Cellular mechanisms of high frequency oscillations in epilepsy: On the diverse sources of pathological activities

Epilepsy Research (2011) 97, 308—317

journal homepage: www.elsevier.com/locate/epilepsyres

Cellular mechanisms of high frequency oscillations in epilepsy: On the diverse sources of pathological activities Liset Menendez de la Prida a,∗, Andrew J. Trevelyan b,∗ a b

Instituto Cajal, CSIC, Ave. Doctor Arce 37, Madrid 28002, Spain Institute for Ageing and Health, Newcastle University, UK

Received 28 January 2011; accepted 20 February 2011 Available online 9 April 2011

KEYWORDS Epilepsy; Fast ripples; High-frequency activity

Summary A major goal in epilepsy research is to understand the cellular basis of pathological forms of network oscillations, particularly those classified as high-frequency activity. What are the underlying mechanisms, and how do they arise? The topic of this review is the pattern of high-frequency oscillations that have been recorded in epileptic tissue, and how they might differ from physiological activity. We discuss recent experimental and clinical data with a major focus on the diverse sources of extracellular signals and the contribution of different neuronal populations, including GABAergic interneurons and glutamatergic principal cells. © 2011 Elsevier B.V. All rights reserved.

Introduction We have come a long way in our efforts to understand electroencephalographic (EEG) recordings from the initial studies of Berger (1929), to more recent studies analysing multielectrode recordings from deep in the brain, and sampling at up to 30 kHz. The computational demands for such recordings are of course large. But from these datasets we can derive a greater range of parameters, extending to higher frequency analyses and identifying spatial patterns, thereby enabling us to distinguish between different neurophysiological events from the local field potentials to large

∗

Corresponding author. Tel.: +91 585 4359; fax: +91 585 4754. E-mail addresses: [email protected] (L. Menendez de la Prida), [email protected] (A.J. Trevelyan).

scale neuronal volumes at a large spectrum of frequencies (Buzsaki, 2006). The relationship of various brain oscillations to particular brain states is well recognized. The method of recording is also important. Anaesthesia is associated with slow waves, similar to those recorded during slow-wave sleep (<2 Hz, Steriade, 2003); alpha rhythms (8—10 Hz) occur during relaxed wakefulness (Adrian and Yamagiwa, 1935) and theta activity (4—12 Hz) is predominant during low level mental activity and in exploratory behaviour and REM sleep (Soltesz and Deschênes, 1993; Bragin et al., 1995; Buzsáki, 2002). Most of these brain oscillations can be detected at the scalp, and are the major sources of surface EEG fluctuations. In contrast, other types of faster neuronal activity are more easily recorded as local field potentials, and exhibit particular spatial distributions in different brain structures. Gamma rhythms (30—80 Hz) in neocortex and the hippocampus are

0920-1211/$ — see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.eplepsyres.2011.02.009

Cellular mechanisms of high frequency oscillations in epilepsy predominant during exploratory behaviour and perceptual tasks (Soltesz and Deschênes, 1993; Bragin et al., 1995; Csicsvari et al., 1999; Gray et al., 1989), whereas hippocampal ripples (80—200 Hz) appear to be instrumental in memory consolidation (Ylinen et al., 1995; Csicsvari et al., 2000; Dupret et al., 2010). Understanding the origin of these rhythms and their functional significance remains a major research goal. A key objective in epileptology is to characterize the pathological forms of network oscillations. High-frequency activity, as noticed in epilepsy research, encompasses a large spectrum of oscillations. It extends from brief highfrequency oscillations termed fast ripples (250—800 Hz), which are recorded locally from the epileptogenic regions of the hippocampus and the temporal cortex of epileptic humans and rodents (Bragin et al., 1999b; Urrestarazu et al., 2007; Worrell et al., 2008), to longer epochs of slower activities (80—250 Hz) recorded at larger neuronal territories (Allen et al., 1992; Worrell et al., 2004; Jacobs et al., 2009). All these forms of activities have particular clinical significance due to their association with the onset of seizures and with interictal events in seizure prone individuals (Bragin et al., 1999a,b; Jacobs et al., 2009; Worrell et al., 2004; Khosravani et al., 2005; Schevon et al., 2009; Jiruska et al., 2010). Here we review recent data obtained by combining single-cell and multisite recording approaches aimed to uncover the cellular and network mechanisms responsible for diverse forms of high-frequency oscillations in epilepsy. We will mainly focus on the diverse sources of extracellular signals and the contribution of different neuronal actors to these rhythmopathies.

Cellular sources of extracellular activity One important step in understanding the basic mechanisms of brain oscillations is to know how they are created at the extracellular field potential. EEG signals recorded at the scalp are thought to represent an average of predominantly synaptic activity over millimetres of cortex (Nunez and Srinivasan, 2006). The local field potentials, which constitute one major component of the extracellular EEG signals, result from transmembrane currents produced by several individual neurons (Nicholson and Freeman, 1975). Ohm’s law tells us that these currents are proportional to the product of the conductance of the channels involved and their driving force (the reversal potential minus the membrane potential, Erev − Em ). Hence, the contribution of the different generators to the ongoing EEG varies substantially depending on the type of activity, the brain state and the circuits involved. Also, while most of these signals may be present in intracranial EEG recordings using either macro(up to 10 mm2 ) or microelectrodes (up to 400 ␮m2 ), action potential firing and decorrelated spontaneous synaptic activity from a large population of cells can only be probed with microelectrodes. Thus our ability to detect different activity patterns depends critically on the tools being used. Stimulation of axonal tracts can elicit an easily recordable field event in the target structure (Lorente de No, 1947; Lomo, 1971). Timing considerations, including the latency and kinetics of this event, indicate that the local field potential corresponds to the postsynaptic compound glutamater-

