Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Feb 1.
Published in final edited form as: Epilepsia. 2023 Dec 14;65(2):362–377. doi: 10.1111/epi.17845

Pathological neurons generate ripples at the UP-DOWN transition disrupting information transfer

Shennan A Weiss 1,2,3, Itzhak Fried 5, Jerome Engel Jr 4,5,6,7,8, Anatol Bragin 4, Shuang Wang 9, Michael R Sperling 10, Robert KS Wong 2, Yuval Nir 11,12,13,14, Richard J Staba 4
PMCID: PMC10922301  NIHMSID: NIHMS1952511  PMID: 38041560

Abstract

Objective:

To confirm and investigate why pathological HFOs (pHFOs), including ripples [80–200 Hz] and fast ripples [200–600 Hz], are generated during the UP-DOWN transition of the slow wave and if information transmission mediated by ripple temporal coupling is disrupted in the seizure onset zone (SOZ).

Methods:

We isolated 217 total units from 175.95 iEEG contact-hours of synchronized macro- and microelectrode recordings from 6 patients. Sleep slow oscillation (0.1–2 Hz) epochs were identified in the iEEG recording. iEEG HFOs that occurred superimposed on the slow wave were transformed to phasors and adjusted by the phase of maximum firing in nearby units (i.e., maximum UP). We tested whether, in the SOZ, HFOs and associated action potentials (AP) occur more often at the UP-DOWN transition. We also examined ripple temporal correlations using cross correlograms.

Results:

At the group level in the SOZ, HFO and HFO-associated AP probability was highest during the UP-DOWN transition of slow wave excitability (p<<0.001). In the non-SOZ, HFO and HFO-associated AP was highest during the DOWN-UP transition (p<<0.001). At the unit level in the SOZ, 15.6% and 20% of units exhibited more robust firing during ripples (Cohen’s d=0.11–0.83) and fast ripples (d=0.36–0.90) at the UP-DOWN transition (p<0.05 f.d.r corrected), respectively. By comparison, also in the SOZ, 6.6% (d=0.14–0.30) and 8.5% (d=0.33–0.41) of units had significantly less firing during ripples and fast ripples at the UP-DOWN transition, respectively. Additional data shows ripple and fast ripple temporal correlations, involving global slow waves, between the hippocampus, entorhinal cortex, and parahippocampal gyrus were reduced by >50% in the SOZ compared to the non-SOZ (N=3).

Significance:

The UP-DOWN transition of slow wave excitability facilitates the activation of pathological neurons to generate pHFOs. Ripple temporal correlations across brain regions may be important in memory consolidation and are disrupted in the SOZ, perhaps by pHFO generation.

Introduction

Many patients with focal epilepsy experience more frequent or more severe seizures during non-REM sleep (NREM)1,2, and exhibit deficits in memory consolidation3. Non-REM sleep is associated with an increase in the rate of inter-ictal epileptiform discharges that may better localize the seizure onset zone (SOZ) 4,5. High-frequency oscillations are brief (10–100 msec) bursts of spectral power subclassified into ripples (80–200 Hz) and fast ripples (200–600 Hz). HFOs occur more frequently during non-REM sleep6. Increased pathological HFO (pHFOs) rates are associated with neuronal injury leading to epileptogenesis7. In particular, increased fast ripple rates can identify brain regions necessary and sufficient for seizure generation8. Fast ripples are also implicated in seizure genesis9–11.

Ripples during NREM sleep are not primarily studied in the context of epilepsy, but rather in the context of facilitating memory consolidation12–14. In the hippocampus, ripples during NREM are superimposed on a sharp wave (SpW-R) in local field potential (LFP) recordings. During the SpW-R neuronal ensembles encode experiences from wakefulness that are replayed compressed forward and reversed15. Furthermore, ripples important for memory consolidation have also been identified during NREM sleep in association neocortex. These ripples are temporally coupled (i.e., coincided) with the hippocampal ripples during sleep13,16–19 and wakefulness20 and promote information transfer and memory. Thus, to understand pathological HFOs (pHFOs) they must first be distinguished from physiological ripples.

During NREM sleep the predominant oscillation is the slow wave21,22, which is generated by synchronized membrane potential changes mainly in deep layer pyramidal neurons that repeatedly alternates between a hyperpolarized (DOWN) state to a depolarized (UP) state. The slow wave propagates from the frontal lobe posteriorly, and into mesial temporal structures. Some slow waves can occur globally across the brain, but particularly in mesial-temporal lobe slow waves can be local too23. Large membrane potential fluctuations are distinctly lacking in the hippocampus24, and it is debated whether hippocampal activity changes there during the slow wave can be classified as DOWN and UP25 or DOWN and SpW-R26. In murine experiments investigating memory, physiological ripples, in both the hippocampus and neocortex, occur mostly during the DOWN to UP (DOWN-UP) transition12,13. In iEEG recordings from patients with epilepsy the relative timing of the UP and DOWN states is inferred by the relative phase of the slow wave, and putatively pathological ripples, in epileptogenic brain regions, are thought to occur relatively more often at the UP to DOWN (UP-DOWN) transition27–29. Although physiological and pathological ripples show differences in the preferred angle of coupling to slow wave excitability, there is substantial overlap. Thus, accounting for slow wave-ripple coupling only marginally improves the localization of epileptogenic regions29,30. Nevertheless, investigating the neuronal and circuit level mechanisms responsible for pHFOs at the UP-DOWN transition could offer important clues to understanding epileptogenesis and seizure genesis.

We hypothesized that in the SOZ, there are pathological neurons that fire more robustly during pHFOs at the UP-DOWN transition than at the DOWN-UP transition of slow wave excitability. We found, at the group level, that HFO and HFO associated action potentials occur with highest probability at the UP-DOWN transition. However, at the individual unit level in the SOZ, only a minority of the units fired more strongly during HFOs at the UP-DOWN transition. These results confirm that pathologic neuron clusters, embedded in pathological circuits, contribute to generating pHFOs at the UP-DOWN transition of slow wave sleep.

Furthermore, in the SOZ where pHFOs occur, ripple temporal coupling was disrupted between the hippocampus, entorhinal cortex, and parahippocampal gyrus. Since ripple temporal coupling has been associated with information transfer in prior studies13,16–19, this observation may indicate a new mechanism contributing to memory disorders in patients with epilepsy.

