Design, total synthesis, and biological evaluation of neodysiherbaine A derivative as potential probes

Article abstract of DOI:10.1016/j.bmcl.2006.08.082

To enable studies to elucidate the detailed biological function of dysiherbaine and neodysiherbaine A, potent and subunit-selective agonists for ionotropic glutamate receptors, the derivative with a hydroxymethyl substituent at the C10 position has been d

Full text of DOI:10.1016/j.bmcl.2006.08.082

Bioorganic & Medicinal Chemistry Letters 16 (2006) 5784–5787
Design, total synthesis, and biological evaluation of
neodysiherbaine A derivative as potential probes
Makoto Sasaki,a,* Koichi Tsubone,a Muneo Shoji,a Masato Oikawa,a
Keiko Shimamotob and Ryuichi Sakaic
aLaboratory of Biostructural Chemistry, Graduate School of Life Sciences, Tohoku University, Sendai 981-8555, Japan
bSuntory Institute for Bioorganic Research, Mishima-gun, Osaka 618-8503, Japan
cSchool of Fisheries Sciences, Kitasato University, Sanriku-cho, Iwate 022-0101, Japan
Received 22 June 2006; revised 14 August 2006; accepted 17 August 2006
Available online 1 September 2006
Abstract—To enable studies to elucidate the detailed biological function of dysiherbaine and neodysiherbaine A, potent and
subunit-selective agonists for ionotropic glutamate receptors, the derivative with a hydroxymethyl substituent at the C10 position
has been developed. Preliminary biological evaluation of the analogue showed that a C10 hydroxymethyl substituent produced sig-
nificant alterations in binding affinities for the ionotropic glutamate receptor subtypes.
Ó 2006 Elsevier Ltd. All rights reserved.
Glutamate receptors (GluRs) play a central role in the
mammalian central nervous system (CNS), not only in
excitatory neurotransmission but also in complex brain
functions such as learning and memory. GluRs are
broadly divided into ionotropic and metabotropic recep-
tors. Ionotropic GluRs (iGluRs) are further subdivided
into three subtypes on the basis of their pharmacological
Figure 1. Structures of dysiherbaine (1), neodysiherbaine A (2), and
10-hydroxymethyl-neodysiherbaine A (3).
preference toward selective agonists: a-amino-3-hy-
droxy-5-methylisoxazole-4-propionic acid (AMPA),
kainate, and N-methyl-D-aspartic acid (NMDA) recep-
tors.1 Molecular cloning studies demonstrated that
iGluRs are encoded by at least six NMDA (NR1,
NR2A-2D, and NR3A), four AMPA (GluR1-4), and
five kainate (GluR5-7 and KA1-2) receptor genes.2
but very low affinity for KA2 receptors, which produced
unusual biophysical behavior from heteromeric kainate
receptors.5 However, exact mode of interaction is still
elusive. Neodysiherbaine A (2, Fig. 1), a closely related
natural congener isolated as a minor constituent from
the same sponge, differs from dysiherbaine in the C8
functional group and is also a selective agonist for
AMPA and kainate receptors.6 Recently, it has been
shown that neodysiherbaine A is similar to dysiherbaine
in its pharmacological activity on kainate receptors,
albeit with slightly different binding affinities for individ-
ual receptor subunits.7
Dysiherbaine (1, Fig. 1),3 isolated from the Micronesian
sponge, Dysidea herbacea, is a remarkable excitatory
amino acid with potent convulsant activity. Dysiherba-
ine activates AMPA and kainate receptors, with a high-
er affinity for kainate receptors, but shows no detectable
affinity for NMDA receptors.4 Furthermore, it has been
revealed that dysiherbaine had extremely high affinity
for recombinant GluR5 or GluR6 kainate receptors
Due to their unprecedented molecular structures and
unique biological profiles, dysiherbaine and neodysiher-
baine A have attracted much attention among organic
and biological communities.6,8–11 Especially, the exceed-
ingly high affinity of dysiherbaines for GluR5 receptors
Keywords: AMPA receptors; Excitatory amino acid; Glutamate recep-
tors; Kainate receptors; Dysiherbaine; Neodysiherbaine A; Total
synthesis.
*
Corresponding author. Tel.: +81 22 717 8828; fax: +81 22 717
8897; e-mail: masasaki@bios.tohoku.ac.jp
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.08.082

M. Sasaki et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5784–5787
5785
suggests that their designed analogues could be a useful
probe in studying previously inaccessible synaptic sites,
at which GluR5 receptors play a functional role.
