Synthetic studies directed towards various homologues of natural Sesquiterpene-Coumarin ethers: The Domino approach

Article abstract of DOI:10.1002/ejoc.200901352

A domino-based strategy was used to construct analogues containing the basic skeleton of the monocyclic sesquiter-pene-coumarin ethers galbanic acid (1) and secodrial (3), through conversion of the domino adduct 19 into 10 and 11, chosen as representative targets. ¹H NMR patterns, corroborated by X-ray crystallographic analysis for two of the four possible diastereomeric arrangements of the six-membered B-ring common to various A-seco terpenes, have been determined. The observed trends help in the design of substituent combinations that provide a tool for diastereomeric recognition, depending on the cis/trans arrangement of the adjacent methyl groups and the adopted conformations.

Full text of DOI:10.1002/ejoc.200901352

FULL PAPER
DOI: 10.1002/ejoc.200901352
Synthetic Studies Directed Towards Various Homologues of Natural
Sesquiterpene-Coumarin Ethers: The Domino Approach
Andrei Corbu,[a] Juan M. Castro,[a] Maurizio Aquino,[a] Zoila Gandara,[a]
Pascal Retailleau,[a] and Siméon Arseniyadis*[a]
Keywords: Configuration determination / Terpenoids / Galbanic acid analogues / Secodriol analogues
A domino-based strategy was used to construct analogues
containing the basic skeleton of the monocyclic sesquiter-
pene-coumarin ethers galbanic acid (1) and secodrial (3),
through conversion of the domino adduct 19 into 10 and 11,
chosen as representative targets. H NMR patterns, corrobo-
rated by X-ray crystallographic analysis for two of the four
possible diastereomeric arrangements of the six-membered
B-ring common to various A-seco terpenes, have been deter-
mined. The observed trends help in the design of substituent
combinations that provide a tool for diastereomeric recogni-
tion, depending on the cis/trans arrangement of the adjacent
methyl groups and the adopted conformations.
1
structure of natural galbanic acid was confirmed as
(8S,9S,10S) through its first asymmetric synthesis in a con-
vergent synthetic pathway. Since then an improved version
of our chiral-pool-based entry into the naturally occurring
sesquitriterpenoid area has been achieved, allowing the first
synthesis of natural galbanic acid [(–)-1], in which Mitsu-
nobu etherification under microwave irradiation conditions
provided a considerable increase in yield while reducing the
route by one step (out of ten linear steps) before coupling.
The synthesis of marneral [(+)-4] was also significantly
shortened in this same work.[8] This and the previous work
from our laboratory were, to the best of our knowledge, the
first published contributions in which the optically pure (or
even racemic) products 1 and 4, as well as analogues of 3
(secodrial) and 6 (secochiliotrin),[9] could be produced.[10]
Initially, the purpose of the research described below was
to synthesize conveniently functionalized six-membered
rings of types 1, 2 (the revised, but erroneous, structure pro-
posed for galbanic acid), 5,[11] 6(8), and 7, possessing all
combinations of the R and S configurations at C-8, C-9,
and C-10, to resolve the still persisting structural doubt and
to provide a prognostic tool for recognizing the substitution
pattern simply from NMR spectroscopic data. Two of the
four possible diastereomers, featuring the trans-methyl (1,
4, 5) and cis-methyl (6) arrangements, occur in natural
products (Figure 1).
Introduction
In the context of our ongoing interest in the synthesis
and biological evaluation of various A-seco terpenes, we
have recently reported a diverging chiral-pool-based ap-
proach to the preparation of the sesquiterpene-coumarin
ether ent-galbanic acid [(+)-1][1] and the monocyclic triter-
pene marneral [(+)-4],[2] starting from (R)-(+)-pulegone.[3]
The structural similarity of galbanic acid and marneral
points to a possible biosynthetic connection through a
series of cyclizations, 1,2-hydride shifts, and methyl mi-
grations. In view of their co-occurrence, Marner proposed
marneral as a biosynthetic precursor of the iridal family.[2b]
A different biogenetic origin was proposed by Appendino
et al.[4] in a revision of the structure of asacoumarin B to
galbanic acid, suggesting mogoltadone[5] as the biogenetic
precursor of galbanic acid.
At the outset of this project, no published syntheses for
any of the A-seco terpene targets existed. Using a chiral
pool approach, we completed the synthesis of (+)-1 by ap-
plying the protocol of Reymond et al.,[6] a modified Wil-
liamson etherification in the presence of 18-crown-6 under
microwave irradiation conditions, with commercially avail-
able umbelliferone and the required B-ring C11-tosylate.
Despite inefficient segment coupling, the synthesis of the
antipodal galbanic acid in that previous work secured the
absolute stereochemistry of the natural galbanic acid while
ending the controversy around its relative stereochemistry.[7]
Of the four plausible pairs of antipodes, the absolute stereo-
For a convenient route to the six-membered B-ring
moiety we envisaged use of the lead-tetraacetate-mediated
oxidative cleavage of readily available bicyclic unsaturated
diols in the Hajos–Parrish series, providing direct access to
stereodefined cyclohexane frameworks.[12] In this context
we had previously described the development of a domino
reaction[13] and its application to the synthesis of bioactive
natural products.[14] Two out of four possible structures
[a] Centre de Recherche de Gif, Institut de Chimie des Substances
Naturelles, CNRS,
91198 Gif-sur-Yvette Cedex, France
E-mail: simeon.arseniyadis@icsn.cnrs-gif.fr
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.200901352.
Eur. J. Org. Chem. 2010, 1483–1493
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1483

S. Arseniyadis et al.
FULL PAPER
Figure 1. Structures of representative A-seco terpenes bearing structurally similar B-ring moieties.
Figure 2. Originally proposed stereochemical assignment of galbanic acid 1 (both antipodes), its previously revised (albeit erroneous)
proposed structure 2, and the remaining two of the four pairs of antipodes, 7 and 8 (only one antipode is represented for 2, 7, and 8).
(type 2 and type 7, Figure 2) could be approached by use logues at C-11 and epimeric to each other at C-8, synthe-
of our domino methodology as a key reaction step, setting sized by the sequence outlined in Scheme 1. The cyclohex-
the configurations at C-9 and C-10, while allowing for the ane derivatives 14 and 15, containing the appropriate C-8,
post-domino construction of the C-8 center both in its R C-9, and C-10 stereogenic centers, were to be accessed from
and in its S configuration.
the known domino product 19, and this in turn was to be
In the light of our previous results relating to the synthe- elaborated from the versatile building block 18.[15] Linking
sis of galbanic acid [(–)-1] and its enantiomer,[3,8] the antici- to the coumarin could subsequently be performed at C-16
pated difficulty in carrying out the coupling of the couma- either directly by a Mitsunobu etherification or by conver-
rin part to the crowded neopentylic center C-11 prompted sion of the free hydroxy functionality into a tosylate leaving
us to investigate a more reactive sesquiterpene partner. We group followed by a Williamson etherification. With the
hence chose to target structures 10 and 11, higher homo- ready accessibility of Hajos–Parrish ketone and pulegone
Scheme 1. Domino and chiral pool approaches for unequivocal structure assignments.
1484
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 1483–1493

Approach to Homologues of Natural Sesquiterpene-Coumarin Ethers
(both antipodes), we established access to six out of eight whereas when H2-Pd/C was replaced by Raney-Ni a slight
possible stereoisomers, although the last enantiomeric pair reversal in the stereomeric ratio occurred, with the C-12α
of type 6(8), characteristic of the secochiliotrin framework, epimer being the major product in the latter case.
still remains inaccessible.