309 gic current. This compound excitatory postsynaptic current (EPSC) is peculiarly visible in the local field potential presumably because many channels in a small volume of tissue are opened almost synchronously after artificial stimulation. The EPSC dynamic of the order of tens of milliseconds is slow enough to facilitate summation of the individual contribution from many parallel-oriented cells. Instead, action potentials are only visible as population spikes if they occur in many cells synchronously with millisecond precision, as might occur following high stimulation strengths (Andersen et al., 1971). Consequently, clinical data on evoked brain activity is thought to represent mainly postsynaptic glutamatergic currents and only rarely population spikes, with the exception of pathological situations of enhanced neuronal firing synchronization such as in epilepsy (Rutecki et al., 1989; Wilson et al., 1990; Valentín et al., 2002). Until recently, the case for GABAergic currents contributing significantly to evoked field events has been less considered — although note the slow ‘‘I-wave’’ induced in the olfactory bulb following stimulation of the olfactory tract (Pickles and Simmonds, 1978), and classical work in the rodent hippocampus (Andersen et al., 1964). The last couple of years, however, has seen the publication of a number of studies showing clearly that spontaneous GABAergic currents are indeed visible in local field recordings using microelectrodes (Glickfeld et al., 2009; Trevelyan, 2009; Bazelot et al., 2010; Oren et al., 2010), and in some circumstances may even be the dominant current during ongoing activity (Trevelyan, 2009; Oren et al., 2010). A notable feature of the spontaneous inhibitory field potentials is that they may arise from the firing of single GABAergic interneurons, or very small populations of interneurons. In contrast, EPSCs elicited by action potentials of single pyramidal cells appear rather less visible in the field potential (Bazelot et al., 2010; Oren et al., 2010), although may be apparent from averaging many repeated events (Nauhaus et al., 2009). The relative field visibility of synaptic potentials elicited by individual pyramidal cells and interneurons may be understood by considering the differences in their respective synaptic contacts. Extracellular EPSCs elicited by a single pyramidal cell are less visible possibly because glutamate is released over a distributed area, at low probability (Prelease typically about 0.4 at 2 mM Ca2+ and 1 mM Mg2+ , Hardingham et al., 2006) and quite low density. It is also relevant that the postsynaptic structures, the dendritic spines, are very small. When glutamate receptors (GluRs) open, the membrane potential collapses quickly on these small structures with only very small conductances being generated thereby reducing the effective driving force (Eglut − Em ) and the subsequent current flow (Jack et al., 1975). The currents consequently are far smaller than if the GluRs were located directly on large dendritic branches. Hence, field potential events resulting from firing from individual glutamatergic cells are less likely to be detected than inhibitory potentials, although spontaneous large events reflecting enhanced synaptic synchrony are recorded by the scalp EEG under certain phases of sleep (Amzica and Steriade, 1998; Cash et al., 2009). Under epileptic conditions, however, unbalanced inhibition and excitation significantly alter the nature of the extracellular signals and pathological synaptic events dominate the ongoing EEG. This is the case of interictal spikes of the order of several hundreds of millivolts, which

310

L. Menendez de la Prida, A.J. Trevelyan

Fig. 1 The contribution of inhibitory synaptic currents (IPSCs) to epochs of high-frequency oscillations at 80—250 Hz. The zero Mg2+ model is used to study situations of extreme discontinuity between active and quiescent cortical regions, as are thought to occur in epileptogenic areas. (A) The local field potential (LFP, black) is filtered in the 80—500 Hz band and scaled to the IPSC filtered signal (red) obtained from the intracellular patch recording (blue). This epoch shows high-frequency field potential oscillations recorded during low levels of neuronal firing. (B) Same as in (A) but for cases of intense neuronal firing. The correlation between the field potential and the IPSC signals was highly dependent on the level of neuronal activity. (C) Plot of the LFP—IPSC correlation indicindexes versus the estimated local activity level. Note the negative correlation between these indices suggesting that during progressive recruitment of neuronal firing IPSCs become less visible at the field potential. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) Modified from Trevelyan (2009) with permission.

presumably reflect a large compound EPSP underlying the paroxysmal depolarizing shift of participating cells (Ayala et al., 1973). In contrast, synaptic contacts established by a single GABAergic basket cell onto a pyramidal cell are typically densely packed at the perisomatic region, and imposes a larger conductance (∼1.7 nS unitary conductance, Kraushaar and Jonas, 2000) compared with pyramid—pyramid connections (∼1 nS unitary conductance, Jonas et al., 1993). Furthermore, basket cells tend to target most, if not all of the pyramidal cells in a given territory of cortex (Freund and Buzsaki, 1996) with a high transmitter release probability (Prelease = 0.55 at 2 mM Ca2+ and 1 mM Mg2+ , Kraushaar and Jonas, 2000), meaning that an action potential is likely to generate a large localized conductance change in many postsynaptic cells. The GABAergic reversal potential (EGABA ), however, is generally quite close to the resting membrane potential, which means that, at rest, even a large conductance will fail to generate much current. If, however, Em becomes shifted from EGABA , this large conductance will suddenly generate a large current. As long as these postsynaptic currents are synchronous in a pool of pyramidal cells a large but localized field potential of the order of tens to hundreds of microvolts is created (Glickfeld et al., 2009; Trevelyan, 2009; Bazelot et al., 2010;

Oren et al., 2010). Other interneuronal types, providing widespread innervation of different dendritic domains, produce less visible synaptic field potential events (Glickfeld et al., 2009; Bazelot et al., 2010). Another potential factor contributing to the large extracellular visibility of inhibitory potentials is that interneuronal firing is coordinated through gap junction coupling (Galarreta and Hestrin, 1999; Gibson et al., 1999). This means that the discharge of a single basket cell may be supplemented by the synchronous discharge of other local interneurons, leading to several fold increase in the number of synapses activated, all of which are densely arranged on the soma and proximal dendrites (Tamás et al., 2000). This role of gap junctions is further supported by reports of reduced power of gamma oscillations recorded in the connexin-36 knock-out mouse (Hormuzdi et al., 2001; Buhl et al., 2003).

A direct relationship between GABAergic currents and pathological high-frequency activities The previous considerations indicate that the field potential generators might dramatically change under conditions of interneuronal loss and circuit reorganization that are known

