A concise [C+NC+CC] coupling-enabled synthesis of kaitocephalin

Article abstract of DOI:10.1039/c4cc01692k

A 15-step synthesis of the iGluR antagonist kaitocephalin from aspartic acid is reported. The linchpin pyrrolidine ring of the target molecule is efficiently assembled with in a single operation via an asymmetric [C+NC+CC] reaction.

Full text of DOI:10.1039/c4cc01692k

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
A concise [C+NC+CC] coupling-enabled synthesis
of kaitocephalin†‡
Cite this: Chem. Commun., 2014,
50, 4908
Philip Garner,* Laksiri Weerasinghe, Ian Van Houten and Jieyu Hu
Received 6th March 2014,
Accepted 27th March 2014
DOI: 10.1039/c4cc01692k
www.rsc.org/chemcomm
A 15-step synthesis of the iGluR antagonist kaitocephalin from
aspartic acid is reported. The linchpin pyrrolidine ring of the target
molecule is efficiently assembled with in a single operation via an
asymmetric [C+NC+CC] reaction.
Kaitocephalin is a natural product isolated from the fungus
Eupenicillium shearii that was found to be a potent competitive
antagonist of ionotropic glutamate receptor (iGluR) activity.1
This property is of interest because glutamate receptors are the
major excitatory neurotransmitter receptors in the vertebrate
brain and are involved in both normal and pathological neuronal
functions.2 Drugs targeting glutamate receptors may have
considerable potential for treating such diverse disorders as
epilepsy, amyotrophic lateral sclerosis, ischemic brain damage,
major depression, drug abuse, and neuropathic pain. Gluta-
mate receptors have also been implicated in non-neurological
Scheme 1 Retrosynthetic analysis of 1.
Our key retrosynthetic analysis of kaitocephalin (1, Scheme 1)
diseases such as diabetes3 and cancer.4 Many of these disease began with the introduction of the C1–C3 moiety via the union of 2
states are associated with the selective expression of iGluR with the serine-derived aldehyde 3 following an established aldol
subtypes, which are broadly classified according to their selective tactic (ref. 8a and e). The pyrrolidine precursor of 2 comprising the
activation by NMDA (N-methyl-D-aspartate), AMPA (2-amino-3-(3- remaining contiguous carbon skeleton of 1 would be assembled via
hydroxy-5-methyl-isoxazol-4-yl)propionate), or kainate. Kaitocephalin a Ag(I)-catalyzed asymmetric [C+NC+CC] coupling of the aspartic
displays differential affinity for iGluR subtypes in the order acid-derived aldehyde 4, (S)-glycyl sultam 5, and vinyl sulphone 6.9
NMDA 4 AMPA c kainate,5 and is thus of interest as a scaffold The key C–C bond-forming step involves a [3+2] cycloaddition
for drug design.6 Recently, the crystal structure of kaitocephalin of a metalated azomethine ylide (formed in situ from 4 and 5)
bound to an iGluR was determined and provides clues for with an electronically-activated dipolarophile 6. The advantage of
the development of subtype-specific antagonists.7 As a first [C+NC+CC] coupling over other [3+2] cycloaddition methodology
step towards this goal, we report a concise total synthesis stems from the ability to employ aliphatic aldehydes such as 4
