliplets in the pyrano[3,2-
b]pyrans. For example, in the 13C NMR of the sulfone 39a C-2 and
The characteristic 1H NMR coupling constants (small J4,4a
,
1
C-2¢ resonated at d = 85.7 and d = 84.6 ppm whereas in the H
J8,8a and J4a,8a, and large J3,4ax) along with 13C NMR chemical
shifts, coupled with X-ray crystallographic analysis of certain
intermediates allowed the confident assignment of the structure
and conformation of the pyrano[3,2-b]pyrans described in this
paper. Furthermore, as alluded to above, the line shape of H-
4a/8a was highly indicative of a pyrano[3,2-b]pyran and became
a useful structure assignment tool. For example, in the sulfone 38,
H-8a and H-4a were narrow triplets with the typical line shape of a
pyrano[3,2-b]pyran and ~3 Hz coupling to their vicinal neighbours
H-8 and H-4 respectively indicating 38 was a pyrano[3,2-b]pyran
predominantly in the O-proximal conformation (Fig. 8); J4a,8a
was too small to be resolved. The anomeric protons H-2 and
H-6 appeared as doublets, indicating they were in equatorial
positions, coupling to one of the vicinal protons being too small
to be resolved. In the 13C NMR C-4a and C-8a resonated at
(interchangeably) d = 67.4 and d = 62.1 ppm which also suggested
a pyrano[3,2-b]pyran structure. The X-ray crystal structure of
the sulfone 3859–61 (Fig. 9) obtained subsequently completely
confirmed the NMR structural assignment. The asymmetric
unit contains two molecules, both with pyrano[3,2-b]pyran rings
NMR spectrum, the corresponding protons had large couplings
between them and to their vicinal neighbours (J2,2¢ 7.0 Hz, J2,3
7.0 Hz, J2¢,3¢ 9.1 Hz).
Reassessment of the structures of elatenyne and the chloroenyne
from L. majuscula
The 13C NMR chemical shifts of the central oxygen-bearing
carbons in elatenyne5 and the chloroenyne from L. majuscula6
resonate at d = 79.5 and 80 ppm, and d = 77.9 and 79.2
respectively, outside the range for a pyrano[3,2-b]pyran. This
initial discrepancy of the 13C NMR chemical shifts alerted us
to the possibility that the structures of elatenyne and the L.
majuscula enyne had been incorrectly assigned. Comparison of
1
the H NMR coupling constants of both the L. majuscula enyne
and elatenyne and closely related derivatives,5 with that of (E)-
and (Z)-dactomelyne led further weight to this proposal (Fig. 10).
In particular, comparison of the 1H NMR coupling constants of
the axial chlorine-containing pyran ring in (E)-dactomelyne with
the corresponding protons in both elatenyne and the L. majuscula
Fig. 8 Structure assignment of the anomeric sulfones by 1H NMR.
244 | Org. Biomol. Chem., 2009, 7, 238–252 This journal is The Royal Society of Chemistry 2009
©

Fig. 10 Comparison of dactomelyne with the L. majuscula enyne and elatenyne (natural product numbering). *Coupling constants are a combination
of those for elatenyne and from closely related derivatives.5
enyne showed considerable differences in the magnitude of the
vicinal couplings.
(Fig. 12). There are striking similarities between the 13C NMR
spectra of elatenyne and the dibromoenyne 52 and we propose
that it is possible that elatenyne is the double bond isomer of the
enyne 52.
The 13C NMR chemical shift pattern we had uncovered and the
1H NMR coupling constants of H-9 and H-10 (natural product
numbering corresponding to H-4a and H-8a in a pyrano[3,2-b]
pyran, and H-2 and H-2¢ in a 2,2¢-bifuranyl) led us to believe that
the correct structures of elatenyne and the chloroenyne from L.
majuscula were the 2,2¢-bifuranyls 50 and 51 respectively (Fig. 11)
related to the natural product notoryne 2.
Notoryne 23 was isolated by Suzuki and co-workers and was
shown to have a 2,2¢-bifuranyl skeleton by chemical correlation
and analysis of fragmentation patterns in the FI and EI mass
spectra. Thus, the EI mass spectrum of notoryne has fragments
at m/z 177/179, 133, 97 and 69 which were assigned to furan
fragments arising from fission of the inter-ring C–C bond (Fig. 11).
Furthermore, the dibrominated 2,2¢-bifuranyl 49, a degradation
product of laurefucin,3,62 shows similar fragmentation under EI
conditions. The EI mass spectra of both elatenyne5 and the
chloroenyne from L. majuscula6 both have ions which can be
readily explained as occurring by the same fragmentation of a
2,2¢-bifuranyl skeleton.
Fig. 12 Proposed structures of a dibromo-enyne from L. majuscula..
We had amassed considerable evidence that the structures
originally proposed for elatenyne and the chlorinated enyne from
L. majuscula were incorrect. However, in order to confirm these
structure misassignments it was necessary to undertake the total
synthesis of the originally proposed structures of these natural
products namely the halogenated pyrano[3,2-b]pyrans 3 and 4.
Total synthesis of the pyrano[3,2-b]pyrans 3 and 4
Having established that the originally proposed structures for
the natural products elatenyne and the chloroenyne from L.
majuscula (3 and 4) were likely to be incorrect, we sought further
confirmation of this by undertaking the total synthesis of these
two halogenated pyrano[3,2-b]pyrans. Given the difficulty we
encountered in preparing the exo-cyclic enol ether 6 we aimed
to synthesise both molecules by using appropriate organometallic
In 1989 Erickson and co-workers reported the isolation and
partial structure determination of a dibrominated 2,2¢-bifuranyl
1
from L. majuscula.63 On the basis of H NMR and 13C NMR J-
value analysis, they proposed a 2,2¢-bifuranyl core structure and
assigned the relative intra-ring stereochemistry but not the relative
inter-ring stereochemistry. The proposed structures (52) are shown
Fig. 11 Fragmentation patterns for a number of halogenated 2,2¢-bifuranyls.