Methods

Dataset collection and delineation of the seizure onset zone (SOZ)

All data were acquired with approval from the local institutional review board (IRB) at University of California Los Angeles (UCLA). Six patients with focal epilepsy who were implanted for the purpose of localization of the SOZ in 2009–2010 were included in the study25,31. The epileptologist defined SOZ contacts were aggregated across all seizures during the entire depth iEEG evaluation for each patient and included regions of earliest propagation. Sleep studies were conducted in the epilepsy monitoring unit 48–72 hr after surgery, prior to the tapering of anti-seizure drugs, and lasted 7 hr on average, and sleep-wake stages were scored according to established guidelines. The montage included two EOG electrodes; two EMG electrodes scalp electrodes at C3, C4 Pz, and Fz; two earlobe electrodes used for reference; and continuous video monitoring. Seizures were not captured during these recordings. In each patient, 8–12 depth electrodes were implanted targeting medial brain areas. Both scalp and depth interictal intracranial EEG (iEEG) data, from the most medial depth electrode macroelectrode contact, were continuously recorded, during slow wave sleep, with a Stellate amplifier at a sampling rate of 2 kHz, bandpass-filtered between 0.1 and 500 Hz, and re-referenced offline to the mean signal recorded from the earlobes. Each electrode terminated in eight 40-μm platinum-iridium microwires from which extracellular signals were continuously recorded (referenced locally to a ninth noninsulated microwire) at a sampling rate of 28 kHz using a Neuralynx Cheetah amplifier and bandpass-filtered between 1 and 6000 Hz.

Spike sorting and characterization of high-frequency oscillations HFOs

Action potentials (APs) were detected by high-pass filtering the LFP recordings above 300 Hz and applying a threshold at 5 SD above the median noise level. Detected events were further categorized as noise, single-unit, or multiunit events using superparamagnetic clustering23. Among the 217 units, a comparison of the AP waveforms and firing characteristics32 identified only 8 putative inhibitory interneurons that were not analyzed separately due to sample size limitations. Unit stability was confirmed by verifying that spike waveforms and inter-spike-interval distributions were consistent and distinct throughout the night. AP time stamps were downsampled to 2000 Hz. HFOs, that were not superimposed on epileptiform spikes, were characterized in the iEEG using previously published methods33–35 (Supplementary Methods, https://github.com/shenweiss). HFOs in the microwire LFP recording were not analyzed, reducing the chance that high-frequency events would be influenced by leakage from APs36.

Analysis of the slow-wave, slow-wave associated unit activity and HFOs, and phase standardization

After downsampling the iEEG recordings to 500 Hz. The slow wave was isolated using an optimized Hamming-windowed FIR band-pass filter between 0.1–2 Hz (eegfitnew.m, https://sccn.ucsd.edu/eeglab). In each iEEG channel, we calculated the normalized instantaneous amplitude of the Hilbert transformed band-pass filtered signal and used independent onset and offset normalized minimum amplitude (z-score) and duration criteria, defined from visual inspection of the algorithm results, to identify epochs in which slow oscillatory epochs appeared29. We then identified the instantaneous phases of the slow wave, in the proximal iEEG channel, during each HFO and at each AP coinciding with a slow wave epoch. The HFOs were then transformed to a HFO phasor with a single corresponding phase angle and vector length value29.

We utilized the circular statistics toolbox for Matlab (https://github.com/circstat/circstat-matlab.git). For each unit, we aggregated its APs instantaneous slow wave phase values across the recording duration and tested for circular non-uniformity using Rayleigh’s test and computed the mean (i.e. preferred) phase angle. The angle of minimum firing was defined by the phase of lowest probability of AP firing. For each macroelectrode, and corresponding bundle of microelectrodes, we assigned a maximum UP (max UP) phase angle by identifying the unit, recorded by that bundle, with the largest Rayleigh Z value. That unit’s preferred slow wave angle of maximal AP firing probability was defined as the max UP. We compared the max UP values across the macroelectrode contacts using a parametric two-way AVOVA for circular data. For all the slow wave HFOs, detected by a single macroelectrode, the max UP value derived for that macroelectrode trigonometrically adjusted the HFO phasor angles such that θHFOadjusted=θmax UP−θHFO. The instantaneous slow wave phase of the APs, for all the units isolated in the macroelectrode’s corresponding bundle of microelectrodes, were similarly adjusted.

We tested for circular non-uniformity of the HFOs max UP adjusted phasor angles in the non-SOZ and SOZ and aggregated across macroelectrode contacts and patients, using Rayleigh’s test and derived the mean (i.e. preferred) phase angle. We compared these distributions using the Kuiper’s test. This methodology was also used to compare the max UP adjusted instantaneous slow wave phases of APs coinciding with HFOs.

Peri-HFO unit activity relative to adjusted slow wave phase angle

All the slow wave HFOs detected in a single macroelectrode contact’s recording were individually compared with the corresponding APs from each individual unit isolated from the LFPs recorded from the bundle of microelectrodes distal to that macroelectrode. APs recorded within the onset and offset of a slow wave HFO were assigned an instantaneous phase of the slow wave that was adjusted by max UP. Within each individual unit, for each macroelectrode HFO event, a two second raster trial was generated centered at the onset of the HFO event with a resolution of 1 ms (Figure 1,2A). This raster was then convolved with a 100 ms Gaussian kernel and the resulting AP train was down sampled to 40 Hz. These individual gaussian smoothed raster plots were then sorted into 12°,22° bins using the corresponding max UP adjusted phasor angle of the trial’s HFO event. Within these bins we then calculated the mean gaussian smoothed adjusted peri-HFO firing rate.

Figure 1: The phase of the maximum of the UP state varies by neuroanatomic location and can be used to adjust and standardize measurements of ripple-slow wave coupling.

Figure 1:

Open in a new tab

(A) Simultaneous recordings from Behnke-Fried hybrid macro- and microelectrodes in the hippocampus (A1) and amygdala (A2) that were clinically designated part of the seizure onset zone (SOZ). The iEEG recordings (top) are colored by the instantaneous phase in the slow band (0.1–2 Hz). Note that the peaks are colored red, and the troughs are colored blue, the trough-peak transition is yellow and green, and the peak-trough transition is blue and purple using a circular color scheme. Some local slow waves differ in the two locations. The corresponding raster plots of the units recorded by the microelectrodes are shown (below) and are color coded by the instantaneous phase of the slow wave at time of occurrence. In the hippocampus (A1) the units fire more often from 0–90° (light-dark blue) whereas in the amygdala (A2) the units fire more often from −90° to −180° (red-yellow-green). (B1) Normalized rose plot of all the phase of firing of all action potentials during the entire recording duration of ~90 min of the most strongly modulated units in A1 (green), and A2 (red) demonstrating the diametrically opposed preferred phase angles of action potential slow wave coupling corresponding to the maximum of the UP state (max UP) and note large Rayleigh Z (Z) scores measuring phase modulation strength. (B2) Polar plot of the max UP phase angle and log10(Z) value for all the units recorded by the Behnke-Fried electrode in (A1) and (A2) showing little variability. (C) Polar plot of the max UP angle and Z value for the unit most strongly modulated by the slow wave isolated from the 37 microelectrode bundles from the Behnke Fried electrodes implanted in six patients. The color of the ray indicates the neuroanatomic location of the microelectrode bundle and solid rays indicate a microelectrode bundle in the SOZ. The neuroanatomic location was significantly correlated with the max UP phase (circular ANOVA, p<0.01), but the SOZ status was not significant (p>0.05). Most neuroanatomic locations, outside the hippocampus, exhibited a max UP from early trough-peak. (D) Illustration of how the max UP phase measured in each bundle can be used to adjust and standardize measurements of ripple-slow wave coupling. (D1) A hypothetical slow wave with two ripple events (red, purple) that occur near the trough, angles shown with the color conventions as in panel A. The green circle indicates the phase of max UP derived from a hypothetical unit, that is set to the trough-peak. (D2) The red ripple can be transformed to a ripple phasor in polar coordinates based on its corresponding slow wave instantaneous phase values. (D3) Polar plot of the two phasors after the transform. (D4) Polar plot of the two ripple phasors with respect to the ray representing that max UP angle. Subtracting the ripple phasor angles from the max UP angle results in an adjusted standardized measurement of the ripple phasor angle with respect to the max UP angle that can be subdivided into stages of transition between actual UP-DOWN and DOWN-UP (D5).

