Design and synthesis of skeletal analogues of gambierol: Attenuation of amyloid-β and tau pathology with voltage-gated potassium channel and N-Methyl-d-aspartate receptor implications

Article abstract of DOI:10.1021/ja300565t

Gambierol is a potent neurotoxin that belongs to the family of marine polycyclic ether natural products and primarily targets voltage-gated potassium channels (K_v channels) in excitable membranes. Previous work in the chemistry of marine polycyclic ethers has suggested the critical importance of the full length of polycyclic ether skeleton for potent biological activity. Although we have previously investigated structure-activity relationships (SARs) of the peripheral functionalities of gambierol, it remained unclear whether the whole polycyclic ether skeleton is needed for its cellular activity. In this work, we designed and synthesized two truncated skeletal analogues of gambierol comprising the EFGH- and BCDEFGH-rings of the parent compound, both of which surprisingly showed similar potency to gambierol on voltage-gated potassium channels (K_v) inhibition. Moreover, we examined the effect of these compounds in an in vitro model of Alzheimer's disease (AD) obtained from triple transgenic (3xTg-AD) mice, which expresses amyloid beta (Aβ) accumulation and tau hyperphosphorylation. In vitro preincubation of the cells with the compounds resulted in significant inhibition of K⁺ currents, a reduction in the extra- and intracellular levels of Aβ, and a decrease in the levels of hyperphosphorylated tau. In addition, pretreatment with these compounds reduced the steady-state level of the N-methyl-d-aspartate (NMDA) receptor subunit 2A without affecting the 2B subunit. The involvement of glutamate receptors was further suggested by the blockage of the effect of gambierol on tau hyperphosphorylation by glutamate receptor antagonists. The present study constitutes the first discovery of skeletally simplified, designed polycyclic ethers with potent cellular activity and demonstrates the utility of gambierol and its synthetic analogues as chemical probes for understanding the function of K_v channels as well as the molecular mechanism of Aβ metabolism modulated by NMDA receptors.

Full text of DOI:10.1021/ja300565t

Article
pubs.acs.org/JACS
Design and Synthesis of Skeletal Analogues of Gambierol:
Attenuation of Amyloid-β and Tau Pathology with Voltage-Gated
Potassium Channel and N-Methyl-D-aspartate Receptor Implications
Eva Alonso,† Haruhiko Fuwa,‡ Carmen Vale,† Yuto Suga,‡ Tomomi Goto,‡ Yu Konno,‡ Makoto Sasaki,‡
Frank M. LaFerla,∥ Mercedes R. Vieytes,§ Lydia Gimenez-Llort,⊥ and Luis M. Botana*,†
́
†Departamento de Farmacología, Facultad de Veterinaria, Universidad de Santiago de Compostela, Lugo, Spain
‡Graduate School of Life Sciences, Tohoku University, 2-1-1 Katahira, Aoba-ku, Sendai 980-8577, Japan
§Departamento de Fisiología, Facultad de Veterinaria, Universidad de Santiago de Compostela, 27003 Lugo, Spain
∥Department of Neurobiology and Behavior, University of California, Irvine, Irvine, California 92697, United States
⊥Departamento de Psiquiatría y Medicina Legal, Instituto de Neurociencias, Universidad Autonoma de Barcelona,
08193 Bellaterra, Spain
S
* Supporting Information
ABSTRACT: Gambierol is a potent neurotoxin that belongs to
the family of marine polycyclic ether natural products and
primarily targets voltage-gated potassium channels (Kv channels)
in excitable membranes. Previous work in the chemistry of marine
polycyclic ethers has suggested the critical importance of the full
length of polycyclic ether skeleton for potent biological activity.
Although we have previously investigated structure−activity
relationships (SARs) of the peripheral functionalities of gambierol,
it remained unclear whether the whole polycyclic ether skeleton is
needed for its cellular activity. In this work, we designed and synthesized two truncated skeletal analogues of gambierol
comprising the EFGH- and BCDEFGH-rings of the parent compound, both of which surprisingly showed similar potency to
gambierol on voltage-gated potassium channels (Kv) inhibition. Moreover, we examined the effect of these compounds in an in
vitro model of Alzheimer’s disease (AD) obtained from triple transgenic (3xTg-AD) mice, which expresses amyloid beta (Aβ)
accumulation and tau hyperphosphorylation. In vitro preincubation of the cells with the compounds resulted in significant
inhibition of K+ currents, a reduction in the extra- and intracellular levels of Aβ, and a decrease in the levels of
hyperphosphorylated tau. In addition, pretreatment with these compounds reduced the steady-state level of the N-methyl-D-
aspartate (NMDA) receptor subunit 2A without affecting the 2B subunit. The involvement of glutamate receptors was further
suggested by the blockage of the effect of gambierol on tau hyperphosphorylation by glutamate receptor antagonists. The present
study constitutes the first discovery of skeletally simplified, designed polycyclic ethers with potent cellular activity and
demonstrates the utility of gambierol and its synthetic analogues as chemical probes for understanding the function of Kv
channels as well as the molecular mechanism of Aβ metabolism modulated by NMDA receptors.
their neurological symptoms resemble those shown by ciguatoxins,
the principal responsible toxin for ciguatera seafood poisoning,
although the natural scarcity of gambierol has hampered for many
years further detailed investigations into its biological activity.
However, recent total syntheses by several groups have made
sufficient quantities of synthetic material available.5−8 Moreover,
preliminary structure−activity relationship (SAR) investigations
involving in vivo evaluation of synthetic analogues have shown
that both the triene side chain and the H-ring functional groups
are essential for the potent toxicity.9 However, there remains an
important question whether the ladder-shaped polycyclic ether
skeleton as a whole is necessary for the biological activity.
INTRODUCTION
■
Ever since the isolation and structure determination of brevetoxin-
B by Nakanishi and his colleagues,1 marine polycyclic ether natural
products have continuously been fascinating to the scientific
community due to their extraordinary complex molecular archi-
tecture and diverse and potent biological activities.2 The re-
presentative members of the family of marine polycyclic ether
natural products are brevetoxins (PbTxs), ciguatoxins,
maitotoxin, and gambierol, which are known to exhibit high
neurotoxicity against mammalians.3 Gambierol (1, Figure 1),
one of the toxic secondary metabolites produced by the
dinoflagellate Gambierdiscus toxicus, was isolated and structurally
characterized by Satake, Yasumoto, and co-workers.4 The Satake/
Yasumoto group has reported that gambierol exhibits potent acute
lethal toxicity against mice (minimal lethal dose: 50 μg/kg, ip) and
Received: January 18, 2012
Published: April 4, 2012
© 2012 American Chemical Society
7467
dx.doi.org/10.1021/ja300565t | J. Am. Chem. Soc. 2012, 134, 7467−7479

