Total synthesis of dinophysistoxin-2 and 2-epi-dinophysistoxin-2 and their PPase inhibition

Article abstract of DOI:10.1002/anie.201101741

The first total syntheses of the title compounds highlight novel assemblies of the C1-C14 and C28-C38 domains, including an unexpected diastereoselectivity in the Sharpless asymmetric dihydroxylation of an alkene at C1iC2. PPase inhibition assays revealed that 2-epi-DTX-2 is at least 1 to 2 orders of magnitude less potent than DTX-2, thus indicating that the configuration at C2 in DTX-2 is crucial for potent inhibition (see picture). Copyright

Full text of DOI:10.1002/anie.201101741

DOI: 10.1002/anie.201101741
Natural Products
Total Synthesis of Dinophysistoxin-2 and 2-epi-Dinophysistoxin-2 and
Their PPase Inhibition**
Yucheng Pang, Chao Fang, Michael J. Twiner, Christopher O. Miles, and Craig J. Forsyth*
Okadaic acid (OA, 1) and its analogues—dinophysistoxin-1
(DTX-1, 2) and -2 (DTX-2, 3)—accumulated in mussels
feeding on certain marine algae species,[1] are causative agents
of “diarrhetic shellfish poisoning” throughout the world.[2]
Okadaic acid and naturally occurring biosynthetic variants
have demonstrated both potent protein serine-threonine
phosphatase (PPase) inhibitory activities and apoptosis-
inducing activities among numerous cancerous cell lines.[3–5]
Circumstantially, these activities appear to be causally
related. Thus, elucidating the structural basis of selective
and potent OA-type PPase inhibition may underlie the
discovery of novel anticancer leads. Herein, we disclose the
first total synthesis of 3 and its nonnatural analogue 2-epi-3,
and their PPase inhibitory activities.
ester at C1)[6] with respect to retaining appreciable PPase
activity. Herein, we show that the configuration at C2 of 3 also
has a dramatic effect on PPase inhibitory activity.
The synthetic design of 3 featured reliable fragment
coupling methods derived from our previous study of 1,[8b] but
also highlights novel approaches to assemble of the C1–C14
and C28–C38 domains. As a result, the overall efficiency of
the assembly of 3 more than doubles that previously
established for 1.[8] Specifically, 1 was prepared in 1.7%
overall yield spanning 26 steps in the longest linear sequen-
ce,[8b] whereas 3 is delivered in 3.6% yield over 21 steps by the
approach outlined herein.
The natural product 3 was designed to derive from enone
4 through a late-stage installation of the spiroketal at C19 by a
reduction/transketalization sequence (Scheme 1).[8b,9] The
two key intermediates, C1–C14 aldehyde 5 and C15–C38 b-
keto phosphonate 6, could be coupled through a Horner–
Wadsworth–Emmons reaction. The spiroketal at C8 in 5 could
be established upon treatment of enone 9 with acid. In turn, 9
could be derived from the opening of lactone 10[10] with the
acetylide derived from alkyne 11. The C15–C38 domain was
À
dissected at the C27 28 bond into aldehyde 7 and bromide 8.
Aldehyde 7 represents the central core of 1-3, whereas 8 is
unique to 3 in its methyl substitution at C35.
The effect of the stereochemistry of the methyl-bearing
carbon 35 of 2 and 3 on PPase inhibitory activity has recently
been described.[5] Compared to 1 and 2, 3 was found to be a
weaker inhibitor of PP2A.[4,5] Thus, variable functionalization
À
at both the C1 C2 region and the C30–C38 domain confers
differential PPase inhibitory activities. Some structural modi-
fications about the C1–C14 domain of 1 have been shown to
be tolerated (e.g. 7-deoxy,[6] C9,10-episulfide[7]), whereas
others are not (2-deoxy, 2-oxo decarboxylate, and methyl
[*] Y. Pang, C. Fang, Prof. C. J. Forsyth
Department of Chemistry, The Ohio State University
100 W. 18th Ave., Columbus, OH 43210 (USA)
E-mail: forsyth@chemistry.ohio-state.edu
Prof. M. J. Twiner
Department of Natural Sciences
The University of Michigan-Dearborn
4901 Evergreen Rd., Dearborn, Michigan 48128 (USA)
Dr. C. O. Miles
Norwegian Veterinary Institute, Chemistry Section
PB 750 Sentrum, NO-0106 Oslo (Norway)
[**] We acknowledge the Ohio State University for financial support, Dr.