Figure 2: The highest probability of ripples and ripple triggered action potentials (AP), in the seizure onset zone (SOZ) at the group level, is during the late UP-DOWN transition of the slow wave. At the individual neuron level, a minority of SOZ units show significantly enhanced ripple triggered APs at the UP-DOWN transition.

Figure 2:

Open in a new tab

(A) Normalized rose plots of the max UP angle adjusted ripple phasors measured during slow wave sleep (left) from electrodes in the SOZ (red) and non-SOZ (blue), also shown are the max UP angle adjusted ripple associated APs instantaneous phase angles with respect to the slow wave (right). The ripple phasors and ripple associated APs in the non-SOZ occur on average during late DOWN-UP, whereas in the SOZ the ripple phasors and ripple associated action potentials occur during late UP-DOWN. (B) Associated, gaussian smoothed peri-slow wave ripple raster plots of units with statistically significant ripple triggered increased firing (p<0.001, f.d.r. corrected) in the non-SOZ (B1) and SOZ (B2). The means of the pooled neuronal firing rate are calculated within 12° bins of the adjusted ripple phasor angle. Mean ripple associated increases in action potential firing rate in the non-SOZ occur at the midway DOWN-UP transition. In the SOZ, ripple associated increases in firing rate are greatest at the late UP-DOWN transition. The number of trials in each phase bin are also shown. (C) For each unit in the non-SOZ (C1) and SOZ (C2), bar plot of the mean ripple triggered firing rate minus baseline firing rate for ripples at DOWN-UP (B1, blue box) and ripples at UP-DOWN (B2, green box). Error bars indicate standard error of the mean (s.e.m). Green asterisks indicate units with significantly (p<0.05, f.d.r. corrected) higher ripple triggered firing during ripples at UP-DOWN compared to ripples at DOWN-UP. Cyan daggers indicate units with significantly higher ripple triggered firing during ripples at DOWN-UP compared to ripples at UP-DOWN.

Comparison of unit HFO triggered firing during DOWN-UP vs. UP-DOWN

Following gaussian smoothing, for each HFO event-unit trial, the pre-event baseline AP rate was defined as the mean firing rate of the Gaussian smoothed AP train rate beginning 750 msec prior to HFO onset until the event (i.e., bl-fr). The peak HFO-AP rate was defined as the maximum of the Gaussian smoothed AP train rate during the duration of the HFO event (i.e., hfo-fr), and another value hfodiff-fr was defined as hfo-fr minus bl-fr. A comparison of the bl-fr and hfo-diff was used to identify the units with significantly elevated firing rates during HFOs. Within each unit we compared the distribution of hfodiff-fr values for ripples that occurred with an adjusted phasor angle of-1.89 >X>−2.30 radians (UP-DOWN) to the hfodiff-fr values for ripples at 1.05<X<1.47 radians (DOWN-UP) using an unpaired t-test with Bonferroni Holms false detection ratio (f.d.r) correction and measured Cohen’s d for effect size. We performed a similar comparison of fast ripples that occurred at −1.57>X>−2.36 radians (UP-DOWN) and −0.4<X<0.4 radians (Max UP).

Cross-correlogram of ripple timing

We computed the slow wave ripple and fast ripple onset times for all events, or only the events occurring during the UP-DOWN or DOWN-UP transition in single macroelectrode contacts. In select pairs of macroelectrode contacts, the ripple onset times were compared in 1 msec bins and no maximum lag. The cross correlations were normalized to calculate a correlation index by setting the autocorrelations of the sequences at zero lag equal to 1. The modulation index was calculated as the ratio of the cross-correlogram peak [maximal value within ±50 ms] to the baseline of the cross-correlogram [midpoint of upper and lower boundary of 95% confidence interval averaged over 2 s of cross-correlation]13.

Results

Patient characteristics and description of data

We analyzed 176 iEEG contact-hours of synchronized macro- and micro-electrode recordings during NREM stage N2/N3 sleep from six patients with medically refractory focal seizures. Of the total patients, 3 were diagnosed with mesial temporal lobe epilepsy (MTLE), one with MTLE and neocortical epilepsy, one with neocortical epilepsy, and one with an unknown site of seizure onset. We detected HFOs in the iEEG, excluding those on epileptiform spikes, and APs from 217 units on the adjacent microelectrodes that occurred on slow waves (Table S1).

Identification and comparison of maximum UP angle across brain regions

Prior work on slow wave-HFO coupling used the instantaneous phase of the scalp slow wave27,37 for comparison to HFO events. This methodology assumes that the slow wave associated changes in excitability during the HFO are global, but slow waves can be locally generated25. Prior work has also used the instantaneous phase of the iEEG slow wave29,30,38 but this methodology assumes that the slow wave associated changes in excitability during the HFO are uniform irrespective of neuroanatomic location. However, in rat neocortex and entorhinal cortex, due to membrane potential bimodality, maximum pyramidal cell firing occurs during the trough to peak transition of the local slow wave, but in the hippocampus where membrane bimodality is absent, maximum firing can occur during the trough to peak transition or at the peak24. Furthermore, in epileptogenic regions, whether neurons maximally fire at other phase angles of the slow wave is unclear.

For reasons noted in the preceding paragraph the first analysis identified for each macroelectrode contact the phase angle of the local iEEG slow wave that was associated with maximum AP normalized probability (Figure 1A,B). This phase was designated as the “max UP” phase (Figure 1B,C). The minimum AP probability, corresponding to the max DOWN state, differed from max UP by 146.96° [95% confidence interval: 118.4° −175.5°]. Then the phase angle of the slow wave-associated ripple and fast ripple on the same contact was adjusted to normalize the HFO phase with respect to the max UP state as well as the transitions between UP-DOWN and DOWN-UP (Figure 1D). A similar approach was used to adjust the phase of the APs during the HFOs. The max UP angle was relatively consistent across neuroanatomic locations and occurred during the early trough to peak transition of the slow wave, except in the hippocampus where max UP often was the early peak to trough transition. The max UP angle did not differ for depth electrodes in the SOZ (Figure 1C, two-way circular ANOVA, d.f.=2,8, SOZ Χ2= 3.693, p=0.16, location Χ2=20.327, p=0.01).

