corrente de calcio e sincronização da da 2008

7/28/2019 corrente de calcio e sincronizao da DA 2008

1/11

Research Report

The slow Ca2+-dependent K+-current facilitates

synchronization of hyperexcitable pyramidal neurons

Jane Skov, Steen Nedergaard, Mogens Andreasen

Department of Physiology and Biophysics, Aarhus University, DK-8000 rhus C, Denmark

A R T I C L E I N F O A B S T R A C T

Article history:

Accepted 13 November 2008

Available online 25 November 2008

Studies on in vivo and in vitro epilepsy models have shown that progression and

maintenance of epileptiform activity can be affected by the slow Ca2+-dependent K+

current (IsAHP). This study aimed to investigate the influence of the IsAHP on population

activity and single cell activity during the transition from the interictal- to the ictal-like

phase of an epileptiform field potential induced by Cs+. Extracellular and intracellular

recordings were performed in area CA1 on 400 m thick hippocampal slices from adult male

Wistar rats. During maintained exposure to Cs+, the transition between the two phases

underwent a slow, time-dependent change, where synchronized population activity

gradually disappeared and a plateau-like prolongation of the interictal-like phase

emerged. In parallel, the size of the ictal-like phase increased. These effects could be

ascribed to a gradual block of the IsAHP and were mimicked by the IsAHP antagonistscarbacholine (2 M), isoproterenol (4 M) and Ba2+ (0.2 mM). Cessation of population activity

generally occurred without a concomitant decrease in firing rate of single CA1 pyramidal

neurons, but was accompanied by the disappearance of hyperpolarizing prepotentials,

indicating a shift in the mechanism of action potential initiation. These findings suggest

that the presence of the IsAHP increases the tendency of hyperexcitable neurons to fire in

synchrony, but at the same time serves to dampen the ictal-like activity that follows the

hyperexcitable state. Our data indicate that both effects can be attributed to the influence of

this current on the steady-state membrane potential in the period of the transition from

interictal- to ictal-like activity.

2008 Elsevier B.V. All rights reserved.

Keywords:

Rat

Hippocampus

Afterhyperpolarization

Cesium

Epileptiform

1. Introduction

Epileptic seizures are paroxysmal events in the central

nervous system caused by abnormal, excessive, synchronized

discharges in aggregates of neurons. From electroencephalo-

graphic recordings of patients with focal epilepsies, brief,sharp spikes can be observed between seizures, and are as

such referred to as interictal activity. The relation between

B R A I N R E S E A R C H 1 2 5 2 ( 2 0 0 9 ) 7 6 8 6

Corresponding author. Fax: +45 86129065.E-mail address: [email protected] (J. Skov).Abbreviations: 4-AP, 4-aminopyridine; AHP, Afterhyperpolarization; APV, DL-2-amino-5-phosphonopentanoic acid; BIC, Bicuculline

methobromide; CNQX, 6-cyano-7-nitroquinoxaline-2,3-dione; Cs-FP, Cesium-induced field potential; ImAHP, The Ca2+-dependent K+-

current underlying the medium afterhyperpolarization; IsAHP, The Ca2+-dependent K+-current underlying the slow afterhyperpolarization;

mGLuR, Metabotropic glutamate receptor; LY341495, (2S)-2-amino-2[1S,2S9-2-carboxycycloprop-1-yl]-3-(xanth-9-yl)propanoic acid; SD,Spreading depression

0006-8993/$ see front matter 2008 Elsevier B.V. All rights reserved.doi:10.1016/j.brainres.2008.11.043

a v a i l a b l e a t w w w . s c i e n c e d i r e c t . c o m

w w w . e l s e v i e r . c o m / l o c a t e / b r a i n r e s
mailto:[email protected]://dx.doi.org/10.1016/j.brainres.2008.11.043http://dx.doi.org/10.1016/j.brainres.2008.11.043mailto:[email protected]


2/11

interictal activity and seizures, also known as ictal activity,

is still a matter of debate, but it seems evident that ictal

activity cannot merely be attributed to a summation of

interictal discharges. The interplay of cellular mechanisms

causing the two different forms of epileptiform activity

appears to be highly complex, and interictal bursts have

been proposed both to precipitate ictal activity (Traynelis and

Dingledine, 1988) and to dampen it (Avoli, 2001).It is well established that extracellular Cs+ (35 mM) can

induce epileptiform activity in neocortical or hippocampal

brain slices (Hwa and Avoli, 1991; D'Ambrosio et al., 1998;

