8. Effects of High-Frequency Stimulations on Epileptiform Discharges

We applied O-HFS trains with biphasic-pulses at 100 Hz to the Schaffer collaterals in two durations (5 s and 1 min) respectively and recorded the CA1 neuronal responses downstream of the stimulation site (Fig. 8.1A). The 5-s O-HFS produced an after-discharge (AD) with large population spikes (PS) that lasted about 10 s after the O-HFS ended (Fig. 8.1A1, upper). However, during the 1-min O-HFS, following the same initial 5 s stimulation, only small PSs occurred, and no AD appeared after the O-HFS ended (Fig. 8.1A1, bottom). We termed this period of small PSs during HFS as the “HFS-discharge”. The neuronal responses were similar during the initial 5 s periods of both O-HFS durations (enlarged views of ① and ② in Fig. 8.1A2), but the subsequent responses differed (enlarged view of ③ in Fig. 8.1A2). The mean duration of the ADs following 5 s O-HFS showed no significant difference from that of the HFS-discharges within 1 min O-HFS (Fig. 8.1B1). However, the mean PS amplitude in the ADs was significantly larger than that in HFS-discharges (Fig. 8.1B2), indicating that the HFS beyond 5 s suppressed PSs. Although the AD-PSs showed larger amplitudes, they had smaller mean full width at half maximum (FWHM, Fig. 8.1B3) and lower mean rate (Fig. 8.1B4). This resulted in similar sums of PS areas per second during both ADs and HFS-discharges (Fig. 8.1B5). Therefore, while the overall firing levels were similar, the HFS-discharge showed lower synchronicity.

Full size image

To further investigate the effect of continuous O-HFS on PS discharges, two 5-s trains of O-HFS were applied with a 5-s interval (Fig. 8.2). The first O-HFS produced an AD with large PSs. The second O-HFS significantly suppressed PS amplitudes, but large PSs reappeared after it ended. This effect was consistent at both 100 Hz and 200 Hz O-HFS frequencies.

Full size image

These results reveal an interesting phenomenon that persisting the excitatory inputs by extending O-HFS can suppress rather than enhance PS discharges. Previous studies have shown that AD stems from impaired inhibitions of GABAergic synapses on pyramidal neurons (Fujiwara-Tsukamoto et al. 2007; Ye and Kaszuba 2017). In the initial period of O-HFS, the strong excitatory inputs—from both O-HFS and produced pyramidal neuron discharges—can activate interneurons in the CA1 local inhibitory circuits through feedforward and feedback connections (Fig. 8.1A, refer to Sect. 2.3 for details). These activations can trigger GABAergic synapses on post-synaptic pyramidal neurons. When these inhibitory synapses are over activated by HFS, excessive chloride ion inflow and bicarbonate ion outflow can result in rapid changes in ion concentrations across post-synaptic membranes (Isomura et al. 2003). This can reverse the normal hyperpolarized synaptic potential into depolarization, converting GABAergic synapses on post-synaptic pyramidal neurons from inhibitory to excitatory (Staley et al. 1995; McCarren and Alger 1985; Pearce et al. 1995). Additionally, brief O-HFS trains on the Schaffer collateral axons can enhance synaptic transmission efficiency through synaptic plasticity (Zucker and Regehr 2002; Rotman et al. 2011; Hennig 2013). These factors can disrupt the balance between inhibition and excitation on pyramidal neurons, leading to AD occurrence. Reports have shown that brief HFS trains can produce ADs in various brain regions of both animals and humans (Jobst et al. 2010; Shigeto et al. 2013; Hannan et al. 2020; Lesser et al. 1999).