of kaitocephalin. Our approach is distinguished from prior without enolization or enamine formation that would result in
syntheses8a–g of kaitocephalin by its effective use of the asym- undesired side reactions. This limitation precludes the applica-
metric [C+NC+CC] coupling reaction to assemble the target’s tion of existing [3+2] dipolar cycloaddition technology to the
core pyrrolidine ring.
kaitocephalin pyrrolidine problem. While essential for activating
the dipolarophile in the underlying cycloaddition process, the
phenylsulfonyl moiety would be removed reductively at a sub-
sequent stage in the synthesis.10
Reduction of this plan to practice began with the synthesis
of the key pyrrolidine 2 (Scheme 2). The [C+NC+CC] coupling
of 4,11 5,12 and 613 proceeded cleanly at ambient temperature
in the presence of Ag(I) ion to give the endo-cycloadduct 13 in
Department of Chemistry, Washington State University, Pullman, WA 99164-4630,
USA. E-mail: ppg@wsu.edu
† This paper is dedicated to Professor Yasufumi Ohfune on the occasion of his
retirement from Osaka City University.
‡ Electronic supplementary information (ESI) available: Experimental procedures
and characterization data for all new compounds are provided. See DOI: 10.1039/
c4cc01692k
4908 | Chem. Commun., 2014, 50, 4908--4910
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
85% isolated yield.14 The pyrrolidine ring stereochemistry of
this cycloadduct was expected based on the (endo-Si) pre-TS
ensemble 12 and was supported by the successful conversion of
13 to the natural product 1. Although the stereochemistry of the
phenylsulphonyl group at C6 is inconsequential because this
functional group is removed downstream, we believe that it is
endo based on our previous experience with the Ag(I)-catalyzed
[C+NC+CC] coupling reaction.15 Post [C+NC+CC] coupling
adjustment of the pyrrolidine substitution pattern began with
methanolysis of the acyl sultam using Mg(OMe)2 to give the
pyrrolidine ester 14 which was N-carbamoylated with methyl
chloroformate to provide the methyl carbamate 15. Chemoselective
reductive cleavage of the phenylsulfonyl group was accomplished
with sodium amalgam to provide 2 in 62% overall yield from 13
concluding the 3-step sequence. Alternatively, the cycloadduct 13
could be converted to 2 in just two steps in comparable overall yield
by exposing 13 to excess magnesium metal in MeOH to produce
the methyl ester (desulphonation and methanolysis) followed
by N-carbamoylation to afford 2.
With the pyrrolidine 2 in hand, we were poised to complete
the synthesis of kaitocephalin (Scheme 3). We found that the
lithium enolate of 2 reacted with the serinal derivative 3 to give
a 2 : 1 mixture of oxazolidinone 16 and undesired (3R) aldol
diastereomer 17 in 73% combined yield. Presumably, 16 arose
from the preferential cyclization of the major (3S) aldolate. This
Scheme 2 Synthesis of pyrrolidine 2.
Scheme 3 Kaitocephalin end-game.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 4908--4910 | 4909