Janigro et al., 1997; Xiong and Stringer, 1999, 2001). Cs+ has

several known, and potentially epileptogenic, effects on

neurons and glia, and has been shown to induce spontaneous

epileptiform activities that depends on both synaptic (Hwa

and Avoli, 1991; Xiong and Stringer, 1999) and non-synaptic

mechanisms (Xiong and Stringer, 2001). In addition,long-term

application of Cs+ promotes an epileptogenic field potential

(Cs-FP) in area CA1of the rathippocampal slice,which requires

stimulation of the Schaffer collateralcommissural fibers but

persists in the presence of ionotropic glutamatergic and

GABAergic receptor antagonists (Skov et al., 2005). The Cs-FP

is biphasic, consisting of an initial positive phase, which seems

to depend on an enhancement of a synaptic signal, as the

response is sensitive to blockade of synaptic transmitter

release, in addition to a general enhancement in excitability

(Skov et al., 2005, 2006). Hitherto, the nature of the synaptic

signal has remained elusive. The positive phase is followed by

a long-lasting negative phase, which likely reflects a transient

increase in extracellular K+, as it is sensitive to alterations in

glial K+ buffering capacity (Andreasen et al., 2007). As the

positive and the negative phase resemble the interictal and

ictal events, respectively, induced by increasing extracellular

K+ (Jensen and Yaari, 1988; Traynelis and Dingledine, 1988),we

have, for descriptive purposes adopted the terms interictal-

like and ictal-like for the positive and negative phase,

respectively.

Preliminary observations have indicated that the transition

from interictal- to ictal-like activity is under the influence of

the Ca2+-dependent K+-current underlying the slow after-

hyperpolarization (IsAHP). Because of its susceptibility toregulation by neurotransmitters and drugs, the potential role

of the IsAHP in the induction and maintenance of ictal activityis of considerable interest. Indeed, the IsAHP has beensuggested to terminate epileptic afterdischarges (Traub and

Jefferys, 1994) andto prevent ictal discharges (Alger andNicoll,

1980). Blockade of the IsAHP has also been put forward as onesource for the progression of epileptiform activity (Martn

et al., 2001; McCormick and Contreras, 2001). A reduced IsAHPappears to be one of the hallmarks of several in vivo models of

epilepsy (Verma-Ahuja et al., 1995;Watts et al., 1993; Wu et al.,

2003) as is indeed found in dentate granula cells from patients

with clinical seizures (Williamson et al., 1993). However,

which specific cellular and network properties that are

influenced by the IsAHP, and how, is still not fully understood.The aim of the present study was therefore to investigate the

possible influences of the IsAHP on the shaping and progressionof the epileptiform activity induced by Cs+, including its

impact on population activity and on discharge of single

pyramidal neurons. By illuminating new aspects of the role of

the IsAHP, this study provides novel insights into the cellularmechanisms important for ictogenesis.

Some of the results have been presented in abstract form.

2. Results

After 4060 min application of Cs+ (5 mM) together with DL-2-amino-5-phosphonopentanoic acid (APV, 50 M), 6-cyano-7-

nitroquinoxaline-2,3-dione (CNQX, 10 M) and bicuculline

methobromide (BIC, 10 M), a fully developed epileptiform

field potential (Cs-FP) was recorded in area CA1 in response to

Schaffer collateralcommissural fiber stimulation. As

described previously (Andreasen et al., 2007; Skov et al., 2005),

when therecording wasmade in stratumpyramidale, theCs-FP

consisted of an initial positive phase followed by a longer

lasting (up to several seconds) negative phase (see Fig. 1).

Synchronous spike activity was present during the positive

phase and continued through the transition period into the

negative phase (Figs.1Aand2A).We monitored theevolvement

of theCs-FP in an 80 minperiod after itsappearance. The initial

burst was maximally expressed in the early stages of the

development of theCs-FP. In thefollowing period, the length of

the burst was gradually reduced. Eventually it became

restricted to the upstroke region of the positive phase. Thus,

after a period of 2060 min, spike activity was abolished in the

transition region between the two phases (Figs. 1A and 2A)

but continued during the initial part of the positive phase

and, usually, during the negative phase. Concomitant with

the alteration in firing, we noted that the decay of the positive

phase became slower, resulting in a progressive broadening

of this part of the Cs-FP to the point where it in some cases

attained the appearance of a plateau (Fig. 1A). Averaged

measurements revealed that this broadening took place

mainly during the later stage of the development of the Cs-

FP (Fig. 1B).

In an attempt to quantify the changes in synchronized

firing during the transition from the positive to the negative

phase of the Cs-FP, we applied a measurement referred to as

the coastline index (Korn et al., 1987; see also Experi-

mental procedures). The limits of the transition period were

tentatively defined as the part of the response stretching

from the peak of the positive phase to 200 ms from the

stimulus (Fig. 2A). As a control measure, we included the

coastline index for the pre-peak region, stretching from the

stimulus to the peak of the positive phase (Fig. 2A). No

significant changes were found in the average coastline

index of the pre-peak region, which amounted to 10814%

of the initial value after 30 min (Fig. 2B, n =7, P =0.57). Incontrast, the coastline index of the transition period showed

a large decrease in all slices (Figs. 2A and B). The decrease

was progressive and reached 846% on average at the end

of the observation period (30 min, Fig. 2B), an effect which

was highly significant (P


3/11

Since population spikes were not generally suppressed

over time (Figs. 2A and B) the decay of spike activity in the

transition period is likely to involve some specific process(es)

related to the evolvement of the Cs-FP.

