Anterior cingulate cortex neuron subtypes differentially regulate seizures

Ziqian Yan; Ting Tang; Kaishan Wang; Bin Fu; Zesheng Li; Mingshan Liu; Hongyi Huang; Yanfeng Yang; Qiang Zheng; Yangmin Zheng; Yongzhi Shan; Yumin Luo; Guoguang Zhao

doi:10.1002/epi.70365

. 2026 Jun 30;67(10):5531–5546. doi: 10.1002/epi.70365

Anterior cingulate cortex neuron subtypes differentially regulate seizures

Ziqian Yan ^1,², Ting Tang ², Kaishan Wang ², Bin Fu ², Zesheng Li ², Mingshan Liu ², Hongyi Huang ², Yanfeng Yang ², Qiang Zheng ², Yangmin Zheng ^2,³, Yongzhi Shan ^2,⁴, Yumin Luo ^2,^3,⁴, Guoguang Zhao ^1,^2,^3,^4,^✉

PMCID: PMC13675779 PMID: 42377056

Abstract

Objective

This study aimed to investigate the regulatory roles of distinct neuronal subtypes within the anterior cingulate cortex (ACC) in acute seizures and to identify cell type‐specific mechanisms underlying seizure modulation in this region.

Methods

Acute seizure models were established in mice via pentylenetetrazol injection. In vivo fiber photometry and miniscope calcium imaging were employed to monitor neuronal calcium activity, and multichannel electroencephalography was used to record brain electrical signals simultaneously. Subsequently, bidirectional chemogenetic and optogenetic manipulations were performed on calcium/calmodulin‐dependent protein kinase II (CaMKII) excitatory neurons and vesicular γ‐aminobutyric acid (GABA) transporter (vGAT)‐expressing GABAergic interneurons, as well as parvalbumin (PV) and somatostatin (SST) interneuron subpopulations.

Results

Calcium recordings demonstrated that both excitatory and inhibitory neurons in the ACC exhibited significant, temporally coordinated hyperactivity during acute seizures. Inhibition of CaMKII neurons significantly reduced seizure severity, whereas their activation induced spontaneous seizurelike activity. Enhancing GABAergic interneuron activity significantly decreased seizure frequency and severity. Notably, inhibition of GABAergic interneurons resulted in markedly more severe seizures. Further examination of interneuron subtypes revealed functional heterogeneity; activation of SST interneurons effectively suppressed seizures, whereas PV activation did not produce significant antiseizure effects. Critically, inhibition of either PV or SST interneurons triggered spontaneous seizurelike activity, indicating that both subtypes are necessary for maintaining network stability.

Significance

This study elucidates the differential roles of ACC neuronal populations in acute seizure dynamics. Activating GABAergic interneurons or inhibiting CaMKII‐positive neurons effectively suppresses seizure activity. Among interneuron subtypes, SST activation reduces seizures, whereas PV activation does not. However, inhibiting either subtype triggers spontaneous epileptiform activity, indicating both are indispensable and play complementary roles in maintaining network homeostasis. These findings reveal the microcircuit mechanisms by which the ACC modulates acute seizures, highlighting that cortical stability relies on both the classical excitation–inhibition balance and the cooperative interplay of multiple interneuron subtypes.

Keywords: CaMKII neurons, excitation–inhibition balance, GABAergic interneurons, somatostatin interneurons

Key points.

Inhibition of CaMKII neurons mitigated seizure severity, whereas their activation induced spontaneous seizures.
Seizures caused by interneuron blockade were more severe than those induced by excitatory neuron activation.
SST activation suppressed seizures more effectively than PV activation, whereas inhibition of either subtype induced seizures.