Accordingly, we have described the total synthesis of
dysiherbaine, neodysiherbaine A, and their structural
analogues, and structure–activity relationship (SAR)
studies.6,8b,10 These SAR studies showed that both the
type and stereochemistry of the C8 and C9 functional
groups affected the subtype selectivity of dysiherbaines
for members of the kainate receptor family. Some ana-
logues showed about 600-fold difference in affinities
for GluR5 over GluR6 in the binding assay.10c Although
these analogues could serve as interesting tools in neuro-
biology, appropriate labeling, preferably, non-radioiso-
tope (RI) labeling would expand their use. However,
selective small-molecule GluR agonists have resisted to
be labeled externally, for example, by fluorescent
groups, due to their small molecular size where no extra
positions are available for labeling. In the case of dysi-
herbaine, however, earlier modeling experiments7 sug-
gested that the terminal methylene unit at C10 of
dysiherbaines appears to contribute little to the binding
for GluRs, and introduction of some functional group
at C10 seems to be achieved without serious loss of its
binding property to GluRs. Hence, we designed the
C10 hydroxymethyl-substituted derivative 3 (Fig. 1) of
Scheme 1. Reagents and conditions: (a) compound 5, Yb(OTf)3,
CH2Cl2, rt, 83%; (b) cat OsO4, (DHQD)2AQN, K2CO3, K2[Fe(CN)6],
MeSO2NH2, t-BuOH/H2O, 0 °C, 83% (8:1 dr); (c) TBSCl, Et3N,
DMAP, CH2Cl2, rt, 92%; (d) Pd2(dba)3, neocuproine, toluene, 60 °C,
98%; (e) cat OsO4, NMO, acetone/H2O, rt, 92%; (f) DMP, CSA,
CH2Cl2, rt, 92%; (g) H2, Pd/C, hexane, rt, 85%; (h) SO3Æpyridine, Et3N,
DMSO, CH2Cl2, rt; (i) NaClO2, NaH2PO4, 2-methyl-2-butene,
t-BuOH/H2O, rt; (j) TMSCHN2, benzene/MeOH, rt, 75% (three steps).
neodysiherbaine
dysiherbaine.
A
as
a
precursor of ‘tagged’
In this letter, we describe a synthesis of the derivative
with a hydroxymethyl substituent at the C10 position
of neodysiherbaine A for the development of potential
molecular probes and show that this molecule still re-
tained high affinity for AMPA/kainate receptors while
some unexpected group selectivity took place.
sively, which was protected as the acetonide and then
subjected to hydrogenolysis to give primary alcohol
11. At this stage, the minor diastereomer at the C4 qua-
ternary center was readily removed by silica gel chroma-
tography. Oxidation of 11 by a two-step procedure led
to the corresponding acid, which was then esterified with
trimethylsilyldiazomethane to afford methyl ester 12.
The stereostructure of 12 was confirmed by NOE exper-
iments as shown in Scheme 1.
According to the previous total synthesis of neodysiher-
baine A (2) and its analogues,10b,c the synthesis of 3
started with C-glycosylation of allylsilane 512 with tri-
O-acetyl-D-glucal (4). Reaction of 4 with 5 in the pres-
ence of Yb(OTf)3 (10 mol %) in CH2Cl2 at room temper-
ature gave the desired C-glycoside 6 in 83% yield as the
sole product (Scheme 1).13 Subsequent asymmetric
dihydroxylation of 6 with (DHQD)2AQN14 as a chiral
ligand proceeded smoothly to afford the desired diol 7
in good yield and diastereoselectivity (80%, 8:1 dr).
Although the stereochemistry at the C4 position could
not be determined at this stage, the major product was
tentatively assigned based on the previous results.10c
Selective protection of the primary hydroxy group of 7
gave the TBS ether 8 in 87% yield. Our first attempt
to construct the bicyclic ether skeleton focused on
stereoselective epoxidation of the double bond followed
by acid-catalyzed 5-exo ring-closure.10b However,
attempted epoxidation of 8 was not successful, resulting
in recovery of the starting material. We next carried out
the formation of a tetrahydrofuran ring by palladi-
um(0)-catalyzed p-allyl ring closure.15 After some exper-
iments, it was found that the best result was obtained by
treatment of 8 with catalytic Pd2(dba)3 in the presence of
neocuproine as a ligand. Thus, the desired bicyclic ether
9 was obtained in excellent yield. Subsequent dihydroxy-
lation of 9 with OsO4/NMO produced cis-diol 10 exclu-
For the construction of the amino acid side chain, 12
was converted to enamide ester 15. After removal of
the TBS ether of 12, the resultant primary alcohol 13
was oxidized under Swern conditions and the derived
aldehyde was subjected to Horner–Wadsworth–Em-
mons (HWE) reaction using phosphonate 1416 in the
presence of tetramethylguanidine (TMG) (Scheme 2).
