Enantioselective synthesis of the protein phosphatase inhibitor (-)-motuporin

Article abstract of DOI:10.1021/ja0206700

A highly convergent asymmetric synthesis of the protein phosphatase inhibitor motuporin 1a is described. Synthesis and coupling of the individual peptide fragments [34 + -35 → 51] followed by macrocyclization afforded the fully protected motuporin precursor 33, which is converted to the natural product by dehydration and ester hydrolysis, Six of the eight stereogenic centers associated with the natural product were introduced using asymmetric crotylsilane bond construction methodology. Our approach features an efficient Pd(0)-catalyzed cross-coupling reaction between a configurationally well-defined vinyl zinc intermediate 22 and an (E)-vinyl iodide 7, which afforded compound 43, resulting in the construction of the trisubstituted (E,E)-diene system of the motuporin side chain. Improved reaction conditions for macrocyclization in the formation of 33 are also detailed.

Full text of DOI:10.1021/ja0206700

Published on Web 08/31/2002
Enantioselective Synthesis of the Protein Phosphatase
Inhibitor (-)-Motuporin
Tao Hu and James S. Panek*,†
Contribution from the Department of Chemistry, Metcalf Center for Science and Engineering,
590 Commonwealth AVenue, Boston UniVersity, Boston, Massachusetts 02215
Received May 9, 2002
Abstract: A highly convergent asymmetric synthesis of the protein phosphatase inhibitor motuporin 1a is
described. Synthesis and coupling of the individual peptide fragments [34 + 35 f 51] followed by
macrocyclization afforded the fully protected motuporin precursor 33, which is converted to the natural
product by dehydration and ester hydrolysis. Six of the eight stereogenic centers associated with the natural
product were introduced using asymmetric crotylsilane bond construction methodology. Our approach
features an efficient Pd(0)-catalyzed cross-coupling reaction between a configurationally well-defined vinyl
zinc intermediate 22 and an (E)-vinyl iodide 7, which afforded compound 43, resulting in the construction
of the trisubstituted (E,E)-diene system of the motuporin side chain. Improved reaction conditions for
macrocyclization in the formation of 33 are also detailed.
(PP2A).11 In vivo, these compounds have been reported to
promote tumor formation,12 suppress cell transformation,13 or
Introduction
It is now widely accepted that modulation of the reversible
phosphorylation of proteins, as catalyzed by protein kinases
(PKs) and protein phosphatases (PPs), is the principal mecha-
nism by which eukaryotic cells respond to external stimuli.1
The balance between phosphorylated and dephosphorylated
proteins, which is controlled by the protein kinase and phos-
phatase activities, is crucial to maintaining proper cellular
function.2 Excessive protein phosphorylation, whether through
the activation of kinases or through the inhibition of phos-
phatases, can lead to uncontrolled cellular proliferation, sug-
gesting the possibility of an active role for the phosphatases in
tumor suppression.3 The controlled disruption of this balance
through the use of PPs inhibitors is a method by which one can
dissect this complex system.
There is a diverse group of structurally interesting natural
products that act by inhibiting certain phosphatases, thereby
disrupting the normal biochemical pathway. Examples include
okadaic acid,4 tautomycin,5 calyculin,6 cantharidine,7 micro-
cystins,8 nodularins,9 and motuporin,10 all of which are potent
competitive inhibitors of two major classes of phosphatases,
protein phosphatase 1 (PP1), and protein phosphatase 2A
induce apoptosis,14 depending on length of exposure and the
cell type.
Motuporin (1a) is a cyclic pentapeptide isolated in 1992 by
Andersen and co-workers from the marine sponge Theonella
swinhoei Gray collected in Papua New Guinea.10 This molecule
is one of the most potent PP1 inhibitors known, inhibiting at
sub-nanomolar concentrations (IC50 < 1.0 nM) and showing
strong in vitro cytotoxicity against a variety of human cancer
cells. The structures of motuporin and cyanobacterial derived
natural product nodularin (1b) show remarkable resemblance
to each other. Specifically, these cyclic peptides contain the
unusual â-amino acid (2S,3S,8S,9S,4E,6E)-3-amino-9-methoxy-
2,6,8-trimethyl-10-phenyldeca-4,6-dienoic acid (Adda),9b an R,â-
unsaturated amino acid N-methyldehydrobutyrine (N-Me∆But),
an isolinked D-glutamate, and a â-methyl D-aspartate (â-MeAsp)
residue (Figure 1).
Motuporin was isolated later than other members of its class
and is derived from a much less readily accessible source.
(8) (a) Botes, D.; Tuinman, A.; Wessels, P.; Viljoen, C.; Kruger, H.; Williams,
D. H.; Santikarn, S.; Smith, R.; Hammond, S. J. Chem. Soc. Perkin Trans.
1 1984, 2311-2318. (b) Painuly, P.; Perez, R.; Fukai, T.; Shimizu, Y.
Tetrahedron Lett. 1988, 29, 11-14.
* To whom correspondence should be addressed. E-mail: Panek@
chem.bu.edu.
(9) (a) Botes, P. B.; Wessels, P. L.; Kruger, H.; Runnegar, M. T. C.; Santikarn,
S.; Smith, R. J.; Barna, J. C. J.; Williams, D. H. J. Chem. Soc. Perkin
Trans. 1 1985, 2747-2748. (b) Rinehart, K. L.; Harada, K.; Namikoshi,
M.; Chen, C.; Harvis, C. A.; Munro, M. H. G.; Blunt, J. W.; Mulligan, P.
E.; Beasley, V. R.; Dahlem, A. M.; Carmichael, W. W. J. Am. Chem. Soc.
1988, 110, 8557-8558.
† Recipient of a 2002 Arthur C. Cope Scholar Award.
(1) Hunter, T. Cell 1995, 80, 225-236.
(2) Mehrotra, A. P.; Webster, K. L.; Gani, D. J. Chem. Soc., Perkin Trans. 1
1997, 2495-2511.
(3) (a) Cohen, P.; Cohen, P. T. W. J. Biol. Chem. 1989, 264, 21435-21438.
(b) Krebs, E. G. Angew. Chem., Int. Ed. Engl. 1993, 32, 1122-1129.
(4) Tachibana, K.; Scheuer, P. J.; Tsukitani, Y.; Kikuchi, Y.; Van Engen, D.;
Clardy, J.; Gopichand, Y.; Schmitz, F. J. J. Am. Chem. Soc. 1981, 103,
2469-2471.
(10) Dilip de Silva, E.; Williams, D. E.; Andersen, R. J.; Klix, H.; Holmes, C.