Differences in occurrence of HFOs, and excitability during HFOs, in the SOZ and non-SOZ relative to the UP and DOWN state

We compared the mean angle of slow wave modulation of ripples and ripple-associated APs in the SOZ and non-SOZ. This comparison was also performed after distinguishing ripples in hippocampal and extrahippocampal regions (Figure S1). In all regions, ripples and ripple-associated AP exhibited statistically significant slow wave phase preferences (SOZ, Rayleigh Z=142 [phasor], 273 [AP]; non-SOZ, Z=200,294, p<<0.001 Figure 2A). In the NSOZ, ripples coupled with the slow wave during the late DOWN-UP transition, and maximal ripple-associated AP probability was during middle of DOWN-UP transition (Figs 2A & 2B). By contrast, in the SOZ, ripple coupling and maximal ripple-associated firing probability occurred during late UP-DOWN transition, which was different than the coupling and firing in NSOZ (Kuiper’s V, p=1e-3). These results included regions with interictal epileptiform discharges as part of the non-SOZ, whereas prior investigations included these regions as part of the SOZ27,30,37. Thus, in the SOZ, pathological ripples occur during the UP-DOWN transition.

To assess the extent of units firing during ripples in relation to UP and DOWN phases, we quantified peak firing rate with respect to baseline for ripple-unit trials that occurred at either DOWN-UP phases (Fig. 2B1, blue box) or UP-DOWN phases (Fig 2B2 green box) of slow wave excitability. Among the 217 units, 174 units (n=77 multi-units, 97 single units) significantly increased firing during ripples (p<0.001 f.d.r. corrected) and were included in the analysis. In the non-SOZ, four multi-units (4.3%, Cohen’s d=0.21–0.27) fired more robustly (t-test, p<0.05, f.d.r. corrected) during ripples at the DOWN-UP transition and three multi-units (3.2%, d=0.26–0.37) fired more robustly during ripples at the UP-DOWN transition (Fig. 2C1). By comparison, in the SOZ, 6 single units and 1 multi-unit (8.5%, d=0.14–0.30) fired more robustly during ripples at the DOWN-UP transition and 4 single units and 9 multi-units (15.6%, d=0.15–0.83) fired more robustly during ripples at the UP-DOWN transition (Fig. 2C2). We analyzed the firing rate properties of the multi-units (Table S2) and single units (Table S3) and found neurons that preferentially fire during ripple at the DOWN-UP or UP-DOWN transition did not always exhibit higher firing rates or bursting behavior.

A similar analysis was performed with fast ripples on slow waves. In the NSOZ, fast ripples coupled with the slow wave during the late DOWN-UP transition and maximal fast ripple-associated AP probability was near max UP (Figs 3A & 3B, p<<0.001). By contrast, in the SOZ, fast ripple coupling (p<0.01) and maximal fast ripple-associated firing (p<<0.001) occurred during late UP-DOWN transition, which was different than the coupling and firing in non-SOZ (Kuiper’s V, p=1e-3).

Figure 3: The highest probability of fast ripples and fast ripple triggered action potentials (AP), in the seizure onset zone (SOZ) at the group level, is during the late UP-DOWN transition of the slow wave. At the individual neuron level, a minority of SOZ units show significantly enhanced fast ripple triggered APs at the UP-DOWN transition.

Figure 3:

Open in a new tab

(A) Normalized rose plots of the max UP angle adjusted fast ripple phasors measured during slow wave sleep (left) from electrodes in the SOZ (red) and non-SOZ (blue), also shown are the max UP angle adjusted ripple associated APs instantaneous phase angles with respect to the slow wave (right). The fast ripple phasors and fast ripple associated APs in the non-SOZ occur on average during late DOWN-UP, whereas in the SOZ the fast ripple phasors and fast ripple associated action potentials occur during late UP-DOWN. (B) Associated, gaussian smoothed peri-slow wave fast ripple raster plots of units with statistically significant fast ripple triggered increased firing (p<0.001, f.d.r. corrected) in the non-SOZ (B1) and SOZ (B2). The means of the pooled neuronal firing rate are calculated within 22° bins of the adjusted fast ripple phasor angle. Mean fast ripple associated increases in action potential firing rate in the non-SOZ occur near max UP. In the SOZ, ripple associated increases in firing rate are greatest at the late UP-DOWN transition. The number of trials in each phase bin are also shown. (C) For each unit in the non-SOZ (C1) and SOZ (C2), bar plot of the mean fast ripple triggered firing rate minus baseline firing rate for fast ripples near max UP (B1, blue box) and fast ripples at UP-DOWN (B2, green box). Error bars indicate standard error of the mean (s.e.m). Green asterisks indicate units with significantly (p<0.05, f.d.r. corrected) higher fast ripple triggered firing during fast ripples at UP-DOWN compared to fast ripples at near max UP. Cyan daggers indicate units with significantly higher fast ripple triggered firing during fast ripples at near max UP compared to fast ripples at UP-DOWN.

With respect to units that fired more robustly during fast ripples at UP-DOWN as compared to during fast ripples at max UP, we analyzed 80 of the 217 units (n= 39 multi-units, 41 single units) that significantly increased in firing during fast ripples (p<0.001 f.d.r corrected). In the non-SOZ, no units showed a statistically significant preference (Fig. 3C1). However, in the SOZ 2 multi-units and 1 single unit (6.6%, d=0.33–0.41) fired more robustly during max-UP fast ripples, whereas 6 multi-units and 3 single units (20.0%, d=0.36–0.90) fired more robustly during fast ripples at the UP-DOWN transition (Fig 3C2).

We also analyzed the differences in ripple and fast ripple rate, spectral power, and spectral frequency in the SOZ and non-SOZ at different phases of slow wave excitability. We found, among the macroelectrode contacts with corresponding units, the ripple and fast ripple rates trended higher in the SOZ at all phases of slow wave excitability, but these differences did not reach statistical significance (unpaired t-test, p>0.05, f.d.r. corrected). Ripple and fast ripple rates at the UP-DOWN transition relative to other phases of the slow wave trended higher in the SOZ than non-SOZ. Also ripple spectral frequency and power were increased especially at the UP-DOWN transition (Figure S2,S3).

Differences in temporal coupling of ripples in the seizure onset zone

We used normalized cross correlograms to examine the coincidence of ripples between the hippocampus and entorhinal cortex (Figs. 4,S5) or between the hippocampus and parahippocampal gyrus (Fig. S4). In two patients (Figs. 4, S4), we compared the ripple onset cross correlogram of events recorded from contacts in the non-SOZ to that of SOZ contacts. We found that ripple temporal coupling between the hippocampus and entorhinal cortex or hippocampus and parahippocampal gyrus was reduced by >50% in the SOZ (Figs 4A, S4A). In another patient with bitemporal epilepsy, we found that ripple temporal coupling was reduced by >50%, bilaterally, relative to the non-SOZ as compared to the aforementioned other two patients (Figs. S5A). We also found that fast ripple temporal coupling was reduced by >50% in the SOZ, relative to the non-SOZ (Figure 5).

