Department of Physiology and Pharmacology, State University of New
York Health Science Center, Brooklyn, New York (K.W., M.D., T.N.S.);
and Graduate School of Pharmaceutical Sciences, Chiba University,
Chiba, Japan (K.K., K.I.)
 |
Introduction |
Several
families of cDNAs encoding subunits of glutamate receptors have been
cloned. These include the NR subunits of NMDA receptors and the GluR
and KA subunits of AMPA and kainate receptors (Hollmann and Heinemann,
1994
; Dingledine et al., 1999). All of these subunits are large (100 to
160 kDa) proteins that share varying degrees of sequence homology and
presumed structural homology. In particular, all glutamate receptor
subunits have three membrane-spanning domains (M1, M3, and M4) and a
pore-forming re-entrant loop (M2), an agonist binding domain formed by
the S1 region preceding M1 and the S2 loop between M3 and M4, and an
amino-terminal domain preceding S1 (Dingledine et al., 1999). Two
orphan subunits, 
1 and 2, which have sequence homology with GluR
and NR subunits, have also been cloned (Yamazaki et al., 1992; Araki et
al., 1993; Lomeli et al., 1993).
Recombinant 1 and 2 receptors expressed in Xenopus
oocytes or in mammalian cells are not activated by glutamate or a
number of other glutamate receptor agonists, nor do they bind
radiolabeled glutamate (Yamazaki et al., 1992; Araki et al., 1993;
Lomeli et al., 1993). Thus, the endogenous ligand (if any) that
activates receptors containing 2 subunits remains unknown. The subunit mRNAs and proteins are expressed in discrete regions of the
nervous system, with 2 being found predominantly in cerebellar
Purkinje cells (Araki et al., 1993; Lomeli et al., 1993; Mayat et al., 1995). Antisense oligonucleotides directed against the 2 subunit were found to selectively reduce the expression of long-term depression (LTD) in cerebellar granule cells (Hirano et al., 1994; Jeromin et al., 1996), and knockout mice with a disrupted 2 gene have deficits in synapse formation and in motor coordination (Kashiwabuchi et al., 1995). This suggests that 2 subunits do, somehow, play an
important role in normal cerebellar functioning and plasticity.
Although it is not known how receptors are normally activated in
vivo, it has been reported that a mutation in the 2 receptor gene
generates constitutively open channels in cerebellar Purkinje cells
(Zuo et al., 1997). This mutation, which generates a threonine instead
of alanine at position 654 in the 2 protein (A654T), leads to
Purkinje cell degeneration and is responsible for the Lurcher mouse phenotype (Zuo et al., 1997). The mutation is
located at the top of the M3 membrane spanning domain, a region
implicated in gating of glutamate receptor channels (Krupp et al.,
1998; Villarroel et al., 1998; Kohda et al., 2000; Jones et al., 2002). In oocytes or mammalian cells expressing recombinant 2 subunits, the
A654T mutation leads to more depolarized resting potentials and larger
holding currents under voltage-clamp than in nontransfected cells or
cells expressing wild-type 2, presumably due to the expression of
constitutively open 2(A654T) channels (Zuo et al., 1997; Wollmuth et
al., 2000). Thus, it is possible to study some properties of 2
channels using the 2(A654T) mutant. In this article, we have
studied some of the pharmacological properties of 2 channels using
the 2(A654T) mutant and have compared them with the properties of
NMDA receptors.
| |
Materials and Methods |
Subunit Clones and Site-Directed Mutagenesis.
The wild-type
1 (pYA91-5) and 2 (pA37-14) clones (Yamazaki et al., 1992;
Araki et al., 1993) were gifts from Dr. M. Mishina (University of
Tokyo, Tokyo, Japan). Some experiments were carried out using a
2(A654T) mutant (Zuo et al., 1997) that was a gift from Drs. J. Zuo
and N. Heintz (Rockefeller University, New York, NY). Most experiments
were carried out using a pSGEM-2(A654T) mutant that we constructed
from the wild-type 2 clone. The plasmids pSGEM-1 and pSGEM-2
were constructed by inserting the EcoRI fragments of pYA91-5
and pA37-14, respectively, into the same site of pSGEM (a gift from Dr.
