Traditional AMPA receptor antagonists partially block Nav1.6-mediated persistent current

Neuropharmacology 55 (2008) 1165–1171

Contents lists available at ScienceDirect

Neuropharmacology journal homepage: www.elsevier.com/locate/neuropharm

Traditional AMPA receptor antagonists partially block Nav1.6-mediated persistent current N.C. Welch a, W. Lin b, P.F. Juranka b, C.E. Morris b, P.K. Stys a, * a b

Department of Clinical Neurosciences, Hotchkiss Brain Institute, University of Calgary, 3330 Hospital Drive NW, Calgary, AB T2N 4N1, Canada Division of Neuroscience, Ottawa Health Research Institute, University of Ottawa, Ottawa, ON K1Y 4E9, Canada

a r t i c l e i n f o

a b s t r a c t

Article history: Received 24 March 2008 Received in revised form 9 July 2008 Accepted 13 July 2008

Voltage-gated Na channels and AMPA receptors play key roles in neuronal physiology. Moreover, both channels have been implicated in the pathophysiology of both grey and white matter in a variety of conditions. Dissecting out the roles of these channels requires specific pharmacological tools. In this study we examined the potential non-specific effects on Nav1.6 channels of five widely used AMPA receptor blockers. Using whole-cell patch clamp electrophysiology, we identified a TTX-sensitive persistent Na channel current in HEK cells stably expressing the Nav1.6 channel. From a holding potential of 120 mV, slow ramp depolarization to þ75 mV generated an inward current that peaked at approximately 15 mV. Superfusion of purportedly specific AMPA antagonists, 1-naphthylacetyl spermine, SYM2206, CP465022, GYKI52466, blocked Nav1.6-mediated persistent currents in a dose-dependent manner. Each of these AMPA receptor blockers significantly inhibited (to z70% of control levels) the persistent Na current at concentrations routinely used to selectively block AMPA receptors. The AMPA/ kainate blocker, NBQX, did not significantly affect persistent Na channel currents. Furthermore, peak transient current was insensitive to NBQX, but was reversibly inhibited by SYM2206, CP465022 and GYKI52466. These results indicate that many commonly used AMPA receptor antagonists have modest but significant blocking effects on the persistent components of Nav1.6 channel activity; therefore caution should be exercised when ascribing actions to AMPA receptors based on use of these inhibitors. Ó 2008 Elsevier Ltd. All rights reserved.

Keywords: CP465022 GYKI52466 1-Naphthylacetyl spermine Nav1.6 Persistent SYM2206

1. Introduction Voltage-gated Na channel activation triggers action potential generation in excitable cells. Nine different subtypes of mammalian voltage-gated Na channels have been identified with a wide variety of developmental, tissue and/or cellular expression (Goldin et al., 2000). The Nav1.6 voltage-gated Na channel is expressed throughout the CNS, including localization at the nodes of Ranvier of mature axons (Caldwell et al., 2000). In addition to the transient Na current, Nav1.6 channels exhibit a persistent current (INap) (Smith et al., 1998), which is biophysically distinct from Nav1.1 and Nav1.2 channel persistent currents (for review, see Goldin, 1999). Nav1.6 INap is resistant to inactivation and important to the highfrequency firing of action potentials (Zhou and Goldin, 2004). However, INap in central axons (Stys et al., 1993) can also drive reverse Na/Ca exchange during injury in white matter, contributing to axonal degeneration. Blocking Na channels with TTX or STX, and more selective blockade of the persistent component with

* Corresponding author. Tel.: þ1 403 210 8646; fax: þ1 403 210 7446. E-mail address: [email protected] (P.K. Stys). 0028-3908/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuropharm.2008.07.015

quaternary amine local anesthetics (Stys et al., 1992a), exerts a significant protective effect (Stys et al., 1992b). Injured axons accumulate Na and lose K, resulting in prolonged axonal depolarization (for review, see Stys, 2004), and providing an ideal ionic environment for the release of glutamate via reversal of the Na-dependent glutamate transporter (Li et al., 1999). AMPA and kainate receptors are expressed on mature oligodendrocytes and astrocytes (for review, see Matute et al., 2006) with overactivation contributing to white matter injury (Yoshioki et al., 1996; Alberdi et al., 2002; Tekko¨k and Goldberg, 2001; Sa´nchez-Gomez and Matute, 1999). Consequently AMPA/kainate receptor blockade protects the functional and structural integrity of mature white matter during ischemic injury (McCarran and Goldberg, 2007; McCracken et al., 2002; Tekko¨k et al., 2007). The blockade of both Na channel and AMPA/kainate receptor activation is protective during white matter injury; however, in order to reliably dissect out the pathophysiological contribution of each pathway, it is important to possess selective tools, such as selective antagonists. In the current study, we explored whether several commonly employed AMPA receptor antagonists exhibit non-specific effects on the main Na channel subtype expressed in axons. We tested the effects of several purportedly selective AMPA blockers on persistent currents carried by Nav1.6 channels