T. Young and A.-K. Yassine for technical help, the Ohio BioProduct
Innovation Consortium for mass spectrometry support, and Prof. J.
Stambuli for helpful discussion.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201101741.
Scheme 1. Retrosynthetic disconnections of DTX-2 (3). Bn=benzyl,
PMB=para-methoxyphenyl, TES=triethylsilyl.
Angew. Chem. Int. Ed. 2011, 50, 7631 –7635
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7631

Communications
Our original synthesis of 1 utilized Seebachꢀs lactate
À
pivalidene acetal (i) as the source of the C1 C2 a-hydroxy, a-
methyl carboxylic acid (Scheme 2).[8b,11] Attempts to imbed
this moiety within the okadaic acid intermediate iv through
late-stage alkylation at C2 of the lactate pivalidene enolate
with C3 halides or sulfonates were uniformly unsuccessful.
Instead, an aldol reaction was used with a C3 aldehyde (ii) for
À
C2 C3 bond formation. The facial selectivity of enolate
addition was modest (2:1) and a subsequent Barton deoxy-
genation of the C3 alcohols (iii) was required to complete the
task. Thus, the combined yield was about 46% for the
formation of C1–C14 intermediate iv from i and ii.
Scheme 3. Synthesis of lactone 10. Reagents and conditions:
a) allylMgBr, CuI (cat.), THF, À408C, then NaH, THF, BnBr, TBAI;
b) O3, CH2Cl2, CH3OH, À788C; then PPh3, 70% (2 steps); c) tributyl(2-
methylallyl)stannane, (+)-binol, Ti(OiPr)4, CH2Cl2, molecular sieves
(4 ꢀ), À208C, 99%, d.r.=6:1; d) DDQ, tBuOH, pH 7 buffer, CH2Cl2,
80%; e) TEMPO, BAIB, CH2Cl2, 83%; f) DIBAL, CH2Cl2, À788C to
408C, then TBSCl, imidazole, DMAP, CH2Cl2, 99%. g) AD-mix-b,
tBuOH, H2O, 08C, d.r.=9:1, 92%; h) AD-mix-a, tBuOH, H2O, 08C,
d.r.=3:1, 91%; i) PPTS, CH2Cl2, 2,2-dimethoxypropane; then TBAF,
THF; then TEMPO, BAIB, NaHCO3, CH2Cl2, 88%. BAIB=iodobenzene
diacetate, binol=1,1’-bi-2-naphthol, DDQ=2,3-dichloro-5,6-dicyano-
1,4-benzoquinone, DIBAL=diisobutylaluminum hydride, DMAP=
4-dimethylaminopyridine, PPTS=pyridinium p-toluenesulfonate,
TBAF=tetra-n-butylammonium fluoride, TBAI=tetra-n-butylammo-
nium iodide, TBS=tert-butyldimethyl, TEMPO=2,2,6,6-tetramethylpi-
peridine-1-oxyl.
Scheme 2. Previous synthesis of the C1–C14 domain of 1.[8b]
AIBN=2,2’-azobisisobutyronitrile, LDA=lithium diisopropylamide,
THF=tetrahydrofuran.
An alternative strategy was adopted for the early incor-
poration of the a-hydroxy, a-methyl carboxylate moiety in 3.