Journal of the American Chemical Society
Article
Aβ deposits.16 The Aβ plaques appear in determinate brain
regions and produce neuronal death, inflammatory response
and a progressive cognitive failure.17 On the other hand,
neurofibrillary tangles are formed predominantly from hyper-
phosphorylated isoforms of the tau protein, a microtubule-
associated protein.18 In normal cells, tau function is to promote
the polymerization of tubulin for the formation of microtubules.19
Phosphorylation of tau can alter the correct axonal transport and
neuronal structure, leading to neuronal degeneration.20
The cellular effects of gambierol on Kv channels and NMDA
receptors21 as well as the involvement of NMDA receptors in
APP expression and Aβ production led us to pursue the
following three objectives in this work: (i) design and synthesis
of truncated skeletal gambierol analogues, (ii) comparison of
the effects of gambierol and its skeletal analogues on Kv
channels, and (iii) analysis of the effects of gambierol and
analogues in an in vitro model of AD obtained from 3xTg-AD
mice with simultaneous overexpression of Aβ and hyper-
phosphorylation of tau.22
RESULTS
■
Chemical Synthesis of Gambierol Analogues. The
synthesis of the heptacyclic analogue 2 (Figure 1) commenced
with reduction of the known ester 423 that corresponds to the
C-ring of 1 (Scheme 1). Wittig methylenation of the derived
aldehyde and removal of the silyl group gave alcohol 6. For
the construction of the B-ring tetrahydropyran, ring-closing
metathesis24 of vinyl ether 7 was envisioned. However, viny-
lation of sterically encumbered tertiary alcohol 6 turned out to
be nontrivial, as the Hg(II)-mediated25 or Pd(II)-catalyzed26
transetherifications only gave disappointing results. We found
that iridium-catalyzed vinylation protocol developed by Ishii
and co-workers27 worked well with 6 to deliver 7, which upon
exposure to the Grubbs second-generation catalyst (G-II)28
(benzene, 60 °C) led to dihydropyran 8. A slight modification
of the Ishii conditions was required to achieve satisfactory
conversion in the vinylation process. Hydrogenation/hydro-
genolysis of 8 followed by acetalization provided p-methox-
ybenzylidene acetal 9. Regioselective reductive opening of the
acetal, iodination, and subsequent base treatment afforded
exocyclic enol ether 11.
Toward the construction of the heptacyclic polyether
skeleton, we exploited the SuzukiMiyaura coupling-based
methodology developed in our laboratory.29,30 Hydroboration
of 11 with 9-BBN-H delivered alkylborane 12, which without
isolation was reacted with the EFGH-ring enol phosphate 135
in the presence of aqueous Cs2CO3 and PdCl2(dppf)·CH2Cl2
catalyst to furnish enol ether 14 in 88% yield. Hydroboration of
14 with BH3·SMe2 followed by oxidative workup gave an inseparable
3:1 mixture of diastereomeric alcohols, which was oxidized31 to
provide ketone 15, after removal of the minor diastereomer
(not shown) by flash chromatography on silica gel. The
stereochemistry of 15 was unambiguously established by an
NOE experiment as shown. Deprotection of the p-methox-
yphenylmethyl (MPM) group, mixed thioacetalization,32 and
acylation of the liberated hydroxy groups afforded mixed
thioacetal 16. Unexpectedly, stereoselective reduction of mixed
thioacetal 16 proved to be a challenging task. Reduction of 16
under tin hydride conditions (n-Bu3SnH, AIBN, toluene, 100 °C)
gave heptacycle 17 in 57% yield, along with the undesired epimer
(not shown) in 24% yield (12% recovery of 16). On the other
hand, oxidation of mixed thioacetal 16 with m-CPBA followed by
in situ treatment with Et3SiH/BF3·OEt233 resulted in a complete
Figure 1. Structures of gambierol 1, heptacyclic analogue 2, and
tetracyclic analogue 3.
Voltage-gated potassium channels (Kv) have been shown to
be the primary molecular targets of the toxin.10 Among the
potassium channel subtypes tested, the evaluation of 1 in
mammalian Kv1.1-Kv1.6, hERG, and insect Shaker IR subtypes
showed Kv1.2 channels as the most sensitive one with an IC50
of about 15 nM and a 98% of inhibition10 and did not affect
TRPV1 channels. However, the compound also exhibited mod-
erate inhibitory activity against voltage-gated sodium channels
(Nav) and modified calcium homeostasis,11,12 but it did not
have effect over the mammalian Nav1.1-Nav1.8 channels and
insect Para at concentrations up to 1 μM, nor on sodium cur-
rents in taste cells at concentrations in the nanomolar range.10
Our previous work has shown that gambierol induced
synchronous calcium oscillations in the presence of extracellular
calcium ion in primary cerebellar neurons, which were com-
pletely abolished by the selective NMDA receptor antagonist D-
(−)-2-amino-5-phosphonopentanoate (APV) and by the
glutamate release inhibitor riluzole.13 Similar effects were
observed with 4-aminopyridine (4-AP) that has been shown
to modulate the function of NMDA receptors via inhibition of
Kv channels. Thus, our results indicated that gambierol-evoked
Ca2+ oscillations in cerebellar neurons would involve mod-
ulation of NMDA receptors, which are most likely secondary to
Kv channel inhibition.
Meanwhile, recent studies on the mechanism of the amyloid-
β (Aβ) metabolism, possibly responsible for the molecular
pathology of Alzheimer’s disease (AD), demonstrated that
NMDA receptors play a significant role in the regulation of
amyloid-β precursor protein (APP) expression and Aβ
production.14 Histopathologically, AD is characterized by two
main hallmarks, deposition of insoluble Aβ peptide in senile
plaques and intracellular neurofibrillary tangles derived from
hyperphosphorylated tau proteins.15 Aβ is produced by the
abnormal proteolytic cleavage of APP, principally by two
enzymes, β-secretase (BACE) and γ-secretase (presenilin 1 and 2).
These enzymes sequentially cleave APP in the N- and
C-terminal end, respectively, generating 39−43 amino acid
polypeptides with limited solubility, which aggregate and form
7468
dx.doi.org/10.1021/ja300565t | J. Am. Chem. Soc. 2012, 134, 7467−7479

Journal of the American Chemical Society
Article
Scheme 1. Synthesis of the Heptacyclic Analogue 2
diastereoselective reduction to give 17 in approximately 73% yield,
albeit an unidentified byproduct contaminated that could not be
separated until the end of the synthesis. Eventually, we resorted to
a new method for the reduction of mixed thioacetal 16. Thus,
activation of 16 with N-iodosuccinimide (NIS)/AgOTf34 in the
presence of Et3SiH (CH2Cl2, −40 °C) cleanly afforded 17 in 90%
yield as a single stereoisomer. The stereochemistry of 17 was
determined by an NOE experiment as shown.
7469
dx.doi.org/10.1021/ja300565t | J. Am. Chem. Soc. 2012, 134, 7467−7479

Journal of the American Chemical Society
Article
Scheme 2. Synthesis of the Tetracyclic Analogue 3
Figure 2. Effect of bath application of 1, 2, or 3 on potassium currents in nontreated neurons. (A−C) Representative electrophysiological recordings
showing the concentration−response effect 5 min after administration of 1 (A), 2 (B), and 3 (C) to the extracellular solution over the IK and IDR
components of K+ currents in 3xTg-AD neurons at concentrations of 0.1, 1, 10, 100, 500, and 1000 nM. The holding potential was −80 mV.
Recordings are from the same cell with the different concentrations of each compound. (D) Quantitative analysis of the inhibition of IK and IDR
current densities (pA/pF) in the presence of 1, 2, or 3 at 100 nM in NonTg and 3xTg-AD neurons. Results are mean SEM of 6−14 cells.
Removal of the acetyl groups and selective silylation of the
liberated primary alcohol gave alcohol 18, which was oxidized
to provide ketone 19. The double bond of the H-ring was
introduced according to the SaegusaIto protocol35 to yield
enone 20, which was reacted with MeMgBr36 to deliver alcohol
21 as a single stereoisomer. The stereochemistry of 21 was
7470
dx.doi.org/10.1021/ja300565t | J. Am. Chem. Soc. 2012, 134, 7467−7479

Journal of the American Chemical Society
Article
Figure 3. Chronic treatments with 1, 2, or 3 reduced the overexpression of the Kv3.1 subunit and the amplitude of potassium currents. (A)
Representative Western blot and the corresponding histogram showing Kv3.1 expression levels in NonTg, 3xTg-AD, and 3xTg-AD neurons treated
with gambierol and the two analogues as measured with the Kv3.1 antibody which recognizes a full-length Kv3.1 protein and does not cross react
with any other potassium channel antigens. Results are mean SEM of 5 experiments, each performed in duplicate. (B) Effect of long-term exposure
from 3 to 7 div to 1, 2, or 3 in the amplitude of IK and IDR currents in 3xTg-AD cortical neurons. (C) Effect of long-term exposure from 3 to 7 div to
1, 2, or 3 in the amplitude of IK and IDR currents in NonTg neurons. Results are mean SEM of 6−14 cells. *p < 0.05 and ***p < 0.001.
established by an NOE experiment. Silylation followed by
selective cleavage of the primary silyl ether under acidic con-
ditions gave alcohol 23, which was oxidized and then converted
to dibromoolefin 24. Selective reduction of 24 (n-Bu3SnH,
Pd(PPh3)4)37 led to (Z)-vinyl bromide 25, which was
desilylated with HF·pyridine to provide alcohol 26. Finally,
Stille coupling38 of 26 with vinyl stannane 27 (Pd(PPh3)4, CuCl,
LiCl, DMSO/THF, 60 °C)9 furnished heptacyclic analogue 2.
The synthesis of the tetracyclic analogue 3 (Figure 1) started
with the known ester 285 that represents the FGH-ring
framework of 1 (Scheme 2). DIBALH reduction of 28 and
ensuing Wittig methylenation gave olefin 29, which was
transformed to diene 30 via desilylation and allylation. Ring-
closing metathesis of 30 proceeded cleanly in the presence of
the Grubbs first-generation catalyst (G-I)40 to deliver oxepene 31.
Hydrogenation of 31 gave pentacycle 32 in nearly quanti-
tative yield. Removal of the acetonide under acidic conditions,
selective silylation of the unmasked primary hydroxy group, and
subsequent oxidation led to ketone 33. This was elaborated to
tetracyclic analogue 3 in a similar manner as that described for 2.
Effect of Gambierol and Its Synthetic Analogues on
Voltage-Gated Potassium Channels. With 2 and 3 in hand
(Figure 1), we evaluated their cellular activity and compared it
to that of 1. Since the effect of acute administration of 1 and
synthetic analogues on K+ currents was not previously
evaluated in cortical neurons, we first analyzed the effect of
the acute application of these compounds in this neuronal
model. As shown in Figure 2A−C, application of gambierol or
analogues to primary cortical neurons produced a concen-
tration-dependent inhibition of K+ current densities. Thus, 1, 2,
and 3 at 100 nM inhibited IK currents in 3xTg-AD neurons to a
69.4 5.1%, 64.8 4.0%, and 56.1 5.0%, respectively, versus
control currents, and the IDR component to a 75.0 7.1%, 63.6
8.1%, and 68.5 7.0%. Similar results were found in NonTg
cortical neurons where, at the same concentration, IK current
density was inhibited to a 75.7 2.0%, 70.7 5.0%, and 68.7
6.1% by 1, 2, and 3, respectively, while the IDR component
was blocked to a 74.6
6.0%, 75.7
9.6%, and 67.3
9.0%, respectively, versus control currents as summarized in
Figure 2D.
7471
dx.doi.org/10.1021/ja300565t | J. Am. Chem. Soc. 2012, 134, 7467−7479