1166

N.C. Welch et al. / Neuropharmacology 55 (2008) 1165–1171

expressed in HEK293 cells, using whole-cell voltage-clamp electrophysiology. Our findings indicate that all four AMPA receptorspecific blockers tested (SYM2206, CP465022-27, GYKI52466, 1-naphthylacetyl spermine) significantly reduced Nav1.6-mediated persistent currents at concentrations routinely used for selective AMPA receptor blockade (Pelletier et al., 1996; Lazzare et al., 2002; Donevan and Rogawski, 1993; Koike et al., 1997). In contrast, the AMPA/kainate receptor antagonist 2,3-dihyroxy-6-nitro-7-sulfamoyl-benzo(F)quinoxaline (NBQX) did not exhibit non-specific effects on Nav1.6 channels.

DMSO (NBQX, SYM2206, CP465022-27 (a kind gift from Dr. Frank Menniti, Pfizer, Groton, CT, USA)), HCl (GYKI52466) or water (1-naphthylacetyl spermine, tetrodotoxin (TTX)) and diluted to the appropriate concentrations in external solution. The pH of final solutions was maintained at 7.3. All non-gifted chemicals were purchased from Sigma Chemical Co. (St. Louis, MO, USA) except GYKI52466 (Tocris Biosciences, Ellisville, MO, USA). 2.3. Data analysis Igor Pro 6.01 (Wavemetrics Inc., Lake Oswego, OR, USA) was used to analyze the raw currents recorded with Clampex 6.0 (Axon Instruments), to calculate INap, and to perform statistical analyses. Current–voltage (I–V) relations were obtained using 50 ms depolarizations from a holding potential of 100 mV to a series of potentials ranging from 65 mV to þ35 mV. The maximum inward current recorded within the first 20 ms of the depolarizing pulse was taken as the transient component of the Na current (INat). INap was calculated from the mean current recorded 35–40 ms into the depolarizing pulse. Both INat and INap values were leak subtracted and plotted against the voltage of the depolarizing step to form I–V curves. Ramp currents were elicited from a holding potential of 120 mV and cells were depolarized to þ75 mV at a rate of 0.975 mV/ms. INap was quantified by extrapolating a line from the linear portion of the each ramp at negative potentials, before any inward currents became apparent, and calculating the difference between this leakage line and the deepest trough of inward current. An example is shown in Fig. 2D. The value of this difference was used as the maximum leak-subtracted amplitude of INap. Ramp current data were normalized to the amplitude of control INap for each individual cell recorded and expressed as the ratio of INap drug/INap vehicle control. Means were calculated from pooled data at each concentration. Data are reported as mean  standard deviation. Statistical analyses were performed using the non-parametric one-tail Wilcoxon–Mann–Whitney two sample rank test, with significance P values  0.05.

2. Methods 2.1. Cell culture The HEK293 cell line stably expressing the human Nav1.6 channel was a kind gift from Dr. Stephen Burbidge (GlaxoSmithKline, UK). Construction of Nav1.6 stably expressing cell lines has been previously described (Burbidge et al., 2002). HEK293 cells were cultured in Dulbecco’s modified eagle medium (DMEM) supplemented with 10% fetal bovine serum, 1% penicillin/streptomycin, 1% non-essential amino acids and 80 ml/10 ml Geneticin at 37  C under an atmosphere of 5% CO2 and 95% air. Cells were passaged by treatment with trypsin EDTA for 5 min, dissociated in supplemented DMEM, and plated on glass coverslips. All supplies were from Gibco, Invitrogen Canada Inc., Burlington, ON, Canada. 2.2. Whole-cell patch clamp electrophysiology Whole-cell voltage-clamp recordings were made from HEK293 cells stably expressing the Nav1.6 channel using an Axopatch 200B amplifier (Axon Instruments, Union City, CA, USA). The pipette solution contained (in mM) 110 CsCl, 10 NaCl, 5 MgCl2, 11 EGTA, 10 HEPES, 2 ATP, 1 GTP at pH 7.3 and adjusted to w300 mOsm l1 with sucrose. The external solution contained (in mM) 110 NaCl, 3 KCl, 1 MgCl2, 1 CaCl2, 20 TEA, 5 CsCl, 10 HEPES at pH 7.3 and adjusted to w310 mOsm l1 with glucose. Capacitance transients were cancelled and series resistance was typically 5– 15 MU. Voltages were not corrected for liquid junction potentials. Solutions, as described by Burbidge et al. (2002), were used to record transient Nav1.6 currents with/without AMPA receptor antagonists displayed in Fig. 4. Currents were recorded using pClamp6 software, filtered at 5 kHz and digitized with a Digidata 1200 interface using an IBM-compatible computer. All experiments were performed at room temperature. In all control recordings, 0.1% DMSO was included in the perfusate to allow direct comparison with DMSO-containing drug solutions. Complete or extensive block of the Na current could be rapidly achieved by addition of 2 mM TTX to the perfusate. Drugs to be tested (Fig. 1) were made up as a stock in

