Total synthesis of (-)-dysiherbaine [4]

Full text of DOI:10.1021/ja000817s

5216
J. Am. Chem. Soc. 2000, 122, 5216-5217
Scheme 1
Total Synthesis of (-)-Dysiherbaine
Hidekazu Masaki, Junji Maeyama, Kazuko Kamada,
Tomoyuki Esumi, Yoshiharu Iwabuchi, and
Susumi Hatakeyama*
Faculty of Pharmaceutical Sciences
Nagasaki UniVersity, Nagasaki 852-8521, Japan
ReceiVed March 7, 2000
(-)-Dysiherbaine (1), a potent neuroexcitotoxin, was isolated
by Sakai and co-workers from a Micronesian sponge Dysidea
herbacea and found to be a selective agonist of non-NMDA (N-
methyl-D-aspartate) type glutamate receptors in the central nervous
system.1 Structurally, this amino acid is characterized by a novel
cis-fused hexahydrofuro[3,2-b]pyran ring system containing a
glutamic acid appendage. The relative configuration was deter-
mined by detailed NMR analysis;1 however, the absolute con-
figuration was not elucidated. Its low availability from natural
sources, combined with the entirely novel molecular architecture
and the potent neuroexcitatory activity, makes 1 an attractive target
for synthetic studies.2 Its potential as a lead for developing
selective agonists or antagonists of glutamate receptors makes
the availability of analogues also an important goal.3 We now
report an enantioselective total synthesis4 of (-)-dysiherbaine (1)
in naturally occurring form.
Scheme 2a
We envisaged tetra-substituted pyran 2 as a precursor of 1 and
postulated that this intermediate could be accessed via palladium-
catalyzed cross-coupling reaction of organozinc reagent 3 and
vinyl triflate 4, accessible from tricyclic lactone 5, based on
Jackson’s protocol5 (Scheme 1). This coupling process is par-
ticularly challenging since Jackson’s methodology has not been
successfully applied to highly functionalized vinyl triflates such
as 4.6 In this synthetic plan, another key issue which must be
addressed is the efficient enantioselective construction of tricyclic
lactone 5 having four contiguous cis stereogenic centers.
Preparation of key compound 5 began with conversion of
σ-symmetrical divinylcarbinol 6 to epoxy alcohol 7 using
methodology we have previously developed7 (Scheme 2). Thus,
Katsuki-Sharpless catalytic asymmetric epoxidation8,9 of 6,
a Reagents: (a) (i) D-DIPT (9 mol %), Ti(O-i-Pr)4 (7 mol %), t-BuOOH
(2 equiv), 4A molecular sieves, CH2Cl2, -25 °C; (ii) DIPAD, Ph3P,
p-(NO2)C6H4CO2H, toluene, -25 °C; (iii) K2CO3, MeOH; (b) (i)
PhCONCO, THF; (ii) K2CO3, n-Bu4NCl (0.2 equiv), MeCN, 0 °C; (c)
NaOMe, MeOH; (d) (i) TBSCl, imidazole, DMF; (ii) NaH, MeI, DMF;
(e) (i) BCl3, CH2Cl2, -60 °C; (ii) TBSCl, imidazole, DMF; (f) (i) O3,
CH2Cl2-MeOH, -78 °C then Me2S; (ii) (PhO)2P(O)CH2CO2Et, NaH,
THF, -78 to 0 °C; (g) 47% HF, MeCN, 70 °C then NaHCO3.
* To whom correspondence should be addressed. Telephone: +81-95-
847-1111 (ext. 2520). Fax: +81-95-848-4286. E-mail: susumi@
net.nagasaki-u.ac.jp.
