Structure–activity relationship study at C9 position of kaitocephalin

Article abstract of DOI:10.1016/j.bmcl.2016.06.026

Kaitocephalin (KCP) isolated from Eupenicillium shearii PF1191 is an unusual amino acid natural product in which serine, proline, and alanine moieties are liked with carbon–carbon bonds. KCP exhibits potent and selective binding affinity for one of the ionotropic glutamate receptor subtypes, NMDA receptors (K_i= 7.8 nM). In this study, new structure–activity relationship studies at C9 of KCP were implemented. Eleven new KCP analogs with different substituents at C9 were prepared and employed for binding affinity tests using native ionotropic glutamate receptors. Replacement of the 3,5-dichloro-4-hydroxybenzoyl group of KCP with a 3-phenylpropionyl group resulted in significant loss of binding affinity for NMDARs (K_i= 1300 nM), indicating an indispensable role of the aromatic ring of KCP in the potent and selective binding to NMDARs. Other analogs showed potent binding affinity in a range of 11–270 nM. These findings would directly link to develop useful chemical tools toward imaging and labeling of NMDARs.

Full text of DOI:10.1016/j.bmcl.2016.06.026

Bioorganic & Medicinal Chemistry Letters 26 (2016) 3543–3546
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Structure–activity relationship study at C9 position of kaitocephalin
Yoko Yasuno a, Makoto Hamada a, Yuya Yoshida a, Keiko Shimamoto b, Yasushi Shigeri c,
Toshifumi Akizawa d, Motomi Konishi d, Yasufumi Ohfune a, Tetsuro Shinada a,
⇑
a Graduate School of Science, Osaka City University, 3-3-138, Sugimoto, Sumiyoshi, Osaka 558-8585, Japan
b Bioorganic Research Institute, Suntory Foundation for Life Sciences, 8-1-1, Seikadai, Seika-cho, Soraku-gun, Kyoto 619-0284, Japan
c National Institute of Advanced Industrial Science and Technology, 1-8-31, Midorigaoka, Ikeda, Osaka 563-8577, Japan
d Analytical Chemistry, Pharmaceutical Science, Setsunan University, 45-1 Nagaotoge-cho, Hirakata, Osaka 573-0101, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Kaitocephalin (KCP) isolated from Eupenicillium shearii PF1191 is an unusual amino acid natural product
in which serine, proline, and alanine moieties are liked with carbon–carbon bonds. KCP exhibits potent
and selective binding affinity for one of the ionotropic glutamate receptor subtypes, NMDA receptors
(Ki = 7.8 nM). In this study, new structure–activity relationship studies at C9 of KCP were implemented.
Eleven new KCP analogs with different substituents at C9 were prepared and employed for binding affin-
ity tests using native ionotropic glutamate receptors. Replacement of the 3,5-dichloro-4-hydroxybenzoyl
group of KCP with a 3-phenylpropionyl group resulted in significant loss of binding affinity for NMDARs
(Ki = 1300 nM), indicating an indispensable role of the aromatic ring of KCP in the potent and selective
binding to NMDARs. Other analogs showed potent binding affinity in a range of 11–270 nM. These find-
ings would directly link to develop useful chemical tools toward imaging and labeling of NMDARs.
Ó 2016 Elsevier Ltd. All rights reserved.
Received 17 May 2016
Revised 8 June 2016
Accepted 9 June 2016
Available online 11 June 2016
Keywords:
Ionotropic glutamate receptors
Kaitocephalin
Natural product
NMDA receptors
Structure–activity relationship study
Ionotropic glutamate receptors (iGluRs) are ligand-gated ion
channels which mediate excitatory neurotransmission in the
mammalian central nervous system.1 Extensive efforts have been
devoted to elucidate the function of iGluRs because of their crucial
role in higher order brain functions such as memory and learning.
iGluRs are grouped into three major subtypes: kainate receptors
remarkable neuroprotecting effect. Miledi et al. investigated phar-
macological details of KCP and proved that KCP was found to be
potent antagonists of NMDARs (IC50 = 75 9 nM) and AMPARs
(IC50 = 242 37 nM) by electrophysiological studies.6 We have
recently evaluated subtype selectivity and binding affinity of KCP
for each iGluR subtype using native iGluRs prepared from rat
synaptic membrane.7 Our study showed that KCP was found to
be a potent and selective ligand for native NMDARs (Ki = 7.8 nM)
over AMPARs (Ki = 590 nM) and KARs (Ki = 14,000 nM).