3. Results Human embryonic kidney epithelial cells (HEK) are commonly used as an expression system to study exogenous ion channel genes. Membrane currents were recorded from HEK293 cells stably expressing the TTX-sensitive human Nav1.6 channel using the whole-cell patch clamp configuration. Previous studies have shown that Nav1.6 can produce a sustained, persistent current that is still detectable tens of milliseconds after the onset of depolarization (Burbidge et al., 2002). We were able to record TTX-sensitive transient (INat) and persistent (INap) Na currents from cells held at

O

.3HCl

H N

N H

N H

NH2 H2NO2S O 2N

NAS

Me

H N

O

N H

O

NBQX

Me

O

O

N

O

N

N H

O

O

Cl

N O

N

F N N

N

.HCl

N

H 2N NH2

SYM 2206

CP 465022 Fig. 1. Structures of the AMPA receptor antagonists tested in this study.

GYKI 52466

N.C. Welch et al. / Neuropharmacology 55 (2008) 1165–1171

A

control

1167

30 µM NAS + 2 µM TTX

20 ms 200 pA

+35 mV

-100 mV

-60

-40

-20

0

20

C

50 ms 100 pA

B

voltage (mV)

30 µM NAS +2 µM TTX

INa (pA)

-500

INap control INap NAS+TTX INat control INat NAS+TTX

30 µM NAS -1000

10 µM NAS 3 µM NAS +75 mV

control -1500 -120 mV

D

E

50 pA

50 ms

*

1.0

INap drug/INap control

0.9

30 µM NBQX control

+75 mV

*

*

0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1

-120 mV

3 µM NAS

10 µM NAS

30 µM NAS

30 µM NBQX

Fig. 2. INap is reduced by NAS but not by NBQX. (A) Na channel currents recorded from a representative cell during depolarizing steps from 65 mV to þ35 mV in 10 mV increments from a holding potential of 100 mV in the absence (left) and in the presence (right) of 30 mM NAS þ 2 mM TTX. Dotted line denotes zero current. (B) I–V relations are shown for the same cell in A depicting INat (closed squares) and INap (closed circles) in control conditions and INat (open squares) and INap (open circles) in the presence of 30 mM NAS þ 2 mM TTX (closed circles). (C) Ramp currents, from the same cell illustrated in A and B, elicited during slow depolarizations from 120 mV to þ75 mV, at a rate of 0.975 mV/ ms, during control conditions (black), in the presence of 3 mM NAS (dark grey), 10 mM NAS (medium grey), 30 mM NAS (light grey), and 30 mM NAS þ 2 mM TTX (black). (D) Representative ramp currents from a different cell recorded during slow depolarizations from 120 mV to þ75 mV, at a rate of 0.975 mV/ms, in control conditions (black), upon application of 30 mM NBQX (medium grey). INap amplitudes were calculated by extrapolating a line (dotted, medium grey) from the linear portion of each ramp and calculating the delta between this line and the deepest trough of inward current. (E) Summary bar graph illustrating the mean ramp current data in the presence of drug, normalized to the amplitude of the control conditions and expressed as a ratio of INap drug to INap control. NAS (3 mM) is shown in dark grey; 10 mM NAS is shown in medium grey; 30 mM NAS is shown in light grey; 30 mM NBQX is shown in medium grey. Error bars represent standard deviation and stars denote significant differences in INap compared to control (P < 0.05).