Ralph Puchalski, Monell Chemical Senses Center, Philadelphia, PA), a
derivative of pGEMHE that contains the 5' and 3' untranslated regions
of Xenopus 
-globin (Liman et al., 1992) flanking the subunit inserts. Site-directed mutagenesis of the subunits was
carried out by the method of Sayers et al. (1992) or Ho et al. (1989).
A similar approach was used to construct the NR1(A653T), NR1(T648A),
and NR2B(A652T) mutants (Kashiwagi et al., 2002). Mutations were
confirmed by DNA sequencing using a Seq 4 × 4 personal sequencing
system (Amersham Biosciences Inc., Piscataway, NJ) over a region of
approximately 300 nucleotides containing the mutation. The NR1 clone
used in these studies is the NR1A variant (Moriyoshi et
al., 1991), which lacks the 21 amino acid insert encoded by exon-5, and
was a gift from Dr. S. Nakanishi (Institute for Immunology, Kyoto
University Faculty of Medicine, Kyoto, Japan). The rat NR2B clone
(Monyer et al., 1992) was a gift from Dr. P.H. Seeburg (Center for
Molecular Biology, University of Heidelberg, Germany). The GluR1 and
GluR2(Q) clones were gifts from Drs. J. Boulter and S. Heinemann (Salk Institute, La Jolla, CA). Amino acids are numbered from the initiator methionine in all subunits.
Expression in Oocytes and Voltage-Clamp Recording.
The
preparation of capped cRNAs and the preparation, injection, and
maintenance of oocytes were carried out as described previously (Williams et al., 1993). Oocytes were injected with 10 ng of 1 or
2 subunits and with 0.1 to 1 ng of NR1 plus 0.5 to 5 ng of NR2B to
study NMDA receptors, and 5 to 10 ng of GluR1 and GluR2(Q) to study
AMPA receptors. Macroscopic currents were recorded with a two-electrode
voltage-clamp using a GeneClamp 500 amplifier (Axon Instruments,
Inc., Union City, CA) as described previously (Williams, 1993).
Electrodes were filled with 3 M KCl and had resistances of 0.4 to 4 M
.
To study currents through constitutively open 2(A654T) channels, we
measured currents in the absence and presence of extracellular Na+. Oocytes were voltage-clamped and initially
superfused with an Na+-free solution that
contained the large impermeant cation NMDG ("NMDG-saline";
composition: 100 mM NMDG, 2 mM KCl, 1.8 mM BaCl2, 10 mM HEPES, pH 7.5). To measure currents through constitutively open
channels, the superfusate was changed to one containing 100 mM NaCl
("Na+-saline"; composition: 100 mM NaCl, 2 mM
KCl, 1.8 mM BaCl2, 10 mM HEPES, pH 7.5). To study
the pH sensitivity of 2(A654T) channels, oocytes were superfused
with NMDG-saline at a given pH (6.0 to 9.0) for 30 to 60 s before
and after superfusion with Na+-saline at that pH.
The difference in current between Na+ and NMDG
was calculated at each pH. For studies of NMDA receptors, oocytes were
injected with K+-BAPTA (100 nl of 40 mM, pH
7.0-7.4) on the day of recording and were continuously superfused with
Na+-saline as described previously (Williams,
1993). NMDA receptors were activated with 10 µM glutamate plus 10 µM glycine. AMPA receptors were activated by 100 µM kainate.
To obtain vales for the IC50 and maximum
inhibition (Inmax), data were fit to logistic
functions using SigmaPlot (SPSS Science, Chicago, IL). Data for
blockers that gave incomplete inhibition (e.g., pentamidine at
2(A654T)) were fit to eq. 1. Data for blockers that gave a complete
inhibition (e.g., pentamidine at NR1/NR2B receptors) were fit to eq. 2:
![I<SUB><UP>+blocker</UP></SUB>=100−(<UP>In<SUB>max</SUB></UP>/1+([<UP>blocker</UP>]<UP>/IC<SUB>50</SUB></UP>)<SUP>n<SUB><UP>H</UP></SUB></SUP>)](/content/vol305/issue2/fulltext/740/img001.gif)
|
(1)
|
![I<SUB><UP>+blocker</UP></SUB>=100−(1+([<UP>blocker</UP>]<UP>/IC</UP><SUB><UP>50</UP></SUB>)<SUP>n<SUB><UP>H</UP></SUB></SUP>)](/content/vol305/issue2/fulltext/740/img002.gif)
|
(2)
|
in which I+blocker is the
current measured in the presence of the antagonist and expressed as a
percentage of the control current.
| |
Results |
Constitutive Activity of Mutant Subunits.