followed by inversion10 of the hydroxy group gave enantiomeri-
cally pure 7 in 64% yield. Reaction of 7 with benzoyl isocyanate,11
followed by treatment of the resulting N-benzoyl carbamate with
K2CO3 allowed stereoselective cyclization accompanied by migra-
tion of the N-benzoyl group to afford cyclic carbamate 8 in 87%
yield. Treatment of 8 with NaOMe in MeOH led to methanolysis
of the benzoate and concomitant migration of the cyclic carbamate
protecting group to give a 1:1 equilibrium mixture of 9 and 10,
quantitatively. These compounds were separated by silica gel
column chromatography and 9 was treated with NaOMe in MeOH
to convert it to the above-mentioned equilibrium mixture. As a
result of this sequence, 10 was obtained in 75% yield from 8. It
is important to note that this sequence could be performed easily
on large scale to provide multigram quantities of 10. Silylation
of 10 with tert-butyldimethylsilyl chloride and N-methylation of
the product generated 11 in 80% yield. Removal of the benzyl
(1) Sakai, R.; Kamiya, H.; Murata, M.; Shimamoto, K. J. Am. Chem. Soc.
1997, 119, 4112-4116.
(2) For the synthesis and biological activity of related model compounds,
see: Sasaki, M.; Maruyama, T.; Sakai, R.; Tachibana, K. Tetrahedron Lett.
1999, 40, 3195-3198.
(3) For reviews on glutamate receptors, see: (a) Ozawa, S.; Kamiya, H.;
Tsuzuki, K. Prog. Neurobiol. 1998, 54, 581-618. (b) Chittajallu, R.;
Braithwaite, S. P.; Clarke, V. R. J.; Henley, J. M. Trends Pharmacol. Sci.
1999, 20, 26-35.
(4) (a) Masaki, H.; Maeyama, J.; Kamada, K.; Iwabuchi, Y.; Hatakeyama,
S. In 41st Symposium on the Chemistry of Natural Products, Symposium
Papers; Nagoya; 1999; pp 13-18. (b) Snider, B. B.; Hawryluk, N. A. Org.
Lett. 2000, 2, 635-638.
(5) (a) Jackson, R. F. W.; James, K.; Wythes, M. J.; Wood, A. J. Chem.
Soc., Chem. Commun. 1989, 644-645. (b) Jackson, R. F. W.; Wythes, M. J.;
Wood, A. Tetrahedron Lett. 1989, 30, 5941-5944. (c) Jackson, R. F. W.;
Wishart, N.; Wood, A.; James, K.; Wythes, M. J. J. Org. Chem. 1992, 57,
3397-3404. (d) Dunn, M. J.; Jackson, R. F. W.; Pietruszka, J.; Turner, D. J.
Org. Chem. 1995, 60, 2210-2215.
(6) For syntheses of amino acids using Jackson’s methodology, see: (a)
Ye, B.; Burke, T. R., Jr. J. Org. Chem. 1995, 60, 2640-2641. (b) Malan, C.;
Morin, C. Synlett 1996, 167-168. (c) Matsubara, J.; Otsubo, K.; Ohtani, S.;
Kawano, Y.; Uchida, M. Heterocycles 1996, 43, 133-140. (d) Torisawa, Y.;
Soe, T.; Katoh, C.; Motohashi, Y.; Nishida, A.; Hino, T.; Nakagawa, M.
Heterocycles 1998, 47, 655-659.
(7) (a) Hatakeyama, S.; Satoh, K.; Takano, S. Tetrahedron Lett. 1993, 34,
7425-7428. (b) Hatakeyama, S.; Kojima, K.; Fukuyama, H.; Irie, H. Chem.
Lett. 1995, 763-764.
(8) Gao, Y.; Hanson, R. M.; Klunder, J. M.; Ko, S. Y.; Masamune, H.;
Sharpless, K. B. J. Org. Chem. 1987, 109, 5765-5780.
(9) For Katsuki-Sharpless asymmetric epoxidation of σ-symmetrical
divinylcarbinols, see: (a) Hatakeyama, S.; Sakurai, K.; Numata, H.; Ochi,
N.; Takano, S. J. Am. Chem. Soc. 1988, 110, 5201-5203 and references
therein. (b) Ja¨ger, V.; Schro¨ter, D.; Koppenhoefer. Tetrahedron 1991, 47,
2195-2210 and references therein. (c) Smith, D. B.; Wang, Z.; Schreiber, S.
L. Tetrahedron 1990, 46, 4793-4808 and references therein.
(10) Martin, S. F.; Dodge, J. A. Tetrahedron Lett. 1991, 32, 3017-3020.