2.1. A slow Ca2+-dependent K+ current is important for

the stability of the transition period

We hypothesized that the observed changes could reflect an

alteration of a membrane conductance which influences the

spiking for a period corresponding to the transition region. A

plausible candidate could be the Ca2+-dependent K+ current

underlying the slow afterhyperpolarization (IsAHP), becausethe time-course of this current is long enough to affect the

region of the Cs-FP in question. We therefore tested the effect

of blockers of this current. Application of carbacholine (2 M)

led to a rapid increase in the duration of the positive phase

(Fig. 3A). As exemplified in Fig. 3B, this effect was marked and

partly reversible upon wash out. On average, carbacholine

increased the duration of the positive phase from 12414 ms

to 25124 ms within 6 min (n =12, P


4/11

A medium duration Ca2+-dependent afterhyperpolariza-

tion has been described in CA1 pyramidal neurons (Halliwell

and Adams, 1982). To block the current (ImAHP) behind thisafterhyperpolarization we applied apamine (0.11 M). This

treatment had no discernible effects on the Cs-FP (n=4, notillustrated), making it unlikely that ImAHP makes any majorcontribution to this potential.

To test if the presence of Cs+ is needed for the pharmaco-

logical effect on the interictal-like phase, we used 4-amino-

pyridine (4-AP, 0.5 mM) to generate a synaptic field potential

which is equivalent to the initial part of the Cs-FP (Andreasen

et al., 2007; Skov et al., 2005). In all slicestested (n=4), wefoundthatcarbacholine (2 M) gave a markedand reversible increase

in the duration of the field potential, from 82 10 ms to 203

24 ms (P=0.02).

2.2. The negative phase is enhanced concomitantly with

IsAHP blockade

In addition to the effects described above, we observed that

the presence of isoproterenol was associated with an enlarge-

ment of the negative, ictal-like phase of the Cs-FP (Fig. 4A).

Averaged measurements of the negativity at 1000, 1500 and

2000 ms after stimulation showed a significant increase after

8 min perfusion with 4 M isoproterenol (Fig. 4B, n =8,P


5/11

continuous adjustment of the holding current. We found that

the amplitude of the slow AHP decreased considerably, and in

two recordings it was reversed to an afterdepolarization. On

average, the AHP was reduced to 416% of the initial value, a

highly significant change (Fig. 5B, P


6/11

2.4. Change in discharge mechanism accompanies

population activity decay

In spite of a similar frequency, there were marked differences

in the properties of the firing before and after decay of

population activity. First, measurements in each neuron of the

mean transmembrane potential in the transition period (see

Experimental procedures) revealed a markedly larger depo-

larization of neurons recorded after the decay (average values

before: 41.4 2.7 mV, after: 27.6 3.8 mV; n =10). Thisdifference was statistically significant (P< 0.01). Since the

Fig. 5 Effects of Cs+ on the slow afterhyperpolarization and

action potential firing. (A) Membrane potential recordings

from a pyramidal neuron before and after 60 min application

of Cs+. The amplitude of the afterhyperpolarization was

measured 200 ms after (marked by an arrowhead) the end of

a 600 ms depolarizing current pulse (0.9 nA). The cell was

manually clamped to the pre-Cs+ membrane potential. (B)

The average amplitude of the slow afterhyperpolarization as

a function of time in Cs+ (n=5). (C and D) Membrane potential

recordings from two different pyramidal neurons during a

Cs-FP. The records were taken at a time where the

extracellular Cs-FP no longer displayed synchronized firing

in the transition region. Note that both cells fire action

potentials during the period corresponding to the transition

period(marked by a white bar) when themembrane potential

is held at the pre-Cs+ value by hyperpolarizing current

injection (lower panels). When the clamp was removed

(upperpanels) thecell in C displayedenhanced firing activity,

whereas the cell in D no longer fired action potentials in this

region.

Fig. 4 Isoproterenol enhances the negative phase. (A) Cs-FP

recorded before and after 8 min application of isoproterenol

(4M). The early parts (marked by double arrows) of the

responses are shown on an expanded time-scale below.Note

thatin addition to the disappearanceof spiking activityin the

transition region and the prolongation of the positive phase,

the negative phase is prolonged. (B) Averageamplitude of thenegative phase measured at four different time points after

the stimulation in control conditions (white columns, n= 8)

and after application of isoproterenol (black columns).

* denotes a statistically significant difference between control

and isoproterenol.

81B R A I N R E S E A R C H 1 2 5 2 ( 2 0 0 9 ) 7 6 8 6


7/11

contribution from action potentials was approximately equal,

thedifferencebetweenthesevalues mainly reflects a difference

in tonic membrane potential. The average resting membrane

potential was not significantly different in the two groups

(before: 60.52.6 mV; after: 58.72.1 mV; P>0.05). Second,beforethe decay, individual action potentialswere often seen to

be triggered abruptly from a transient hyperpolarizing

prepotential. Such prepotentials coincided with a larger

negative deflection in extracellular voltage (Fig. 6A), which

indicated that these action potentials were triggered by a

depolarization of the transmembrane potential caused by

transfer of current from neighboring neurons (field effects,

see Jefferys, 1995; Taylor and Dudek, 1984). Conversely, after

decay of the population activity, action potentials were usually

preceded by a depolarizing prepotential (Fig. 6B). In order to

obtain a quantitative estimate of this possible change in

discharge property, we counted, in one sweep for each neuron,

the number of action potentials which displayed a hyper-

polarizing prepotential. The prepotential was defined as the

mean potential 02 ms before the action potential minus the

mean potential in the preceding 2 ms, i.e. 24 ms before the

action potential. Application of this analysis in 10 neurons,

recorded before the decay of population activity, revealed that

hyperpolarizing prepotentials were present in 41 out of 62

action potentials (66%). In 8 neurons, recorded after the decay,

only one action potential out of 46 had a hyperpolarizing

prepotential (2%).