Cellular mechanisms of high frequency oscillations in epilepsy to occur in epilepsy (Sloviter, 1987; Cossart et al., 2001; Ferrer et al., 1994; Marco et al., 1996). Such alterations are typically spatially inhomogeneous so that extremely active foci are typically flanked by quiescent territories of reduced synchronization levels (Towle et al., 1999; Ortega et al., 2008). Therefore, it is expected that the expression of different forms of high-frequency activity might be also spatially dependent in recordings from epileptic patients. These various considerations indicate that there is a particular network state in which epochs of high-frequency local field potential oscillations directly reflect the pattern of inhibitory postsynaptic currents (IPSCs) in pyramidal cells. This state occurs when focal discharges are sharply delineated from quiescent adjacent cortical territories by the high-frequency discharge of interneurons (most likely basket cells, Menendez de la Prida and Gal, 2004). This situation, of an extreme discontinuity in activity between active and quiescent cortical territories, occurs routinely in the zero Mg model of epilepsy (Menendez de la Prida and Gal, 2004; Trevelyan et al., 2006, 2007). It produces a particular network state in which high-frequency local field oscillations ranging from 80 to 250 Hz directly reflect the pattern of inhibitory postsynaptic currents (IPSCs) in pyramidal cells. In the quiescent territory, pyramidal cells are experiencing a large excitatory drive from the adjacent active zone, and so are relatively depolarized, but are prevented from firing by the coincident high-frequency volley of IPSCs (Trevelyan et al., 2006). Using this model, Trevelyan (2009) derived an estimate of the GABAergic current, to show that it correlated directly with the field. However, and this is a crucial point, this correlation only held when the inhibition was effective; as soon as inhibition weakened, and breakthrough action potentials started occurring, the correspondence between the IPSCs and the local field collapsed (Fig. 1). This suggests that during the progressive recruitment of neurons to a seizure, IPSCs become less and less visible and the field potential appear to be more dominated by excitatory synaptic activity and synchronous neuronal firing. Thus, the major cellular generators contributing to the high-frequency activities could evolve dynamically as a seizure progresses. We suggest that this particular pattern of high-frequency activity is likely to have an upper frequency limit reflecting the maximal basket cell firing rate at about 300 Hz. It is associated, at least in vitro, with a restraint of pyramidal cell activity, suggesting a possible physiological involvement of similar processes in cortical topographic maps, keeping them tightly focussed. Significantly, they seem to represent a situation very like the widely accepted model of gamma rhythms (Traub et al., 1999; Hasenstaub et al., 2005), namely excitatory activity, rhythmically discontinued by basket cell firing, but in which the interneuronal frequency is increased, such that the window for pyramidal firing between IPSCs is closed. This pattern of high-frequency activity may also be apparent during the secondary generalization of seizures into normal cortical territories (Khalilov et al., 2003, 2005), which may be the in vivo situation best matched by the zero Mg2+ model when ictal events are induced in brain slices taken from normal animals. In short, pathological high-frequency activities at 80—250 Hz, which are the direct corollary of high-frequency IPSCs are likely

311 to represent a more physiological type of high-frequency oscillation exacerbated by pathological forms of activation.

Other ways in which GABAergic currents may influence pathological high-frequency activity Intense interneuronal discharges may also trigger a different pattern of high-frequency activity arising because of heterogeneity in levels of intracellular chloride, [Cl− ]i , within the pyramidal cell population of epileptogenic regions. This is of particular relevance since the level of intracellular chloride determines the sing of the GABA-mediated extracellular current flows. Cohen et al. (2002) studied tissue resected during epilepsy surgery, and showed that epileptiform discharges could occur with glutamatergic neurotransmission reduced, but that further addition of GABAergic blockers stopped these events. They subsequently showed that only some pyramidal cells participated in these events, and the defining feature of these cells was that they had peculiarly low levels of the K+ /Cl− cotransporter KCC2 (Huberfeld et al., 2007). They concluded that these cells had a defective [Cl− ]i clearance mechanism, which meant that EGABA was at a relatively depolarized level. Other cells, with ostensibly normal KCC2 levels, remained quiescent during these epileptiform discharges. Thus subpopulations of pyramidal cells in the epileptogenic regions may be defined by whether or not GABA is depolarizing. Staley’s group has recently extended these observations using sophisticated imaging techniques to follow intracellular [Cl− ]i in cells with a genetically expressed chloride sensor (Glykys et al., 2009; Dzhala et al., 2010). Consistent with previous work, they showed that repeated epileptiform discharges led to a progressive build up of [Cl− ]i in all cells, but a particularly striking feature of their imaging was the large range of [Cl− ]i levels seen in different cells (Dzhala et al., 2010). It is interesting to consider what might be the consequences of an extreme variability of [Cl− ]i levels in different groups of pyramidal cells for events detected at the field potential. Basket cell firing is particularly evident both during gamma frequency oscillations and seizures, when they may fire at much higher rates (Kawaguchi, 2001). In both situations, the gamma rhythm and the epileptic crisis, there is also a strong, although not necessarily structured glutamatergic drive. The important point though is that the structure imposed on this activity by basket cell discharge will depend critically on the distribution of pyramidal intracellular [Cl− ]i levels. With a physiological EGABA somewhere close to resting Em , basket cells effectively inhibit pyramidal firing, and this is the mechanism by which pyramidal populations are entrained into a gamma rhythm activity clearly detected extracellularly (Cobb et al., 1995; Whittington et al., 1995). In contrast, in cells with relatively high [Cl− ]i however, firing may variably contribute to the ongoing oscillation; Grenier et al. showed this by artificially loading cells through an intracellular recording electrode (Grenier et al., 2003). We suggest that the same may occur naturally in the epileptogenic regions where some cells apparently lack KCC2. This would be probably a dynamic process during seizure progression because [Cl− ]i will tend to increase in all cells, exacerbated both by the large currents flowing through the GABA channels, and the increasing extracellular

312 K+ (Perreault and Avoli, 1992; Kaila et al., 1997). Therefore, inhibitory currents will no longer prevail in the field recordings and extracellular oscillations will necessarily accelerate thus representing an intermixed contribution of glutamatergic, GABAergic and neuronal firing activity.

The case of fast ripples One important milestone in the study of the relationship between high-frequency activity and epilepsy came from the discovery of a peculiar form of high-frequency oscillation of the order of several hundreds of microvolts to up to millivolts, being recorded close to epileptogenic foci (Bragin et al., 1999b). These short transient oscillations, termed fast ripples, are observed in the local field potential recorded from the hippocampus and the temporal cortex of epileptic humans and rodents (Bragin et al., 1999b; Urrestarazu et al., 2007; Worrell et al., 2008). In contrast to the previously described role of GABAergic currents in longer epochs of high-frequency activities, fast ripples appear to reflect brief pathological bursting from principal neurons caused by abnormal forms of synchrony (Bragin et al., 2000, 2011; Ibarz et al., 2010). In classical experiments with penicillin, most cells in the focus burst synchronously during population discharges, presumably reflecting the result of a pharmacologically induced disinhibition (Matsumoto and Ajmone Marsan, 1964). In the hippocampus, synchronous bursting seems to require disinhibition (Miles and Wong, 1987) to recruit a sufficient number of pyramidal cells for burst initiation (Menendez de la Prida et al., 2006). Data also suggest that more physiological forms of population burst, like the hippocampal sharp-waves, appear to require prolonged periods of neuronal silence followed by dendritic excitation (Kamondi et al., 1998; Harris et al., 2001; Ellender et al., 2010). Thus, in regions generating fast ripples, the inhibitory restraint mechanism just described has probably failed to keep neuronal activity controlled (Trevelyan et al., 2007). Regions producing fast ripples typically occupy small volumes of less than 1 mm3 in TLE patients (Bragin et al., 2002; Crépon et al., 2010); they are closely related with epileptogenic areas and with regions actively involved in the spontaneous generation of seizures (Bragin et al., 2000; Staba et al., 2007; Jiruska et al., 2010). Direct evidence that fast ripples reflect neuronal spiking recently came from simultaneous recordings of single cells and field potentials in anesthetized epileptic rodents: juxtacellular recordings revealed bursts of action potentials aligned with individual fast ripple cycles (Ibarz et al., 2010) and with large population spikes (Bragin et al., 2011). Hence, the extracellular currents formed by nearly synchronous action potentials of parallel-oriented principal cells summate to produce consecutive population spikes that reflect individual neuronal firing. However, such a high-level of synchrony and bursting is not reproduced in many focal models of epilepsy and is controversial in human TLE (Wyler et al., 1975; Colder et al., 1996). Indeed, the oscillatory spectral content of fast ripples (250—800 Hz) is well above the maximal firing frequency of most hippocampal pyramidal neurons.