1168

N.C. Welch et al. / Neuropharmacology 55 (2008) 1165–1171

a holding potential of 100 mV and then stepped from 65 mV to þ35 mV in 10 mV increments (Fig. 2A). The maximum INat recorded and the mean INap measured at 35–40 ms after the depolarizing step were plotted against membrane potential resulting in typical INa current–voltage (I–V) plots (Fig. 2B). In control conditions, INat was activated below 45 mV and was maximal near 15 mV, while INap was evident at membrane potentials positive to 35 mV and was maximal near 5 mV. Both transient and persistent Na currents were abolished in the presence of 2 mM TTX (Fig. 2A, B). During slow ramp depolarizations, slowly inactivating Na channels, such as Nav1.6, generate an inward current. Using the same cell whose currents are shown in Fig. 2A, the w1 mV/ms ramp from 120 mV to þ75 mV produced a persistent inward current (Fig. 2C) of comparable amplitude to that elicited by step depolarizations. The ramp current is deemed to represent INap, the component(s) of the Na current that persists after the rapid inactivation process is completed, and thus INap was studied via ramp clamp for the remainder of this study. 1-Naphthylacetyl spermine (NAS; Fig. 1) is a competitive AMPA receptor antagonist selective for calcium-permeable receptors, i.e. those lacking the GluR2 subunit (Iino et al., 1996; Koike et al., 1997). The representative INap ramp currents displayed in Fig. 2C were recorded in the presence of increasing concentrations of NAS with

50 pA

10 µM CP 3 µM CP

30 µM SYM 10 µM SYM 3 µM SYM control

50 µM GYKI +75 mV

-120 mV

D

*

1.0

*

0.9

50 ms

100 µM GYKI

0.3 µM CP control

+75 mV

-120 mV

INap drug/INap control

C

50 ms

20 pA

B

50 ms

50 pA

A

significant decreases in INap observed. The mean ratio of INap NAS/ INap control represents the % of control INap remaining after application of drug. In the cells tested (n ¼ 5), the mean ratio was 82  7% in 3 mM NAS, 79  20% in 10 mM NAS, and 68  13% in 30 mM NAS. As expected, application of 2 mM TTX in the presence of 30 mM NAS blocked INap further. The quinoxaline NBQX (Fig. 1) is a competitive antagonist which acts on both AMPA and kainate receptors (Sheardown et al., 1990; Wilding and Huettner, 1996). INap was not significantly altered by the application of even relatively high concentrations of NBQX (30 mM). The mean ratio of INap NBQX/INap control (n ¼ 4) was 92  9% (P ¼ 0.16 vs. control). The bar graph in Fig. 2E summarizes results with the two competitive receptor antagonists. We also tested three putatively selective non-competitive AMPA receptor antagonists to determine whether they exhibit non-specific effects on INap. The 2,3-benzodiazepines SYM2206 (SYM) and GYKI52466 (GYKI) have similar structures (Fig. 1), whereas the quinazolinone CP465022-27 (CP) was chosen for its different structure (Fig. 1). Surprisingly, all three agents exhibited significant blockade of persistent Nav1.6 currents at concentrations typically used to inhibit AMPA receptors. We tested SYM2206 at 3 mM, 10 mM and 30 mM; the decreasing INap trend became statistically significant at 30 mM (Fig. 3A, D). This concentration, which is commonly used for

10 µM GYKI control

+75 mV

-120 mV

*

*

0.8 0.7

*

0.6 0.5 0.4 0.3 0.2 0.1 3 µM SYM

10 µM SYM

30 µM SYM

0.3 µM CP

3 µM CP

10 µM CP

10 µM GYKI

50 µM GYKI

100 µM GYKI

Fig. 3. Non-competitive AMPA receptor blockers act non-selectively on INap. (A) Ramp currents from a representative cell recorded during slow depolarizations from 120 mV to þ75 mV, at a rate of 0.975 mV/ms, in control conditions (black), upon application of 3 mM SYM (dark grey), 10 mM SYM (medium grey), and 30 mM SYM (light grey). (B) Ramp currents elicited during slow depolarizations from 120 mV to þ75 mV, at a rate of 0.975 mV/ms, from a representative cell during control conditions (black), in the presence of 0.3 mM CP (dark grey), 3 mM CP (medium grey), and 10 mM CP (light grey). (C) From a representative cell, ramp currents were recorded during depolarizations from 120 mV to þ75 mV, at a rate of 0.975 mV/ms, in the absence of drug (control; black), in the presence of 10 mM GYKI (dark grey), 50 mM GYKI (medium grey), and 100 mM GYKI (light grey). (D) Summary bar graph detailing the mean ramp current data in the presence of drug, normalized to the amplitude of the control conditions and expressed as a ratio of INap drug to INap control. SYM (3 mM) is shown in dark grey; 10 mM SYM is shown in medium grey; 30 mM SYM is shown in light grey; 0.3 mM CP is shown in dark grey; 3 mM CP is shown in medium grey; 10 mM CP is shown in light grey. 10 mM GYKI is shown in dark grey; 50 mM GYKI is shown in medium grey; 100 mM GYKI is shown in light grey. Error bars represent standard deviation and stars denote significant differences in INap compared to control (P < 0.05).