2.5. The possible involvement of metabotropic glutamate

receptors in the development of the Cs-FP

Finally, we sought to address the cause of the Cs+-induced

block of the IsAHP. Since activation of metabotropic glutamatereceptors (mGluRs) is known to block the IsAHP (for review, seeAnwyl, 1999), we hypothesized that Cs+-induced tonic release

of glutamate onto mGluRs is the reason for the block of the

IsAHP and hence for the observed changes in the transitionperiod. In support of such a mechanism, we have previously

observed that the nonselective mGluR antagonist (2S)-2-

amino-2[1S,2S9-2-carboxycycloprop-1-yl]-3-(xanth-9-yl)

propanoic acid (LY341495, 100 M) reduces the duration of the

positive phase by 21% on average (Skov et al., 2005). Our

hypothesis therefore predicts that co-application of Cs+ and

LY341495 will prevent the decay of synchronized activity. We

therefore appliedLY341495 (100M) after 20 minperfusionwith

Cs+, i.e. at a time long before theCs-FP had fully developed. We

found that the decrease in coastline index of the transition

period was 7815% in the presence of LY341495 (n=4, not

Fig. 6 Change in neuronal firing properties during decay of population activity. (A and B) Simultaneous intracellular

(upper sweep) and extracellular (lower sweep)recordings of the Cs-FP in its early phase (A) and, in the same neuron, after decay

of population spikes in the transition period (B). The horizontal broken lines indicate 50 mV. The baseline Vm was at resting

potential in both records. Note larger membrane depolarization during the action potential burst in B compared to A. Records

from the periods of action potential firing in the transition period (marked a and b) are shown enlarged in right panels.

Arrowheads in a mark the transient, hyperpolarizing deflections that precede action potentials. These show exact

correspondence in time to negative deflections in the extracellular potential, indicative of field interactions being the trigger

source for the action potentials. After decay of population firing the action potentials were preceded by a depolarizing

prepotential (b). Note lack of coincidence of action potentials with population spikes. Calibrations in B also apply to A. All

voltages are recorded with respect to a distant ground electrode.

82 B R A I N R E S E A R C H 1 2 5 2 ( 2 0 0 9 ) 7 6 8 6


8/11

illustrated), a value which is not substantially different from

the 846% decrease observed in the absence of LY341495. An

unpaired t-test assuming equal variances (F-test: P=0.25) didnot show any statistical significance (P=0.42) between theslices exposed to LY341495 and the controls. Similar results

were obtained when Cs+ wasco-appliedwith atropine(n=3,notshown).

3. Discussion

The present study suggests that blockade of the IsAHP i) causesa reduction in population activity specifically during the

transition from interictal- to ictal-like activity, ii) induces a

plateau-like prolongation of the positive phase, and iii) leads

to an enhancement of the negative phase. These changes

occurred rapidly in response to carbacholine, isoproterenol or

Ba2+, which have the IsAHP as common target, but they alsodeveloped gradually during prolonged application of Cs+. The

latter effect seems to be ascribed to a slow antagonizing effect

of Cs+ on the IsAHP.

3.1. How does Cs+ block the IsAHP?

The reduction in AHP amplitude observed in Cs+ is unlikely to

result from a reduced driving force for the underlying K+

current, because the membrane conductance during the AHP

was severely reduced after application of Cs+, which is

strongly indicative of a channel block. The mechanism of

this blocking effect was not pursued in the present study, but

results from the literature and previous studies on our lab,

points to several possible causes.

Blockade of the IsAHP, through activationof mGluRs has beenreported in 4-AP induced epileptiform activity (Martn et al.,

2001). We found that prior blockade of mGluRs or muscarinic

receptorsdid notprevent theeffects of Cs+, indicatingthat these

receptors were not involved. Since the IsAHP is also sensitive tootherneurotransmitters,e.g. noradrenalinand 5-HT(Sah,1996),

it is conceivable that the observed effect results from a

combined activation of several metabotropic receptor types.

Alternatively, intracellularly accumulating Cs+ in pyramidal

neurons could exert a direct block of the Ca2+-dependent K+

channels (de Sevilla et al., 2006) or indirectly affect the channels

through interference with signal transduction mechanisms

(Brette et al., 2003). Such mechanism seems plausible, since it

requires transport of Cs+ across the membrane, which would

explain the slow time course of the block. Intracellular

accumulation of high concentrations of Cs+ has been demon-

strated in other tissues in response to long-term perfusion of

the ion (Schornack et al., 1997), and our previous studies show

that Cs-FP development is sensitive to blockade of the Na/K

pump (Skov et al., 2006), which is a known route for trans-

membrane transport of Cs+ (Akera et al., 1979). Finally, it cannot

be excluded that the epileptiform activity, in itself, somehow

could affect the IsAHP. In favor of the latter explanation is theobservation that wash out of Cs+ leads to a relatively fast

recovery of synchronized activity during the transition period

which occurs in parallel with the disappearance of the ictal-like

negative phase (Andreasen et al., 2007). Further studies will be

needed to fully resolve this issue.

