Total synthesis of kaitocephalin, the first naturally occurring AMPA/KA receptor antagonist [17]

Full text of DOI:10.1021/ja016403w

9706
J. Am. Chem. Soc. 2001, 123, 9706-9707
Scheme 1. Structure of Kaitocephalin and Synthetic Analysis
Total Synthesis of Kaitocephalin, the First Naturally
Occurring AMPA/KA Receptor Antagonist
Dawei Ma* and Jiade Yang
State Key Laboratory of Bioorganic and
Natural Products Chemistry
Shanghai Institute of Organic Chemistry
Chinese Academy of Sciences, 354 Fenglin Lu
Shanghai 200032, China
Scheme 2a
ReceiVed June 13, 2001
ReVised Manuscript ReceiVed August 10, 2001
Excitatory neuronal transmission within the central nervous
system (CNS) is mediated predominantly by L-glutamate, which
plays a role of utmost importance in many physiological processes
such as neural plasticity, memory, and learning.1 However,
excessive of L-glutamate release can result in neuronal cell death,
a phenomenon that has been termed excitotoxcity.2 It is now
commonly accepted that excitotoxic cell death substantially
contributes to the pathophysiology of both acute and chronic
neurodegenerative disorders in the CNS. These disorders include
epilepsy, focal and global ischaemia, stroke, pain, and neuro-
degenerative diseases.3 These observations have stimulated con-
siderable research efforts on the development of selective and
potent antagonists for glutamate receptors, particularly compounds
acting at the N-methyl-D-aspartate (NMDA), R-amino-3-hydroxy-
5-methyl-4-isoxazolepropionate (AMPA), and kainic acid (KA)
receptor subtypes.4 Some AMPA/KA receptor antagonists have
shown promising usage for treatment of epilepsy and cerebro-
vascular ischemia.5
In 1997, kaitocephalin, the first natural AMPA/KA antagonist,
was isolated from Eupenicillium shearii. In the models of chick
primary telencephalic and rat hippocampal neurons, this com-
pound showed protection from kainate toxicity at 500 µM with
EC50 values being 0.68 and 2.4 µM, respectively, and from
AMPA/cyclothiazide (500 µM/50‚M) toxicity with EC50 values
0.6 and 0.4 µM, respectively. Unlike the known AMPA/KA
antagonists with a quinoxalinedione skeleton, kaitocephalin does
not have any cytotoxic effect.6 Thus, SAR studies on this
compound may open a new avenue to the development of
therapeutic tools for protection of excitotoxicity. However, before
comprehensive SAR studies become a reality, an efficient route
to kaitocephalin is required. Toward this goal we report here the
first total synthesis of kaitocephalin.
a Reagents and conditions: (a) i) BnOH, SOCl2, 0 °C to rt, 96%; ii)
(Boc)2O or ClCO2Me, Et3N, DMAP, 86-95%; iii) DIBAL-H, THF, -78
°C, then MeOH, TsOH, rt, 95%; (b) allyltrimethylsilane, TiCl4, -78 °C,
95% for 2a, 77% for 2b; (c) LiHMDS, then (R)-Garner aldehyde, -78
°C, 60% for 3a, 86% for 3b; (d) LiHMDS, -42 °C, 100%; (e) i) TsOH,
MeOH, 85%; ii) TBSCl, Et3N; 97%; (f) i) DMSO, (COCl)2 then Et3N,
95%; ii) NaBH4, MeOH; 90%; (g) i) TBAF, HOAc, 98%, ii) Ac2O, Et3N,
95%; (h) i) K2OsO2(OH)4 1%, (DHQD)2PHAL 5%, K3Fe(CN)6 3 equiv,
K2CO3 3equiv, t-BuOH/H2O (1/1), rt ii) TPSCl, DMAP, Et3N, 92% for
two steps.
On the basis of the proposed stereochemistry of kaitocephalin,7
we described a synthetic plan as shown in Scheme 1. The 2,5-
disubstituted pyrrolidines A were envisioned to be ideal building
blocks for synthesizing the target molecule because its 2-position
could be lithiated and then coupled with (R)-Garner aldehyde to
assemble the right-hand part of kaitocephalin, and its C-C double
bond could be converted into the left-hand moiety through
Sharpless asymmetric dihydroxylation.8 The detailed studies were
outlined in Scheme 2. Protection of the acid and amide groups
of (S)-pyroglutamic acid under ordinary conditions followed by
reduction with DIBAL-H provided 1a and 1b.9 Treatment of 1
with allyltrimethylsilane and TiCl4 in methylene chloride afforded
cis-2 as the major product, together with some separable trans-
isomer.10 The aldol condensation reaction was obviously a key
step for this synthesis and therefore tested under many conditions.