F. B.; Allen, T. M. Tetrahedron Lett. 1992, 33, 1561-1564.
(11) Takai, A.; Sasaki, K.; Nagai, H.; Mieskes, G.; Isobe, M.; Isobe, M.; Isons,
K.; Yasumoto, T. Biochem. J. 1995, 306, 657-662.
(12) Nishiwaki, M. R.; Ohta, T.; Nishiwaki, S.; Suganuma, M.; Kohyama, K.;
Ishikawa, T.; Carmichael, W. W.; Fujiki, H. J. Cancer Res. Clin. Oncol.
1992, 118, 420-424.
(5) Cheng, X. C.; Ubukata, M.; Isono, K. J. Antibiot. 1990, 43, 809-819.
(6) Kato, Y.; Fusetani, N.; Matsunaga, S.; Hashimoto, K.; Fujita, S.; Furuya,
T. J. J. Am. Chem. Soc. 1986, 108, 2780-2881.
(7) Li, Y. M.; Casida, J. E.; Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 11867-
11870.
(13) Mordon, L. J. Science 1991, 251, 142-143.
(14) Boee, R.; Gjertsen, B. T.; Vintermyr, O. K.; Houge, G.; Lanotte, M.;
Doeskeland, S. O. Exp. Cell Res. 1991, 195, 237-246.
9
11368
J. AM. CHEM. SOC. 2002, 124, 11368-11378
10.1021/ja0206700 CCC: $22.00 © 2002 American Chemical Society

Protein Phosphatase Inhibitor (−)-Motuporin
A R T I C L E S
Scheme 1
Figure 1.
Consequently much of the information concerning its biology
and chemistry is extrapolated from data on nodularin and the
structurally related heptapeptide microcystins. Of the nanomolar-
binding agents of small molecule protein phosphatase inhibitors,
1a is an especially attractive target for total synthesis for several
reasons, including its relative scarcity, its reported high potency,
and its unique structure.
The biological activity and the structural complexity exhibited
by the cyclic peptide PPs inhibitors have stimulated much
activity within the synthetic community. Thus far, four asym-
metric syntheses of motuporin have been accomplished inde-
pendently, with the first total synthesis being reported in 1995
from Schreiber’s group.15 In addition, Chamberlin and co-
workers have recently accomplished a total synthesis of
structural related natural product microcystin.16 Here we de-
scribe, in full, the design and execution of a convergent and
flexible synthetic strategy for motuporin.
Scheme 2. Synthetic Plan for Adda and â-MeAsp Residues
Results and Discussion
After a careful analysis of the amino acid constituents of
motuporin, one recognizes a pair of common R-amino acids
and three unusual residues (Scheme 1). Our approach to the
synthesis of 1a would necessarily employ L-valine and D-
glutamate as starting materials and require development of
efficient syntheses of three unnatural amino acids, Adda,
â-MeAsp, and N-Me∆But.
We envisioned that the Adda fragment, possessing syn-related
methyl-oxygen stereogenic centers at C8-C9, could be
constructed using asymmetric crotylation methodology (Scheme
2).17 In addition, the crotylation to the dimethyl acetal allows
the direct introduction of the C9-methyl ether. Further examina-
tion of the Adda and â-MeAsp residues identified a common
stereochemical triad; a pair of carbon atoms bearing vicinal
amino and methyl groups in an anti relationship is shared by
both â-amino acids. Utilizing a chiral silane reagent bearing an
azide alpha to the ester,18 the â-methyl aspartate equivalent could
also be synthesized with high selectivity.
Retrosynthetic Analysis. On the basis of isolation and
biosynthetic predictions of the natural product,19 we anticipated
that cyclization of the 19-membered macrocycle could be
effected at the C21 of the valine-Adda amide bond (Scheme
3). With further disconnection of the linear pentapeptide 2 at
the C1-carboxyl of the Adda residue, a convergent approach
was then developed which involved the fragment coupling of
two advanced intermediates, the N-Boc Adda 3 and C21-C40
tetrapeptide fragments 4. This plan allowed the incorporation
of the valuable Adda residue late in the synthesis. The orthog-
onal protecting group scheme of the tetrapeptide 4 was com-
posed of a N-Boc group, 2-trimethylsilylethyl ester (TMSE),
and the dehydroamino acid moiety (N-Me∆But), which was
masked as a â-hydroxy amide until late in the synthesis. In the
(15) (a) Valentekovich, R. L.; Schreiber, S. L. J. Am. Chem. Soc. 1995, 117,
9069-9070. (b) Bauer, S. M.; Armstrong, R. W. J. Am. Chem. Soc. 1999,
121, 6355-6366. (c) Samy, R.; Kim, H. Y.; Brady, M.; Toogood, P. J.
Org. Chem. 1999, 64, 2711-2728. (d) Hu, T.; Panek, J. S. J. Org. Chem.
1999, 64, 3000-3001.
(16) Humphrey, J. M.; Aggen, J. B.; Chamberlin, A. R. J. Am. Chem. Soc. 1996,
118, 11759-11770.
(17) (a) Panek, J. S.; Yang, M. J. Am. Chem. Soc. 1991, 113, 6594-6600. (b)
Panek, J. S.; Yang, M. J. Org. Chem. 1991, 56, 5755-5758. (c) Panek, J.
S.; Yang, M. J. Am. Chem. Soc. 1991, 113, 9868-9870. (d) For review,
see: Masse, C. E.; Panek, J. S. Chem. ReV. 1995, 95, 1293-1316.
(18) Panek, J. S.; Beresis, R.; Xu, F.; Yang, M. J. Org. Chem. 1991, 56, 7341-
7344.
(19) Choi, B. W.; Namikoshi, F. S.; Rinehart, K. L.; Carmichael, W. W.; Kaup,
A. M.; Evans, W. R.; Beasley, V. R. Tetrahedron Lett. 1993, 34, 7881-
7884.
9
J. AM. CHEM. SOC. VOL. 124, NO. 38, 2002 11369

A R T I C L E S
Hu and Panek
Scheme 5. Retrosynthetic Analysis of Tetrapeptide Fragment
Scheme 3. Retrosynthetic Analysis of Motuporin
Scheme 4. Retrosynthetic Analysis of N-Boc Adda Fragment
(E)-vinyl iodide 6 (C1-C5 Subunit). A palladium(0)-mediated
cross-coupling reaction could be used to construct the (E,E)-
diene employed a vinylmetal intermediate derived from acety-
lene 5. Further analysis of the individual subunits produced two
silane reagents, of which the more functionalized anti-azido
silane 8 was obtained from the (S)-silane reagent 7 through the
stereoselective azidation of a derived â-silyl enolate.18 In this
approach, the C1-C5 and C6-C10 subunits were ultimately
derived from the same chiral silane reagent 7.