This approach involved targeting lactone 10 (Scheme 1), the
masked vicinal diol of which could be ultimately oxidized into
the a-hydroxy carboxylic acid of 3. In the seminal total
synthesis of 1, Isobe et al. installed the core vicinal diol of 10
through a substrate chelation-controlled hydroxymercuration
parison of spectroscopic data with those obtained previously
by Isobe and co-workers[17] The SAD was repeated with 17
but using AD-mix-a, which inverted the sense of diastereo-
selectivity to give a 3:1 diastereomeric ratio of diols 18 and
18a, respectively. The major diastereomer was converted into
a lactone (10) as before (Scheme 3). As a control reaction to
probe potential inherent substrate bias, dihydroxylation of 17
using OsO4 and pyridine/NMO in THF/H2O gave a 1:1 ratio
of diastereomeric diols in good yield.
The differential levels of diastereoselectivity with 17 and
AD-mix-a versus AD-mix-b may represent mismatched
versus matched diastereomeric transition states involving
the pseudoenantiomeric AD-mix ligands dihydroquinine and
dihydroquinidine, respectively. The unexpected sense of
diastereoselectivity in the SAD of 17 can be ascribed to the
homoallylic trisubstituted oxane moiety, which reverses the p-
facial selectivity generally predicted by the Sharpless empiri-
cal model for the AD-mix reagents. These results provide a
caveat to the Sharplessꢀ empirical rules[12] for the SAD facial
selectivity with 17 and perhaps additional olefins of its type.
Hale et al. had reported modest levels of anomalous enantio-
selectivity in the SAD of achiral 1,1-disubstituted methyallyl
alcohol derivatives.[11]
[10]
À
process using a C2 C3 alkene.
Since then, reagent-con-
trolled Sharpless asymmetric dihydroxylation (SAD) has
been well-established to provide reliable and empirically
predictable facial selective vicinal dihydroxylations of
alkenes.[12] We aimed to apply the SAD process with the
À
C1 C2 alkene of 17 en route to 10, which in turn would serve
as a precursor to the C1–C14 domain 5. However, an
unexpected outcome was obtained.
The synthesis of 10 began with conversion of epoxide 12[13]
into aldehyde 13 (Scheme 3). Keck methallylation[14] of 13
installed the stereogenic center at C4 of 3 in homoallylic
alcohol 14. The PMB ether of 14 was cleaved to generate 1,5-
diol 15, which was oxidatively lactonized to 16 with TEMPO
and BAIB.[15] The SAD process was originally applied to 16
but undesired saponification occurred under the basic AD-
mix conditions. Alternatively, lactone 16 was converted into
mixed acetal 17 through a reduction/silylation sequence.
Application of AD-mix-b[12a] dihydroxylaton to 17 gen-
erated the vicinal primary/tertiary diol 18a in 9:1 diastereo-
selectivity. It was anticipated that diol 18 would be the major
diastereomer according to Sharplessꢀ empirical rules[12] and
most previous applications of SAD with terminal unsym-
metrical disubstituted alkenes.[12a,16] However, after convert-
ing the major diol diastereomer 18a into the corresponding d-
lactone 10a it was found that the SAD reaction had given the
opposite diastereoselectivity. This was determined by com-
The newly devised synthesis of the C1–C14 domain 5
continued with opening of lactone 10 with the lithium
acetylide generated from 19[8b] to give ynone 21 (Scheme 4).
After silylation of the residual alcohol, conjugate addition of
methylcuprate afforded an E/Z mixture of enones 22. Cleav-
age of the silyl ether groups and spiroketalization using PPTS
in CH2Cl2 and methanol yielded thermodynamically favored
23. To complete the synthesis of 5, the PMB ether of 23 was
7632
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 7631 –7635

dehydratively spiroketalized the in situ generated keto
triol to 30 in 93% yield of isolated product. Thereafter,
the residual alcohol was converted into bromide 8.[22]
The C16–C27 aldehyde 7, prepared from 31,[23] was
coupled with the alkyl lithium derived from
8
(Scheme 6).[24] The resultant mixture of epimeric alco-
hols 32a and 32 were separated and 32a was converted
into 32 through oxidation/reduction.[8a,d] After benzyla-
tion of alcohol 32, the less-hindered alkene in 33 was
oxidized to aldehyde 34 by treatment with OsO4 and
NaIO4.[23] A sequence of phosphonate anion addition
and oxidation using Dess–Martin periodinane con-
verted 34 into b-ketophosphonate 6.
