A subtype-selective, use-dependent inhibitor of native AMPA receptors

Article abstract of DOI:10.1021/ja0705801

AMPA (α-amino-3-hydroxy-5-methyl-4-isooxazole) receptors (AMPARs) are glutamate-gated ion channels that play central roles in the mammalian brain, mediating fast excitatory synaptic transmission and underlying several forms of synaptic plasticity. Two sub

Full text of DOI:10.1021/ja0705801

Published on Web 03/29/2007
A Subtype-Selective, Use-Dependent Inhibitor of Native AMPA Receptors
Aaron Nilsen and Pamela M. England*
Departments of Pharmaceutical Chemistry and Cellular & Molecular Pharmacology, UniVersity of California,
San Francisco, California 94158-2517
Received January 25, 2007; E-mail: england@picasso.ucsf.edu
Scheme 1
AMPA (R-amino-3-hydroxy-5-methyl-4-isooxazole) receptors
(AMPARs) are glutamate-gated ion channels that play central roles
in the mammalian brain, mediating fast excitatory synaptic trans-
mission and underlying several forms of synaptic plasticity.1 In
particular, activity-dependent changes in the number of synaptic
AMPARs modulate the strength of synaptic transmission. For
example, the insertion of additional AMPARs into synapses
following brief periods of high-frequency activity underlies hip-
pocampal long-term potentiation (LTP), a strengthening of synaptic
transmission thought to share processes related to learning and
memory.2 Discreet, subtype-specific phases of AMPAR trafficking
may underlie this potentiation.
Two types of AMPARs are primarily expressed on excitatory
neurons in adult animalssheteromeric channels composed of GluR1
PhTx-74 in 78% overall yield, a marked improvement over the
and GluR2 subunits (GluR1/2 receptors) and channels composed
previously reported synthesis6 in terms of overall yield, cost per
of GluR2 and GluR3 subunits (GluR2/3 receptors).3 Efforts to
gram, and scalability.
distinguish between the roles these receptors play at synapses (the
To evaluate the selectivity of PhTx-74, we carried out standard
subcellular fraction of AMPARs mediating communication between
two-electrode voltage clamp current recordings from Xenopus
neurons) have relied on genetic approaches and have produced
oocytes heterologously expressing individual AMPAR subtypes.
conflicting results. Experiments utilizing virus-driven overexpres-
First, we examined the effect of PhTx-74 on homomeric AMPARs
sion of mutant receptor subunits in brain slices, providing a unique
by injecting oocytes with GluR1, GluR2, or GluR3 subunit mRNA.
biophysical and/or optical signature for AMPARs containing
Generally, polyamine toxins are potent inhibitors of AMPARs
lacking the GluR2 subunit.1b,7 Indeed, we found that PhTx-74 blocks
glutamate-evoked currents from homomeric GluR1 and GluR3
receptors, having no effect on homomeric GluR2 receptors (Figure
1). The observed inhibition is use-dependent, consistent with the
generally accepted binding mode of polyamine toxinsswithin the
pore of open channels.8 Next, we examined the effect of PhTx-74
on heteromeric GluR1/2 and GluR2/3 receptors by co-injecting
oocytes with GluR2 and either GluR1 or GluR3 subunit mRNAs
in a 1:1 ratio. In these experiments, we observed that PhTx-74 (100
µM) inhibits GluR1/2 receptors, but not GluR2/3 receptors even at
concentrations as high as 500 µM (Figure 1). Unfortunately, efforts
to measure an IC50 for PhTx-74 at GluR1/2 receptors met with little
success (see Supporting Information for a detailed discussion).
GluR1 and GluR3 AMPAR subunits appear to preferentially
assemble with GluR2 subunits to form heteromeric GluR2-
containing receptors.9 The subunit stoichiometry (fixed versus
variable) is a matter of debate, though significant evidence points
to assembly of tetrameric receptors as dimers of heteromeric subunit
dimers.10 Nonetheless, to rule out the possibility that the observed
block is due to the formation of a substantial population of
homomeric GluR1 receptors in oocytes co-injected with GluR1 and
GluR2 subunits, we monitored current-voltage (I/V) curves in these
cells. Linear I/V curves are indicative of the formation of GluR2-
containing receptors, whereas inwardly rectifying I/V curves are
diagnostic of receptors lacking the GluR2 subunit (e.g., GluR1
homomers).1b We consistently observed linear I/V curves in these
experiments, confirming the predominant formation of GluR2-
containing AMPA receptors (Figure 1). Moreover, 1-naphthyl acetyl
spermine (NASPM, 100 µM), a potent inhibitor of GluR1 ho-
specific combinations of subunits, suggested that GluR1/2 and
GluR2/3 receptors play distinct roles at synapses.4 However, in
genetically engineered mice lacking any single AMPAR subunit
or combinations of subunits, synaptic transmission and plasticity
are intact, implying redundant roles for GluR1/2 and GluR2/3
receptors.5 These observations underscore the importance of
developing a subtype-selective AMPAR antagonist to probe native
receptors and, along with an intriguing report describing the solid-
phase synthesis of polyamine toxins shown to be active against
AMPARs,6 prompted us to synthesize and characterize the selectiv-
ity of philanthotoxin-7,4 against a panel of AMPAR subtypes.
Philanthotoxin-7,4 (PhTx-74) is a synthetic analogue of the
naturally occurring wasp venom toxin philanthotoxin-4,3,3 (PhTx-
433), differing in the number of amines and intervening methylene
units contained in the polyamine tail.7 Here we show that PhTx-74
is a subtype-selective AMPAR antagonist, inhibiting GluR1/2, but
not GluR2/3 receptors. Inhibition is use-dependent (only open
channels are blocked), allowing for the selective block of synaptic
receptors activated by neurotransmitter released at nerve terminals.
The molecular basis for the observed selectivity and a highly
efficient synthesis of this toxin are also presented.
PhTx-74 was synthesized in six steps, as shown in Scheme 1.
Acid 1 was activated as the p-nitrophenol ester and coupled to
7-aminoheptanol to form alcohol 2, which was subsequently
converted to the mesylate and treated with N-Boc-1,4-diaminobu-
tane to form the protected toxin 3. The benzyloxy ether protecting
group was removed using hydrogenolysis and the tert-butyl
carbamate hydrolyzed with hydrochloric acid, to provide PhTx-74
4. Notably, this solution-phase route provides gram quantities of
9
4902
J. AM. CHEM. SOC. 2007, 129, 4902-4903
10.1021/ja0705801 CCC: $37.00 © 2007 American Chemical Society