The
Lurcher mutation, A654T, is located at the top of the M3
segment in the 2 subunit (Fig. 1A). As
reported previously (Zuo et al., 1997), oocytes expressing 2(A654T)
had more positive resting membrane potentials and larger holding
currents under voltage-clamp in Na+-saline than
did uninjected oocytes or oocytes expressing the wild-type 2
subunit. This is due to the expression of constitutively open
2(A654T) channels that gate Na+. The large
holding current in oocytes expressing 2(A654T) could be reduced by
replacing external Na+ with NMDG. In the
remainder of this article, we define currents through the 2(A654T)
channels as the difference between the holding currents measured in
Na+-saline and NMDG-saline. An example of this
current is shown in the inset to Fig. 1B, in which the superfusion is
switched from NMDG-saline to Na+-saline. In
oocytes expressing wild-type 2, switching from NMDG-saline to
Na+-saline had only a very small effect on the
holding current (5-20 nA) (Fig. 1B). A similar effect was seen in
uninjected oocytes, indicating that this small shift in the holding
current is due to a change in the background current and is unrelated
to wild-type 2 channels.
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 1.
Constitutive activity of 2(A654T) channels. A,
schematic of a glutamate receptor subunit showing the position of the
2(A654T) mutation, located at the top of the M3 membrane-spanning
domain. The amino acid sequences of the 1, 2, NR1, and NR2B
subunits in this region are also shown. The 2(A654) residue and the
corresponding alanine residues in other subunits (A654 in 1, A653 in
NR1, and A652 in NR2B) are shown in bold. The position of the Q/R/N
site, at the tip of the M2 helix, is also shown. B, the difference in
current (I; nanoamps) measured in NMDG-saline versus
Na+-saline was measured in oocytes expressing 1,
1(A654T), 2, and 2(A654T) subunits and voltage-clamped at  70
mV. Values are mean ± S.E. from 10 to 11 oocytes for each
receptor type, all recorded on the same day from the same batch of
oocytes. Examples of these currents are shown in the inset, in which
the perfusion was switched from NMDG-saline to Na+-saline
during the times shown by the horizontal bars.
|
|
The 1 subunit has sequence similarity to 2 and to NMDA and AMPA
receptor subunits. However, 1, like 2, is an "orphan" subunit
and no activation of 1 by glutamate or other agonists has been
reported. Because of the similarity of 1 and 2, we made a mutant
1 subunit, 1(A654T), which has an A-to-T mutation in the position
equivalent to the 2(A654T) mutation (Fig. 1A). Oocytes expressing
either the wild-type or mutant 1 subunit had only small holding
currents when voltage-clamped at 70 mV, and the differences in the
holding current measured in Na+-saline versus
NMDG-saline were similar to those seen with the wild-type 2 subunit
(Fig. 1B). Thus, unlike the 2(A654T) mutant, the equivalent mutation
in 1 does not generate constitutively open channels.
Inhibition of 2(A654T) Receptors.
The 2 subunit has
homology with subunits of NMDA and AMPA receptors. Several classes of
compounds that are open-channel blockers of NMDA and/or AMPA receptors
have been identified. These include dissociative anesthetics such as
ketamine and MK-801, adamantine derivatives such as memantine,
acetylcholinesterase inhibitors such as 9-aminoacridine, the antiviral
compound pentamidine, and a variety of polyamines and polyamine
derivatives (Collingridge and Lester, 1989; Dingledine et al., 1999).