The requisite enamide ester 15 was obtained in 77%
yield over two steps. Subsequent asymmetric hydrogena-
tion using Burk’s (S,S)-EtDuPHOS Rh(I) catalyst17
under high pressure conditions proceeded in a highly
stereoselective manner to afford the desired amino acid
derivative 16 in 74% yield as the sole product. Finally,
the Cbz group was replaced with the Boc group (H2,
Pd/C, (Boc)2O, MeOH, 85%) to afford intermediate 17.
Selective removal of the acetonide of 17 with DDQ18
provided diol 18 in modest yield (49%), which was then

5786
M. Sasaki et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5784–5787
Table 1. Epileptogenic activity of natural dysiherbaines (1 and 2) and
derivative 3
Compound
ED50 pmol/mouse (icv)
1a
2b
3
13
16
110
a Ref. 4.
b Ref. 6.
ly different from those of the natural products 1 and 2.
Stereotyped behaviors, such as persistent scratching or
clonic convulsions, frequently observed after adminis-
tration of 1 and 2, were absent in the case of 3. Instead,
transient jumping and running behaviors were apparent.
Scheme 2. Reagents and conditions: (a) TBAF, THF, rt, 100%; (b)
(COCl)2, DMSO, Et3N, CH2Cl2, ꢀ78 ! 0 °C; (c) compound 14,
TMG, CH2Cl2, 0 °C, 77% (two steps); (d) H2 (0.8 MPa), [Rh(I)(-
COD)(S,S)-Et-DuPHOS]+OTfꢀ (5 mol %), THF, rt, 74%; (e) H2, Pd/
C, (Boc)2O, MeOH, rt, 85%.
Next, the binding affinity of 3 was evaluated with native
ionotropic glutamate receptors by radioligand binding
assays using rat synaptic membrane preparation.4 The
results are summarized in Table 2. It is noteworthy that
3 displaced [3H]AMPA more potently than [3H]kainic
acid from the receptors. Derivative 3 displaced the
[3H]AMPA with Ki value of 153 35 nM, which is com-
parable to that of the natural neodysiherbaine A (2),
whereas 3 displaced [3H]kainic acid with 15-fold less
potency than 2. As expected, 3 did not exhibit detectable
affinity for NMDA receptors.
converted to the cyclic sulfonate 19 by a two-step proce-
dure in 69% overall yield (Scheme 3).19 Treatment of 19
with cesium acetate effected regioselective substitution at
the C9 position to generate alcohol 20 after acid hydro-
lysis of the derived sulfonate ester in 85% yield over two
steps. Inversion of the hydroxy group of 20 was next
carried out by nucleophilic substitution. Thus, alcohol
20 was converted to the triflate and subsequently treated
with cesium acetate to afford diacetate 21 in 51% overall
yield. Finally, acid hydrolysis of 21 (6M HCl, 80 °C)
completed the synthesis of the targeted 10-hydroxy-
methyl-neodysiherbaine A (3) in 90% yield.20 The pres-
ent synthesis is sufficiently efficient and allows us to
prepare derivative 3 in several hundred milligrams.
Not expectedly, the affinity of 3 for kainate receptors
was attenuated significantly, whereas that for AMPA
receptors retained. This drastic shift in subtype selectiv-
ity was not predictable from our earlier model.7 Howev-
er, the present result suggested that introduction of an
additional polar group created another site for hydrogen
bonding with some amino acid residue in AMPA recep-
tors. Thus, the C10 position of dysiherbaines is another
interesting site for further modification as analogous re-
sult has been recently reported by Chamberlin and co-
workers.11
The in vivo toxicity of derivative 3 against mice was test-
ed by intracerebroventricular injection (Table 1). Com-
pound 3 induced dose-dependent behavioral changes
in mice with the ED50 value of 110 pmol/mouse, which
is 7-fold less active than that for 2 (16 pmol/mouse).
