Mechanisms of fast ripples in the hippocampus

Reduced spike-timing reliability correlates with the emergence of fast ripples in the rat epileptic hippocampus

Ripples are sharp-wave-associated field oscillations (100-300 Hz) recorded in the hippocampus during behavioral immobility and slow-wave sleep. In epileptic rats and humans, a different and faster oscillation (200-600 Hz), termed fast ripples, has been described. However, the basic mechanisms are unknown. Here, we propose that fast ripples emerge from a disorganized ripple pattern caused by unreliable firing in the epileptic hippocampus. Enhanced synaptic activity is responsible for the irregular bursting of CA3 pyramidal cells due to large membrane potential fluctuations. Lower field interactions and a reduced spike-timing reliability concur with decreased spatial synchronization and the emergence of fast ripples. Reducing synaptically driven membrane potential fluctuations improves both spike-timing reliability and spatial synchronization and restores ripples in the epileptic hippocampus. Conversely, a lower spike-timing reliability, with reduced potassium currents, is associated with ripple shuffling in normal hippocampus. Therefore, fast ripples may reflect a pathological desynchronization of the normal ripple pattern.

Long-term potentiation of high-frequency oscillation and synaptic transmission characterize in vitro NMDA receptor-dependent epileptogenesis in the hippocampus

The implication of high-frequency network oscillations (HFOs) in brain pathology resides in as yet unclear mechanisms. Employing field recordings from ventral hippocampal slices and two models of epileptogenesis (i.e. establishment of interictal-like persistent bursts), we found that HFOs associated with epileptiform bursts and excitatory synaptic transmission were co-modulated during epileptogenesis. NMDA receptor-dependent epileptogenesis in CA3 was consistently accompanied by long-lasting strengthening in synaptic transmission (by 94+/-17%, n=5) and HFOs (frequency, power and duration increased by 24+/-8%, 57+/-18% and 33+/-10%, respectively). Co-modulation of synaptic transmission and HFOs was also observed in NMDA receptor-independent epileptogenesis, although in individual experiments either enhancement or depression of both phenomena was observed. Pathological HFOs >200 Hz were unequivocally present in persistent bursts induced by NMDA receptor-dependent but not NMDA receptor-independent mechanisms. The duration of pathological HFOs associated with persistent bursts but not of HFOs associated with bursts before the establishment of epileptogenesis was linearly and strongly correlated with the duration of bursts (r=0.58, P<0.0001). We propose that interplay between spontaneous synchronous bursting and long-lasting synaptic potentiation accompanying certain forms of epileptogenesis may underlie long-lasting potentiation of HFOs, whose quantitative aspects may reliably signal the degree of network changes involved in epileptogenesis.

Late-onset epileptogenesis and seizure genesis: lessons from models of cerebral ischemia

Patients surviving ischemic stroke often express delayed epileptic syndromes. Late poststroke seizures occur after a latency period lasting from several months to years after the insult. These seizures might result from ischemia-induced neuronal death and associated morphological and physiological changes that are only partly elucidated. This review summarizes the long-term morphofunctional alterations observed in animal models of both focal and global ischemia that could explain late-onset seizures and epileptogenesis. In particular, this review emphasizes the change in GABAergic and glutamatergic signaling leading to hyperexcitability and seizure genesis.

Spike Timing of Lacunosom-Moleculare Targeting Interneurons and CA3 Pyramidal Cells During High-Frequency Network Oscillations In Vitro

Network activity in the 200- to 600-Hz range termed high-frequency oscillations (HFOs) has been detected in epileptic tissue from both humans and rodents and may underlie the mechanism of epileptogenesis in experimental rodent models. Slower network oscillations including theta and gamma oscillations as well as ripples are generated by the complex spike timing and interactions between interneurons and pyramidal cells of the hippocampus. We determined the activity of CA3 pyramidal cells, stratum oriens lacunosum-moleculare (O-LM) and s. radiatum lacunosum-moleculare (R-LM) interneurons during HFO in the in vitro low-Mg2+ model of epileptiform activity in GIN mice. In these animals, interneurons can be identified prior to cell-attached recordings by the expression of green-fluorescent protein (GFP). Simultaneous local field potential recordings from s. pyramidale and on-cell recordings of individual interneurons and principal cells revealed three primary firing behaviors of the active cells: 36% of O-LM interneurons and 60% of pyramidal cells fired action potentials at high frequencies during the HFO. R-LM interneurons were biphasic in that they fired at high frequency at the beginning of the HFO but stopped firing before its end. When considering only the highest frequency component of the oscillations most pyramidal cells fired on the rising phase of the oscillation. These data provide evidence for functional distinction during HFOs within otherwise homogeneous groups of O-LM interneurons and pyramidal cells.