Although a method for selective production either of dia-
Here we report the domino-based approach involving the stereomer 20a or of diastereomer 20b would ultimately be
stereocontrolled formation of natural product analogues. desired, a 1:1 mixture of epimers was tolerable at this point
Two of them are umbelliferone-derived ethers (10, 11) and because we needed both diastereomeric types for structure
the other two are isofraxidine-derived ethers (30a, 30b) with elucidation.
the same sesquiterpenoid unit. The goal was to enable a
Thus, after the preparation of 19, in which the olefin
combination of the two routes, the former chiral pool ap- serves as a C-12 methyl precursor, a Raney-Ni reduction
proach and the current domino approach, to provide both afforded the C-12α- and -β-methyl groups, providing a
a means to access three out of the four possible cyclohexane nearly quantitative yield of 20a and 20b as an inseparable
core structures 1, 2, 7, and 8 (Figure 2) and a reliable diag- mixture. En route to the isopropylidene alcohols 21a, 21b,
nostic model for structure assignment by NMR techniques. and 21c, separation of the C-8 epimers was achieved
through lithium aluminium hydride reduction of the thus
obtained 20a and 20b and subsequent selective acetonide
formation by standard procedures (Scheme 3). The C-8 epi-
Results and Discussion
The required unsaturated diols 18 and the domino prod- meric diastereomers 21a and 21b/21c were easily separated
uct 19 were prepared straightforwardly, in their optically by silica gel column chromatography and readily differen-
1
pure forms, by published procedures.[15] Reduction of the tiated by H NMR spectroscopy.
exocyclic alkene was carried out first, with the installment
With success in the separation of C-8 epimers and large
of the coumarin moiety being left for later stages. Thus, amounts available with which to press forward, the targeted
with the large-scale acquisition of the domino derivative 19, B-ring building block 14, together with its C-8 epimer 15,
the crucial issue of setting up the C-12 methyl with the ap- was readily accessed by carrying out the sequences de-
propriate stereochemistry was addressed. An initial investi- scribed in Scheme 4 and Scheme 5 in parallel. After the dia-
gation for this study, based on palladium-catalyzed hydro- stereomeric separations it becomes necessary temporarily to
genation of the double bond, revealed moderate selectivity block the secondary hydroxy function at C-16. An ideal
(4:1) with the major product being the C-12b epimer 20b blocking group would be benzyl, which could be removed
(Scheme 2). We then briefly studied the factors influencing later by Raney nickel catalytic hydrogenolysis. Treatment of
the stereochemistry of the C8–C12 olefin reduction. Cata- 21 with benzyl bromide in the presence of sodium hydride
lyst support and added hydrogen was investigated, the tem- in dry DMF afforded the corresponding benzyl ethers, and
perature was varied between 25 °C and 60 °C, and various subsequent treatment with HCl/THF (5%), selective protec-
types of solvents were employed. Although the hydrogena- tion of the resulting diols 22 as the corresponding tert-but-
tion was effective in terms of yield (Ն98%), the formation yldimethylsilyl ethers, and Swern oxidation took the route
of the C-8 methyl group proceeded with only moderate dia- as far as the TBS-protected aldols 23 (Scheme 4). These
stereoselection. To summarize, the domino intermediate 19 now require construction of the exocyclic gem-dimethyl ole-
exhibited modest selectivity towards hydrogenation, fin and further each require an additional two-carbon ho-
Scheme 2. Exploring the facial selectivity of the C8–C12 exocyclic olefin hydrogenation.
Scheme 3. Formation of the central B-ring.
Eur. J. Org. Chem. 2010, 1483–1493
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1485

S. Arseniyadis et al.
FULL PAPER
mologation. Insofar as the dimethyl olefin introduction was
concerned, it was initially hoped that this could be ac-
complished either directly through a Wittig olefination or
indirectly through a Knoevenagel condensation[16] with one
of the many adapted modifications for tetrasubstituted ole-
fin formation. Attempted olefinations with 23 under a vari-
ety of conditions proved fruitless, however, and as a result
an organolithium addition/oxidative rearrangement se-
quence, inspired by our previous iridal synthesis,[14b] was
investigated. Thus, with the aim of introducing the C-4/C-
13/C-14 unit, as a potential formylolefin precursor, 23 was
first treated with (prop-2-enyl)magnesium bromide to af-
ford a 54% yield of the corresponding carbinol along with
unreacted starting material (23%). Alternatively, addition
of isopropenyllithium (prepared in situ from 2-bromo-
propene and tBuLi in THF, –78 °C) raised the yield of the
required carbinol to 87% (4% of recovered starting mate-
rial). After deprotection (fluoride) and Swern oxidation,
homologation of 24 was achieved by means of a Wittig-
based olefination with ethyl (triphenylphosphoranylidene)-
acetate, which furnished the α,β-unsaturated esters 25a and
25b (70% and 81% yields, respectively, over three steps,
Scheme 4).
Scheme 5. Elaboration of the exocyclic dimethyl olefin and C-10
side chain (brsm = based on recovered starting material).
however, because the Z/E geometry would be destroyed in
the unveiling of the gem-dimethyl group through removal
of the dithiane at C-13 for the conversion of 27a and 27b
into 14 and 15. This conversion proved more beneficial than
originally anticipated, because among the standard condi-
tions surveyed, the use of Raney nickel in EtOH at 25 °C
directly afforded high yields of 14 (67%) and 15 (77%). In
this three-step one-pot reaction, treatment of 27a and 27b
with Raney nickel in ethanol at ambient temperature led to
debenzylation, dethioketalization, and selective reduction
of the conjugated olefins (Scheme 5).
Interestingly, the Reymond protocol, which in our pre-
vious work had given low-yielding coupling with the tosyl-
ate (+)-i for the methylene-homologated ent-galbanic acid
(+)-iii,[18] shows a configuration-controlled selectivity. Thus,
once the synthesis of the B-ring moiety had been achieved,
we proceeded in parallel with the two experimental cou-
pling protocols – namely the Mitsunobu and the William-
son etherification – with the aim of investigating the reactiv-
ity patterns of various diastereomeric B-ring frameworks.
Firstly, Mitsunobu couplings in which the commercially
available coumarins 28 (Scheme 6) and 29 (Scheme 7) were
allowed to react with the B-ring precursors 14 and 15 were
tried, affording the desired coumarin ethers 10 and 11 in
81% and 74% yields, respectively.
A Williamson etherification by the protocol of Reymond
et al.[6] followed, with the same coumarins 28 and 29 being
treated with tosylates derived from the one-carbon-homolo-
gated alcohols 14 and 15 as shown in Scheme 6 and
Scheme 7. The fully elaborated skeletons of the galbanic
acid analogues 10 and 11 were produced uneventfully upon
Scheme 4. Addition of the missing carbons at C-1 and C-5.