3.2. Influences of IsAHP blockade on neuronal and

population discharge

Our data from intracellular recordings showed that the

majority of neurons (eight out of ten) recorded after block of

population activity, firedat frequencies thatwere not different

from those recorded before the block. These data therefore

point to desynchronization as the major cause of the block ofpopulation spike firing. The mechanism by which the IsAHPcontributes to spike synchronization is obviously different

from its influence on burst synchronization between CA3

pyramidal neurons (de Sevilla et al., 2006), which occurs at a

time-scale compatible with the slow kinetics of the IsAHP.Instead, the effect observed here is likely to be a more indirect

one, related to the effect of the IsAHP on neuronal membranepotential in the period after the initial burst (see below). The

frequency of population spikes was found to be significantly

higher than the discharge frequency of single neurons,

suggesting that it represents the composite activity of several

smaller neuronal aggregates (see Bikson et al., 2003a). Since

most of our recordings showed that during the loss of

synchronized activity there was no concomitant change in

the population spike frequency, it seems unlikely that

desynchronization involves a change in the number of

neuronal aggregates, but rather a gradual reduction of their

size. To explain this result, we suggest that membrane

depolarization following the IsAHP block could increase theprobability that some neurons would begin to fire indepen-

dently (i.e. fall out of phase with their neighbors) and hence be

dismembered from the aggregate. This interpretation is in

accordance withthe observeddifference in average membrane

potential andin the frequencyof prepotentials before andafter

suppression of population activity. It furthermore implies that

thereasonwhy thedischarge of individualcells could be largely

constant through this period, in spite of a pronounced

membrane depolarization, was because it shifted from an

external drive (field effect) to an internal drive (membrane

potential). In addition however, we observed in two out of ten

neurons a transient cessation of firing corresponding to the

transition period. This block was associated with plateau-like

depolarizationpositiveto actionpotentialthreshold,suggesting

that depolarization-induced inactivation of Na+ channels could

be the cause. If a considerable fractionof neurons enters such a

state of depolarization block, this will contribute further to a

reduced population activity, and hence partially explain the

observed run down.

3.3. IsAHP and epileptiform activity

Conclusive evidence for the role of the IsAHP under hyper-excitable conditions is still rather limited, although it has long

been proposed that blockade of this current is a key feature in

transition from interictal to ictal discharges (Alger and Nicoll,

1980; Dichter and Ayala, 1987). IsAHP blockade seems anecessary mechanism for 4-AP induced spontaneous epilepti-

form activity (Martn et al., 2001), and by reducing the slow

AHP, isoproterenol has been shown to increase the rate of

epileptiform discharges induced by either GABAA-blockade or

an increased extracellular concentration of K+ (Rutecki, 1994).

In agreement with these studies, the present results suggest

83B R A I N R E S E A R C H 1 2 5 2 ( 2 0 0 9 ) 7 6 8 6


9/11

that in the Cs+ model of epileptiform activity, blockade of the

IsAHP contributes to the hyperexcitable state that leads to theepileptiform burst activity. Specifically, we observed that the

negative, ictal-like phase of the Cs-FP was amplified in

response to IsAHP blockade. Furthermore, our data indicatethat the size of the ictal-like phase is correlated to the level of

tonic membrane depolarization during the transition period.

Discharge-dependent release of K+ is believed to be a majorcause of non-synaptic seizure induction (Yaari et al., 1986).

During ictal-like activity, however, the extracellular K+

concentration has been shown to remain elevated in periods

of interruptions of neuronal firing (due to a depolarizing

block), and the continued K+ release in that situation is most

likely promoted by persistent inward currents that depolarize

the membrane to maintain the driving force for K+-efflux

(Bikson et al., 2003b). In line with this notion, we propose that

under our experimental conditions (i.e. with action potential

activity preserved), an increase in steady-state membrane

depolarization can still favor such discharge-independent K+-

release and hence contribute to the progression of interictal to

ictal activity. More detailed studies of this particular aspect of

ictogenesis will be needed before its overall significance can be

assessed.

4. Experimental procedures

All experimental protocols were in accordance with university

guidelines for animal research and complied with Danish and

European law on the care and use of laboratory animals.

Experiments were performed on hippocampal slices prepared

from 41 male Wistar rats (250300 g). The rats were anesthe-

tized with isoflurane and decapitated. The brain was quickly

removed and placed in a dissection medium (see below) at

4 C. The hippocampus was dissected free and 400 m thick

slices were cut on a McIlwan tissue chopper. One slice was

immediately transferred to a recording chamber, where it was

placed on a nylon-mesh grid at the interface between warm

(3132 C) standard perfusion medium (see below) and warm

humidified carbogen (95% O2, 5% CO2). Perfusion flow rate was

1 ml/min. The slice was allowed to rest for 1 h before

recordings were started. The remaining slices were kept in a

storage chamber at room temperature.

