| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Neurology, Children's Hospital; and 2 Program in Neuroscience, Harvard Medical School, Boston Massachusetts 02115
 |
ABSTRACT |
|---|
 Abstract
 Introduction
Methods
Results
Discussion
References
|
|---|
Segal, Michael M. and Andrea F. Douglas. Late sodium channel openings underlying epileptiform activity are preferentially diminished by the anticonvulsant phenytoin. J. Neurophysiol. 77: 3021-3034, 1997. Late openings of sodium channels were observed in outside-out patch recordings from hippocampal neurons in culture. In previous studies of such neurons, a persistent sodium current appeared to underlie the ictal epileptiform activity. All the channel currents were blocked by tetrodotoxin. In addition to the transient openings of sodium channels making up the peak sodium current, there were two types of late channel openings: brief late and burst openings. These late channel openings occurred throughout voltage pulses that lasted 750 ms, producing a persistent sodium current. At 
30 mV, this current was 0.4% of the peak current. The late channel openings occurred throughout the physiological range of trans-membrane voltages. The anticonvulsant phenytoin reduced the late channel openings more than the peak currents. The effect on the persistent current was greatest at more depolarized voltages, whereas the effect on peak currents was not substantially voltage dependent. In the presence of 60 µM phenytoin, peak sodium currents at 30 mV were 40-41% of control, as calculated using different methods of analysis. Late currents were 22-24% of control. Phenytoin primarily decreased the number of channel openings, with less effect on the duration of channel openings and no effect on open channel current. This set of findings is consistent with models in which phenytoin binds to the inactivated state of the channel. The preferential effect of phenytoin on the persistent sodium current suggests that an important pharmacological mechanism for a sodium channel anticonvulsant is to reduce late openings of sodium channels, rather than reducing all sodium channel openings. We hypothesize that pharmacological interventions that are most selective in reducing late openings of sodium channels, while leaving early channel openings relatively intact, will be those that produce an anticonvulsant effect while interfering minimally with normal function.
The persistent sodium current is made up of rare "late" openings of sodium channels that continue to occur many milliseconds after the beginning of a depolarization of cellular voltage. In contrast, the peak sodium current is made up of the more numerous transient openings of the sodium channels. It appears that both the late and transient channel openings reflect different states of the same channel molecule, because both types of openings are produced by sodium channels encoded by the same brain (Moorman et al. 1990 Culture methods
Cultures were prepared according to the microculture method of Segal and Furshpan (1990) Recording methods
Recordings were made 14-30 days after plating. Neurons in microcultures or mass cultures that were >17 µm diam were chosen for recording. Such cells are highly likely to release a glutamate-like transmitter (Segal 1991 Solutions
The extracellular perfusion solution was based on Ca2+/Mg2+-free Hanks' balanced salt solution, to which was added (in mM) 25 glucose, 5 N-2-hydroxyethylpiperazine-N Drugs
Phenytoin was dissolved to make 10-60-mM stock solutionsin a solvent vehicle consisting of water (50%), propylene glycol (40%), and ethanol (10%). The stock solutions were diluted 1:1000 in perfusion solution minutes before perfusion so as to minimize loss of drug by adsorption. Solvent vehicle was added to control solutions, resulting in the same solvent concentration in all drug and control solutions. Similar channel openings were seen with or without the vehicle.
Data analysis
Peak currents were assessed using blank traces for subtracting capacitive currents. Because there were usually no records without any transient channel openings, blanks were obtained using one of the following two procedures: 1) tetrodotoxin blanks were recorded directly using the drug (1 µM) to block all channel currents or 2) calculated blanks were constructed for each run of 56 pulses. Channel-free segments were combined using the segmented average feature of the Fetchan program of pClamp. Such segmented averages are not accurate in the first ~20 ms of the pulse due to the presence of transient channel openings in virtually every record. Accordingly, the opening-free end part of the pulse was copied, inverted, and used to form a symmetric beginning of the pulse, with appropriate correction for leak current. The calculated blanks appeared the same as tetrodotoxin blanks, but calculated blanks were used preferentially because they could be computed for each run of 56 pulses, thus minimizing the effects of drift of leakage current during the experiment. The blank construction procedure was automated using a program written in the programming language C, for which the source code is available on request from the authors, and also has been provided to Axon Instruments. Calculated blanks then were used in the pClamp software to generate blank-subtracted records.