NMDA receptor-dependent high-frequency network oscillations (100-300 Hz) in rat hippocampal slices

High-frequency oscillations (HFOs or ripples, >or=100 Hz) appear to be important expressions of cortical circuits, characterizing physiological and pathological functional states. Synaptic and non-synaptic mechanisms are involved in their generation. This study shows that spontaneous N-methyl-D-aspartate receptor (NMDAR) mediated potentials, recorded in dorsal and ventral hippocampal slices perfused with magnesium-free medium and antagonists of non-NMDARs and GABA receptors were associated with high-frequency oscillations (100-300 Hz), recorded in all hippocampal subregions. Both CA3 and CA1 regions displayed HFOs at the range of 180-300 Hz with oscillations in CA3 being significantly faster than in CA1 (232+/-22 Hz, n=64 slices versus 206+/-18 Hz, n=24, P<0.001). Moreover, in most of the slices (39/63) the CA1 network oscillated also at a lower frequency (121.8+/-2.45 Hz). Simultaneous recordings showed that activity was most often initiated in CA3 region; however, dentate gyrus and CA1 were potential sites of generation as well. The incidence of spontaneous events was significantly higher in ventral than in dorsal slices (20+/-1.6/min versus 5.4+/-0.3/min, P<0.001). The competitive and non-competitive NMDAR antagonists, d-AP5 (50 microM) and MK 801 (50 microM), respectively abolished spontaneous activity. The gap-junction blocker carbenoxolone significantly suppressed spontaneous activity in a concentration-dependent manner. These data indicate that synaptic transmission provided by solely NMDARs can sustain the generation of high-frequency network oscillations, which display distinct characteristics in CA3 and CA1 subregions.

High-frequency intracerebral EEG activity (100-500 Hz) following interictal spikes

PURPOSE

High-frequency activity has been recorded with intracerebral microelectrodes in epileptic patients and related to seizure genesis. Our goal was to analyze high-frequency activity recorded with electroencephalograph (EEG) macroelectrodes during the slow wave immediately following interictal spikes, given the potential importance of this presumed hyperpolarization in transforming spikes into seizures.

METHODS

Depth electrode EEG recordings from 10 patients with intractable focal epilepsy were low-pass filtered at 500 Hz and sampled at 2,000 Hz. Spikes were categorized according to localization and morphology. Segments of 256 ms were selected immediately following (postspike), and 2 s before each spike (baseline). Power was estimated in subgamma (0-40 Hz), gamma (40-100 Hz), high frequency (100-200 Hz), and very high frequency (250-500 Hz) bands.

RESULTS

Changes in power above 100 Hz were seen in 22 of 29 spike categories, consisting primarily of a widespread decrease in frequencies above 100 Hz. This decrease became spatially more restricted as frequencies increased, and coincided with the localization of largest spikes for the highest frequencies. High-frequency power decreases were prominent in the hippocampus but less common in amygdala and neocortex. High-frequency power increases were observed in the amygdala.

CONCLUSIONS

Thus high-frequency EEG activity can be recorded with macroelectrodes in humans and may provide insights on neuronal mechanisms related to human epilepsy. This activity undergoes consistent modifications after EEG spikes. We propose that the reduction in high frequencies reflects a postspike depression in neuronal activity that is more pronounced in the region of spike generation. This depression is almost always seen in hippocampus but less in amygdala.

Fast oscillations trigger bursts of action potentials in neocortical neurons in vitro: A quasi-white-noise analysis study

PURPOSE

Recent evidence supports the importance of action potential bursts in physiological neural coding, as well as in pathological epileptogenesis. To better understand the temporal dynamics of neuronal input currents that trigger burst firing, we characterized spectral patterns of stimulation current that generate bursts of action potentials from regularly spiking neocortical neurons in vitro.

METHODS

Sharp microelectrodes were used for intracellular recording and stimulation of cortical neurons in rat brain slices. Quasi-white-noise (0-2 kHz) and "chirp" sine wave currents of decreasing wavelength were applied to represent a broad spectrum of stimulation frequencies. Action potential-related averaging of the stimulation current variations preceding bursting was used to characterize stimulation current patterns more likely to result in a burst rather than a single-spike response.