Journal of the American Chemical Society
Article
Figure 4. Gambierol (1) and its analogues 2 and 3 decreased intra- and extracellular accumulation of Aβ. (A) Confocal microscopy images of the
immunoreactivity for the 6E10 antibody, which reacts with abnormally processed isoforms and precursor forms of the protein, employed to measure
intracellular Aβ levels and corresponding quantitative analysis in NonTg, 3xTg-AD, and 3xTg-AD neurons treated with 1, 2, or 3 from third to
seventh div. Scale bar is 20 μm. (B) Representative Western blot bands indicating intracellular Aβ levels in NonTg neurons, 3xTg-AD neurons, and
3xTg-AD neurons treated with 1 and probed with the 6E10 antibody. Quantification of Western blot band intensities for Aβ levels as obtained from
3 independent experiments, performed in duplicate. ***p < 0.0001 versus 3xTg-AD neurons. (C) Quantitative analysis of the levels of extracellular
Aβ measured in the culture medium after treatment of the neurons with 1, 2, or 3 from the third to the seventh div. Results are mean SEM of 3−5
independent experiments each performed in duplicate. *p < 0.05, **p < 0.01 and ***p < 0.001.
Cytotoxic Evaluation of Gambierol and Its Synthetic
Analogues in Cortical Neurons. Previous work in our
laboratory has shown that 1 modified the intracellular calcium
concentration in a way dependent on glutamate signaling and
that the toxin also inhibited K+ currents in cerebellar granule
neurons.13 Glutamate signaling is related to neurodegenerative
diseases, and glutamate receptors and their excitotoxic effects
are being widely studied in AD.41 Therefore, the effect of 1, 2,
and 3 was evaluated in an in vitro model of primary cortical
neurons obtained from 3xTg-AD mice, a model known to
overexpress Aβ and hyperphosphorylated tau.22 First, the in vitro
toxicity of the compounds was evaluated by the MTT assay after
exposing neuronal cultures to the compounds from 3 to 7 days in
vitro. 1 and 3 did not show any cytotoxicity even at 10 μM
79.5
effects at concentrations below 500 nM, showing a viability of
124.4 8.1% with respect to control cells. In the following
experiments, we used nontoxic concentrations of the com-
pounds. Thus, 3xTg-AD neurons were pretreated with 1 at
10 μM, 2 at 100 nM, and 3 at 5 μM. Treatments were made in
the culture medium from the third div (days in vitro) to the
seventh div. Control cells were treated with the corresponding
concentration of the vehicle, DMSO.
Long-Term Exposure of Cortical Neurons to Gambier-
ol and Its Truncated Skeletal Analogues Decreases
Voltage-Gated Potassium Currents. Since it has been
observed that acute application of 1 at nanomolar concen-
trations inhibited K+ currents and that the toxin strongly
suppressed Kv3.1 dependent currents with higher potency than
Kv2 or Kv4 currents,21 we analyzed the long-term effects of the
exposure of cortical neurons to 1 on Kv3.1 expression and K+
10.0% for 0.5 μM. However, it showed no cytotoxic
(105.8
control), respectively. In contrast, 2 showed a decrease in cellular
viability, a 46.2 0.5% of cellular viability for 1 μM and a
6.27% of control) and at 5 μM (101.5
4.6% of
7472
dx.doi.org/10.1021/ja300565t | J. Am. Chem. Soc. 2012, 134, 7467−7479

Journal of the American Chemical Society
Article
Figure 5. Chronic exposure of 3xTg-AD neurons to 1, 2, or 3 decreased tau phosphorylation. (A) Representative Western blot bands probed with
the AT8 antibody (which recognizes a tau epitope doubly phosphorylated at serine 202 and threonine 205 and does not cross-react with normal tau
sequences) in control, 3xTg-AD, and 3xTg-AD treated neurons. Quantitative analysis of AT8 levels in control, 3xTg-AD neurons, and 3xTg-AD
treated neurons. (B) Representative Western blot bands indicating phospho-tau levels in control, 3xTg-AD, and 3xTg-AD treated neurons probed
with the AT100 antibody (which recognizes tau phosphorylated at serine 212 and threonine 214). Quantification of AT100 levels in control, 3xTg-
AD neurons, and 3xTg-AD treated neurons. Results are mean SEM of 3 experiments, each performed in duplicate. **p < 0.01 versus 3xTg-AD
neurons. ***p < 0.001 versus 3xTg-AD neurons.
current amplitude in this 3xTg-AD primary cortical model. As
shown in Figure 3A, 3xTg-AD neurons in culture presented an
overexpression of the Kv3.1 channel. Immunoblotting analysis
with a specific antibody for this Kv subtype on cellular lysates
and its corresponding densitometry showed a significant in-
crease in the band intensity corresponding to 3xTg-AD cortical
neurons lysates versus control NonTg neurons (control, 100
23.6%; 3xTg-AD, 290.3 35.2% (p = 0.0012)), as shown in
Figure 3A. Treatment of the 3xTg-AD neurons with 1, 2, or 3
from the third to the seventh div decreased the Kv3.1 up-
regulation (Figure 3A). Next, electrophysiological recordings
were performed to evaluate the functional relevance of
potassium channels in the in vitro AD model. The IK current
in 3xTg-AD neurons was partially abolished by the skeletal
analogues but not significantly by 1. As shown in Figure 3B, in
3xTg-AD neurons treated with 3, the total outward current IK
was 26.3 6.0% smaller than that in nontreated neurons and
22.4 4.5% for 2, while the inhibition of IK by 1 was 16.8
9.1%. On the other hand, exposure of cortical neurons to 1, 2,
or 3 showed a similar effect on the IDR component. Thus, in
neurons treated with 1, IDR was 47.5 6.8% smaller than that
in nontreated 3xTg-AD neurons, whereas in 3xTg-AD neurons
cultured in the presence of 2 or 3, IDR was decreased by 32.4
6.5% or 22.5 5.1%, respectively. In agreement with the
overexpression of Kv3.1 channels in 3xTg-AD neurons observed
by the Western blot experiments, the size of K+ currents was
smaller in NonTg cells with respect to 3xTg-AD. Thus, the
amplitude of the IK current was 607.8 44.1 pA in nontreated
3xTg-AD neurons and 471.3 79.6 pA in nontreated NonTg
neurons, while for the IDR component, the amplitude of the
current was 450.6 39.6 pA in 3xTg-AD neurons and 316.2
37.7 pA in NonTg cells. Unexpectedly, only long-term exposure
to 1 inhibited the IDR current in normal neurons, whereas the
total outward current was unaffected (Figure 3C).
Gambierol Effects in Aβ and Tau Cellular Pathologies.
In view of the effects of 1−3 on the expression of Kv3.1
channels and on the K+ currents, we analyzed if these com-
pounds could affect the intra- and extracellular Aβ accumu-
lation levels observed in 3xTg-AD neurons in culture. Western
blot and confocal microscopy employing the 6E10 antibody
were used for quantifying intracellular Aβ levels. As shown in
Figure 4A, all the compounds significantly reduced the
immunoreactivity of 3xTg-AD neurons to 6E10 antibody. To
confirm these data, Western blot experiments were performed
7473
dx.doi.org/10.1021/ja300565t | J. Am. Chem. Soc. 2012, 134, 7467−7479