Late sodium channel openings
Outside-out patches from all neurons studied (n = 17) displayed both early and late openings of channels displaying inward current. Early openings were taken as those openings occurring in the first few milliseconds, comprising the peak current, and late openings were taken as those after 50 ms (for 100-ms pulses) or after 100 ms (for 750-ms pulses). There were no patches with only a single channel. Potassium channel currents in these patches were blocked by the cesium (144 mM) in the patch pipette. No outward currents were seen in these recordings, using depolarizations to a range of voltages from
Phenytoin reduces late channel openings more than early openings
Phenytoin (10-60 µM) reduced both late and peak sodium currents in all patches (n = 17). Traces from one trial with 60 µM phenytoin are shown in Fig. 5, using a series of consecutive current traces to show the channel activity. The records are stacked to show many current traces, although the stacking obscures part of the peak currents in all but the final record in each panel.
Analysis of the phenytoin effect using ensemble currents
Single channel events were studied in most detail at a holding potential of Analysis of the phenytoin effect using idealized traces
Idealized traces were used to analyze late sodium channel openings in seven patches in which all trials used 750-ms pulses to a holding potential of
Concentration dependence of the phenytoin effect
The experiments described above used 60 µM phenytoin, which is above the therapeutic range of free phenytoin concentrations of 4-8 µM (McLean and Macdonald 1983
Voltage dependence of the phenytoin effect
The effect of phenytoin on peak and persistent sodium currents also was tested at different voltages from
Late sodium channel openings
This study adds to the growing body of evidence for the presence of persistent sodium currents in neurons (Alonso and Llinás 1989 Late sodium channel openings and epileptiform activity
The demonstration of late sodium channel openings in these neurons is relevant to our understanding of the currents that generate epileptiform activity. In this culture system, some neurons display "interictal" epileptiform activity [with wedge-shaped Paroxysmal Depolarizing Shifts (PDSs)], whereas others display "ictal" epileptiform activity (with runs of PDSs and a sustained depolarization) (Segal 1991 Phenytoin effects on sodium currents
The outside-out patch sodium channel studies reported here demonstrate a preferential effect of phenytoin on late openings of sodium channels. This is in accord with the known use-dependent action of phenytoin (Kuo and Bean 1994 Implications for development of anticonvulsant drugs
These results suggest that a persistent sodium current is important in initiating and continuing ictal epileptiform activity in this system. The action of the anticonvulsant phenytoin is to reduce this persistent current underlying seizures more effectively than it reduces the peak current underlying normal action potentials.
INTRODUCTION
Abstract
Introduction
Methods
Results
Discussion
References
) or muscle (Ukomadu et al. 1992; Zhou et al. 1991) sodium channel mRNA expressed in cells.
). Three lines of evidence, taken together, suggest that persistent sodium currents may be important in the treatment of convulsions and are a possible site of abnormalities producing convulsive seizures.
). Among the best studied of these drugs are phenytoin (Kuo and Bean 1994; Matsuki at al. 1984; Willow et al. 1985), carbamazepine (Willow et al. 1985), and lamotrigine (Lang et al. 1993).
; Matsumoto and Ajmone Marsan 1964). Evidence from a simplified culture system displaying epileptiform activity suggests that a persistent sodium current may be an important component of this sustained depolarization (Segal 1994). In that simplified system, neurons with the ictal sustained depolarizations also have endogenous bursts of action potentials and depolarization even when synaptic transmission is blocked, which is evidence for a nontransmitter persistent inward current.
) and paramyotonia congenita (Chahine et al 1994) and the cardiac long QT syndrome (Bennett et al. 1995) are caused by mutations, leading to an increase in late openings of sodium channels.
; Segal and Furshpan 1990) and also explores the actions of the anticonvulsant phenytoin on these currents.
; Lang et al. 1993; Matsuki at al. 1984; Willow et al. 1985), which is easier to study because it is two orders of magnitude larger than the persistent sodium current. However, if a persistent sodium current is crucial in maintaining ictal depolarization, it may be the most relevant portion of the sodium current to study.
). This study found a reduction in channel open times and number of channel openings, but the relevance to intact channels was unclear because of the proteolytic digestion of the channel.
).
