Catalysis-based and protecting-group-free total syntheses of the marine oxylipins hybridalactone and the ecklonialactones A, B, and C

Article abstract of DOI:10.1021/ja204027a

Concise and protecting-group-free total syntheses of the marine oxylipins hybridalactone (1) and three members of the ecklonialactone family (2-4) were developed. They deliver these targets in optically pure form in 14 or 13 steps, respectively, in the longest linear sequence; five of these steps are metal-catalyzed and four others are metal-mediated. The route to either 1 or 2-4 diverges from the common building block 22, which is accessible in 7 steps from 2[5H]furanone by recourse to a rhodium-catalyzed asymmetric 1,4-addition reaction controlled by the carvone-derived diene ligand 35 and a ring-closing alkene metathesis (RCM) catalyzed by the ruthenium indenylidene complex 17 as the key operations. Alternatively, 22 can be made in 10 steps from furfural via a diastereoselective three-component coupling process. The further elaboration of 22 into hybridalactone as the structurally most complex target with seven contiguous chiral centers was based upon a sequence of cyclopropanation followed by a vanadium-catalyzed epoxidation, both of which were directed by the same free hydroxy group at C15. The macrocyclic scaffold was annulated to the headgroup by means of a ring-closing alkyne metathesis reaction (RCAM). In response to the unusually high propensity of the oxirane of the targeted oxylipins for ring opening, this transformation had to be performed with complexes of the type [(Ar₃SiO)₄Mo≡CPh] [K·OEt₂] (43), which represent a new generation of exceedingly tolerant yet remarkably efficient catalysts. Their ancillary triarylsilanolate ligands temper the Lewis acidity of the molybdenum center but are not sufficiently nucleophilic to engage in the opening of the fragile epoxide ring. A final semireduction of the cycloalkyne formed in the RCAM step to the required (Z)-alkene completed the total synthesis of (-)-1. The fact that the route from the common fragment 22 to the ecklonialactones could follow a similar logic showcased the flexibility inherent to the chosen approach.

Full text of DOI:10.1021/ja204027a

ARTICLE
pubs.acs.org/JACS
Catalysis-Based and Protecting-Group-Free Total Syntheses of the
Marine Oxylipins Hybridalactone and the Ecklonialactones A, B, and C
Volker Hickmann, Azusa Kondoh, Barbara Gabor, Manuel Alcarazo, and Alois F€urstner*
Max-Planck-Institut f€ur Kohlenforschung, D-45470 M€ulheim/Ruhr, Germany
S Supporting Information
b
ABSTRACT: Concise and protecting-group-free total syntheses of
the marine oxylipins hybridalactone (1) and three members of the
ecklonialactone family (2ꢀ4) were developed. They deliver these
targets in optically pure form in 14 or 13 steps, respectively, in the
longest linear sequence; five of these steps are metal-catalyzed and
four others are metal-mediated. The route to either 1 or 2ꢀ4 diverges
from the common building block 22, which is accessible in 7 steps
from 2[5H]furanone by recourse to a rhodium-catalyzed asymmetric
1,4-addition reaction controlled by the carvone-derived diene ligand
35 and a ring-closing alkene metathesis (RCM) catalyzed by the ruthenium indenylidene complex 17 as the key operations.
Alternatively, 22 can be made in 10 steps from furfural via a diastereoselective three-component coupling process. The further
elaboration of 22 into hybridalactone as the structurally most complex target with seven contiguous chiral centers was based upon a
sequence of cyclopropanation followed by a vanadium-catalyzed epoxidation, both of which were directed by the same free hydroxy
group at C15. The macrocyclic scaffold was annulated to the headgroup by means of a ring-closing alkyne metathesis reaction
(RCAM). In response to the unusually high propensity of the oxirane of the targeted oxylipins for ring opening, this transformation
had to be performed with complexes of the type [(Ar3SiO)4MotCPh][K OEt2] (43), which represent a new generation of
3
exceedingly tolerant yet remarkably efficient catalysts. Their ancillary triarylsilanolate ligands temper the Lewis acidity of the
molybdenum center but are not sufficiently nucleophilic to engage in the opening of the fragile epoxide ring. A final semireduction of
the cycloalkyne formed in the RCAM step to the required (Z)-alkene completed the total synthesis of (ꢀ)-1. The fact that the route
from the common fragment 22 to the ecklonialactones could follow a similar logic showcased the flexibility inherent to the chosen
approach.
’ INTRODUCTION
As is now well documented in the literature, this conjecture also
turned out to be correct.7,8
Hybridalactone (1) was the first marine oxylipin to be isolated
that contained a cyclopropane and a macrolactone moiety.1
Derived from antimicrobial extracts of the red alga Laurencia
hybrida, the constitution of this then very unusual secondary
metabolite was established by the isolation team, and a partial
assignment of its relative stereochemistry could be made on the
basis of the recorded spectral data.1 However, it was a captivating
biosynthetic hypothesis, which allowed Corey and co-workers to
anticipate the entire stereostructure as well as the absolute
configuration of 1 (Scheme 1).2
According to this pathway, the hydroperoxide formed upon S-
selective 12-lipoxygenation3 of eicosapentaenoic acid enters a
cationic manifold, which engenders the selective formation of
both carbocyclic rings and the macrolide backbone as shown in
Scheme 1. This prediction was confirmed by the total synthesis of
(ꢀ)-14 as well as by a crystal structure analysis of a bromohydrine
derived from an authentic sample.2,5 From the conceptual view-
point, Corey’s biosynthetic proposal innately linked the meta-
bolic oxygenation pathways of polyunsaturated fatty acids with
the intriguing chemistry of “nonclassical” carbocations of the
homoallyl/cyclopropylmethyl type,6 which implied the occur-
rence of many more cyclopropyl-containing oxylipins in nature.
