Formal total synthesis of the algal toxin (-)-polycavernoside A

Article abstract of DOI:10.1002/chem.201204551

A concise and largely catalysis-based approach to the potent algal toxin polycavernoside A (1) is described that intercepts a late-stage intermediate of a previous total synthesis; from there on, this challenging target can be reached in a small number of steps. Key to success was a sequence of a molybdenum-catalyzed ring-closing alkyne metathesis (RCAM) reaction to forge the macrocyclic frame, followed by a gold-catalyzed and strictly regioselective transannular hydroalkoxylation of the resulting cycloalkyne that allows the intricate oxygenation pattern of the macrolactone ring of 1 to be properly set. The required cyclization precursor 5 was assembled by the arguably most advanced fragment coupling process based on an Evans-Tishchenko redox esterification known to date, which was optimized to the extent that the precious coupling partners could be used in an almost equimolar ratio (6/7 1:1.3). The preparation of these building blocks features, inter alia, the power of the Sc(OTf) ₃-catalyzed Leighton crotylation as well as the superb selectivities of alkene cross metathesis, asymmetric keto-ester hydrogenation, and the Jacobsen epoxidation/epoxide resolution technologies. Copyright

Full text of DOI:10.1002/chem.201204551

DOI: 10.1002/chem.201204551
Formal Total Synthesis of the Algal Toxin (ꢀ)-Polycavernoside A
Lennart Brewitz, Josep Llaveria, Akira Yada, and Alois Fꢀrstner*[a]
Abstract: A concise and largely cataly-
of the resulting cycloalkyne that allows
the intricate oxygenation pattern of the
macrolactone ring of 1 to be properly
set. The required cyclization precursor
5 was assembled by the arguably most
advanced fragment coupling process
based on an Evans–Tishchenko redox
esterification known to date, which was
optimized to the extent that the pre-
sis-based approach to the potent algal
toxin polycavernoside A (1) is descri-
bed that intercepts a late-stage inter-
mediate of a previous total synthesis;
from there on, this challenging target
can be reached in a small number of
steps. Key to success was a sequence of
cious coupling partners could be used
in an almost equimolar ratio (6/7
1:1.3). The preparation of these build-
ing blocks features, inter alia, the
power of the ScACHTNUTRGENNG(U OTf)3-catalyzed
Leighton crotylation as well as the
superb selectivities of alkene cross
metathesis, asymmetric keto-ester hy-
drogenation, and the Jacobsen epoxida-
tion/epoxide resolution technologies.
a
molybdenum-catalyzed ring-closing
alkyne metathesis (RCAM) reaction to
forge the macrocyclic frame, followed
by a gold-catalyzed and strictly regiose-
lective transannular hydroalkoxylation
Keywords: alkynes · gold · meta-
thesis · molybdenum · natural prod-
ucts · samarium
Introduction
five-membered hemiketal contained within the macrocyclic
frame of 1 while elaborating the lateral chain of the target
(Scheme 1). Compound 3 derives from cycloalkyne 4 by a
transannular hydroalkoxylation with the aid of a carbophilic
catalyst.[14] We surmised that product 4, in turn, provides an
excellent opportunity to scrutinize ring-closing alkyne meta-
thesis (RCAM)[15—17] in a challenging synthetic context.
Cases of deadly foodborne intoxication reported in Guam in
the Spring of 1991 were traced back to the red alga Graci-
laria edulis (Polycavernosa tsudai), even though this widely
consumed seaweed had never before caused any harm.[1]
This sudden outburst seemed to be a tragic episode until
similarly fatal incidents occurred in the Philippines about a
decade later.[2,3] A masterful investigation allowed Yasumo-
to and co-workers to identify the glycosylated macrolides of
the polycavernoside series as the causative toxins contained
in the poisonous algal crops.[4–6] For their potent bioactivity
and intricate structures, these targets spurred considerable
synthetic efforts worldwide. The interest does not seem to
cease even many years after the first conquest of polycaver-
noside A (1), by which the absolute configuration of this
parent family member had been formally established.[7–13}
We now present a concise alternative entry and formal
total synthesis of this challenging marine natural product,
which intercepts a key intermediate of the successful ap-
proach reported by Woo and Lee.[10] Specifically, compound
3 had been converted by these authors into 1 by a short and
productive sequence based on the oxidation of the endocy-
clic enol ether to the masked a-dicarbonyl derivative 2; this
particular product cleanly rearranged to the conspicuous
[a] M. Sc. L. Brewitz, Dr. J. Llaveria, Dr. A. Yada, Prof. A. Fꢀrstner
Max-Planck-Institut fꢀr Kohlenforschung
45470 Mꢀlheim/Ruhr (Germany)
Fax : (+49)208306-2994
E-mail: fuerstner@kofo.mpg.de
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201204551.