N.C. Welch et al. / Neuropharmacology 55 (2008) 1165–1171

near-maximal inhibition of AMPA receptors (Pelletier et al., 1996), significantly inhibited INap (72  22% of control, n ¼ 5, P ¼ 0.05). From these same cells the mean ratios for 3 mM and 10 mM were 89  13% and 82  19%, respectively. In the presence of successively increasing concentrations, CP significantly inhibited INap in a dosedependent manner (Fig. 3B). The mean ratio of INap CP/INap control (n ¼ 5 cells) was as follows: 75  26% in 0.3 mM; 63  28% in 3 mM; 38  19% in 10 mM CP. GYKI significantly inhibited INap only at the highest concentration tested (100 mM; Fig. 3C). The mean ratio of INap GYKI/INap control (n ¼ 5 cells) was 98  20% in 10 mM GYKI; 82  24% in 50 mM; 70  23% in 100 mM. The bar graph in Fig. 3D summarizes the mean ratio observations using these antagonists. As illustrated in Fig. 4, effects on peak inward current I–V relations paralleled those seen with the ramp-induced currents. Again, 30 mM NBQX (5 cells) had no effect (INat drug/INat control ¼ 109  14%, P ¼ 0.97; Fig. 4E). For both 30 mM CP (n ¼ 5 cells) and 30 mM SYM (n ¼ 7 cells) (Fig. 4B, C) the ratio (INat drug/INat control) was reversibly reduced to 67  23% (P ¼ 0.031) and 41  21% (P ¼ 0.008) (respectively). For 100 mM GYKI (Fig. 4D) this ratio fell reversibly to 84  12% (P ¼ 0.002, n ¼ 9 cells).

A

B

1169

4. Discussion In this study, we investigated the potential effects of AMPA/ kainate and more selective AMPA receptor antagonists on other ion channels important in neuronal signaling and pathophysiology, specifically, the Nav1.6 Na channel. We were most interested in exploring the persistent component of this Na current, as it plays a major role in white matter injury mechanisms (Craner et al., 2004); mature white matter axons express high densities of Nav1.6 at nodes of Ranvier (Boiko et al., 2001), and exhibit a prominent persistent Na current (Dubois and Bergman, 1975; Stys et al., 1993). The Na currents that we observed, both transient and persistent components, were biophysically consistent with Nav1.6 currents previously recorded from transfected HEK293 cells (Burbidge et al., 2002), from transfected Xenopus oocytes (Smith et al., 1998), and from non-transfected motorneurons expressing Nav1.6 as well as other endogenous ion channels (Maurice et al., 2001; Harvey et al., 2006). In our study, the I–V relations of the persistent component of Nav1.6 currents were comparable when evoked from step depolarizations or voltage ramps (Fig. 2A–C). After confirming the

control V [mV] -50

voltage (mV)

100

50

50

control CP (3µ M) washout

-200

0 -50

3 µ M CP

-400 washout

-100 0

10

20

200 pA

time (ms)

-50

400

D

I [pA]

C

I [pA]

5 ms

50 -50

control SYM (30µ M) washout

50 -400

-500

E

V [mV] -50

I [pA]

-800

control GYK (100µM) washout

50

-500 500 pA -1000 control NBQX (30mM) washout

1 ms

Fig. 4. Some AMPA receptor antagonists reversibly inhibit transient Nav1.6 current. (A) Family of depolarizing steps used to generate I–V relations from a representative cell (Vhold ¼ 90 mV, steps applied at 1 Hz). (B) An example of I–V relations for a representative cell exposed to 30 mM CP, plus three sets of the raw current traces, as indicated. (C–E) Sample I–V relations for representative cells exposed to SYM, GYKI and NBQX (including sample traces at 0 mV), as indicated.