Recently, we examined the structure analysis of 1 bound to the
ligand binding domain (LBD) of iGluR subtype, GluA2 by X-ray
analysis (Fig. 2A)8 and computational docking studies of 1 with a
NMDAR subtype, GluN2A. These results provided several interest-
ing conformational and binding features of the KCP-iGluR com-
plexes: (i) the ligand–protein complex forms an open-bilobed
structure in which the structural feature is found in other
ligand–protein complexes of iGluR antagonists bound to iGluRs,
(ii) the glutamate like moiety (C1–C4 in KCP) is buried in the glu-
tamate binding site, (iii) the benzoyl alanine side chain at C7 lies in
the cavity of the opened bilobe far from the glutamate binding site,
(iv) interactions between the side chain at C7 and the amino acid
residues G486, H485, R692, and E517 of GluN2A are predicted by
the docking model of KCP bound to GluN2A (Fig. 2B).
(KARs), N-methyl-D-aspartate receptors (NMDARs), and
a-amino-
3-hydroxy-5-methyl-4-isoxazolepropionic
acid
receptors
(AMPARs) by their selective binding abilities of kainic acid (KA:
3), NMDA (4), and AMPA (5) for each iGluR subtype.2 These sub-
type selective ligands as well as naturally occurring iGluR ligands
such as domoic acid (6), quisqualate (7), dysiherbaine (8), and
NPTX-594 (9) have been widely employed as a useful tool for neu-
robiological studies to investigate functions of each iGluR subtype
by selective activation and inactivation of iGluRs.3 In addition,
these ligands have successfully served for X-ray structure analysis
to understand the gating mechanism how these ligands open or
close iGluRs in the molecular structure level (Fig. 1).4
Kaitocephalin (1: KCP) was isolated form Eupenicillium shearii
PF1191 by a screening program of naturally occurring neuropro-
tecting agents which suppressed KA-induced excitotoxicity.5 It
was speculated that KCP would antagonize iGluRs to display the
Further structure–activity relationship (SAR) studies using KCP
analogs 10–12 were performed to screen the key pharmacophore
⇑
Corresponding author. Tel.: +81 6 6605 2570; fax: +81 6 6605 3153.
E-mail address: shinada@sci.osaka-cu.ac.jp (T. Shinada).
http://dx.doi.org/10.1016/j.bmcl.2016.06.026
0960-894X/Ó 2016 Elsevier Ltd. All rights reserved.

3544
Y. Yasuno et al. / Bioorg. Med. Chem. Lett. 26 (2016) 3543–3546
Figure 3. Binding affinity of KCP and its analogs for NMDARs prepared from rat
synaptic membranes (Ki values). [3H]CGP 39653 was used as a competitive ligand.
Figure 1. Structures of ionotropic glutamate receptor ligands: kaitocephalin (1),
glutamate (2), kainic acid (3), NMDA (4), AMPA (5) and other naturally occurring
ligands (6)–(9).
at C7 (Fig. 3).7 Analogs 10–12 revealed much lower affinity for
NMDARs than KCP (1), indicating that powerful contribution of
the aromatic ring and the carboxylic acid at C9 for the potent bind-
ing to NMDARs. These experimental results are consisted with the
assumption obtained by docking model study in which these func-
tional groups would cooperatively interact with NMDARs as shown
in Figure 2B.
Figure 4. Structures of KCP analogs.
In this study, we performed an advance SAR study focusing on
the benzoyl group at C9 of KCP (Fig. 4). New analogs 14a–17a in
which the hydroxyl and chlorides on the benzoyl group were
removed or replaced with other functional groups were employed
to assess effects of these functional groups on the potency and
selectivity. The corresponding (9R)-analogs 13 and 14b–17b were
also employed to evaluate tolerance of the cavity of the binding
site where the benzoyl group fits. 3-Phenylpropionamide analogs
18a and 18b were designed to test whether the distal aromatic ring
could take part in the preferential ligand–protein interactions or
not.
KCP analogs 14a and 14b were prepared in one step by a reduc-
tive dechlorination reaction of a mixture of KCP (1) and (9R)-KCP
(13) under the hydrogenation reaction conditions in the presence
of i-Pr2Net (Scheme 1). The diastereomeric mixture was purified
by a C18-reversed phased HPLC to give (9S)-14a and (9R)-14b,
respectively.