Scheme 1. Retrosynthetic analysis of polycavernoside A (1).
4532
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Chem. Eur. J. 2013, 19, 4532 – 4537

FULL PAPER
Moreover, the 1,3-anti-diol monoester motif present in the
prospective cyclization precursor 5 invites the use of an
Evans–Tishchenko coupling reaction.[18,19] It is of note, how-
ever, that this redox-esterification typically mandates the
use of a (large) excess of the aldehyde component relative
to the hydroxy-ketone partner. This unfavorable stoichiom-
etry may well explain why Evans–Tishchenko reactions—
despite an excellent track record—have hardly ever been
applied to advanced fragment coupling processes.[20] In any
case, the use of this transformation in the projected synthe-
sis of 1 is only warranted if the rather advanced building
blocks 6 and 7 can be used in roughly equimolar ratio; this
challenge has yet to be met. Furthermore, we wished to
strictly base the material supply on asymmetric synthesis
and catalysis but avoid any recourse to the “chiral pool”,
which had served as the dominating source of chirality in
the reported total syntheses of 1 (pantolactone, citronellal,
malic acid).[7—11]
Results and Discussion
The preparation of the tetrahydropyran segment 7 com-
menced with an asymmetric hydrogenation of the commer-
cial b-ketoester 8 using P-Phos (18) as the chiral ligand,
which delivered multigram quantities of the corresponding
chlorohydrin in excellent optical purity (94%, 96% ee)
(Scheme 2).[21] A copper-catalyzed PMB protection allowed
any premature epoxide ring formation to be avoided and
set the basis for a DIBAl-H reduction of the ester group in
9 to the corresponding aldehyde 10. Exposure of this com-
pound to the Z-crotylsilane reagent 19 in the presence of
Scheme 2. Preparation of the aldehyde building block: a) [RuCl2((R)-P-
Phos)]
(96% ee); b) PMBOC(NH)CCl3, Cu
08C, 82%; c) DIBAl-H, toluene, ꢀ788C, 88%; d) 19, Sc
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
(4.5 mol%), CH2Cl2, 08C, 78% (d.r. ꢁ28:1); e) TBSCl, imidazole,
DMAP cat., CH2Cl2, 85%; f) DDQ, CH2Cl2, pH 7 buffer, 99%; g) KOH,
EtOH, 08C ! RT, 94%; h) methyl acrylate, 20 (3.5 mol%), toluene,
708C, 75%; i) nBuLi, propyne, BF3·Et2O, THF, ꢀ788C, 89%; j) KOtBu
(10 mol%), THF, ꢀ108C, 58% (15) + 9% (17); k) DBU, LiCl, MeCN,
1008C (sealed tube), 84% (17/15 ꢁ20:1); l) DIBAl-H, toluene, ꢀ788C,
catalytic amounts of ScACTHNUTRGNE(UGN OTf)3 furnished the desired homo-
allyl alcohol 11 in good yield and excellent syn-selectivity
(d.r. ꢁ28:1),[22] which clearly surpassed the results obtained
by a classical Brown crotylation (d.r. 9:1);[23] the ready sep-
aration and recovery of the chiral diamine ligand to silicon
is an additional bonus of this method. Protection of the
newly formed secondary alcohol as a TBS-ether followed
by cleavage of the PMB-group and treatment of the result-
ing material with KOH in EtOH gave epoxide 12 without
incident. This compound underwent a clean cross metathe-
sis with methyl acrylate to give product 13, without the reac-
tive oxirane interfering, which was best effected by rutheni-
um carbene complex 20 (Zhan catalyst 1B).[24–27]
Originally, it was hoped that the opening of the oxirane in
13 with 1-propynyllithium might result in a spontaneous clo-
sure of the pyran ring by an oxy-Michael reaction of the in-
termediate lithium alkoxide; in practice, however, this over-
all transformation would proceed only when carried out in
two separate operations. Best results were obtained when
the epoxide opening was carried out in the presence of
BF3·Et2O. As expected, the subsequent 1,4-addition of the
resulting alcohol to the E-configured enoate function in 14
furnished the undesired 2,6-trans-disubstituted pyran 15 as
the major product,[28] when performed with catalytic
amounts of KOtBu in THF.[29] Gratifyingly though, the use
86%; PMB = p-methoxybenzyl; DBU = 1,8-diazabicycloACHTUNTRGNEU[GN 5.4.0]undec-7-
ene; DIBAl-H = diisobutylaluminum hydride; DDQ = 2,3-dichloro-5,6-
dicyano-p-benzoquinone; TBS = tert-butyldimethylsilyl; OTf = trifluoro-
methanesulfonate.