RESULTS

Bursts of action potentials were most reliably generated by a preceding series of > or = 2 positive current transients at 164+/-37 Hz of the quasi-white-noise, and to sine wave currents with frequencies greater than 90 Hz. The intraburst action potential rate was linearly related to the frequency of the input sine wave current.

CONCLUSIONS

This study demonstrates that regularly spiking cortical neurons in vitro burst in response to fast oscillations of input currents. In the presence of positive cortical feedback loops, encoding input frequency in the intraburst action potential rate may be safer than producing a high-frequency regular output spike train. This leads to the experimentally testable and therapeutically important hypothesis that burst firing could be an antiepileptogenic and/or anti-ictogenic mechanism.

Ongoing epileptiform activity in the post-ischemic hippocampus is associated with a permanent shift of the excitatory-inhibitory synaptic balance in CA3 pyramidal neurons

Ischemic strokes are often associated with late-onset epilepsy, but the underlying mechanisms are poorly understood. In the hippocampus, which is one of the regions most sensitive to ischemic challenge, global ischemia induces a complete loss of CA1 pyramidal neurons, whereas the resistant CA3 pyramidal neurons display a long-term hyperexcitability several months after the insult. The mechanisms of this long-term hyperexcitability remain unknown despite its clinical implication. Using chronic in vivo EEG recordings and in vitro field recordings in slices, we now report spontaneous interictal epileptiform discharges in the CA3 area of the hippocampus from post-ischemic rats several months after the insult. Whole-cell recordings from CA3 pyramidal neurons, revealed a permanent reduction in the frequency of spontaneous and miniature GABAergic IPSCs and a parallel increase in the frequency of spontaneous and miniature glutamatergic postsynaptic currents. Global ischemia also induced a dramatic loss of GABAergic interneurons and terminals together with an increase in glutamatergic terminals in the CA3 area of the hippocampus. Altogether, our results show a morpho-functional reorganization in the CA3 network several months after global ischemia, resulting in a net shift in the excitatory-inhibitory balance toward excitation that may constitute a substrate for the generation of epileptiform discharges in the post-ischemic hippocampus.

The dark side of high-frequency oscillations in the developing brain

Adult brain networks generate a wide range of oscillations. Some of these are behaviourally relevant, whereas others occur during seizures and other pathological conditions. This raises the question of how physiological oscillations differ from pathogenic ones. In this review, this issue is discussed from a developmental standpoint. Indeed, both epileptic and physiological high-frequency oscillations (HFOs) appear progressively during maturation, and it is therefore possible to determine how this program corresponds to maturation of the neuronal populations that generate these oscillations. We review here important differences in the development of neuronal populations that might contribute to their different oscillatory properties. In particular, at an early stage, the density of glutamatergic synapses is too low for physiological HFOs but an additional drive can be provided by excitatory GABA, triggering epileptic HFOs and the cascades involved in long-lasting epileptogenic transformations. This review is part of the INMED/TINS special issue "Nature and nurture in brain development and neurological disorders", based on presentations at the annual INMED/TINS symposium (http://inmednet.com/).

Contralateral 80–280Hz EEG ripples and hippocampal single unit discharge inhibition in response to acute tetanization of rat right caudate putamen in vivo

Clinically, 4-8 Hz (or 30-80 Hz) stimulation of the caudate nucleus ceases (or enhances) the neocortical and hippocampal epileptiform activities of the epilepsy patients. Possibly, electric stimulation of the caudate nucleus could produce epilepsy. In order to prove this point we delivered the acute tetanization (60 Hz, 2s, 0.4-0.6 mA) into the rat right caudate putamen nucleus (ATRC) and examined bilateral neocortical EEG and hippocampal unit discharges in vivo. The results demonstrated that: (1) 80-280 Hz EEG ripples could be evoked bilaterally, and more stronger on the contralateral side. And the maximum amplitudes of the power spectra (microV2/Hz) have higher shifting variability among multiple contralateral EEG ripples. (2) The EEG ripples were coupled contralaterally with the hippocampal neuronal firing inhibition. (3) An episode of 10-15 Hz EEG oscillations was ipsilaterally coupled with rhythmic hippocampal neuronal bursts. It suggested that the hemispheric reactions of neocortical EEG and hippocampal neuronal discharges are lateralized in response to the stimulation. It implies that the epileptic network activities were reorganized by the ATRC. Neocortical EEG ripples, called as seizure-like fast oscillations, were repetitively evoked by the ATRC.