1170

N.C. Welch et al. / Neuropharmacology 55 (2008) 1165–1171

presence of INap, we focused on persistent currents by studying ramp currents, which minimized the contribution of any transiently activated currents, either from Nav1.6 itself or from other rapidly inactivating ion channels. Surprisingly, with all presumably selective AMPA receptor blockers that we studied, the inward currents carried by Nav1.6 in response to voltage ramp stimulation was decreased in a dose-dependent manner (Figs. 2 and 3), with statistically significant reductions of Nav1.6-mediated persistent currents induced by drug concentrations commonly used to specifically antagonize AMPA receptors. In contrast, the AMPA/ kainate receptor blocker NBQX was the only agent with no significant effect on either the persistent or transient components of Nav1.6 current (Fig. 2D; Fig. 4E). Our findings have important implications for studies relying on the reputed selectivity of AMPA antagonists. Nav1.6 channels are the most abundant Na channel subtype in the CNS, but Nav1.1 and Nav1.2 are also present and both have transient and persistent components (for review, see Goldin, 1999). Nav1.1 channels are primarily localized in the soma of neurons, including hippocampus, cerebellum and spinal cord, while Nav1.2 and Nav1.6 are found along axons with developmental-dependent expression (Westenbroek et al., 1989). The persistent current of Nav1.6 channels, but not Nav1.2 channels, is strongly linked to axonal injury (Craner et al., 2004). The constant influx of Na ions reverses the Nadependent glutamate transporter releasing excitotoxic levels of glutamate (Li and Stys, 2001). Persistent glutamate activation of Capermeable AMPA receptors on oligodendrocytes has been shown to cause injury to the glial cell as well as its myelin sheath (Sa´nchezGomez and Matute, 1999; Li and Stys, 2000; Tekko¨k and Goldberg, 2001; Alberdi et al., 2002). The actions of AMPA receptor activation and persistent Na channel opening are closely related, and thus, discerning the role of each ion channel in cellular injury is critically linked the specificity of pharmacological tools. The precise mechanism of AMPA antagonist blockade of Nav1.6 channels is unknown. Structurally, TTX is distinct from the AMPA blockers tested here as the toxin does not contain any aromatic rings and is not lipid soluble. The binding of TTX occurs from the extracellular side of the Na channel and does not interfere with gating. The Na channel need not be in the open configuration for TTX to block INa (for review, see Hille, 2001). In our study, maximal AMPA antagonist blockade was observed at the peak of INap hinting that AMPA blockers may act in a voltage-dependent manner. Local anesthetics, another class of Na channel blockers, act on Na channels by altering gating kinetics (for review, see Hille, 2001). Most local anesthetics, such as lidocaine, are lipid soluble and cross the cell membrane to reach intracellular site(s) of action. SYM2206, CP465022 and GYKI52466 are also lipid soluble and, thus, may cross the membrane to reach the local anesthetic binding site(s) of the Na channel. However, at the AMPA receptor, SYM2206, CP456022, and GYKI52466 are all non-competitive antagonists. Unlike NBQX, these blockers do not compete with glutamate for the agonist binding site on the receptor and are selective for AMPA over kainate receptors. It is interesting to note therefore, that all three non-competitive ‘‘selective’’ AMPA antagonists tested exhibited non-specific effects on the Nav1.6 channel, whereas the competitive, less selective AMPA/kainate blocker did not affect Na channels. This may provide clues about the mechanisms of lack of selectivity of one class of agents over the other. Although the mechanism of action of AMPA blockers on Na channels is still unresolved, our findings have important implications for studies relying on the reputed selectivity of AMPA antagonists. Our results raise a caution when using presumably selective AMPA antagonists in cells and tissues where physiological and pathophysiological responses additionally depend on Nav1.6 activity; assigning observations to AMPA receptors based solely on the actions of these traditional antagonists may not be warranted