Initially, lithiation of 2a with LDA at -78 °C followed by
trapping the resultant anion with (R)-Garner aldehyde gave the
condensation products quantitatively. However, it was found that
four isomers were formed in this reaction in almost equal amounts.
(1) (a) Bortolotto, Z. A.; Bashir, Z. I.; Davies, C. H.; Collingridge, G. L.
Nature 1994, 386, 740. (b) Kennedy, M. B. Cell, 1989, 59, 777. (c) Muller,
D.; Joly, M.; Lynch, G. Science 1988, 242, 1694.
(2) For reviews, see: (a) Nakanishi, S. Science 1992, 258, 597. (b) Medrum
B. Brain Res. ReV. 1993, 18, 293.
(3) Doble, A. Pharmacol. Ther. 1999, 81, 163.
(4) For reviews, see: (a) Brauner-Osborne, H.; Egebjerg, J.; Nielsen, E.;
Madsen, U.; Krogsgaard-Larsen, P. J. Med. Chem. 2000, 43, 2609. (b)
Kozikowski, A. P. Drug Design for Neuroscience; Raven Press Ltd.: New
York, 1993.
(5) (a) Lee, J. M.; Zipfel, G. J.; Choi, D. W. Nature 1999, 339 (supp.), 47.
(b) Parson, C. G.; Danyssz, W.; Quack, G. Drug News Perspect. 1998, 11,
523, and references therein.
(6) Shin-ya, K.; Kim, J.-S.; Furihata, K.; Hayakawa, Y.; Seto, H.
Tetrahedron Lett. 1997, 40, 7079.
(8) (a) Sharpless, K. B.; Amberg, W.; Bennani, Y. L.; Crispino, G. A.;
Hartung, J.; Jeong, K.-S.; Kwong, H.-L.; Morikawa, K.; Wang, Z.-M.; Xu,
D.; Zhang, X.-L. J. Org. Chem. 1992, 57, 2768. (b) Kolb, H. C.; VanNieu-
wenhze, M. S.; Sharpless, K. B. Chem. ReV. 1994, 94, 2483.
(9) Li, H.; Sakamoto, T.; Kikugawa, Y. Tetrahedron Lett. 1997, 38, 6677.
(10) (a) David, M.; Dhimane, H.; Vanucci-Bacque, C.; Lhommet, G. Synlett
1998, 206. (b) Chiesa, M. V.; Manzoni, L.; Scolastico, C. Synlett 1996, 441.
(7) The stereochemistry was proposed to be 2S,3S,4R,7R,9S or 2S,3S,4R,-
7R,9R. Shin-Ya, K. Personal communication. During the preparation of this
manuscript, a communication regarding the stereochemistry of kaitocephalin
appeared, see: Kobayashi, H.; Shin-ya, K.; Furihata, K.; Hayakawa, Y.; Seto,
H. Tetrahedron Lett. 2001, 42, 4021.
10.1021/ja016403w CCC: $20.00 © 2001 American Chemical Society
Published on Web 09/07/2001

Communications to the Editor
J. Am. Chem. Soc., Vol. 123, No. 39, 2001 9707
Switching the base to LiHMDS provided improved results, and
a major isomer 3a was isolated in 60% yield (with its stereo-
chemistry assigned by single-crystal X-ray analysis). Fortunately,
when 2b was used as substrate with LiHMDS as base, the aldol
reaction provided alcohol 3b as a single product in 86% yield.
Raising the temperature of aldol reaction from -78 °C to -42
°C or treating 3b with LiHMDS at -42 °C produced the bicyclic
oxazolidinone 4 in excellent yields. The structure of 4 was
established by its single-crystal X-ray analysis, thereby confirming
the stereochemistry of 3b. Although we obtained 3b in high
diastereoselectivity, the configuration of its 3-OH was wrong for
synthesizing kaitocephalin. Considering that our product was a
Cram-selective isomer, we tried to add anhydrous ZnBr2 or CuI
to the aldol reaction to reach anti-Cram selectivity to get 3S-
isomer.11 However, this attempt failed to obtain any coupling
products. At this moment we planned to continue the synthesis
by reverting the configuration of the 3-OH of 3b. Accordingly,
the oxazolidine ring of 3b was opened, and the primary hydroxy
group was selectively protected with TBSCl to afford the alcohol
5. Swern oxidation of 5 followed by reduction with NaBH4 in
THF/MeOH provided 3S-isomer 6 as a major product in 86%
yield. The alcohol 5 was also isolated in <10% yield after
reaction, which implied that the diastereoselectivity for the
reduction step was 9/1. It was notable that, when 3b was subjected
to this oxidation/reduction method, no desired 3S-isomer was
obtained at all after reduction. After the cleavage of the silyl ether
in 6, the diol was protected with acetic anhydride to afford 7,
which was submitted to Sharpless asymmetric dihydroxylation,
and then protecting of the primary hydroxy group with TPSCl
provided 8a and 8b. We found the ratio of 8a to 8b was about
2.2/1 when commercial AD-mix-â was used, which indicated that
the chiral centers in 7 had some mismatched effect on the
diastereoselectivity in the dihydroxylation. Moreover, by using
enriched AD-mix-â, we raised the ratio of 8a to 8b to 6.8/1.