of DBU/LiCl in MeCN at elevated temperature (1008C,
sealed tube) allowed the outcome to be rectified.[30] Under
these conditions, compound 17 was obtained in high yield,
virtually as a single isomer (cis/trans ꢁ20:1). We assume
that the equilibration of the minor isomer 15 in combination
with a favorable chelate transition state 16 for the forward
process accounts for this favorable outcome. DIBAl-H re-
duction of 17 then completed the synthesis of the required
aldehyde building block 7 in readiness for the envisaged
Evans–Tishchenko reaction.
Access to the Northern hemisphere was secured by an
aldol addition of ethyl isobutyrate 21 to acrolein[31] followed
by a lipase-catalyzed kinetic resolution of the resulting prod-
uct 22 (Scheme 3).[32] Acetate (+)-23 thus available in large
quantities and optically pure form (ꢁ98% ee)[33] was readily
converted into acid 24 by routine protecting-group manipu-
Chem. Eur. J. 2013, 19, 4532 – 4537
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4533

A. Fꢀrstner et al.
to-handle tosylate 30. Gratifyingly, product 31 formed by
this route had an impeccable ee of ꢁ 99%. Exposure to acti-
vated zinc dust in dimethylacetamide at 758C followed by
acylation of the resulting organozinc reagent by freshly pre-
pared acid chloride 25 in the presence of CuCN·2LiCl[37] at
ꢀ308C furnished the sterically hindered hydroxy ketone 6 in
appreciable overall yield after cleavage of the silyl protect-
ing group with trifluoroacetic acid (Scheme 5).
Scheme 3. a) LDA, acrolein, THF, ꢀ788C, 94%; b) Novozyme 435, vinyl
acetate, toluene, 708C, 33% (ꢁ98% ee); c) NaOH, MeOH, 08C ! RT;
d) TBSCl, imidazole, DMF; e) K2CO3, THF/H2O/MeOH 2:1:1, 67%
(over three steps); f) (COCl)2, CH2Cl2, quant.
lations. It was planned to couple an activated derivative
thereof (acid chloride, Weinreb amide etc.) with a suitable
homo-propargyl donor derived from bromide 31. The devel-
opment of a scalable route to the latter building block,
though small in size, was actually non-trivial and required
careful optimization.
A satisfactory solution was found by Sonogashira cou-
pling[34] of propyne with vinyl bromide (26) to give enyne 27
in good yield (Scheme 4). The corresponding epoxide 28
was obtained in almost perfect optical purity (ee ꢁ99%) by
a sequence of Mn–salen catalyzed epoxidation followed by a
Scheme 5. Fragment coupling by an Evans–Tishchenko reaction: a)
CuCN·2LiCl, THF, ꢀ308C; b) TFA, CH2Cl2, 67% (over both steps); c)
7, SmI2 (50 mol%), THF, ꢀ508C, 68 % (5a), see text; d) dichloroacetic
acid anhydride, pyridine, CH2Cl2, 08C, 94%; TFA = trifluoroacetic acid.