The retrosynthesis of the C21-C40 tetrapeptide 4 is shown
in Scheme 5. Amide bond cleavage at C31 carbonyl afforded
two dipeptide segments 9 and 10, which can be further
disconnected to give four protected amino acids or their
equivalents 11-14. As described earlier in Scheme 2, the
erythro-(D)-â-Me-Asp equivalency 13 would be derived from
the azido silane 8.
Synthesis of N-Boc Adda Fragment 3. During the course
of our studies, several reports had appeared describing the
synthesis of Adda.21 The majority of these approaches rely on
the use of phosphorus-based olefination procedures to assemble
the trisubstituted (E,E)-diene, which often produces mixtures
of isomers. To complement that strategy, we have designed a
highly convergent and stereocontrolled synthesis of Adda which
is based on a Pd(0) cross-coupling strategy.22 The preparation
of the C6-C10 subunit 5 relied on a syn-crotylation for the
installation of the C8-C9 stereochemical relationship. The
synthesis was initiated with a BF3‚OEt2 (1.2 equiv) promoted
condensation of silane (S)-7 with phenyl acetaldehyde dimethyl
synthesis reported by Schreiber and Valentekovich it was shown
that a Ba(OH)2 catalyzed saponification of the C29 and C40
methyl esters also resulted in stereoselective dehydration of the
threonine residue to afford the natural product motuporin.15a The
stereochemical course of the dehydration is consistent with NMR
studies on the solution structure of motuporin which, suggested
that the (Z)-geometry of the olefin is preferred on thermody-
namic considerations.20
The retrosynthesis of N-Boc protected Adda fragment 3 is
shown in Scheme 4. The first disconnection at the C5-C6 bond
produced two subunits including the syn-homopropargylic ether
5 (C6-C10 subunit) and secondary allylic amine bearing an
(21) (a) Namikoshi, M.; Rinehart, K. L.; Dahlem, A. M.; Beasley, V. R.;
Carmichael, W. W. Tetrahedron Lett. 1989, 30, 4349-4352. (b) Chakraborty,
T. K.; Joshi, S. P. Tetrahedron Lett. 1990, 31, 2043-2046. (c) Beatty, M.
F.; White, C. J.; Avery, M. A. J. Chem. Soc., Perkin Trans. 1 1992, 1637-
1641. (d) Kim, H. Y.; Toogood, P. L. Tetrahedron Lett. 1996, 37, 2349-
2352. (e) D’Aniello, F.; Mann, A.; Taddei, M. J. Org. Chem. 1996, 61,
4870-4871. (f) Sin, N.; Kallmerten, J. Tetrahedron Lett. 1996, 37, 5645-
5648. (g) Cundy, D. J.; Donohue, A. C.; McCarthy, T. D. Tetrahedron
Lett. 1998, 39, 5125-5128.
(22) Preliminary communication on the synthesis of 3, see: Panek, J.; Hu, T.
J. Org. Chem. 1997, 62, 4914-4915.
(20) See ref 15 for detailed discussion.
9
11370 J. AM. CHEM. SOC. VOL. 124, NO. 38, 2002

Protein Phosphatase Inhibitor (−)-Motuporin
A R T I C L E S
Scheme 7. Strategy for Introducing the C4, C5 Stereocenters
Scheme 6. Synthesis of the C6-C10 Subunit
Table 1. Lewis Acid Promoted Azide-Silane Addition to
(s)-Trioxane (eq 1)
Lewis acid
time (h)/temp (°C)
dr of 20a
yield of20b (%)
yield of 21b (%)
SbCl5
TiCl4
SnCl4
AlCl3
TMSOTf
BF3‚OEt2
2/-78
>30:1
>30:1
>30:1
>30:1
>30:1
>30:1
34
56
38
30
0
0
33
25
48
83
91
acetal 15 (-78 °C f -50 °C) to afford the homoallylic ether
16 in 92% yield, with 10:1 (syn/anti) diastereoselectivity. The
stereochemical course of the crotylation is based on an anti
SE′-open transition state model illustrated in Scheme 6.23 Two
important features that may account for the useful level of
diastereo- and enantioselectivty bear mention. First, the large
silicon group is positioned anti-periplanar to the π-orbital of
the oxonium ion, so maximum stabilization of the emerging
secondary carbocation (â to the silyl group) could be achieved.
Second, the proper orientation of the vinyl methyl group is
required to minimize steric destabilization among the diaster-
eomeric transition states. Oxidative cleavage of the trans-double
bond of 16 gave the desired aldehyde 17, which was found to
be prone to epimerization when subjected to purification on SiO2
(Scheme 6). Accordingly, this material was used directly without
purification in the next reaction after extractive isolation. The
dibromoolefination24 of the aldehyde 17 gave the desired
acetylene precursor 18 in 88% yield (two steps from 16).
Exposure of the 18 to Corey-Fuchs conditions24 (2.2 equiv of
nBuLi, THF, -78 °C) followed by trapping of the intermediate
acetylenic anion with MeI (5.0 equiv) led to the substituted
acetylene 5 in 98% yield.
The major challenge in the synthesis of C1-C5 subunit 6
was to control the stereochemistry of the adjacent methyl and
amine groups. We had anticipated that a crotylation using 8 to
formaldehyde would produce allylic azide 19. This initial
intermediate was anticipated to undergo a facile, and stereose-
lective allylic azide isomerization25 to generate 1,3-azido alcohol
20 (Scheme 7), bearing a vicinal anti methyl-azide relation-
ship.26 This transformation was especially attractive since the
stereocontrolled allylic azide isomerization is a highly under-
developed reaction and capable of rapidly building molecular
complexity. Further transformation of intermediate 20 into the
requisite C1-C5 subunit 6 would follow a straightforward
sequence.
14/-35
14/-35
14/-35
14/-50
14/-50
0
a Ratio of diastereomers was determined by 1H NMR (400 MHz) analysis
of the crude products. b Yield refer to pure materials isolated after
chromatography on SiO2.
was evaluated for its feasibility to access 20 (Table 1). In the
presence of BF3‚OEt2, or TMSOTf, this reaction produced
stereochemically pure tetrahydrofuran 21 as sole product; while
20 or a mixture of 20/21 was obtained when AlCl3, TiCl4, SnCl4,
or SbCl5 was used to catalyze the reaction. The formation of
tetrahydrofuran 21 exists as the major competing reaction
pathway with elimination of the silyl group to form homoallylic
alcohol 19 after initial condensation with the carbonyl carbon.