The decreased PS amplitude during long O-HFS indicated a suppression from sustained O-HFS. An amplitude decrease in PSs can stem from two factors: fewer neurons participating in the discharge or/and reduced firing synchronicity among neurons. The PSs in HFS-discharge exhibited a larger mean FWHM in waveform (Fig. 8.1B3) and a higher firing rate (Fig. 8.1B4), showing more frequent but less synchronized in the discharges. Additionally, the mean durations of HFS-discharges and ADs were similar (Fig. 8.1B1) with comparable sum areas of PSs (Fig. 8.1B5), suggesting that the overall firing levels during the two discharge types were similar. Therefore, decreased firing synchronicity can be a possible mechanism of the PS suppression during HFS-discharges (Wang et al. 2021). The unit spike analysis in Sect. 5.3 also supports this point, showing that sustained O-HFS has an excitatory effect on downstream postsynaptic neurons, but with reduced synchronicity.

The above experiments showed that prolonged O-HFS can suppress PSs; however, the AD epileptiform activity used as a control was itself induced by the initial O-HFS. To validate the suppressive effect of O-HFS on epileptiform discharges, we tested it further using other epileptiform models as below.

We used 4-aminopyridine (4-AP), an epileptogenic agent, to create an epileptiform model. 4-AP is a potassium channel blocker that can slow the repolarization of action potentials, prolong the depolarization state and increase the excitability of neuronal membrane (Bean 2007; Storm 1987). Additionally, the prolonged depolarization caused by 4-AP can increase the release of excitatory neurotransmitters like glutamates from presynaptic membranes (Tibbs et al. 1989; Thomsen and Wilson 1983). It can also increase the release of inhibitory neurotransmitters like GABA (Peña and Tapia 1999, 2000). However, overactivation of GABAergic synapses can weaken and reverse their inhibitory effect. Collectively, these effects enable 4-AP to generate epileptiform activity (Perreault and Avoli 1992; Lévesque et al. 2013; Chiang et al. 2013).

To apply 4-AP locally in the rat hippocampal CA1 region, we made a drug delivery device by using a segment of model 7 puncture needle as an injection tubing, then connected its tail to a microinjector through silicone tubing (Fig. 8.3). The needle tip was implanted alongside the recording electrode at the same posterior bregma coordinate, but closer to the midline—about 1 mm from the recording electrode (Fig. 8.4A). To induce epileptiform discharges, 4-AP solution (40 mM, 0.5–1 μL) was injected into the hippocampal CA1 region near the recording site.

Full size image

Full size image

In baseline recording before 4-AP administration, no PS occurred (Fig. 8.4B). After applying 4-AP, epileptiform discharges gradually developed (Fig. 8.4C), producing waveforms consistent with previous studies (Chiang et al. 2013). Two types of epileptiform waveforms appeared: large-amplitude PSs and small positive wavelets (SPW) at a frequent around 100 Hz. Some PSs were immediately followed by SPW (see the enlarged insets in Fig. 8.4C). During the epileptiform period, a 2-min O-HFS was applied on the Schaffer collaterals. While the first O-HFS pulse produced a large PS, only small PSs and SPWs occurred during subsequent O-HFS (Fig. 8.4D). Statistical data (Fig. 8.4E) showed that both 100 Hz and 200 Hz O-HFS significantly reduced PS rates and amplitudes (Fig. 8.4E1, E2). Consequently, the sum of PS amplitudes per second decreased significantly compared to pre-O-HFS control values (Fig. 8.4E3). After the end of O-HFS, all these measurements returned to their control levels (Fig. 8.4E1–E3).

Figure 8.5 further shows the suppressive effect of O-HFS on PSs by intermittent 5-s O-HFS trains applied at 5-s intervals. During the total 2-min period of 24 trains at 200 Hz O-HFS, PSs rarely occurred except at O-HFS onsets. However, large PSs occurred during all of the intervals (Fig. 8.5A1, A2). The PSs at the O-HFS onsets showed significant variations (Fig. 8.5A3–A5). When O-HFS started during SPW, it produced continuous SPW without obvious PS (Fig. 8.5A1, A3). When O-HFS started immediately after a large PS, the first O-HFS pulse produced either a small PS (Fig. 8.5A2, A4) or a large PS (Fig. 8.5A5), with no more obvious PSs following. Among the 24 O-HFS trains shown in Fig. 8.5A, the first pulses produced large PSs in 15 trains (highlighted in orange in the aligned plots of 24 onset signals in Fig. 8.5B).