Cell-attached voltage-clamp and current-clamp recording and stimulation techniques in brain slices

Cell-attached recording provides a way to record the activity of - and to stimulate - neurons in brain slices without rupturing the cell membrane. This review uses theory and experimental data to address the proper application of this technique and the correct interpretation of the data. Voltage-clamp mode is best-suited for recording cell firing activity, and current-clamp mode is best-suited for recording resting membrane potential and synaptic potentials. The magnitude of the seal resistance determines what types of experiments can be accomplished with a cell-attached recording: a loose seal is adequate for recording action potential currents, and a tight seal is required for evoking action potentials in the attached cell and for recording resting and synaptic potentials. When recording action potential currents, if the researcher does not want to change the firing activity of the cell, then it is important that no current passes from the amplifier through the patch resistance. In order to accomplish this condition, the recording pipette should be held at the potential that gives a holding current of 0. An advantage of cell-attached current-clamp over whole-cell recording is that it accurately depicts whether a synaptic potential is hyperpolarizing or depolarizing without the risk of changing its polarity.

Online Prediction of Onsets of Seizure-like Events in Hippocampal Neural Networks Using Wavelet Artificial Neural Networks

It has been previously shown that wavelet artificial neural networks (WANNs) are able to classify the different states of epileptiform activity and predict the onsets of seizure-like events (SLEs) by offline processing (Ann. Biomed. Eng. 33(6):798-810, 2005) of the electrical data from the in-vitro hippocampal slice model of recurrent spontaneous SLEs. The WANN design entailed the assumption that time-varying frequency information from the biological recordings can be used to estimate the times at which onsets of SLEs would most likely occur in the future. Progressions of different frequency components were captured by the artificial neural network (ANN) using selective frequency inputs from the initial wavelet transform of the biological data. The training of the WANN had been established using 184 SLE episodes in 34 slices from 21 rats offline. Nine of these rats also exhibited periods of interictal bursts (IBs). These IBs were included as part of the training to help distinguish the difference in dynamics of bursting activities between the preictal- and interictal type. In this paper, we present the results of an online processing using WANN on 23 in-vitro rat hippocampal slices from 9 rats having 93 spontaneous SLE episodes generated under low magnesium conditions. Over the test cases, three of the nine rats exhibited over 30 min of IB activities. We demonstrated that the WANN was able to classify the different states, namely, interictal, preictal, ictal, and IB activities with an accuracy of 86.6, 72.6, 84.5, and 69.1%, respectively. Prediction of state transitions into ictal events was achieved using regression of initial "normalized time-to-onset" estimates. The SLE onsets can be estimated up to 36.4 s ahead of their actual occurrences, with a mean error of 14.3 +/- 27.0 s. The prediction errors decreased progressively as the actual time-to-onset decreased and more initial "normalized time-to-onset" estimates were used for the regression procedure.

The effects of high-frequency oscillations in hippocampal electrical activities on the classification of epileptiform events using artificial neural networks

The existence of hippocampal high-frequency electrical activities (greater than 100 Hz) during the progression of seizure episodes in both human and animal experimental models of epilepsy has been well documented (Bragin A, Engel J, Wilson C L, Fried I and Buzsáki G 1999 Hippocampus 9 137-42; Khosravani H, Pinnegar C R, Mitchell J R, Bardakjian B L, Federico P and Carlen P L 2005 Epilepsia 46 1-10). However, this information has not been studied between successive seizure episodes or utilized in the application of seizure classification. In this study, we examine the dynamical changes of an in vitro low Mg2+ rat hippocampal slice model of epilepsy at different frequency bands using wavelet transforms and artificial neural networks. By dividing the time-frequency spectrum of each seizure-like event (SLE) into frequency bins, we can analyze their burst-to-burst variations within individual SLEs as well as between successive SLE episodes. Wavelet energy and wavelet entropy are estimated for intracellular and extracellular electrical recordings using sufficiently high sampling rates (10 kHz). We demonstrate that the activities of high-frequency oscillations in the 100-400 Hz range increase as the slice approaches SLE onsets and in later episodes of SLEs. Utilizing the time-dependent relationship between different frequency bands, we can achieve frequency-dependent state classification. We demonstrate that activities in the frequency range 100-400 Hz are critical for the accurate classification of the different states of electrographic seizure-like episodes (containing interictal, preictal and ictal states) in brain slices undergoing recurrent spontaneous SLEs. While preictal activities can be classified with an average accuracy of 77.4 +/- 6.7% utilizing the frequency spectrum in the range 0-400 Hz, we can also achieve a similar level of accuracy by using a nonlinear relationship between 100-400 Hz and <4 Hz frequency bands only.