Scheme 4. Synthesis of aldehyde 5. Reagents and conditions: a) TESCl,
imidazole, DMAP, CH2Cl2, 99%; b) nBuLi, THF, À788C; then 10, 95%;
c) TMSCl, imidazole, CH2Cl2. d) MeLi, CuI, Et2O, À408C; e) PPTS, CH2Cl2,
MeOH, 38% (3 steps); f) DDQ, tBuOH, H2O, pH 7 buffer; g) DMP, NaHCO3,
CH2Cl2, 82% (2 steps). DMP=Dess–Martin periodinane, TMS=trimethylsilyl.
oxidatively cleaved and the resultant primary alcohol was
oxidized using the Dess–Martin periodinane reagent.[18] This
preparation of the C1–C14 domain of 1–3 is unprecedented in
its overall efficiency (8.0% overall yield over 15 steps from
12).
The synthesis of the C15–C38 domain 6 relied upon
generation of the common central C15–C27 intermediate 7
and the unique C28–C38 bromide 8. The newly developed
synthesis of the C30–C40 terminal spiroketal domain of 3
necessarily deviated from that previously used for 1.
Although a Keck crotylation was used to set the configu-
rations at vicinal C30 and C31 stereogenic centers in our
original synthesis of 1;[8b] the stereogenic centers at C29, C30,
and C35 in 3 were established reliably by Evans asymmetric
aldol and alkylation processes. Imide 24[19] provided access to
the C33–C38 phosphonate 25, and the complementary C28–
C32 aldehyde 28 was obtained from aldol adduct 26[20]
(Scheme 5). Reduction of 26 afforded a diol that was
protected as dibenzyl ether 27. The PMB ether of 27 was
selectively cleaved and the resultant alcohol was oxidized to
aldehyde 28. b-Ketophosphonate 25 was then coupled with
28.[21] The resulting enone 29 was converted into the
thermodynamically favored spiroketal 30 in a single process.
Hydrogenation of 29 using Pearlmanꢀs catalyst in ethanol
saturated the alkene, cleaved the three benzyl ethers, and
Scheme 6. Synthesis of 6. Reagents and conditions: a) TBAF, THF, 100%;
b) DMP, NaHCO3, CH2Cl2; c) 8, tBuLi, Et2O À788C to 258C; then 7, À788C,
48%, d.r.=1:1; d) DMP, NaHCO3, CH2Cl2; e) NaBH4, MeOH, À208C, 90%
(2 steps); f) NaH, THF; then BnBr, TBAI, 93%; g) OsO4, NaIO4, THF, H2O, 77%;
h) dimethyl methylphosphonate, tBuLi, THF, À788C; i) DMP, NaHCO3, CH2Cl2,
70% (2 steps).
Only six steps from 5 and 6 were required to complete the
first total synthesis of 3. First, aldehyde 5 and b-ketophosph-
onate 6 were coupled in high yield to give (E)-enone 4
(Scheme 6). Diastereoselective reduction of the ketone using
(S)-CBS catalyst[25] converted 4 into the (16R)-alcohol, which
upon treatment with acetic acid in THF and water underwent
intramolecular transketalization and acetonide hydrolysis to
yield diol 35 (Scheme 7). Sequential Parikh–Doering[26] and
Pinnick[27] oxidations converted 35 into a-hydroxy acid 36.
Finally, the three benzyl groups were cleaved in THF at
À788C using freshly prepared LiDBB[28] to provide 3.