unless the actions of Na channels are explicitly excluded (for instance by combining with TTX blockade). While the sensitivity of Nav1.1 and Nav1.2 channels to AMPA receptor antagonists was not determined in this study, our results raise the possibility that other Na channels may be similarly affected. This will require further study in cell systems expressing these and possibly other Na channels as well. Paradoxically, however, although the usefulness of these AMPA antagonists as selective tools may be diminished, their neuroprotective actions may be enhanced by such a combined block of AMPA and Nav1.6 channels, as both channels are heavily implicated in the pathophysiology of both grey and white matter in response to ischemia and other insults (Lynch et al., 1995; Taylor and Meldrum, 1995; Tekko¨k et al., 2007). Acknowledgements This work was supported by grants from the MS Society of Canada and the Canadian Institutes for Health Research. NCW was supported by a fellowship award from the Paralyzed Veterans of America. PKS was supported by a Heart and Stroke Foundation of Canada Career Investigator Award and more recently by an AHFMR Scientist Award. Work by CEM’s group was supported by a grant from the Canadian Institutes for Health Research. The authors thank G. Zamponi for helpful discussion. References Alberdi, E., Sa´nchez-Gomez, M.V., Marino, A., Matute, C., 2002. Ca(2þ) influx through AMPA or kainate receptors alone is sufficient to initiate excitotoxicity in cultured oligodendrocytes. Neurobiology of Disease 9, 234–243. Boiko, T., Rasband, M.N., Levinson, S.R., Caldwell, J.H., Mandel, G., Trimmer, J.S., Matthews, G., 2001. Compact myelin dictates the differential targeting of two sodium channel isoforms in the same axon. Neuron 30, 91–104. Burbidge, S.A., Dale, T.J., Powell, A.J., Whitaker, W.R.J., Xie, X.M., Romanos, M.A., Clare, J.J., 2002. Molecular cloning, distribution and functional analysis of the NaV1.6 voltage-gated sodium channel from human brain. Molecular Brain Research 103, 80–90. Caldwell, J.H., Schaller, K.L., Lasher, R.S., Peles, E., Levinson, S.R., 2000. Sodium channel NaV1.6 is localized at nodes of Ranvier, dendrites, and synapses. Proceedings of the National Academy of Sciences of the United States of America 97, 5616–5620. Craner, M.J., Newcombe, J., Black, J.A., Hartle, C., Cuzner, C.H., Waxman, S.G., 2004. Molecular changes in neurons in multiple sclerosis: altered axonal expression of NaV1.2 and NaV1.6 sodium channels and Na/Ca exchanger. Proceedings of the National Academy of Sciences of the United States of America 101, 8168–8173. Donevan, S.D., Rogawski, M.A., 1993. GYKI 52466, a 2,3-benzodiazepine, is a highly selective, noncompetitive antagonist of AMPA/kainate receptor responses. Neuron 10, 51–59. Dubois, J.M., Bergman, C., 1975. Late sodium current in the node of Ranvier. Pflugers Archiv: European Journal of Physiology 357, 145–148. Goldin, A.L., 1999. Diversity of mammalian voltage-gated sodium channels. Annals of the New York Academy of Sciences 868, 38–50. Goldin, A.L., Barchi, R.L., Caldwell, J.H., Hofmann, F., Howe, J.R., Hunter, J.C., Kallen, R.G., Mandel, G., Meisler, M.H., Metter, Y.B., Noda, M., Tamkun, M.M., Waxman, S.G., Wood, J.N., Catterall, W.A., 2000. Nomenclature of voltage-gated sodium channels. Neuron 28, 365–368. Harvey, P.J., Li, X., Li, Y., Bennett, D.J., 2006. Endogenous monoamine receptor activation is essential for enabling persistent sodium currents and repetitive firing in rat spinal motoneurons. Journal of Neurophysiology 96, 1171–1186. Hille, B., 2001. Ion Channels of Excitable Membranes, third ed. Sinauer Associates, Inc., Sunderland, MA. Iino, M., Koike, M., Isa, T., Ozawa, S., 1996. Voltage-sensitive blockage of Ca(2þ)permeable AMPA receptors by Joro spider toxin in cultured rat hippocampal neurones. Journal of Physiology 496, 431–437. Koike, M., Iino, M., Ozawa, S., 1997. Blocking effect of 1-naphthyl acetyl spermine on Ca(2þ) permeable AMPA receptors in cultured rat hippocampal neurons. Neuroscience Research 29, 27–36. Lazzare, J.T., Paternain, A.V., Lerma, J., Chenard, B.L., Ewing, F.E., Huang, J., Welch, W.M., Ganong, A.H., Menniti, E.S., 2002. Functional characterization of CP-465,022, a selective, noncompetitive AMPA receptor antagonist. Neuropharmacology 42, 143–153. Li, S., Mealing, G.R., Morley, P., Stys, P.K., 1999. Novel injury mechanism in anoxia and trauma of spinal cord white matter: glutamate release via reverse Nadependent glutamate transport. Journal of Neuroscience 19 (RC16), 1–9. Li, S., Stys, P.K., 2000. Mechanisms of ionotropic glutamate receptor-mediated excitotoxicity in isolated spinal cord white matter. Journal of Neuroscience 20, 1190–1198.