(11) Hirai, Y.; Watanabe, J.; Nozaki, T.; Yokoyama, H.; Yamaguchi, S. J.
Org. Chem. 1997, 62, 776-777.
10.1021/ja000817s CCC: $19.00 © 2000 American Chemical Society
Published on Web 05/12/2000

Communications to the Editor
J. Am. Chem. Soc., Vol. 122, No. 21, 2000 5217
Scheme 3a
With the advanced intermediate 2 in hand, the stage was set
for completion of the total synthesis. Upon desilylation and
epoxidation using mCPBA, 2 gave a 1:1 separable mixture of
4S-epoxide 17 and 4R-epoxide 18 in 70% yield. Treatment of 17
with PPTS (1 equiv) in acetone at room temperature, followed
by methanolysis17 produced 19 in 58% yield. In this particular
case, the cyclization turned out to take place with complete
inversion of the stereochemistry of the quaternary center even
though the reaction was very sluggish. Interestingly, when CSA
was used in place of PPTS in CH2Cl2, the cyclization18 occurred
with poor stereoselectivity to give a 3:2 mixture of 19 and its
C4-epimer in 41% yield after methanolysis. It is important to note
that 19 was also obtained from 4R-epoxide 18 by CSA catalyzed
cyclization although the yield was moderate.19
Oxidation of 19 with TPAP-NMO20 provided lactam 20 in 76%
yield, the stereochemistry of which was confirmed via NOE
experiments by comparison with the C4-epimer of 20 derived
from 18. Finally, upon alkaline hydrolysis and neutralization using
ion-exchange resin (IRC-76), 20 furnished (-)-dysiherbaine (1)
which was purified by reverse-phase HPLC. The synthetic
substance, [R]24D -3.5° (c 0.23, H2O), was identical with natural
dysiherbaine, [R]26 -3.5° (c 0.4, H2O), by spectroscopic (1H
D
a Reagents: (a) (i) MeONHMe‚HCl, Me2AlCl, CH2Cl2; (ii) TESCl,
imidazole, DMF; (b) MeMgBr, THF, 0 °C; (c) LDA, 2-[N,N-bis(trifluo-
romethylsulfonyl)amino]-5-chloropyridine, THF, -78 °C; (d) 3, (Ph3P)4Pd
(10 mol %), LiCl (4 equiv), benzene-DMA-HMPA (20:1:1), 80 °C;
(e) (i) 1 M HCl, THF; (ii) mCPBA, CH2Cl2; (f) (i) PPTS, acetone; (ii)
NaOMe, MeOH; (g) n-Pr4NRuO4 (20 mol %), NMO, CH2Cl2; (h) 41%
NaOH, reflux.
and 13C NMR)21 and chromatographic (reverse-phase TLC and
HPLC) comparisons. 4-Epidysiherbaine was also synthesized from
18 in 45% overall yield in the same manner as that described for
the synthesis of 1 from 17. The radioligand binding assay of
synthetic 1 and 4-epidysiherbaine toward ionotropic glutamate
receptors on rat brain synaptic membranes showed that synthetic
1 inhibited binding of [3H]-kainic acid and [3H]-1-amino-3-
hydroxy-5-methyl-4-isoxazolepropionic acid to the same degree
as that observed for natural dysiherbaine but 4-epidysiherbaine
did not. These results further confirmed the absolute structure of
natural dysiherbaine as depicted in 1.
ethers with BCl3, followed by silylation with tert-butyldimeth-
ylsilyl chloride gave a 4:1 mixture of 12 and 13 in 74% yield.
Without separation, this mixture was subjected to ozonolysis and
Horner-Emmons olefination using ethyl diphenylphosphono-
acetate12 to provide unsaturated ester 14 with complete Z-
selectivity in 90% yield. Exposure of 14 to HF in acetonitrile led
to formation of the corresponding butenolide which, upon
basification with NaHCO3, underwent intramolecular Michael
addition of the primary hydroxy group to produce tricyclic lactone
5 as the sole product in 71% yield. Single-crystal X-ray analysis
as well as NOE experiments secured the stereostructure of 5.