1. INTRODUCTION

Epilepsy is a chronic central nervous system (CNS) disorder arising from abnormal, excessive, and synchronized neuronal discharges, clinically manifesting as recurrent unprovoked seizures. ¹ Current epidemiological estimates indicate that approximately 70 million individuals worldwide are affected by epilepsy. ² The etiology of this condition is highly heterogeneous, encompassing genetic, structural, metabolic, infectious, and immune etiologies. ³ Although a variety of antiepileptic drugs are widely used in clinical practice, approximately one third of patients still manifest drug‐resistant epilepsy, ⁴ , ⁵ which accounts for substantial disability‐adjusted life‐year reduction and elevated premature mortality risks, thereby imposing a considerable global disease burden. ⁶

Normal CNS function is critically dependent upon the dynamic equilibrium between excitatory and inhibitory synaptic transmission. A shift in the excitation/inhibition (E/I) balance toward hyperexcitability represents a fundamental mechanism underlying epileptogenesis and seizure generalization. ⁷ Prior investigations have elucidated the cellular substrates governing this balance within discrete brain regions. For instance, cholinergic projections from the septum to the hippocampus can suppress epileptic activity through recruitment of somatostatin (SST)‐expressing interneurons, ⁸ and optogenetic activation of the thalamic reticular nucleus attenuates limbic seizures through modulation of midline thalamic output. ⁹ Therefore, elucidating the network substrates of E/I imbalance at the microcircuit level and identifying cell type‐specific targets capable of modulating this equilibrium may provide insights for developing antiseizure strategies.

However, the precise circuit mechanisms underlying E/I imbalance remain incompletely understood, particularly with regard to specific cortical regions and neuronal subtypes. Although hippocampal circuits have been extensively investigated, emerging evidence indicates that cortical structures play a crucial yet underappreciated role in seizure generation and propagation. The anterior cingulate cortex (ACC), functioning as a central hub of the limbic system, is extensively implicated in higher order brain functions including emotion regulation, cognitive control, and pain processing. ¹⁰ Importantly, ACC is widely connected with key epileptogenic regions, suggesting its role in seizure regulation. Clinical evidence further supports its involvement in the seizure network. Neuroimaging studies have revealed significant gray matter volume reduction in the ACC of patients with genetic generalized epilepsy. ¹¹ Furthermore, functional magnetic resonance imaging and intracranial electroencephalographic (EEG) data indicate that network activity within the ACC and medial prefrontal cortex is strongly correlated with impaired consciousness during seizures. ¹² At the molecular level, aberrant expression of glutamate ionotropic receptor kainate type subunit 1 in ACC projection neurons can induce epileptiform activity. ¹³ Although the ACC is implicated in epilepsy and its comorbidities, ¹⁴ the specific contribution of local E/I dynamics remains elusive. In particular, it is unknown how the interplay between excitatory and inhibitory neurons within these microcircuits ¹⁵ modulates seizures.

In this study, using a pentylenetetrazol (PTZ)‐induced acute seizure model in mice, we integrated behavioral analysis, in vivo fiber photometry calcium imaging, EEG, and chemogenetic and optogenetic approaches to examine the roles of ACC excitatory and inhibitory neurons in acute seizure regulation, and further dissected the contributions of γ‐aminobutyric acidergic (GABAergic) interneuron subtypes. This study aims to reveal the ACC microcircuit mechanism underlying acute seizure regulation and provide new evidence for understanding the role of cortical nodes in the seizure network.

2. MATERIALS AND METHODS

2.1. Animals

Male wild‐type C57BL/6J mice and transgenic mice, including CaMKII‐Cre (C57BL/6J‐Tg[Camk2a‐Cre]), Vgat‐Cre (C57BL/6J‐Slc32a1 ^em[Cre]), SST‐Cre (C57BL/6J‐Sst ^em[Cre]), parvalbumin (PV)‐Cre (C57BL/6J‐Pvalb ^em[Cre]), and glutamate decarboxylase 67 (Gad67)‐GFP (C57BL/6J‐Tg[Gad1‐EGFP]) mice (6–8 weeks old, 22–24 g) were purchased from Genepax Biotechnology Co. and housed at the Animal Center of Xuanwu Hospital, Capital Medical University. All transgenic lines were maintained as heterozygous breeding colonies. Animals were housed in a specific pathogen‐free facility under a 12‐h light/dark cycle (lights on from 8:00 to 20:00), with temperature at 22 ± 2°C and humidity at 40%–60%. All experimental procedures were approved by the animal ethics committee of Xuanwu Hospital (approval number: XW‐20230605‐1) and performed in accordance with the National Institutes of Health guidelines.