With aldol 6 and aldehyde 7 in hand, the stage was set for
the key fragment coupling via an Evans–Tishchenko reac-
tion (Scheme 5). Although the formation of the desired 1,3-
anti-diol monoester 5a could be observed under the stand-
ard reaction conditions (35 mol% of SmI2 in THF at
ꢀ108C),[18] a by-product was formed in almost equal quanti-
ty to which structure 36 was assigned based on spectroscopic
evidence. Its formation is best explained by a retro-aldol/
aldol sequence via a transient samarium enolate of type 35
as shown in Scheme 5.[38] Although this parasitic process
could be largely suppressed by lowering the temperature to
ꢀ508C, an increase of the catalyst loading was then manda-
tory to ensure complete conversion. The use of 50 mol% of
SmI2 was the best compromise, while a larger amount nega-
tively impacted on the mass balance. Under these condi-
tions, the two reaction partners 6 and 7 could be used in a
1:1.3 ratio to give the anti-configured product 5a with a dia-
stereoselectivity of ꢁ10:1 (HPLC); no more than 2% of the
byproduct 36 were detected in the crude material. Even
though the minor syn-isomer could only be separated by
preparative HPLC, the desired compound 5a was still ob-
tained in respectable 68% yield.[39] We believe that this fa-
Scheme 4. a) Propyne, [(PPh3)2PdCl2] (1.5 mol%), CuI (3 mol%), Et3N,
ꢀ788C ! RT (sealed tube), 81%; b) 33 (5 mol%), NaOCl, CH2Cl2, 08C;
c) 34 (0.5 mol%), H2O, Et2O, 73% (over both steps) (ꢁ99% ee); d)
MeLi, BF3·Et2O, ꢀ788C; e) TsCl, Et3N, DMAP cat., CH2Cl2, 70% (over
both steps); f) LiBr, DMF, 608C, 86% (ꢁ99% ee); g) Zn, dimethyl-
ACHTUNGTRENNUNGacetamide, 758C; LDA = lithium diisopropylamide; Ts = tosyl; DMAP
= 4-dimethylaminopyridine.
Co–salen catalyzed hydrolytic kinetic resolution, as previ-
ously described for the other enantiomer.[35] The subsequent
ring opening with a methyl donor turned out to be surpris-
ingly capricious: after some experimentation, it was found
that the use of MeLi in combination with BF3·Et2O ensured
a reliable and clean SN2 pathway.[36] In contrast, MeMgX (X
= Cl, Br) or Me2Mg, under a variety of experimental condi-
tions, led to a noticeable and—in part—even very significant
erosion of the optical purity of the resulting a-branched al-
cohol 29 (36–97% ee) and/or to the formation of product
mixtures. The conversion of this volatile material into the
equally volatile bromide 31 was best achieved via the easy-
4534
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 4532 – 4537

Algal Toxin (ꢀ)-Polycavernoside A
FULL PAPER
vorable outcome should encourage a wider use of the
Evans–Tishchenko redox-esterification reaction as a means
for the coupling of elaborate fragments in advanced organic
synthesis.
Gratifyingly though, both secondary processes could be
completely suppressed and the desired enol ether formation
accomplished in good yield with the aid of the cationic gold
complex 40 bearing a bulky phosphine ligand.[46] Although
we have no reason to believe that compound 3b could not
be elaborated into polycavernoside A (1) according to the
published literature route,[10] it should be noticed that this
particular product differs from the key intermediate used by
Woo and Lee in that it carries a TBS-ether rather than an
acetyl group on the pyran moiety. To exclude any doubts
and ensure full identity, we bothered to swap this protecting
group after the macrocyclization (4b ! 4a). As expected,
the transannular functionalization by the alkynophilic gold
catalyst was not affected by this peripheral modification; ac-
tually, the isolated yield of the resulting enol ether 3a was
even better. In any case, compound 3a is identical in all re-
gards with the key intermediate that had previously been
elaborated into polycavernoside A (1) in eight straightfor-
ward operations with an average yield of 74% per step.[10]
Compound 5a undergoes a surprisingly facile intramolec-
ular transesterification, leading to an equilibrium mixture
with the corresponding 1,3-transposed ester derivative.