C O M M U N I C A T I O N S
investigate this possibility, we replaced the GluR3 C-terminal
domain with the terminus from GluR1 and coexpressed the resulting
GluR3 subunit chimera (GluR3CT1) with GluR2 in oocytes. In these
experiments, we found that, unlike wild-type GluR2/3 receptors,
GluR2/3CT1 receptors are partially inhibited by PhTx-74, consistent
with the idea that the C-terminal domains impact AMPAR
biophysics, including the binding to polyamine toxins within the
channel pore (Figure 1).
In summary, we have identified and developed a highly efficient
synthesis of the first subtype-selective inhibitor of natiVe AMPARs.
Notably, since PhTx-74 is a use-dependent inhibitor, it provides a
means of selectively blocking synaptic GluR1/2 receptors and thus
directly monitoring subtype-specific changes in the composition
of synaptic AMPA receptors proposed to underlie synaptic plastic-
ity, including hippocampal LTP.
Figure 1. PhTx-74 is a subtype-selective inhibitor of AMPARs. (A) Effect
of PhTx-74 on homomeric AMPARs (GluR1, GluR2, and GluR3). (A1)
Bar graph showing PhTx-74 inhibits GluR1 and GluR3 receptors, but not
GluR2 receptors. Percent inhibition (mean ( SEM) was calculated for each
cell (n g 5) by measuring the steady-state glutamate-evoked (100 µM)
current before and after co-applying PhTx-74 (100, 500 µM). (A2)
Representative trace showing the inhibitory effect of PhTx-74 (100 µM)
on steady-state glutamate-evoked (100 µM) currents from GluR1 receptors.
(A3) Representative trace from the same cell in (A2) showing the effect of
co-applying glutamate (100 µM) and PhTx-74 (100 µM) on GluR1 receptors.
The peak current is comparable to the steady-state current observed in (A1),
consistent with use-dependent, open-channel block. (A4) Representative
inwardly rectifying current-voltage curve measured from the same cell as
in (A2) and (A3). (B) Effect of PhTx-74 on heteromeric AMPARs (GluR1/
2, GluR2/3, GluR2/3CT1). (B1) Bar graph showing that PhTx-74 inhibits
GluR1/2, but not GluR2/3 receptors. Partial inhibition is observed at GluR2/
Acknowledgment. We thank Ning Bao and Ming An for
technical assistance. This work was supported by grants from the
Sandler Foundation and McKnight Foundation.
Supporting Information Available: Synthetic and electrophysi-
ological methods. This material is available free of charge via the
Internet at http://pubs.acs.org.
References
(1) (a) Hollmann, M.; Heinemann, S. Annu. ReV. Neurosci. 1994, 17, 31-
108. (b) Dingledine, R.; Borges, K.; Bowie, D.; Traynelis, S. F. Pharmacol.
ReV. 1999, 51, 7-61. (c) Palmer, C. L.; Cotton, L.; Henley, J. M.
Pharmacol. ReV. 2005, 57, 253-277.
(2) (a) Malinow, R.; Malenka, R. C. Annu. ReV. Neurosci. 2002, 25, 103-
126. (b) Bredt, D. S.; Nicoll, R. A. Neuron 2003, 40, 361-379. (c)
Collingridge, G. L.; Isaac, J. R.; Wang, Y. T. Nat. ReV. Neurosci. 2004,
5, 952-961. (d) Malenka, R. C.; Bear, M. F. Neuron 2004, 44, 5-21.
(3) See ref 1c and references cited therein.
(4) (a) Shi, S. H.; Hayashi, Y.; Petralia, R. S.; Zaman, S. H.; Wenthold, R.
J.; Svoboda, K.; Malinow, R. Science 1999, 284, 1811-1816. (b) Hayashi,
Y.; Shi, S. H.; Esteban, J. A.; Piccini, A.; Poncer, J. C.; Malinow, R.
Science 2000, 287, 2262-2267. (c) Shi, S.; Hayashi, Y.; Esteban, J. A.;
Malinow, R. Cell 2001, 105, 331-343.