Various brown algae were later found to produce closely
related oxylipins with C-18 carbon chains (rather than the
C-20 skeleton of 1).9ꢀ11 Although formally classified into three
different series, the individual members of the ecklonialactone,
eiseniahalide, and egregiachloride families likely derive from a
common biosynthetic pathway; this path must closely follow the
initial steps outlined above for hybridalactone but is less complex
in that it lacks the “nonclassical cation” component (Scheme 2).12
Whereas the ecklonialactones A (2) and B (3) originate from an
intramolecular attack of the carboxylic acid terminus on the
cation generated upon lipoxygenation of the fatty acid and
subsequent cyclopentane formation, interception of the very
same intermediate by external chloride leads to the egregiachlor-
ides 8 and 9.13 Ecklonialactones C (4) and D (5) as well as the
eiseniahalides 6 and 7 derive from the parent compounds 2 and 3
by nucleophilic opening of the epoxide ring.9b,11 Another inter-
esting variation is borne out in oxylipin 10 isolated from Eisenia
bicyclis, in which reaction of the proximate double bond in 2 with
Received: May 2, 2011
Published: July 22, 2011
r
2011 American Chemical Society
13471
dx.doi.org/10.1021/ja204027a J. Am. Chem. Soc. 2011, 133, 13471–13480
|

Journal of the American Chemical Society
ARTICLE
Scheme 1. Proposed Biosynthetic Pathway Explaining the
Constitution and Stereostructure of Hybridalactone (ꢀ)-1
Scheme 2. Biosynthesis of the Ecklonialactones and Related
Oxylipins with a C-18 Chain
the labile epoxide ring gives rise to an annulated cyclopropane
motif; the configuration of this particular compound at C9,
however, has not yet been elucidated.11
The fascinating structural diversity of these and related marine
oxylipins originating from now well-understood biosynthetic
machinery stands in marked contrast to the lack of information
about their biological role in the producing organisms. It is
generally assumed that they are part of the chemical defense
mechanism of the algae against herbivores and/or microbial
pathogens and could therefore be interesting lead structures of
proven low cytotoxicity.9b,14 Yet, the limited supply of any of
these marine lipids hampered more detailed assessments in the
past. We now present a concise entry into hybridalactone and the
ecklonialactone family via a common advanced precursor.15,16
The stunning chemoselectivity of contemporary homogeneous
catalysis in general and the ability to rigorously distinguish
between alkenes and alkynes in oxidative, reductive, as well as
metathetic maneuvers in particular constitute the key enabling
features. As a result, the final routes to these unusual lipids of
marine origin are short, productive, enantioselective, and devoid
of any protecting group manipulations.
total synthesis has to cope, more often than not, with protecting
group chemistry as a necessary evil. Such manipulations are
structurally unproductive and hence adversely affect all desirable
“economies” of synthesis.18 The pursuit of increasingly complex
targets in a protecting group-free format poses considerable
practical and intellectual challenges yet constitutes an inspira-
tional framework for chemical invention.19 A closer analysis of
the published success stories, however, shows that a significant
part of them deals with alkaloid or terpenoid targets, whereas
polyketides and carbohydrates are rare, and even lipids are
underrepresented.19ꢀ21
Following up on our previous investigations on prostanoids of
marine origin,22,23 we were committed to developing a concise
approach to hybridalactone under the premise of minimizing
protecting group manipulations; in the ideal case, none would be
required en route to this lipidic target containing seven contig-
uous chiral centers. At the same time, the chosen route should be
flexible enough to allow for a late-stage bifurcation toward the
ecklonialactone series as well.
’ RESULTS AND DISCUSSION
Strategic Considerations. Despite the many tremendous
advances during the last decades,17 contemporary natural product
13472
dx.doi.org/10.1021/ja204027a |J. Am. Chem. Soc. 2011, 133, 13471–13480

Journal of the American Chemical Society
ARTICLE
Scheme 3. Retrosynthetic Analysis of Hybridalactone and
the Ecklonialactones, Diverging at the Common Synthon E
Scheme 4a
a Reagents and conditions: a) O2, rose Bengal, hν, MeOH, 79%; b) L-
menthol, CSA cat., toluene, reflux, then recrystallization from hexanes,
cf. ref 33; c) H2CdCHMgBr, CuI, THF, ꢀ78 °C, then allyl iodide, ꢀ60 °C,
2 h, 86%; d) aq. TFA, 16 h, 95%; e) NaBH4, MeOH, 2 h, then HCl in
Et2O, reflux, 1 h, 75%; f) Me(MeO)NH HCl, Me3Al, CH2Cl2, 4 h; g)
3
17 (4 mol %), CH2Cl2, 16 h, 74% (over both steps); h) Dess-Martin
periodinane, NaHCO3, CH2Cl2, 1.5 h, 73%; i) 2-chloro-4-nitrobenzoyl
chloride, Et3N, DMAP cat., CH2Cl2, 16 h, 65%; j) 21, K2CO3, MeOH,
16 h, 75%; k) LiHMDS, MeOTf, THF, ꢀ78 °C, 2.5 h, 80%.
Our retrosynthetic analysis (Scheme 3) suggested that the
proven orthogonal character of ring-closing alkene metathesis
(RCM, F f E)24 and ring-closing alkyne metathesis (RCAM,
A f 1, B f 2,3)25,26 could be favorably exploited for the
assembly of the carbon skeleton,27 whereas a properly configured
C15-OH group in D might serve as a strategic relay to direct both
the asymmetric cyclopropanation of a cis-configured allylic
double bond and the epoxidation of the more distant olefinic
site within the five-membered headgroup. Fragment D was
envisaged to derive from E; this building block would also serve
as the cornerstone en route to the ecklonialactones, which have a
shorter carbon chain but a larger, 14-membered lactone ring.
Further analysis then leads back via F to commercial 2[5H]-
furanone (G, X = H) or derivatives thereof (X ¼ H), which
ideally might be functionalized with the required olefinic appen-
dices in a one-pot, three-component coupling process.28,29
Original Auxiliary-Based Route to the Headgroups. In an
initial foray, recourse was made to the proven versatility of L-(ꢀ)-
menthol as a chiral auxiliary30 since butenolide 12 is known to
participate well in conjugate addition reactions of various nu-
cleophiles including organocopper reagents (Scheme 4).31
Moreover, 12 is available from cheap furfural (11) in two high-
yielding operations without need for a chromatographic
purification.32,33 Addition of the copper reagent derived from
vinylmagnesium bromide and CuI to a solution of 12 in THF
Figure 1. Structure of compound 20 in the solid state.34
followed by an allyl iodide quench furnished the desired three-
component adduct 13 in 86% yield as a single diastereomer on a
multigram scale. Acid-catalyzed cleavage of the auxiliary and
subsequent reduction of the resulting hemiacylal 14 with NaBH4
gave butanolide 15 in excellent overall yield.
This product was then ring opened to the corresponding
Weinreb amide 16,35 which was found to revert to lactone 15
13473
dx.doi.org/10.1021/ja204027a |J. Am. Chem. Soc. 2011, 133, 13471–13480

Journal of the American Chemical Society
ARTICLE
Scheme 5a
Scheme 6a
a Reagents and conditions: a) [Rh(cod)(MeCN)][BF4] (1.5 mol %),
(S)-BINAP (1.6 mol %), THF, H2O, 65 °C, 31a (65%, 82% ee), 31b
(17%, 86% ee), 31c (64%, 73% ee), see Text; b) [Rh(C2H4)2Cl]2
(1.5 mol %), 35 (3.3 mol %), 1,4-dioxane, aq. KOH, 72 h, 31c (52%, 80%
ee, 93% ee after recryst.); c) LDA, THF, ꢀ78 °C, then allyl iodide, 80 min,
87%; d) HN(OMe)Me HCl, Me3Al, CH2Cl2, 2 h, 0 °C; e) 17 (8 mol %),
3
CH2Cl2, 16 h, 75% (over both steps); f) see Schemes 4 and 5.
with L-selectride at ꢀ78 °C to deliver the required alcohol
segment 24 as a single isomer. The stereochemical outcome is
in line with a prototype FelkinꢀAhn transition state and was
confirmed by Mosher ester analysis at a later stage of the
synthesis after epoxidation of the double bond.34
The formation of the analogous allylic alcohol 28 required for
the synthesis of hybridalactone turned out to be more delicate.