Mesylation of 8a followed by removal of the silyl-protecting
group provided mesylate 9 (Scheme 3). Hydrogenolysis of 9
followed by Jones oxidation and subsequent esterification with
diazomethane gave diester 10. To distinguish 1-OAc and 3-OAc
and convert the N-methoxycarbonyl group to the oxazolidinone
ring that was easier to be cleavaged, the diester 10 was treated
with 3 equiv of magnesium in methanol to afford 11 in 78%
yield.12 This step was essential for this total synthesis because
we found that the N-methoxycarbonyl-protecting group could not
be removed under various conditions in the final stage. The
solvent in this transformation was critical, because we observed
that when 10 was treated with 3 equiv of magnesium in ethanol
either no reaction occurred (at 40 °C) or another oxazolidinone
13 was formed (at 50 °C). Clearly, 13 was produced by an
intramolecular SN2 reaction. Its X-ray structure supported the
configuration of 9-OH in 8a. Next, reaction of the mesylate 11
with sodium azide followed by PtO2-catalyzed hydrogenation
produced the primary amine, which was treated with 3,5-dichloro-
4-benzoxybenzoic chloride to afford an amide, which was
submitted to Jones oxidation to give acid 12. Finally, 12 was
Scheme 3a
a Reagents and conditions: (a) i) MsCl, Et3N, DMAP, CH2Cl2, 0 °C,
ii) TBAF, HOAc, 90% for two steps; (b) i) PtO2, H2, 100%, ii) Jones
oxidation then CH2N2, 85%; (c) 3 equiv Mg, MeOH, 35-40 °C, 78%;
(d) i) NaN3, DMF, 30 °C, ii) Pd/C, H2, iii) 3,5-dichloro-4-benzoxybenzoic
chloride, Et3N, CH2Cl2, rt, iv) Jones oxidation, 53% for four steps; (e) i)
Pd/C, H2, 100%, ii) 2 N HCl, 100%, iii) 1 N NaOH, 92%.
transformed into kaitocephalin by following deprotection steps:
(1) hydrogenolysis to remove benzyl group, (2) treatment with 2
N HCl to remove Boc group, and (3) hydrolysis with 1 N NaOH
to cleave the ester and oxazolidinone ring. Comparison of the
spectroscopic data of the product with those of authentic kaito-
cephalin confirmed the identity of our synthetic kaitocephalin.
This first total synthesis of kaitocephalin includes a number
of key steps which are of broad interest, including the highly
diastereoselective aldol reaction of 2b with (R)-Garner aldehyde
to provide 3b and various functional group manipulations
involving internal protection and group selectivity. In addition,
This synthetic approach (25 linear steps, 8% overall yield) should
permit the preparation of useful quantities of kaitocephalin and
its analogues, greatly facilitating the ongoing pharmacological
studies of AMPA/KA receptors.
Acknowledgment. We are grateful to the National Natural Science
Foundation of China (Grant 29725205), Shanghai Unilever Research &
Development Foundation, and Qiu Shi Science & Technologies Founda-
tion for financial support. We thank Professor K. Shin-ya of Tokyo
University for providing the stereochemistry information of kaitocephalin.
Supporting Information Available: Experimental procedures and
characterizations for kaitocephalin and compounds 1-12 (PDF). This
material is available free of charge via the Internet at http://pubs.acs.org.
(11) Herold, P. HelV. Chim. Acta 1988, 71, 354.
(12) (a) Xu, Y.-C.; Bizuneh, A.; Walker, C. J. Org. Chem. 1996, 61, 9086.
(b) Xu, Y.-C.; Lebeau, E.; Walker, C. Tetrahedron Lett. 1994, 35, 6207.
JA016403W