From our previous experience in the iridal synthesis, it attachment of umbelliferone (28) to 14 and 15, respectively
was initially assumed that the regiochemical outcome of the (Scheme 6). The ethyl homogalbanates 10 and 11 were sa-
Dauben–Michno rearrangement[17] would give a Z/E mix- ponified by treatment with LiOH in THF and then con-
ture of enals 26. This would have no long-term significance, verted into their corresponding methyl esters 10b and 11b
1486
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 1483–1493

Approach to Homologues of Natural Sesquiterpene-Coumarin Ethers
Scheme 6. Completion of the synthesis of galbanic acid analogues.
with TMSCHN2, in methanol, thus affording two more an- diates involved in the sequence can also clear the way for
alogues, because methyl galbanate is also a natural prod- the final assignment. The task is even more arduous, how-
uct.[19]
ever, when conformational equilibria have to be taken into
In the same manner as above, isofraxidine (29) and the account for the structural assignment, especially for small
same sesquiterpenoid units 14 and 15 afforded the secodrial molecules that can adopt various conformations. In our
analogues 30a and 30b, respectively (Scheme 7), thus com- case we dealt with the unknown conformation of galbanic
pleting the synthesis of the targeted compounds.
acid, which subsequently generated confusion in its relative
stereoassignment. The assignment of the vicinal stereocen-
ters in the trans (2) and the cis (7) series as (8R,9R) and
(8S,9R), respectively, had largely been based on spatial
proximity effect measurements based on the 2D NOESY
technique and J-analysis, with subsequent corroboration of
two key intermediates (22bT, 25a) and a target analogue
(10) by X-ray crystallographic analysis.
The C-12/C-15 cis-Dimethyl Series
The triol 22bT obtained by acetonide deprotection gave
single crystals suitable for X-ray crystallographic analysis
(Figure 4, below, ORTEP drawing). The depicted relative
stereochemistry and preferred conformations were first de-
duced from the magnitudes of coupling constants and cor-
roborated by 2D NOESY experiments (Figure 3). The ring
protons have 3JH,H values consistent with a chair conforma-
tion, as can be seen in both Figure 3 and Figure 4.
The axially oriented Me15 group of 27bZ displays corre-
lation peaks with the protons resident on the β face of the
molecule (7-H, 10-H), but also to the 13-H proton, an inter-
action that would not be observed if the thioketal group
were in the opposite geometry. Conversely, a strong NOE
was observed as a consequence of the syn relationship of
the protons of Me14 and 6βeq-H. This assignment was also
confirmed by the dipolar coupling between the olefinic pro-
ton 1-H and the axial protons 8α-H and 6α-H. Insofar as
the B-ring right-half residue 15 is concerned, it can immedi-
ately be seen that the β-cis relationship for Me12/Me15 is
in fact correct, with the six-membered ring also disposed in
chair form, as in the precursor molecule 22bT. Observation
of correlation peaks between the signals for the C-13
Scheme 7. Completion of the synthesis of secodrial analogues.
As demonstrated by the above results, the segment cou-
pling by the Raymond protocol could be accomplished
cleanly and in good yields with the diastereomeric B-ring
precursors 14 and 15, as well as with several model sub-
strates,[3] whereas with (+)-i as substrate it had failed,[18]
which thus corroborated the influence of the configuration
for a successful segment coupling pathway. Finally, al-
though etherification with coumarins at the congested neo-
pentylic C-11 needs further efforts, the segment coupling
proceeds in good yields on moving only one carbon further
away from the neopentylic center.
Stereochemical Assignments
Synthesis can bring essential insight into the stereochem- methyl/10-H and C-14 methyl/6β-H protons (15) confirmed
ical problems that may arise from the final relative configu- a syn geometry in each case. The map of diagnostic NOEs
ration assignment of the target.[20] Analysis of the interme- is depicted in Figure 3 for the most significant effects.
Eur. J. Org. Chem. 2010, 1483–1493
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1487

S. Arseniyadis et al.
FULL PAPER
Figure 3. Key NOESY correlations used to establish the relative stereochemistries of the A-secoterpenoid B-ring residues in the Me12/
Me15 cis series (arrows indicate diagnostic NOESY correlations for products 22bT, 27bZ, and 15). Benzyl and ethyl ester components in
15 and 27bZ have been removed to permit a clearer view of the framework conformation (Chem3D output of PM3 minimized structures).
Figure 4. Ortep view of the molecular structure of triol 22bT.
The C-12/C-15 trans-Dimethyl Series
The missing stereochemical evidence for the target ana-
logue 10 (Me12/Me15 trans relationship) was deduced after
further elaboration of the “south” part, subsequent to at-
tachment of the formyl-olefin moiety at C-5 and the homol-
ogated side chain at C-1. Upon recrystallization, the thus
obtained conjugated ester 25a afforded single crystals suit-
Figure 5. Ortep view of the molecular structure of conjugated ester
25a.
The illustrated diagnostic NOESY interactions, together
able for X-ray crystallography, which confirmed the C-8(S) with the small vicinal coupling constants between 8β-H and
absolute configuration for the 21a series (Figure 5). 7-H (upon irradiation of the Me12 group 8β-H collapses
At this stage all three asymmetric centers of the targeted into a triplet, J = 3.6 Hz), establish the configurations of
analogues 10 and 11 were established. The same chair con- the C-8, C-9, and C-10 stereocenters of 25a, featuring a
formation is adopted both in the solid state and in solution, Me12/Me15 trans relationship, as (8R,9R,10R). Finally, ad-
as demonstrated by X-ray and J-analysis, as well as by fa- ditional support was also obtained from an X-ray crystallo-
cial proximity effects measured by 2D NOESY spec- graphic analysis of the target analogue 10 (Figure 7).
troscopy. The stereochemistry of the 21a series featuring the
In summary, the assignment of the trans and cis methyl
trans-methyl relationship was further confirmed with two relationships in 14 and 15 was initially deduced from the
more intermediates by J-analysis and the observed NOEs NMR spectra and particularly from the J-analysis and the
of two later intermediates (27aZ and 14, Figure 6).
NOE data. Single-crystal X-ray diffraction analyses of the
1488
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 1483–1493

Approach to Homologues of Natural Sesquiterpene-Coumarin Ethers
Figure 6. Key NOESY correlations used to establish the relative stereochemistries of the A-secoterpenoid B-ring residues in the Me12/
Me15 trans series (arrows indicate diagnostic NOESY correlations for products 25a, 27aZ, and 14). Benzyl and ethyl ester parts (as well
as the thioacetal moiety for 27aZ) have been removed to permit a clearer view of the framework conformation (Chem3D output of PM3
minimized structures).
natural galbanic acid. The patterns of the cyclohexane core
could then be correlated with the relative configuration of
Me12 and Me15 in a specific conformation (eq-eq, ax-ax,
ax-eq, and eq-ax). A very important fixation point of our
conformation analysis is the relative relationship between
the lateral chain C1–C3 and the methyl group 13, because
allylic 1,3-strain[21] pushes this lateral chain into an axial
position, a fact confirmed by all our previously presented
experimental data. Once the validity of this approach had
Figure 7. ORTEP drawing of the X-ray structure of 10; the C1–C3
side chain prefers an axial orientation (allylic A1,3 strain).
been established, we were able to plan its use in structure
determination in the series.