Gram quantities of the necessary alkenyl bromide 25 were
conveniently obtained in isomerically pure form (Z:E g 96:4)
by a base-promoted decarboxylative elimination of 2,3-dibromo-
pentanoic acid, which itself is available on large scale by addition
of Br2 to 2E-pentenoic acid in CH2Cl2.42 However, the Grignard
reagent derived from 25 reacted with Weinreb amide 22 to give
the desired enone 26 only in 56% yield together with appreciable
amounts of the more stable E-isomer (19%). Although these
compounds could be separated by careful flash chromatography,
product 26 isomerizes fairly quickly, thus rendering any lengthy
purification protocol counterproductive. Gratifyingly, though,
the lithium reagent formed on exposure of 25 to Li sand in Et2O
at low temperature led to a more favorable outcome, furnishing
26 in 83% yield with only minute amounts of the corresponding
E-isomer being detectable.
Enone 26 thus formed was immediately subjected to 1,2-
reduction to avoid any scrambling of the conjugated alkene. The
use of L-selectride, which had been highly appropriate in the case
of the analogous ethyl ketone 23, gave the desired alcohol 28 as a
single isomer but in a disappointing yield of only 31%. It required
an extensive screening of chiral and achiral reducing agents
until a satisfactory solution could be found.34 The best results
were obtained with a combination of LiAlH4, (R)-BINOL
(1 equiv), and EtOH (1 equiv) in THF at ꢀ78 °C.43 Under
these conditions, the desired allylic alcohol 28 was obtained in
90% yield with a diastereomeric ratio of >10:1.44 The outcome
is consistent with a transition state model,43 in which the
a Reagents and conditions: a) EtMgBr, THF, 0 °C, 30 min, 93%; b)
LiBH(sec-Bu)3, THF, ꢀ78 °C, 2 h, 69%; c) NaHCO3, DMF, 70 °C,
130 mbar, 1 h, 87%; d) (i) 25, Li sand, Et2O, 80 min, ꢀ50 °C; (ii) 22,
THF, 0 °C, 1 h, 83%; e) LiAlH4, (R)-BINOL, EtOH, THF, ꢀ78 °C,
72 h, 90% (dr >10:1), see text.
fairly quickly. Therefore, this diene was immediately subjected to
ring-closing olefin metathesis using the ruthenium indenylidene
complex 17 previously described by our group as a cost-effective
alternative to the classical Grubbs catalyst.36ꢀ39 Cyclopentene 18
was thus obtained in a well-reproducible yield of 74% over both
steps. To confirm the integrity of this key intermediate, 18 was
transformed into the crystalline ester derivative 20. Not only did
the structure of 20 in the solid state show the trans disposition of
the lateral chains, but the chloride substituent also facilitated the
determination of the absolute configuration of this product
(Figure 1). As expected, the elaboration of 18 into enyne 22 as
the key building block of the projected syntheses by oxidation
with Dess-Martin periodinane,40 OhiraꢀBestmann reaction,41
and end-capping of the resulting terminal alkyne with a methyl
group was straightforward and high yielding.
The syntheses of the ecklonialactones on the one hand and
hybridalactone on the other hand diverge at this point
(Scheme 5). Entry into the former series was gained upon
reaction of 22 with EtMgBr to give ketone 23, which was reduced
13474
dx.doi.org/10.1021/ja204027a |J. Am. Chem. Soc. 2011, 133, 13471–13480

Journal of the American Chemical Society
ARTICLE
n/π-repulsion between the conjugated double bond of enone 26
and the oxygen lone pair of the axially disposed binaphtholate
oxygen atom in 270 is much more destabilizing than the steric
effects between this oxygen and the axially oriented group R in
the diastereomeric array 27.
Scheme 7a
Auxiliary-Free and Catalysis-Based Approach. Although
the route described above was highly productive and could be
conveniently performed on a large scale and—strictly speaking—
the use of an auxiliary does not fall into the category of
protecting group chemistry, an attempt was made to find an
even more direct entry into the headgroups of the targeted
oxylipins. To this end, the asymmetric 1,4-addition of alkenyl-
boronic acid derivatives to 2[5H]-furanone in the presence of a
chiral rhodium catalyst was considered, even though successful
applications of this methodology to R,β-unsaturated esters or
lactones are still scarce.45,46
To ensure crystallinity of the downstream products as well as
for convenience of preparation and handling, various substituted
alkenylboronates 30 (R ¼ H) were tested and found to add to
butenolide 29 in moderate to good yields in the presence of a
catalyst formed in situ from [Rh(cod)(MeCN)2]BF4 and (S)-
BINAP (34) in aqueous THF at g60 °C (Scheme 6). Even
though the desired 1,4-adducts 31 were obtained with an
encouraging ee of up to 86%, the fairly harsh conditions led to
partial migration of the alkene into conjugation with the lactone
carbonyl. Gratifyingly though, the use of the carvone-derived
diene ligand 35 instead of BINAP allowed the rhodium-catalyzed
conjugate addition to be carried out at ambient temperature
without isomerization interfering.46 As the resulting adduct 31c
derived from styrenylboronate 30c (R = Ph) turned out to be
nicely crystalline, the optical purity of the compound could be
increased from 80% ee to a workable 93% ee by a single
recrystallization.
GC analysis on a chiral stationary phase proved that the use of
either (S)-BINAP (34) or the (ꢀ)-carvone-derived diene 35
furnished the same enantiomer of 31c; its configuration was
assigned according to the transition state models proposed in the
literature47 and confirmed by comparison with literature data.48
Moreover, its identity with the material obtained via the auxiliary-
based route described above, the configuration of which had been
proven by X-ray diffraction, was firmly established after R-
allylation. This step was accomplished by deprotonation of 31c
with LDA and trapping of the resulting enolate with allyl iodide.
The subsequent opening of lactone 32 to the corresponding
Weinreb amide 33 followed by RCM furnished cyclopentene 18,
which was identical to the material described above in all regards.
It is of note that the metathetic ring closures of the styryl
derivative 33 and the vinyl derivative 16 (see Scheme 4) with
the aid of the ruthenium indenylidene complex 1736 are equally
productive. Compound 18 thus obtained intercepts the previous
route and could be elaborated into the key building blocks 24 and
28 as outlined above.