Full size image

To analyze the PS changes during 4-AP-induced epileptiform discharges, we used a bandpass filter (10–3000 Hz) to remove frequency components outside the PS waveform range. A threshold of − 2 mV was then set to detect PSs (Fig. 8.5A). Consistent with the results from the 2-min continuous O-HFSs shown in Fig. 8.4E, the mean PS rate was significantly lower during the intermittent 5-s O-HFS periods than 5-s interval periods (Fig. 8.5C1). While PS amplitudes showed no significant difference between O-HFS and interval periods—due to single large PSs at O-HFS onsets (Fig. 8.5C2), the sum of PS amplitudes per second was significantly smaller during O-HFS periods than interval periods (Fig. 8.5C3).

These results showed that O-HFS at afferent fiber axons can rapidly and persistently suppress large PSs in 4-AP-induced epileptiform discharges. This suppression could occur due to a depolarization block of downstream CA1 pyramidal neurons, caused by overexcitation from O-HFS excitatory inputs. To test this hypothesis, we applied paired-pulse antidromic test stimulations (ATS) to the alveus to determine whether the CA1 pyramidal neurons had lost their firing ability during O-HFS.

As shown in Fig. 8.6A, during baseline recording, paired-pulse ATS with a 5-ms interval produced two large APSs (APS₁ and APS₂) with similar amplitudes. After 4-AP administration, during an interval of epileptiform discharges without PSs, an ATS also produced two large APSs followed by SPW (ATS #1 in Fig. 8.6B1). ATS paired-pulses were then applied every 5 s (marked by orange dots). The second ATS, applied during SPW period, produced no obvious APS (ATS #2 in Fig. 8.6B1). The onset of O-HFS during this SPW period also failed to produce large PS (Fig. 8.6B2), consistent with the pattern shown in Fig. 8.5A3. During the 2-min continuous O-HFS period, the ATS pulses produced only small APSs at most (Fig. 8.6B3). Throughout the O-HFS period, the APSs induced by ATSs were significantly suppressed (Fig. 8.6B, bottom). Moreover, this APS suppression was particularly pronounced during SPW periods. Unlike the baseline situation shown in Fig. 8.6A, APS₂ induced by the second pulse of ATS was significantly smaller than APS₁. After the end of O-HFS, ATSs produced large APSs again (Fig. 8.6B4).

Full size image

The effects of O-HFS at 100 and 200 Hz were similar. The pooled data from both frequencies showed that during the 4-AP-induced epileptiform period, the amplitudes of ATS-induced APS₁ and APS₂ were significantly smaller during O-HFS than before O-HFS (Fig. 8.6C). Additionally, the amplitude ratio of APS₂/APS₁ decreased progressively from baseline control (before 4-AP administration), to the 4-AP-induced epileptiform period, and finally during the O-HFS application in the epileptiform period (Fig. 8.6D). The decreased APS₂ indicated an extension of the APS₁ refractory period. These substantial APS suppressions indicated that synchronous firing was prevented by depolarization block due to excessive neuronal excitation. Therefore, O-HFS can suppress 4-AP-induced epileptiform PSs by driving downstream neurons into a depolarization block state.