The Royal Society of Chemistry 2009 Org. Biomol. Chem., 2009, 7, 238–252 | 245
This journal is
©

Scheme 8 Opening of the epoxide 18. Reagents and conditions: (a) diallylmagnesium, THF, Et2O, 57%.
reagents to open the bis-epoxide 18 at the anomeric centres with
Having developed an efficient synthesis of the C2-symmetric
diol 57a it was necessary to introduce the two side chains which
required the differentiation of the two primary alcohols. Our first
approach towards this goal was to convert the alcohols into the
corresponding benzylidene acetal and then cleave the resulting
acetal with DIBAL-H.72 Exposure of the diol 57a to benzaldehyde
dimethylacetal with PPTS as the acid catalyst, delivered the
desired benzylidene acetal 58 in low yield (Scheme 10).73 The
use of stronger acids such as PTSA resulted in extensive silyl
group migration and cleavage. Disappointingly, treatment of the
acetal 58 with DIBAL-H gave a very poor yield of the desired
differentially protected tetrol derivative 59 and hence this method
of desymmetrising the diol 57a was not pursued further.
inversion of configuration thus setting the required stereochem-
istry for the synthesis of 3 and 4. The synthesis of C-glycosides by
the opening of 1,2-anhydrosugars with organometallic reagents
is well precedented64–68 and treatment of the epoxide 18 with
allylmagnesium chloride or bromide gave the desired pyrano[3,2-
b]pyran 53a in moderate yields along with the inseparable
diastereomer 53b (Scheme 8).30 The highest yields were obtained
using diallylmagnesium which Rainier has used extensively for the
opening of similar epoxides in the synthesis of the ladder toxins.67
The improved yields using diallylmagnesium may be due to the
removal of Lewis acidic magnesium bromide or chloride from the
reaction mixture where it may catalyse side reactions of 18. If
the epoxide was not purified by crystallisation prior to treatment
with the organomagnesium reagents, then the tertiary alcohols 55
were formed as a side product as a 4 : 1 mixture of inseparable
diastereomers. Such products have previously been observed by
Rainer who proposed that they arise from opening of the epoxide
to give an oxocarbenium ion (e.g. 54) followed by 1,2-hydride shift,
to give a ketone which is then attacked by the allylmagnesium
reagent.69
Scheme 10 Synthesis of the benzyl ether 59. Reagents and conditions:
˚
(a) PhCH(OMe)2, PTSA, 4 A molecular sieves, toluene, reflux, 41%;
The diols 53 were persilylated and the terminal alkenes cleaved
by ozonolysis of the mixture of silylethers 56 with a reductive
workup (PPh3 and NaBH4) to give the separable diols 57 in
excellent overall yield (Scheme 9).30 Use of other methods of
(b) DIBAL, toluene, 0 ◦C, 17%.
Our next approach to desymmetrising the diol 57a involved
monotosylation or monoiodination such that the resulting prod-
ucts could be reduced to install the necessary ethyl side chain
of 3 and 4. Unfortunately, under a large number of reaction
conditions neither monotosylation nor monoiodination of the diol
57a could be achieved in yields above 30%. Furthermore, we could
only oxidise the diol 57a to the monoaldehyde in sub-statistical
yield and further functionalisation of the monoaldehyde was low
yielding (see ESI‡). The low efficiency of these transformations
led us to investigate an alternative desymmetrisation procedure.
Schreiber has shown that mono-silylation is an efficient method
for desymmetrising C2-symmetric intermediates and this proce-
dure proved effective in our system.74 Thus, treatment of diol
57a with 1 equivalent of chlorotriethylsilane gave 48% of the
desired alcohol 61, along with 17% of the bis-triethylsilyl ether
60 and 35% recovered starting material 57a (Scheme 11). Selective
deprotection of the two triethylsilyl groups in 60 by treatment
with K2CO3 in methanol gave quantitative recovery of the diol
57a, which was combined with the diol recovered from the initial
70
double bond cleavage (RuCl3/NaIO4 or OsO4/NaIO471) was far
less satisfactory.
Scheme 9 Synthesis of the diols 57. Reagents and conditions: (a) TBSOTf,
Et3N, CH2Cl2, 99%; (b) O3/O2, CH2Cl2, MeOH, -78 ◦C, then PPh3,
-78 ◦C, 2 h, then NaBH4, -78 ◦C → RT, 2 h, 82%.
246 | Org. Biomol. Chem., 2009, 7, 238–252
This journal is
The Royal Society of Chemistry 2009
©

Scheme 11 Synthesis of the pyrano[3,2-b]pyran 66. Reagents and conditions: (a) TESCl, Et3N, CH2Cl2, 61 48%, 57a 35%, 60 17%; (b) K2CO3, MeOH,
˚
100%; (c) TsCl, Et3N, DMAP, CH2Cl2, 93%; (d) Et3BHLi, Et2O, 91%; (e) K2CO3, MeOH, 98%; (f) TPAP, NMO, 4 A molecular sieves, CH2Cl2, 95%;
(g) Me3SiC CCH2SiMe2tBu, tBuLi, Ti(OiPr)4, THF, -78 ◦C, add 64, -78 ◦C → RT, 0.5 h, then (Me3Si)2NK, 75%; (h) TsOH, MeOH, 22 h, 75%.