Whereas the auxiliary-based approach is inherently longer, it
enjoys the practical convenience of being readily scalable and
features the beauty of a concise three-component coupling event,
which is not yet possible in the rhodium-catalyzed sequence. The
latter, however, is distinguished by its directness, avoiding any
steps that do not contribute straight to the build-up of the carbon
framework of the target molecule. In any case, the effectiveness of
either route to the common intermediate 18 secured a good
material supply and therefore formed a sound basis for the
completion of the projected total syntheses.
a Reagents and conditions: a) CH2I2 (4 equiv), Et2Zn (2 equiv),
CH2Cl2, ꢀ20 °C, 16 h, 65%; b) tBuOOH, VO(acac)2 (15 mol %),
MS 3 Å, CH2Cl2, 5 h, 69%; c) 39, acid 40, DMAP cat., CH2Cl2, 7 d, 77%;
d) 43b (15 mol %), MS 5 Å, toluene, 70 °C, 24 h, 79%; e) P2ꢀNi (cat.)
[prepared from Ni(OAc)2 4H2O, NaBH4, ethylenediamine], H2
3
(1 atm), EtOH, 3 h, 84%.
Total Synthesis of Hybridalactone. With compound 28 in
hand, the project entered the critical phase of diastereoselective
formation of both sensitive three-membered ring motifs decorat-
ing the backbone of hybridalactone with the aid of the 15-OH
group as a key relay substituent.49 To this end, the Z-configured
allylic alcohol 28 was engaged in a syn-selective cyclopropanation
on reaction with the zinc carbenoid generated in situ from CH2I2
and Et2Zn (Scheme 7). However, the very fragile nature of our
substrate and the presence of a second double bond as well as of
an alkyne, both of which must survive unchanged, precluded
direct application of the conditions described in the literature.50
In particular, the use of a prescribed 1:1 ratio between CH2I2 and
Et2Zn and of a fairly large excess of the active component at 0 °C
led to extensive degradation. After some experimentation, it was
found that the combination of 4 equiv of CH2I2 with only 2 equiv
of Et2Zn in CH2Cl2 gave the desired cyclopropane 36 in well
reproducible 65% yield as a single isomer, provided that the
reaction was carried out at ꢀ20 °C. Likewise, the subsequent
hydroxyl-directed epoxidation of the remaining olefinic site in 36
with a combination of tBuOOH and catalytic VO(acac)2 turned
13475
dx.doi.org/10.1021/ja204027a |J. Am. Chem. Soc. 2011, 133, 13471–13480

Journal of the American Chemical Society
ARTICLE
out to be exquisitely selective,51 delivering the rather labile product
38, in which all seven contiguous chiral centers of hybridalactone
are appropriately set. Assuming the usual trigonal bipyramidal
coordination geometry for the loaded vanadium complex and a
backside displacement of the OꢀO bond51 allows a reasonable
transition state model 37 to be formulated, which nicely accounts
for the observed outcome. The analysis of the derived Mosher
esters confirmed the stereochemistry assigned to this key
intermediate.34,52
Scheme 8a
Not unexpectedly, the esterification of 38 with acid 4053 was
rather challenging. The proclivity of the epoxide ring of such
oxylipins to undergo ring opening had been anticipated from the
host of known natural products formed upon nucleophilic attack
of water or halide onto the oxirane of the parent ecklonialactones
(see Scheme 2). This pronounced bias pertains to 38 and related
building blocks (see below) and precluded the use of all
esterification methods that contain or produce any nucleophilic
species in the reaction mixture (acid chloride, acid fluoride,
mixed anhydrides, N-methyl 2-chloropyridinium chloride, etc).
Moreover, Lewis-acid-catalyzed transesterification methods are
not applicable as they destroy the very sensitive cyclopropylcar-
binol motif.54 Further complications arose from the fact that the
dicyclohexylurea derived from DCC proved to be inseparable by
flash chromatography from the desired ester 41. Gratifyingly,
however, it was found that the modified carbodiimide 39
escorted by an only weakly nucleophilic tosylate anion furnished
product 41 in respectable 77% yield.55
Exposure of this compound to catalytic amounts of the
molybdenum alkylidyne complex 43a as the prototype member
of the latest generation of alkyne metathesis catalysts effected
gentle closure of the macrocyclic scaffold (Scheme 7).56 How-
ever, the resulting product 42 and the triphenylsilanol released
from the catalyst upon workup were difficult to separate by
conventional means. This problem could be fixed by using
complex 43b bearing p-methoxyphenyl groups on the silanolate
ligands, which could be easily removed by standard flash
chromatography; the activity of the catalyst remains largely
unchanged by this peripheral modification.57 Although the strain
of the emerging 13-membered cycle containing a stiff 1,4-enyne
moiety mandated a reaction temperature of 70 °C and applica-
tion of high dilution conditions (0.002 M), the desired product
42 was obtained in well reproducible 79% yield. As previously
described by our group, the silanolate ligand sphere of the
molybdenum alkylidene catalyst is key to success:56 it imposes
a well-balanced level of Lewis acidity onto the Mo(+6) center,
which is high enough to ensure excellent reactivity in alkyne
metathesis yet sufficiently tempered to guarantee outstanding
compatibility with a host of polar substituents; the current
example extends this list to the fragile cyclopropylcarbinol ester,
oxirane, and skipped enyne moieties featured in 42. Moreover,
the silanolates themselves do not engage in nucleophilic opening
of the unusually reactive epoxide rings of the substrate and/or the
resulting product even at elevated temperatures. The rigorous
distinction of the catalyst between the alkynes and the pre-
existing Z-alkene in 42 was also imperative for the present
application. The superior performance of complexes 43 will
become even more obvious by the comparative investigation
conducted in the ecklonialactone series (see below).
a Reagents and conditions: a) VO(acac)2 (8 mol %), tBuOOH, CH2Cl2,
3 h, 94%; b) 9-undecynoic acid, 39, DMAP cat., CH2Cl2, 16 h, 61%; c)
43a (4 mol %), toluene, MS 5 Å, 3 h, 80%, or: 43b (4 mol %), toluene,
MS 5 Å, 3 h, 94%; d) Lindlar catalyst, H2 (1 atm), CH2Cl2, 2.5 h, 90%.
agreement with those of the natural product reported in the
literature.1 Overall, (ꢀ)-1 was obtained in 14 (Rh-catalyzed
route) or 17 steps (auxiliary-based route) over the longest linear
sequences, respectively, without any protecting group chemistry
whatsoever in either route. Except for the final Lindlar-type
reduction of the cyclic alkyne, the sequence also adheres to the
logic of a linearly escalating oxidation state management set forth
in the recent literature.59
As an initial foray into the preparation of analogues, we
targeted desethyl-hybridalactone (44) differing from (ꢀ)-1 only
in the absence of the lateral ethyl substituent on the cyclopropyl unit.