Therefore the secondary hydroxyl group of 5a was tempora-
rily protected as an orthogonal dichloroacetate and the re-
sulting diyne 5b subjected to macrocyclization by ring-clos-
ing alkyne metathesis (RCAM). In line with our expecta-
tions, the reaction proceeded smoothly when catalyzed by
the molybdenum alkylidyne ate-complex 38 endowed with
triarylsilanolate ligands (Scheme 6).[40] This example is an-
other important entry in the rapidly growing list of success
stories of this new type of catalysts,[41] which are well acces-
sible, highly active and remarkably selective at the same
time. It is emphasized that the neutral alkylidyne species
ꢂ
[(Ar3SiO)3Mo CC6H4OMe] (Ar = p-MeOC6H4-), which is
formed in situ as the active principle from the ate-complex
38 by loss of Ar3SiOK, can also
be generated from the indefi-
nitely bench-stable phenanthro-
line adduct 39 on treatment
with either MnCl2 or ZnCl2, as
previously reported by our
group; this procedure is partic-
ularly user friendly.[40]
Next, cycloalkyne 4b thus
formed was subjected to
a
transannular hydroalkoxylation
with formation of enol ether
3b. Although Woo and Lee had
previously effected a closely re-
lated transformation with the
aid of PtCl2/CO,[10,42] we ob-
served
when 4b was exposed to cata-
lytic amounts of [PtCl2A(C2H4)]2
as more soluble platinum
a
secondary process
CTHUNGTRENNUNG
a
source. Although the alkyne
was selectively attacked at the
C9 position by the secondary
OH group, the intermediate
enol ether was hydrolyzed to
the
corresponding
hemi-
ketal[43,44] and the allylic ester
concomitantly rearranged. It is
the relieve of strain caused by
the congested neopentylic envi-
ꢀ
ronment about the C O bond
in 4b that provides the driving
force for the formation of the
primary allylic ester 37 by this
platinum-catalyzed Overman-
type [3,3]-sigmatropic rear-
rangement.[45]
Scheme 6. Completion of the (formal) total synthesis of polycavernoside A: a) 38 (5 mol%), toluene, MS 5 ꢂ,
808C, 91%; b) K2CO3, MeOH, 408C, 80%; c) [PtCl2ACHTUNTRGNEUNG(C2H4)]2 (10 mol%), Et2O, 77%; d) 40 (10 mol%), MS
4 ꢂ, CH2Cl2, 67% (3b), 84% (3a); e) TFA, CH2Cl2, 408C, 76%; f) Ac2O, pyridine, CH2Cl2, 408C, 97%; for
the completion of the total synthesis from 3a onward, see ref. [10]; DMDO = dimethyldioxirane; PPTS = pyr-
idinium p-toluenesulfonate; TPAP = tetra-n-propylammonium perruthenate; NMO = N-methylmorpholine-
N-oxide.
Chem. Eur. J. 2013, 19, 4532 – 4537
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4535

A. Fꢀrstner et al.
[9] a) J. D. White, P. R. Blakemore, C. C. Browder, J. Hong, C. M. Lin-
coln, P. A. Nagornyy, L. A. Robarge, D. J. Wardrop, J. Am. Chem.
Soc. 2001, 123, 8593–8595; b) P. R. Blakemore, C. C. Browder, J.
Hong, C. M. Lincoln, P. A. Nagornyy, L. A. Robarge, D. J. Wardrop,
J. D. White, J. Org. Chem. 2005, 70, 5449–5460.
Conclusion
A formal total synthesis of the challenging algal toxin poly-
cavernoside A (1) is presented, which is concise, productive
and largely catalysis-based. Our entry features the arguably
most advanced case of an Evans–Tishchenko reaction for
the coupling of two elaborate building blocks and showcases
the maturity of the latest generation of catalysts for alkyne
metathesis. This result highlights the notion that RCAM
provides ample opportunity for advanced organic synthesis,
in particular when combined with the power of noble metal
catalysis for the selective activation of the resulting acety-
lenic products. Although the use of carbophilic Lewis acids
in general has seen an explosive growth during the last
decade, late-stage applications to the targeted pursuit of
complex natural products are still scarce.[47] It is hoped that
the polycavernoside case study will encourage further inves-
tigations along these lines, although it also illustrates the
subtle differences in the behavior of platinum- and gold cata-
lysts. Ongoing projects in this laboratory intend to contrib-
ute to a better understanding of the underpinning reactivity
patterns and try to find new applications of carbophilic acti-
vation in concert with alkyne metathesis.
[10] S. K. Woo, E. Lee, J. Am. Chem. Soc. 2010, 132, 4564–4565.
[11] Y. Kasai, T. Ito, M. Sasaki, Org. Lett. 2012, 14, 3186–3189.
[12] For studies towards the polycavernosides, see: a) N. Hayashi, T.
Mine, K. Fujiwara, A. Murai, Chem. Lett. 1994, 2143–2146; b) K.
Fujiwara, S. Amano, T. Oka, A. Murai, Chem. Lett. 1994, 2147–
2150; c) K. Fujiwara, S. Amano, A. Murai, Chem. Lett. 1995, 191–
192; d) K. Fujiwara, S. Amano, A. Murai, Chem. Lett. 1995, 855–
856; e) L. A. Paquette, D. Pissarnitski, L. Barriault, J. Org. Chem.