Moreover, 2-epi-DTX-2 was also prepared in the same
fashion as 3 by replacing 10 with 10a in the synthetic
sequence. Confirmation of the identity of synthetic 3 with
naturally occurring DTX2 was obtained by comparison of
LC-HRMS and 1H NMR data.[29]
Naturally isolated 1 and 3[30] were tested in parallel
with synthetic 3 and 2-epi-3 for their PP2A and PP1
inhibition activity. All toxins were shown to inhibit each of
these enzymes but with various degrees of potency
(Figure 1). Although PP2A was always more sensitive
than PP1, the relative order of potency of the toxins
remained the same for both enzymes: nat. 1 > nat. 3 ꢀ
synthetic 3 > synthetic 2-epi-3. Respective IC50 values
(nm) for PP2A are 0.47, 0.99, 1.35, and 137, and for PP1
are 25.2, 76.4, 82.6, and 3114 (Table 1 in the Supporting
Information). The IC50 value for 1 on PP2A and PP1 is
Scheme 5. Synthesis of bromide 8. Reagents and conditions: a) MeMgBr,
MeOH, 08C; b) dimethyl methylphosphonate, nBuLi, THF, À788C, 76%
(2 steps); c) NaBH4, THF, H2O; d) NaH, BnBr, TBAI, THF, 67% (2 steps);
e) DDQ, tBuOH, pH 7 buffer, CH2Cl2; f) py·SO3, DMSO, iPr2NEt, CH2Cl2,
87% (2 steps); g) 25, LiCl, iPr2NEt, CH3CN, 97%; h) H2, Pd(OH)2/C,
EtOH, 93%; i) CBr4, PPh3, Et3N, CH2Cl2, 95%. DMSO=dimethyl sulfoxide,
py=pyridine.
Angew. Chem. Int. Ed. 2011, 50, 7631 –7635
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7633

Communications
the active sites of the respective enzymes in either similar
or considerably different conformations. For the former,
the hydroxy and methyl substituents at C2 of 3 and 2-epi-
3 would be projected in opposite directions. For 2-epi-3
this may cause the loss of potentially favorable contacts
between the hydroxyl group at C2 and Tyr272 and
Arg96, and between the methyl group at C2 and His125
in PP1.[33] Similar disruption of hydrogen bonding
contacts between the hydroxyl group at C2 and Tyr265
and Arg89 in PP2A may occur.[32] The marriage of novel
fragment assemblies with reliably established tricompo-
nent couplings described here will provide access to
further structural analogues derived from both 3 and the
novel nonnatural analogue 2-epi-3. These unique mole-
cules will facilitate studies to further elucidate the
structural basis of selective PPase regulation and asso-
ciated biological consequences.
Scheme 7. Total synthesis of 3. Reagents and conditions: a) LiCl, iPr2NEt,
CH3CN, 93%; b) (S)-Me-CBS, BH3·THF, THF, 08C; c) AcOH, THF, H2O,
558C, 36 h, 80%; d) py·SO3, DMSO, iPr2NEt, CH2Cl2; e) NaClO2, NaH2PO4,
2-methyl-2-butene, tBuOH, H2O, 08C, 89%; f) LiDBB, THF, À788C, 69%.
DBB=4,4’-di-tert-butylbiphenyl, (S)-Me-CBS=Corey–Bakshi–Shibata oxaboro-
lidine.
Received: March 10, 2011
Revised: May 27, 2011
Published online: July 1, 2011
Keywords: natural products · phosphatase inhibitors ·
.
structure–activity relationships · total synthesis
[1] a) K. Tachibana, P. J. Scheuer, Y. Tsukitani, H. Kikuchi, D.
Van Engen, J. Clardy, Y. Gopichand, F. J. Schmitz, J. Am. Chem.
Soc. 1981, 103, 2469 – 2471; b) M. Murata, M. Shimatani, H.
Sugitani, Y. Oshima, T. Yasumoto, Bull. Jpn. Soc. Sci. Fish. 1982,
48, 549 – 552; c) T. Hu, J. Doyle, D. Jackson, J. Marr, E. Nixon, S.
Pleasance, M. A. Quilliam, J. A. Walter, J. L. C. Wright, J. Chem.
Soc. Chem. Commun. 1992, 39 – 41.
[2] C. F. B. Holmes, H. A. Luu, F. Carrier, F. J. Schmitz, FEBS Lett.
1990, 270, 216 – 218.