L. Menendez de la Prida, A.J. Trevelyan Recently, the group of Menendez de la Prida showed that transient high-frequency field potential oscillations similar to fast ripples occur at about the double the firing rate of individual neurons in an in vitro model of population discharges (Foffani et al., 2007). They proposed the idea that fast ripples not only represent hypersynchronous bursting from a single cellular population but also result from the out-of-phase firing of several groups of neurons. This was further confirmed in vivo by following up the firing dynamic of individual cells and the local field potentials over successive fast ripple events (Ibarz et al., 2010). A lack of correlation between the firing rate of single cells and the dominant spectral peak of the nearby field potential further suggest that fast ripples have an emergent dynamic. In this scenario, each fast ripple cycle reflects a population spike created by firing from individual neurons but not all neurons contribute to consecutive cycles similarly — especially for frequencies faster than 300 Hz — hence the emergent character. This is also evident when two consecutive population spikes are recorded at short intervals and individual cell firing is typically aligned with one of them only (Bragin et al., 2011). The emergent dynamic of fast ripples is reflected in large variability from event to event, with spectral fluctuations between the low- and the high-frequency bands. This is consistent with the presence of pathological high-frequency transient oscillations in the 150—200 Hz band being typically intermixed with fast ripples of >250 Hz (Bragin et al., 1999b; Staba et al., 2002; Worrell et al., 2008). The power spectrum of several individual fast ripple events reflects such variability (Fig. 2), being difficult to classify them based only on frequency criteria (Engel et al., 2009; Ibarz et al., 2010). Hence, the averaged power spectra typically exhibit a dominant fundamental frequency and other emergent peaks, including harmonics, suggesting that most of the time participating cells organize in pools that mainly fire at the fundamental frequency but out-of-phase between each other (Foffani et al., 2007; Ibarz et al., 2010). Such an oscillatory variability can be characterized by different statistical measurements, including the spectral entropy, the power and a fast ripple index (Ibarz et al., 2010). The participation rate, the evolution of different active pools of neurons, and the organization of firing within these groups (discussed variably in terms of synchrony and hypersynchrony, and fluctuations in the spectral entropy) will all determine the dominant frequencies and their power (Ibarz et al., 2010, Fig. 2C). The statistical behaviour of several fast ripple events can thus be further exploited to extract useful information on the underlying dynamic and to improve classification of diverse forms of high-frequency oscillations (Ibarz et al., 2010; Blanco et al., 2010).

Mechanisms of out-of-phase firing during fast ripples The apparent ‘‘harmonic’’ nature of these very fast ripples should perhaps not be taken too literally (Engel et al., 2009). There appears to be some jitter of the precise action potential timing with respect to the dominant frequency (Netoff and Schiff, 2002; Foffani et al., 2007), and so the precise phase shift of different active populations of pyramidal cells

Cellular mechanisms of high frequency oscillations in epilepsy

313

A2

A1

500 μV

normalizedspec. (x10 -4)

50 ms

0.3

0.3

0.2

0.2

0.1

0.1

0.0

200 300 400 500 600 700 800

0.0

200 300 400 500 600 700 800

frequency(Hz)

A3

500 μV

B

50 ms

500 μV 50 ms

0.8

0.3

fastrippleindex

-4

normalizedspec. (x10 )

C

0.2

0.1

0.0

200 300 400 500 600 700 800

frequency(Hz)

0.6 0.4

CA1 DG simulation

0.2 0.0 3.0 3.5 4.0

4.5 5.0 5.5

entropy

Fig. 2 Emergent dynamic of fast ripple events. Fast ripples are brief transients of very high-frequency activity in the range of 250—800 Hz, that can be recorded from epileptogenic regions. Recordings of successive fast ripples reflect large spectral fluctuations suggesting that neuronal participation varies from event to event. (A1—A3) Three individual fast ripples recorded in vivo from an epileptic rat. Note the variability in the number of cycles and the multi-peaked character of the individual power spectrum. (B) Some events reflect single population spikes and hence no high-frequency oscillations are detected (black arrow). Very interestingly, these isolated population spikes resemble first cycles of fast ripples (black arrow in (A3)), suggesting that they might be built from the out-of-phase firing from independent pool of neurons. (C) The spectral variability of individual events can be further exploited to extract further information from the dynamic organization of fast ripples. Here, the spectral entropy and a fast ripple index are used to look at the different dynamic of fast ripples recorded at the CA1 and the dentate gyrus of the epileptic hippocampus. Panel C modified from Ibarz et al. (2010) with permission.

will vary quite widely around 180◦ . Realistic computer simulations showed that pools of neurons firing each other with lags of less than 1.5 ms would contribute together to the same population spike, whereas those firing more separately would generate an independent spike (Ibarz et al., 2010). This reflects the requirement for millisecond precision of simultaneous action potentials to be visible at the extracellular field (Andersen et al., 1971). As such, out-of-phase

populations might simply reflect a barely regulated firing pattern, driven by an escalating level of excitation which creates functional clusters at the field potential, similar to the aggregates being described in non-synaptic models of seizures (Bikson et al., 2003). Simulations of epileptiform discharges which evolve from a handful of trigger neurons indicate that the apparent network frequencies are very sensitive to changes in the strength of both glutamatergic