METHODS
Abstract
Introduction
Methods
Results
Discussion
References
, with minor modifications (Segal 1994). Briefly, drops of collagen were sprayed onto a dried film of agarose to define microislands. Glial cells and then neurons adhered to form microcultures on these islands. The cultures were grown in medium containing synaptic antagonists (11 mM Mg2+ and 1 mM kynurenate), conditions that led to epileptiform activity in mass cultures (Furshpan and Potter 1989) and microcultures (Segal and Furshpan 1990) when the cultures were returned to solutions with standard magnesium (1 mM) and no kynurenate.
). Cultures were continually perfused during recording and drugs were added via perfusion with delay times of 
1 min. The temperature was maintained at 27.5°C in the central well of the dish using a heating coil.
). The signals were recorded using an Axopatch 200 or 200A amplifier and an IBM-type computer running pClamp 6.0 software (Axon Instruments). RMS current was 0.25-0.40 pA using pipettes that were coated with Sylgard (Dow-Corning) such that the visible component of the Sylgard was ~200 µm from the tip. Pipettes were fire polished to 5-10 M
. Holding potential was 100 mV.
-2-ethanesulfonic acid (HEPES), 1.5 Ca2+, and 1.0 Mg2+. The pipette solution contained (in mM) 60 CsCl, 60 CsF, 5 MgCl2, 10 Cs ethylene glycol-bis(
-aminoethyl ether)-N,N,N,N-tetraacetic acid (has Cs+ 20), and 10 Cs HEPES (has Cs+ 4), pH 7.4.
values.
View larger version (18K):
[in a new window]
FIG. 6.
Ensemble sodium currents have a reduced persistent current in phenytoin. Ensemble average current in phenytoin (60 µM; from 495 pulses) was scaled up (inset) to same peak size as control (from 297 pulses). Magnified view (main figure) illustrates that persistent current in phenytoin was reduced proportionately more than peak current. Dotted line: 0 current. (Baselines before and after 100-ms pulse are virtually noiseless because segmented average subtraction subtracts both signal and noise away from regions in which there were no channel openings at all.)
. Baselines for drug applications were taken as the average of the values before the drug and after washout of the drug so as to control for the slow decreases in sodium currents which were a factor more for peak currents than for late currents (e.g., Fig. 7).
View larger version (23K):
[in a new window]
FIG. 7.
Time course of phenytoin effects on currents. Both peak and late currents were reduced by applications of phenytoin (60 µM), with late current reduced to a disproportionately greater extent. Each point represents a run of 56 pulses lasting 750 ms.
RESULTS
Abstract
Introduction
Methods
Results
Discussion
References
60 to +30 mV.
62 mV for these neurons) (Segal 1994), there were brief, punctuated channel openings, often occurring in bursts (Fig. 1). At more depolarized potentials, the channel openings were longer and more "box-like", not interrupted by frequent closings.
View larger version (19K):
[in a new window]
FIG. 1.
Sodium currents recorded during pulses to different voltages in an outside-out patch. Records are selected traces that illustrate openings at different voltages. Late channel openings at hyperpolarized pulse voltages are more punctuated than "box-like" openings typically seen at more depolarized voltages.
). These open-channel bursts were rare: in 16 of the 17 patches, the bursts occurred in <1% of depolarizing pulses, and in many of these 16 patches, the bursts were seen far more rarely or not at all. Some of this variability may be due to differences between different neurons, but some of the variability appeared to be from the tendency of channels to go into long-lasting states that were predisposed to bursting. The sequence of bursts shown in Fig. 5 was the longest encountered. It demonstrates that these bursts often recur in consecutive pulses despite the channel being closed in the 1.25-s period at the holding potential between subsequent pulses. There was no absolute demarcation between the duration of brief late openings and the long burst openings, because events of intermediate duration often were seen. The rarity of the burst openings makes quantitative studies of the bursts difficult to do, because typically only several bursts of >100 ms would be seen in the ~1,000 pulses obtained during 30 min of recording. In addition, statistical analysis of the bursts was made more difficult by the tendency of bursts to occur in clusters and thus not represent statistically independent events.
View larger version (50K):
[in a new window]
View larger version (31K):
[in a new window]
FIG. 5.