Scheme 1. Synthesis of 14ab from 1 and its (9R)-isomer (13).
Analogs 17a and 17b were prepared in a similar manner to our
previous synthetic method for the efficient total synthesis of KCP
(1)9 (Scheme 2). Commercially available 3-chloro-4-hydroxybenzoic
acid (19) was converted to allyl benzoate 20 in 82% yield via the Clai-
sen rearrangement reaction. Compound 20 was subjected to benzyl
etherification and hydrolysis reactions to provide the acid 21 which
was condensed with ammonium salt 23 prepared from 229 to give
phosphonate 24 in 66% yield. Aldehyde 259 was subjected to the ole-
fination reaction with 24 in the presence of DBU/NaI to give E-26 in a
highly selective manner (E:Z = > 95:5). Hydrogenation of E-26 in the
presence of a Rh catalyst under H2 allowed the selective reduction of
the dehydroamino acid and allyl group to afford a 7:3 mixture of
(9S)- and (9R)-27 in 72% yield without the loss of the chloride atom.
Global deprotection of 27 was achieved under the conditions using
AlCl3 and Me2S. Theresulting crude mixture of17a and17bwas puri-
fied by an ion-exchange chromatography using Dowex 50W
followed by a C18-reversed phased HPLC to give (9S)-17a and (9R)-
17b, respectively. The stereochemistry at C9 of 17a and 17b was
Figure 2. (A) KCP bound to GluA2 ligand binding domain (LBD), (B) Docking model
of KCP bound to GluN2A.

Y. Yasuno et al. / Bioorg. Med. Chem. Lett. 26 (2016) 3543–3546
3545
Table 1
Ki values of 1, 13, 14–18 for NMDA, AMPA, and KA receptors using 3H-labelled
ligandsa and rat brain synaptic membrane11
Compound
Ki valueb (nM)
NMDA
7.8 0.8
AMPA
KA
14,000 3300
14,000 1600
15,000 2000
7800 3100
18,000 7300
25,000 9300
>105c
47,000 11000
28,000 4300
19,000 9700
>105
KCP (1)
14a
15a
16a
17a
590 54
1600 230
1900 270
610 240
22
6
8
2
1
49
11
17
260 32
18a
1300 300
60
3700 870
2500 560
6100 800
4700 830
790 130
(9R)-KCP (13)
(9R)-14b
(9R)-15b
(9R)-16b
(9R)-17b
(9R)-18b
9
270 68
270 50
29
73
5
9
2000 150
18,000 1300
4100 1300
10,000 1900
a
3H-labelled ligands: [3H]CGP 39653 for NMDARs, [3H]AMPA for AMPARs, and
[3H]KA for KARs.
b
Ki values are indicated as the mean SE for three determinations.
Binding affinity at 100,000 nM was not detected.
c
(9R)-analog of KCP 13 and (9R)-14b–17b displayed similar pattern
of the selectivity and potent binding affinity for NMDARs in a range
of 29–270 nM (Ki values). (9R)-Analog 18b displayed weak binding
affinity for NMDARs (Ki = 4100 nM) and the others.
It is interesting to note that analogs 18a and 18b having a distal
benzene ring displayed significant loss of the binding affinity
(Table 1 and Fig. 5). Moreover, configurationally different (9R)-
isomers 13 and 14b–17b showed potent binding affinity for
NMDARs at a nano molar range though these binding affinities
were lower than those of the (9S)-series. The indispensable role
of the benzoyl group and remarkable tolerance of NMDARs which
flexibly accommodate the sterically different (9R)-analogs are
unveiled for the first time by these experiments.
Screening of the key functional group using 1 and 14a–17a
revealed several new insights into the binding mechanism of KCP
to NMDARs. The chloride atoms could be replaced with iodide
atoms or hydrogen atoms and one of the chloride atoms could be
replaced with a propyl side chain without significant loss of bind-
ing affinity (Fig. 5). The binding affinity decreased when 15a was
employed. It was 6.7-fold less potent than that of KCP and 2.2-fold
less potent than 14a. These results provide supporting evidence of
the interaction between the phenolic hydroxy group and the bind-
ing pocket of NMDARs. It is also consisted with the computational
model predicted by the docking study shown in Figure 2B.