Interestingly, the behavioral profile of 3 was substantial-
In conclusion, we have developed the derivative 3 with a
hydroxymethyl substituent at the C-10 position of
neodysiherbaine A. Preliminary biological evaluation re-
vealed that the hydroxymethyl derivative 3 shows in vivo
epileptogenic activity in mice with about seven times less
potency than the natural product 2 with significant shift
in receptor selectivity. This result may suggest that the
C10 position of dysiherbaines is another interesting site
for modification. Nevertheless, this molecule could still
Table 2. Receptor binding affinities of natural dysiherbaines (1 and 2)
and derivative 3a
Compound
Ki (nM)
[3H]AMPA
[3H]kainic acid
1b
2c
3
153 11
151 32
153 35
26
52
4
4
790 100
Scheme 3. Reagents and conditions: (a) DDQ, MeCN/H2O, 40 °C,
49%; (b) SOCl2, Et3N, CH2Cl2, ꢀ20 °C; (c) RuCl3, NaIO4, CCl4/
MeCN/H2O, rt, 69% (two steps); (d) CsOAc, DMF, rt; (e) cat H2SO4,
THF, rt, 85% (two steps); (f) Tf2O, pyridine, DMAP, CH2Cl2, ꢀ20 °C;
(g) CsOAc, DMF, rt, 51% (two steps); (h) 6 M HCl, 80 °C, 90%.
a Affinities for receptors (Ki, nM) were determined by the displacement
of [3H]AMPA and [3H]kainic acid.
b Ref. 4.
c IC50 value from Ref. 6 was converted to the Ki value.

M. Sasaki et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5784–5787
5787
neodysiherbaine A, see, Ref. 6; (f) Lygo, B.; Slack, D.;
Wilson, C. Tetrahedron Lett. 2005, 46, 6629; (g) Takah-
ashi, K.; Matsumura, T.; Corbin, G. R. M.; Ishihara, J.;
Hatakeyama, S. J. Org. Chem. 2006, 71, 4227.
be considered as an interesting precursor for the probe
as only a few examples of non-RI probes for GluRs
are known. Detailed neurophysiological studies on 3
and preparation of its biotinylated or fluorescently la-
beled probes are in progress and will be reported in
due course.
9. For synthetic studies on dysiherbaine, see (a) Naito, T.;
Nair, J. S.; Nishiki, A.; Yamashita, K.; Kiguchi, T.
Heterocycle 2000, 53, 2611; (b) Huang, J.-M.; Xu, K.-C.;
Loh, T.-P. Synthesis 2003, 755; (c) Miyata, O.; Iba, R.;
Hashimoto, J.; Naito, T. Org. Biomol. Chem. 2003, 1, 772;
(d) Kang, S. H.; Lee, Y. M. Synlett 2003, 993.
Acknowledgments
10. (a) Sasaki, M.; Maruyama, T.; Sakai, R.; Tachibana, K.
Tetrahedron Lett. 1999, 40, 3195; (b) Shoji, M.; Shiohara,
K.; Oikawa, M.; Sakai, R.; Sasaki, M. Tetrahedron Lett.
2005, 46, 5559; (c) Shoji, M.; Akiyama, N.; Lash, L. L.;
Sanders, J. M.; Swanson, G. T.; Sakai, R.; Shimamoto,
K.; Oikawa, M.; Sasaki, M. J. Org. Chem. 2006, 71, 5208.
11. Cohen, J. L.; Limon, A.; Miledi, R.; Chamberlin, A. R.
Bioorg. Med. Chem. Lett. 2006, 16, 2189.
This work was financially supported by a Grant-in-Aid
for Scientific Research on Priority Area ‘Creation of
Biologically Functional Molecules’ (No. 16073202) from
the Ministry of Education, Culture, Sports, Science and
Technology, Japan.
12. Konosu, T.; Furukawa, Y.; Hata, T.; Oida, S. Chem.
Pharm. Bull. 1991, 39, 2813.
References and notes
13. Takhi, M.; Rahman, A. A.-H. A.; Schmidt, R. R.
Tetrahedron Lett. 2001, 42, 4053.
14. Becker, H.; Sharpless, K. B. Angew. Chem., Int. Ed. Engl.
1996, 35, 448.
15. For a recent example, see: Uenishi, J.; Ohmi, M. Angew.
Chem., Int. Ed. 2005, 44, 2756.