1998, 63, 7389–7398; f) J. N. Johnston, L. A. Paquette, Tetrahedron
Lett. 1995, 36, 4341–4344; g) R. Dumeunier, I. E. Markꢃ, Tetrahe-
dron Lett. 2000, 41, 10219–10222; h) C. Pꢄrez-Balado, I. E. Markꢃ,
Tetrahedron Lett. 2005, 46, 4887–4890; i) C. Pꢄrez-Balado, I. E.
Markꢃ, Tetrahedron 2006, 62, 2331–2349; j) P. Jourdain, F. Philip-
part, R. Dumeunier, I. E. Markꢃ, Tetrahedron Lett. 2009, 50, 3366–
3370; k) C. S. Barry, N. Bushby, J. R. Harding, C. L. Willis, Org. Lett.
2005, 7, 2683–2686.
[13] a) For studies on synthetic analogues, see: L. Barriault, S. L. Boulet,
K. Fujiwara, A. Murai, L. A. Paquette, M. Yotsu-Yamashita, Bioorg.
Med. Chem. Lett. 1999, 9, 2069–2072; b) for additional physiological
studies, see: E. Cagide, M. C. Louzao, I. R. Ares, M. R. Vieytes, M.
Yotsu-Yamashita, L. A. Paquette, T. Yasumoto, L. M. Botana, Cell
Physiol. Biochem. 2007, 19, 185–194.
[14] a) D. J. Gorin, F. D. Toste, Nature 2007, 446, 395–403; b) A. Fꢀrst-
ner, P. W. Davies, Angew. Chem. 2007, 119, 3478–3519; Angew.
Chem. Int. Ed. 2007, 46, 3410–3449; c) E. Jimꢄnez-NfflÇez, A. M.
Echavarren, Chem. Rev. 2008, 108, 3326–3350; d) A. S. K. Hashmi,
Chem. Rev. 2007, 107, 3180–3211.
Experimental Section
[15] a) A. Fꢀrstner, G. Seidel, Angew. Chem. 1998, 110, 1758–1760;
Angew. Chem. Int. Ed. 1998, 37, 1734–1736; b) A. Fꢀrstner, O.
Guth, A. Rumbo, G. Seidel, J. Am. Chem. Soc. 1999, 121, 11108–
11113; c) A. Fꢀrstner, C. Mathes, C. W. Lehmann, Chem. Eur. J.
2001, 7, 5299–5317.
All experimental details can be found in the Supporting Information.
The material includes compound characterization and copies of spectra
of new compounds.
[16] A. Fꢀrstner, Angew. Chem. 2013, 125; Angew. Chem. Int. Ed. 2013,
52, 2794–2819.
[17] a) A. Fꢀrstner, P. W. Davies, Chem. Commun. 2005, 2307–2320;
b) W. Zhang, J. S. Moore, Adv. Synth. Catal. 2007, 349, 93–120;
c) R. R. Schrock, C. Czekelius, Adv. Synth. Catal. 2007, 349, 55–77.
[18] D. A. Evans, A. H. Hoveyda, J. Am. Chem. Soc. 1990, 112, 6447–
6449.
[19] Review: K. J. Ralston, A. N. Hulme, Synthesis 2012, 2310–2324.
[20] We are aware of only two cases in which an Evans–Tishchenko reac-
tion was used for fragment coupling purposes. The first one was re-
ported by Schreiber and co-workers as part of their rapamycin syn-
thesis, although the aldehyde component was rather simple and
could hence be used in excess (4 equiv). The other example relates
to a study towards the octalactone core reported by Hulme et al.,
wherein 2 equivalents of the aldehyde component were employed,
see: a) D. Romo, S. D. Meyer, D. D. Johnson, S. L. Schreiber, J. Am.
Chem. Soc. 1993, 115, 7906–7907; b) J. I. Aird, A. N. Hulme, J. W.
White, Org. Lett. 2007, 9, 631–634.
[21] a) C.-C. Pai, C.-W. Lin, C.-C. Lin, C.-C. Chen, A. S. C. Chan, W. T.
Wong, J. Am. Chem. Soc. 2000, 122, 11513–11514; b) J. Wu, A. S. C.
Chan, Acc. Chem. Res. 2006, 39, 711–720.