Journal of the American Chemical Society
Article
after treatment of cortical neurons with 1. As shown in Figure 4B,
cortical 3xTg-AD cultures grown in the presence of 1 from the
third to seventh div, showed a significant decrease in
intracellular Aβ accumulation after treatment with 1 (45.2
9.3%; p = 0.003). Since extracellular Aβ40−42 is also increased in
3xTg-AD neurons in culture,22 we checked if the intracellular
decrease of Aβ was accompanied with a decrease in Aβ
accumulation in the extracellular medium. Collected samples of
culture medium with and without compounds were analyzed by
BetaMark x-42 ELISA kit (SIGNET), a colorimetric ELISA kit.
As it can be seen in Figure 4C, the compounds slightly reduced
the amount of Aβ40−42 in the extracellular culture medium,
that is, 1 in 23.0
10.1% (p = 0.003), 2 in 26.18
1.8%
(p < 0.001), and 3 in 15.7 3.1% (p < 0.01).
Inhibition of Aβ accumulation delays the progression of tau
pathology in vivo.42 Moreover, Aβ oligomers can stimulate tau
phosphorylation in vitro43 and Aβ and tau can also form
complexes that enhance tau phosphorylation by GSK-3β.44
Therefore, we evaluated whether gambierol and its analogues
could modify tau expression. As shown in Figure 5, all the
compounds decreased the immunoreactivity levels of AT8 and
AT100 antibodies by 30−50% after exposure of 3xTg-AD
cultures to the compounds. AT8 recognizes tau proteins
phosphorylated at Ser 202, while AT100 binds to tau proteins
phosphorylated at Thr 212 and Ser 214. These are the two of
representative antibodies used in the study of tau hyper-
phosphorylation in AD. As shown in Figure 5A, treatment of
3xTg-AD cortical cultures with 1, 2, or 3 decreased
phosphorylated tau levels for the two antibodies. For AT8,
the immunoreactivity decreased in 51.2 11.4% (p < 0.001),
Figure 6. Effect of 1, 2, and 3 on glutamate-induced neurotoxicity and
the calcium response elicited by glutamate. (A) Chronic exposure of
cortical neurons to 10 μM of 1 did not modify the calcium response
elicited by addition of 50 μM glutamate (indicated by the arrow) when
compared with the response evoked by glutamate in nontreated
neurons. (B) Glutamate-induced toxicity in primary cortical neurons
from control and 3xTg-AD mice was not affected by pretreatment of
the neurons with either 1, 2, or 3 as evaluated by the MTT reduction
28.3
8.2% (p < 0.01), and 26.6
6.3% (p < 0.05) by
assay. *p < 0.05 versus NonTg cells. Data are mean
SEM of 3
exposure to 1, 2, and 3, respectively. For AT100, the decreases
were 45.4 11.4% (p = 0.0037), 56.5 19% (p < 0.05), and
42.4 10.4% (p = 0.0034) for 1, 2, and 3, respectively (Figure 5B).
Involvement of Glutamate Signaling on Gambierol
Effects against Aβ and Tau Pathologies. We have
previously shown that, in cerebellar granule neurons, acute
administration of 1 elicited calcium oscillations that were
abolished by the application of APV, a specific NMDA receptor
antagonist, and by riluzole, a glutamate release inhibitor,
suggesting that glutamate signaling was involved in the effect
of 1.13 To evaluate the involvement of glutamate signaling in
the beneficial effect of 1 against Aβ and tau pathology, we
examined the effects of the pretreatment of 3xTg-AD cortical
neurons with 1 in the intracellular calcium response elicited by
application of 50 μM glutamate. In 3xTg-AD neurons, long-
term exposure to 1 showed no difference in the glutamate
induced calcium response respect to 3xTg-AD nontreated neurons
(Figure 6A). In addition, the effect of 1−3 on glutamate excito-
toxicity was evaluated by adding 100 μM glutamate to the culture
medium after a 24 h-pretreatment with 1, 2, or 3. Thus, the cells
were incubated for additional 48 h and then cellular viability was
evaluated by the MTT test. As expected, glutamate application
reduced cellular viability by 25.4 6.8%,45,46 but neither 1, 2, nor
3 modified the decrease in cell viability produced by glutamate in
control NonTg and 3xTg-AD neurons (Figure 6B).
independent experiments, each performed in duplicate.
induced a decrease in band intensity obtained by Western blot
with the NMDA receptor subunit N2A antibody. Although no
differences in N2A expression between control and 3xTg-AD
neurons were found, a diminishment in N2A expression of 37.7
2.9% was found after exposure of cortical neurons to 1. Similar
decreases of N2A expression were observed with 2 by 33.08
15.5% (p = 0.09) and with 3 by 43.1
7.6% (p < 0.05).
Moreover, no significant differences in the expression of the
NMDA receptor subunit N2B were found between control and
3xTg-AD neurons and none of the compounds affected the N2B
subunit expression as shown in Figure 7B. Also, the compounds
did not induce any changes in the mGlu5 receptor expression as
shown in Figure 7C, where the upper band corresponds to the
receptor dimers.47 The decrease in the expression of the NMDA
2A subunit induced by 1−3 suggest the involvement of this
receptor in the gambierol effects. Moreover, our previous work
had shown a relationship between NMDA receptors and the
effects of 1 over intracellular calcium concentration.13 In fact, we
have previously shown that the calcium oscillations induced by 1
and 4-AP in cerebellar neurons occurred mainly through
NMDA receptor activation. Although we cannot rule out the
possibility of direct interaction of 1 with glutamate receptors, it is
known that potassium channel blockers can stimulate NMDA
receptors.14,48 Therefore, we investigated if the co-treatment with
the well-known NMDA receptor antagonist APV could block the
observed effects. For this purpose, 3xTg-AD primary neurons were
pretreated with APV at 100 μM either alone or in combination with
1, and also with 20 μM of 6-cyano-7-nitroquinoxaline-2,3-dione
(CNQX), an AMPA/kainate receptor antagonist, to evaluate the
Previous studies in our laboratory have shown that calcium
effects induced by 1 were abolished by NMDA modulators.13
Therefore, we also investigated if the cellular expression of the
NMDA receptor and the closely linked mGlu5 receptor
(metabotropic glutamate receptor 5) were affected by long-term
exposure of cortical 3xTg-AD neurons to 1, 2, or 3. As shown in
Figure 7A, pretreatment of 3xTg-AD cells with the compounds
7474
dx.doi.org/10.1021/ja300565t | J. Am. Chem. Soc. 2012, 134, 7467−7479

Journal of the American Chemical Society
Article
Figure 7. Chronic treatment with 1 induced changes in glutamate receptors expression. (A) Representative experiment showing Western blot bands
for NMDA 2A subunit levels in control, 3xTg-AD, and treated 3xTg-AD neurons. Quantitative analysis of the effect on NMDA subunit 2A levels as
obtained from 3 independent experiments. (B) Representative Western blot bands showing the effect of 1, 2, and 3 in the steady-state level of
NMDA 2B subunit. Quantification of the data as obtained from five independent experiments indicating no effect of chronic pretreatment on N2B
expression. (C) Representative experiment showing Western blot bands for mGlu5 levels in control, 3xTg-AD, and treated 3xTg-AD neurons.
Quantitative analysis showing no effect of the pretreatments in mGlu5 levels expression as obtained from 3 independent experiments each performed
in duplicate. (D) Representative Western blot bands for AT8 immunoreactivity (tau phosphorylated at Ser 199 and Ser 202) showing that
pretreatment with APV but not with CNQX blocked the effect of 1 over AT8 levels expression as obtained from 3 independent experiments.
possible interaction of these glutamate receptors. In fact, the
extracellular presence of APV fully blocked the effect of 1
over AT8 expression, while CNQX failed to block the action
of 1 (Figure 7D), pointing to NMDA receptor implications
in the effects of 1.
DISCUSSION
■
The past decade has seen significant advances in total synthesis
of marine polycyclic ether toxins.49 However, skeletal SARs of
marine polycyclic ether natural products have been underex-
plored because chemoselective modification/alteration of their
7475
dx.doi.org/10.1021/ja300565t | J. Am. Chem. Soc. 2012, 134, 7467−7479