16. Schmidt, U.; Lieberknecht, A.; Wild, J. Synthesis 1984, 53.
17. (a) Burk, M. J.; Feaster, J. E.; Nugent, W. A.; Harlow, R.
L. J. Am. Chem. Soc. 1993, 115, 10125; (b) Burk, M. J.
Acc. Chem. Res. 2000, 33, 363.
1. Watkins, J. C.; Evans, R. H. Annu. Rev. Pharmacol.
Toxicol. 1981, 21, 165.
2. (a) Seeburg, P. Trends Neurosci. 1993, 16, 359; (b)
Hollmann, M.; Heinemann, S. Annu. Rev. Neurosci.
1994, 17, 31; (c) Dingledine, R.; Borges, K.; Bowie, D.;
Traynelis, S. F. Pharmacol. Rev. 1999, 51, 7.
3. Sakai, R.; Kamiya, H.; Murata, M.; Shimamoto, K. J.
Am. Chem. Soc. 1997, 119, 4112.
4. Sakai, R.; Swanson, G. T.; Shimamoto, K.; Contractor,
A.; Ghetti, A.; Tamura-Horikawa, Y.; Oiwa, C.; Kamiya,
H. J. Pharmacol. Exp. Ther. 2001, 296, 655.
5. Swanson, G. T.; Green, T.; Sakai, R.; Contractor, A.;
Che, W.; Kamiya, H.; Heinemann, S. F. Neuron 2002, 34,
589.
6. Sakai, R.; Koike, T.; Sasaki, M.; Shimamoto, K.; Oiwa,
C.; Yano, A.; Suzuki, K.; Tachibana, K.; Kamiya, H. Org.
Lett. 2001, 3, 1479.
7. Sanders, J. M.; Ito, K.; Settimo, L.; Pentikainen, O. T.;
Shoji, M.; Sasaki, M.; Jonson, M. S.; Sakai, R.; Swanson,
G. T. J. Pharmacol. Exp. Ther. 2005, 314, 1068.
8. For total synthesis of dysiherbaine, see (a) Snider, B. B.;
Hawryluk, N. A. Org. Lett. 2000, 2, 635; (b) Sasaki, M.;
Koike, T.; Sakai, R.; Tachibana, K. Tetrahedron Lett.
2000, 41, 3923; (c) Masaki, H.; Maeyama, J.; Kamada, K.;
Esumi, T.; Iwabuchi, Y.; Hatakeyama, S. J. Am. Chem.
Soc. 2000, 122, 5216; (d) Phillips, D.; Chamberlin, A. R. J.
Org. Chem. 2002, 67, 3194; (e) For total synthesis of
18. (a) Fernandez, J. M. G.; Mellet, C. O.; Martin, A. M.;
Fuentes, J. Carbohydr. Res. 1995, 274, 263; (b) Tu, Y.;
Wang, Z.-X.; Frohn, M.; He, M.; Yu, H.; Tang, Y.; Shi,
Y. J. Org. Chem. 1998, 63, 8475; (c) Tian, H.; She, X.; Yu,
H.; Shu, L.; Shi, Y. J. Org. Chem. 2002, 67, 2435.
19. For a review of cyclic sulfates, see: Byun, H.-S.; He, L.;
Bittman, R. Tetrahedron 2000, 56, 7051.
20
1
20. Data for compound 3: ½aꢁD +1.4 (c 0.07, H2O); H NMR
(600MHz, D2O) d 4.28 (br s, 1H), 4.14 (br s, 1H), 3.94 (m,
1H), 3.80 (dd, J = 3.6, 3.6 Hz, 1H), 3.68 (dd, J = 12.0,
9.6 Hz, 1H), 3.57 (br m, 1H), 3.49 (d, J = 12.0 Hz, 1H),
3.48 (dd, J = 12.6, 1.8 Hz, 1H), 2.58 (dd, J = 15.0, 1.8 Hz,
1H), 2.50 (d, J = 14.4 Hz, 1H), 2.12 (dd, J = 14.4, 3.6 Hz,
1H), 1.87 (dd, J = 15.0, 12.6 Hz, 1H); 13C NMR
(125MHz, D2O/CD3OD = 15:1) d 179.4, 173.2, 87.1,
82.6, 82.5, 73.3, 69.0, 66.1, 59.4, 53.2, 46.4, 39.8; HRMS
(FAB) calcd for C12H18NO9 [(MꢀH)ꢀ] 320.0982, found
320.0982.