∫
silylation reaction and resubjected to the silylation conditions to
provide a further 22% of 61 (total 70% of 61 after one recycle).
The alcohol 61 was readily converted into the corresponding
tosylate 62 which was reduced with lithium triethylborohydride75,76
to give the ethyl-substituted pyrano[3,2-b]pyran 63 in excellent
overall yield. The primary silyl protecting group of 63 was
removed under basic conditions and the resulting alcohol oxidised
to the corresponding aldehyde 64 using TPAP and NMO.77
The (Z)-enyne was introduced in a highly selective manner
using a Yamamoto–Peterson reaction.78,79 Thus, addition of the
allenyltitanium reagent derived from 3-(tbutyldimethylsilyl)-1-
Fig. 13 Conformations of the diol 53a.
trimethylsilylpropyne to the aldehyde 64 gave an intermediate
A variety of methods (CBr4/P(oct)3,83 PBr3,84 the Ghosez
silanol which, on the addition of a potassium base, was readily
reagent,85 SOBr2,8,86 Mitsunobu reaction with ZnBr2,87 triflate with
converted into the desired (Z)-enyne 65 in good yield and
LiBr in HMPA,88 imidazolylsulfonate with TBABr in toluene89)
selectivity (>10 : 1, (Z) : (E)). The remaining oxygen protecting
failed to give the desired dibromide 68. Ultimately, we found that
groups were removed under acidic conditions to give the diol 66 in
heating the bis-triflate 67 with tetrabutylammonium bromide in
readiness for the proposed double bromination for the synthesis
toluene under reflux gave the desired dibromide 68 in low yield
of 3.
(Scheme 12). Proof that the installation of the bromine atoms in
The introduction of bromine atoms with inversion of con-
68 had occurred with inversion of configuration followed from
figuration to the more hindered face of the pyrano[3,2-b]pyran
66 was expected to prove challenging and we first investigated
this transformation on the allyl-substituted pyrano[3,2-b]pyrans
53.80 SN2 reaction of the activated hydroxyl groups would be
impossible in the ground state O-proximal conformation of 53a-
O-prox since the trajectory for backside attack is completely
blocked (Fig. 13). The reactive O-distal conformation 53a-O-
dist would place all the substituents in axial positions, and
furthermore, in this conformation the nucleophilic bromide ion
must attack a secondary centre past an axial substituent. In
addition, the SN2 reaction is at a carbon atom bearing a b-oxygen
substituent, a situation which is known to yield slow rates of SN2
reactions.81 Kozikowski’s synthesis of the dactomelynes stalled at
the introduction of the corresponding chlorine substituent8 and
Scheme 12 Synthesis of the dibromide 68. Reagents and conditions:
Murai has implied that the halogen substituents in molecules such
as the dactylenes are best introduced prior to ring-formation.82
(a) Tf2O, pyridine, CH2Cl2; (b) nBu4NBr, toluene, reflux 2 h, 17% from
53.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 238–252 | 247
©

J-value analysis. Furthermore, H-4a and H-8a in 68 appeared as a
multiplet with the typical line shape for a pyrano[3,2-b]pyran with
the 13C NMR chemical shift of C-4a,8a being 71.4 ppm, consistent
with the pyrano[3,2-b]pyran structure.
Mixtures of other unidentifiable compounds were also isolated
from the reaction mixture which had NMR spectra consistent
with elimination and/or rearranged products, however, pure
material could never be obtained. Attempts at optimising the
reaction did not prove fruitful. Extended reaction times resulted
in decomposition of the product dibromide. Use of more polar
solvents such as DMF in place of toluene gave the formate ester
of the starting material and elimination products in low yield and
purity.
Pleasingly, treatment of the triflate 69 derived from the diol 66
under the same reaction conditions (tetrabutylammonium bro-
mide in toluene at reflux) delivered the corresponding dibromide
70 again in low yield (14%) (Scheme 13). The dibromide was
relatively unstable, however, deprotection with TBAF gave the
stable enyne 3 in quantitative yield. The 3JH,H coupling constants
between the ring protons in 70 and synthetic 3 were very similar
to those observed in the model brominated compound 68 (vide
supra) which strongly suggested that bromination with inversion of
configuration had occurred to give a pyrano[3,2-b]pyran with the
desired stereochemistry. The 1H NMR coupling constant between
the bridgehead protons H-4a and H-8a in synthetic 3 was 1.8 Hz
(in 70 it was also 1.8 Hz), which is in excellent agreement with
a pyrano[3,2-b]pyran in the O-proximal conformation. The 13C
NMR chemical shifts of C-4a and C-8a in 3 were 71.4 and 71.2
ppm again consistent with a pyrano[3,2-b]pyran structure. Thus,
we are confident that the synthetic enyne 3 had the structure and
conformation shown in Scheme 13. The 1H NMR and 13C NMR
chemical shifts and 3JH-H coupling constants were compared with
those reported for elatenyne5 and many discrepancies were noted.
In particular, the 1H NMR chemical shifts for H-6 and H-7, and
H-2 and H-3 were different by >0.2 ppm between the synthetic
and natural material with differences in the 13C NMR chemical
shifts of up to 8 ppm.90 Thus, we have confirmed that 3 is not the
structure of natural elatenyne.