Journal of the American Chemical Society
Article
Table 1. List of Antibodies and Dilutions Used
antibody
6E10
immunogen
host
dilution
source
Signet
Aa 1−16 of Aβ
mouse
mouse
mouse
rabbit
mouse
rabbit
mouse
1:1000
1:1000
1:1000
1:1000
1:500
AT8
Peptide with phospho-S199/S202/T205
Peptide with phospho-S212/T214
C-terminal fusion protein rat NMDAR2A aa.1253−1391
NMDAR2B aa. 892−1051
Pierce
AT100
Pierce
NMDA 2A
NMDA 2B
mGluR5
Actin
Millipore
BD Bioscience
Millipore
Shyntetic peptide from rat mGluR5
1:1000
1:20000
C-terminal actin fragment, clone C4
Millipore
highly complicated skeletal structure is assumed to be virtually
impossible. In this context, “diverted total synthesis”50 is be-
lieved to be the only way to access designed skeletal analogues.
Unfortunately, drastic simplification of the skeletal structure of
marine polycyclic ether natural products without affecting their
biological function has been a significant challenge at the
interface of chemistry and biology. In fact, previous studies have
not been successful in yielding truncated analogues with cellular
activity.51−56
It is important to remark that we show for the first time that
cortical 3xTg-AD neurons endogenously overexpress Kv3.1
channels, which may be a consequence of the intra- and
extracellular Aβ overexpression in these cultures. In this sense,
it has been shown that extracellular administration of Aβ1−42 in
human microglia and hippocampal primary neurons produces
and upregulates Kv3.1 and Kv3.4 channels.57,58 Yu and
colleagues showed that treatment of mouse cortical neurons
with Aβ25−35 or Aβ1−42 fragments caused an enhancement of
the outward K+ current, with a special upregulation of the
delayed current and that the specific IDR blocker tetraethy-
lammonium (TEA) was able to protect these cells from Aβ
induced cell death.59
Kopljar et al. showed that 1 μM of gambierol (1, Figure 1)
had no effect on Kv2 or Kv4 channels but it fully suppressed
Kv3.1 currents, showing a nanomolar affinity to Kv3.1 and Kv3.3
channels in transfected mouse fibroblasts.21 Furthermore, they
also profoundly studied the binding of 1 to the chimeric Kv2.1/
Kv3.1 channels and found that 1 stabilized the closed state of
potassium channels without acting as an external pore blocker
or as an internal cavity blocker. It seems that 1 links to its
binding site through the plasma membrane on the lipid-
exposed face of the channel pore, stabilizing the closed state of
the channel. Our previous SAR study has elucidated that the
H-ring functionalities, including the partially skipped triene side
chain, were found to be important for the potent acute toxicity
against mice, while the A-ring functionalities had little influence
on the activity. Taken together with the recent studies on the
effects of gambierol on Kv channels,10−13,21 we hypothesized
that the minimal skeletal structure required for potent Kv
channel binding ability would reside in the right-half of the
molecule. Accordingly, in the present study, we designed the
heptacyclic and tetracyclic analogues (2 and 3, respectively,
Figure 1), both of which contain all of the H-ring functionalities
required for the potent toxicity, while their chemical synthesis is
significantly easier than that of the natural product. Although it
has long been believed that the full molecular length of the
polycyclic ethers that reaches several nanometers in general
may be critically important for their potent biological
activity,51−56 the present study clearly demonstrates that only
the right-half domain of 1 is necessary for the inhibition of K+
current and its beneficial effects in the 3xTg-AD cortical model.
In this work, we showed that the upregulation of Kv3.1
subunit expression found in 3xTg-AD neurons is inhibited by
long-term pretreatment with 1, 2, or 3. Moreover, this decrease
in Kv3.1 expression in neurons cultured in the presence of the
compounds was linked to a reduction in the peak amplitude of
Kv currents in this cellular model. However, this effect was not
observed in the corresponding wild-type nontransgenic cortical
culture, where only long-term exposure of the neurons to 1
showed an inhibition of the IDR component of the potassium
current. In addition, acute administration of these compounds
decreased the outward IK current and also the delayed IDR
current by 35−50% at compound concentrations in the nano-
molar range.
We have previously linked the gambierol-induced neuronal
effects to functional modulation of NMDA receptors in primary
cultures of cerebellar neurons.13 Although in this study 1 did
not modify the glutamate-induced calcium increase, the
compound causes a sustained increase in the intracellular
calcium concentration in cortical neurons (Alonso et al.,
unpublished observations) which could be probably attributed
to the interaction of the toxin with glutamate receptors.
Overactivation of NMDA receptors can lead to neuronal
damage through calcium overload,60 but a high blockage also
provokes neuronal dysfunction.61 Memantine is a low affinity
NMDA antagonist widely used for the treatment of moderate
to severe AD,62 which showed NMDA antagonism in
micromolar concentrations63 with beneficial effects in 3xTg-
AD mice,64 resulting in a decrease in the levels of Aβ and
hyperphosphorylated tau. Furthermore, recent studies have
indicated the involvement of NMDA receptors in APP
expression and Aβ production.14 As shown in this work, 1−3
were also able to reduce these two cellular pathologies,
decreasing both intra- and extracellular Aβ levels together
with the amount of hyperphosphorylated tau recognized by
AT8 and AT100 antibodies. As occurred with the intracellular
calcium oscillations induced by 1 in cerebellar granule cells,13
the effects of 1 over AT8 expression were blocked by the
preincubation of the cells with APV and also diminished in the
presence of CNQX, pointing again to NMDA and AMPA/
kainate receptors as mediators of the beneficial effect of these
compounds on the AD pathology. Both Aβ peptides and even
phosphorylated tau proteins are supposed to interact with
NMDA receptors and to be implicated in the glutamate
enhancement of AD pathology.65 Although coincubation with
NMDA and AMPA receptor modulators blocked the effects of
1 in this cellular system, further detailed investigation into the
molecular mechanism of NMDA receptor modulation by 1
remains as a future challenge. Even though bioavailability
studies to evaluate the ability of these compounds to cross the
blood brain barrier are not available yet, a previous report with
modified analogues of gambierol showed that these analogues
produced neurological symptoms resembling those of 1.9b
7476
dx.doi.org/10.1021/ja300565t | J. Am. Chem. Soc. 2012, 134, 7467−7479