Comparison of the binding affinity of 1, 16a, and 14a indicate a
possibility that the chloride atoms on the benzoyl group would
contribute to the potent binding to NMDARs. The halogen atom
effect was also observed in the (9R)-analogs 13 and 14b–18b.
Scheme 2. Synthesis of 17a and 17b.
determined by comparison of their NMR data and HPLC profile with
those of KCP and (9R)-KCP (Supporting information). In a similar
manner, analogs 15ab, 16ab, and 18ab were prepared from the cor-
responding benzoyl phosphonates 28–30, respectively (Scheme 3).
Binding affinity of the above new analogs for each iGluR subtype
was evaluated using native iGluRs prepared by rat synaptic mem-
branes (Table 1, SI-Figs. 1 and 2).7,10 [3H]KA for the KARs, [3H]AMPA
for the AMPARs, and [3H]CGP 39653 for the NMDARs were used for
subtype selective radiolabeled ligands.11 (9S)-Analogs 14a–17a dis-
played selective binding affinity for NMDARs over AMPARs and
KARs. These selectivity patterns are the same as that of KCP. Biding
affinity of these analogs (Ki = 11–49 nM) was found to be slightly
lower than that of KCP (Ki = 7.8 nM). In contrast, remarkable loss
of binding affinity for iGluRs was observed when analog 18a having
a distal benzene ring instead of the benzoyl group was employed. It
displayed a 170-fold lower potency than that of KCP. Interestingly,
Scheme 3. Synthesis of 15ab, 16ab, and 18ab.
Figure 5. Comparison of binding affinity for NMDARs. aKi value.

3546
Y. Yasuno et al. / Bioorg. Med. Chem. Lett. 26 (2016) 3543–3546
Interestingly, (9R)-iodo analog 16b was 2-fold more potent than
that of (9R)-KCP. Although the binding mechanism of 16b to
NMDARs has not been unclear, these results suggest the presence
of a preferential fitting site of the iodine atom in the binding pocket
of NMDARs.
References and notes
1. (a) Traynelis, S. F.; Wollmuth, L. P.; McBain, C. J.; Manniti, F. S.; Vance, K. M.;
Ogden, K. K.; Hansen, K. B.; Yuan, H.; Myers, S. J.; Dingledine, R. Pharmacol. Rev.
2010, 62, 405; (b) Kew, J. N. C.; Kemp, J. A. Psychopharmacology 2005, 179, 4; (c)
Dingledine, R.; Borges, K.; Bowie, D.; Traynelis, S. F. Pharmacol. Rev. 1999, 51, 7.
2. (a) Li, S.; Wong, A. H. C.; Liu, F. Front. Cell. Neurosci. 2014, 8, 125; (b) Lai, T. W.;
Zhang, S.; Wang, Y. T. Prog. Neurobiol. 2014, 115, 157; (c) Paoletti, P.; Bellone, C.;
Zhou, Q. Nat. Rev. Neurosci. 2013, 14, 383; (d) Lakhan, S. E.; Caro, M.;
Hadzimichalis, N. Front. Psychiatry 2013, 4, 52; (e) Monaghan, D. T.; Jane, D.
E. Pharmacology of NMDA Receptors. In Biology of the NMDA Receptor; Van
Dongen, A. M., Ed.; CRC Press, 2009. Chapter 12.
3. (a) Sakai, R.; Swanson, G. T. Nat. Prod. Rep. 2014, 31, 273; (b) Swanson, G. T.;
Sakai, R. Prog. Mol. Subcell. Biol. 2009, 46, 123; (c) Regan, M. C.; Romero-
Hernandez, A.; Furukawa, H. Curr. Opin. Struct. Biol. 2015, 33, 68.
4. (a) Kumar, J.; Mayer, M. L. Annu. Rev. Physiol. 2013, 75, 313; (b) Unno, M.;
Shinohara, M.; Takayama, K.; Tanaka, H.; Teruya, K.; Doh-ura, K.; Sakai, R.;
Sasaki, M.; Ikeda-Saito, M. J. Mol. Biol. 2011, 413, 667; (c) Frydenvang, K.; Lash,
L. L.; Naur, P.; Postila, P. A.; Pickering, D. S.; Smith, C. M.; Gajhede, M.; Sasaki,
M.; Sakai, R.; Pentikäinen, O. T.; Swanson, G. T.; Kastrup, J. S. J. Biol. Chem. 2009,
284, 14219; (d) Mayer, M. L. Neuron 2005, 45, 539; (e) Nanao, M. H.; Green, T.;
Stern-Bach, Y.; Heinemann, S. F.; Choe, S. Proc. Natl. Acad. Sci. U.S.A. 2005, 102,
1708; (f) Jin, R.; Horning, M.; Mayer, M. L.; Gouaux, E. Biochemistry 2002, 41,
15635; (g) Armstrong, N.; Sun, Y.; Chen, G.-Q.; Gouaux, E. Nature 1998, 395,
913.