Scheme 13 Synthesis of the dibromide 3. Reagents and conditions:
(a) Tf2O, pyridine, CH2Cl2; (b) nBu4NBr, toluene, reflux 2 h, 14% from
66; (c) TBAF, THF, 100%.
silylation procedure we had used previously74 to give the mono-
protected alcohol 72 in 64% yield after one recycling sequence
(Scheme 14).91 The required chlorine atom was readily introduced
by conversion of the alcohol 72 to the corresponding triflate
followed by heating with tetrabutylammonium chloride in toluene
under reflux to give the desired chloride 73 in 43% yield after
complete removal of the silyl protecting group. It is interesting to
note that the attempted replacement of an axial hydroxy group by a
chlorine atom in studies towards the synthesis of the dactomelynes
failed completely.8 The remaining secondary hydroxy group in 73
was inverted by an oxidation77/reduction sequence to give the
alcohol 75 as a single diastereomer; the stereochemistry of 75 was
confirmed by X-ray crystal structure analysis.9
Synthesis of the proposed structure of the enyne from L. majuscula
Differentiation of the two terminal alkenes was now required
such that the C-2 and C-6 side chains of 4 could be intro-
duced selectively. Exposure of the alcohol 75 to iodine gave the
We aimed to synthesise the chlorinated enyne from L. majuscula
from the diols 53. The diols 53 were desymmetrised by the
Scheme 14 Synthesis of the pyrano[3,2-b]pyran 75. Reagents and conditions: (a) TESCl, Et3N, CH2Cl2, 72 42%, 71 38%, 53a 20%; (b) Tf2O, pyridine,
CH2Cl2; (c) nBu4NCl, toluene, reflux, 2 h, then AmberliteTM resin IR-120, MeOH, 43% from 72; (d) nPr4NRuO4, NMO, CH2Cl2, 4 A molecular sieves,
˚
65%; (e) NaBH4, MeOH, 87%.
248 | Org. Biomol. Chem., 2009, 7, 238–252
This journal is
The Royal Society of Chemistry 2009
©

Scheme 15 Iodoetherification of the alcohol 75. Reagents and conditions: (a) I2, CH2Cl2, NaHCO3; 76a 85%, 76b 11%.
corresponding tricyclic iodides 76 in 96% yield as an 8 : 1 mixture
of diastereomers (Scheme 15). The configuration at the iodomethyl
1
bearing stereocentre was tentatively assigned on the basis of H
NMR NOESY experiments.30
Having successfully differentiated the two terminal alkenes in 75
we aimed to eliminate HI from the iodides 76 to give an enol ether
which on ozonolysis would deliver a lactone aldehyde in readiness
for introduction of the enyne side chain. We conducted a number
of exploratory experiments to test the validity of this approach. On
a small scale, exposure of the major diastereomer of the iodides 76a
to DBU in toluene at reflux gave the unstable enol ether 77 which
readily hydrolysed to the keto alcohol 78 on silica gel (Scheme 16).
Disappointingly, the elimination reaction to form the enol ether
was somewhat capricious. Furthermore, although ozonolysis of
the enol ether 77 did generate the lactone aldehyde 79 (nmax 1775,
1723 cm-1), the reaction was not clean even under a number of
reaction conditions. We were able to conduct a Wittig reaction on
<1 mg of the lactone aldehyde 79 which did give rise to material
1
with a H NMR in accord with the desired enyne 80; however,
given the capricious nature of both the formation of the enol ether
77 from the iodide 76a and its subsequent ozonolysis, this route
was not going to be able to supply sufficient quantities of material
for completion of the synthesis. Having demonstrated that the
Wittig reaction on the lactone aldehyde 79 was indeed possible,
Scheme 17 Completion of the synthesis of 4. Reagents and conditions:
we proposed to synthesise this intermediate from the chloride
(a) Et3SiOTf, Et3N, CH2Cl2, 99%, (b) O3/O2, CH2Cl2, -78 ◦C, then Ph3P,
75 by ozonolysis of the terminal olefins followed by oxidation
of the intermediate lactol to the corresponding g-lactone 79. We
were disappointed to discover that ozonolysis of the chloride 75
under a range of conditions destroyed the substrate and gave
unidentifiable material. The use of potassium osmate and sodium
periodate were equally ineffective.71 We suspected that the axial
alcohol in 75 might be interfering with the cleavage of the olefins.
Indeed exposure of the alcohol 75 to triethylsilyl triflate gave the
corresponding silyl ether 81 which on ozonolysis under standard
conditions gave the dialdehyde 82 in excellent yield after reductive
workup with triphenylphosphine (Scheme 17).
◦
+
-
∫
-78 C, 0.5 h, then RT, 8 h, 99%; (c) Ph3P CH2C C-SiMe3 Br , nBuLi,
THF, add 82, -78 → 0 ◦C, 45%, (15% recovered 82); (d) NaBH4, MeOH,
100%; (e) I2, PPh3, imidazole, Et2O, MeCN, RT, 72%; (f) Zn, AcOH,
MeOH, Et2O, RT, then HCl, 96%; (g) nBu4NF, THF, RT, 92%.
We were delighted to find that addition of one equiv-
alent of the ylide derived from (3-trimethylsilyl-2-propynyl)-
triphenylphosphonium bromide to a cold solution of the dialde-
hyde 82 delivered the desired enyne 83 with high E-selectivity (E/Z
> 7 : 1). A small quantity of the bis-enyne was also formed in the
reaction and 25% of the dialdehyde 82 was recovered; however,
Scheme 16 Exploratory transformations of the iodide 76a. Reagents and conditions: (a) DBU, toluene, reflux; (b) SiO2; (c) O3/O2, CH2Cl2, -78 ◦C, then
PPh3, -78 ◦C → RT; (d) Ph3P CH2C C-SiMe3 Br , nBuLi, THF, add 79, -78 → 0 ◦C.