Journal of the American Chemical Society
Article
bromophenol blue, pH 6.8) by SDS−PAGE and transferred onto
PVDF membranes (Millipore). The Snap i.d. protein detection system
was used for blocking and antibody incubation as previously
described.22 The concentrations of primary antibodies employed in
this work are shown in Table 1.
The immunoreactive bands were detected using the Supersignal
West Pico chemiluminescent substrate (Pierce) and the Diversity 4 gel
documentation and analysis system (Syngene, Cambridge, U.K.).
Chemiluminescence was measured with the Diversity GeneSnap
software (Syngene). β-Actin was used as control for lane loading and
to normalize chemiluminescence values.
ELISA. The amount of amyloid β in the culture medium was
measured with the Colorimetric BetaMark x-42 ELISA kit (SIGNET)
following the protocol indicated by the manufacturer. The human
Aβ1−42 peptide used in this kit shows an estimated 21% cross-
reactivity with rodent Aß1−42 peptide. In all the experiments, culture
medium samples were obtained at the same days in vitro from NonTg
and 3xTg-AD cultures after the treatment with the compounds from
third to seventh div. The optical density was measured at 620 nm in a
Syngene multiwell plate reader.
Determination of the Cytosolic Calcium Concentration
[Ca2+]c. Cultured cortical neurons from NonTg and 3xTg-AD mice
treated with the compounds from third to seventh div were loaded
with the Ca2+ sensitive fluorescent dye Fura-2 AM at 2.5 μM for 10
min at 37 °C. Then, cells were washed 3 times with cold buffer. The
coverslips were inserted into a thermostatted chamber at 37 °C (Life
Science Resources, Royston, Herts, U.K.) and viewed with a Nikon
Diaphot 200 microscope equipped with epifluorescence optics (Nikon
40×-immersion UV-Fluor objective). The [Ca2+]c images were
collected by doubled excitation fluorescence with a Life Science
Resources equipment. The light source was a 175 W xenon lamp. The
calibration of the fluorescence was made by the Grynkiewicz method.
For the calcium experiments, the extracellular medium contained (in
mM): 154 NaCl, 5.6 KCl, 1.3 CaCl2, 1 MgCl2, 5.6 glucose, and 10
HEPES, pH 7.4 adjusted with Tris. All experiments were carried out in
duplicate.
Electrophysiology. Potassium currents were recorded from
primary cortical neurons at room temperature (22−25 °C), using a
computer-controlled amplifier (Multiclamp 700B, Molecular Devices;
Sunnyvale, CA). Signals were recorded and analyzed using a Pentium
computer equipped with the Digidata 1440 data acquisition system
and pClamp10 software (Molecular Devices). Signals were prefiltered
at 5 kHz and digitized using a Digidata 1200 interface (Axon
Instruments).
In most experiments, the patch-clamp technique in whole-cell
configuration was performed using borosilicate glass micropipets of
5−10 MΩ resistance. The internal pipet solution contained (in mM):
132.5 KCl, 0.6 EGTA, 10 HEPES, 2 MgCl2, 2 ATP, and 0.3 GTP, pH
7.2 adjusted with KOH. The extracellular solution contained (in mM):
154 NaCl, 5.6 KCl, 3.6 NaHCO3, 1.3 CaCl2, 1 MgCl2, 5 glucose, and
10 HEPES, pH 7.2. To eliminate sodium currents, saxitoxin at 1 μM
was added to the extracellular solution in all the experiments testing
potassium currents.
To isolate the inactivating component IA and the delayed-rectifier
noninactivating component IDR of K+ currents, previously described
electrophysiological protocols were used.58 Total outward current
(IK = IA+ IDR) was measured by applying a depolarizing voltage step
from −100 to +40 mV of 250 ms duration, preceded by a preconditioning
pulse at −100 mV of 1.5 s from a holding potential of −80 mV. Then,
after a conditioning pulse at −40 mV during 100 ms, for full inacti-
vation of the IA component, the IDR current was isolated by applying a
depolarizing step from −40 to +40 mV.
Determination of Cell Viability. Cell viability was assessed by the
MTT (3- [4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide)
test, as previously described.45,46 This test, which measures
mitochondrial function, was used to assess cell viability as it has
been shown that in neuronal cells there is a good correlation between a
drug-induced decrease in mitochondrial activity and its cytotoxicity.
The assay was performed in cultures grown in 96 well plates and
exposed to different concentrations of 1, 2, or 3 added to the culture
Moreover, the results described here showing that two different
analogues with different lengths showed the same effects as 1
and even with lower doses are important for the development
of further studies to investigate the capacity of these
compounds to cross the blood brain barrier.
CONCLUSIONS
■
In the present study, we designed and synthesized truncated
skeletal analogues of gambierol (1) and showed that the
inhibitory activity of 1 against Kv channels resides in the right-
wing domain of the polycyclic ether skeleton. Our successful
elaboration of structurally simplified skeletal analogues, that is,
2 and 3, with potency comparable to that of 1 demonstrates the
power of the concept of diverted total synthesis.66 Furthermore,
we have shown that compounds 1−3 lowered both intra- and
extracellular Aβ levels and tau hyperphosphorylation via
modulation of NMDA receptors that is possibly secondary to
Kv channel inhibition in an in vitro mouse model of AD. Thus,
1 and its skeletal analogues should serve as useful chemical
probes for understanding the function of Kv channels and for
elucidating the molecular mechanism of Aβ metabolism
modulated by NMDA receptors.
EXPERIMENTAL SECTION
■
Synthetic Chemistry. For details, see the Supporting Information.
Primary Cortical Neurons. Two colonies of homozygous 3xTg-
AD mice and wild nontransgenic (NonTg) were established at the
animal facilities of the University of Santiago de Compostela, Spain,
where animals were used to obtain primary cultures of cortical neurons
from 3xTg-AD and Nontransgenic mice. All protocols described in this
work were revised and authorized by the University of Santiago de
Compostela Institutional animal care and use committee.
Primary cortical neurons were obtained from embryonic day 15−17
NonTg and homozygous 3xTg-AD mice fetuses as recently described
elsewhere.22,45,46 Briefly, cerebral cortex was removed and neuronal
cells were dissociated by trypsinization followed by mechanical
titration in DNase-containing solution (0.005% w/v) with a soybean
trypsin inhibitor (0.05% w/v) at 37 °C. After dissociation, the cells
were suspended in Dulbecco’s Modified Eagle’s medium (DMEM)
supplemented with p-amino benzoic acid, insulin, penicillin, and 10%
fetal calf serum. The cell suspension was seeded in 12 or 96 multiwell
plates precoated with poly-D-lysine and incubated for 7−10 days in
vitro (div) in a humidified 5% CO2/95% air atmosphere at 37 °C.
Cytosine arabinoside, 20 μM, was added before 48 h of culture to
prevent growing of non-neuronal cells. In all the experiments, cortical
neurons from NonTg and 3xTg-AD mice were prepared and pro-
cessed simultaneously.
Chemicals and Solutions. Plastic tissue-culture dishes were
obtained from Falcon (Madrid, Spain). Fetal calf serum was purchased
from Gibco (Glasgow, U.K.) and DMEM was from Biochrom (Berlin,
Germany). Fura-2 acetoxymethyl ester (Fura 2-AM) was from
Molecular Probes (Leinden, The Netherlands). All other chemicals
were reagent grade and purchased from Sigma-Aldrich (Madrid,
Spain).
Synthetic gambierol and analogues were purified by HPLC prior to
biological experiments. Copies of HPLC traces are included in the
Supporting Information. Dimethylsulfoxide (DMSO) was used for the
preparation of gambierol and analogue stock solutions. The final
DMSO concentration in the extracellular culture medium was always
lower than 0.05%.
Western Blotting. Cultured neurons pretreated with the
compounds from the third to the seventh div were lysed in 50 mM
Tris-HCl buffer (pH 7.4) containing a phosphatase/protease cocktail
inhibitor (Roche). The protein concentration in the lysates was
determined by the Bradford assay. Samples of cell lysates containing
20 μg of total protein were resolved in gel loading buffer (50 mM Tris-
HCl, 100 mM dithiotreitol, 2% SDS, 20% glycerol, 0.05%
7477
dx.doi.org/10.1021/ja300565t | J. Am. Chem. Soc. 2012, 134, 7467−7479