In the case examined, the allylic azide group of 19 underwent
a stereoselective 1,3-allylic azide rearrangement to give the
desired 1,3-azido alcohol 20 bearing an R,â-unsaturated ester.
For this reaction, the stereochemistry of the Me-bearing center
of 20 is determined by the configuration of the C-Si bond of
8 and is rationalized through an anti SE′ open TS model. The
relative stereochemistry of tetrahydrofuran 21 was assigned by
1H NMR experiments, 2D COSY, and difference NOE mea-
surements and by analogy with closely related structures
obtained from earlier experiments.27 The absolute stereochem-
istry is consistent with the stepwise mechanism proposed to
rationalize the previously reported [3 + 2]- and [4 + 2]-annu-
lations resulting in the production of tetrahydrofurans,27 cyclo-
pentanes,28 ∆2-isoxazoline,29 and dihydropyrans.30 As illustrated
in Scheme 7, the initial C-C bond construction occurs by an
anti-SE′-mode of addition and the emerging â-silyl carbocation
is stabilized through the σ f π conjugation of the adjacent C-Si
bond. A 1,2-cationic silyl migration then proceeds through a
bridged carbocation, followed by heterocyclization producing
the tetrahydrofuran 21. A synclinal or antiperiplanar transition
With the first synthetic plan established, the Lewis acid
promoted condensation reaction of silane 8 with (s)-trioxane
(27) (a) Panek, J. S.; Yang, M. J. Am. Chem. Soc. 1991, 26, 9868-9870. (b)
Beresis, R. T.; Panek, J. S. Bioorg. Med. Chem. Lett. 1993, 1 (3), 1609-
1613.
(28) Panek, J. S.; Jain, N. F. J. Org. Chem. 1993, 58, 2345-2348.
(29) Panek, J. S.; Beresis, R. T. J. Am. Chem. Soc. 1993, 115, 7898-7899.
(30) (a) Huang, H.; Panek, J. S. J. Am. Chem. Soc. 2000, 122, 9836-9837. (b)
Huang, H.; Panek, J. S. Org. Lett. 2001, 3, 1693-1696.
(23) Hayashi, T.; Konishi, M.; ito, H.; Kumada, M. J. Am. Chem. Soc. 1982,
104, 4962-4963.
(24) Corey, E. J.; Fuchs, P. L. Tetrahedron Lett. 1972, 3769-3772.
(25) Gagneux, A.; Winstein, S.; Young, W. G. J. Am. Chem. Soc. 1960, 82,
5956-5957.
(26) Panek, J. S.; Yang, M.; Muler, I. J. Org. Chem. 1992, 57, 4063-4064.
9
J. AM. CHEM. SOC. VOL. 124, NO. 38, 2002 11371

A R T I C L E S
Hu and Panek
Scheme 8. Preparation of the C1-C5 Subunit
state can be invoked for the furan formation, but the proximity
of the oxygen functionality to the silylcarbocation prior to
cyclization in the synclinal orientation lends support to the
former.
tected amine 23 in 84% yield over two steps (Scheme 8).
Oxidative cleavage (O3/Me2S) of the trans double bond of 23
gave the corresponding unstable aldehyde, which was im-
mediately transformed into vinyl iodide 6 using Takai’s CrCl2-
mediated homologation procedure33 to afford exclusively the
(E)-alkene in 74% yield (two steps), completing the preparation
of the C1-C5 subunit.
Hydrozirconation of Internal Alkyne 5. To establish
palladium(0)-mediated cross-coupling to assemble the trisub-
stituted (E,E)-diene of the Adda fragment, the unsymmetrical
acetylene 5 was to be converted to a terminal (E)-vinylmetal
species. Literature precedent regarding hydrometalation of
substrates similar to 5 suggested that a hydroboration34 or
hydrozirconation35 might be the best options for this cross-
coupling reaction. Our initial efforts were directed on the
hydroboration of alkyne 5 using conventional hydroborating
reagents such as 9-BBN,36 dibromoborane dimethyl sulfide (Br2-
BH‚SMe2),37 and catecholborane:38 all gave high conversion of
the starting alkyne; however, poor regioselectivity in the range
of a 2-4:1 ratio was not acceptable. Accordingly, we began to
explore conditions for regioselective hydrozirconation of un-
symmetrical acetylene 5, with the anticipation that the resulting
Although the furan was produced in 91% yield in the presence
of BF3‚OEt2 the desired 1,3-azido alcohol 20 could only be
isolated in yields ranging from 30 to 56%. We then explored
conditions for transforming the tetrahydrofuran 21 to the key
1,3-azido alcohol 20. After considerable effort,31 we had learned
that clean tetrahydrofuran ring opening and subsequent suprafa-
cial allylic azide isomerization could be effected by treatment
of 21 with SbCl5 (1.3 equiv, -50 °C, 14 h) in dilute CH2Cl2
(0.05 M). This action resulted in formation of alcohol 20 as a
single diastereomer with an isolated yield of 96%. A proposed
mechanism for the SbCl5 promoted ring opening may involve
coordination of the Lewis acid to the furan oxygen facilitating
an E2-like elimination process followed by azide isomerization.
This one-pot process delivered the thermodynamically more
stable R,â-unsaturated (E)-olefin 20 as the only observed product
(Scheme 7). Several attempts to convert 20 back to 19 by
treatment with a range of conventional Lewis acids were
unsuccessful.
With the establishment of an anti-relationship between the
methyl and azide groups, the transformation of the primary
alcohol to the 1,3-azido alcohol and eventually to a vinyl iodide
for coupling with the C6-C10 subunit 5 was addressed.
Accordingly, the primary hydroxyl was first protected using
TBSCl/imidazole to give TBS ether 22 in nearly quantitative
yield. The azide functional group of 22 was reduced with SnCl2
in anhydrous methanol (0 °C f room temperature, 4 h), and
the resulting primary amine was then acylated with (Boc)2O in
dioxane-aqueous sodium bicarbonate32 to give the Boc pro-
(32) Evans, D. A.; Britton, T. C.; Ellman, J. A.; Dorow, R. L. J. Am. Chem.
Soc. 1990, 112, 4011-4030.
(33) Takai, K.; Nitta, K.; Utimoto, K. J. Am. Chem. Soc. 1986, 108, 7408-
7410.
(34) (a) Brown, H. C.; Campbell, J. B. Aldrichim. Acta 1981, 14, 3-11. (b)
Pelter, A.; Smith, K.; Brown, H. C. Borane Reagent; Academic Press:
London, 1988; p 98. (c) Smith, K.; Pelter, A. In ComprehensiVe Organic
Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon Press: Oxford, U.K.,
1991; Vol. 8, p 703.