In addition to the biphasic-pulse O-HFS stimulations used in the above experiments, sinusoidal stimulations at the Schaffer collaterals can also modulate the firing in downstream neurons (Wang et al. 2020). We investigated whether sinusoidal stimulations could suppress 4-AP-induced epileptiform discharges (Guo et al. 2016). We first determined the appropriate sinusoidal intensity (peak-to-peak current amplitude). As shown in Fig. 8.7A, the baseline recording showed spontaneous unit spikes of neurons (marked by the arrows). When 50 Hz sinusoidal stimulation was applied with increasing intensities, its modulation on neuronal firing gradually increased (Fig. 8.7B). At 15 μA intensity (Fig. 8.7B1), unit spikes appeared randomly across different phases of sinusoidal waves, indicating minimal effect on neuronal firing. At 30μA intensity (Fig. 8.7B2), the firing showed a clear phase-locked pattern, with most spikes occurring at the beginning of sinusoidal descending phases. When the intensity increased to 50 μA, the PSs formed by synchronous firing of neuronal populations appeared (marked by orange dots in Fig. 8.7B3).

Full size image

We set the sinusoidal stimulation at an intensity that was able to modulate neuronal firing but without triggering large PS discharges. As shown in Fig. 8.8A, we applied a 40 μA 50 Hz sinusoidal stimulation for 2 min during the 4-AP-induced epileptiform period. Under the stimulation, 4-AP-induced large PSs disappeared while SPW remained. Statistical data showed that during the 2-min stimulation at a mean intensity of 32 ± 9.7 µA (n = 5 rats), the PS rate, PS amplitude, and sum of PS amplitudes were significantly smaller than their pre-stimulation levels (Fig. 8.8B). The results indicate that sinusoidal stimulation can suppress 4-AP-induced epileptiform discharges.

Full size image

The experiment results from both pulse and sinusoidal stimulations show that orthodromic high-frequency stimulations can suppress the synchronous discharges generated by 4-AP. The underlying mechanism may involve the stimulation excitatory inputs further elevating the depolarization of downstream CA1 pyramidal neurons under 4-AP-induced discharges, leading to depolarization block and preventing neuronal synchronous firing, as verified by ATS experiments (Fig. 8.6). We then investigated whether the O-HFS excitation could also suppress epileptiform discharges induced by another epileptogenic agent—picrotoxin.

Picrotoxin (PTX) is an antagonist of the GABA_A receptor. By blocking GABAergic inhibitory synapses in local circuits, it can destroy the balance between neuronal excitation and inhibition, leading to epileptiform discharges.

In baseline recording, no PSs occurred in normal LFP signal (Fig. 8.9A). After injecting PTX (4 mM, 2–4 μL) through the drug tubing into the CA1 region near recording site, epileptiform discharges gradually appeared in the cell layer recording. The discharges were characterized by continuous bursts, each consisting of multiple PSs (Fig. 8.9B). During the PTX-induced epileptiform period, we applied 100 Hz O-HFS at the Schaffer collaterals for 1 min, which significantly increased the mean PS rate (Fig. 8.9C) from 0.68 ± 0.27 counts/s before O-HFS to 5.7 ± 1.0 counts/s during the O-HFS periods (P < 0.01, paired t-test, n = 6 rats). However, mean PS amplitude before stimulation (7.1 ± 1.2 mV) was not significantly different from that during O-HFS (7.3 ± 1.1 mV, P = 0.58, paired t-test, n = 6).

Full size image

This result suggests that sustained O-HFS can enhance rather than suppress PTX-induced epileptiform PSs. However, during the early stage of O-HFS, there was always a PS suppression period (Fig. 8.9C), indicating that short O-HFS might suppress PSs. Clinical studies have shown that adaptive neural stimulation (RNS), using sub-second brief stimulations, can suppress epileptiform seizures (Geller et al. 2017; Nune et al. 2015). Based on this clue, we designed a closed-loop stimulation system (see Sect. 3.5.6 for details) and applied 1-s brief O-HFS trains immediately following the first PSs of PTX-induced bursts. As shown in Fig. 8.10A, these brief O-HFS trains significantly suppressed the subsequent PSs in bursts, though SPW still occurred. The sum of PS amplitudes in each burst was significantly smaller with O-HFS than without O-HFS (Fig. 8.10B), with the first PS in bursts before O-HFS initiation being the main contributor to the PS amplitudes during O-HFS. Similarly, applying a 0.3-s brief O-HFS train triggered by the first PS of burst consistently suppressed subsequent PSs in each burst (Fig. 8.10C).