[22] a) H. Kim, S. Ho, J. L. Leighton, J. Am. Chem. Soc. 2011, 133, 6517–
6520; b) review: J. L. Leighton, Aldrichim. Acta 2010, 43, 3–12.
[23] H. C. Brown, K. S. Bhat, J. Am. Chem. Soc. 1986, 108, 293–294.
[24] Z.-Y. Zhan, PCT WO 2007/003135 A1, 2007.
Acknowledgements
Generous financial support by the MPG, the Alexander-von-Humboldt
Foundation (stipend to J.L.), and the Uehara Memorial Foundation (sti-
pend to A.Y.) is gratefully acknowledged. We thank all analytical depart-
ments of our Institute for excellent support.
[1] R. L. Haddock, O. L. T. Cruz, Lancet 1991, 338, 195–196.
[2] M. Yotsu-Yamashita, T. Yasumoto, S. Yamada, F. F. A. Bajarias,
M. A. Formeloza, M. L. Romero, Y. Fukuyo, Chem. Res. Toxicol.
2004, 17, 1265–1271.
[3] The reasons for these sudden outbreaks remained largely obscure;
however, it is reasonable to assume that the polycavernosides as the
causative toxins are actually of microbial origin.
[4] The polycavernosides were isolated in minute quantities from the
natural sources; for example, the amount of 1 in intoxicated algal
samples was estimated as 72–84 nmolkgꢀ1, cf. ref. [2].
[5] M. Yotsu-Yamashita, R. L. Haddock, T. Yasumoto, J. Am. Chem.
Soc. 1993, 115, 1147–1148.
[6] a) M. Yotsu-Yamashita, T. Seki, V. J. Paul, H. Naoki, T. Yasumoto,
Tetrahedron Lett. 1995, 36, 5563–5566; b) M. Yotsu-Yamashita, K.
Abe, T. Seki, K. Fujiwara, T. Yasumoto, Tetrahedron Lett. 2007, 48,
2255–2259-
[7] K. Fujiwara, A. Murai, M. Yotsu-Yamashita, T. Yasumoto, J. Am.
Chem. Soc. 1998, 120, 10770–10771.
[8] a) L. A. Paquette, L. Barriault, D. Pissarnitski, J. Am. Chem. Soc.
1999, 121, 4542–4543; b) L. A. Paquette, L. Barriault, D. Pissarnit-
ski, J. N. Johnston, J. Am. Chem. Soc. 2000, 122, 619–631.
[25] a) R. H. Grubbs, Angew. Chem. 2006, 118, 3845–3850; b) A. Fꢀrst-
ner, Angew. Chem. 2000, 112, 3140–3172; Angew. Chem. Int. Ed.
2000, 39, 3012–3043; c) S. J. Connon, S. Blechert, Angew. Chem.
2003, 115, 1944–1968; d) K. C. Nicolaou, P. G. Bulger, D. Sarlah,
Angew. Chem. 2005, 117, 4564–4601; e) A. H. Hoveyda, A. R. Zhu-
gralin, Nature 2007, 450, 243–251.
4536
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 4532 – 4537

Algal Toxin (ꢀ)-Polycavernoside A
FULL PAPER
[26] Use of the Hoveyda–Grubbs catalyst gave a lower yield, a less fa-
vorable isomer ratio, and a higher amount of dimethyl fumarate,
which was difficult to separate from the desired product.
[27] A. Fꢀrstner, Chem. Commun. 2011, 47, 6505–6511.
[28] The assignment was based on NOESY data and the analysis of the
3JH,H coupling constants.
[29] The preference of E-configured enoates to give trans-substituted
pyran rings in oxy-Michael type reactions has ample precedent; for
a recent review see: H. Fuwa, Heterocycles 2012, 85, 1255–1298.
[30] a) For the use of DBU at elevated temperatures to induce a cis-se-
lective oxy-Michael addition to an E-configured enoate, see: H.
Fuwa, H. Yamaguchi, M. Sasaki, Org. Lett. 2010, 12, 1848–1851;
b) for the use of LiCl/DBU at ambient temperature, see: M. Kane-
matsu, M. Yoshida, K. Shishido, Angew. Chem. 2011, 123, 2666–
2668.
[31] P. G. Bulger, M. G. Moloney, P. C. Trippier, Org. Biomol. Chem.
2003, 1, 3726-3737.