314 and GABAergic synapses (Foffani et al., 2007; Ibarz et al., 2010). Models of naturally occurring oscillations have demonstrated that if the individual components have a natural rhythm, then extremely weak interactions between these components will tend to synchronize them in different fashions. Examples include audiences synchronizing their claps after a theatre show, or the synchronized flashing of fireflies (Pikovsky et al., 2001). With respect to epileptic high-frequency oscillations (Draguhn et al., 1998), these weak interactions have been ascribed to various mechanisms, including ephaptic interactions (Jefferys, 1995) and axonal gap junctions (Traub and Bibbig, 2000). All these factors would play a role in out-of-phase clustering of action potentials and hence would correlate with the existence of emergent spectral components of fast ripple events. This would include neuronal loss (Staba et al., 2007), elevated synaptic noise (Esclapez et al., 1999; Foffani et al., 2007; Ibarz et al., 2010) or circuit reorganization (Nadler et al., 1980; Kobayashi and Buckmaster, 2003). None of these factors is necessary per se but each of them might suffice to promote firing clustering and emergent high-frequency fast ripples. This is clearly evident for neuronal loss and ephaptic coupling for instance. We previously showed that, in an in vitro model of population discharges, the expression of emergent fast ripples were correlated with the degree of neuronal loss and reduced ephaptic interactions (Foffani et al., 2007; see also Staba et al., 2007). However, fast ripples can also emerge in the tetanus toxin model, which exhibits minor cell loss (Jiruska et al., 2010).

Clinical implications: extending the parameter space The initial efforts to characterize clinically high-frequency activity patterns used single electrodes, and such recordings may not provide enough information to distinguish between the underlying mechanisms. As we clearly point out, the analysis of the oscillatory frequency alone is unlikely to be able to distinguish between different pathological patterns of high-frequency oscillations. Also, physiological high-frequency oscillations extend up to at least 250 Hz, and while epileptogenic zones do appear to show uniquely high frequencies, the oscillations there also have dominant components below 250 Hz — that is within the acceptable physiological range. Are these oscillations, in undoubtedly pathological foci, physiological, or do they share the same basic mechanism as the very high pathological frequencies but spectrally shifted? Real insights are likely to be forthcoming as the parameter space of our recordings increase in clinical settings. Recently, several groups have recorded epileptic discharges in humans prior to epilepsy surgery, using arrays of up to 100 electrodes organized in a 10 × 10 matrix at hundreds micrometers spacing (Brinkmann et al., 2009; Schevon et al., 2009). Such approaches would allow a more sophisticated spatiotemporal analysis of evolving activity patterns. The same electrodes have also been used to record physiological activity patterns in monkeys (Baker et al., 2009), which may provide further insights into the differences between physiological and pathological oscillations. Finally,

L. Menendez de la Prida, A.J. Trevelyan there is the prospect of extending our parameter space outside of the frequency domain analyses altogether, to incorporate information on [Cl− ]i handling within the network, or other measures of the brain state. These advances will enable a more subtle subdivision of high-frequency activities, and may yet provide the means for the holy grail of epileptology, to be able to predict seizure activity with confidence.

Acknowledgements This work was supported by grants from the Spanish Ministry MICINN (BFU2009-07989) and the Spanish National Research Council (CSIC 200720I023) to LMP. AJT currently holds an Epilepsy Research UK fellowship.

References Adrian, E.B., Yamagiwa, K., 1935. The origin of the Berger rhythm. Brain 58, 323—351. Allen, P.J., Fish, D.R., Smith, S.J., 1992. Very high-frequency rhythmic activity during SEEG suppression in frontal lobe epilepsy. Electroencephalogr. Clin. Neurophysiol. 82 (2 (February)), 155—159. Amzica, F., Steriade, M., 1998. Cellular substrates and laminar profile of sleep K-complex. Neuroscience 82 (3 (February)), 671—686. Andersen, P., Bliss, T.V.P., Skrede, K.K., 1971. Unit analysis of hippocampal population spikes. Exp. Brain Res. 13, 208—221. Andersen, P., Eccles, J.C., Loyning, Y., 1964. Location of postsynaptic inhibitory synapses on hippocampal pyramids. J. Neurophysiol. 27 (July), 592—607. Ayala, G.F., Dichter, M., Gumnit, R.J., Matsumoto, H., Spencer, W.A., 1973. Genesis of epileptic interictal spikes. New knowledge of cortical feedback systems suggests a neurophysiological explanation of brief paroxysms. Brain Res. 52 (March), 1—17. Baker, J., Bishop, W., Kellis, S., Levy, T., House, P., Greger, B., 2009. Multi-scale recordings for neuroprosthetic control of finger movements. Conf. Proc. IEEE Eng. Med. Biol. Soc., 4573—4577. Bazelot, M., Dinocourt, C., Cohen, I., Miles, R., 2010. Unitary inhibitory field potentials in the CA3 region of rat hippocampus. J. Physiol. 588, 2077—2090. Berger, H., 1929. Uber das Elektroenekephalogramm des Menschen. Arch. Psychiatr. Nervenkr. 87, 527—570. Bikson, M., Fox, J.E., Jefferys, J.G., 2003. Neuronal aggregate formation underlies spatiotemporal dynamics of nonsynaptic seizure initiation. J. Neurophysiol. 89, 2330—2333. Blanco, J.A., Stead, M., Krieger, A., Viventi, J., Marsh, W.R., Lee, K.H., Worrell, G.A., Litt, B., 2010. Unsupervised classification of high-frequency oscillations in human neocortical epilepsy and control patients. J. Neurophysiol. 104, 2900—2912. Bragin, A., Benassi, S.K., Kheiri, F., Engel Jr., J., 2011. Further evidence that pathologic high-frequency oscillations are bursts of population spikes derived from recordings of identified cells in dentate gyrus. Epilepsia 52 (1 (January)), 45—52. Bragin, A., Engel Jr., J., Wilson, C.L., Fried, I., Buzsáki, G., 1999a. High-frequency oscillations in human brain. Hippocampus 9, 137—142. Bragin, A., Engel Jr., J., Wilson, C.L., Fried, I., Mathern, G.W., 1999b. Hippocampal and entorhinal cortex high-frequency oscillations (100—500 Hz) in human epileptic brain and in kainic acid-treated rats with chronic seizures. Epilepsia 40, 127—137. Bragin, A., Jandó, G., Nádasdy, Z., Hetke, J., Wise, K., Buzsáki, G., 1995. Gamma (40—100 Hz) oscillation in the hippocampus of the behaving rat. J. Neurosci. 15, 47—60.