+
-
∫
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 238–252 | 249
©

we were unable to find any of the mono-enyne corresponding to
reaction of the aldehyde proximal to the silyl protecting group. The
origin of the regioselectivity of this Wittig reaction may be steric in
nature due to the larger volume of an OTES group compared with
a chlorine atom. Reduction of the remaining aldehyde in 83 gave
the alcohol 84. The alcohol was converted into the corresponding
iodide 85 which on treatment with zinc dust and a small quantity
of acetic acid,92 followed by the addition of aqueous hydrochloric
acid gave the ethyl-substituted pyrano[3,2-b]pyran 86 in excellent
yield. The acetylene protecting group was removed with TBAF to
complete the synthesis of the originally proposed structure (4) of
the enyne from L. majuscula.
elatenyne and the enyne isolated from L. majuscula respectively;
Dr David Fox for helpful discussions; GlaxoSmithKline (CASE
award to HMS), the Royal Society (University Research Fellow-
ships to JWB and SIP) and the EPSRC for funding. AstraZeneca
UK are gratefully acknowledged for some unrestricted funds.
Notes and references
1 For an excellent review regarding recent misassigned natural products
and the role of chemical synthesis in structure determination see: K. C.
Nicolaou and S. A. Snyder, Angew. Chem., Int. Ed., 2005, 44, 2050.
2 For the isolation and structure determination of the dactomelynes see:
Y. Gopichand, F. J. Schmitz, J. Shelly, A. Rahman and D. Vanderhelm,
J. Org. Chem., 1981, 46, 5192; reisolation:; Z. Aydogmus, S. Imre, L.
Ersoy and V. Wray, Nat. Prod. Res., 2004, 18, 43.
3
The 1H and 13C NMR chemical shifts and JH,H coupling
constants of the enyne 4 were compared with this reported for
the chloroenyne from L. majuscula, and many discrepancies were
observed. In particular, in the 1H NMR of the L. majuscula enyne,
the resonances corresponding to H-6, H-8a, H-4a and H-12 were
all at >d = 4 ppm whereas the corresponding protons in the
synthetic material were all well below <d = 4 ppm. For the
synthetic material, the central oxygen-bearing carbons resonated
at d = 70.5 and 73.9 ppm whereas for the natural product these
carbons resonated at d = 79.2 and 77.9. It was clear that the
spectroscopic data for the natural material did not match that of
the synthetic pyrano[3,2-b]pyran confirming that the structure of
the natural product had been originally misassigned.
3 For the isolation and structure determination of notoryne see: H.
Kikuchi, T. Suzuki, E. Kurosawa and M. Suzuki, Bull. Chem. Soc.
Jpn., 1991, 64, 1763.
4 K. L. Erickson, in Marine Natural Products, ed. P. J. Scheuer, Academic
Press, New York, 1983 p. 131.
5 J. G. Hall and J. A. Reiss, Aust. J. Chem., 1986, 39, 1401.
6 A. D. Wright, G. M. Konig, R. Denys and O. Sticher, J. Nat. Prod.,
1993, 56, 394.
7 For a previous synthesis of the dactomelynes see: E. Lee, C. M. Park
and J. S. Yun, J. Am. Chem. Soc., 1995, 117, 8017.
8 For a previous attempted synthesis of the dactomelynes see: A. P.
Kozikowski and J. M. Lee, J. Org. Chem., 1990, 55, 863.
9 H. M. Sheldrake, C. Jamieson and J. W. Burton, Angew. Chem., Int.
Ed., 2006, 45, 7199.
10 It should be noted that at the time of the isolation of elatenyne, no
C15 halogenated 2,2¢-bifuranyl natural products had been reported. A
2,2¢-bifuranyl structure for elatenyne was considered less likely than the
corresponding pyrano[3,2-b]pyran: J. G. Hall, Ph. D. Thesis, La Trobe
University (Australia), 1984.
Conclusions
In this study we explored a number of routes for the preparation
of the bis-exo-cyclic enol ether 6 en route to the synthesis of the
originally proposed structures of elatenyne 3 and the chloroenyne
from L. majuscula 4. This resulted in the preparation of a
large number of cis-fused pyrano[3,2-b]pyrans and 2,2¢-bifuranyls.
Although we ultimately did not synthesise the desired exo-cyclic
enol ether 6 the large number of pyrano[3,2-b]pyrans and 2,2¢-
bifuranyls we had made led to us uncovering a 13C NMR chemical
shift/structure correlation and to postulate that the originally
proposed structure for elatenyne and the chloroenyne from L.
majuscula were incorrect. This proposal was confirmed by the two-
directional total synthesis of both of these halogenated pyrano[3,2-
b]pyrans. On the basis of our chemical shift model and reanalysis
of all of the spectroscopic data of both natural products, we
have proposed that the gross structures of the natural products
is based upon a central 2,2¢-bifuranyl core (the 2,2¢-bifuranyls 50
and 51). Reisolation of the natural products would allow further
spectroscopic analysis to aid full structure determination. In the
meantime, work is underway to predict the structures of elatenyne
and the chloroenyne from L. majuscula on the basis of DFT
calculations of 13C NMR chemical shifts93 and using a rational
biosynthetic pathway,3,94 and to confirm the stereochemistry of
the natural products by stereoselective total synthesis.
11 For a summary of the advantages of two-directional synthesis see: C. S.
Poss and S. L. Schreiber, Acc. Chem. Res, 1994, 27, 9; for an excellent
review on desymmetrisation and two-directional synthesis of natural
products see: S. R. Magnuson, Tetrahedron, 1995, 51, 2167.
12 J. W. Burton, J. S. Clark, S. Derrer, T. C. Stork, J. G. Bendall and A.
B. Holmes, J. Am. Chem. Soc., 1997, 119, 7483; J. W. Burton, E. A.
Anderson, P. T. O’Sullivan, I. Collins, J. E. Davies, A. D. Bond, N.