Epileptogenic Actions of GABA and Fast Oscillations in the Developing Hippocampus

GABA excites immature neurons and inhibits adult ones, but whether this contributes to seizures in the developing brain is not known. We now report that in the developing, but not the adult, hippocampus, seizures beget seizures only if GABAergic synapses are functional. In the immature hippocampus, seizures generated with functional GABAergic synapses include fast oscillations that are required to transform a naive network to an epileptic one: blocking GABA receptors prevents the long-lasting sequels of seizures. In contrast, in adult neurons, full blockade of GABA(A) receptors generates epileptogenic high-frequency seizures. Therefore, purely glutamatergic seizures are not epileptogenic in the developing hippocampus. We suggest that the density of glutamatergic synapses is not sufficient for epileptogenesis in immature neurons; excitatory GABAergic synapses are required for that purpose. We suggest that the synergistic actions of GABA and NMDA receptors trigger the cascades involved in epileptogenesis in the developing hippocampus.

NKCC1 transporter facilitates seizures in the developing brain

During development, activation of Cl(-)-permeable GABA(A) receptors (GABA(A)-R) excites neurons as a result of elevated intracellular Cl(-) levels and a depolarized Cl(-) equilibrium potential (E(Cl)). GABA becomes inhibitory as net outward neuronal transport of Cl(-) develops in a caudal-rostral progression. In line with this caudal-rostral developmental pattern, GABAergic anticonvulsant compounds inhibit motor manifestations of neonatal seizures but not cortical seizure activity. The Na(+)-K(+)-2Cl(-) cotransporter (NKCC1) facilitates the accumulation of Cl(-) in neurons. The NKCC1 blocker bumetanide shifted E(Cl) negative in immature neurons, suppressed epileptiform activity in hippocampal slices in vitro and attenuated electrographic seizures in neonatal rats in vivo. Bumetanide had no effect in the presence of the GABA(A)-R antagonist bicuculline, nor in brain slices from NKCC1-knockout mice. NKCC1 expression level versus expression of the Cl(-)-extruding transporter (KCC2) in human and rat cortex showed that Cl(-) transport in perinatal human cortex is as immature as in the rat. Our results provide evidence that NKCC1 facilitates seizures in the developing brain and indicate that bumetanide should be useful in the treatment of neonatal seizures.

Prediction of Seizure Onset in an In-Vitro Hippocampal Slice Model of Epilepsy Using Gaussian-Based and Wavelet-Based Artificial Neural Networks

We propose that artificial neural networks (ANNs) can be used to predict seizure onsets in an in-vitro hippocampal slice model capable of generating spontaneous seizure-like events (SLEs) in their extracellular field recordings. This paper assesses the effectiveness of two ANN prediction schemes: Gaussian-based artificial neural network (GANN) and wavelet-based artificial neural network (WANN). The GANN prediction system consists of a recurrent network having Gaussian radial basis function (RBF) nonlinearities capable of extracting the estimated manifold of the system. It is able to classify the underlying dynamics of spontaneous in-vitro activities into interictal, preictal and ictal modes. It is also able to successfully predict the onsets of SLEs as early as 60 s before. Improvements can be made to the overall seizure predictor design by incorporating time-varying frequency information. Consequently, the idea of WANN is considered. The WANN design entails the assumption that frequency variations in the extracellular field recordings can be used to compute the times at which onsets of SLEs are most likely to occur in the future. Progressions of different frequency components can be captured by the ANN using appropriate frequency band adjustments via pruning, after the initial wavelet transforms. In the off-line processing comprised of 102 spontaneous SLEs generated from 14 in-vitro rat hippocampal slices, with half of them used for training and the other half for testing, the WANN is able to predict the forecoming ictal onsets as early as 2 min prior to SLEs with over 75% accuracy within a 30 s precision window.