[32] For a related case, see: T. Oshitari, Saiyinbilige, T. Mandai, Hetero-
cycles 2004, 62, 185–190.
[33] For the assignment of the configuration by Mosher ester analysis,
see the Supporting Information.
[34] R. Chinchilla, C. Nµjera, Chem. Rev. 2007, 107, 874–922.
[35] M. A. McGowan, C. P. Stevenson, M. A. Schiffler, E. N. Jacobsen,
Angew. Chem. 2010, 122, 6283–6286; Angew. Chem. Int. Ed. 2010,
49, 6147–6150.
[39] This outcome must also be seen in the light of a related Evans–Tish-
chenko reaction previously used in a total synthesis of polycaverno-
side (ref. [9b]), which was low yielding despite the use of a large
excess of acetaldehyde as one of the coupling partners, see:
[40] a) J. Heppekausen, R. Stade, R. Goddard, A. Fꢀrstner, J. Am.
Chem. Soc. 2010, 132, 11045–11057; b) J. Heppekausen, R. Stade,
A. Kondoh, G. Seidel, R. Goddard, A. Fꢀrstner, Chem. Eur. J. 2012,
18, 10281–10299.
[41] a) K. Micoine, A. Fꢀrstner, J. Am. Chem. Soc. 2010, 132, 14064–
14066; b) S. Benson, M.-P. Collin, A. Arlt, B. Gabor, R. Goddard,
A. Fꢀrstner, Angew. Chem. 2011, 123, 8898–8903; Angew. Chem.
Int. Ed. 2011, 50, 8739–8744; c) K. Lehr, R. Mariz, L. Leseurre, B.
Gabor, A. Fꢀrstner, Angew. Chem. 2011, 123, 11575–11579; Angew.
Chem. Int. Ed. 2011, 50, 11373–11377; d) V. Hickmann, M. Alcara-
zo, A. Fꢀrstner, J. Am. Chem. Soc. 2010, 132, 11042–11044; e) V.
Hickmann, A. Kondoh, B. Gabor, M. Alcarazo, A. Fꢀrstner, J. Am.
Chem. Soc. 2011, 133, 13471–13480; f) W. Chaładaj, M. Corbet, A.
Fꢀrstner, Angew. Chem. 2012, 124, 7035–7039; Angew. Chem. Int.
Ed. 2012, 51, 6929-6933; g) K. Lehr, A. Fꢀrstner, Tetrahedron 2012,
68, 7695–7700.
[42] For the development of this reagent combination, see: a) A. Fꢀrst-
ner, P. W. Davies, T. Gress, J. Am. Chem. Soc. 2005, 127, 8244–8245;
b) A. Fꢀrstner, P. W. Davies, J. Am. Chem. Soc. 2005, 127, 15024–
15025; c) A. Fꢀrstner, C. Aïssa, J. Am. Chem. Soc. 2006, 128, 6306–
6307; d) A. Fꢀrstner, E. K. Heilmann, P. W. Davies, Angew. Chem.
2007, 119, 4844–4847; Angew. Chem. Int. Ed. 2007, 46, 4760–4763.
[43] Adventitious moisture is likely responsible for the hydration of the
intermediate enol ether.
[44] Within the limits of detection, compound 37 was formed as a single
isomer, but the configuration of the anomeric center was not deter-
mined.
[45] L. E. Overman, Angew. Chem. 1984, 96, 565–573; Angew. Chem.
Int. Ed. Engl. 1984, 23, 579–586.
[36] F. Ferreira, A. Denichoux, F. Chemla, J. Bejjani, Synlett 2004, 2051–
2065.
[37] P. Knochel, M. C. P. Yeh, S. C. Berk, J. Talbert, J. Org. Chem. 1988,
53, 2390-2392.
[38] Related retro-aldol/aldol processes were reported for Evans–Tish-
chenko reactions of heteroaromatic aldehydes, see: P. D. Dorgan, J.
Durrani, M. J. Cases-Thomas, A. N. Hulme, J. Org. Chem. 2010, 75,
7475–7478.
[46] C. Nieto-Oberhuber, S. Lꢃpez, A. M. Echavarren, J. Am. Chem.
Soc. 2005, 127, 6178–6179.
[47] A. Fꢀrstner, Chem. Soc. Rev. 2009, 38, 3208–3221.
Received: December 21, 2012
Published online: February 18, 2013
Chem. Eur. J. 2013, 19, 4532 – 4537
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4537