2.2. Acute seizure model and behavioral assessment

PTZ (Sigma‐Aldrich) was dissolved in .9% saline as a 50 mg/mL stock (stored at −20°C in the dark) and diluted to 2 mg/mL on the day of the experiment. Acute seizures were induced by intraperitoneal PTZ injection (60 mg/kg). Immediately after injection, mice were placed in an observation box, and behavior was recorded by a video camera for 30 min.

2.3. EEG recording and analysis

Recording electrode screws were fixed to the skull above the ACC, and reference and ground electrodes were implanted over the cerebellum. EEG signals were acquired at 2000 Hz using a multichannel acquisition system (Neuracle), with simultaneous video recording of behavior. Raw signals were band‐pass filtered (1–100 Hz) and notch filtered at 50 Hz. Power spectral density was calculated using Welch's method, with frequency bands as follows: delta (1–4 Hz), theta (4–8 Hz), alpha (8–12 Hz), beta (12–30 Hz), and gamma (30–100 Hz). During behavioral scoring and EEG analysis, the experimenter was blinded to the experimental conditions.

2.4. Stereotaxic surgery

Under pentobarbital sodium anesthesia (50 mg/kg ip), mice were head‐fixed in a stereotaxic instrument (RWD Life Science), and body temperature was maintained at 37–38°C using a heating pad. A glass micropipette was used to inject 180 nL of virus into the ACC (coordinates relative to bregma: anteroposterior +.8 mm, mediolateral ±.3 mm, dorsoventral [DV] −1.75 mm) at a rate of 35 nL/min, with the pipette left in place for 10 min postinjection. For optogenetics and fiber photometry experiments, a ceramic ferrule optical fiber cannula (diameter = 200 μm, numerical aperture = .37) was implanted with the tip at DV −1.55 mm; for miniscope imaging experiments, a gradient‐index (GRIN) lens was implanted. All implants were secured with dental cement. Mice were allowed to recover for 3–4 weeks to ensure sufficient viral expression. Adeno‐associated virus (AAV) vectors were obtained from BrainCase with titers ≥ 1 × 10¹² vg/mL, including the following: rAAV2/9‐hSyn‐DIO‐GCaMP6s, rAAV2/9‐hSyn‐DIO‐hM3D(Gq)‐mCherry, rAAV2/9‐hSyn‐DIO‐hM4D(Gi)‐mCherry, rAAV2/9‐hSyn‐DIO‐mCherry, rAAV2/9‐EF1α‐DIO‐eNpHR3.0‐mCherry, rAAV2/9‐EF1α‐DIO‐ChR2(H134R)‐mCherry, and rAAV‐EF1α‐DIO‐taCasp3‐T2A‐TEVp‐P2A‐EGFP.

2.5. In vivo fiber photometry

Population calcium signals were acquired with a fiber photometry system (Thinkertech). A 470‐nm light‐emitting diode (LED) was used to excite GCaMP6s. The two excitation lights were delivered interleaved at 40 Hz, and fluorescence signals were detected by a photomultiplier tube at 40 Hz. Data were processed with MATLAB. The fluorescence change was calculated as ΔF/F₀ = (F − F₀)/F₀ × 100%. Metrics included mean ΔF/F₀ and the area under the curve (AUC) during seizures.