Feeder and A. B. Holmes, Org. Biomol. Chem., 2008, 6, 693; S. Y. F.
Mak, N. R. Curtis, A. N. Payne, M. S. Congreve, A. J. Wildsmith, C.
L. Francis, J. E. Davies, S. I. Pascu, J. W. Burton and A. B. Holmes,
Chem.–Eur. J., 2008, 14, 2867.
13 For reviews see: B. Marciniec, Coord. Chem. Rev., 2005, 249, 2374; B.
Marciniec, Silicon Chem., 2002, 1, 155; I. Ojima, Z.-J. Li, and J. Zhu,
in The Chemistry of Organic Silicon Compounds, ed. Z. Rappoportand
Y. Apeloig, John Wiley & Sons Ltd., New York, 1998, vol. 2 p. 1688.
14 In model studies related to this project we have demonstrated that
the intramolecular hydrosilation of glucose and galactose-derived exo-
cyclic enol ethers gives rise to the corresponding 1,3-diols.
15 For the trans-selective hydroboration/oxidation of 6-membered exo-
cyclic enol ethers see: T. V. RajanBabu and G. S. Reddy, J. Org. Chem.,
1986, 51, 5458; Y. Ichikawa, R. Monden and H. Kuzuhara, Tetrahedron
Lett., 1986, 27, 611; Y. Ichikawa, R. Monden and H. Kuzuhara,
Carbohydr. Res., 1988, 172, 37; A. D. Campbell, D. E. Paterson, T.
M. Raynham and R. J. K. Taylor, Chem. Commun., 1999, 1599; D. E.
Paterson, F. K. Griffin, M. L. Alcaraz and R. J. K. Taylor, Eur. J. Org.
Chem., 2002, 1323; H. Tanaka, K. Kawai, K. Fujiwara and A. Murai,
Tetrahedron, 2002, 58, 10017.
16 For the stability of g and d-lactones see: J. M. Brown, A. D. Conn,
G. Pilcher, M. L. P. Leitao and M. Y. Yang, J. Chem. Soc., Chem.
Commun., 1989, 1817; K. B. Wiberg and R. F. Waldron, J. Am. Chem.
Soc., 1991, 113, 7697.
Experimental
17 P. A. Grieco, T. Oguri and Y. Yokoyama, Tetrahedron Lett., 1978, 419.
18 M. E. Maier and S. Reuter, Synlett, 1995, 1029.
See ESI.‡
19 (a) A. Krief, W. Dumont, P. Pasau and P. Lecomte, Tetrahedron, 1989,
45, 3039; (b) S. Saito, O. Narahara, T. Ishikawa, M. Asahara, T.
Moriwake, J. Gawronski and F. Kazmierczak, J. Org. Chem., 1993,
58, 6292; (c) H. Naito, E. Kawahara, K. Maruta, M. Maeda and S.
Sasaki, J. Org. Chem., 1995, 60, 4419.
Acknowledgements
We thank Prof. James Reiss (La Trobe University) and Prof.
Gabriele Ko¨nig (Universita¨t Bonn) for supplying spectral data for
20 D. J. Kopecky and S. D. Rychnovsky, J. Org. Chem., 2000, 65, 191.
250 | Org. Biomol. Chem., 2009, 7, 238–252
This journal is
The Royal Society of Chemistry 2009
©

21 The equilibrium position was readily determined by 1H NMR by
allowing various mixtures of 15 and 9 to equilibrate in d4-methanol
to which 10 mol% acetyl chloride had been added. Beginning either
with a mixture of the 2,2¢-bifuranyls 15 or a 1 : 1 mixture of 15 : 9 gave
the equilibrium ratios after 10 hours at room temperature which did
not change after one week. In a separate experiment, a mixture of the
pyrano[3,2-b]pyrans 9 also readily equilibrated to the same mixture of
15 : 9 on stirring in acidic methanol. The pyrano[3,2-b]pyrans 9 exist as
a 1 : 1.3 mixture of diastereomers at equilibrium which is in keeping with
the magnitude of the anomeric effect for 2-methoxytetrahydropyran in
methanol see: R. U. Lemieux, A. A. Pavia, J. C. Martin and K. A.
Watanabe, Can. J. Chem., 1969, 47, 4427.
22 Prior to this work, Nelson reported the equilibration of a 2,2¢-bifuranyl
dimethyl acetal to the corresponding trans-fused pyrano[3,2-b]pyran in
a beautiful synthesis of a ladder toxin intermediate see: J. M. Holland,
M. Lewis and A. Nelson, Angew. Chem., Int. Ed., 2001, 40, 4082; J. M.
Holland, M. Lewis and A. Nelson, J. Org. Chem., 2003, 68, 747.
23 Warren has studied the equilibration of a 2,2¢-bifuranyl with a cis-fused
pyrano[3,2-b]pyran via intermediate epi-sulfonium ions see: J. Carlisle,
D. J. Fox and S. Warren, Chem. Commun., 2003, 2696.
52 S. Abramson, E. Ashkenazi, K. Frische, I. Goldberg, L. Golender,
M. Greenwald, N. G. Lemcoff, R. Madar, S. Weinman and B. Fuchs,
Chem.–Eur. J., 2003, 9, 6071.
53 M. Grabarnik, N. G. Lemcoff, R. Madar, S. Abramson, S. Weinman
and B. Fuchs, J. Org. Chem., 2000, 65, 1636.
54 H. Jatzke, K. Frische, M. Greenwald, L. Golender and B. Fuchs,
Tetrahedron, 1997, 53, 4821.
55 H. Senderowitz, L. Golender and B. Fuchs, Tetrahedron, 1994, 50,
9707.