Figure 4: The temporal coupling of ripples during global slow waves is reduced in the seizure onset zone (SOZ).

Figure 4:

Open in a new tab

Data from patient 423. Slow wave ripple onset times from the hippocampus and ipsilateral entorhinal cortex were compared using a cross-correlogram. Left sided structures were in the SOZ (red) and right sided structures were in the non-SOZ (blue). (A) Comparison of the normalized cross-correlogram of slow wave ripple onset times between the hippocampus and entorhinal cortex reveals decreased coupling in the SOZ. (B-E) Cross-correlograms calculated after separation of the slow wave ripples by onset times during DOWN-UP transition and UP-DOWN transition. The strongest reduction in ripple coupling in the SOZ, relative to the non-SOZ, was found for ripples when both occurred during the DOWN-UP transition (B), or both occurred during the UP-DOWN transition (E). A smaller reduction was seen in C and D which correspond to ripple temporal coupling involving local slow waves.

Figure 5: The temporal coupling of fast ripples during global slow waves is reduced in the seizure onset zone (SOZ).

Figure 5:

Open in a new tab

Comparison of the normalized cross-correlogram of slow wave fast ripple onset times between the hippocampus and entorhinal cortex in patients 423 (A) and 416 (C) with unilateral and bilateral seizure onset zones, respectively. (B) Comparison of the normalized cross-correlogram of slow wave fast ripple onset times between the hippocampus and parahippocampal gyrus in patients 406 with bilateral mesial temporal lobe epilepsy.

After assigning each ripple event to either the DOWN-UP or UP-DOWN transition, in the non-SOZ, we found that ripple temporal coupling was strongest when the events occurred at the same phase of slow wave excitability whether DOWN-UP (Figs 4B,S24B) or UP-DOWN (Figs. 4E,S4E). In the SOZ, ripple temporal coupling was reduced more for these DOWN-UP/DOWN-UP (Figs 4B,S4B) or UP-DOWN/UP-DOWN pairs (Figs. 4E,S4E) than the temporal coupling of ripples that occurred at opposing phases of slow wave excitability (Fig 4B,E,S4B,E).

Discussion

To better understand how the UP-DOWN transition of slow wave excitability facilitates pHFO generation, we first established that the phase of the slow wave corresponding to the maximum UP state was similar for units in the SOZ and non-SOZ. Next, we showed at the group level in the SOZ, the probability of both HFOs and HFO-associated APs was highest at the UP-DOWN transition. Then, at the unit level in the SOZ, we found a subpopulation of units that fire more robustly during HFOs at the UP-DOWN transition. Lastly, ripple temporal correlations between the hippocampus and entorhinal cortex, or hippocampus and parahippocampal gyrus, showed that ripples coincided less often in the SOZ. These results suggest that the UP-DOWN transition facilitates the activation of pathological neurons to generate pHFOs, and that pHFOs are potentially involved in disrupting ripple temporal correlations which may be important in information transfer and memory13,16–20.

Potential Mechanisms Generating pHFOs at the UP/DOWN transition

Physiological ripples are generated by a combination of synchronized inhibitory post-synaptic potentials (IPSPs) and APs12. A well-supported theory is pHFOs are generated solely by out of phase39,40 APs from clusters of pathologically interconnected neurons (PIN)41–43. PIN clusters are found following exposure to chemoconvulsants41,44–46, but the properties and characteristics that endow PIN clusters to produce pHFOs and seizures are not well understood. We hypothesize that the neurons that fired more vigorously during HFOs at the UP-DOWN transition are part of the PIN cluster. Our rationale is based on the ground truth that the SOZ is epileptogenic, and HFOs at UP-DOWN occur at higher probability and higher rates in the SOZ 27,29,30,37,38 than non-SOZ. Moreover, neurons that preferentially fire during HFO at UP-DOWN with large effect size are more likely to be found in SOZ than non-SOZ. Challenges and limitations to this interpretation are: 1) HFOs at the DOWN-UP transition increase as well, but to a lesser degree, in the SOZ than non-SOZ; 2) fast ripples are believed to be pathological irrespective of the phase of slow wave excitability; and 3) it is unproven whether all HFOs in the non-SOZ are definitively physiological.

The pathological neurons that fired more during HFOs at the UP-DOWN transition may be facilitated by reduced inhibitory neuron firing47 and inhibitory conductances47,48 on pyramidal cell soma characterized during the UP-DOWN transition. A decreased inhibitory tone paradoxically shortens the UP state but can, in case of GABAA-R blockade, trigger epileptiform discharges at the UP-DOWN transition49. An alternative mechanism of pathological HFO generation is that ID2/NKx2.1 DOWN state active neurogliaform inhibitory interneurons may themselves be released from inhibition at the UP-DOWN transition50. If pyramidal neurons in epileptogenic tissue exhibit chloride dysregulation and depolarizing inhibitory post-synaptic potentials, as shown previously11,51–53, GABA release by these ID2/NKx2.1 at DOWN state onset may contribute to generating pHFOs at the UP/DOWN transition (Figure 6). Also, sleep deprivation induces seizures, and sleep pressure results in depolarizing shifts in the GABAA reversal potential54. The steep slope of the UP-DOWN transition may correspond with synchronized IPSPs55 that, if depolarizing, could elicit pHFOs. Experimental epilepsy models are better suited to investigate the role of inhibitory neurons and chloride ion dysregulation in pHFO generation since patch clamp recordings, microscopy, and optogenetic experiments are likely required50.

Figure 6: An illustration of how excitatory-inhibitory imbalance at the UP-DOWN transition results in epileptiform abnormalities which interfere with normal ripple coupling.

Figure 6:

Open in a new tab

(A) In healthy brain tissue balanced excitation and inhibition persists during the UP state and ripples occur at the late DOWN-UP transition (A1) and normal ripple coupling across brain regions is observed (A2). In epileptogenic brain (B-C), inhibition is reduced (B1: purple, C1: blue) during the UP-DOWN transition and at the DOWN state when ID2/NKX2.1 neurogliaform interneurons in layer 6 are active. Increased excitability at the UP-DOWN transition increases the probability of pHFO generation and may disrupt ripple coupling across brain regions.

Disrupted ripple temporal coupling due to pHFOs may interfere with memory consolidation

We observed, in the SOZ, ripple and fast ripple coincidence between the hippocampus and entorhinal cortex, or hippocampus and parahippocampal gyrus, was reduced by >50% as compared to the non-SOZ. Ripple temporal coupling has been investigated in murine models between the hippocampus and association neocortex. Hippocampal SpW-R (i.e., DOWN-UP ripples) induce a DOWN state in association neocortex56,57 and the hippocampal SpW-R are also synchronized with neocortical ripples there that occur at the corresponding neocortical DOWN-UP transition13,19. This form of ripple temporal coupling has been associated with information transfer and improved memory consolidation13,56,57. Ripple temporal coupling has not been investigated within the mesial-temporal lobe in behavioral experiments so the functional implications of reduced ripple coupling in the SOZ is less clear.