2.6. In vivo miniscope calcium imaging

A head‐mounted microscope system (nVista HD, Inscopix) was used to monitor single‐cell activity of ACC neurons. Two to three weeks after viral expression, a GRIN lens (1.0 mm) was implanted and secured with dental cement. After 2–3 weeks of recovery and fluorescence stabilization, a baseplate was attached. The internal 475‐nm LED served as the excitation source, with a sampling rate of 20 Hz. Image data were preprocessed with Inscopix Data Processing Software, downsampled to 10 Hz, and neurons were extracted with the principal component analysis–independent component analysis algorithm.

2.7. Chemogenetic manipulation

Clozapine‐N‐oxide (CNO; Tocris) was dissolved in .9% saline to prepare a .5‐mg/mL working solution. Mice received an intraperitoneal injection of CNO (5 mg/kg) 1 h prior to PTZ injection.

2.8. Optogenetic manipulation

Laser stimulation was generated by a laser source (Newdoon) and delivered via a patch cord connected to the implanted fiber cannula. Laser stimulation commenced 5 min prior to PTZ injection and continued until the end of the observation period. For inhibitory manipulation (eNpHR3.0), 594 nm yellow light (10 mW at fiber tip, continuous mode) was used. For excitatory manipulation Channelrhodopsin‐2 (ChR2), 473 nm blue light (8 mW at fiber tip, pulsed mode, 20 Hz, 5‐ms pulse width) was used. Control mice received light stimulation with identical parameters.

2.9. Immunofluorescence staining

Sixty minutes after PTZ injection, mice were anesthetized and transcardially perfused with saline followed by 4% paraformaldehyde. Brains were harvested and postfixed at 4°C overnight. Coronal sections (30 μm) were obtained with a cryostat. Sections were permeabilized with .3% Triton X‐100, blocked with 3% donkey serum, and incubated with primary antibodies at 4°C overnight. Primary antibodies were as follows: guinea pig anti‐c‐Fos (1:1000; Synaptic Systems), rabbit anti‐c‐Fos (1:500; Abcam), rabbit anti‐NeuN (1:500; Abcam), rabbit anti‐CaMKII (1:200; Abcam), rabbit anti‐Gad67 (1:200; Invitrogen), guinea pig anti‐PV (1:200; Synaptic Systems), and mouse anti‐SST (1:50; GeneTex). Sections were incubated with Alexa Fluor‐conjugated secondary antibodies (1:500; Invitrogen) for 2 h at room temperature and mounted with 4,6‐diamidino‐2‐phenylindole mounting medium. Fluorescence images were acquired with a confocal microscope (Leica).

2.10. Statistical analysis

All statistical analyses and plotting were performed using GraphPad Prism 9.0. Data are presented as mean ± SEM. The Shapiro–Wilk test and F‐test were used to evaluate normality and homogeneity of variance. The two groups were compared using the unpaired t‐test, Welch t‐test, or Mann–Whitney U‐test. Two‐way analysis of variance was used for multigroup comparisons, followed by Bonferroni or Tukey post hoc tests; p < .05 was considered statistically significant. Given the highly exploratory nature of this study, no formal a priori power analysis was conducted prior to the experiments. The sample size was determined based on comparable studies in the field, effect sizes observed in pilot experiments, and principles of animal ethics.

3. RESULTS

3.1. PTZ induces robust behavioral and electrophysiological seizures in mice

We evaluated PTZ‐induced acute seizures using behavioral scoring and EEG recordings in freely moving mice (Figure 1A). Following intraperitoneal injection of PTZ, mice exhibited characteristic seizure behaviors as assessed using the modified Racine scale (Figure 1B). Behaviors progressed from facial twitching to limb clonus to generalized tonic–clonic seizures; stages I–III were classified as focal and IV–VI as generalized seizures (GSs). The PTZ group exhibited marked cortical hypersynchronous discharges characterized by repetitive high‐amplitude spike–wave complexes and polyspike bursts (Figure 1C). Power spectral density analysis showed that absolute power across the δ, θ, α, β, and γ frequency bands was significantly elevated in the PTZ group compared to controls (Figure 1D). Behavioral assessment confirmed significantly elevated seizure stages in the PTZ group (Figure 1E), with every PTZ‐treated mouse experiencing at least one GS episode (Figure 1F). Following PTZ injection, mice exhibited EEG seizure onset at 2.62 ± .40 min and generalized seizures at 3.79 ± .65 min (Figure 1G,H). Together, these data confirm that PTZ effectively triggers cortical hypersynchronous discharges and seizure behaviors in mice.