We reasoned that 2 channels likely have structural features in
common with NMDA and/or AMPA channels and that some compounds that
block these channels may also block 2 channels. Thus we screened a
number of potential channel blockers for activity at 2(A654T)
receptors (Table 1). At concentrations of
1 to 10 µM, most compounds had little or no effect on 2(A654T)
channels, but inhibition by 20 to 50% was seen with 9-aminoacridine,
9-tetrahydroaminoacridine (9-THA), N1-dansyl-spermine
(N1-DnsSpm),
N1-dansyl-spermidine, and pentamidine
(Fig. 2A; Table 1). We compared the
effects of several of these blockers at 2(A654T), NMDA, and AMPA
channels (Fig. 2). Memantine, TB-3-4, and MK-801 are potent blockers of
NMDA channels but, at micromolar concentrations, have no effect on
2(A654T) or AMPA channels (Fig. 2B). Pentamidine, 9-THA and
N1-DnsSpm block 2(A654T) channels
but are also potent blockers of NMDA channels. Notably, 9-THA and
pentamidine have little effect at AMPA channels, although
N1-DnsSpm is a potent blocker of these
channels (Fig. 2B).
View this table:
[in this window]
[in a new window]
|
TABLE 1
Effects of various compounds on currents through 2(A654T) channels
The effects of each compound were tested at concentrations of 1 to 100 µM on currents through 2(A654T) channels (in Na+-saline;
background currents were measured in NMDG-saline) in oocytes
voltage-clamped at 70 mV. Values are the mean for two to six oocytes
for each compound. MGBG = methylglyoxyl-bis-guanylhydrazone. For
the structures of bis-ethyl- and benzyl-polyamines (BE4444, MB-3,
4-MB-3-3, DB-7, DB-10, DB-3-3, 1,12-D-B-3-4-3, 4,9-DB-3-4-3, and
TB-3-4) see Igarashi and Williams (1995),
Igarashi et al. (1997).
|
|
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 2.
Effects of channel blockers at 2(A654T), NMDA, and
AMPA receptors. A, representative traces showing the effects of
pentamidine and 9-THA on currents through 2(A654T). Oocytes were
voltage-clamped at 70 mV and superfused with NMDG-saline. The
superfusion was switched to Na+-saline (Na+)
and pentamidine was applied during the times shown by the horizontal
bars. B, 2(A654T) currents were measured as illustrated in A; NMDA
receptors were activated by 10 µM glutamate + 10 µM glycine; AMPA
receptors were activated by 100 µM kainate. The effects of six
channel blockers (1 or 10 µM) were measured at each receptor subtype,
and data are expressed as the percentage inhibition of currents through
each receptor. Thus, 10 µM pentamidine inhibits 2(A654T) and NMDA
receptors but has almost no effect at AMPA receptors. Oocytes were
voltage-clamped at 70 mV for all three receptor types. Values are
mean ± S.E. from four to eight oocytes.
|
|
The block of 2(A654T) currents by
N1-DnsSpm was slow in onset and very
slowly or incompletely reversible (data not shown), suggesting that
N1-DnsSpm would be of limited use as a
tool to study 2 channels. In contrast, block by 9-THA and
pentamidine was rapid in onset and rapidly and fully reversible (see
Fig. 2A). Subsequently, the effects of 9-THA and pentamidine were
studied in detail at 2(A654T) channels. For comparison, we also
studied the effects of these blockers on NMDA receptors expressed from
NR1/NR2B subunits.
Pentamidine and 9-THA inhibited 2(A654T) channels in a
concentration-dependent manner. In oocytes voltage-clamped at 70 mV,
IC50 values were 5 and 27 µM for pentamidine
and 9-THA, respectively (Fig. 3A;
Table 2). At 70 mV, pentamidine was about 10-fold and
9-THA about 3-fold more potent at NMDA receptors than at 2(A654T) receptors (Table 2). To determine whether the effects of the blockers
were voltage-dependent, we measured concentration-inhibition curves at
different holding potentials (40 to 100 mV). Block of 2(A654T)
currents by pentamidine was weakly voltage-dependent, the
IC50 decreasing 4-fold between 40 and 100 mV
(Fig. 3A; Table 2). Notably, however, block of 2(A654T) currents by
pentamidine was incomplete and the maximum inhibition increased as the
membrane potential decreased from 40 to 100 mV (Fig. 3A). Block was
incomplete even at membrane potentials more negative than 100 mV.
Thus, 10 µM pentamidine inhibited currents by 2 ± 1% at 20
mV, 42 ± 2% at 70 mV, 58 ± 1% at 100 mV, and by
57 ± 2% at 130 mV (mean ± S.E., four oocytes). Block of
2(A654T) receptors by 9-THA showed little or no voltage-dependence.