Journal of the American Chemical Society
Article
medium. Cultures were maintained in the presence of the compounds
at 37 °C in humidified 5% CO2/95% air atmosphere for 120 h.
Sodium azide was used as cellular death control and its fluorescence
was subtracted to the other data. After the exposure time, cells were
rinsed and incubated for 60 min with a solution of MTT (500 μg/mL)
dissolved in Locke’s buffer containing (in mM): 154 NaCl, 5.6 KCl,
1.3 CaCl2, 1 MgCl2, 5.6 glucose, and 10 HEPES, pH 7.4 adjusted with
Tris. After washing off excess MTT, the cells were disaggregated with
5% sodium dodecyl sulfate and the colored formazan salt was
measured at 590 nM in a spectrophotometer plate reader.
Statistical Analysis. All data are expressed as means SEM of
three or more experiments (each performed in duplicate). Statistical
comparison was by nonpaired Student’s t test or ANOVA with
Dunnett’s multiple comparison test. P-values <0.05 were considered
statistically significant.
Natural Products”. From Japan Society for the Promotion of
Science (JSPS), Japan: Grants-in-Aid for Young Scientists (A) and
for Scientific Research (A). Eva Alonso is recipient of a predoctoral
fellowship from Fondo de Investigaciones Sanitarias (pFIS),
Ministerio de Sanidad y Consumo, Spain.
REFERENCES
■
(1) Lin, Y.-Y.; Risk, M.; Ray, S. M.; Van Engen, D.; Clardy, J.; Golik,
J.; James, J. C.; Nakanishi, K. J. Am. Chem. Soc. 1981, 103, 6773.
(2) For review see: (a) Yasumoto, T.; Murata, M. Chem. Rev. 1993,
93, 1897. (b) Murata, M.; Yasumoto, T. Nat. Prod. Rep. 2000, 17, 293.
(3) Ito, E.; Suzuki-Toyota, F.; Toshimori, K.; Fuwa, H.; Tachibana,
K.; Satake, M.; Sasaki, M. Toxicon 2003, 42, 733.
(4) Satake, M.; Murata, M.; Yasumoto, T. J. Am. Chem. Soc. 1993,
Abbreviations. 3xTg, Triple transgenic; 4-AP, 4-aminopyridine;
Aβ, amyloid β peptide; Ac, acetyl; AD, Alzheimer’s Disease; AIBN,
α,α′-azobisisobutyronitrile; APP, amyloid-β precursor protein; APV,
D-(−)-2-amino-5-phosphonopentanoate; BACE, β-secretase; CNQ
X, 6-cyano-7-nitro-quinoxaline-2,3-dione; COD, 1,5-cyclooctadiene;
CSA, ( )-10-camphorsulfonic acid; div, days in vitro; DIBALH,
diisobutylaluminum hydride; DMEM, Dulbecco’s modified Eagle’s
medium; DMAP, 4-dimethylaminopyridine; DMF, N,N-dimethylfor-
mamide; DMSO, dimethylsulfoxide; Kv, voltage-gated potassium
channel; LHMDS, lithium bis(trimethylsilyl)amide; mGluR, metabo-
tropic glutamate receptor; MPM, p-methoxyphenylmethyl; MS,
molecular sieves; MTT, 3-[4,5dimethylthiazol-2-yl]-2,5-diphenyltetra-
zolium bromide, Nav, voltage-gated sodium channel; NaHMDS,
sodium bis(trimethylsilyl)amide; NIS, N-iodosuccinimide; NMDA,
N-methyl-D-aspartate; NMO, N-methylmorpholine N-oxide; nonTg, non
Triple transgenic; OTf, trifluoromethanesulfonate; PbTxs, brevetoxins;
PPTS, pyridinium p-toluenesulfonate; Py, pyridine; SAR, structure−
activity relationship; TBAF, tetra-n-butylammonium fluoride; TBS,
tert-butyldimethylsilyl; TEA, tetraethylammonium; THF, tetrahydrofuran;
TMS, trimethylsilyl; TPAP, tetra-n-propylammonium perruthenate.
115, 361.
(5) (a) Fuwa, H.; Sasaki, M.; Satake, M.; Tachibana, K. Org. Lett.
2002, 4, 2981. (b) Fuwa, H.; Kainuma, N.; Tachibana, K.; Sasaki, M. J.
Am. Chem. Soc. 2002, 124, 14983.
(6) (a) Kadota, I.; Takamura, H.; Sato, K.; Ohno, A.; Matsuda, K.;
Yamamoto, Y. J. Am. Chem. Soc. 2003, 125, 46. (b) Kadota, I.;
Takamura, H.; Sato, K.; Ohno, A.; Matsuda, K.; Satake, M.;
Yamamoto, Y. J. Am. Chem. Soc. 2003, 125, 11893.
(7) (a) Johnson, H. W.; Majumder, U.; Rainier, J. D. J. Am. Chem.
Soc. 2005, 127, 848. (b) Majumder, U.; Cox, J. M.; Johnson, H. W.;
Rainier, J. D. Chem.Eur. J. 2006, 12, 1736. (c) Johnson, H. W.;
Majumder, U.; Rainier, J. D. Chem.Eur. J. 2006, 12, 1747.
(8) Furuta, H.; Hasegawa, Y.; Mori, Y. Org. Lett. 2009, 11, 4382.
(b) Furuta, H.; Hasegawa, Y.; Hase, M.; Mori, Y. Chem.Eur. J. 2010,
16, 7586.
(9) (a) Fuwa, H.; Kainuma, N.; Satake, M.; Sasaki, M. Bioorg. Med.
Chem. Lett. 2003, 13, 2519. (b) Fuwa, H.; Kainuma, N.; Tachibana, K.;
Tsukano, C.; Satake, M.; Sasaki, M. Chem.Eur. J. 2004, 10, 4894.
(10) Ghiaroni, V.; Sasaki, M.; Fuwa, H.; Rossini, G. P.; Scalera, G.;
Yasumoto, T.; Pietra, P.; Bigiani, A. Toxicol. Sci. 2005, 85, 657.
Cuypers, E.; Yanagihara, A.; Rainier, J. D.; Tytgat, J. Biochem. Biophys.
Res. Commun. 2007, 361, 214−17. Cuypers, E.; Abdel-Mottaleb, Y.;
Kopljar, I.; Rainier, J. D.; Raes, A. L.; Snyders, D. J.; Tytgat, J. J.
Toxicon 2008, 51 (6), 974−83.
ASSOCIATED CONTENT
* Supporting Information
Synthetic procedure, compound characterization data, and
■
S
1
copies of H and 13C NMR spectra for all new compounds.
(11) Louzao, M. C.; Cagide, E.; Vieytes, M. R.; Sasaki, M.; Fuwa, H.;
Yasumoto, T.; Botana, L. M. Cell. Physiol. Biochem. 2006, 17, 257.
(12) LePage, K. T.; Rainier, J. D.; Johnson, H. W.; Baden, D. G.;
Murray, T. F. J. Pharmacol. Exp. Ther. 2007, 323, 174.
(13) Alonso, E.; Vale, C.; Sasaki, M.; Fuwa, H.; Konno, Y.; Perez, S.;
Vieytes, M. R.; Botana, L. M. J. Cell. Biochem. 2010, 110, 497.
(14) (a) Verges, D. K.; Restivo, J. L.; Goebel, W. D.; Holtzman, D.
M.; Cirrito, J. R. J. Neurosci. 2011, 31, 11328. (b) Bordji, K.; Becerril-
Ortega, J.; Nicole, O.; Buisson, A. J. Neurosci. 2010, 30, 15927.
(c) Hoey, S. E.; Williams, R. J.; Perkinton, M. S. J. Neurosci. 2009, 29,
4442.
(15) Ballatore, C.; Lee, V. M.; Trojanowski, J. Q. Nat. Rev. Neurosci.
2007, 8, 663.
(16) Hardy, J. A.; Higgins, G. A. Science 1992, 256, 184.
(17) Selkoe, D. J. Science 1997, 275, 630.
(18) Goedert, M.; Wischik, C. M.; Crowther, R. A.; Walker, J. E.;
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Luis.Botana@usc.es
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
This work was funded with the following grants. From
Ministerio de Ciencia y Tecnologia, Spain: AGL2007-60946/
■
́
ALI, SAF2009-12581 (subprograma NEF), AGL2009-13581-
CO2-01, TRA2009-0189, AGL2010-17875. From Xunta de
Galicia, Spain: GRC 2010/10, PGDIT 07MMA006261PR,
PGIDIT (INCITE) 09MMA003261PR, PGIDIT (INCITE)
Klug, A. Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 4051.
(19) Pendlebury, W. W.; Solomon, P. R. Clin. Symp. 1996, 48, 2.
(20) Spillantini, M. G.; Goedert, M. Trends Neurosci. 1998, 21, 428.
(21) Kopljar, I.; Labro, A. J.; Cuypers, E.; Johnson, H. W.; Rainier, J.
D.; Tytgat, J.; Snyders, D. J. Proc. Natl. Acad. Sci. U.S.A. 2009, 106,
9896.
(22) Vale, C.; Alonso, E.; Rubiolo, J. A.; Vieytes, M. R.; LaFerla, F.
M.; Gimenez-Llort, L.; Botana, L. M. Cell. Mol. Neurobiol. 2010, 30,
577.
(23) Fuwa, H.; Noji, S.; Sasaki, M. J. Org. Chem. 2010, 75, 5072.
(24) For reviews see: (a) Hoveyda, A. H.; Zhugralin, A. R. Nature
2007, 450, 243. (b) Gradillas, A.; Perez-Castells, J. Angew. Chem., Int.
Ed. 2006, 45, 6086. (c) Nicolaou, K. C.; Bulger, P. G.; Sarlah, D.
09261080PR, 2009/XA044, 2009/053 (Consell. Educacion
́
),
2008/CP389 (EPITOX, Consell. Innovacion e Industria,
́
programa IN.CI.TE.), 10PXIB261254 PR. From EU VIIth
Frame Program: 211326 - CP (CONffIDENCE), 265896
BAMMBO, 265409 μAQUA, and 262649 BEADS. From the
Atlantic Area Programme (Interreg IVB Trans-national): 2008-
1/003 (Atlantox) and 2009-1/117 Pharmatlantic. From the
Ministry of Education, Culture, Sports, Science and Technol-
ogy (MEXT), Japan: Grants-in-Aid for Young Scientists (B) and
for Scientific Research on Innovative Areas “Chemical Biology of
7478
dx.doi.org/10.1021/ja300565t | J. Am. Chem. Soc. 2012, 134, 7467−7479