56 H. Senderowitz, A. Linden, L. Golender, S. Abramson and B. Fuchs,
Tetrahedron, 1994, 50, 9691.
57 L. Norskov, R. B. Jensen and G. Schroll, Acta Chem., Scand. Ser. B:
Org. Chem. Biochem., 1983, 37, 133.
58 The 1H NMR spectra for the C2-symmetric pyrano[3,2-b]pyrans are
not first order; however, the narrow line-width of H-4a and H-8a in all
of the pyrano[3,2-b]pyrans synthesised in this work implies that they
exist predominantly in the O-proximal conformation and hence J4a,8a is
small. Because J4a, 8a is small the 1H NMR spectra of these pyrano[3,2-
b]pyrans is pseudo-first order and meaningful coupling constants can be
extracted. The validity of this assumption was confirmed by simulating
the 1H NMR spectra of a number of the C2-symmetric pyrano[3,2-
b]pyrans reported in this work (see ESI).
24 K. Iijima, W. Fukuda and M. Tomoi, J. Macromol. Sci. Pure Appl.
Chem., 1992, A29, 249.
25 L. A. Paquette and J. A. Oplinger, J. Org. Chem., 1988, 53, 2953.
26 D. Craig, M. W. Pennington and P. Warner, Tetrahedron, 1999, 55,
13495.
59 Crystal structure determination: crystallographic data of sulfone 38 was
collected on the synchrotron radiation source at Station 9.8, Daresbury
SRS, UK, on a Bruker SMART CCD diffractometer. The structures
were solved by direct methods using the program SIR92 (ref. 65). The
refinement (on F) and graphical calculations were performed using the
CRYSTALS (ref. 66) program suite. Crystal data: C15H20O5S, M =
27 Some of the resonances corresponding to 29 in the 1H NMR and 13
C
C
NMR were broad and there were some resonances missing from the 13
NMR which indicated that the anomeric iodides were in intermediate
exchange on the NMR timescale; cooling the sample did not result in
significant sharpening of the resonances.
˚
312.38, Z = 4, monoclinic, space group P21, a◦= 5.5615(17) A, b =
3
˚
˚
˚
27.699(8) A, c = 10.094(3) A, b = 105.644(6) , V = 1497.4(8) A ,
T = 293 K, m = 0.235 mm-1. Of 10048 reflections measured, 6771 were
independent (Rint = 0.02). Final R = 0.0464 (4429 reflections with I >
3s (I)) and wR = 0.0493. Crystallographic data (excluding structure
factors) for this structure have been deposited with the Cambridge
Crystallographic Data Centre as supplementary publication no. CCDC
698760. Copies of the data can be obtained free of charge on application
to the CCDC, 12 Union Road, Cambridge CB2 1EZ, U.K. [Fax:
(internat.) + 44-1223/336-033; E-mail: deposit@ccdc.cam.ac.uk].
60 A. Altomare, G. Cascarano, C. Giacovazzo and A. Guagliardi, J. Appl.
Crystallogr., 1993, 26, 343.
28 C. Baylon, M. P. Heck and C. Mioskowski, J. Org. Chem., 1999, 64,
3354.
29 K. Kadota, T. Kurusu, T. Taniguchi and K. Ogasawara, Adv. Synth.
Catal., 2001, 343, 618.
30 For full structure determination see ESI.
31 The acetates were formed using the method of Gin: L. Shi, Y. J. Kim
and D. Y. Gin, J. Am. Chem. Soc., 2001, 123, 6939.
32 J. Lussmann, D. Hoppe, P. G. Jones, C. Fittschen and G. M. Sheldrick,
Tetrahedron Lett., 1986, 27, 3595.
33 R. L. Halcomb and S. J. Danishefsky, J. Am. Chem. Soc., 1989, 111,
6661.
34 W. Adam, J. Bialas and L. Hadjiarapoglou, Chem. Ber., 1991, 124, 2377;
R. W. Murray and M. Singh, Org. Synth., 1997, 74, 91.
35 M. Hatanaka and H. Nitta, Tetrahedron Lett., 1987, 28, 69.
36 B. Bessieres and C. Morin, J. Org. Chem., 2003, 68, 4100.
37 L. Alcaraz, A. Cridland and E. Kinchin, Org. Lett., 2001, 3,
4051.
38 We orginally attempted to methylenate the bis-d-lactone 8 with dimethyl
titanocene to give the corresponding bis-exo-cyclic enol ether which
would under go allylic oxidation to give 6 directly. However, exposure
of 8 to dimethyltitanocene delivered solely the bis-endo-cyclic enol
ether 42 in variable yield; the use of the Tebbe reagent resulted
in decomposition. Attempted allylic oxidation of the enol ether 42
with either stoichiometric or catalytic selenium dioxide was also
unsuccessful.
61 D. J. Watkin, C. K. Prout, J. R. Carruthers and P. W. Betteridge,
CRYSTALS Issue 11, Chemical Crystallography Laboratory, Oxford
Uk, 2001; P. W. Betteridge, J. R. Carruthers, R. I. Cooper, K. Prout
and D. J. Watkin, J. Appl. Crystallogr., 2003, 36, 1487.
62 A. Furusaki, E. Kurosawa, A. Fukuzawa and T. Irie, Tetrahedron Lett.,
1973, 4579.
63 I. K. Kim, M. R. Brennan and K. L. Erickson, Tetrahedron Lett., 1989,
30, 1757.
64 S. P. Allwein, J. M. Cox, B. E. Howard, H. W. B. Johnson and J. D.
Rainier, Tetrahedron, 2002, 58, 1997.