Thus, the IC50 for block by 9-THA was not
different between 40 and 100 mV, although there was a small
increase in the maximum inhibition at more negative potentials (Fig.
3A; Table 2). In contrast, block of NMDA receptors by both pentamidine
and 9-THA was voltage-dependent, with the potency increasing about
10-fold between 40 and 100 mV, and both antagonists produced a
complete block of NMDA currents at concentrations of 30 to 100 µM
(Fig. 3B).
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 3.
Inhibition of 2(A654T) and NMDA receptors by
pentamidine and 9-THA. Concentration-inhibition curves were constructed
using various concentrations of pentamidine and 9-THA in oocytes
expressing 2(A654T) (A) and NR1/NR2B NMDA receptors (B). At
2(A654T) receptors, currents were measured as the difference between
NMDG-saline and Na+-saline; NMDA receptors were activated
by 10 µM glutamate + 10 µM glycine. Oocytes were voltage-clamped at
40 to 100 mV. Currents measured in the presence of pentamidine or
9-THA are expressed as a percentage of the control current in the
absence of antagonist. Values are mean ± S.E. from four to seven
oocytes. The structures of pentamidine and 9-THA are shown above the
top panels.
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 2
Effects of pentamidine and 9-THA at 2(A654T) and NMDA receptors
IC50 values are the geometric mean (S.E.M., +S.E.M.), and
values for the maximum inhibition (Inmax) are the mean ± S.E.M. from four to seven oocytes for each receptor type.
Vh is the holding potential in millivolts.
|
|
We studied the effects of A-to-T mutations at positions in the NMDA
subunits, NR1(A653T) and NR2B(A652T), that are equivalent to position
A654 in 2 (see Fig. 1A). These experiments had two goals. First, to
determine whether the mutations in NMDA subunits would produce
constitutively open channels like those seen with the 2(A654T)
mutant. Second, to determine whether mutations at this position had
effects on the potency or degree of block by pentamidine and 9-THA. The
NR1(A653T) and NR2B(A652T) mutants, either alone or in combination, did
not produce constitutively active channels. The mutations, either alone
or in combination, did not affect block by pentamidine or 9-THA, with
IC50 values at 70 mV being, at most, 2-fold
lower than in the wild-type NR1/NR2B receptors and both antagonists
producing a complete block of macroscopic currents at these receptors
(Table 2).
A striking difference between the effects of pentamidine at NMDA
receptors and 2(A654T) receptors is the incomplete block at
2(A654T) receptors (Fig. 3A). This could be due to a different site
and/or mechanism of block by pentamidine at 2(A654T) compared with
NMDA receptors. Alternatively, the incomplete block could be a
facet of the 2(A654T) channels being constitutively open rather than being gated by a ligand. One way to address this question would be to study ligand-gated 2 channels but there are, of course, no ligands that have been shown to gate 2 receptors. Another approach would be to study the effects of pentamidine at constitutively active NMDA receptors and to compare the effects with wild-type NMDA
receptors and with 2(A654T) receptors. Although mutations at the
"Lurcher" positions in NMDA receptor subunits (NR1 A653T and NR2B
A652T) do not produce constitutively open channels, we have recently
found that several other mutations in this region of the NMDA receptor
NR1 subunit do produce constitutively open channels (Kashiwagi et al.,
2002). These mutants, which include NR1(T648A) (Fig.
4A), when coexpressed with NR2B generate
relatively large holding-currents in oocytes. The holding currents are
reduced by replacing extracellular Na+ with NMDG
or by adding extracellular Mg2+, which presumably
blocks the constitutively open NMDA channels (Kashiwagi et al., 2002).
The NR1(T648A)/NR2B receptors generated constitutive currents of
237 ± 51 nA (n = 10) and glutamate + glycine gate
only a small additional current (53 ± 15 nA) at these receptors.
In the same batch of oocytes, we also studied constitutive currents in
NR1(T648A)/NR2B(A654T) receptors, but these receptors generated only
small constitutive currents (59 ± 9 nA) and small glutamate + glycine currents (15 ± 4 nA; n = 10). In
wild-type NR1/NR2B receptors studied in the same batch of oocytes,
glutamate + glycine induced currents of 1247 ± 129 nA
(n = 10).
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 4.