Cellular mechanisms of high frequency oscillations in epilepsy Bragin, A., Mody, I., Wilson, C.L., Engel Jr., J., 2002. Local generation of fast ripples in epileptic brain. J. Neurosci. 22, 2012—2021. Bragin, A., Wilson, C.L., Engel Jr., J., 2000. Chronic epileptogenesis requires development of a network of pathologically interconnected neuron clusters: a hypothesis. Epilepsia 41 (Suppl. 6), S144—S152. Brinkmann, B.H., Bower, M.R., Stengel, K.A., Worrell, G.A., Stead, M., 2009. Large-scale electrophysiology: acquisition, compression, encryption, and storage of big data. J. Neurosci. Methods 180, 185—192. Buhl, D.L., Harris, K.D., Hormuzdi, S.G., Monyer, H., Buzsáki, G., 2003. Selective impairment of hippocampal gamma oscillations in connexin-36 knock-out mouse in vivo. J. Neurosci. 23 (3 (February)), 1013—1018. Buzsáki, G, 2002. Theta oscillations in the hippocampus. Neuron 33, 325—340. Buzsaki, G., 2006. Rhythms in the Brain. Oxford UP, Oxford. Cash, S.S., Halgren, E., Dehghani, N., Rossetti, A.O., Thesen, T., Wang, C., Devinsky, O., Kuzniecky, R., Doyle, W., Madsen, J.R., Bromfield, E., Eross, L., Halász, P., Karmos, G., Csercsa, R., Wittner, L., Ulbert, I., 2009. The human K-complex represents an isolated cortical down-state. Science 324 (5930 (May)), 1084—1087. Cobb, S.R., Buhl, E.H., Halasy, K., Paulsen, O., Somogyi, P., 1995. Synchronization of neuronal activity in hippocampus by individual GABAergic interneurons. Nature 378, 75—78. Cohen, I., Navarro, V., Clemenceau, S., Baulac, M., Miles, R., 2002. On the origin of interictal activity in human temporal lobe epilepsy in vitro. Science 298, 1418—1421. Colder, B.W., Wilson, C.L., Frysinger, R.C., Chao, L.C., Harper, R.M., Engel Jr., J., 1996. Neuronal synchrony in relation to burst discharge in epileptic human temporal lobes. J. Neurophysiol. 75, 2496—2508. Cossart, R., Dinocourt, C., Hirsch, J.C., Merchan-Perez, A., De Felipe, J., Ben-Ari, Y., Esclapez, M., Bernard, C., 2001. Dendritic but not somatic GABAergic inhibition is decreased in experimental epilepsy. Nat. Neurosci. 4 (1), 52—62. Crépon, B., Navarro, V., Hasboun, D., Clemenceau, S., Martinerie, J., Baulac, M., Adam, C., Le Van Quyen, M., 2010. Mapping interictal oscillations greater than 200 Hz recorded with intracranial macroelectrodes in human epilepsy. Brain 133 (Pt 1 (January)), 33—45. Csicsvari, J., Hirase, H., Czurkó, A., Mamiya, A., Buzsáki, G., 1999. Oscillatory coupling of hippocampal pyramidal cells and interneurons in the behaving rat. J. Neurosci. 19, 274—287. Csicsvari, J., Hirase, H., Mamiya, A., Buzsáki, G., 2000. Ensemble patterns of hippocampal CA3—CA1 neurons during sharp waveassociated population events. Neuron 28, 585—594. Draguhn, A., Traub, R.D., Schmitz, D., Jefferys, J.G., 1998. Electrical coupling underlies high-frequency oscillations in the hippocampus in vitro. Nature 394, 189—192. Dupret, D., O’Neill, J., Pleydell-Bouverie, B., Csicsvari, J., 2010. The reorganization and reactivation of hippocampal maps predict spatial memory performance. Nat. Neurosci. 13 (8), 995—1002. Dzhala, V.I., Kuchibhotla, K.V., Glykys, J.C., Kahle, K.T., Swiercz, W.B., Feng, G., Kuner, T., Augustine, G.J., Bacskai, B.J., Staley, K.J., 2010. Progressive NKCC1-dependent neuronal chloride accumulation during neonatal seizures. J. Neurosci. 30 (35 (September)), 11745—11761. Ellender, T.J., Nissen, W., Colgin, L.L., Mann, E.O., Paulsen, O., 2010. Priming of hippocampal population bursts by individual perisomatic-targeting interneurons. J. Neurosci. 30 (17 (April)), 5979—5991. Engel Jr., J., Bragin, A., Staba, R., Mody, I., 2009. High-frequency oscillations: what is normal and what is not? Epilepsia 50, 598—604.