In the mesial-temporal non-SOZ, we found substantial ripple temporal correlations for the ripples that occurred either both at the DOWN-UP or UP-DOWN transitions of slow wave excitability. Notably ripple temporal coupling between hippocampus and association neocortex involved only the DOWN-UP ripples, but physiological ripples can be generated in the hippocampus at the UP-DOWN transition too24. Weaker coupling was seen for ripples that occurred at opposing phases of slow wave excitability (i.e., DOWN-UP/UP-DOWN).

In the SOZ, relative to the non-SOZ, temporal coupling between ripples at the same phase of slow wave excitability were reduced most. Ripples at opposing phases of slow wave excitability exhibited a smaller reduction in the SOZ relative to the non-SOZ. One explanation is that ripple temporal coupling during global slow waves may be selectively disrupted. More work is needed to understand how the pHFOs in the SOZ mechanistically disrupt the ripple temporal coupling. The pHFOs occur with the highest probability during the UP-DOWN transition but in the SOZ temporal coupling between ripples at the DOWN-UP transition was also reduced.

Additionally, our prior work, studying fast ripple temporal correlational networks found that effective epilepsy surgery requires targeting electrode contacts (i.e., nodes) which generate fast ripples relatively desynchronously with other nodes58–60. Herein, we again found, using a different methodology, lower fast ripple temporal correlations in the SOZ, than non-SOZ. Increased fast ripple temporal correlations in the non-SOZ suggest that fast ripples could possibly serve a role in normal sleep as they may in cognition61.

Limitations

Comparing the unit activity in the microelectrodes with the slow wave iEEG recorded by the most proximal microelectrode may be inaccurate, because the macroelectrode may be positioned outside the neuroanatomic structure the microwires are recording from, and slow waves may propagate25 and exhibit different fields24. Analyzing pHFOs in LFP recordings could resolve this issue but introduce AP leakage. Additionally, some local slow wave activity could be inherently pathological62. Ripple temporal coupling was analyzed in only three patients because of the requirement of establishing the max UP phase of each macroelectrode contact, and a larger study is required to confirm these findings.

Conclusion

Our observations can be used to improve the HFO diagnostic value to mark the ictogenic tissue. Mechanistically we found that in epileptogenic regions, the UP-DOWN transition of slow wave excitability facilitates the activation of pathological neurons to generate pHFOs. We found that these pathological neurons represent only 15–20% of the units sampled in the SOZ. Such pathological neurons may be excited at the UP-DOWN transition by decreased inhibitory tone or possibly depolarizing inhibition. Epileptogenic mesial-temporal regions also exhibit decreased physiological and pathological ripple coincidences. Ripple temporal correlations between brain areas may enable information transfer and memory consolidation. Decreased ripple temporal correlations within epileptogenic regions, and across structures, may be due to pathological neurons that abnormally generate pHFOs at the UP-DOWN transition or the pHFOs themselves. Decreased ripple temporal correlations may be an important mechanism in memory impairment in patients with epilepsy.

Supplementary Material

Supinfo

Key Points.

  1. In the SOZ, HFO probability is highest during the UP-DOWN transition of slow wave excitability.

  2. In the SOZ, action potentials associated with HFOs occur at the highest probability at the UP-DOWN transition.

  3. In the SOZ, a subpopulation of individual units fire more robustly during HFOs at UP-DOWN compared to HFOs at DOWN-UP.

  4. Ripple temporal correlations between the hippocampus and other mesial-temporal structures are reduced in the SOZ.

Acknowledgements:

We thank Dr. Gyorgy Buzsaki, Dr. Anli Liu, and Dr. Simon Henin for their helpful suggestions prior to drafting the manuscript.

Funding:

This work was fully supported by the National Institute of Health K23 NS094633, a Junior Investigator Award from the American Epilepsy Society (S.A.W.), R01 NS106957(R.J.S.) and R01 NS033310 (J.E.), the Resnick family foundation (J.E.), and the Christina Louise George Trust (R.J.S., J.E.), European Research Council ERC-2019-CoG 864353 (Y.N.). The views, opinions and/or findings contained in this material are those of the authors and should not be interpreted as representing the official views or policies of the U.S. Government or the American Epilepsy Society.

Footnotes

Epilepsia ethical publishing statement: “We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.”

Disclosures Statement: S.A.W., I.F., J.E., A.B., S.W., R.K.S.W, Y.N has nothing to disclose, M.R.S has received compensation for speaking at continuing medical education (CME) programs from Medscape, Projects for Knowledge, International Medical Press, and Eisai. He has consulted for Medtronic, Neurelis, and Johnson & Johnson. He has received research support from Eisai, Medtronic, Neurelis, SK Life Science, Takeda, Xenon, Cerevel, UCB Pharma, Janssen, and Engage Pharmaceuticals. He has received royalties from Oxford University Press and Cambridge University Press.

Data availability statements:

The processed data used to generate the figures and statistics in this paper are available at https://zenodo.org/record/8127018, the code is available at https://github.com/shenweiss/publishedcode/tree/master/UPDOWNHFO. The raw data is available upon reasonable request to Dr. Richard Staba or Dr. Yuval Nir.