65 J. D. Rainier and J. M. Cox, Org. Lett., 2000, 2, 2707.
66 U. Majumder, J. M. Cox and J. D. Rainier, Org. Lett., 2003, 5, 913.
67 J. D. Rainier, S. P. Allwein and J. M. Cox, J. Org. Chem., 2001, 66, 1380.
68 D. A. Evans, B. W. Trotter and B. Cote, Tetrahedron Lett., 1998, 39,
1709.
39 D. S. Brown, M. Bruno, R. J. Davenport and S. V. Ley, Tetrahedron,
69 We attempted to convert the asymmetric diol 53b into the symmetric
diol 53a by an oxidation, epimerisation, reduction sequence. Treatment
of the mixture of diols 53 with Jones’ reagent gave the corresponding
separable diketones. Disappointingly, attempted epimerisation of the
resulting diketones under basic conditions resulted in decomposition
(see ESI).
1989, 45, 4293.
40 S. V. Ley, B. Lygo, F. Sternfeld and A. Wonnacott, Tetrahedron, 1986,
42, 4333.
41 J. Sisko, M. Mellinger, P. W. Sheldrake and N. H. Baine, Org. Synth.,
2000, 77, 198.
42 M. Gibert, M. Ferrer, F. SanchezBaeza and A. Messeguer, Tetrahedron,
1997, 53, 8643.
43 M. Ando, A. Akahane, H. Yamaoka and K. Takase, J. Org. Chem.,
1982, 47, 3909.
44 V. S. Joshi, N. p. Damodara and S. Dev, Tetrahedron, 1968, 24, 5817.
45 G. A. Kraus and K. Frazier, J. Org. Chem., 1980, 45, 2579.
46 H. R. Kricheldorf, G. Morber and W. Regel, Synthesis, 1981, 383.
47 R. J. Giguere, G. Von Ilsemann and H. M. R. Hoffmann, J. Org. Chem.,
1982, 47, 4948.
48 S. Tchilibon and R. Mechoulam, Org. Lett., 2000, 2, 3301.
49 R. W. Rickards and W. P. Watson, Aust. J. Chem., 1980, 33, 451.
50 R. W. Hoffmann and I. Munster, Liebigs Ann.-Recl., 1997, 1143.
51 A. G. Santos and R. W. Hoffmann, Tetrahedron: Asymmetry, 1995, 6,
2767.
70 D. Yang and C. Zhang, J. Org. Chem., 2001, 66, 4814.
71 R. Pappo, D. S. Allen, R. U. Lemieux and W. S. Johnson, J. Org. Chem.,
1956, 21, 478.
72 S. L. Schreiber, Z. Y. Wang and G. Schulte, Tetrahedron Lett., 1988, 29,
4085.
73 The acetal carbon in 58 is a chirotopic non-stereogenic centre according
the classification of Mislow see: K. Mislow and J. Siegel, J. Am. Chem.
Soc., 1984, 106, 3319.
74 C. S. Poss, S. D. Rychnovsky and S. L. Schreiber, J. Am. Chem. Soc.,
1993, 115, 3360.
75 H. C. Brown and S. Krishnamurthy, J. Am. Chem. Soc., 1973, 96, 1669.
76 S. Krishnamurthy, J. Org. Chem., 1980, 45, 2550.
77 S. V. Ley, J. Norman, W. P. Griffith and S. P. Marsden, Synthesis, 1994,
639.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 238–252 | 251
©

78 K. Furuta, M. Ishiguro, R. Haruta, N. Ikeda and H. Yamamoto, Bull.
Chem. Soc. Jpn., 1984, 57, 2768.
79 Y. Yamakado, M. Ishiguro, N. Ikeda and H. Yamamoto, J. Am. Chem.
Soc., 1981, 103, 5568.
80 The model bromination studies were conducted on the inseparable 5 :
1 mixture of diastereomers 60a : 60b.
81 A. Streitwieser, Chem. Rev., 1956, 56, 571.
82 L. X. Gao and A. Murai, Heterocycles, 1996, 42, 745.
83 K. Fujiwara, M. Kobayashi, D. Awakura and A. Murai, Synlett, 2000,
1187.
88 R. Ranganathan, Tetrahedron Lett., 1977, 1291.
89 S. Hanessian and J. M. Vatele, Tetrahedron Lett., 1981, 22, 3579.
90 For a side by side comparison of the 1H NMR and 13C NMR chemical
shifts for the natural and synthetic material see the ESI.
91 Compounds 71 and 72 were formed as mixtures (and yields are for the
mixture) along with their corresponding diastereomers arising from
silylation of the diol 53b. The minor diastereomer was removed after
chlorination to give 73.
92 The reported conditions were particularly effective: Z. H. Peng and K.
A. Woerpel, J. Am. Chem. Soc., 2003, 125, 6018.
84 N. Petragnani, H. M. C. Ferraz and M. Yonashiro, Synthesis, 1985, 27.
85 L. Ghosez, I. George-Koch, L. Patiny, M. Houtekie, P. Bovy, P.
Nshimyumukiza and T. Phan, Tetrahedron, 1998, 54, 9207.
86 A. P. Kozikowski and J. Lee, Tetrahedron Lett., 1988, 29, 3053.
87 P. T. Ho and N. Davies, J. Org. Chem., 1984, 49, 3027.
93 S. G. Smith, R. S. Paton, J. W. Burton and J. M. Goodman, J. Org.
Chem., 2008, 73, 4053.
94 A. Murai, in Comprehensive Natural Products Chemistry, ed. D. H.
R. Barton, O. Meth-Cohn, and K. Nakinishi, Elsevier, Oxford, 1999,
p. 301.
252 | Org. Biomol. Chem., 2009, 7, 238–252
This journal is
The Royal Society of Chemistry 2009
©