315 Esclapez, M., Hirsch, J.C., Ben-Ari, Y., Bernard, C., 1999. Newly formed excitatory pathways provide a substrate for hyperexcitability in experimental temporal lobe epilepsy. J. Comp. Neurol. 408, 449—460. Ferrer, I., Oliver, B., Russi, A., et al., 1994. Parvalbumin and calbindin-D28k immunocytochemistry in human neocortical epileptic foci. J. Neurol. Sci. 123, 18—25. Foffani, G., Uzcategui, Y.G., Gal, B., Menendez de la Prida, L., 2007. Reduced spike-timing reliability correlates with the emergence of fast ripples in the rat epileptic hippocampus. Neuron 55, 930—941. Freund, T.F., Buzsaki, G., 1996. Interneurons of the hippocampus. Hippocampus 6, 347—470. Galarreta, M., Hestrin, S., 1999. A network of fast-spiking cells in the neocortex connected by electrical synapses. Nature 402, 72—75. Gibson, J.R., Beierlein, M., Connors, B.W., 1999. Two networks of electrically coupled inhibitory neurons in neocortex. Nature 402, 75—79. Glickfeld, L.L., Roberts, J.D., Somogyi, P., Scanziani, M., 2009. Interneurons hyperpolarize pyramidal cells along their entire somatodendritic axis. Nat. Neurosci. 12 (1 (January)), 21—23, Epub 2008 November 23. Glykys, J., Dzhala, V.I., Kuchibhotla, K.V., Feng, G., Kuner, T., Augustine, G., Bacskai, B.J., Staley, K.J., 2009. Differences in cortical versus subcortical GABAergic signaling: a candidate mechanism of electroclinical uncoupling of neonatal seizures. Neuron 63 (5 (September)), 657—672. Gray, C.M., König, P., Engel, A.K., Singer, W., 1989. Oscillatory responses in cat visual cortex exhibit inter-columnar synchronization which reflects global stimulus properties. Nature 338 (6213 (March)), 334—337. Grenier, F., Timofeev, I., Steriade, M., 2003. Neocortical very fast oscillations (ripples, 80—200 Hz) during seizures: intracellular correlates. J. Neurophysiol. 89, 841—852. Hardingham, N.R., Bannister, N.J., Read, J.C., Fox, K.D., Hardingham, G.E., Jack, J.J., 2006. Extracellular calcium regulates postsynaptic efficacy through group 1 metabotropic glutamate receptors. J. Neurosci. 26, 6337—6345. Hasenstaub, A., Shu, Y., Haider, B., Kraushaar, U., Duque, A., McCormick, D.A., 2005. Inhibitory postsynaptic potentials carry synchronized frequency information in active cortical networks. Neuron 47, 423—435. Harris, K.D., Hirase, H., Leinekugel, X., Henze, D.A., Buzsáki, G., 2001. Temporal interaction between single spikes and complex spike bursts in hippocampal pyramidal cells. Neuron 32, 141—149. Hormuzdi, S.G., Pais, I., LeBeau, F.E., Towers, S.K., Rozov, A., Buhl, E.H., Whittington, M.A., Monyer, H., 2001. Impaired electrical signaling disrupts gamma frequency oscillations in connexin 36-deficient mice. Neuron 31 (3 (August)), 487— 495. Huberfeld, G., Wittner, L., Clemenceau, S., Baulac, M., Kaila, K., Miles, R., Rivera, C., 2007. Perturbed chloride homeostasis and GABAergic signaling in human temporal lobe epilepsy. J. Neurosci. 27, 9866—9873. Ibarz, J.M., Foffani, G., Cid, E., Inostroza, M., Menendez de la Prida, L., 2010. Emergent dynamics of fast ripples in the epileptic hippocampus. J. Neurosci. 30, 16249—16261. Jack, J.J.B., Noble, D., Tsien, R.W., 1975. Electric Current Flow in Excitable Cells. Oxford University Press. Jacobs, J., Zelmann, R., Jirsch, J., Chander, R., Dubeau, C.E., Gotman, J., 2009. High frequency oscillations (80—500 Hz) in the preictal period in patients with focal seizures. Epilepsia 50 (7 (July)), 1780—1792. Jefferys, J.G., 1995. Nonsynaptic modulation of neuronal activity in the brain: electric currents and extracellular ions. Physiol. Rev. 75, 689—723.

316 Jiruska, P., Finnerty, G.T., Powell, A.D., Lofti, N., Cmejla, R., Jefferys, J.G., 2010. Epileptic high-frequency network activity in a model of non-lesional temporal lobe epilepsy. Brain 133, 1380—1390. Jonas, P., Major, G., Sakmann, B., 1993. Quantal components of unitary EPSCs at the mossy fibre synapse on CA3 pyramidal cells of rat hippocampus. J. Physiol. 472, 615—663. Kaila, K., Lamsa, K., Smirnov, S., Taira, T., Voipio, J., 1997. Longlasting GABA-mediated depolarization evoked by high-frequency stimulation in pyramidal neurons of rat hippocampal slice is attributable to a network-driven, bicarbonate-dependent K+ transient. J. Neurosci. 17 (20 (October)), 7662—7672. Kawaguchi, Y., 2001. Distinct firing patterns of neuronal subtypes in cortical synchronized activities. J. Neurosci. 21, 7261—7272. Kamondi, A., Acsády, L., Buzsáki, G., 1998. Dendritic spikes are enhanced by cooperative network activity in the intact hippocampus. J. Neurosci. 18, 3919—3928. Khalilov, I., Holmes, G.L., Ben-Ari, Y., 2003. In vitro formation of a secondary epileptogenic mirror focus by interhippocampal propagation of seizures. Nat. Neurosci. 6 (10 (October)), 1079—1085. Khalilov, I., Le Van Quyen, M., Gozlan, H., Ben-Ari, Y., 2005. Epileptogenic actions of GABA and fast oscillations in the developing hippocampus. Neuron 48 (5 (December)), 787—796. Khosravani, H., Pinnegar, C.R., Mitchell, J.R., Bardakjian, B.L., Federico, P., Carlen, P.L., 2005. Increased high-frequency oscillations precede in vitro low-Mg seizures. Epilepsia 46, 1188—1197. Kobayashi, M., Buckmaster, P.S., 2003. Reduced inhibition of dentate granule cells in a model of temporal lobe epilepsy. J. Neurosci. 23, 2440—2452. Kraushaar, U., Jonas, P., 2000. Efficacy and stability of quantal GABA release at a hippocampal interneuron—principal neuron synapse. J. Neurosci. 20, 5594—5607. Lomo, T., 1971. Patterns of activation in a monosynaptic cortical pathway: the perforant path input to the dentate area of the hippocampal formation. Exp. Brain Res. 12 (1), 18—45. Lorente de No, R., 1947. A study of nerve physiology. Stud. Rockefeller Inst. Med. Res. Repr. 132, 1—548. Marco, P., Sola, R.G., Pulido, P., et al., 1996. Inhibitory neurons in the human epileptogenic temporal neocortex. An immunocytochemical study. Brain 119, 1327—1347. Matsumoto, H., Ajmone Marsan, C., 1964. Cortical interictal phenomena in experimental epilepsy: interictal manifestations. Exp. Neurol. 9, 286—304. Menendez de la Prida, L., Gal, B., 2004. Synaptic contributions to focal and widespread spatiotemporal dynamics in the isolated rat subiculum in vitro. J. Neurosci. 24, 5525—5536. Menendez de la Prida, L., Huberfeld, G., Cohen, I., Miles, R., 2006. Threshold behavior in the initiation of hippocampal population bursts. Neuron 49, 131—142. Miles, R., Wong, R.K., 1987. Inhibitory control of local excitatory circuits in the guinea-pig hippocampus. J. Physiol. 388, 611—629. Nadler, J.V., Perry, B.W., Cotman, C.W., 1980. Selective reinnervation of hippocampal area CA1 and the fascia dentata after destruction of CA3—CA4 afferents with kainic acid. Brain Res. 182, 1—9. Nauhaus, I., Busse, L., Carandini, M., Ringach, D.L., 2009. Stimulus contrast modulates functional connectivity in visual cortex. Nat. Neurosci. 12, 70—76. Netoff, T.I., Schiff, S.J., 2002. Decreased neuronal synchronization during experimental seizures. J. Neurosci. 22, 7297—7307. Nicholson, C., Freeman, J.A., 1975. Theory of current sourcedensity analysis and determination of conductivity tensor for anuran cerebellum. J. Neurophysiol. 38 (2 (March)), 356—368. Nunez, P.L., Srinivasan, R., 2006. Electric Fields of the Brain: the Neurophysiology of EEG, second ed. Oxford University Press. Oren, I., Hájos, N., Paulsen, O., 2010. Identification of the current generator underlying cholinergically induced gamma frequency