The most definitive distinguishing feature of the B-ring
1
isomers is the H NMR shifts for the Me12 and Me15 sig-
nals, with those of 1 appearing at δ = 0.91 ppm (Me12
doublet) and 1.15 ppm (Me15 singlet) whereas those of 10
appear at δ = 1.01 ppm (Me15 singlet) and 1.13 ppm (Me12
doublet) and those of 11 at δ = 0.79 ppm (Me12 doublet)
and 0.81 ppm (Me15 singlet). It is noteworthy that com-
pound 11, of (8S,9R) chirality, has a more strongly upfield
shift for the Me12 doublet, a trend opposite to that found
in 10, in natural secochiliotrin [6, at δ = 1.03 ppm (Me12
doublet), 0.97 ppm (Me15 singlet)], or in suaveolindole [5
(see ref.[11]), at δ = 1.03 ppm (Me12 doublet), 1.01 ppm
(Me15 singlet)].
crystalline galbanate analogue 10 (Figure 7), of its precur-
sor derivative 25a (Figure 5), and of the triol 22bT (Fig-
ure 4), precursor of the targeted galbanate analogue 11 with
a cis-dimethyl relationship, enabled unequivocal stereo-
chemical assignment and further confirmed the correctness
of previous assignments by NMR techniques and molecular
mechanics calculations. By comparing the 1H NMR spectra
of natural galbanic acid and its higher homologue with
those of the analogues prepared in this work, we have been
able to draw an empirical correlation between the structure
of the central ring and the proton NMR spectrum. We fo-
1
cused our comparison on the H NMR window that in-
As portrayed in Figure 8, this reversal of trends for the
cludes the chemical shifts of both Me12 and Me15, because Me12/Me15 signals in A-seco terpenoids appears to be reli-
their chemical environments vary as a function of their cy- able for a fast structure assignment, including for higher
clohexane conformations. Our main objective was to be homologue series, because the ∆δ values for natural gal-
able to compare the structures of the homologated ana- banic acid (1) and for its higher methylene homologue
logues 10 and 11 with that of the galbanic acid. The first homogalbanic acid (δ = 0.83 ppm Me12 doublet, 1.03 ppm
validation of this analysis was provided by the resemblance Me15 singlet) are only 0.08 and 0.12 ppm respectively. The
of the surveyed patterns of the homogalbanic acid[8] and of striking similarities between analogue 10 and secochiliotrin
Eur. J. Org. Chem. 2010, 1483–1493
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1489

S. Arseniyadis et al.
FULL PAPER
1
Figure 8. H NMR spectra (CDCl3, 500 MHz): Me12 (doublet) and Me15 (singlet) signals showing the upfield methyl region of natural
galbanic acid (1) versus those of its C-11 homologue, the trans-dimethyl analogue 10 (“Lee-type”), the cis-dimethyl analogue 11, and
secochiliotrin (6), covering all of the four possible steroeisomeric possibilities. The conformational lock imposed by the exocyclic olefin
accounts for the portrayed conformations.
(6), the structure of which was proposed by Bohlmann
As a final note, the cell growth inhibitory activities of the
et al.,[9] call for a synthesis of the latter in order to establish synthesized analogues were evaluated against KB cells.
its structure unequivocally.
Their insignificant levels of cytotoxicity (see the Supporting
Information) make them potentially useful as biological
probes.
Conclusions
The most important consequence of this work is that
three out of four possible diastereomers could be produced
either by the chiral pool route[8] or by the domino reaction
pathway. Although the latter approach does not afford an
opening to control the C-8 configuration, access to both
configurational forms was gained.[22] Significant advantages
are that the compounds can be obtained in high purity and
that analogues can be prepared for testing, thus enabling
structure/activity studies. Further, the tested synthetic
routes should allow definitive verification of the proposed
structures, including relative and absolute stereochemistry
for the two still missing diastereomeric dispositions of the
B-ring in new A-secoterpenoids awaiting isolation.
Experimental Section
General: For general methods and standard procedures see the Sup-
porting Information.
Raney Nickel Reduction. Completion of the Synthesis of B-Ring Pre-
cursors 14 and 15: A solution of 27aZ/E (238 mg, 0.50 mmol, mix-
ture of isomers) in absolute EtOH (14 mL) was added dropwise at
room temperature under argon to a stirred suspension of Raney
nickel (50% w/v in water, 7 mL, previously washed three times with
absolute EtOH) in absolute EtOH (10 mL). The mixture was
stirred for 2 h at the same temperature. The suspension was then
diluted with EtOAc and filtered through a silica gel pad, with elu-
tion with further EtOAc. The solvent was evaporated under
1490
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 1483–1493

Approach to Homologues of Natural Sesquiterpene-Coumarin Ethers
vacuum and the residue was chromatographed on silica gel (hep-
(C-16), 101.4 (C-8Ј), 112.4 (C-4aЈ), 112.9 (C-6Ј), 113.1 (C-3Ј), 124.9
(C-4), 128.7 (C-5Ј), 131.0 (C-5), 143.4 (C-4Ј), 156.0 (C-8aЈ), 161.2
(C-2Ј), 162.4 (C-7Ј), 174.0 (C-3) ppm. IR (film): ν = 1731, 1613,
˜
tane/EtOAc 3:1) to yield 14 (100 mg, 67%) as a colorless oil. [α]2D0
1
= –30 (c = 0.9, CHCl3). H NMR (500 MHz, CDCl3): δ = 0.91 (s,
3 H, Me-15), 1.10 (d, J = 7.5 Hz, 3 H, Me-12), 1.24 (t, J = 7.1 Hz, 1279, 1230, 1122, 835, 615 cm–1. ESIMS (MeOH+CH2Cl2): m/z =
3 H, EtO), 1.35 (br. d, J = 13.1 Hz, 1 H, 7-H), 1.44–1.55 (m, 1 H, 463.2 (100) [M + Na]+. HRESIMS: calcd. for C27H36O5Na
8-H), 1.58 (d, J = 1.3 Hz, 3 H, 13-H), 1.63–1.69 (m, 1 H, 11-H),
1.70 (br. s, 3 H, Me-14), 1.73–1.87 (m, 3 H, 1-H, 2-H, 7-H), 1.92–
2.09 (m, 3 H, 1-H, 2-H, 6α-H), 2.13–2.21 (m, 1 H, 2-H), 2.26–2.35
(m, 2 H, 10-H, 6β-H), 3.73 (td, J = 5.5, 10.1 Hz, 1 H, 16-H), 3.78
(td, J = 5.7, 10.1 Hz, 1 H, 16-H), 4.09 (qd, J = 1.6, 7.1 Hz, 2 H,
EtO) ppm. 13C NMR (125 MHz, CDCl3): δ = 14.2 (EtO), 17.1 (C-
12), 20.3 (2 C, C-6, C-13), 20.5 (C-14), 24.4 (C-1), 27.0 (C-15), 30.3
(C-7), 33.2 (C-2), 36.7 (C-8), 39.6 (C-9), 41.1 (C-11), 47.0 (C-10),
59.3 (C-16), 60.2 (EtO), 124.5 (C-4), 131.3 (C-5), 174.2 (C-3) ppm.
463.2460; found 463.2461. Calcd. for C27H36O5 (440.57): C 73.61,
H 8.24; found C 73.54, H 8.36.
Formula: C27H36O5. Unit cell parameters: a = 7.588(3), b =
6.995(3), c = 22.868(5), β = 95.888(5), space group P21.
Compound 15 (55 mg, 0.18 mmol) was tosylated by the general
procedure to afford the crude tosylate (70 mg), which was used as
such in the next reaction. The tosylate (70 mg, 0.15 mmol) was cou-
pled with 28 by the general procedure for Williamson etherification
(the Reymond protocol) to afford the expected ether 11 (54 mg,
77%) after chromatography of the residue on silica gel (heptane/
EtOAc 5:1 to 3:1). Mitsunobu etherification of 15 (47 mg,
0.16 mmol) was achieved by the general procedure to give, after
flash chromatography (SiO2, heptane/EtOAc 5:1 to 3:1), the desired
ether 11 (57 mg, 81%) as a colorless oil. [α]2D0 = –55 (c = 1.3,
IR (film): ν = 3422, 1733, 1454, 1372, 1164, 1032 cm–1. ESIMS
˜
(MeOH): m/z = 319.2 (100) [M + Na]+. HRESIMS: calcd. for
C18H32O3Na 319.2249; found 319.2214. Calcd. for C18H32O3
(296.44): C 72.93, H 10.88; found C 72.89, H 10.92.