Journal of the American Chemical Society
Article
Angew. Chem., Int. Ed. 2005, 44, 4490. (d) Deiters, A.; Martin, S. F.
Chem. Rev. 2004, 104, 2199.
(25) Watanabe, W. H.; Conlon, L. E. J. Am. Chem. Soc. 1957, 79,
(55) Oguri, H.; Tanabe, S.; Oomura, A.; Umetsu, M.; Hirama, M.
Tetrahedron Lett. 2006, 47, 5801. (b) Oguri, H.; Oomura, A.; Tanabe,
S.; Hirama, M. Tetrahedron Lett. 2005, 46, 2179.
(56) Candenas, M. L.; Pinto, F. M.; Cintado, C. G.; Morales, E. Q.;
Brouard, I.; Díaz, M. T.; Rico, M.; Rodríguez, E.; Rodríguez, R. M.;
Perez, R.; Perez, R. L.; Martín, J. D. Tetrahedron 2002, 58, 1921.
́ ́
(57) Franciosi, S.; Ryu, J. K.; Choi, H. B.; Radov, L.; Kim, S. U.;
McLarnon, J. G. J. Neurosci. 2006, 26, 11652.
2828.
(26) Bosch, M.; Schlaf, M. J. Org. Chem. 2003, 68, 5225.
(27) Okimoto, Y.; Sakaguchi, S.; Ishii, Y. J. Am. Chem. Soc. 2002, 124,
1590.
(28) Scholl, M.; Ding, S.; Lee, C. W.; Grubbs, R. H. Org. Lett. 1999,
(58) Pannaccione, A.; Boscia, F.; Scorziello, A.; Adornetto, A.;
Castaldo, P.; Sirabella, R.; Taglialatela, M.; Di Renzo, G. F.;
Annunziato, L. Mol. Pharmacol. 2007, 72, 665.
(59) Yu, S. P.; Farhangrazi, Z. S.; Ying, H. S.; Yeh, C. H.; Choi, D. W.
Neurobiol. Dis. 1998, 5, 81.
(60) Albers, G. W.; Goldberg, M. P.; Choi, D. W. Arch. Neurol. 1992,
49, 418.
(61) Collingridge, G. L.; Bliss, T. V. Trends Neurosci. 1995, 18, 54.
(62) Reisberg, B.; Doody, R.; Stoffler, A.; Schmitt, F.; Ferris, S.;
Mobius, H. J. N. Engl. J. Med. 2003, 348, 1333.
(63) Parsons, C. G.; Stoffler, A.; Danysz, W. Neuropharmacology
2007, 53, 699.
(64) Martinez-Coria, H.; Green, K. N.; Billings, L. M.; Kitazawa, M.;
Albrecht, M.; Rammes, G.; Parsons, C. G.; Gupta, S.; Banerjee, P.;
LaFerla, F. M. Am. J. Pathol. 2010, 176, 870.
(65) Hu, N. W.; Ondrejcak, T.; Rowan, M. J. Pharmacol., Biochem.
Behav. 2012, 100, 855.
1, 953.
(29) (a) Sasaki, M.; Fuwa, H.; Inoue, M.; Tachibana, K. Tetrahedron
Lett. 1998, 39, 9027. (b) Sasaki, M.; Fuwa, H.; Ishikawa, M.;
Tachibana, K. Org. Lett. 1999, 1, 1075. (c) Sasaki, M.; Ishikawa, M.;
Fuwa, H.; Tachibana, K. Tetrahedron 2002, 58, 1889.
(30) For reviews see: (a) Fuwa, H. Synlett 2011, 6. (b) Fuwa, H.. Bull.
Chem. Soc. Jpn. 2010, 83, 1401. (c) Sasaki, M.; Fuwa, H. Nat. Prod.
Rep. 2008, 25, 401. (d) Sasaki, M. Bull. Chem. Soc. Jpn. 2007, 80, 856.
(31) Ley, S. V.; Norman, J.; Griffith, W. P.; Marsden, S. P. Synthesis
1994, 1994, 639.
(32) Fuwa, H.; Sasaki, M.; Tachibana, K. Org. Lett. 2001, 3, 3549.
(33) Nicolaou, K. C.; Prasad, C. V. C.; Hwang, C. K.; Duggan, M. E.;
Veale, C. A. J. Am. Chem. Soc. 1989, 111, 5321.
(34) Kobayashi, S.; Hori, M.; Hirama, M. Carbohydr. Res. 2008, 343,
443.
(35) Ito, Y.; Hirao, T.; Saegusa, T. J. Org. Chem. 1978, 43, 1011.
(36) Feng, F.; Murai, A. Chem. Lett. 1992, 1587.
(66) For successful examples of skeletal truncation of structurally
complex natural products, see: (a) Wender, P. A.; Baryza, J. L.;
Brenner, S. E.; DeChristopher, B. A.; Loy, B. A.; Schrier, A. J.; Verma,
V. A. Proc. Natl. Acad. Sci. U.S.A. 2011, 108, 6721. (b) Schnermann, M.
J.; Beaudry, C. M.; Egorova, A. V.; Polishchuk, R. S.; Sutterlin, C.;
Overman, L. E. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 6158.
(c) Bonazzi, S.; Eidam, O.; Guttinger, S.; Wach, J. Y.; Zemp, I.; Kutay,
U.; Gademann, K. J. Am. Chem. Soc. 2010, 132, 1432. (d) Shan, D.;
Chen, L.; Njardarson, J. T.; Gaul, C.; Ma, X.; Danishefsky, S. J.;
Huang, X. Y. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 3772. (e) Kishi, Y.;
Fang, F.; Forsyth, C.; Scola, P.; Yoon, S. Patent, U., Ed. 1995; Vol.
5436238.
(37) Uenishi, J. i.; Kawahama, R.; Yonemitsu, O.; Tsuji, J. J. Org.
Chem. 1998, 63, 8965.
(38) Stille, J. K. Angew. Chem., Int. Ed. Engl. 1986, 25, 508.
(39) Han, X.; Stoltz, B. M.; Corey, E. J. J. Am. Chem. Soc. 1999, 121,
7600.
(40) Schwab, P.; Grubbs, R. H.; Ziller, J. W. J. Am. Chem. Soc. 1996,
118, 100.
(41) Lau, A.; Tymianski, M. Pflugers Arch. 2010, 460, 525.
(42) Oddo, S.; Caccamo, A.; Tseng, B.; Cheng, D.; Vasilevko, V.;
Cribbs, D. H.; LaFerla, F. M. J. Neurosci. 2008, 28, 12163.
(43) De Felice, F. G.; Wu, D.; Lambert, M. P.; Fernandez, S. J.;
Velasco, P. T.; Lacor, P. N.; Bigio, E. H.; Jerecic, J.; Acton, P. J.;
Shughrue, P. J.; Chen-Dodson, E.; Kinney, G. G.; Klein, W. L.
Neurobiol. Aging 2008, 29, 1334.
(44) Guo, J. P.; Arai, T.; Miklossy, J.; McGeer, P. L. Proc. Natl. Acad.
Sci. U.S.A. 2006, 103, 1953.
(45) Alonso, E.; Vale, C.; Vieytes, M. R.; Laferla, F. M.; Gimenez-
Llort, L.; Botana, L. M. Neurochem. Int. 2011, 59, 1056.
(46) Alonso, E.; Vale, C.; Vieytes, M. R.; Laferla, F. M.; Gimenez-
Llort, L.; Botana, L. M. Cell. Physiol. Biochem. 2011, 27, 783.
(47) Romano, C.; Yang, W. L.; O’Malley, K. L. J. Biol. Chem. 1996,
271, 28612.
(48) Hardingham, G. E.; Fukunaga, Y.; Bading, H. Nat. Neurosci.
2002, 5, 405.
(49) For reviews see: (a) Nicolaou, K. C.; Aversa, R. J. Isr. J. Chem.
2011, 51, 359. (b) Nicolaou, K. C.; Frederick, M. O.; Aversa, R. J.
Angew. Chem., Int. Ed. 2008, 47, 7182. (c) Inoue, M. Chem. Rev. 2005,
105, 4379. (d) Nakata, T. Chem. Rev. 2005, 105, 4314.
(50) For review see: Szpilman, A. M.; Carreira, E. M. Angew. Chem.,
Int. Ed. 2010, 49, 9592.
(51) (a) Rein, K. S.; Baden, D. G.; Gawley, R. E. J. Org. Chem. 1994,
59, 2101. (b) Gawley, R. E.; Rein, K. S.; Jeglitsch, G.; Adams, D. J.;
Theodorakis, E. A.; Tiebes, J.; Nicolaou, K. C.; Baden, D. G. Chem.
Biol. 1995, 2, 533.
(52) (a) Inoue, M.; Hirama, M.; Satake, M.; Sugiyama, K.; Yasumoto,
T. Toxicon 2003, 41, 469. (b) Inoue, M.; Lee, N.; Miyazaki, K.; Usuki,
T.; Matsuoka, S.; Hirama, M. Angew. Chem., Int. Ed. 2008, 47, 8611.
(53) (a) Torikai, K.; Oishi, T.; Ujihara, S.; Matsumori, N.; Konoki,
K.; Murata, M.; Aimoto, S. J. Am. Chem. Soc. 2008, 130, 10217.
(b) Torikai, K.; Yari, H.; Mori, M.; Ujihara, S.; Matsumori, N.; Murata,
M.; Oishi, T. Bioorg. Med. Chem. Lett. 2006, 16, 6355. (c) Ujihara, S.;
Oishi, T.; Torikai, K.; Konoki, K.; Matsumori, N.; Murata, M.;
Oshima, Y.; Aimoto, S. Bioorg. Med. Chem. Lett. 2008, 18, 6115.
(54) Sasaki, M.; Tachibana, K. Tetrahedron Lett. 2007, 48, 3181.
7479
dx.doi.org/10.1021/ja300565t | J. Am. Chem. Soc. 2012, 134, 7467−7479