Whereas this formal deletion is thought to be a minor change in
functional regard, it greatly simplifies the synthesis by avoiding the
delicate stereoselective cyclopropanation. The synthesis of 44 is
outlined in detail in the Supporting Information.
Ecklonialactone A, B and C. Along similar lines, the total
syntheses of the C18 oxylipins ecklonialactone A and B could be
readily completed.15,60 Starting from building block 24, a hydro-
xyl-directed vanadium-catalyzed epoxidation gave 45, the con-
figuration of the secondary alcohol of which was confirmed by
analysis of the derived Mosher esters.34 As in the previous cases,
it was necessary to effect the esterification of this alcohol with
9-undecynoic acid under the aegis of carbodimide 39 since all
other methods investigated led to side reactions of the epoxide
ring or even complete destruction of this valuable material
(Scheme 8).61
Final semihydrogenation of cycloalkyne 42 with the aid of
colloidal P2-nickel58 in the presence of ethylenediamine gave
hybridalactone (ꢀ)-1 in 84% yield with <5% overreduction. The
analytical and spectral data of the final product were in excellent
13476
dx.doi.org/10.1021/ja204027a |J. Am. Chem. Soc. 2011, 133, 13471–13480

Journal of the American Chemical Society
ARTICLE
Scheme 9a
a Reagents and conditions: a) undec-6Z-en-9-ynoic acid, 39, DMAP cat.,
CH2Cl2, 16 h, 65%; b) 43b (5 mol %), MS 5 Å, toluene, 7 h, 90%; c)
P2ꢀNi (25 mol %), H2 (1 atm), EtOH, 2 h, 69%.
With good amounts of diyne 46 in hand, a comparative
investigation was carried out, which showcased the superiority
of complexes of type 43 in the subsequent ring-closing alkyne
metathesis reaction. As anticipated, 43a and 43b both furnished
the desired cycloalkyne 47 in yields of 80% and 94%, respec-
tively; once again, complex 43b had the practical advantage of
easier separation of the product from the silanolate ligands
released from the catalyst during work up. In contrast to the
hybridalactone series, however, the ring closure occurred smo-
othly even at ambient temperature in 0.02 M toluene solution.
The lower ring strain of the 14-membered macrolactone 47
as compared to the 13-membered ring of 42 is thought to
account for the milder conditions and the higher applicable
concentration.
Figure 2. Geometry-optimized structure of ecklonialactone A (2),
which is in excellent agreement with the experimentally observed NOE’s
(indicated as red arrows).
Scheme 10. Two Different Acid-Promoted Oxirane Opening
Reactions of Ecklonialactone A and Stereostructure of the
Derived Product 53 (Indicative NOE’s Are Shown As Red
Arrows)
More significant is the observation that the classical tungsten
Schrock alkylidyne complex (tBuO)3WtCCMe3 (48), which
serves as the benchmark in alkyne metathesis chemistry ever
since its discovery,62 afforded none of the desired product due to
instantaneous decomposition of the starting material. We tenta-
tively ascribe this failure to the pronounced Lewis acidity of the
high valent tungsten center in this complex. Likewise, the
performance of [(tBu)ArN]3Mo (49, Ar = 3,5-dimethylphenyl),
after in situ activation with CH2Cl2 as previously described,63 was
unsatisfactory. Unusually high loadings (20ꢀ40 mol %) of the
precatalyst were necessary, and the yields of 47 obtained in
different runs were variable (50ꢀ89%). Although this system had
previously been found compatible with oxiranes,64 GC/MS data
indicated that the amide ligands present in [(tBu)ArN]3Mo
partly reacted with substrate 46, most likely by opening of its
unusually sensitive epoxide ring. Likewise, precatalyst 50 cannot
be applied because the nitrido function is known to react with
oxiranes.65,66 Collectively, these results showcase the superior
application profile of the readily available molybdenum alkyli-
dynes of type 43 endowed with triarylsilanolate ligands, which
are the most effective and, at the same time, most tolerant alkyne
metathesis catalysts currently available.56 Since these species
need not be handled per se but can be conveniently released
in situ from the corresponding phenanthroline adduct precur-
sors, which for their part are fully air stable and can be stored and
handled without any precautions,56 alkylidynes of type 43 are
also practical to use and hence deserve further consideration in
the realm of advanced organic synthesis.67
Diyne 51 containing an extra double bond was prepared
analogously from the same starting material 45 and found to
cyclize equally well on treatment with catalytic amounts of 43b
(Scheme 9). Cycloalkynes 52 and 47 were then transformed into
ecklonialactone A (2) and B (3), respectively, by semireduction.
Lindlar hydrogenation was effective in the latter case, whereas 52
required the use of P2-nickel58 to avoid substantial over-reduction.
13477
dx.doi.org/10.1021/ja204027a |J. Am. Chem. Soc. 2011, 133, 13471–13480

Journal of the American Chemical Society
ARTICLE
The recorded data of our samples matched those of the natural
products.9ꢀ11 Since 2 is known to undergo clean opening of the
epoxide ring on treatment with aqueous HClO4 to give ecklonia-
lactone C (4),10 a formal synthesis of this congener has also been
accomplished.
’ AUTHOR INFORMATION
Corresponding Author
fuerstner@kofo.mpg.de
Follow-Up Chemistry of Ecklonialactone A. Careful NMR
spectroscopic investigations of our samples of ecklonialactone A
(2) revealed strong transannular interactions across the macro-
cycle between the protons at the ring junction, in particular H11,
and the methylene groups at C4 and C5 on the back side of the
ring. These tight contacts are in good agreement with the com-
pact geometry-optimized structure of 2 calculated with the OM3
semiempirical method (Figure 2).68
’ ACKNOWLEDGMENT
Generous financial support by the MPG and the Fonds der
Chemischen Industrie is gratefully acknowledged. We thank
Dr. C. W. Lehmann and J. Rust for solving the crystal structure and
Dr. T. Benighaus for the calculated structure of ecklonialactone A.
’ REFERENCES
When a sample of 2 was kept in CDCl3, which had not been
dried and distilled immediately prior to use, a fairly rapid and
clean epoxide opening was observed, the outcome of which could
be deduced from the characteristic spectral fingerprints of the
resulting products (Scheme 10). Specifically, direct attack of
chloride at the oxirane furnished the corresponding chlorohy-
drine, the data of which matched those of eiseniachloride A (7) in
every regard.11 Moreover, the formation of a second product
derived from a more involved epoxide opening process was
detected, to which we confidently ascribe structure 53; whereas
the constitution of this compound is definite, the stereochemistry
of the chiral center at C9 bearing the chloride substituent could
not be unambiguously determined, even though the observed
NOE’s strongly suggest the depicted configuration. This product
originates from attack of the C9ꢀC10 double bond onto the
adjacent epoxide ring, which is favored by the close spatial
proximity of these groups evident from Figure 2. Except for
the chloride residue, product 53 corresponds to the unnamed
oxylipin 10, which was coextracted with ecklonialactone A (2)
from the brown alga Eisenia bicyclis.11
(1) Higgs, M. D.; Mulheirn, L. J. Tetrahedron 1981, 37, 4259.