In this study, the utility of KCP as a potent platform for the
development of NMDAR ligands was displayed. However, the KCP
analogs prepared in this study showed moderate NMDAR selectiv-
ity than KCP. Previously, we discovered that the 7-epimer of KCP
[(7S)-KCP] was found to be the superior NMDAR selective ligand.7
It is conceivable that the benzoyl group modification of (7S)-KCP
would be a promising approach to improve the ligand selectivity.
We found that the propyl side chain of 17a is tolerated. This
result suggests that one of the ortho-positions to the hydroxy
group would be a preferred point for further attachment of various
substituent or a tether for the development of more potent NMDA
ligands and chemical probe for functional studies of NMDARs from
the view point of chemical biology.12 Further studies including the
synthesis of fluorescent chemical probe by taking advantage of KCP
as a promising platform is ongoing.
5. Shin-ya, K.; Kim, J.-S.; Furihata, K.; Hayakawa, Y.; Seto, H. Tetrahedron Lett.
1997, 38, 7079.
6. Limon, A.; Reyes-Ruiz, J. M.; Vaswani, R. G.; Chamberlin, A. R.; Miledi, R. ACS
Chem. Neurosci. 2010, 1, 175.
Acknowledgments
7. Yasuno, Y.; Hamada, M.; Kawasaki, M.; Shimamoto, K.; Shigeri, Y.; Akizawa, T.;
Konishi, M.; Ohfune, Y.; Shinada, T. Org. Biomol. Chem. 2016, 14, 1206.
8. Ahmed, A. H.; Hamada, M.; Shinada, T.; Ohfune, Y.; Weerasinghe, L.; Garner, P.
P.; Oswald, R. E. J. Biol. Chem. 2012, 287, 41007.
9. Hamada, M.; Shinada, T.; Ohfune, Y. Org. Lett. 2009, 11, 4664.
10. (a) Sakai, R.; Swanson, G. T.; Shimamoto, K.; Green, T.; Contractor, A.; Ghetti, A.;
Tamura-Horikawa, Y.; Oikawa, C.; Kamiya, H. J. Pharmacol. Exp. Ther. 2001, 296,
650; (b) Shimamoto, K.; Lebrun, B.; Yasuda-Kamatani, Y.; Sakaitani, M.; Shigeri,
Y.; Yumoto, N.; Nakajima, T. Mol. Pharmacol. 1998, 53, 195; (c) Murphy, D. E.;
Snowhill, E. W.; Williams, M. Neurochem. Res. 1987, 12, 775.
TS gratefully thanks to financial supports from Scientific
Research of Innovative Areas, Chemical Biology of Natural Prod-
ucts, the Ministry of Education, Culture, Sports, Science and Tech-
nology (No. 23102009) and the Japan Society for the Promotion
of Science (KAKENHI Nos. 23228001, 25282233). YY gratefully
thanks to financial supports by the Sasagawa Scientific Research
Grant from The Japan Science Society.
11. The preparation of rat synaptic membrane was carried out in accordance with
the National Institute of Advanced Industrial Science and Technology (AIST)
guidelines for life science experiments. The experimental procedure of
receptor binding assay and calculation of Ki values are described in details
on Supporting information.
Supplementary data
Supplementary data (experimental details and spectra (1H, 13C
NMR) for all products) associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.bmcl.2016.06.
026.
ˇ
12. (a) Schenone, M.; Dancík, V.; Wagner, B. K.; Clemons, P. A. Nat. Chem. Biol. 2014,
9, 232; (b) Rizzo, S.; Waldmann, H. Chem. Rev. 2014, 114, 4621; (c) Ueda, M.
Chem. Lett. 2012, 41, 658; (d) Nicolaou, K. C.; Hale, C. R. H.; Nilewski, C.;
Loannidou, H. A. Chem. Soc. Rev. 2012, 41, 5185.