Effects of pentamidine at NR1(T648A)/NR2B and at
2(A654T) receptors. A, schematic of the NR1 subunit showing the
position of residue T648 and the position of A653, which corresponds to
2A654. B, the effects of pentamidine on currents through
constitutively active NR1(T648A)/NR2B receptors were determined in
oocytes voltage-clamped at 70 mV. Values for current (nA) are the
difference between currents measured in Na+-saline and
NMDG-saline. In the absence of pentamidine this current was 122 ± 17 nA, and in the presence of 30 µM pentamidine it was 26 ± 5 nA. In control oocytes injected with wild-type NR1/NR2B (filled
circle), the current measured as the difference between
Na+-saline and NMDG-saline was 26 ± 7 nA. Inset, the
effects of pentamidine were determined in oocytes voltage-clamped at
40 mV and at 70 mV, and currents are expressed as a percentage of
the control current in the absence of pentamidine. IC50
values were 0.56 µM (40 mV) and 0.13 µM (70 mV). C, the effects
of pentamidine were measured on oocytes expressing 2(A654T) and
voltage-clamped at 70 mV. Values for current (nanoamps) are the
difference between currents measured in Na+-saline and
those measured in NMDG-saline. In the absence of pentamidine, this
current was 318 ± 27 nA, and in the presence of 300 µM
pentamidine it was 77 ± 8 nA. The IC50 value for
pentamidine was 6.8 µM. In control oocytes injected with wild-type
2 (filled circle) the current measured as the difference between
Na+-saline and NMDG-saline was 14 ± 1 nA. Values are
mean ± S.E. from five to eight oocytes.
|
|
Thus, we studied block by pentamidine at constitutively active
NR1(T648A)/NR2B receptors. The protocol was similar to that used to
study 2(A654T) receptors; currents through NR1(T648A)/NR2B channels
were defined as the difference between the holding currents measured in
Na+-saline and NMDG-saline in oocytes
voltage-clamped at 70 mV. We also measured this current in control
oocytes (from the same frog, and recorded on the same day) expressing
wild-type (i.e., not constitutively active) NR1/NR2B receptors.
NR1(T648A)/NR2B receptors generated currents of 122 ± 17 nA
(n = 5) compared with 26 ± 7 nA in control
oocytes (Fig. 4B). Pentamidine reduced the constitutive NR1(T648A)/NR2B
current to 26 ± 5 nA at 30 µM pentamidine, but did not affect
currents in oocytes expressing wild-type NR1/NR2B receptors (24 ± 6 nA with 30 µM pentamidine). Thus, pentamidine reduces currents at
NR1(T648A)/NR2B receptors to the level seen in control oocytes,
suggesting that pentamidine completely inhibits these constitutively
active NR1(T648A)/NR2B channels (Fig. 4B). The
IC50 for pentamidine at constitutively active
NR1(T648A)/NR2B receptors was 0.12 µM, similar to the value (0.5 µM) at wild-type NR1/NR2B receptors gated by glutamate and glycine.
This suggests that block by pentamidine is mechanistically similar in
the wild-type and constitutively active NMDA receptors. We also
determined whether this block was voltage-dependent. In another batch
of oocytes, we measured inhibition of NR1(T648A)/NR2B receptors at 40
mV and, in the same oocyte, at 70 mV (Fig. 4B, inset). The
IC50 values for pentamidine were 0.56 µM (40
mV) and 0.13 µM (70 mV). The 4-fold shift in sensitivity to
pentamidine between 40 and 70 mV seen at constitutively active
NR1(T648A)/NR2B receptors is similar to the shift seen over the same
voltage range at wild-type NR1/NR2B receptors activated by glutamate
(Table 1). We measured the concentration-inhibition curve for
pentamidine at 2(A654T) channels at 70 mV and compared the
currents in these oocytes to currents in control oocytes (from the same
frog, and measured on the same day) expressing the wild-type 2
subunit (Fig. 4C). Block by pentamidine was incomplete at 2(A654T)
channels, and the residual current (77 ± 8 nA in the presence of
300 µM pentamidine; n = 8) was much greater than that
in control oocytes expressing wild-type 2 (14 ± 1 nA;
n = 8). Thus, in constitutively active NMDA channels
block by pentamidine is complete, as it is at glutamate-gated NMDA
channels, but in constitutively active 2 channels the block is incomplete.