N.C. Welch et al. / Neuropharmacology 55 (2008) 1165–1171 Li, S., Stys, P.K., 2001. Na–K-ATPase inhibition and depolarization induce glutamate release via reverse Na-dependent transport in spinal cord white matter. Neuroscience 107, 675–683. Lynch, J.L., Yu, S.P., Canzoniero, L.M.T., Sensi, S.L., Choi, D.W., 1995. Sodium channel blockers reduce oxygen–glucose deprivation-induced cortical neuronal injury when combined with glutamate receptor antagonists. Journal of Pharmacology and Experimental Therapeutics 273, 554–560. Matute, C., Domercq, M., Sa´nchez-Go´mez, M.V., 2006. Glutamate-mediated glial injury: mechanisms and clinical importance. Glia 53, 212–224. Maurice, N., Tkatch, T., Meisler, M., Sprunger, L.K., Surmeier, D.J., 2001. D1/D5 dopamine receptor activation differentially modulates rapidly inactivating and persistent sodium currents in prefrontal cortex pyramidal neurons. Journal of Neuroscience 21, 2268–2277. McCarran, W.J., Goldberg, M.P., 2007. White matter axon vulnerability to AMPA/ kainate receptor-mediated ischemic injury is developmentally regulated. Journal of Neuroscience 27, 4220–4229. McCracken, E., Fowler, J.H., Dewar, D., Morrison, S., McCulloch, J., 2002. Grey matter and white matter ischemic damage is reduced by the competitive AMPA receptor SPD 502. Journal of Cerebral Blood Flow & Metabolism 22, 1090–1097. Pelletier, J.C., Hesson, D.P., Jones, K.A., Costa, A.M., 1996. Substituted 1,2-dihydrophthalazines: potent, selective, and noncompetitive inhibitors of the AMPA receptor. Journal of Medicinal Chemistry 39, 343–346. Sa´nchez-Gomez, M.V., Matute, C., 1999. AMPA and kainate receptors each mediate excitotoxicity in oligodendroglial cultures. Neurobiology of Disease 6, 475–485. Smith, M.R., Smith, R.D., Plummer, N.W., Meisler, M.H., Goldin, A.L., 1998. Functional analysis of the mouse Scn8a sodium channel. Journal of Neuroscience 18, 6093–6102. Sheardown, M.J., Nielsen, E.O., Hansen, A.J., Jacobsen, P., Honore´, T., 1990. 2,3Dihydroxy-6-nitro-7-sulfamoylbenzo(F)quinoxaline: a neuroprotectant for cerebral ischemia. Science 247, 571–574.

1171

Stys, P.K., 2004. White matter injury mechanisms. Current Molecular Medicine 4, 109–126. Stys, P.K., Sontheimer, H., Ransom, B.R., Waxman, S.G., 1993. Noninactivating, tetrodotoxin-sensitive Na conductance in rat optic nerve axons. Proceedings of the National Academy of Sciences of the United States of America 90, 6976– 6980. Stys, P.K., Waxman, S.G., Ransom, B.R., 1992a. Tertiary and quaternary local anesthetics protect CNS white matter from anoxic injury at concentrations that do not block excitability. Journal of Neurophysiology 67, 236–240. Stys, P.K., Waxman, S.G., Ransom, B.R., 1992b. Ionic mechanisms of axonic injury in mammalian CNS white matter: role of Na channels and Na–Ca exchanger. Journal of Neuroscience 12, 430–439. Taylor, C.P., Meldrum, B.S., 1995. Naþ channels as targets for neuroprotective drugs. Trends in Pharmacological Sciences 16, 309–316. Tekko¨k, S.B., Goldberg, M.P., 2001. AMPA/kainate receptor activation mediates hypoxic oligodendrocyte death and axonal injury in cerebral white matter. Journal of Neuroscience 21, 4237–4248. Tekko¨k, S.B., Ye, Z., Ransom, B.R., 2007. Excitotoxic mechanisms of ischemic injury in myelinated white matter. Journal of Cerebral Blood Flow & Metabolism 27, 1540–1552. Westenbroek, R.E., Merrick, D.K., Catterall, W.A., 1989. Differential subcellular localization of the RI and RII Naþ channel in central neurons. Neuron 3, 695–704. Wilding, T.J., Huettner, J.E., 1996. Antagonist pharmacology of kainate- and alphaamino-3-hydroxy-5-methyl-4-isoxazolepropionic acid-preferring receptors. Molecular Pharmacology 49, 540–546. Yoshioki, A., Bacskai, B., Pleasure, D., 1996. Pathophysiology of oligodendroglial excitotoxicity. Journal of Neuroscience Research 46, 427–437. Zhou, W., Goldin, A.L., 2004. Use-dependent potentiation of the Nav1.6 sodium channel. Biophysical Journal 87, 3862–3872.