Phenytoin reduces sodium channel openings. Each block of records represents consecutive traces, stacked so closely that peak currents are obscured for all but last trace in each block. Phenytoin (60 µM) reduced peak currents as well as both types of late currents, bursts and brief late openings, with no change in open channel amplitude. Control record displays by far most active channel bursting seen in any patch in this study; typically <1% of traces contain such burst openings. This cluster of bursts, however, serves to illustrate the tendency for bursts to recur in consecutive pulses despite fact that channel closed in each case for 1.25-s time at holding potential between subsequent pulses. Short segments of this 1.25-s interpulse time are seen at beginning and end of each displayed trace.
). Because sodium was present in the extracellular solution but absent in the intracellular solution, there was no actual reversal potential. The amplitudes for single channel currents were obtained from 6-12 single channel openings at each voltage, only measuring openings that were box-like enough to determine channel open current reliably. Because the large majority of open channel events were very brief or interrupted by many channel closings, this choice of box-like events is likely to be a more accurate method of estimating open channel current than using a total amplitude histogram. At voltages more hyperpolarized than 40 mV, the channel openings were too brief to measure the current reliably.
View larger version (10K):
[in a new window]
FIG. 2.
Single channel current-voltage curve for sodium channels in an outside-out patch. Points reflect amplitudes of 6-to 12-channel openings, selected for their box-like character so as to measure amplitude accurately.
View larger version (12K):
[in a new window]
FIG. 3.
Peak current-voltage curve for an outside-out patch.
View larger version (16K):
[in a new window]
FIG. 4.
Tetrodotoxin (TTX; 1 µM) blocked all channel currents in outside-out patches. No channel currents were seen in TTX traces, but channel currents were present in almost all control traces before and after TTX. Cesium (144 mM) in the patch pipette blocked any potassium channels.
30 mV and a phenytoin concentration of 60 µM. Analysis using the ensemble current method in other systems typically consists of a straightforward process of adding individual patch current traces such as those in Fig. 5. However, in the case of late openings of sodium channels, additional caution is needed. The late current is <1% of the peak current, similar in magnitude to typical changes in leakage current that occur between runs of voltage pulses. To add single channel events to get an ensemble current, it is best to control for baseline changes by subtracting all nonchannel segments of the record using a blank trace constructed from each run of traces (see METHODS). This process is time-consuming, but it removes all leak current, which changes significantly over time. Because this technique uses a mathematical subtraction of traces that includes the noise in the recording, the segments of the records free of channel openings can be entirely free of noise and baseline drift (e.g., the opening-free segments of Fig. 6).
100 mV and a test potential of 30 mV, late sodium currents (50-100 ms) averaged 0.48 ± 0.27% of peak current (mean ± SE; n = 3).
30 mV and a phenytoin concentration of 60 µM. The idealized traces were constructed as described in METHODS. The peak currents were measured after subtraction of calculated blank pulses as described in METHODS.
30 mV and 60 µM drug concentration.
(~0.3 ms) of the late openings was the same as control (4% increase, n.s.) and the slow (~2 ms) was decreased to 73% of control (P < 0.005). A pair of histograms from this analysis is shown in Fig. 8. These findings indicate that the major effect of the drug was to decrease the number of late channel openings with both fast and slow time courses, but there may in addition have been a small amount of additional reduction in the duration or numbers of the slow events.
View larger version (15K):
[in a new window]
View larger version (15K):
[in a new window]
FIG. 8.
Histograms of channel open time in vehicle control and phenytoin. More test depolarizations were included for the phenytoin condition than control (448 vs. 168; all from the same patch) to give comparable numbers of channel openings (2,119 vs. 2,572). Exponential curves were fit using the simplex least square algorithm in pClamp software using 2 exponentials.
1% of traces or 0-2 bursts per drug condition) and the lack of statistical independence of successive bursts in those traces that have clusters of successive bursts (e.g., Fig. 5). In contrast, the numerous brief late events were reduced by phenytoin to a degree similar to the effect on the total late events, as determined by mathematically excluding burst openings >5 ms from the analysis (cf. concentration dependence studies below) and by the similarity of results in patches in which the bursts were rare or absent.
). The 60-µM dose was high enough to obtain results that were robust to variations in channel opening such as those produced by channels bursting.
30 mV; n=4; Fig. 9). At each concentration, phenytoin reduced both the peak and persistent currents.
View larger version (16K):
[in a new window]
FIG. 9.
Dependence of peak and persistent currents on phenytoin concentration. Holding potential was 100 mV; test pulses were to 30 mV. Error bars are ±SE for n = 4 patches, with 112 depolarizations per patch at each point.