(35) For recent reviews on hydrozirconation, see: (a) Labinger, J. A. In
ComprehensiVe Organic Synthesis: Hydrozirconation of CdC and CtC
and Hydrometalation by Other Metals; Trost, B. M., Fleming, I., Eds.;
Pergamon: Oxford, U.K., 1991; Vol. 8; p 667-702. (b) Wipf, P.; Jahn,
H. Tetrahedron 1996, 52, 12853-12910.
(36) Brown, H. C.; Scouten, C. G.; Liotta, R. J. Am. Chem. Soc. 1979, 101,
96-99.
(37) Brown, H. C.; Campbell, J. B. J. Org. Chem. 1980, 45, 389-395.
(38) (a) Brown, H. C.; Gupta, S. K. J. Am. Chem. Soc. 1975, 97, 5249-5255.
(b) Brown, H. C.; Gupta, S. K. J. Am. Chem. Soc. 1972, 94, 4370-4371.
(31) Furan ring opening proved unsuccessful with Lewis acids such as TiCl4,
ZnCl2, MgBr2, and SnCl4 at various reaction temperatures. After the BF3‚
OEt2 promoted furan reaction went to completion, attempts to effect ring
opening by reaction with SbCl5 failed. Reaction under basic conditions
with TBAF/THF was also carried out, but no reaction ensued.
9
11372 J. AM. CHEM. SOC. VOL. 124, NO. 38, 2002

Protein Phosphatase Inhibitor (−)-Motuporin
A R T I C L E S
Table 2. Regioselective Hydrozirconation of Unsymmetrical Internal Alkyne 5
c
entry
solvent
Cp2ZrHCl (equiv)a
temp (°C)
45
45
45
time (h)
conversionb
A/B (%)
1
2
3
4
5
6
7
8
9
toluene
toluene
benzene
benzene
CH2Cl2
CH2Cl2
(CH2)2Cl2
THF
1.2
2.0
1.5
2.0
2.0
2.0
2.0
1.3
2.0
2.0
2.0
3.0
0.3
4.0
5.0
3.0
2.0
0.5
0.4
1.0
92
100
90
100
100
100
100
87
63:37
92:8
75:25
100:0
85:15
91:9
87:13
63:37
87:13
100:0
50
room temp
40
60
50
50
50
THF
THF
100
100
10
1
a For the preparation of Schwartz’s reagent, see ref 40b. b Percent conversion was measured by H NMR analysis of the crude reaction mixtures and is
1
based on remaining alkyne. c Regioselectivity (A/B ratio) was determined by H NMR analysis.
Scheme 9. Completion of the N-Boc Adda Synthesis
Scheme 10. Preparation of l-Val-â-Me-Asp Dipeptide 10
vinylmetal species (or its derivatives) could participate in Pd(0)-
catalyzed cross-coupling reaction with the C1-C5 vinyl iodide
subunit.39
sion of alkyne 5 to the desired vinyl zirconium product A could
be achieved when the reaction was conducted in THF at 50 °C
(entry 10). This crucial result would allow one to execute the
hydrozirconation and subsequent cross-coupling reaction in
THF as a one-pot reaction sequence thereby simplifying the
operational aspects. Our experiments have shown that the
hydrozirconation of alkyne 5 was governed by the amount of
Schwartz’s reagent employed. With 1 equiv, the kinetic product
was obtained as a mixture of regioisomers A and B, and with
an excess of reagent the thermodynamic product A was obtained.
This may be rationalized by the reversible addition of a second
equivalent of reagent to the initially formed vinylzirconium
adduct.39 Having achieved complete conversion and regiose-
lectivity in the hydrozirconation of 5, the vinylzirconium
intermediate was used for the subsequent cross-coupling reaction
with the vinyl iodide fragment to access the diene.
Completion of the N-Boc Adda Synthesis (3). The key sp2-
sp2 bond construction was carried out using a modified Negishi
type coupling and involved the reaction of a vinylzinc inter-
mediate with branched (E)- and (Z)-vinyl halides, catalyzed with
Pd(PPh3)4. The feasibility of this method was tested in our
stereoselective synthesis of Adda (Scheme 9).42 The coupling
process was initiated with the regioselective formation of the
(E)-vinylzinc intermediate. Alkyne 5 was hydrozirconated using
Schwartz’s reagent (2.0 equiv of Cp2Zr(H)Cl, THF, 50 °C, 1.0
In 1975 Schwartz documented the syn-addition of zirconocene
hydrochloride (Cp2Zr(H)Cl, Schwartz’s reagent)40 to terminal
or internal alkynes followed by treatment with electrophiles
provided trans-alkenes with a high selectivity (>98%).41
Although useful levels of regio- and stereoselectivity can be
achieved on unfunctionalized alkynes, the hydrozirconation of
branched internal alkynes reveals a more complex picture for
regioselective product formation and is highly underdeveloped.34
Experimental results aimed at optimizing the regioselectivity
of the hydrozirconation with R,â-branched alkyne 5 are sum-
marized in Table 2.
In agreement with literature precedent,35 our experiments have
also shown that excess Schwartz’s reagent is necessary to
achieve useful selectivity in the generation of the vinylmetal
species. Using conventional solvents for hydrozirconation such
as benzene or dichloromethane, high selectivity could be
obtained (entries 4 and 6, Table 2); however, neither is suitable
for the subsequent cross-coupling reaction. Quantitative conver-
(39) Negish, E.; Okukado, N.; King, A. O.; Van Horn, D. E.; Spiegel, B. I. J.
Am. Chem. Soc. 1978, 100, 2254-2256.
(40) (a) Hart, D. W.; Schwartz, J. J. Am. Chem. Soc. 1974, 96, 8115-8116.
For preparation of Schwatz’s reagent, see: (b) Buchwald, S. L.; LaMaire,
S. J.; Nielsen, R. B.; Watson, B. T.; King, S. M. Org. Synth. 1992, 71,
77-80. (c) Buchwald, S. L.; La Maire, S. J.; Nielsen, R. B.; Watson, B.
T.; King, S. M. Tetrahedron Lett. 1987, 28, 3895-3898.
(41) Hart, D. W.; Blackburn, T. F.; Schwartz, J. J. Am. Chem. Soc. 1975, 97,
679-680.