NMDA receptors are inhibited by protons, with a tonic inhibition of
about 50% at physiologic pH (Tang et al., 1990; Traynelis and
Cull-Candy, 1990). Protons also inhibit AMPA and kainate receptors, but
much less potently than NMDA receptors, with a pH
IC50 of 5.5 to 6.3 (Tang et al., 1990; Traynelis
and Cull-Candy, 1990). The effects of pH on 2 receptors have not
been reported. Therefore, we examined the influence of extracellular pH
on currents through 2(A654T) channels. In these experiments, we
measured the difference in the holding current between
Na+-saline and NMDG-saline at different
extracellular pH. Currents through 2(A654T) were sensitive to
extracellular pH, being small at acidic pH and larger at alkaline pH
(Fig. 5A). Inhibition by protons was
incomplete, reaching a maximum of 69 ± 2% between pH 6.5 and 6.0 and with a pH IC50 of 7.47 ± 0.04 (Fig.
5A). We also determined whether the effects of pH were
voltage-dependent by measuring pH inhibition curves in oocytes
voltage-clamped at different holding potentials. Between 20 and 100
mV, the effects of pH showed little or no voltage-dependence; neither
the pH IC50 nor the maximum inhibition was
affected by membrane potential (Fig. 5B).
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 5.
Influence of extracellular pH on 2(A654T)
receptors. A, currents through 2(A654T) receptors were recorded at
different extracellular pH (6.0 to 9.0) in oocytes voltage-clamped at
70 mV. Oocytes were superfused with NMDG-saline at a given pH for 30 to 60 s before and after switching to Na+-saline at
that pH. Currents (I) are expressed as a fraction of the
current at pH 9.0 (I9.0), and values are
mean ± S.E. from 15 oocytes. The inset shows representative
currents induced by switching from NMDG-saline to
Na+-saline at different pH values in one oocyte. B, the pH
IC50 and the maximum inhibition were measured at 20 to
100 mV using protocols similar to those shown in A.
|
|
NMDA receptors are modulated by polyamines such as spermine (Williams,
1997). Spermine has several effects on NMDA receptors including
"glycine-independent" stimulation, seen with saturating concentrations of glycine, which may involve a relief of tonic proton
inhibition (Traynelis et al., 1995; Williams, 1997). This effect is
mediated by an extracellular spermine binding site located in the
amino-terminal domain of the NR1 subunit (Masuko et al., 1999). In
light of the pH sensitivity of 2 receptors and their sequence
homology with other glutamate receptors, we carried out experiments to
determine whether 2(A654T) receptors were modulated by polyamines.
Spermine (100 µM) had no effect on currents through 2(A654T) in
oocytes voltage-clamped at 20 or 70 mV measured at pH 6.5, 7.5, and
8.5 (data not shown). Thus, unlike NMDA receptors, 2 receptors are
not stimulated by spermine.
| |
Discussion |
In this report we have studied some of the pharmacological
properties of 2 receptors using the constitutively active
2(A654T) mutant. The A654T mutation is located in a highly conserved
motif at the top of the M3 region, and A-to-T mutations at the
equivalent positions in the GluR1 and GluR6 subunits of AMPA and
kainate receptors have been reported to generate constitutively open
channels (Kohda et al., 2000). Interestingly, mutations at the
equivalent positions in NR1 and NR2 subunits of NMDA receptors do not
generate constitutively active channels (this report and Kohda et al., 2000) and we found that an equivalent mutation in the 1 subunit, which has no known function but, among the glutamate receptor family,
is most homologous to 2, did not generate constitutively active 1
channels. In the case of the 1 subunit, it is not known whether this
subunit can actually form functional homomeric channels because it is
not gated by glutamate or by the 1(A654T) mutation. The results with
the NMDA receptor NR1(A653T) and NR2B(A652T) mutants suggest that other
structural determinants, which are presumably different in NMDA, AMPA,
and subunits, influence gating in subunits that carry an A-to-T
mutation at the "Lurcher" position because the mutation is
sufficient to produce constitutive activity in 2 and AMPA channels,
but not in NMDA channels.