4.1. Electrophysiology

We used borosilicate glass microelectrodes (1.2 mm o.d,

Harvard Apparatus, Edenbridge, UK) filled with 1 M NaCl (tip

resistance 525 M) for extracellular recordings, or 3 M KCland

0.1 M K+ acetate (tip resistance 5070 M) for intracellular

recordings. In combined recordings, the electrodes were

placed as closely as possible. A bipolar teflon-insulated

platinum electrode (tip diameter 50 m, intertip distance

25 m) placed in stratum radiatum at the border between area

CA3and area CA1was used for orthodromic stimulation of the

Schaffer collateralcommisural fibers with constant-current

pulses (50 s, 0.20.4 mA) at a frequency of 0.05 Hz.

Conventional recording techniques were employed, using a

high input impedance amplifier (Axoclamp 2A, Axon Instru-

ments) with bridge-balance and current-injection facilities.

Signals were digitized on-line using a labmaster A/D converter

and transferred to a PC employing pCLAMP acquisition

software (Axon Instruments). Signal analysis was performed

using pCLAMP analysis software.

Slices were accepted if they, during control conditions,

displayed a normal orthodromic field excitatory postsynaptic

potential with a single population spike (amplitude between 5

and 15 mV) and showed no additional spikes with supra-maximal stimulation.

Application of Cs+ sometimes leads to the occurrence of

spreading depression (Skov et al., 2005; Xiong and Stringer,

1999). After a spreading depression episode (SD), the field

potential was monitored closely and recordings were stopped

if the field potential did not fully return to its pre-SD level.

Recordings were also stopped if more than three SDs occurred.

With those criteria, we found no differences between results

obtained in the absence or presence of SDs, in agreement with

previous findings (Skov et al., 2005). In the data set where the

temporal changes during wash in of Cs+ were evaluated, only

experiments free of SDs were included.

When intracellular recordings were used, cells were

accepted if they had a resting membrane potential 15 M and action potential height >80 mV.

4.2. Data analyses

In extracellular recordings, all amplitudes weremeasured with

respect to the prestimulus baseline on three to eight averaged

traces. Since population spikes occurred during the rising

phase of the Cs-FP,measurements of therate-of-rise could not

be performed. The duration of the Cs-FP was measured from

the onset of the positive phase to the time where the potential

reached or crossed the pre-stimulus baseline level.

We used coastline index as a quantitative approximation

of the amount of synchronized spike activity during the field

potential. The coastline index express the total length of a line

between two points (Korn et al., 1987). We accomplished this

by using the Pythagorean Theorem to calculate the distances

between individual points on the traces digitized at 0.55 kHz.

The sum of the individual distances gives the coastline index

and has the unit:ffiffiffiffiffiffiffiffiffiffi ffiffiffiffiffiffiffiffiffiffiffi ffiffiffiffiffiffims2 +mV2

q. This parameter is influenced

by any perturbation in the recorded voltage; however fast,

repetitive signals, such as population spikes and high

frequency noise have far more impact on coastline index

than an underlying slow waveform. Therefore, for the coast-

line index to selectively reflect theamount of population spike

activity, the contribution from noise needed to be subtracted.

This was done for each experiment by obtaining the coastline

index from a section of the prestimulus baseline voltage,

where no population spikes were present (ie only containing

noise signals), and subtracting this index from the coastline

indexes obtained during the Cs-FPs from the same experiment

(it was assumed that the level of noise was constant during

each experiment). It should be noted that after noise filtration

the coastline index is highly sensitive to changes in both

frequency and amplitude of population spikes, and does not

discriminate between the two. To be able to pool data from

different experiments, it was necessary to compensate for

variation in the coastline index between experiments. This

was accomplished by normalizing the coastline index in each

84 B R A I N R E S E A R C H 1 2 5 2 ( 2 0 0 9 ) 7 6 8 6


10/11

experiment with respect to the value recorded at time zero

(see below) set to 100%. All values of coastline index reported

here reflect the average of four individual traces.

When temporal development was considered, time zero

was set to be the time after startingperfusion with Cs+ where a

clear biphasic potential had developed with population spikes

occurring both during and after the positive phase (see Fig. 2).

This procedure was adopted in order to compensate for thedifferences in induction time of the Cs-FP.

The amplitude of the AHP was measured as the difference

between the pre-stimulus membrane potential and the

membrane potential measured 200 ms after the end of a

600 ms depolarizing pulse (0.9 nA).

The transmembrane potential was calculated as the

difference between the extracellular potential, measured by

an electrode placed close to the recorded neuron, and the

intracellular potential.

4.3. Statistical analysis

Values are given as meanSEM unless otherwise indicated.

For statistical evaluation the paired or unpaired t-test wasused as appropriate with a level of significance set at 5% (two-

sided value). Before the unpaired t-test was applied, theassumption of equal variances was checked using the F-test.n denotes the number of slices or cells used. For eachexperimental approach, slices from more than one animal

were used. In intracellular experiments, one cell from each

slice was studied.