[3] T. Yasumoto, M. Murata, Chem. Rev. 1993, 93, 1897 – 1909.
[4] T. Aune, S. Larsen, J. A. B. Aasen, N. Rehmann, M. Satake, P.
Hess, Toxicon 2007, 49, 1 – 7.
[5] a) K. Larsen, D. Petersen, A. L. Wilkins, I. A. Samdal, M.
Sandvik, T. Rundberget, D. Goldstone, V. Arcus, P. Hovgaard, F.
Rise, N. Rehmann, P. Hess, C. O. Miles, Chem. Res. Toxicol.
2007, 20, 868 – 875; b) J. Huhn, P. D. Jeffrey, K. Larsen, T.
Rundberget, F. Rise, N. R. Cox, V. Arcus, Y. Shi, C. O. Miles,
Chem. Res. Toxicol. 2009, 22, 1782 – 1786.
Figure 1. Inhibitory effects of natural OA (1), natural and synthetic
DTX-2 (3), and synthetic 2-epi-DTX-2 on (A) PP2A and (B) PP1 activity.
[6] A. Takai, M. Murata, K. Torigoe, M. Isobe, G. Mieskes, T.
Yasumoto, Biochem. J. 1992, 284, 539 – 544.
[7] F. J. Schmitz, R. S. Prasad, Y. Gopichaud, M. B. Hossain, D.
van der Helm, P. Schmidt, J. Am. Chem. Soc. 1981, 103, 2467 –
2469.
within the range of other published reports,[5b,6,31] as is the
IC50 value for 3 (both natural and synthetic) on PP2A
(3.5 nm).[32]
[8] a) M. Isobe, Y. Ichikawa, D. L. Bai, H. Masaki, T. Goto,
Tetrahedron 1987, 43, 4767 – 4776; b) C. J. Forsyth, S. F. Sabes,
R. A. Urbanek, J. Am. Chem. Soc. 1997, 119, 8381 – 8382; c) S. F.
Sabes, R. A. Urbanek, C. J. Forsyth, J. Am. Chem. Soc. 1998, 120,
2534 – 2542; d) R. A. Urbanek, S. F. Sabes, C. J. Forsyth, J. Am.
Chem. Soc. 1998, 120, 2523 – 2533; e) S. V. Ley, A. C. Humphries,
H. Eick, R. Downham, A. R. Ross, R. J. Boyce, J. B. J. Pavey, J.
Pietruszka, J. Chem. Soc. Perkin Trans. 1 1998, 3907 – 3911.
[9] A. B. Dounay, R. A. Urbanek, S. F. Sabes, C. J. Forsyth, Angew.
Chem. 1999, 111, 2403 – 2406; Angew. Chem. Int. Ed. 1999, 38,
2258 – 2262.
These studies indicate that the nonnatural synthetic
analogue 2-epi-3 is at least 1 to 2 orders of magnitude less
potent than 3 (101- and 38-fold for PP2A and PP1, respec-
tively). The fact that the C2 epimer of DTX2 has such a
remarkably decreased potency (IC50 = 137 and 3114 nm for
PP2A and PP1, respectively) indicates biogenetic optimiza-
tion of PPase inhibitory activity and the essential roles of
detailed structural features within the C1–C12 domain for
conferring biological activity of okadaic acid and its con-
geners.[6]
[10] Compound 10 was previously prepared by Isobe using an
oxymercuration/demercuration protocol to set the tertiary
stereogenic center: M. Isobe, Y. Ichikawa, T. Goto, Tetrahedron
Lett. 1986, 26, 5199 – 5202.
The structural basis for the differential PPase inhibition
by 3 and 2-epi-3 may involve docking of both compounds at
7634
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 7631 –7635

[11] K. F. Hale, S. Manaviazar, S. A. Peak, Tetrahedron Lett. 1994, 35,
425 – 428.
[21] M. A. Blanchette, W. Choy, J. T. Davis, A. P. Essenfeld, S.
Masamune, W. R. Roush, T. Sakai, Tetrahedron Lett. 1984, 25,
2183 – 2186.