Compound 15: Raney nickel reduction of 27b (112 mg, 0.24 mmol,
mixture of E/Z isomers) as described above afforded 15 (54 mg,
77%) as a colorless oil after silica gel chromatography (heptane/
EtOAc 3:1 to 2:1): [α]2D0 = –40 (c = 0.9, CHCl3). 1H NMR
(500 MHz, CDCl3): δ = 0.73 (s, 3 H, Me-15), 0.75 (d, J = 6.7 Hz,
3 H, Me-12), 1.14–1.26 (m, 1 H, 7-H), 1.24 (t, J = 7.1 Hz, 3 H,
EtO), 1.41–1.47 (m, 1 H, 7-H), 1.57 (d, J = 1.9 Hz, 3 H, Me-13),
1.58–1.77 (m, 3 H, 8-H, 2ϫ11-H), 1.66 (d, J = 0.9 Hz, 3 H, Me-
14), 1.78–1.90 (m, 3 H, 6α-H, 2ϫ1-H), 2.03 (dt, J = 8.1, 16.2 Hz,
1 H, 2-H), 2.14 (ddd, J = 5.8, 8.1, 16.2 Hz, 1 H, 2-H), 2.38–2.45
(m, 2 H, 6β-H, 10-H), 3.71 (td, J = 5.5, 10.2 Hz, 1 H, 16-H), 3.85
(td, J = 5.6, 10.2 Hz, 1 H, 16-H), 4.09 (q, J = 7.1 Hz, 2 H,
EtO) ppm. 13C NMR (125 MHz, CDCl3): δ = 14.2 (EtO), 15.7 (C-
12), 17.8 (C-15), 20.2 (C-13), 20.3 (C-14), 21.7 (C-1), 24.5 (C-6),
31.6 (C-2), 31.8 (C-7), 35.0 (C-8), 39.4 (C-9), 41.0 (C-11), 44.5 (C-
10), 58.8 (C-16), 60.2 (EtO), 124.6 (C-4), 130.1 (C-5), 174.3 (C-
1
CHCl3). H NMR (500 MHz, CHCl3): δ = 0.80 (d, J = 6.7 Hz, 3
H, Me-12), 0.82 (s, 3 H, Me-15), 1.24 (t, J = 7.1 Hz, 3 H, OEt),
1.23–1.25 (m, 1 H, 7-H), 1.46–1.51 (m, 1 H, 7-H), 1.61 (s, 3 H, Me-
13), 1.69 (s, 3 H, Me-14), 1.72–1.78 (m, 1 H, 8-H), 1.82–1.90 (m, 4
H, 2ϫ1-H, 6-H, 11-H), 1.96 (ddd, J = 5.2, 9.7, 14.5, Hz, 1 H, 11-
H), 2.08 (td, J = 7.9, 16.3 Hz, 1 H, 2-H), 2.18 (ddd, J = 5.7, 7.6,
16.4, 7.6 Hz, 1 H, 11-H), 2.46 (dd, J = 3.1, 14.5 Hz, 1 H, 6-H),
2.51 (dd, J = 4.7, 11.0 Hz, 1 H, 10-H), 4.08–4.13 (m, 3 H, 16-H,
OEt), 4.32 (dt, J = 6.0, 9.4 Hz, 1 H, 16-H), 6.24 (d, J = 9.5 Hz, 1
H, 3Ј-H), 6.87 (dd, J = 2.0, 8.6 Hz, 1 H, 6Ј-H), 6.89 (m, 1 H, 8Ј-
H), 7.36 (d, J = 8.5 Hz, 1 H, 5Ј-H), 7.63 (d, J = 9.5 Hz, 1 H, 4Ј-
H) ppm. 13C NMR (75 MHz, CDCl3): δ = 14.2 (EtO), 15.7 (C-12),
17.9 (C-15), 20.3 (C-14), 20.4 (C-13), 21.7 (C-1), 24.5 (C-6), 31.5
(C-2), 31.6 (C-7), 35.0 (C-8), 36.8 (C-11), 39.4 (C-9), 44.5 (C-10),
60.2 (OEt), 65.1 (C-16), 101.4 (C-8Ј), 112.4 (C-4aЈ), 112.9 (C-6Ј),
113.1 (C-3Ј), 125.1 (C-4), 128.7 (C-5Ј), 129.8 (C-5), 143.4 (C-4Ј),
156.0 (C-8aЈ), 161.3 (C-2Ј), 162.4 (C-7Ј), 174.1 (C-3) ppm. IR
3) ppm. IR (film): ν = 3397, 1734, 1458, 1373, 1174, 1036 cm–1.
˜
ESIMS (MeOH): m/z = 319.1 (100) [M + Na]+. HRESIMS: calcd.
for C18H32O3Na 319.2249; found 319.2236. Calcd. for C18H32O3
(296.44): C 72.93, H 10.88; found C 72.23, H 10.97.
(film): ν = 1730, 1395, 1278, 1121, 1011, 834 cm–1. ESIMS
˜
(MeOH+CH2Cl2): m/z = 463.2 (100) [M + Na]+. HRESIMS: calcd.
Preparation of Coupling Partners and Segment Coupling: Com-
pound 14 (54 mg, 0.18 mmol) was tosylated by the general pro-
cedure to afford the crude tosylate (80 mg), which was used as such
in the next reaction. The resulting tosylate (52 mg, 0.12 mmol) was
coupled with 28 by the general procedure for Williamson etherifi-
cation (the Reymond protocol) to afford the expected ether 10
(42 mg, 80%) after chromatography of the residue on silica gel
(heptane/EtOAc 5:1 to 3:1).
for C27H36O5Na 463.2460; found 463.2475.