9
J. AM. CHEM. SOC. VOL. 124, NO. 38, 2002 11373

A R T I C L E S
Hu and Panek
Scheme 11. Preparation of Tetrapeptide 4
Scheme 12. Fragment Coupling and Macrocyclization
h) to produce the (E)-zirconate as a single regioisomer. Sub-
sequent transmetalation with anhydrous ZnCl2 (3.0 equiv, room
temperature, 2.0 min) afforded the vinylzinc species 24, which
was used directly in the coupling with (E)-vinyl iodide 6 in the
presence of Pd(Ph3P)4 (0.05 equiv, THF, room temperature, 5
min), completing the assembly of the configurationally pure
(E,E)-diene (84% yield). This one-pot sequence gave the fully
functionalized precursor to Adda 25, which was converted to
N-Boc Adda 3. Treatment of a solution of the silyl ether 25 in
THF with nBu4NF (1.0 equiv, THF, room temperature, 30 min)
resulted in clean conversion to the primary alcohol, which was
oxidized to the corresponding carboxylic acid by treatment with
PDC (7.0 equiv, DMF, 22 h, 86%), completing the synthesis
of N-Boc Adda 3. The spectroscopic and physical properties of
this material were identical in all respects (1H, 13C, IR, [R]D,
and HRMS) with those previously reported.21
Synthesis of Tetrapeptide Fragment 4. Preparation of
dipeptide 10 began with the modification of 1,3-amino alcohol
20, as illustrated in Scheme 10. This intermediate was oxidized
with Jones' reagent (1.9 M, acetone/H2O, 0 °C) to provide the
â-azido acid 13 in 88% yield. It was coupled directly to
compound 14a, the Boc-deprotected TFA salt of amine 1443
(42) Our preliminary results of the hydrozirconation and subsequent cross-
coupling strategy have been published. See: Panek, J. S.; Hu, T. J. Org.
Chem. 1997, 62, 4912-4913.
(43) Compound 14 was prepare by DCC mediated coupling between comme-
cially available N-Boc-(L)-Val and trimethylsilyl ethanol in the presence
of catalytic amount of DMAP.
9
11374 J. AM. CHEM. SOC. VOL. 124, NO. 38, 2002

Protein Phosphatase Inhibitor (−)-Motuporin
A R T I C L E S
Scheme 13. Revised Retosynthetic Analysis of Motuporin
(1.1 equiv) by treatment with excess DIPEA (3.0 equiv) and
BOP reagent (1.1 equiv) to afford dipeptide 26 in 89% overall
yield. The trans double bond in 26 was oxidized by using a
catalytic amount of RuCl3 with NaIO4 (4.0 equiv) as cooxidant,44
and the resulting crude carboxylic acid was immediately
converted to its methyl ester by treatment with CH2N2, affording
10 in 78% yield over two steps.
Selective N-methylation of commercially available N-Boc-
O-benzyl-D-threonine 12, under conditions developed by Benoi-
ton,45 gave N-methylated product 27 in 85% yield (Scheme 11).
The carboxylic acid 27 was protected as its benzyl ester 28,
followed by removal of the Boc group to generate the TFA salt
of amine 29. This residue was used directly in peptide coupling
with carboxylic acid 30, which was prepared in two steps from
N-Boc-D-Glu R-methyl ester 11. The acylation of N-terminal
N-methyl amino acids is generally a sluggish process,46 and in
the present case, the reaction rate was enhanced by the fact that
the acylating carbonyl of 30 is unhindered. The coupling
between 29 and 30 was carried out under standard peptide
coupling conditions (1.1 equiv of 30, 1.0 equiv of BOP, 3.0
equiv of DIPEA) afforded 9 in 83% yield over two steps.
Coupling of the dipeptides 9 and 10 was carried out using
Carpino’s racemization-supressing reagent HATU.47 Accord-
ingly, dipeptides 9 and 10 (1:1 mole ratio) were dissolved in
EtOAc and exposed to an atmosphere of H2 in the presence of
catalytic amounts of Pd/C effected simultaneous debenzylation
of compound 9 and reduction of the azide group of 10 (Scheme
11). The resulting reaction intermediates were then subjected
to peptide coupling conditions as described by Carpino (1.3
equiv of HATU, 3.0 equiv of collidine, DMF) for 15 h at room
temperature. This reaction afforded the fully protected tetrapep-
tide fragment 4 as a 2.5:1 mixture of diastereomers as
determined by 1H NMR analysis of the crude reaction mixtures.
Presumably, epimerization at the threonyl R-center (C32) is
responsible for the resulting mixture.46 After separation from
the minor isomer, the desired tetrapeptide 4 was obtained in
55% yield.48 This one-pot deprotection-reduction-coupling
sequence completed the synthesis of fragment 4 and produced
the material to be used in subsequent fragment coupling with
Boc-Adda fragment 3.
Scheme 14. Preparation of Trisubstituted (E)-Vinyl Iodide 36
slow addition to a CH2Cl2 solution of HATU/DIPEA. Unfor-
tunately, only 8-12% of the desired macrocycle 33 was
obtained under these cyclization conditions, as well as extensive
epimerization at the valine R-carbon was observed. When the
cyclization was performed in the presence of pentafluorophenyl
diphenylphosphinate (FDPP),49 a slightly higher yield of the
macrocycle was achieved. However, due to epimerization and
the slow rate of cyclization, we were unable to obtain useful
amounts of the macrocycle for transformation to motuporin. At
this point several concerning issues had arisen that caused us
to abandon the present route. [At this point we chose to modify
our route by selecting an alternative cyclization site along with
minor alterations in the protecting group strategy.]
Revised Retrosynthetic Analysis. In the revised synthetic
plan, we anticipated that the macrocyclization could be carried
out at C26 that contains a carboxylate group and also lacks a
deactivating R-amide substituent. In addition, fragment coupling
would still be conducted at C1-C39 amide bond (Scheme 13).
In our original plan, the deprotection step for the trimethyl-
silylethyl ester (TMSE) had become unmanageable, we then
chose to protect the C26 carboxylic acid as a trichloroethyl (Tce)
ester whose removal would be carried out under mild reductive
Fragment Coupling and Macrocyclization. The N-Boc
Adda intermediate was activated by HATU (1.5 equiv) and
coupled with the Boc-deprotected tetrapeptide of 4 in the
presence of collidine (3.0 equiv), affording the desired pen-
tapeptide 31 in 70% yield with no trace of epimerization
(Scheme 12). Surprisingly, sequential removal of the TMSE
ester of 31 proved troublesome. After surveying a variety of
reaction conditions (TBAF, HF‚Py, TBAF/AcOH, KF/nBu4NCl,
and TAS-F), the most effective reagent was TBAF‚SiO2. Under
these reaction conditions (TBAF‚SiO2 2.0 equiv, 5:1 THF/
DMF), the desired TMSE-deprotected product was obtained in
50-65% yield. After removal of the Boc group, the resulting
crude residue 32 was subjected to the cyclization conditions by
(44) Carlsen, Per H. J.; Katsuki, T.; Martin, V. S.; Sharpless, K. B. J. Org.