Full size image

Similar to the method described in Sect. 8.2, we used ATS at the axons of CA1 pyramidal neurons to test their ability to generate synchronous firing (see Fig. 8.6A), and used single-pulse ATSs here. As shown in Fig. 8.11A, a PTX-induced burst began with a large initial PS of multiple spikes, followed by subsequent PSs and SPW waves. During such a burst, the application of ATS pulses at 10 Hz persistently induced large APSs and SPW waves (Fig. 8.11B, orange dots indicate ATSs). However, when we simultaneously applied a 1-s O-HFS, ATSs failed to induce large APSs during the O-HFS period, and only SPW waves appeared (Fig. 8.11C). The mean APS amplitude decreased significantly during O-HFS compared to the control period before O-HFS. After O-HFS ended, ATSs again induced large APSs comparable to those in the control period (Fig. 8.11D). The result indicates that O-HFS application during bursts can impair the synchronous firing ability of pyramidal neurons.

Full size image

These results show that brief O-HFS trains can suppress large PSs within PTX-induced bursts. Based on the changes in ATS-induced APSs, we can infer the underlying mechanisms: When PTX blocks the inhibitory GABA_A synapses and increases the excitability of pyramidal neurons, epileptiform bursts occur and ATSs can induce large APSs. The addition of brief O-HFS application can further elevate the neuronal membrane depolarization, resulting in a depolarization block in the post-synaptic pyramidal neurons that prevents their firing. The decrease or absence of ATS-induced APSs confirmed the present of this block (Fig. 8.11C). In contrast, when long rather than brief O-HFS is applied, the sustained stimulation can produce intermittent depolarization block in the directly stimulated axons. This can attenuate the O-HFS excitation to downstream pyramidal neurons—not enough to cause depolarization block but sufficient to enhance synchronous discharges in these neurons, as shown in Fig. 8.9C.

This chapter shows the effects of electrical stimulations on epileptiform discharges. We induced these discharges locally in the rat hippocampal CA1 region using three methods: short O-HFS trains and two epileptogenic agents (4-AP and PTX). We investigated whether applying O-HFS to the Schaffer collaterals—the CA1 afferent fiber—could suppress the epileptiform discharges of pyramidal neurons downstream from the stimulation site. Our results demonstrate that both pulse and sinusoidal O-HFS can suppress epileptiform PSs induced by these methods, although the effective stimulation patterns vary according to the epileptogenic method used. The underlying mechanisms can involve desynchronization and depolarization block, as analyzed below.

Under normal physiological conditions, a single-pulse stimulation with sufficient intensity at the Schaffer collaterals can strongly excite downstream neuron populations and induce their synchronous firing to form a large population spike (PS) (see Sect. 2.3). A brief train of HFS pulses can also induce epileptiform firing—after-discharge (AD) consisting of PSs (Leung 1987; Bragin et al. 1997a, 1997b). Although sustained HFS can lead to axonal block and weaken the HFS excitation on downstream neurons, this excitation can still increase the firing of these neurons (see Sect. 5.3 for details). This excitatory effect can explain why animal and clinical trials have shown that HFS can enhance epileptiform activity (Gloor et al. 1982; Lado 2006; Feddersen et al. 2007). However, experiments on in-vitro brain slices have shown that HFS using both sinusoidal and pulse waveforms can suppress epileptiform activity in the hippocampus. Additionally, these stimulations increased the extracellular potassium concentration ([K⁺]_o), suggesting that the suppress may result from desynchronized activation and/or depolarization block of neuronal membranes (Bikson et al. 2001; Durand and Bikson 2001). These two mechanisms can also explain the results from the three types of epileptiform models shown in this chapter.