60 to 0 mV. These tests were done as multiple voltage pulses for each patch (n = 3; all at 60 µM phenytoin; Fig. 10) so as to control for possible variations due to different states or types of sodium channels in a patch. The reduction of peak currents by phenytoin was similar at all voltages tested (
), while the effect on the late currents was greater at more depolarized voltages (
).
View larger version (19K):
[in a new window]
FIG. 10.
Voltage-dependence of block of peak and persistent sodium currents by phenytoin. Persistent current ( ) was reduced more by phenytoin at depolarized potentials than at hyperpolarized potentials. In contrast, effect of phenytoin on peak currents () was similar at all voltages. Holding potential was 100 mV; 60 µM phenytoin was used. Error bars are ±SE for n = 3 patches, with 112 depolarizations per patch at each point.
DISCUSSION
Abstract
Introduction
Methods
Results
Discussion
References
; Alzheimer et al. 1993; French and Gage 1985; French et al. 1990; Masukawa et al. 1991; Stafstrom et al. 1985; reviewed in Taylor 1993). Such persistent currents appear to account for the pacemaker potentials associated with repetitive firing (Hotson et al. 1979; Stafstrom et al. 1982, 1984) and appear to underlie the ictal activity in a culture system with epileptiform activity (Segal 1994 and these studies).
) neurotransmitters (Guérineau et al. 1995; Haj-Dahmane and Andrade 1996) or calcium (Partridge and Swandulla 1988). Such currents have reversal potentials ranging from 60 to 9 mV (Alzheimer 1994; Guérineau et al. 1995). However, additional studies would be required before reaching firm conclusions as to the absence of cation channels in the system. Some posited cofactors for cation currents, such as cyclic-nucleotides, calcium and neurotransmitters, were reduced or absent in the outside-out patch recordings, so one cannot rule out the presence of cation channels from these experiments that were designed to examine sodium channels. It is also possible that the posited cation channels were blocked by the cesium used in these experiments, though described currents have been permeable to cesium (Alzheimer 1994; Guérineau et al. 1995). It is also possible that a small population of tetrodotoxin (TTX)-sensitive cation channels is buried among a more numerous population of sodium channels, though described cation currents appear to be TTX insensitive (Alzheimer 1994). Experiments designed specifically to test for cation channels are needed to resolve these questions. However, this study was aimed to identify the currents underlying endogenous bursts of action potentials and depolarization associated with ictal activity, which are TTX-sensitive events that occur in the absence of neurotransmitter action or calcium entry (Segal 1994), so those endogenous bursts likely are to be driven largely or completely by a persistent sodium current.
) or neuronal channels (Aldrich and Stevens 1987). However, some have found little difference between cell-attached patch and off-cell patch recordings in neurons (Kirsch and Brown 1989), and "box-like" channel openings that look very similar to those found here (e.g., Fig. 5) have been recorded using cell-attached patch recording (Alzheimer et al. 1993). In studies in progress in this culture system, we also have used cell-attached patch recordings and found instances in which only punctuated sodium channel openings were found as well as instances at the same voltage in which the common type of openings was box-like (M. Segal, S. Yilma, and A. Douglas, unpublished data). This indicates that the outside-out channel recordings shown here are representative of at least some physiological circumstances present in neurons. The nature of these conditions governing channel open properties is being investigated.
). The ictal neurons generally had endogenous (nonsynaptic) bursts of action potentials and depolarization, whereas interictal neurons lacked such endogenous bursts (Segal 1994). Because the endogenous bursts of action potentials and depolarization underlying the ictal epileptiform activity were TTX sensitive and occurred in the absence of calcium or neurotransmitter action, it was hypothesized that these endogenous events were driven by a persistent sodium current (Segal 1994), but this remained to be tested by voltage-clamp studies as done here.
62 mV) (Segal 1994) and at the voltage needed to sustain plateau depolarizations ( 30 mV) (Segal 1994) suggests that these late channel openings can be physiologically significant in producing epileptiform activity. Because application of the sodium channel blocking TTX abolishes the ramp depolarization that occurs at resting potential and leads to initiation of the ictal-associated endogenous bursts in neurons (Segal 1994), it is likely that these late sodium channel openings at the resting potential play an important role in initiating ictal epileptiform activity. Because the plateau depolarizations appear to underlie the continuation of ictal epileptiform activity in these neurons (Segal 1994), it is likely that the late sodium channel openings seen at approximately equal to 30 mV contribute to the current underlying the continuation of the ictal activity in this system.