Chem. 1981, 46, 3936-3938.
(45) Cheung, S. T.; Benoiton, N. L. Can. J. Chem. 1977, 55, 906-910.
(46) For review on peptide coupling strategy, see: Humphrey, J. M.; Chamberlin,
A. R. Chem. ReV. 1997, 97, 2243-2266.
(47) (a) Carpino, L. A.; El-Faham, A. J. Org. Chem. 1994, 59, 695-698. (b)
Carpino, L. A. J. Am. Chem. Soc. 1993, 115, 4397-4398. (c) Carpino, L.
A.; El-Faham, A. J. Org. Chem. 1995, 60, 3561.
(49) (a) Chen, S.; Xu, J. Tetrahedron Lett. 1991, 32, 6711-6714. (b) Dudash,
(48) Although other peptide coupling conditions were also attempted, similar
levels of racemization were observed in this segment coupling reaction.
J., Jr.; Jiang, J.; Mayer, S. C.; Joullie, M. M. Synth. Commun. 1993, 23,
349.
9
J. AM. CHEM. SOC. VOL. 124, NO. 38, 2002 11375

A R T I C L E S
Hu and Panek
Scheme 16. Synthesis of the N-Boc-Valine-Adda Fragment
Scheme 15. Preparation of (E)-Vinylstannane for Stille Coupling
Scheme 17. Preparation of Tripeptide 35
Table 3. Summary of the Stille Coupling Reaction between 36
and 42
a
catalyst
solvent/temp (°C)
base
additives
E/Z yieldb (%)
Pd(PPh3)4
Pd(MeCN)2Cl2 THF/room temp
DMF/50
1:1
2:1
2:1
2:1
3:2
2:1
4:1
22
39
45
30
46
35
50
55
CdCl2
ZnCl2
Pd(PPh3)4
Pd(PPh3)Cl2
Pd2(dba)3
Pd2(dba)3
Pd(PPh3)4
Pd(PPh3)4
THF/50
DMF/room temp DIPEA
THF/50 DIPEA AsPh3
DMF/room temp DIPEA TFP
DMF/room temp DIPEA CuI/CsF
NMP/room temp DIPEA CuI (0.1 equiv) 4:1
a Ratio was determined by 1H NMR (400 MHz) analysis of the crude
products. b Yield refer to combined yields of E- and Z-isomers isolated after
chromatography on SiO2.
alkyne 5, followed by iodination of the alkenyl zirconium
intermediate (Scheme 14).
The primary hydroxyl of 20 was protected as TBDPS ether,
and the azide reduced using SnCl2 (1.5 equiv) in anhydrous
methanol (0 °C f room temperature, 4 h) to provide primary
amine 37, which was condensed with pentafluorophenyl ester
activated N-Boc L-valine 38 in dioxane/aqueous NaHCO3 using
a biphasic reaction system to afford dipeptide 39 in 84% yield
over three steps (Scheme 14). Oxidative cleavage of the olefin
gave the aldehyde 40, which was immediately subjected to
Takai’s homologation protocol32 to afford the pure (E)-vinyl
iodide 41 (74%, Scheme 15). This material was transformed to
the corresponding stannane by reaction with bis(trimethyltin)
in the presence of catalytic Pd(PPh3)4, cleanly generating the
(E)-vinylstannane 42 in 90% yield, ready for the subsequent
Stille coupling.50
conditions using activated zinc dust. Literature precedence has
documented that similar conditions have been routinely used
in structurally complex peptide syntheses. By dividing motu-
porin into two advanced intermediates, the fragment 34 and the
tripeptide fragment 35, the level of convergence is maintained.
As illustrated in Scheme 13, the fragment 34 could, in principle,
be assembled through two different sets of coupling condi-
tions: the conventional Stille coupling or a modified Negishi
coupling.
Stille Coupling Strategy. The construction of the (E,E)-
trisubstituted diene using Stille coupling conditions required the
preparation of a trisubstituted (E)-vinyl iodide and a (E)-
vinylstannane coupling partner. The (E)-vinyl iodide 36 was
readily prepared through regioselective hydrozirconation of
The Stille coupling between 36 and 42 was evaluated, and
those results are summarized in Table 3. Despite extensive
9
11376 J. AM. CHEM. SOC. VOL. 124, NO. 38, 2002

Protein Phosphatase Inhibitor (−)-Motuporin
A R T I C L E S
Scheme 18. Completion of the Synthesis of (-)-Motuporin
efforts to optimize this reaction, it remained plagued by poor
double bond ratios at C4-C5 (1-4:1, E/Z), protodestannylation,
and homocoupling byproducts.
of Adda. When the identical reaction conditions were applied
to the cross-coupling between alkyne 7 and dipeptide vinyl
iodide 41, configurationally pure (E,E)-diene 43 was produced
in 81% yield (Scheme 16). The overall efficiency of the Negishi
coupling process may be attributed to the enhanced electrophi-
licity and accessibility of the vinyl iodide in combination with
good nucleophilicity of the vinyl zinc species (facile transmeta-
lation). The success of this cross-coupling relied on the use of
a double transmetalation process (Zr f Zn f Pd) of low kinetic
barrier that replaced a single transmetalation process (Zr f Pd)
of high activation energy, which leads to an overall rate
enhancement.38
With the cross-coupling chemistry established as an efficient
approach to the diene, the final stage of the valine-Adda
synthesis was carried out. Deprotection of the silyl group using
TBAF‚SiO2 (2.5 equiv, room temperature, 4 h) afforded the
primary alcohol 44 in 95% yield. With the presence of
stoichiometric amount of H2O this material was directly oxidized
to carboxylic acid 34 using Ley’s oxidation protocol (TPAP/
NMO, CH3CN, 1 h; then H2O, room temperature, 18 h),
completing the synthesis of valine-Adda dipeptide.55
The addition of a noncoordinating base, diisopropylethyl-
amine (DIPEA), markedly reduced the amount of protodestan-
nylation.51 However, we were unable to prevent double bond
isomerization and the formation of products from homocoupling
that might be attributed to the slow transmetalation of the
vinylstannane to the Pd(II) intermediate. Working under the
assumption that the transmetalation was the rate-determining
step in the catalytic cycle, the addition of facilitators, such as
AsPh3,52 CdCl2,53 or ZnCl238 were also explored; however, only
slight improvements were achieved.