[22] R. Appel, Angew. Chem. 1975, 87, 863 – 874; Angew. Chem. Int.
Ed. Engl. 1975, 14, 801 – 811.
[12] a) H. C. Kolb, M. S. VanNieuwenhz, K. B. Sharpless, Chem. Rev.
1994, 94, 2483 – 2547; b) K. B. Sharpless, W. Amberg, Y. L.
Bennani, G. A. Crispino, J. Hartung, K. S. Jeong, H.-L. Kwong,
K. Morikawa, Z.-M. Wang, D. Xu, X.-L. Zhang, J. Org. Chem.
1992, 57, 2768 – 2771.
[23] A. B. Dounay, R. A. Urbanek, V. A. Frydrychowski, C. J.
Forsyth, J. Org. Chem. 2001, 66, 925 – 938.
[13] M. E. Furrow, S. E. Schaus, E. N. Jacobsen, J. Org. Chem. 1998,
63, 6776 – 6777.
[14] G. E. Keck, K. H. Tarbet, L. S. Gareci, J. Am. Chem. Soc. 1993,
115, 8467 – 8468.
[15] T. M. Hansen, G. J. Florence, P. Lugo-Mas, J. Chen, J. N. Abrams,
C. J. Forsyth, Tetrahedron Lett. 2003, 44, 57 – 59.
[16] A. B. Smith III, C. Sfouggatakis, C. A. Risatti, J. B. Sperry, W.
Zhu, V. A. Doughty, T. Tomioka, D. B. Gotchev, C. S. Bennett, S.
Sakamoto, O. Atasoylu, S. Shirakami, D. Bauer, M. Takeuchi, J.
Koyanagi, Y. Sakamoto, Tetrahedron 2009, 65, 6489 – 6509.
[17] Y. Ichikawa, M. Isobe, D. L. Bai, T. Goto, Tetrahedron 1987, 43,
4737 – 4748.
[24] C. Fang, Y. Pang, C. J. Forsyth, Org. Lett. 2010, 12, 4528 – 4531.
[25] E. J. Corey, R. K. Bakshi, S. Shibata, C. P. Chen, V. K. Singh, J.
Am. Chem. Soc. 1987, 109, 7925 – 7926.
[26] J. R. Parikh, W. v. E. Doering, J. Am. Chem. Soc. 1967, 89, 5505 –
5507.
[27] B. S. Bal, W. E. Childers, Jr., H. W. Pinnick, Tetrahedron 1981,
37, 2091 – 2096.
[28] P. K. Freeman, L. L. Hutchinson, J. Org. Chem. 1980, 45, 1924 –
1930.
[29] See the Supporting Information.
[30] A sample of naturally occurring DTX2 was obtained from the
National Research Council (Canada).
[18] D. B. Dess, J. C. Martin, J. Org. Chem. 1983, 48, 4155 – 4156.
[19] K. H. Altmann, G. Bold, G. Caravatti, D. Denni, A. Florsheimer,
A. Schmidt, G. Rihs, M. Wartmann, Helv. Chim. Acta 2002, 85,
4086 – 4110.
[20] L. C. Dias, S. M. Pinheiro, V. M. de Oliveira, M. A. B. Ferreira,
C. F. Tormena, A. M. Aguilar, J. Zukerman-Schpector, E. R. T.
Tiekink, Tetrahedron 2009, 65, 8714 – 8721.
[31] C. Bialojan, A. Takai, Biochem. J. 1998, 256, 283 – 290.
[32] Y. Xing, Y. Xu, Y. Chen, P. D. Jeffrey, Y. Chao, Z. Lin, Z. Li, S.
Strack, J. B. Stock, Y. Shi, Cell 2006, 127, 341 – 353.
[33] J. T. Maynes, K. S. Bateman, M. M. Cherney, A. K. Das, H. A.
Luu, C. F. B. Holmes, M. N. G. James, J. Biol. Chem. 2001, 276,
44078 – 44882.
Angew. Chem. Int. Ed. 2011, 50, 7631 –7635
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7635