Preparation of the Higher Homologue Acid 10a and the Correspond-
ing Methyl Ester 10b: An aqueous solution of LiOH·H2O (21 mg
in 0.5 mL) was added at 0 °C to a solution of 10 (41 mg,
0.093 mmol) in THF (2.5 mL) and the mixture was stirred over-
night at 25 °C. It was then extracted with Et2O and the organic
layer was discarded. The aqueous layer was acidified to pH 1–2
and extracted with Et2O. The combined organic layers were washed
with brine and dried with MgSO4 and the solvent was removed
under vacuum to afford the expected acid 10a (28 mg, 74%) as a
Mitsunobu etherification of 14 (47 mg, 0.16 mmol) was achieved
by the general procedure to give, after flash chromatography (SiO2,
heptane/EtOAc 5:1 to 3:1) the desired ether 10 (57 mg, 81%) as
colorless crystals; m.p. 72.3–73.0 °C (hexane). [α]2D0 = –11 (c = 0.9,
CHCl3). 1H NMR (500 MHz, CDCl3): δ = 1.01 (s, 3 H, Me-15),
1
glassy solid. [α]2D0 = –13.0 (c = 1.4, CHCl3). H NMR (500 MHz,
CDCl3): δ = 1.02 (s, 3 H, Me-15), 1.14 (d, J = 7.5 Hz, 3 H, Me-
1.14 (d, J = 7.5 Hz, 3 H, Me-12), 1.24 (t, J = 7.1 Hz, 3 H, EtO), 12), 1.23–1.29 (m, 1 H, 7-H), 1.38–1.45 (m, 1 H, 7-H), 1.56–1.63
1.41 (br. d, J = 13.0 Hz, 1 H, 7-H), 1.58–1.62 (m, 1 H, 8-H), 1.61 (m, 1 H, 8-H), 1.62 (d, J = 1.5 Hz, 1 H, Me-13), 1.73 (br. s, 3 H,
(br. s, 3 H, Me-13), 1.73 (br. s, 3 H, Me-14), 1.76–1.94 (m, 3 H, 1- Me-14), 1.78–1.94 (m, 3 H, 1-H, 6-H, 11-H), 1.96–2.19 (m, 3 H, 1-
H, 7-H, 11-H), 1.96–2.14 (m, 4 H, 1-H, 2-H, 4-H, 6α-H), 2.15–2.23 H, 2-H, 11-H), 2.26 (ddd, J = 5.6, 8.5, 16.4 Hz, 1 H, 2-H), 2.34–
(m, 1 H, 2-H), 2.33–2.39 (m, 2 H, 6β-H, 10-H), 4.04–4.16 (m, 3 H, 2.41 (m, 2 H, 6-H, 10-H), 4.11 (td, J = 5.5, 9.3 Hz, 1 H, 16-H),
16-H, EtO), 4.21 (td, J = 5.8, 9.3 Hz, 1 H, 16-H), 6.24 (d, J =
9.5 Hz, 1 H, 3Ј-H), 6.83–6.87 (m, 2 H, 6Ј-H, 8Ј-H), 7.36 (d, J =
4.21 (td, J = 5.8, 9.3 Hz, 1 H, 16-H), 6.24 (d, J = 9.5 Hz, 1 H, 3Ј-
H), 6.82–6.87 (m, 2 H, 6Ј-H, 8Ј-H), 7.36 (d, J = 9.2 Hz, 1 H, 5Ј-
9.1 Hz, 1 H, 5Ј-H), 7.63 (d, J = 9.5 Hz, 1 H, 4Ј-H) ppm. 13C NMR H), 7.63 (d, J = 9.5 Hz, 1 H, 4Ј-H) ppm. 13C NMR (125 MHz,
(125 MHz, CDCl3): δ = 14.3 (EtO), 17.2 (C-12), 20.2 (C-6), 20.4
CDCl3): δ = 17.2 (C-12), 20.2 (C-6), 20.3 (C-13), 20.6 (C-14), 24.3
(C-13), 20.5 (C-14), 24.4 (C-1), 27.0 (C-15), 30.2 (C-7), 33.1 (C-2),
(C-1), 27.0 (C-15), 30.2 (C-7), 32.5 (C-2), 36.7 (C-8), 36.8 (C-4),
36.6 (C-8), 36.8 (C-11), 39.6 (C-9), 47.1 (C-10), 60.2 (EtO), 65.5 39.6 (C-9), 47.0 (C-10), 65.5 (C-16), 101.4 (C-8Ј), 112.4 (C-4aЈ),
Eur. J. Org. Chem. 2010, 1483–1493
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1491

S. Arseniyadis et al.
FULL PAPER
112.9 (C-3Ј), 113.1 (C-6Ј), 125.2 (C-4), 128.7 (C-5Ј), 130.8 (C-5),
after chromatography of the residue on silica gel (heptane/EtOAc
143.4 (C-4Ј), 156.0 (C-8aЈ), 161.3 (C-2Ј), 162.4 (C-7Ј), 178.5 (C- 4:1 to 1:1). Mitsunobu etherification of 14 (5.0 mg, 0.017 mmol)
3) ppm. IR (film): ν = 1731, 1705, 1612, 1280, 1230, 1124, 834 cm–1.
was achieved by the general procedure to give, after flash
chromatography (SiO2, heptane/EtOAc 4:1 to 1:1) the desired ether
30a (5.1 mg, 61%) as colorless crystals; m.p. 72.3–73.0 °C (hexane).
[α]2D0 = –10 (c = 0.6, CHCl3). 1H NMR (500 MHz, CDCl3): δ =
0.95 (s, 3 H, Me-15), 1.14 (d, J = 7.5 Hz, 3 H, Me-12), 1.24 (t, J =
7.2 Hz, 3 H, EtO), 1.37 (br. d, J = 13.3 Hz, 1 H, 7-H), 1.50–1.57
(m, 1 H, 8-H), 1.59 (br. s, 3 H, 13-H), 1.71 (br. s, 3 H, Me-14), 1.78
(tt, J = 4.6, 13.9 Hz, 1 H, 7-H), 1.83–1.95 (m, 2 H, 1-H, 11-H),
1.95–2.10 (m, 3 H, 1-H, 2-H, 6-H), 2.11–2.22 (m, 2 H, 2-H, 11-H),
2.28–2.37 (m, 2 H, 6-H, 10-H), 3.89 (s, 3 H, MeO), 4.04 (s, 3 H,
MeO), 4.08 (q, J = 7.2 Hz, 2 H, EtO), 4.15–4.26 (m, 2 H, 16-H),
6.34 (d, J = 9.6 Hz, 1 H, 3Ј-H), 6.66 (s, 1 H, 5Ј-H), 7.61 (d, J =
9.6 Hz, 1 H, 4Ј-H) ppm. 13C NMR (125 MHz, CDCl3): δ = 14.2
(EtO), 17.1 (C-12), 20.3 (2 C, C-6, C-13), 20.5 (C-14), 24.5 (C-1),
27.0 (C-15), 30.3 (C-7), 33.4 (C-2), 36.5 (C-8), 37.9 (C-4), 39.5 (C-
9), 47.3 (C-10), 56.3 (MeO), 60.1 (EtO), 61.8 (MeO), 71.3 (C-16),
103.7 (C-5Ј), 114.4 (C-4aЈ), 115.2 (C-3Ј), 124.6 (C-4), 131.3 (C-5),
141.7 (C-8Ј), 143.1 (C-8aЈ), 143.4 (C-4Ј), 145.4 (C-7Ј), 150.6 (C-6Ј),
˜
ESIMS (MeOH+CH2Cl2): m/z = 435.2 (100) [M + Na]+. HRES-
IMS: calcd. for C25H32O5Na 435.2147; found 435.2182.
Preparation of the Higher Homologue 10b:
A solution of
TMSCHN2 in Et2O (2 , 0.22 mL, 0.44 mmol) was added at 0 °C
to a solution of the carboxylic acid 10a (28 mg, 0.068 mmol) in
methanol (1 mL) and the mixture was stirred for 2 h at room tem-
perature. The mixture was concentrated under vacuum and
chromatography of the residue on silica gel (heptane/EtOAc 5:1 to
2:1) afforded the methyl ester 10b (20 mg, 69%) as colorless crys-
1
tals. M.p. 111–112 °C. [α]2D0 = –12.1 (c = 1.01, CHCl3). H NMR
(500 MHz, CDCl3): δ = 1.01 (s, 3 H, Me-15), 1.13 (d, J = 7.5 Hz,
3 H, Me-12), 1.40 (br. d, J = 14.1 Hz, 1 H, 7-H), 1.55–1.60 (m, 1
H, 8-H), 1.59 (d, J = 1.2 Hz, 3 H, Me-13), 1.72 (br. s, 3 H, Me-
14), 1.76–1.93 (m, 3 H, 1-H, 7-H, 11-H), 1.95–2.13 (m, 4 H, 1-H,
2-H, 4-H, 6α-H), 2.17–2.25 (m, 1 H, 2-H), 2.31–2.39 (m, 2 H, 6β-
H, 10-H), 3.63 (s, 3 H, OMe), 4.11 (td, J = 5.4, 9.4 Hz, 1 H, 16-
H), 4.19 (td, J = 5.9, 9.4 Hz, 1 H, 16-H), 6.23 (d, J = 9.5 Hz, 1 H,
3Ј-H), 6.82–6.88 (m, 2 H, 6Ј-H, 8Ј-H), 7.36 (d, J = 9.3 Hz, 1 H, 5Ј-
H), 7.63 (d, J = 9.5 Hz, 1 H, 4Ј-H) ppm. 13C NMR (125 MHz,
CDCl3): δ = 17.2 (C-12), 20.2 (C-6), 20.3 (C-13), 20.5 (C-14), 24.4
(C-1), 27.0 (C-15), 30.2 (C-7), 32.9 (C-2), 36.6 (C-8), 36.8 (C-11),
39.6 (C-9), 47.1 (C-10), 51.4 (OMe), 65.5 (C-16), 101.4 (C-8Ј), 112.4
(C-4aЈ), 112.9 (C-6Ј), 113.0 (C-3Ј), 124.9 (C-4), 128.7 (C-5Ј), 130.9
(C-5), 143.4 (C-4Ј), 155.9 (C-8aЈ), 161.2 (C-2Ј), 162.4 (C-7Ј), 174.4
160.5 (C-2Ј), 174.0 (C-3) ppm. IR (film): ν = 1729, 1605, 1563,
˜
1455, 1408, 1290, 1152, 1124, 1042, 983, 845 cm–1. ESIMS
(MeOH+CH2Cl2): m/z (%) = 523.2 (100) [M + Na]+. HRESIMS:
calcd. for C29H40O7Na 523.2672; found 523.2662.