(2) Corey, E. J.; De, B.; Ponder, J. W.; Berg, J. M. Tetrahedron Lett.
1984, 25, 1015.
(3) Corey, E. J. Experientia 1983, 39, 1084.
(4) Corey, E. J.; De, B. J. Am. Chem. Soc. 1984, 106, 2735.
(5) The alcohol formed by reduction of the initially formed hydro-
peroxide was later isolated from related red algae, thus providing
additional support for the proposed biosynthesis. See: (a) Bernart, M.;
Gerwick, W. H. Tetrahedron Lett. 1988, 29, 2015. (b) Likewise, the
cymathere ethers likely originate from a closely related lipoxygenase
pathway. See: Proteau, P. J.; Gerwick, W. H. Tetrahedron Lett. 1992,
33, 4393.
(6) Olah, G. A.; Reddy, V. P.; Prakash, G. K. S. Chem. Rev. 1992, 92,
69.
(7) (a) Gerwick, W. H. Chem. Rev. 1993, 93, 1807.(b) Gerwick,
W. H.; Singh, I. P. In Lipid Biotechnology; Kuo, T. M., Gardner, H. W.,
Eds.; Marcel Dekker: New York, 2002; p 249.
(8) White, J. D.; Yang, J. Synlett 2009, 1713.
(9) (a) Kurata, K.; Taniguchi, K.; Shiraishi, K.; Hayama, N.; Tanaka,
I.; Suzuki, M. Chem. Lett. 1989, 18, 267. (b) Kurata, K.; Taniguchi, K.;
Shiraishi, K.; Suzuki, M. Phytochemistry 1993, 33, 155.
(10) Todd, J. S.; Proteau, P. J.; Gerwick, W. H. J. Nat. Prod. 1994, 57,
171.
(11) Kousaka, K.; Ogi, N.; Akazawa, Y.; Fujieda, M.; Yamamoto, Y.;
Takada, Y.; Kimura, J. J. Nat. Prod. 2003, 66, 1318.
(12) Simple reduction products from the proposed hydroperoxide
formed in the initial lipoxygentaion at C13 have been characterized. See:
Proteau, P. J.; Gerwick, W. H. Lipids 1993, 28, 783.
’ CONCLUSIONS
The use of protecting groups implicitly acknowledges an
inability to solve (chemo)selectivity issues in a more direct
manner. Although there is no reason to believe that the need
to do so in the realm of natural product synthesis will cease
anywhere close, an increasing number of case studies in the
recent literature highlight the fact that this exigent goal can
ultimately be reached.19 The work described in this paper adds
some decently complex lipidic targets to this growing list. The
challenges posed by the ecklonialactones and hybridalactone,
which contain up to seven contiguous chiral centers and display
quite reactive functional groups, were met by a sequence of
metal-catalyzed and metal-mediated reactions, which exhibits
respectable levels of atom-, redox-, and step-economy. Concep-
tually speaking, it is the ability to rigorously distinguish between
the π-systems of alkenes and alkynes in reductive, oxidative, and
metathetic transformations which formed the basis of the
successful route and therefore represents an essential lesson to
be learnt from this venture.
(13) Todd, J. S.; Proteau, P. J.; Gerwick, W. H. Tetrahedron Lett.
1993, 34, 7689.
(14) (a) Moreover, the ecklonialactones showed no significant
activity in a cell adhesion assay. See: Takamatsu, S.; Nagle, D. G.;
Gerwick, W. H. Planta Med. 2004, 70, 127. (b) For a discussion of the
antioxidant and antibacterial properties of extracts of edible algae of the
Ecklonia family, see: Kuda, T.; Kunii, T.; Goto, H.; Suzuki, T.; Yano, T.
Food Chem. 2007, 103, 900.
(15) For a preliminary communication of our work on the
ecklonialactones, see: Hickmann, V.; Alcarazo, M.; F€urstner, A. J. Am.
Chem. Soc. 2010, 132, 11042.
(16) For a write-up on the work disclosed in our previous
Communication, see: Roulland, E. Angew. Chem., Int. Ed. 2011, 50, 1226.
(17) (a) Corey, E. J.; Cheng, X.-M. The Logic of Chemical Synthesis;
Wiley: New York, 1989. (b) Nicolaou, K. C.; Vourloumis, D.; Winssinger,
N.; Baran, P. S. Angew. Chem., Int. Ed. 2000, 39, 44.
(18) (a) Burns, N. Z.; Baran, P. S.; Hoffmann, R. W. Angew. Chem.,
Int. Ed. 2009, 48, 2854. (b) Wender, P. A.; Miller, B. L. Nature 2009,
460, 197. (c) For the “economy of steps” as a priority issue, see: F€urstner,
A. Synlett 1999, 1523 and literature cited therein.
’ ASSOCIATED CONTENT
(19) (a) Hoffmann, R. W. Synthesis 2006, 3531. (b) Young, I. S.;
Baran, P. S. Nat. Chem. 2009, 1, 193.
(20) (a) For tropinone as a classical example of a protecting group-
free synthesis of an alkaloid, see: Robinson, R. J. Chem. Soc., Trans. 1917,
111, 762. (b) For leading cases from the recent literature, see: Baran,
S
Supporting Information. Experimental Section, pre-
b
paration of desethyl-hybridalactone, and NMR spectra of new
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
13478
dx.doi.org/10.1021/ja204027a |J. Am. Chem. Soc. 2011, 133, 13471–13480

Journal of the American Chemical Society
ARTICLE
P. S.; Maimone, T. J.; Richter, J. M. Nature 2007, 446, 404. (c) Gu, Z.;
Zakarian, A. Angew. Chem., Int. Ed. 2010, 49, 9702. (d) Gerfaud, T.; Xie,
C.; Neuville, L.; Zhu, J. Angew. Chem., Int. Ed. 2011, 50, 3954 and
literature cited therein.
O. R. J. Org. Chem. 2000, 65, 1738. (c) F€urstner, A.; Jeanjean, F.; Razon,
P. Angew. Chem., Int. Ed. 2002, 41, 2097. (d) F€urstner, A.; Leitner, A.
Angew. Chem., Int. Ed. 2003, 42, 308. (e) Scheiper, B.; Glorius, F.;
Leitner, A.; F€urstner, A. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 11960.
(f) F€urstner, A.; Jeanjean, F.; Razon, P.; Wirtz, C.; Mynott, R. Chem.—
Eur. J. 2003, 9, 320. (g) F€urstner, A.; Radkowski, K.; Wirtz, C.; Goddard,
R.; Lehmann, C. W.; Mynott, R. J. Am. Chem. Soc. 2002, 124, 7061.