References

  • 1.Herman ST, Walczak TS, Bazil CW. Distribution of partial seizures during the sleep--wake cycle: differences by seizure onset site. Neurology. 2001; 56(11):1453–9. [DOI] [PubMed] [Google Scholar]
  • 2.Loddenkemper T, Vendrame M, Zarowski M, Gregas M, Alexopoulos AV, Wyllie E, et al. Circadian patterns of pediatric seizures. Neurology. 2011; 76(2):145–53. [DOI] [PubMed] [Google Scholar]
  • 3.Lambert I, Tramoni-Negre E, Lagarde S, Roehri N, Giusiano B, Fonseca AT, et al. Hippocampal Interictal Spikes during Sleep Impact Long-Term Memory Consolidation. Ann Neurol. 2020; 87(6):976–87. [DOI] [PubMed] [Google Scholar]
  • 4.Conrad EC, Tomlinson SB, Wong JN, Oechsel KF, Shinohara RT, Litt B, et al. Spatial distribution of interictal spikes fluctuates over time and localizes seizure onset. Brain. 2019; 143(2):554–69. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Conrad EC, Revell AY, Greenblatt AS, Gallagher RS, Pattnaik AR, Hartmann N, et al. Spike patterns surrounding sleep and seizures localize the seizure-onset zone in focal epilepsy. Epilepsia. 2023; 64(3):754–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Staba RJ, Wilson CL, Bragin A, Jhung D, Fried I, Engel J. High-frequency oscillations recorded in human medial temporal lobe during sleep. Ann Neurol. 2004; 56(1):108–15. [DOI] [PubMed] [Google Scholar]
  • 7.Li L, Kumar U, You J, Zhou Y, Weiss SA, Engel J, et al. Spatial and temporal profile of high-frequency oscillations in posttraumatic epileptogenesis. Neurobiol Dis. 2021; 161:105544. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Nevalainen P, Ellenrieder N von, Klimeš P, Dubeau F, Frauscher B, Gotman J. Association of fast ripples on intracranial EEG and outcomes after epilepsy surgery. Neurology. 2020; 95(16):e2235–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Bragin A, Azizyan A, Almajano J, Wilson CL, Engel J. Analysis of Chronic Seizure Onsets after Intrahippocampal Kainic Acid Injection in Freely Moving Rats. Epilepsia. 2005; 46(10):1592–8. [DOI] [PubMed] [Google Scholar]
  • 10.Weiss SA, Alvarado-Rojas C, Bragin A, Behnke E, Fields T, Fried I, et al. Ictal onset patterns of local field potentials, high frequency oscillations, and unit activity in human mesial temporal lobe epilepsy. Epilepsia. 2016; 57(1):111–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Pallud J, Quyen MLV, Bielle F, Pellegrino C, Varlet P, Labussiere M, et al. Cortical GABAergic excitation contributes to epileptic activities around human glioma. Sci Transl Med. 2014; 6(244):244ra89–244ra89. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Buzsáki G. Hippocampal sharp wave-ripple: A cognitive biomarker for episodic memory and planning. Hippocampus. 2015; 25(10):1073–188. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Khodagholy D, Gelinas JN, Buzsáki G. Learning-enhanced coupling between ripple oscillations in association cortices and hippocampus. Science. 2017; 358(6361):369–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Liu AA, Henin S, Abbaspoor S, Bragin A, Buffalo EA, Farrell JS, et al. A consensus statement on detection of hippocampal sharp wave ripples and differentiation from other fast oscillations. Nat Commun. 2022; 13(1):6000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Wilson MA, McNaughton BL. Reactivation of Hippocampal Ensemble Memories During Sleep. Science. 1994; 265(5172):676–9. [DOI] [PubMed] [Google Scholar]
  • 16.Helfrich RF, Lendner JD, Mander BA, Guillen H, Paff M, Mnatsakanyan L, et al. Bidirectional prefrontal-hippocampal dynamics organize information transfer during sleep in humans. Nat Commun. 2019; 10(1):3572. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Dickey CW, Verzhbinsky IA, Jiang X, Rosen BQ, Kajfez S, Stedelin B, et al. Widespread ripples synchronize human cortical activity during sleep, waking, and memory recall. Proc National Acad Sci. 2022; 119(28):e2107797119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Dickey CW, Verzhbinsky IA, Jiang X, Rosen BQ, Kajfez S, Eskandar EN, et al. Cortical Ripples during NREM Sleep and Waking in Humans. J Neurosci. 2022; 42(42):7931–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Dahal P, Rauhala OJ, Khodagholy D, Gelinas JN. Hippocampal–cortical coupling differentiates long-term memory processes. Proc National Acad Sci. 2023; 120(7):e2207909120. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Vaz AP, Inati SK, Brunel N, Zaghloul KA. Coupled ripple oscillations between the medial temporal lobe and neocortex retrieve human memory. Science. 2019; 363(6430):975–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Steriade M, Nuñez A, Amzica F. Intracellular analysis of relations between the slow (< 1 Hz) neocortical oscillation and other sleep rhythms of the electroencephalogram. J Neurosci Official J Soc Neurosci. 1993; 13(8):3266–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Steriade M, Nuñez A, Amzica F. A novel slow (< 1 Hz) oscillation of neocortical neurons in vivo: depolarizing and hyperpolarizing components. J Neurosci Official J Soc Neurosci. 1993; 13(8):3252–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Nir Y, Mukamel R, Dinstein I, Privman E, Harel M, Fisch L, et al. Interhemispheric correlations of slow spontaneous neuronal fluctuations revealed in human sensory cortex. Nat Neurosci. 2008; 11(9):1100–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Isomura Y, Sirota A, Özen S, Montgomery S, Mizuseki K, Henze DA, et al. Integration and Segregation of Activity in Entorhinal-Hippocampal Subregions by Neocortical Slow Oscillations. Neuron. 2006; 52(5):871–82. [DOI] [PubMed] [Google Scholar]
  • 25.Nir Y, Staba RJ, Andrillon T, Vyazovskiy VV, Cirelli C, Fried I, et al. Regional Slow Waves and Spindles in Human Sleep. Neuron. 2011; 70(1):153–69. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Levenstein D, Buzsáki G, Rinzel J. NREM sleep in the rodent neocortex and hippocampus reflects excitable dynamics. Nat Commun. 2019; 10(1):2478. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Frauscher B, Ellenrieder N von, Ferrari-Marinho T, Avoli M, Dubeau F, Gotman J. Facilitation of epileptic activity during sleep is mediated by high amplitude slow waves. Brain. 2015; 138(6):1629–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Ellenrieder N von, Dubeau F, Gotman J, Frauscher B. Physiological and pathological high-frequency oscillations have distinct sleep-homeostatic properties. Neuroimage Clin. 2017; 14:566–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Song I, Orosz I, Chervoneva I, Waldman ZJ, Fried I, Wu C, et al. Bimodal coupling of ripples and slower oscillations during sleep in patients with focal epilepsy. Epilepsia. 2017; 58(11):1972–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Ellenrieder N, Frauscher B, Dubeau F, Gotman J. Interaction with slow waves during sleep improves discrimination of physiologic and pathologic high-frequency oscillations (80–500 Hz). Epilepsia. 2016; 57(6):869–78. [DOI] [PubMed] [Google Scholar]