The smaller and more frequent PSs during “HFS-discharge” than during AD indicated a desynchronization effect of HFS. As shown in Sect. 5.3.1, large APSs induced by ATSs during sustained O-HFS suggested that this desynchronization originated from the pre-synaptic asynchronous excitations of the Schaffer collaterals, not from post-synaptic neurons. The asynchronous inputs resulted from O-HFS-induced intermittent depolarization block at stimulated axons. Additionally, sustained O-HFS can eliminate synaptic potentiation caused by initial brief stimulation (Zhou et al. 2017). During sustained O-HFS, the induced axonal block can also avoid excessive excitations to downstream interneurons, allowing GABAergic synapses to restore their inhibitory effect and prevent large PS events, as shown in Sect. 5.3. Therefore, unlike brief O-HFS which has an epileptogenic effect, sustained O-HFS can prevent the formation of large epileptiform PSs under certain conditions.

Epileptogenic agent 4-AP can increase neuronal excitability by blocking potassium ion channels, resulting in epileptiform activity. When injected locally into the hippocampus, 4-AP produced two types of epileptiform waveforms—large PSs and small positive waves (SPW) in our experiments, similar to those reported in previous studies (Perreault and Avoli 1992; Chiang et al. 2013). Brain slice experiments with both intracellular and extracellular recordings have shown that PS results from synchronized action potentials of neuronal populations, while SPW represents incomplete action potentials appearing as fluctuations on extended depolarization potentials (Avoli et al. 1993). This indicates that during SPW, neuronal membranes stay in a depolarized state. During 4-AP-induced epileptiform discharges, both pulse and sinusoidal O-HFS maintained SPW activity but did not induce large PSs, likely because the stimulation further excited the downstream neurons to a level sufficient to reach depolarization block (refer to Figs. 8.4, 8.5, 8.6 and 8.8). The significant decrease in ATS-induced APSs confirmed the reduced synchronous firing ability in downstream neurons. The extension of refractory period shown by paired-pulse ATS tests further supported that the post-synaptic downstream neurons were in a prolonged depolarized state. Therefore, the suppression of O-HFS on 4-AP-induced large PSs can be caused by excessive depolarization of downstream neurons to an inactive state. Even when directly stimulated axons experience intermittent blocking, sustained O-HFS can still deliver sufficient excitation to create a depolarization block in post-synaptic pyramidal neurons during 4-AP-induced discharges.

Previous literature has reported that HFS pulses can significantly shorten epileptiform discharge duration and reduce discharge intensity (Schiller and Bankirer 2007; McIntyre et al. 2004). Several possible mechanisms have been proposed to explain the results. First, HFS may enhance the activity of inhibitory neurons, particularly GABAergic interneurons, thereby enhancing synaptic inhibitions on principal neurons and suppressing their epileptiform discharges (Schiller and Bankirer 2007; Dostrovsky et al. 2000; Chiken and Nambu 2013). Second, sustained HFS may lead to axonal conduction block and depletion of synaptic neurotransmitters, preventing the spreading of epileptiform activity (Jensen and Durand 2009). Third, HFS may cause neurons to remain in a depolarized state, leading to ion channel inactivation (depolarization block) and preventing neuronal membranes from generating action potentials (Bikson et al. 2001; Durand and Bikson 2001). However, the first two mechanisms cannot explain the suppressive effect of brief O-HFS trains on epileptiform bursts shown in Sect. 8.4. This is because our epileptiform model was induced by the GABA_A antagonist PTX, and GABA_A is the main neurotransmitter of inhibitory synapses in the hippocampal region (Andersen et al. 2007). Since these inhibitory synapses were blocked by PTX, O-HFS could not have acted through them. Moreover, our experiments in Sect. 8.4 show that brief O-HFS trains lasting only 1 s or less suppressed PSs, while sustained O-HFS lasting for 1 min—sufficient time to cause axonal block—produced strong epileptiform discharges. Therefore, the suppression likely did not occur through axonal conduction block or synaptic neurotransmitter depletion. Instead, the combination of inhibitory synapse block by PTX and additional excitation from O-HFS can lead to excessive excitation, causing depolarization block in neurons that can prevent the generation of large epileptiform spikes.