Compound 15 (36 mg, 0.12 mmol) was tosylated by the general
procedure to afford the crude tosylate (52 mg), which was used as
such in the next reaction. The resulting tosylate (14 mg,
0.031 mmol) was coupled with 29 by the general procedure for Wil-
liamson etherification (the Reymond protocol) to afford the ex-
pected ether 30b (11 mg, 69%) after chromatography of the residue
on silica gel (heptane/EtOAc 4:1 to 1:1). Mitsunobu etherification
of 15 (5.0 mg, 0.017 mmol) was achieved by the general procedure
to give, after flash chromatography (SiO2, heptane/EtOAc 4:1 to
(C-3) ppm. IR (film): ν = 1732, 1613, 1279, 1230, 1122, 835 cm–1.
˜
ESIMS (MeOH+CH2Cl2): 449.2 (100) [M + Na]+. HRESIMS:
calcd. for C26H34O5Na 449.2304; found 449.2280.
Preparation of the Higher Homologue Methyl Ester 11b: Saponifica-
tion/re-esterification of 11 (41 mg, 0.093 mmol) as described above
afforded the methyl ester 11b (20 mg, 69%) as colorless crystals
1:1), the desired ether 30b (5.2 mg, 61%) as a colorless oil. [α]2D0
=
after silica gel chromatography (heptane/EtOAc 5:1 to 2:1). [α]2D0
=
1
–29 (c = 1.0, CHCl3). H NMR (500 MHz, CDCl3): δ = 0.79 (s, 3
H, Me-15), 0.80 (d, J = 7.1 Hz, 3 H, Me-12), 1.22 (t, J = 7.2 Hz,
3 H, OEt), 1.22–1.26 (m, 1 H, 7-H), 1.45–1.49 (m, 1 H, 7-H), 1.58
(br. s, 3 H, Me-13), 1.67 (br. s, 3 H, Me-14), 1.71–1.78 (m, 1 H, 8-
H), 1.81–1.93 (m, 4 H, 2ϫ1-H, 6-H, 11-H), 1.99–2.05 (m, 2 H, 2-
H, 11-H), 2.13 (ddd, J = 5.5, 9.2, 16.2 Hz, 1 H, 2-H), 2.41–2.46
(m, 2 H, 6-H, 10-H), 3.89 (s, 3 H, OMe), 4.04 (s, 3 H, OMe), 4.07
(q, J = 7.1 Hz, 2 H, OEt), 4.18 (ddd, J = 5.2, 8.6, 10.5 Hz, 1 H,
16-H), 4.29 (ddd, J = 5.8, 9.0, 10.3 Hz, 1 H, 16-H), 6.34 (d, J =
9.5 Hz, 1 H, 3Ј-H), 6.66 (s, 1 H, 5Ј-H), 7.6 (d, J = 9.5 Hz, 1 H, 4Ј-
H) ppm. 13C NMR (125 MHz, CDCl3): δ = 14.2 (OEt), 15.7 (C-
12), 17.7 (C-15), 20.2 (C-14), 20.4 (C-13), 21.9 (C-1), 24.6 (C-6),
31.6 (C-7), 32.0 (C-2), 35.1 (C-8), 37.9 (C-11), 39.4 (C-9), 44.7 (C-
10), 56.3 (OMe), 60.1 (OEt), 61.8 (OMe), 71.0 (C-16), 103.6 (C-5Ј),
114.4 (C-4aЈ), 115.2 (C-3Ј), 124.7 (C-4), 130.1 (C-5), 141.7 (C-8Ј),
143.1 (C-8aЈ), 143.4 (C-4Ј), 145.4 (C-7Ј), 150.6 (C-6Ј), 160.5 (C-2Ј),
1
–29 (c = 1.2, CHCl3). H NMR (500 MHz, CHCl3): δ = 0.79 (d, J
= 6.5 Hz, 3 H, Me-12), 0.81 (s, 3 H, Me-15), 1.22–1.28 (m, 1 H, 7-
H), 1.46–1.51 (m, 1 H, 7-H), 1.59 (d, J = 2.0 Hz, 3 H, Me-13), 1.69
(s, 3 H, Me-14), 1.72–1.78 (m, 1 H, 8-H), 1.82–1.90 (m, 4 H, 2ϫ1-
H, 6-H, 11-H), 1.96 (ddd, J = 5.2, 10.2, 13.7, Hz, 1 H, 11-H), 2.09
(dt, J = 8.1, 16.5 Hz, 1 H, 2-H), 2.19 (ddd, J = 5.4, 7.8, 16.6 Hz, 1
H, 11-H), 2.45 (br. d, J = 14.3 Hz, 1 H, 6-H), 2.50 (dd, J = 4.2,
11.2 Hz, 1 H, 10-H), 3.64 (s, 3 H, OMe), 4.10–4.13 (td, J = 5.1,
9.6 Hz, 1 H, 16-H), 4.31 (td, J = 5.9, 9.5 Hz, 1 H, 16-H), 6.23 (d,
J = 10.0 Hz, 1 H, 3Ј-H), 6.87 (dd, J = 2.4, 8.4 Hz, 1 H, 6Ј-H), 6.89
(m, 1 H, 8Ј-H), 7.36 (d, J = 8.5 Hz, 1 H, 5Ј-H), 7.63 (d, J = 9.4 Hz,
1 H, 4Ј-H) ppm. 13C NMR (75 MHz, CDCl3): δ = 15.7 (C-12), 17.8
(C-15), 20.2 (C-14), 20.4 (C-13), 21.7 (C-1), 24.4 (C-6), 31.3 (C-2),
31.6 (C-7), 35.0 (C-8), 36.7 (C-11), 39.3 (C-9), 44.5 (C-10), 51.5
(OMe), 65.1 (C-16), 101.4 (C-8Ј), 112.3 (C-4aЈ), 112.8 (C-6Ј), 113.0
(C-3Ј), 125.1 (C-4), 128.7 (C-5Ј), 129.7 (C-5), 143.4 (C-4Ј), 155.9
174.0 (C-3) ppm. IR (film): ν = 1732, 1559, 1457, 1409, 1289, 1150,
˜
1124, 1042, 668 cm–1. ESIMS (MeOH+CH2Cl2): m/z = 523.2 (100)
[M + Na]+. HRESIMS: calcd. for C29H40O7Na 523.2672; found
523.2670.