Best results for this Stille coupling were achieved using
Liebeskind’s procedure.54 In the presence of a catalytic amount
of CuI and Pd(PPh3)4, using NMP (N-methylpyrrolidinone) as
the solvent, this Stille coupling went to completion in 30 min
at room temperature, providing the coupling product 43 in 55%
yield with a 4:1 E/Z ratio at the C4-C5 double bond. Reaching
only moderate yield and poor stereoselectivity in the olefination,
we then focused on the modified Negishi coupling strategy.
Modified Negishi Coupling Strategy. The establishment of
an efficient method for the formation of the trisubstituted (E,E)-
diene of N-Boc Adda 2 using a one-pot sp2-sp2 Pd(0)-catalyzed
coupling has provided the basis for a highly convergent synthesis
Preparation of Tripeptide 35. The preparation of the
tripeptide fragment once again began with the oxidation of 1,3-
azido alcohol 20 followed by protection of the carboxylic acid
as a trichloroethyl ester gave 45 in 84% yield in two steps.
Olefin oxidation (catalyst RuCl3‚nH2O, NaIO4) of 45, followed
by exposure of the resulting R-azido acid to diazomethane (Et2O,
0 °C) afforded methyl ester 47 in 86% yield (two steps) as
(50) For reviews on Stille coupling reaction, see: (a) Stille, J. K. Angew. Chem.,
Int. Ed. Engl. 1986, 25, 508-524. (b) Farina, V.; Krishnamurthy, V.; Scott,
W. J. Org. React. 1998, 50, 1-652.
(51) Evans, D. A.; Gage, J. R.; leighton, J. L. J. Am. Chem. Soc. 1992, 114,
9434-9453.
(52) Farina, V.; Krishnan, B. J. Am. Chem. Soc. 1991, 113, 9585-9595.
(53) Evans, D. A.; Black, W. C. J. Am. Chem. Soc. 1993, 115, 4497-4513.
(54) Leibeskind, L. S.; Fengl, R. W. J. Org. Chem. 1990, 55, 5359-5364.
(55) Decreased reaction times resulted in isolation of mixtures of aldehyde and
desired carboxylic acid. For a review, see: Ley, S.; Norman, J.; Griffith,
W.; Marsden, S. Synthesis 1994, 639-666.
9
J. AM. CHEM. SOC. VOL. 124, NO. 38, 2002 11377

A R T I C L E S
Hu and Panek
described in Scheme 17. The azide group could be selectively
reduced with SnCl2 (1.5 equiv) to provide the free amine,
without effecting the trichloroethyl ester. The crude amine was
used without purification and coupled with N-Boc(Me)-D-
threonine 4656 (1.1 equiv) under standard conditions to give
dipeptide 49 in 89% yield. After the removal of Boc protecting
group, the resulting TFA salt of 49 underwent BOP induced
amide bond formation with 11, completing the formation of
tripeptide 35 in 83% yield.
Fragment Coupling and Completion of the Synthesis. With
the tripeptide 35 and N-Boc-valine-Adda fragments 34 in our
possession, the crucial fragment coupling reaction was ef-
fectively carried out using Carpino’s protocol.46 First, the Boc
protecting group was removed with TFA and the resulting salt
was washed with aqueous Na2CO3 to give free amine 50.
Without purification, the crude amine was treated with 34 in
the presence of HATU/collidine in DMF to provide the protected
pentapeptide 51 in 93% yield (Scheme 18). In contrast to the
lower yields and epimerization problems with the BOP reagent,
reaction using HOBt/EDCI, or the pentaflurophenol activated
ester proceeded without any detectable amounts of epimeriza-
tion.57
In the final stage of the synthesis, removal of the trichloroethyl
group of 51 was accomplished under reductive conditions using
zinc dust/acetic acid (room temperature, 4 h) which gave the
carboxylic acid 52 in 95% yield. Removal of the N-Boc
protecting group of 52 was carried out using TFA/CH2Cl2 (0
°C, 30 min) to afford the amine as its TFA salt, which was set
for macrocyclization. This intermediate underwent an efficient
cyclization in the presence of HATU and N-ethylmorpholine,
yielding macrocycle 33 in 79% yield.58 Treatment of 33 with
barium hydroxide (2 N Ba(OH)2, H2O/MeOH, 10:1) resulted
in simultaneous hydrolysis of both methyl esters and a complete
in situ dehydration of N-methylthreonine.15a The resulting (Z)-
olefin geometry is thermodynamically controlled, which was
previously observed by Schreiber and Valentekovich (Scheme
18).15a Upon acidification with 1 N HCl to pH 2 and reverse-
phase HPLC purification, synthetic motuporin (1) was obtained
as its free dicarboxylic acid form in 52% yield. The spectro-
scopic and analytical properties of this material were identical
in all respects with the reported data.59 Treatment of 1 with
NaHCO3 /MeOH followed by gel filtration using Sephadex LH-
20 provided the disodium salt of motuporin, which agreed in
all respects (1H and 13C NMR, [R]D, IR, FAB-HRMS, rpHPLC)
with the data reported earlier.
Conclusion
We have described a highly convergent, enantioselective
synthesis of the (-)-motuporin based on the use of chiral or-
ganosilane reagents. Six of the eight stereogenic centers were
established utilizing asymmetric crotylsilane bond construction
methodology. An efficient Pd(0)-catalyzed cross-coupling reac-
tion for the construction of the trisubstituted (E,E)-diene system
of the (-)-motuporin side chain is also featured in this synthesis.
The route was completed in a total of 31 steps with the longest
linear sequence of 16 steps (from chiral crotylsilane reagents)
commencing with an overall yield of 15.8%.
Acknowledgment. Financial support was obtained from NIH/
NCI (Grant RO1 CA56304).
Supporting Information Available: Complete experimental
procedures and spectral data for all new compounds (PDF). This
material is available free of charge via the Internet at
http://pubs.acs.org.
(56) This compound was prepared through hydrogenolysis of substrate 27 using
H2/Pd-C/EtOH at room temperature for 15 h.
JA0206700
(57) Satisfactory spectroscopic and analytic data (1H and 13C NMR, IR, [R]D,
MS, HRMS) were obtained for all new compounds. Ratios of diastereomers
were determined by 1H NMR (400 MHz) analysis on crude reaction
mixtures.
(58) Hale, K. J.; Cai, J.; Williams, G. Synlett 1998, 149-152.
(59) Robert J. Valentekovich, Ph.D. Thesis, Harvard University, June 1995.
9
11378 J. AM. CHEM. SOC. VOL. 124, NO. 38, 2002