4.4. Drugs and solutions

The composition of the dissection medium was (in mM): NaCl,

120; KCl, 2;KH2PO4, 1.25; HEPES acid, 6.6; NaHEPES, 2.6; NaHCO3,

20; D-glucose, 10;CaCl2,2;MgSO4, 2; bubbledwith carbogen. The

composition of the standard perfusion medium was (in mM):

NaCl, 124; KCl, 3.25; NaH2PO4, 1.25; NaHCO3, 20; CaCl2,2;MgSO4,

2; D-glucose, 10; bubbled with carbogen (pH 7.3). In experiments

where BaCl2 was applied,phosphate and sulphate was omitted

from the standard perfusion medium, in order to prevent

precipitation. Unless otherwise noted, the experiments with

Cs+ were all performed in the presence of CNQX (10 M), APV

(50 M) and BIC (10 M).

The pharmacological compounds were made up in stock

solutions of 1001000 times the required final concentration

and diluted in the standard perfusion medium as appropriate.

4-aminopyridine (4-AP), atropine, BIC, carbacholine, CNQX

and isoproterenol were purchased from Sigma, apamine from

Alomone Labs, APV from Ascent Scientific and LY341495 from

Tocris.

Conflict of interest statementNone of the authors has any conflict of interest to disclose.

Acknowledgments

This project was supported by grants from the Lundbeck

Foundation, The Aarhus University Research Foundation and

The Danish Medical Research Council. The authors would like

to express their gratitude to Bertha P. B Mortensen and Neven

Akrawi for excellent technical assistance.

R E F E R E N C E S

Akera, T., Gubitz, R.H., Brody, T.M., Tobin, T., 1979. Effects ofmonovalent cations on (Na++ K+)-ATPase in rat brain slices.Eur. J. Pharmacol. 55, 281292.

Alger, B.E., Nicoll, R.A., 1980. Epileptiform burstafterhyperpolarization: calcium-dependent potassiumpotential in hippocampal CA1 pyramidal cells. Science 210,11221124.

Andreasen, M., Skov, J., Nedergaard, S., 2007. Inwardly rectifying K+

(Kir) channels antagonize ictal-like epileptiform activity in areaCA1 of the rat hippocampus. Hippocampus 17, 10371048.

Anwyl, R., 1999. Metabotropic glutamate receptors:electrophysiological properties and role in plasticity. Brain Res.Rev. 29, 83120.

Avoli, M., 2001. Do interictal discharges promote or control

seizures? Experimental evidence from an in vitro model ofepileptiform discharge. Epilepsia 42 (Suppl. 3), 24.

Bikson, M., Fox, J.E., Jefferys, J.G.R., 2003a. Neuronal aggregateformation underlies spatiotemporal dynamics of nonsynapticseizure initiation. J. Neurophysiol. 89, 23302333.

Bikson, M., Hahn, P.J., Fox, J.E., Jefferys, J.G.R., 2003b.Depolarization block of neurons during maintenance ofelectrographic seizures. J. Neurophysiol. 90, 24022408.

Brette, F., Lacampagne, A., Sall, L., Findlay, I., Le Guennec, J.Y.,2003. Intracellular Cs+ activates the PKA pathway, revealing afast, reversible, Ca2+-dependent inactivation of L-type Ca2+

current. Am. J. Physiol. Cell Physiol. 285, C310C318.Colino, A., Halliwell, J.V., 1993. Carbachol potentiates

Q current and activates a calcium-dependent non-specificconductance in rat hippocampus in vitro. Eur. J. Neurosci. 5,

11981209.Congar, P., Leinekugel, X., Ben-Ari, Y., Crpel, V., 1997. A

long-lasting calcium-activated nonselective cationic current isgenerated by synaptic stimulation or exogenous activation ofgroup I metabotropic glutamate receptors in CA1 pyramidalneurons. J. Neurosci. 17, 53665379.

D'Ambrosio, R., Wenzel, J., Schwartzkroin, P.A., McKhann, G.M.,Janigro, D., 1998. Functional specialization and topographicsegregation of hippocampal astrocytes. J. Neurosci. 18,44254438.

Dichter, M.A., Ayala, G.F., 1987. Cellular mechanism of epilepsy: astatus report. Science 237, 157164.

de Sevilla, D.F., Garduo, J., Galvn, E., Buo, W., 2006.Calcium-activated afterhyperpolarizations regulatesynchronization and timing of epileptiform bursts inhippocampal CA3 pyramidal neurons. J. Neurophysiol. 96,30283041.

Fraser, D.D., MacVicar, B.A., 1996. Cholinergic-dependent plateaupotential in hippocampal CA1 pyramidal neurons. J. Neurosci.16, 41134128.

Halliwell, J.V., Adams, P.R., 1982. Voltage-clamp analysis ofmuscarinic excitation in hippocampal neurons. Brain Res. 250,7192.

Hess, P., Tsien, R.W., 1984. Mechanism of ion permeation throughcalcium channels. Nature 309, 453456.

Hotson, J.R., Prince, D.A., 1980. A calcium-activatedhyperpolarization follows repetitive firing in hippocampalneurons. J. Neurophysiol. 43, 409419.