Having achieved construction of the cis-fused hexahydrofuro-
[3,2-b]pyran core, we then proceeded to install the glutamic acid
appendage (Scheme 3). Tricyclic lactone 5 was homologated to
methyl ketone 16 via Weinreb amide 15 in 82% overall yield.13
Deprotonation of 16 with LDA, followed by triflation14 of the
resulting lithium enolate with 2-[N,N-bis(trifluoromethylsulfonyl)-
amino]-5-chloropyridine allowed preferential formation of vinyl
triflate 4 in 79% yield.15 To attach the alanine moiety to 4, the
crucial palladium-catalyzed cross-coupling process was then
investigated according to Jackson’s protocol.5 After considerable
experimentation, we eventually found the conditions where the
desired cross-coupling took place cleanly. Thus, upon reaction
of 4 with 3, prepared in situ from N-methoxycarbonyl-â-
iodoalanine (5 equiv)16 by the action of Zn-Cu couple under
sonication, in the presence of (PPh3)4Pd (10 mol %) and LiCl (4
equiv) in benzene-dimethylacetamide-HMPA (20:1:1) at 80 °C,
coupling product 2 was obtained in 73% yield.
In conclusion, we have achieved a total synthesis of (-)-
dysiherbaine (1) from 6 in enantiocontrolled manner, thereby
rigorously establishing its absolute structure. We demonstrated
Jackson’s cross-coupling methodology to be a powerful tool in
the synthesis of highly functionalized amino acids for the first
time. The synthetic route established herein should enable us to
synthesize a variety of interesting analogues.
Acknowledgment. We thank Professor R. Sakai (Kitasato University)
for kindly suplying natural dysiherbaine and its spectra. We also thank
Dr. K. Shimamoto (Suntory institute for Bioorganic Research) for the
radioligand binding assay and Professor M. Nishizawa (Tokushima Bunri
University) for X-ray crystallographic analysis. Partial financial support
(No. 11170243) from the Ministry of Education, Science, Sports and
Culture of Japan is gratefully acknowledged.
Supporting Information Available: Experimental procedures, spec-
1
tral data, and H and 13C NMR spectra for new compounds, and tables
of crystal data of 5 (PDF). This material is available free of charge via
the Internet at http://pubs.acs.org.
JA000817S
(17) Since the corresponding δ-lactone was partially formed under the
cyclization conditions, methanolysis was carried out to convert it to 19.
(18) This reaction posssibly took place via a SN1-like reaction process.
However, it is also assumed that the cyclization occurred via a double inversion
process involving participation of the carbamate group in competition with a
SN2-like reaction process because treatment of the N-Boc derivative of 17
with CSA in CH2Cl2 resulted in formation of the corresponding six-membered
cyclic carbamate rather than the desired cyclization.
(12) Ando, K. J. Org. Chem. 1997, 62, 1934-1939.
(13) (a) Nahm, S.; Weinreb, S. M. Tetrahedron Lett. 1991, 22, 3815-
3818. (b) Shimizu, T.; Osako, K.; Nakata, T. Tetrahedron Lett. 1997, 38,
2685-2688.
(19) The CSA catalyzed cyclization provided a 2:3 mixture of 19 and its
C4-epimer in 51% yield, and the PPTS catalyzed cyclization gave the C4-
epimer of 19 exclusively in 52% yield after methanolysis.
(14) Comins, D. L.; Dehghani, A. Tetrahedron Lett. 1992, 33, 6299-6302.
(15) A 1:2 E/Z-mixture of the corresponding trisubstituted alkenyl triflates
was also obtained in ∼11% yield.
(20) Griffith, W. P.; Ley, S. V.; Whitcombe, G. P.; White, A. D. J. Chem.
Soc., Chem. Commun. 1987, 1625-1627.
(16) Since methyl 2-(methoxycarbonylamino)acrylate was also formed
under these conditions, the use of 3∼5 equiv of N-methoxycarbonyl-â-
iodoalanine was required for the sufficient production of the organozinc
reagent.
(21) In 1H NMR spectra (D2O), chemical shifts of NCH3 and CH-NMe
varied markedly depending on the concentration. However, the 1H NMR
spectrum (4.5 × 10-2 M DCl) of synthetic dysiherbaine was found to be
completely identical with that of natural specimen.