The schematic diagrams in Fig. 8.12 summarize the distinct excitation levels of neurons and their local neural circuits in the CA1 region under different conditions. Pink and blue colors represent excitatory and inhibitory synapses, respectively. The symbol “−” represents inhibitory effects, while different “ + “ symbols represent various excitatory inputs and enhanced excitations. The sum of “−” and “ + “ symbols within the cell body of pyramidal neuron (depicted as a triangle) determines its excitation level. As shown in Fig. 8.12A, under normal physiological conditions (baseline), pyramidal neurons maintain a balance between excitatory and inhibitory inputs and exhibit “random” spontaneous firing without synchronous epileptiform discharges. Figure 8.12B illustrates the situation during brief O-HFS-induced AD when excitatory synapses at the terminals of the Schaffer collaterals are enhanced and inhibitory synapses of interneurons are transformed into excitatory ones. These combined changes increase the excitability of pyramidal neurons, resulting in epileptiform discharges. Figure 8.12C shows the “HFS-discharge” period during sustained O-HFS when inhibitory synapses tend to recovery and O-HFS continues to enhance excitatory synapses (though with a weaker strength than during initial O-HFS). These changes result in pyramidal neurons firing with less synchronized small PSs and increased unit spikes. Figure 8.12D shows the situation of sustained O-HFS application during 4-AP-induced epileptiform discharges. Both 4-AP and O-HFS enhance excitatory synapses. Additionally, 4-AP transforms inhibitory synapses into excitatory ones and directly prolongs neuronal depolarization. These combined effects greatly enhance the depolarization of pyramidal neurons, leading to depolarization block. Figure 8.12E illustrates the application of brief O-HFS during PTX-induced bursts. PTX blocks inhibitory synapses. When combined with strong excitatory inputs from brief O-HFS, the changes cause depolarization block in pyramidal neurons. In each subgraph of Fig. 8.12, the net count of “+” symbols (after subtracting “−” symbols) in the cell body of pyramidal neuron can indicate its overexcitation level, from lowest to highest: (C) < (B) < (E) < (D). Figure 8.12C and B respectively show neuron states with increased firing and epileptiform discharge. Figure 8.12E, D show overexcitation states that lead to depolarization block. These analyses contain inferences that require further validations.

Full size image

The experiment results presented in this chapter show that high-frequency stimulations can suppress epileptiform discharges through the mechanisms of desynchronization and depolarization block. In our experiments, we mainly used electrical stimulations of narrow pulses. The direct effect of these pulses on neuronal membranes is depolarization (Lowet et al. 2022)—an excitatory rather than inhibitory effect. Especially, depolarization block—which results from further increase of excitation beyond already abnormal levels—can suppress epileptiform discharges. Such suppressive effect through depolarization block acts like “fighting fire with fire” (Gwinn and Spencer 2004). Similar phenomena have been observed in clinical studies. For instance, brief HFS can induce intense after-discharges in human brains (Lesser et al. 1999; Motamedi et al. 2002), which is typically epileptogenic. However, this same stimulation pattern can also suppress epileptiform activity in the human brain (Lesser et al. 1999; Sun and Morrell 2014).

Our findings demonstrate that suppressing epileptiform discharges requires stimulation patterns matching the underlying pathological mechanisms. Different epileptiform situations require different stimulation parameters (e.g., short versus long duration) and approaches (e.g., open-loop versus closed-loop stimulation). In addition, epileptiform PSs—which represent synchronized action potentials from neuronal populations—are characteristic waveforms in epileptiform activity. Continuous PS discharges indicate severe seizures that can cause symptoms such as brain dysfunction and myoclonus. Thus, suppressing PS discharges is crucial for effective epilepsy treatments. Our findings provide new insights for developing DBS treatments for epilepsy.