(5) (a) Jensen, V.; Katharina, M. M. K.; Borchardt, T.; Adelmann, G.; Rozov,
A.; Burnashev, N.; Brix, C.; Frotscher, M.; Andersen, P.; Hvalby, O.;
Sakmann, B.; Seeburg, P. H.; Sprengel, R. J. Physiol. 2003, 553.3, 843-
856. (b) Jia, Z. P.; Agopyan, N.; Miu, P.; Xiong, Z. G.; Henderson, J.;
Gerlai, R.; Taverna, F. A.; Velumian, A.; MacDonald, J.; Carlen, P.;
Abramow-Newerly, W.; Roder, J. Neuron 1996, 17, 945-956. (c) Meng,
Y. H.; Zhang, Y.; Jia, Z. P. Neuron 2003, 39, 163-176.
(6) Kromann, H.; Krikstolaityte, S.; Andersen, A. J.; Andersen, K.; Krogs-
gaard-Larsen, P.; Jaroszewski, J. W.; Egebjerg, J.; Stromgaard, K. J. Med.
Chem. 2002, 45, 5745-5754.
(7) (a) Mellor, I. R.; Usherwood, P. N. R. Toxicon 2004, 43, 493-508. (b)
Stromgaard, K.; Mellor, I. Med. Res. ReV. 2004, 24, 589-620.
(8) See: Andersen, T. F.; Tikhonov, D. B.; Bolcho, U.; Bolshakov, K.; Nelson,
J. K.; Pluteanu, F.; Mellor, I. R.; Egebjerg, J.; Stromgaard, K. J. Med.
Chem. 2006, 49, 5414-5423 and references cited therein.
(9) Mansour, M.; Nagarajan, N.; Nehring, R. B.; Clements, J. D.; Rosenmund,
C. Neuron 2001, 32, 841-853.
3CT1 receptors comprised of GluR2 and chimeric GluR3 subunits in which
the GluR3 C-terminal domain has been replaced by the GluR1 C-terminus.
Percent inhibition was calculated for each cell (n g 10, across two different
batches of oocytes) as in (A1). (B2) Representative trace showing the
inhibitory effect of PhTx-74 (100 µM) on steady-state glutamate-evoked
(100 µM) currents from GluR1/2 receptors. See Figure S2 for a representa-
tive trace showing the block is use-dependent. (B3) Representative trace
from the same cell in (B2), showing the absence of an effect of NASPM
(100 µM) on glutamate-evoked currents from GluR1/2 receptors. (B4)
Representative linear current-voltage curve measured from the same cell
as in (B2) and (B3).
momers (IC50 ) 0.3 µM), had an insignificant effect on AMPA
receptor-mediated currents from these oocytes (Figure 1).
The apparent subtype-selective block of GluR1/2 but not GluR2/3
receptors by PhTx-74 is quite remarkable. Sequences lining the
pore region of AMPARs are generally thought to be the key
determinant of polyamine toxin binding (e.g., the mRNA-edited
Q/R site residue).1b However, the membrane spanning sequences
of GluR1 and GluR3 (i.e., GluR1/2 and GluR2/3 receptors),
including those lining the channel pore, are identical (Figure S1).
In contrast, the GluR1 and GluR3 C-terminal domains, which have
been shown to influence the channel conductance (i.e., rate of ion
permeation),1c are quite dissimilar and could give rise to quaternary
structural differences between the GluR1/2 and GluR2/3 open-
channel pores, thus resulting in distinct affinities for PhTx-74. To
(10) (a) Washburn, M. S.; Numberger, M.; Zhang, S.; Dingledine, R. J.
Neurosci. 1997, 17, 9393-9406. (b) Ayalon, G.; Stern-Bach, Y. Neuron
2001, 13, 103-113. (c) Greger, I. H.; Khatri, L.; Kong, X. P.; Ziff, E. B.
Neuron 2003, 40, 763-774. (d) Brorson, J. R.; Li, D. D.; Suzuki, T. J.
Neurosci. 2004, 24, 3461-3470. (e) Gouaux, E. J. Physiol. 2004, 554
(Pt 2), 249-253. (f) Madden, D. R. Nat. ReV. Neurosci. 2002, 3, 91-
101. (g) Mayer, M. L. Curr. Opin. Neurobiol. 2005, 15, 282-288.
JA0705801
9
J. AM. CHEM. SOC. VOL. 129, NO. 16, 2007 4903