; Lang et al. 1993; Matsuki at al. 1984; Willow et al. 1985). However, the use-dependence studies were done by analyzing peak sodium currents so the nature of effects on late sodium channel openings was unclear (Taylor 1993). The direct examination of persistent sodium currents allows a direct test of the effect of an anticonvulsant on a current that appears to be important in initiating and continuing epileptiform activity.
), one other study of macroscopic persistent currents has demonstrated that phenytoin reduces persistent sodium currents at low drug concentrations (Chao and Alzheimer 1995). This study, using whole cell recording, found inhibition of the persistent current by phenytoin with an EC50 of 34 µM, lower than the 70-260 µM EC50 values for peak currents reported in the literature (reviewed in Chao and Alzheimer 1995). The suppression of persistent currents to ~23% with 60 µM phenytoin reported in this paper falls directly on the macroscopic whole-neuron dose response curve in the Chao and Alzheimer (1995) study, providing confirmation for both the single channel and the whole neuron methods for analyzing late sodium currents.
at that concentration. The results could differ for a variety of reasons, such as the pulse parameters or the interpulse intervals (1.25 s here). Also, single channel methods may be inherently more accurate because one can correct for changes in leak current down to the level of single channel openings as was done here. Because 8 µM phenytoin abolishes the plateau depolarizations in those neurons that are associated with ictal activity (Segal 1994), the effect of phenytoin on the persistent sodium current appears to be sufficient to account for the anti-epileptic action of the drug in this system.
proposing that phenytoin acts primarily by stabilizing inactivated states of the sodium channel, which would lead to fewer channel openings. Because the voltage dependence of the phenytoin effect on the persistent sodium current was substantial, and no voltage dependence of drug effect on peak currents was found (Fig. 10), this suggests that one of the steps involved in phenytoin's effect on the late channel openings is voltage dependent.
| |
ACKNOWLEDGEMENTS |
|---|
We thank D. Zurakowski for assistance with statistical tests and N. Gatt for expert technical assistance and preparation of cultures. We also thank B. Bean, S. Lipton, and Z.-H. Pan for assistance with single channel recording techniques and V. Chen, S. Lipton, Z.-H. Pan, and C. Stafstrom for comments on earlier versions of this manuscript.
These studies were supported by grants from The New York Community Trust and the Klingenstein Fund to M. M. Segal as well as by grants from the William Randolph Hearst Fund and the William F. Milton Fund.
| |
FOOTNOTES |
|---|
Address for reprint requests: M. Segal, Enders 3 Neurology, Children's Hospital, 320 Longwood Ave., Boston MA 02115.
Received 18 October 1995; accepted in final form 4 February 1997.
| |
REFERENCES |
|---|
|
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
This article has been cited by other articles:
|
|
 |
|
  M. M. Segal, G. F. Rogers, H. L. Needleman, and C. A. Chapman Hypokalemic Sensory Overstimulation J Child Neurol, December 1, 2007; 22(12): 1408 - 1410. [Abstract] [PDF]
|
|
|
|
 |
|
 C. Yvon, A. Czarnecki, and J. Streit Riluzole-Induced Oscillations in Spinal Networks J Neurophysiol, May 1, 2007; 97(5): 3607 - 3620. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Castro-Alamancos, P. Rigas, and Y. Tawara-Hirata Resonance (~10 Hz) of excitatory networks in motor cortex: effects of voltage-dependent ion channel blockers J. Physiol., January 1, 2007; 578(1): 173 - 191. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Aracri, E. Colombo, M. Mantegazza, P. Scalmani, G. Curia, G. Avanzini, and S. Franceschetti Layer-Specific Properties of the Persistent Sodium Current in Sensorimotor Cortex J Neurophysiol, June 1, 2006; 95(6): 3460 - 3468. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Remy and H. Beck Molecular and cellular mechanisms of pharmacoresistance in epilepsy Brain, January 1, 2006; 129(1): 18 - 35. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. G. Vanoye, C. Lossin, T. H. Rhodes, and A. L. George Jr. Single-channel Properties of Human NaV1.1 and Mechanism of Channel Dysfunction in SCN1A-associated Epilepsy J. Gen. Physiol., December 27, 2005; 127(1): 1 - 14. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 |