(h) F€urstner, A.; Nagano, T.; M€uller, C.; Seidel, G.; M€uller, O. Chem.—
Eur. J. 2007, 13, 1452. (i) F€urstner, A.; Korte, A. Chem. Asian J. 2008,
3, 310.
(39) For a related transformation used en route to agardhilactone, see:
Miyaoka, H.; Hara, Y.; Shinohara, I.; Kurokawa, T.; Yamada, Y. Tetrahedron
Lett. 2005, 46, 7945.
(40) Meyer, S. D.; Schreiber, S. L. J. Org. Chem. 1994, 59, 7549.
(41) (a) Ohira, S. Synth. Commun. 1989, 19, 561. (b) M€uller, S.;
Liepold, B.; Roth, G. J.; Bestmann, H. J. Synlett 1996, 521.
(42) (a) Mori, K.; Brevet, J.-L. Synthesis 1991, 1125. (b) Bayer, A.;
Maier, M. E. Tetrahedron 2004, 60, 6665.
(21) For selected examples of protecting group-free syntheses of
nonalkaloid targets, see: (a) Danishefsky, S.; Dumas, D. Chem. Commun.
1968, 1287. (b) Chapman, O. L.; Engel, M. R.; Springer, J. P.; Clardy,
J. C. J. Am. Chem. Soc. 1971, 93, 6696. (c) Wender, P. A.; Dreyer, G. B.
Tetrahedron 1981, 37, 4445. (d) Pfeiffer, M. W. B.; Phillips, A. J. Am.
Chem. Soc. 2005, 127, 5334. (e) McFadden, R. M.; Stoltz, B. M. J. Am.
Chem. Soc. 2006, 128, 7738. (f) Trost, B. M.; Weiss, A. H. Org. Lett. 2006,
8, 4461. (g) Enquist, J. A., Jr.; Stoltz, B. M. Nature 2008, 453, 1228.
(h) Sofiyev, V.; Navarro, G.; Trauner, D. Org. Lett. 2008, 10, 149.
(i) Tiefenbacher, K.; Mulzer, J. Angew. Chem., Int. Ed. 2008, 47, 6199.
(j) Hayashida, J.; Rawal, V. H. Angew. Chem., Int. Ed. 2008, 47, 4373.
(k) Pardeshi, S. G.; Ward, D. E. J. Org. Chem. 2008, 73, 1071. (l) Sirasani,
G.; Paul, T.; Andrade, R. B. J. Org. Chem. 2008, 73, 6386. (m) Gaich, T.;
Mulzer, J. J. Am. Chem. Soc. 2009, 131, 452. (n) Zhou, Q.; Chen, X.; Ma,
D. Angew. Chem., Int. Ed. 2010, 49, 3513.
(43) Noyori, R.; Tomino, I.; Tanimoto, Y.; Nishizawa, M. J. Am.
Chem. Soc. 1984, 106, 6709.
(44) It is interesting to note that the reduction is substrate controlled
since the use of (S)-BINOL did not override its inherent bias; for further
details, see the Supporting Information.
(22) (a) F€urstner, A.; Grela, K.; Mathes, C.; Lehmann, C. W. J. Am.
Chem. Soc. 2000, 122, 11799. (b) F€urstner, A.; Grela, K. Angew. Chem.,
Int. Ed. 2000, 39, 1234. (c) F€urstner, A.; Mathes, C. Org. Lett. 2001,
3, 221.
(23) F€urstner, A.; Schlede, M. Adv. Synth. Catal. 2002, 344, 657.
(24) (a) Trnka, T. M.; Grubbs, R. H. Acc. Chem. Res. 2001, 34, 18.
(b) F€urstner, A. Angew. Chem., Int. Ed. 2000, 39, 3012. (c) Connon, S. J.;
Blechert, S. Angew. Chem., Int. Ed. 2003, 42, 1900. (d) Nicolaou, K. C.;
Bulger, P. G.; Sarlah, D. Angew. Chem., Int. Ed. 2005, 44, 4490.
(e) Deiters, A.; Martin, S. F. Chem. Rev. 2004, 104, 2199. (f) Hoveyda,
A. H.; Zhugralin, A. R. Nature 2007, 450, 243.
(25) (a) F€urstner, A.; Seidel, G. Angew. Chem., Int. Ed. 1998,
37, 1734. (b) F€urstner, A.; Guth, O.; Rumbo, A.; Seidel, G. J. Am. Chem.
Soc. 1999, 121, 11108.
(45) Reviews:(a) Hayashi, T.; Yamasaki, K. Chem. Rev. 2003,
103, 2829. (b) Shintani, R.; Hayashi, T. Aldrichimica Acta 2009, 42, 31.
(46) (a) The closest precedent is the rhodium-catalyzed addition of
PhB(OH)2 to 2[5H]-furanone. See: Defieber, C.; Paquin, J.-F.; Serna,
S.; Carreira, E. M. Org. Lett. 2004, 6, 3873. (b) See also: Paquin, J.-F.;
Stephenson, C. R. J.; Defieber, C.; Carreira, E. M. Org. Lett. 2005,
7, 3821. (c) Defieber, C.; Gr€utzmacher, H.; Carreira, E. M. Angew.
Chem., Int. Ed. 2008, 47, 4482.
(47) (a) Takaya, Y.; Ogasawara, M.; Hayashi, T.; Sakai, M.; Miyaura,
N. J. Am. Chem. Soc. 1998, 120, 5579. (b) Hayashi, T.; Ueyama, K.;
Tokunaga, N.; Yoshida, K. J. Am. Chem. Soc. 2003, 125, 11508.
(48) The enantiomer of 31c was previously prepared by an inde-
pendent route. See: Kim, S.-G. Tetrahedron Lett. 2008, 49, 6148.
(49) For a review on directed functionalizations, see: Hoveyda,
A. H.; Evans, D. A.; Fu, G. C. Chem. Rev. 1993, 93, 1307.
(50) Charette, A. B.; Lebel, H. J. Org. Chem. 1995, 60, 2966.
(51) (a) Sharpless, K. B.; Michaelson, R. C. J. Am. Chem. Soc. 1973,
95, 6136. (b) Sharpless, K. B.; Verhoeven, T. R. Aldrichimica Acta 1979,
12, 63.
(52) (a) Seco, J. M.; Qui~noꢀa, E.; Riguera, R. Chem. Rev. 2004, 104, 17.
(b) Hoye, T. R.; Jeffrey, C. S.; Shao, F. Nat. Protoc. 2007, 2, 2451.
(53) Two different routes to acid 40 have been worked out. The first
one starts from cheap δ-valerolactone, which is opened with acidic
MeOH to methyl 5-hydroxypentanoate, followed by oxidation and
Wittig reaction to deliver the Z-alkene selectively; finally, the methyl
ester is saponified to give the required acid. Strictly speaking, however,
the use of the methyl ester may be regarded as a protecting group
manipulation. Therefore an alternative sequence starting from 3,4-
dihydro-2H-pyran was devised, which is truly protecting group-free;
for the details, see the Supporting Information.