  • 31.Andrillon T, Nir Y, Staba RJ, Ferrarelli F, Cirelli C, Tononi G, et al. Sleep Spindles in Humans: Insights from Intracranial EEG and Unit Recordings. J Neurosci. 2011; 31(49):17821–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Elahian B, Lado NE, Mankin E, Vangala S, Misra A, Moxon K, et al. Low-Voltage Fast Seizures in Humans Begin With Increased Interneuron Firing. Ann Neurol. 2018; 84(4):588–600. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Shimamoto S, Waldman ZJ, Orosz I, Song I, Bragin A, Fried I, et al. Utilization of independent component analysis for accurate pathological ripple detection in intracranial EEG recordings recorded extra- and intra-operatively. Clin Neurophysiol. 2018; 129(1):296–307. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Waldman ZJ, Shimamoto S, Song I, Orosz I, Bragin A, Fried I, et al. A method for the topographical identification and quantification of high frequency oscillations in intracranial electroencephalography recordings. Clin Neurophysiol. 2018; 129(1):308–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Weiss SA, Berry B, Chervoneva I, Waldman Z, Guba J, Bower M, et al. Visually validated semi-automatic high-frequency oscillation detection aides the delineation of epileptogenic regions during intra-operative electrocorticography. Clin Neurophysiol. 2018; 129(10):2089–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Buzsáki G, Anastassiou CA, Koch C. The origin of extracellular fields and currents — EEG, ECoG, LFP and spikes. Nat Rev Neurosci. 2012; 13(6):407–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Chen C, Wang Y, Ye L, Xu J, Ming W, Liu X, et al. A region-specific modulation of sleep slow waves on interictal epilepsy markers in focal epilepsy. Epilepsia. 2023; . [DOI] [PubMed] [Google Scholar]
  • 38.Weiss SA, Sheybani L, Seenarine N, Fried I, Wu C, Sharan A, et al. Delta oscillation coupled propagating fast ripples precede epileptiform discharges in patients with focal epilepsy. Neurobiol Dis. 2022; :105928. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Foffani G, Uzcategui YG, Gal B, Prida LM de la. Reduced Spike-Timing Reliability Correlates with the Emergence of Fast Ripples in the Rat Epileptic Hippocampus. Neuron. 2007; 55(6):930–41. [DOI] [PubMed] [Google Scholar]
  • 40.Ibarz JM, Foffani G, Cid E, Inostroza M, Prida LM de la. Emergent Dynamics of Fast Ripples in the Epileptic Hippocampus. J Neurosci. 2010; 30(48):16249–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Bragin A, Mody I, Wilson CL, Engel J. Local Generation of Fast Ripples in Epileptic Brain. J Neurosci. 2002; 22(5):2012–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Bragin A, Wilson CL, Engel J. Chronic Epileptogenesis Requires Development of a Network of Pathologically Interconnected Neuron Clusters: A Hypothesis. Epilepsia. 2000; 41(s6):S144–52. [DOI] [PubMed] [Google Scholar]
  • 43.Bragin A, Benassi SK, Kheiri F, Engel J. Further evidence that pathologic high-frequency oscillations are bursts of population spikes derived from recordings of identified cells in dentate gyrus. Epilepsia. 2011; 52(1):45–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Bragin A, Wilson CL, Almajano J, Mody I, Engel J. High-frequency Oscillations after Status Epilepticus: Epileptogenesis and Seizure Genesis. Epilepsia. 2004; 45(9):1017–23. [DOI] [PubMed] [Google Scholar]
  • 45.Bragin A, Wilson CL, Engel J. Spatial Stability over Time of Brain Areas Generating Fast Ripples in the Epileptic Rat. Epilepsia. 2003; 44(9):1233–7. [DOI] [PubMed] [Google Scholar]
  • 46.Bragin A, Wilson CL, Engel J. Voltage Depth Profiles of High-frequency Oscillations after Kainic Acid-induced Status Epilepticus. Epilepsia. 2007; 48(s5):35–40. [DOI] [PubMed] [Google Scholar]
  • 47.Neske GT, Patrick SL, Connors BW. Contributions of Diverse Excitatory and Inhibitory Neurons to Recurrent Network Activity in Cerebral Cortex. J Neurosci. 2015; 35(3):1089–105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Haider B, Duque A, Hasenstaub AR, McCormick DA. Neocortical Network Activity In Vivo Is Generated through a Dynamic Balance of Excitation and Inhibition. J Neurosci. 2006; 26(17):4535–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Sanchez-Vives MV, Mattia M, Compte A, Perez-Zabalza M, Winograd M, Descalzo VF, et al. Inhibitory Modulation of Cortical Up States. J Neurophysiol. 2010; 104(3):1314–24. [DOI] [PubMed] [Google Scholar]
  • 50.Valero M, Viney TJ, Machold R, Mederos S, Zutshi I, Schuman B, et al. Sleep down state-active ID2/Nkx2.1 interneurons in the neocortex. Nat Neurosci. 2021; 24(3):401–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Cohen I, Navarro V, Clemenceau S, Baulac M, Miles R. On the Origin of Interictal Activity in Human Temporal Lobe Epilepsy in Vitro. Science. 2002; 298(5597):1418–21. [DOI] [PubMed] [Google Scholar]
  • 52.Huberfeld G, Prida LM de la, Pallud J, Cohen I, Quyen MLV, Adam C, et al. Glutamatergic pre-ictal discharges emerge at the transition to seizure in human epilepsy. Nat Neurosci. 2011; 14(5):627–34. [DOI] [PubMed] [Google Scholar]
  • 53.Weiss SA. Chloride ion dysregulation in epileptogenic neuronal networks. Neurobiol Dis. 2023; 177:106000. [DOI] [PubMed] [Google Scholar]
  • 54.Alfonsa H, Burman RJ, Brodersen PJN, Newey SE, Mahfooz K, Yamagata T, et al. Intracellular chloride regulation mediates local sleep pressure in the cortex. Nat Neurosci. 2023; 26(1):64–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Vyazovskiy VV, Olcese U, Lazimy YM, Faraguna U, Esser SK, Williams JC, et al. Cortical Firing and Sleep Homeostasis. Neuron. 2009; 63(6):865–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Peyrache A, Khamassi M, Benchenane K, Wiener SI, Battaglia FP. Replay of rule-learning related neural patterns in the prefrontal cortex during sleep. Nat Neurosci. 2009; 12(7):919–26. [DOI] [PubMed] [Google Scholar]
  • 57.Maingret N, Girardeau G, Todorova R, Goutierre M, Zugaro M. Hippocampo-cortical coupling mediates memory consolidation during sleep. Nat Neurosci. 2016; 19(7):959–64. [DOI] [PubMed] [Google Scholar]
  • 58.Weiss SA, Eliashiv D, Stern J, Rubinstein D, Fried I, Wu C, et al. Stimulation better targets fast-ripple generating networks in super responders to the responsive neurostimulator system. Epilepsia. 2023; . [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Weiss SA, Fried I, Wu C, Sharan A, Rubinstein D, Engel J, et al. Graph theoretical measures of fast ripple networks improve the accuracy of post-operative seizure outcome prediction. Sci Rep-uk. 2023; 13(1):367. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Weiss SA, Pastore T, Orosz I, Rubinstein D, Gorniak R, Waldman Z, et al. Graph theoretical measures of fast ripples support the epileptic network hypothesis. Brain Commun. 2022; . [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61.Kucewicz MT, Cimbalnik J, Matsumoto JY, Brinkmann BH, Bower MR, Vasoli V, et al. High frequency oscillations are associated with cognitive processing in human recognition memory. Brain. 2014; 137(8):2231–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62.Sheybani L, Mierlo P van, Birot G, Michel CM, Quairiaux C. Large-scale 3–5 Hz oscillation constrains the expression of neocortical fast-ripples in a mouse model of mesial temporal lobe epilepsy. Eneuro. 2019; 6(1):ENEURO.0494-18.2019. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supinfo
NIHMS1952511-supplement-Supinfo.pdf (604.3KB, pdf)

Data Availability Statement

The processed data used to generate the figures and statistics in this paper are available at https://zenodo.org/record/8127018, the code is available at https://github.com/shenweiss/publishedcode/tree/master/UPDOWNHFO. The raw data is available upon reasonable request to Dr. Richard Staba or Dr. Yuval Nir.

RESOURCES