L. Menendez de la Prida, A.J. Trevelyan field potential oscillations in the hippocampal CA3 region. J. Physiol. 588 (Pt 5 (March)), 785—797. Ortega, G.J., Menendez de la Prida, L., Sola, R.G., Pastor, J., 2008. Synchronization clusters of interictal activity in the lateral temporal cortex of epileptic patients: intraoperative electrocorticographic analysis. Epilepsia 49, 269—280. Perreault, P., Avoli, M., 1992. 4-Aminopyridine-induced epileptiform activity and a GABA-mediated long-lasting depolarization in the rat hippocampus. J. Neurosci. 12, 104—115. Pickles, H.G., Simmonds, M.A., 1978. Field potentials, inhibition and the effect of pentobarbitone in the rat olfactory cortex slice. J. Physiol. 275, 135—148. Pikovsky, A., Rosenblum, M., Kurths, J., 2001. Synchronization: A Universal Concept in Nonlinear Sciences. Cambridge University Press. Rutecki, P.A., Grossman, R.G., Armstrong, D., Irish-Loewen, S., 1989. Electrophysiological connections between the hippocampus and entorhinal cortex in patients with complex partial seizures. J. Neurosurg. 70, 667—675. Schevon, C.A., Trevelyan, A.J., Schroeder, C.E., Goodman, R.R., McKhann Jr., G., Emerson, R.G., 2009. Spatial characterization of interictal high frequency oscillations in epileptic neocortex. Brain 132 (Pt 11 (November)), 3047—3059. Sloviter, R.S., 1987. Decreased hippocampal inhibition and a selective loss of interneurons in experimental epilepsy. Science 235 (4784), 73—76. Soltesz, I., Deschênes, M., 1993. Low- and high-frequency membrane potential oscillations during theta activity in CA1 and CA3 pyramidal neurons of the rat hippocampus under ketamine—xylazine anesthesia. J. Neurophysiol. 70, 97— 116. Staba, R.J., Frighetto, L., Behnke, E.J., Mathern, G.W., Fields, T., Bragin, A., Ogren, J., Fried, I., Wilson, C.L., Engel Jr., J., 2007. Increased fast ripple to ripple ratios correlate with reduced hippocampal volumes and neuron loss in temporal lobe epilepsy patients. Epilepsia 48, 2130—2138. Staba, R.J., Wilson, C.L., Bragin, A., Fried, I., Engel Jr., J., 2002. Quantitative analysis of high-frequency oscillations (80—500 Hz) recorded in human epileptic hippocampus and entorhinal cortex. J. Neurophysiol. 88 (4), 1743—1752. Steriade, M, 2003. Neuronal Substrates of Sleep and Epilepsy. Cambridge UP, Cambridge, UK. Tamás, G., Buhl, E.H., Lörincz, A., Somogyi, P., 2000. Proximally targeted GABAergic synapses and gap junctions synchronize cortical interneurons. Nat. Neurosci. 3 (4 (April)), 366—371. Towle, V.L., Carder, R.K., Khorasani, L., Lindberg, D., 1999. Electrocorticographic coherence patterns. J. Clin. Neurophysiol. 16, 528—547. Traub, R.D., Bibbig, A, 2000. A model of high-frequency ripples in the hippocampus based on synaptic coupling plus axon—axon gap junctions between pyramidal neurons. J. Neurosci. 20, 2086—2093. Traub, R.D., Jefferys, J.G., Whittington MA, 1999. Fast Oscillations in Cortical Circuits. Massachussetts Institute of Technology. Trevelyan, A.J., Sussillo, D., Yuste, R., 2007. Feedforward inhibition contributes to the control of epileptiform propagation speed. J. Neurosci. 27, 3383—3387. Trevelyan, A.J., Sussillo, D., Watson, B.O., Yuste, R., 2006. Modular propagation of epileptiform activity: evidence for an inhibitory veto in neocortex. J. Neurosci. 26, 12447—12455. Trevelyan, A.J., 2009. The direct relationship between inhibitory currents and local field potentials. J. Neurosci. 29, 15299—15307. Urrestarazu, E., Chander, R., Dubeau, F., Gotman, J., 2007. Interictal high-frequency oscillations (100—500 Hz) in the intracerebral EEG of epileptic patients. Brain 130, 2354—2366. Valentín, A., Anderson, M., Alarcón, G., Seoane, J.J., Selway, R., Binnie, C.D., Polkey, C.E., 2002. Responses to single pulse

Cellular mechanisms of high frequency oscillations in epilepsy electrical stimulation identify epileptogenesis in the human brain in vivo. Brain 125 (Pt 8 (August)), 1709—1718. Whittington, M.A., Traub, R.D., Jefferys, J.G., 1995. Synchronized oscillations in interneuron networks driven by metabotropic glutamate receptor activation. Nature 373, 612—615. Wilson, C.L., Isokawa, M., Babb, T.L., Crandall, P.H., 1990. Functional connections in the human temporal lobe. I. Analysis of limbic system pathways using neuronal responses evoked by electrical stimulation. Exp. Brain Res. 82, 279—292. Worrell, G.A., Gardner, A.B., Stead, S.M., Hu, S., Goerss, S., Cascino, G.J., Meyer, F.B., Marsh, R., Litt, B., 2008. High-frequency oscillations in human temporal lobe: simultaneous microwire and clinical macroelectrode recordings. Brain 131, 928—937.

317 Worrell, G.A., Parish, L., Cranstoun, S.D., Jonas, R., Baltuch, G., Litt, B., 2004. High-frequency oscillations and seizure generation in neocortical epilepsy. Brain 127 (Pt 7 (July)), 1496—1506, Epub 2004 May 20. Wyler, A.R., Fetz, E.E., Ward Jr., A.A., 1975. Firing patterns of epileptic and normal neurons in the chronic alumina focus in undrugged monkeys during different behavioral states. Brain Res. 98, 1—20. Ylinen, A., Bragin, A., Nádasdy, Z., Jandó, G., Szabó, I., Sik, A., Buzsáki, G., 1995. Sharp wave-associated highfrequency oscillation (200 Hz) in the intact hippocampus: network and intracellular mechanisms. J. Neurosci. 15, 30— 46.