(54) Lin, M.-H.; RajanBabu, T. V. Org. Lett. 2000, 2, 997.
(55) Szeto, P. 1-Cyclohexyl-3-(2-morpholinoethyl)carbodiimide
Metho-p-toluenesulfonate. In e-EROS Encyclopedia of Reagents for
Organic Synthesis; Wiley: New York, 2010, DOI: 10.1002/047084289X.
(56) Heppekausen, J.; Stade, R.; Goddard, R.; F€urstner, A. J. Am.
Chem. Soc. 2010, 132, 11045.
(26) (a) F€urstner, A.; Davies, P. W. Chem. Commun. 2005, 2307.
(b) Zhang, W.; Moore, J. S. Adv. Synth. Catal. 2007, 349, 93. (c) Schrock,
R. R.; Czekelius, C. Adv. Synth. Catal. 2007, 349, 55. (d) Mortreux, A.;
Coutelier, O. J. Mol. Catal. A: Chem. 2006, 254, 96.
(27) For instructive examples showing that alkyne metathesis is
orthogonal to conventional alkene metathesis, see: (a) F€urstner, A.;
Dierkes, T. Org. Lett. 2000, 2, 2463. (b) F€urstner, A.; Mathes, C.; Grela,
K. Chem. Commun. 2001, 1057. (c) F€urstner, A.; Turet, L. Angew. Chem.,
Int. Ed. 2005, 44, 3462. (d) F€urstner, A.; Bindl, M.; Jean, L. Angew.
Chem., Int. Ed. 2007, 46, 9275. (e) F€urstner, A.; De Souza, D.; Turet, L.;
Fenster, M. D. B.; Parra-Rapado, L.; Wirtz, C.; Mynott, R.; Lehmann,
C. W. Chem.—Eur. J. 2007, 13, 115. (f) F€urstner, A.; Kirk, D.; Fenster,
M. D. B.; Aïssa, C.; De Souza, D.; Nevado, C.; Tuttle, T.; Thiel, W.;
M€uller, O. Chem.—Eur. J. 2007, 13, 135.
(28) (a) Noyori, R.; Suzuki, M. Angew. Chem., Int. Ed. Engl. 1984,
23, 847. (b) Guo, H.-C.; Ma, J.-A. Angew. Chem., Int. Ed. 2006, 45, 354.
(29) For examples from our group, see: (a) F€urstner, A.; Langemann,
K. J. Org. Chem. 1996, 61, 8746. (b) F€urstner, A.; M€uller, T. Synlett
1997, 1010.
(30) Feringa, B. L.; de Jong, J. C. J. Org. Chem. 1988, 53, 1125.
(31) (a) Robertson, J.; Mꢀenard, M.; Ford, R.; Bell, S. Org. Biomol.
Chem. 2004, 2, 2988. (b) Mulzer, J.; Riether, D. Tetrahedron Lett. 1999,
40, 6197.
(32) Feringa, B. L.; de Lange, B.; de Jong, J. C. J. Org. Chem. 1989,
54, 2471.
(33) Moradai, O. M.; Paquette, L. A. Org. Synth. 2003, 80, 66.
(34) For details, see the Supporting Information.
(35) Nahm, S.; Weinreb, S. M. Tetrahedron Lett. 1981, 22, 3815.
(36) F€urstner, A.; Guth, O.; D€uffels, A.; Seidel, G.; Liebl, M.; Gabor,
B.; Mynott, R. Chem.—Eur. J. 2001, 7, 4811.
(37) Ruthenium-indenylidene-based metathesis catalysts are sold
under the tradenames “Neolyst” (Umicore, Hanau) and “catMETium”
(Degussa).
(57) A more detailed investigation on structural variants of the
parent complex 43a will be disclosed shortly.
(58) Brown, C. A.; Ahuja, V. K. J. Org. Chem. 1973, 38, 2226.
(59) Burns, N. Z.; Baran, P. S.; Hoffmann, R. W. Angew. Chem., Int.
Ed. 2009, 48, 2854.
(60) For a study towards ecklonialactone B, see: Wang, Q.; Millet,
A.; Hiersemann, M. Synlett 2007, 1683.
(38) Applications in total synthesis: (a) F€urstner, A.; Grabowski, J.;
Lehmann, C. W. J. Org. Chem. 1999, 64, 8275. (b) F€urstner, A.; Thiel,
(61) To avoid such problems, it was also tried to change the
sequence by running the esterification and RCAM prior to epoxidation
13479
dx.doi.org/10.1021/ja204027a |J. Am. Chem. Soc. 2011, 133, 13471–13480

Journal of the American Chemical Society
ARTICLE
of the double bond. In this case, however, the epoxidation occurred
preferentially from the wrong π-face, delivering 12,13-epi-ecklonialac-
tone B as the major product; for details, consult ref 15.
(62) (a) Wengrovius, J. H.; Sancho, J.; Schrock, R. R. J. Am. Chem.
Soc. 1981, 103, 3932. (b) Schrock, R. R. Angew. Chem., Int. Ed. 2006,
45, 3748.
(63) (a) F€urstner, A.; Mathes, C.; Lehmann, C. W. J. Am. Chem. Soc.
1999, 121, 9453. (b) F€urstner, A.; Mathes, C.; Lehmann, C. W. Chem.—
Eur. J. 2001, 7, 5299.
(64) (a) F€urstner, A.; Larionov, O.; Fl€ugge, S. Angew. Chem., Int. Ed.
2007, 46, 5545. (b) F€urstner, A.; Fl€ugge, S.; Larionov, O.; Takahashi, Y.;
Kubota, T.; Kobayashi, J. Chem.—Eur. J. 2009, 15, 4011.
(65) Bindl, M.; Stade, R.; Heilmann, E. K.; Picot, A.; Goddard, R.;
F€urstner, A. J. Am. Chem. Soc. 2009, 131, 9468.
(66) For recent applications of this complex in total synthesis, see:
(a) Smith, B. J.; Sulikowski, G. A. Angew. Chem., Int. Ed. 2010, 49, 1599.
(b) Nilson, M. G.; Funk, R. L. Org. Lett. 2010, 12, 4912.
(67) For an exigent case, see: Micoine, K.; F€urstner, A. J. Am. Chem.
Soc. 2010, 132, 14064.
(68) (a) Silva-Junior, M. R.; Thiel, W. J. Chem. Theory Comput. 2010,
6, 1546. (b) Scholten, M. Dissertation, Universit€at D€usseldorf, 2003.
13480
dx.doi.org/10.1021/ja204027a |J. Am. Chem. Soc. 2011, 133, 13471–13480