The results of previous studies have shown that 2(A654T) channels
have some properties in common with AMPA channels, including permeability to Ca2+ and double rectification
(Kohda et al., 2000; Wollmuth et al., 2000). In this study, we
investigated the effects of a number of channel blockers, some of which
differentially block NMDA and AMPA channels. Among a number of
structurally diverse compounds, we found that pentamidine and 9-THA
were the most potent blockers of currents through 2(A654T) channels.
These compounds also block NMDA channels, at which they are 3- to
10-fold more potent than at 2 channels, but do not block AMPA
channels. The polyamine derivative
N1-DnsSpm (Chao et al., 1997), which
is a potent blocker of both NMDA and AMPA channels, also blocks
2(A654T) currents, but the block is slow and poorly reversible.
Several NMDA receptor blockers, including MK-801 and memantine, which
are inactive at AMPA channels, had no effect on 2(A654T) channels.
These results suggest that the channel pore of 2 receptors has
features in common with both NMDA and AMPA channels, but also has
properties distinct from these other classes of channel. The 2
subunit contains a glutamine (Q) residue at the Q/R/N site in the M2
loop (see Fig. 1A), as do the GluR1 and GluR2(Q) AMPA receptor subunits
used in this study. NMDA receptor subunits contain an asparagine (N)
residue at this position. The nature of the residue at the Q/R/N site is known to affect divalent cation permeability and sensitivity to
channel blockers (Dingledine et al., 1999; Kashiwagi et al., 2002), and
the presence of a Q residue in the 2 subunit may account for the
lack of sensitivity to MK-801, memantine, and some other NMDA channel
blockers. The differences in sensitivity to 9-THA, pentamidine, and
N1-DnsSpm among 2, AMPA, and NMDA
receptors suggests that there are other important determinants for
block in addition to the Q/R/N site.
Although pentamidine and 9-THA block both 2(A654T) and NMDA
receptors, there are some notable differences in their effects at these
two classes of receptor. At NMDA receptors, block of macroscopic
currents by pentamidine and 9-THA was complete and the block was
strongly voltage-dependent. The potency of block increased as the
membrane potential was made more negative, a hallmark feature of a
typical open-channel blocker that binds to a site within the
membrane-spanning region of the channel pore. In contrast, block by
pentamidine and 9-THA was only weakly voltage-dependent at 2(A654T)
channels. Block by pentamidine, in particular, was incomplete although
the degree of block was also voltage-dependent. We found that block by
pentamidine at a constitutively active NMDA receptor, NR1(T648A)/NR2B,
was complete and had characteristics similar to block at wild-type
NR1/NR2B receptors. Thus, the different profile for pentamidine at 2
and NMDA receptors is not merely a reflection of one receptor type
(2) being constitutively active and the other (NMDA) being gated by
a ligand. One possible explanation is that pentamidine binds to a site
outside the channel pore of 2(A654T) receptors and does not act as
an open-channel blocker. Another possible explanation, compatible with
the incomplete and shallow block at 2(A654T) channels, is that
pentamidine can easily permeate these channels (and cannot easily
permeate NMDA channels). Permeation of NMDA channels by other classes
of blockers at very negative membrane potentials has been previously
reported (Igarashi and Williams, 1995; Chao et al.,
1997; Igarashi et al., 1997).
We found that 2(A654T) receptors are, like NMDA receptors, sensitive
to changes in extracellular pH around the physiological range. In
contrast to NMDA receptors, which are inhibited completely at acidic pH
(Traynelis and Cull-Candy, 1990; Traynelis et al., 1995), 2(A654T)
channels were inhibited by a maximum of 70% at pH 6.5 to 6.0. Proton
inhibition of 2(A654T) currents was not voltage-dependent,
suggesting that it involves protonation of an extracellular site
outside the ion channel pore. The location of the proton sensor on NMDA
receptors is not yet known, but it is conceivable that 2 and NMDA
receptors share a similar pH-dependent gating mechanism and/or proton
sensor. The pH sensitivity of 2(A654T) currents again suggests that
2 subunits have some properties in common with NMDA receptors and
other properties in common with AMPA receptors.
This work was supported by U.S. Public Health Service Grant NS35047.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
2 Glutamate Receptors: Effects of Pentamidine
and Protons
Top
Abstract
Introduction
-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid) receptors. Block of 