(C-8aЈ), 161.2 (C-2Ј), 162.4 (C-7Ј), 174.4 (C-3) ppm. IR (film): ν =
˜
1731, 1612, 1279, 1122, 1012, 834 cm–1. ESIMS (MeOH+CH2Cl2):
m/z = 449.2 (100) [M + Na]+. HRESIMS: calcd. for C26H34O5Na
449.2304; found 449.2300.
CCDC-746923 (for 10), -746924 (for 22bT), and -746925 (for 25a)
contain the supplementary crystallographic data for this paper.
These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
Synthesis of Secodrial Analogues
Preparation of Coupling Partners and Segment Coupling: Com-
pound 14 (54 mg, 0.18 mmol) was tosylated by the general pro-
cedure to afford the crude tosylate (52 mg), which was used as such
in the next reaction. The tosylate (14 mg, 0.031 mmol) was coupled
with 29 by the general procedure for Williamson etherification (the
Reymond protocol) to afford the expected ether 30a (11 mg, 69%)
Supporting Information (see also the footnote on the first page of
this article): Complete characterization data and 1H and 13C NMR
spectra for all new compounds.
1492
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 1483–1493

Approach to Homologues of Natural Sesquiterpene-Coumarin Ethers
[13] a) L. F. Tietze, U. Beifuss, Angew. Chem. 1993, 105, 115–136;
Angew. Chem. Int. Ed. Engl. 1993, 32, 131–163; b) L. F. Tietze,
Chem. Rev. 1996, 96, 115–136; c) L. F. Tietze, F. Haunert, in:
Stimulating Concepts in Chemistry (Eds.: M. Shibasaki, J. F.
Stoddart, F. Vögtle), Wiley-VCH, Weinheim, 2000, pp. 39–64;
d) Domino Reactions in Organic Synthesis (Eds.: L. F. Tietze,
G. Brasche, K. M. Gericke), Wiley-VCH, 2006; ISBN 3-527-
29060-5.
[14] a) For an application to the taxoid ABC diterpene core see:
J. I. Martín Hernando, J. Quílez del Moral, M. Rico Ferreira,
J. I. Candela Lena, S. Arseniyadis, Tetrahedron: Asymmetry
1999, 10, 783–797; b) For an application to the first enantiose-
lective synthesis of iridal see: A. Corbu, M. Aquino, T. V. Pra-
tap, P. Retailleau, S. Arseniyadis, Org. Lett. 2008, 10, 1787–
1790.
Acknowledgments
Financial support from the Centre National de la Recherche Sci-
entifique (CNRS) and fellowship awards from the Institut de Chi-
mie des Substances Naturelles (ICSN) are gratefully acknowledged.
[1] Structure and stereochemistry of galbanic acid from galbanum:
V. Y. Bagirov, V. I. Sheichenko, N. V. Veselovskaya, Y. E.
Sklyar, A. A. Savina, I. Kir’yanova, Khim. Prir. Soedin. 1980,
16, 620–623.
[2] Matsuda et al., see ref.[8], named this compound marneral (and
its corresponding alcohol marnerol) in recognition of Marner’s
pioneering work on iridals. Marneral has never been isolated
but was hypothesized to be the first carbocyclic precursor in
the biosynthesis of iridals; a) F.-J. Marner, W. Krick, B.
Gellrich, L. Jaenicke, J. Org. Chem. 1982, 47, 2531–2538; b) F.-
J. Marner, Curr. Org. Chem. 1997, 1, 153–186; c) M. Lamshöft,
H. Schmickler, F. J. Marner, Eur. J. Org. Chem. 2003, 727–733.
[3] A. Corbu, M. Perez, M. Aquino, P. Retailleau, S. Arseniyadis,
Org. Lett. 2008, 10, 2853–2856.
[15] A. Corbu, G. Gauron, J. M. Castro, M. Dakir, S. Arseniyadis,
Tetrahedron: Asymmetry 2008, 19, 1730–1743.
[16] W. Lehnert, Tetrahedron 1973, 29, 635–638. Several protocols
were examined; all conditions failed in our hands.
[17] a) W. G. Dauben, D. M. Michno, J. Org. Chem. 1977, 42, 643–
685; b) J. H. Babler, D. O. Olsen, W. H. Arnold, J. Org. Chem.
1974, 39, 1656–1658; c) G. Büchi, B. Egger, J. Org. Chem. 1971,
36, 2021–2023.
[4] Isolation from asafetida and biogenetic derivation: a) G. Ap-
pendino, S. Tagliapietra, G. Nano, J. Jakupovic, Phytochemistry
1994, 35, 183–186; b) G. Appendino, L. Maxia, M. Bascope,
P. J. Houghton, G. Sanchez-Duffhues, E. Muñoz, O. Sterner, J.
Nat. Prod. 2006, 69, 1101–1104.
[18] See ref.[8]
[5] T. K. Khasanov, A. I. Saidkhodzhaev, G. K. Nikonov, Chem.
Nat. Compd. 1974, 10, 20–23.
[6] E. M. González-García, J. Grognux, D. Wahler, J. L. Reymond,
Helv. Chim. Acta 2003, 86, 2458–2470.
[7] For the revised, though erroneous, structure of galbanic acid
suggested by Lee et al., see: S. G. Lee, S. Ryu, J. W. Ahn, Bull.
Korean Chem. Soc. 1998, 19, 384–386.
[8] A. Corbu, M. Aquino, M. Perez, Z. Gandara, S. Arseniyadis,
Eur. J. Org. Chem. 2009, 6386–6392.
[9] J. Jakupovic, S. Banerjee, F. Bohlmann, R. M. King, Tetrahe-
dron 1986, 42, 1305–1313.
[10] With the exception of the contribution reported recently by our
group there appear to have been no prior synthetic studies.
[11] a) H. D. Yoo, P. A. Cremin, L. Zeng, E. Garo, C. T. Williams,
C. M. Lee, M. G. Goering, M. O’Neil-Johnson, G. R. Eldridge,
[19] Methyl galbanate and galbanic acid were screened in an in vi-
tro antiinfluenza A viral (H1N1) assay, with amantadine as a
positive control. They are reported to show good potency
against influenza A virus H1N1 (IC50 0.26 and 0.45 mgmL–1),
with the C-3 methyl ester showing better potency than its cor-
responding free (galbanic) acid: C. L. Lee, L. C. Chiang, L. H.
Cheng, C. C. Liaw, M. H. Abd El-Razek, F. R. Chang, Y. C.
Wu, J. Nat. Prod. 2009, 72, 1568–1572.
[20] K. C. Nicolaou, S. Snyder, Angew. Chem. Int. Ed. 2005, 44,
1012.
[21] F. Johnson, Chem. Rev. 1968, 68, 375–413.
J. F. Hu, J. Nat. Prod. 2005, 68, 122–124; b) E. J. Velthuisen, [22] The numbering throughout the text is intended to simplify dis-
S. J. Danishefsky, J. Am. Chem. Soc. 2007, 129, 10640–10641.
[12] a) L. Finet, J. I. Candela Lena, T. Kaoudi, N. Birlirakis, S. Ar-
seniyadis, Chem. Eur. J. 2003, 9, 3813–3820; b) A. Corbu, G.
Gauron, J. M. Castro, M. Dakir, S. Arseniyadis, Org. Lett.
2007, 9, 4745–4748.
cussion by maintaining a fixed designation for a particular car-
bon throughout a series of structures. It follows the numbering
adopted by Bagirov[1] and Appendino.[4]
Received: November 24, 2009
Published Online: January 29, 2010
Eur. J. Org. Chem. 2010, 1483–1493
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1493