Hwa, G.G.C., Avoli, M., 1991. Cesium potentiates epileptiformactivities induced by bicuculline methiodide in rat neocortexmaintained in vitro. Epilepsia 32, 747754.

85B R A I N R E S E A R C H 1 2 5 2 ( 2 0 0 9 ) 7 6 8 6


11/11

Janigro, D., Gasparini, S., D'Ambrosio, R., McKhann, G.,DiFrancesco, D., 1997. Reduction of K+ uptake in glia preventslong-term depression maintenance and causes epileptiformactivity. J. Neurosci. 17, 28132824.

Jefferys, J.G.R., 1995. Nonsynaptic modulation of neuronal activityin the brain: electric currents and extracellular ions. Physiol.Rev. 75, 689723.

Jensen, M.S., Yaari, Y., 1988. The relationship between interictal

and ictal paroxysms in an in vitro model of focal hippocampalepilepsy. Ann. Neurol. 24, 591598.

Korn, S.J., Giacchino, J.L., Chamberlin, N.L., Dingledine, R., 1987.Epileptiform burst activity induced by potassium in thehippocampus and its regulation by GABA-mediated inhibition.

J. Neurophysiol. 57, 325340.Madison, D.V., Nicoll, R.A., 1986. Actions of noradrenaline

recorded intracellularly in rat hippocampal CA1 pyramidalneurones, in vitro. J. Physiol. 372, 221244.

Martn, E.D., Araque, A., Buo, W., 2001. Synaptic regulation of theslow Ca2+-activated K+ current in hippocampal CA1 pyramidalneurons: implication in epileptogenesis. J. Neurophysiol. 86,28782886.

McCormick, D.A., Contreras, D., 2001. On the cellular and networkbases of epileptic seizures. Annu. Rev. Physiol. 63, 815846.

Rutecki, P.A., 1994. Noradrenergic modulation of epileptiformactivity in the hippocampus. Epilepsy Res. 20, 125136.

Sah, P., 1996. Ca2+-activated K+ currents in neurones: types,physiological roles and modulation. Trends Neurosci. 19,150154.

Schornack, P.A., Song, S.-K., Ling, C.S., Hotchkiss, R., Ackerman, J.J.H., 1997. Quantification of ion transport in perfused rat heart:133Cs+ as an NMR K+ analog. Am. J. Physiol. 272, C1618C1634.

Skov, J., Nedergaard, S., Andreasen, M., 2005. New type ofsynaptically mediated epileptiform activity independent ofknown glutamate and GABA receptors. J. Neurophysiol. 93,18451856.

Skov,J., Andreasen, M., Nedergaard, S., 2006.Postnatal development

of a new type of epileptiform activity in the rat hippocampus.Brain Res. 1096, 6169.

Storm, J.F., 1990. Potassium currents in hippocampal pyramidalcells. Prog. Brain Res. 83, 161187.

Taylor, C.P., Dudek, F.E., 1984. Excitation of hippocampalpyramidal cells by an electrical field effect. J. Neurophysiol. 52,126142.

Traub, R.D., Jefferys, J.G., 1994. Are there unifying principles

underlying the generation of epileptic afterdischarges in vitro?Prog. Brain Res. 102, 383394.

Traynelis, S.F., Dingledine, R., 1988. Potassium-inducedspontaneous electrographic seizures in the rat hippocampalslice. J. Neurophysiol. 59, 259276.

Verma-Ahuja, S., Evans, M.S., Pencek, T.L., 1995. Evidence for adecreased calcium dependent potassium conductance inhippocampal CA3 neurons of genetically epilepsy-prone rats.Epilepsy Res. 22, 137144.

Watts, A.E., Whittington, M.A., Jefferys, J.G.R., 1993. Epileptic focusinduced in rat by intrahippocampal cholera toxin: neuronalproperties in vitro. Neuroscience 55, 4556.

Williamson, A., Spencer, D.D., Shepherd, G.M., 1993. Comparisonbetween the membrane and synaptic properties of human androdent dentate granula cells. Brain Res. 622, 194202.

Wu, H.-M., Huang, C.-C., Chen, S.-H., Liang, Y.-C., Tsai, J.-J., Hsieh,C.-L., Hsu, K.-S., 2003. Herpes simplex virus type 1 inoculationenhances hippocampal excitability and seizure susceptibilityin mice. Eur. J. Neurosci. 18, 32943304.

Xiong, Z.Q., Stringer, J.L., 1999. Cesium induces spontaneousepileptiform activity without changing extracellular potassiumregulation in rat hippocampus. J. Neurophysiol. 83, 33393346.

Xiong, Z.Q., Stringer, J.L., 2001. Prolonged bursts occur in normalcalcium in hippocampal slices after raising excitability andblocking synaptic transmission. J. Neurophysiol. 86, 26252628.

Yaari, Y., Konnerth, A., Heinemann, U., 1986. Nonsynapticepileptogenesis in the mammalian hippocampus in vitro. II.Role of extracellular potassium. J. Neurophysiol. 56, 424438.

86 B R A I N R E S E A R C H 1 2 5 2 ( 2 0 0 9 ) 7 6 8 6

corrente de calcio e sincronização da da 2008

Documents