View Article Online
ChemComm
Communication
J. Yang and D. Ma, Tetrahedron, 2011, 67, 1673; ( f ) K. Takahashi,
D. Yamaguchi, J. Ishihara and S. Hatakeyama, Org. Lett., 2012,
14, 1644; (g) W. Lee, J. H. Youn and S. H. Kang, Chem. Commun.,
2013, 49, 5231.
9 P. Garner, H. U. Kaniskan, J. Hu, W. J. Youngs and M. Panzner, Org.
Lett., 2006, 8, 3647.
10 A related approach to 1 had been suggested but never implemented,
apparently due to the poor yield and stereoselectivity associated with
the underlying azomethine ylide cycloaddition (K. V. Kudryavtsev,
N. V. Nukolova, O. V. Kokoreva and E. S. Smolin, Russ. J. Org. Chem.,
2006, 42, 412).
result was in line with Ma’s report (ref. 8e). Because of the
difficulty in separating 16 from 17, the mixture was used for the
next reaction. Following Dondoni’s protocol,16 exposure of this
mixture to Jones reagent resulted in clean conversion of both
N-protected oxazolidines to their N-protected a-amino acids. The
resulting diacid 18 was treated directly with excess diazomethane to
give the triester 19 in 36% overall yield from 2. This three-step
reaction sequence accomplished a total of five synthetic transfor-
mations (aldol addition, oxazolidinone formation, bis-oxazolidine
hydrolysis, diol oxidation, and bis-esterification), affording pure 19
after a single chromatography.17 Hydrogenolysis of 19 released the
amine 20, which was acylated with the known acid chloride 21 to
give the amide 22 in good overall yield. Synthesis of 22 constituted
a formal synthesis of the target molecule 1 (ref. 8e). Removal of the
Boc, benzyl ether, and three methyl ester protecting groups was
effected by treating 22 with the Fujita reagent18 to afford 23. This
was followed by hydroxide-mediated hydrolysis of the oxazolidi-
none to give kaitocephalin (1). Our material was identical to an
authentic sample of 1.
¨
11 Aldehyde 4 was synthesized from aspartic acid 7 as shown below.
This short (15 steps from aspartic acid) total synthesis of
kaitocephalin demonstrates the value of the asymmetric
[C+NC+CC] coupling reaction for the construction of complex
pyrrolidine structures and paves the way for the generation of
novel kaitocephalin analogues for biological studies.
Our research was supported by a grant from the National
Science Foundation (CHE-1149327). We thank Vertex Pharma-
ceuticals for a supply of 5ÁHCl and the Ohfune lab at Osaka City
University for providing us with a sample of kaitocephalin for
comparison.
12 Glycyl sultam 5 can be prepared by a route involving azide reduction
¨
(H. U. Kaniskan, A Formal Total Synthesis of Bioxalomycin Beta 2,
PhD dissertation, Case Western Reserve University, 2007) or the
´
Delepine reaction (A. Isleyen, C. Gonsky, R. C. Ronald and P. Garner,
Synthesis, 2009, 1261).
13 L. A. Paquette and R. V. C. Carr, Org. Synth., 1990, 7, 453.
14 This multicomponent reaction could be performed on a 10 g scale
(aldehyde 4 is the limiting reagent) but with a slight reduction in
yield of 13 to 70%.
Notes and references
1 K. Shin-ya, J.-S. Kim, K. Furihata, Y. Hayakawa and H. Seto, Tetra-
hedron Lett., 1997, 38, 7079.
15 Indirect evidence for the endo-stereochemistry in 13 comes from the fact
that the Cu(I)-catalyzed [C+NC+CC] coupling of 4, 5 and 6, conditions
that favour the exo-product (P. Garner, J. Hu, C. G. Parker, W. J. Youngs
and D. Medvetz, Tetrahedron Lett., 2007, 48, 3867), produced a cyclo-
adduct that was isomeric to 13. (See ESI‡).
16 A. Dondoni, A. Marra and A. Massi, Tetrahedron, 1998, 54, 2827.
17 Compound 25 was also isolated in 11% yield, presumably arising from
a retro-aldol reaction at some stage during the reaction sequence.
2 S. F. Traynelis, L. P. Wollmuth, C. J. McBain, F. S. Menniti,
K. M. Vance, K. K. Ogden, K. B. Hansen, H. Yuan, S. J. Myers and
R. Dingledine, Pharmacol. Rev., 2010, 62, 405.
3 P. E. Cryer, Endocrinology, 2012, 153, 1039.
4 W. Rzeski, C. Ikonomidou and L. Turski, Biochem. Pharmacol., 2002,
64, 1195.
5 A. Limon, J. M. Reyes-Ruiz, R. G. Vaswani, A. R. Chamberlin and
R. Miledi, ACS Chem. Neurosci., 2009, 19, 132.
6 R. G. Vaswani, A. Limon, J. M. Reyes-Ruiz, R. Miledi and
A. R. Chamberlin, Bioorg. Med. Chem. Lett., 2009, 19, 132.
7 A. H. Ahmed, M. Hamada, T. Shinada, Y. Ohfune, L. Weerasinghe,
P. P. Garner and R. E. J. Oswald, Biol. Chem., 2012, 287, 41007.
8 (a) H. Watanabe, M. Okue, H. Kobayashi and T. Kitahara, Tetrahedron
Lett., 2002, 43, 861; (b) M. Kawasaki, T. Shinada, M. Hamada and
Y. Ohfune, Org. Lett., 2005, 7, 4165; (c) R. G. Vaswani and A. R. J.
Chamberlin, J. Org. Chem., 2008, 73, 1661; (d) M. Hamada, T. Shinda 18 M. Node, K. Nishide, M. Sai, K. Fuji and E. J. Fujita, J. Org. Chem.,
and Y. Ohfune, Org. Lett., 2009, 11, 4664; (e) S. Yu, S. Zhu, X. Pan, 1981, 46, 1991.
4910 | Chem. Commun., 2014, 50, 4908--4910
This journal is ©The Royal Society of Chemistry 2014