Pentylenetetrazol (PTZ) induces robust behavioral and electrophysiological seizures in mice. (A) Schematic of the experimental timeline. Mice were implanted with electroencephalographic (EEG) recording electrodes targeting the anterior cingulate cortex. Following a 7‐day recovery period, animals received a single intraperitoneal injection of PTZ (60 mg/kg) or an equivalent volume of saline, and EEG signals were continuously recorded for 30 min. (B) Schematic illustration of the modified Racine scale used to evaluate behavioral seizure severity. Stage 1, freezing and facial twitching; Stage 2, head nodding; Stage 3, unilateral forelimb clonus; Stage 4, rearing with bilateral forelimb clonus; Stage 5, generalized tonic–clonic seizure with rearing and falling; Stage 6, death. (C) Representative EEG spectrograms and corresponding raw EEG traces recorded from saline and PTZ mice. The boxed region in the PTZ spectrogram is magnified below to illustrate the seizure discharge pattern. Color scale indicates power spectral density (μV²/Hz). Scale bars: 100 μV (vertical), 1 s (horizontal). (D) Absolute EEG power across five frequency bands in saline and PTZ mice. PTZ administration significantly increased spectral power in the δ band (saline, n = 6; PTZ, n = 6; two‐way repeated measures analysis of variance [RMANOVA] followed by Bonferroni post hoc test, p = .0422), θ band (two‐way RMANOVA followed by Bonferroni post hoc test, p = .013), α band (two‐way RMANOVA followed by Bonferroni post hoc test, p < .0001), β band (two‐way RMANOVA followed by Bonferroni post hoc test, p < .0001), and γ band (two‐way RMANOVA followed by Bonferroni post hoc test, p = .0149). (E) Maximum behavioral seizure stage attained during the 30‐min observation window, scored according to the modified Racine scale (saline, n = 5; PTZ, n = 5; Mann–Whitney U‐test, p = .0079). (F) Total number of generalized seizures (GSs) recorded during the observation period (saline, n = 5; PTZ, n = 5; Mann–Whitney U‐test, p = .0079). (G) Latency to the first electrographic seizure onset. Animals in the saline group that did not display any electrographic seizure activity were assigned the maximum observation duration of 30 min (saline, n = 5; PTZ, n = 5; Mann–Whitney U‐test, p = .0079). (H) Latency to the first GS. Saline‐treated animals without GS were assigned the maximum observation duration of 30 min (saline, n = 5; PTZ, n = 5; Mann–Whitney U‐test, p = .0079). Data are presented as mean ± SEM, with individual data points representing single animals. *p < .05, **p < .01, ****p < .0001.

3.2. Widespread activation of excitatory and inhibitory neurons in the ACC during seizures

Immunofluorescence revealed a significant increase in c‐Fos‐positive cells in the ACC following PTZ‐induced seizures compared with saline controls (Figure 2A,B). The colocalization rate of c‐Fos with calcium/calmodulin‐dependent protein kinase II (CaMKII) neurons was significantly higher in the PTZ group than in controls (Figure 2C,D). Inhibitory neurons were also significantly activated, as shown by a markedly higher colocalization rate of c‐Fos with Gad67 in the PTZ group (Figure 2E,F). PV and SST neurons comprised 43.8% and 38.3% of Gad67‐positive cells, respectively (Figure S1A,B), and both subtypes showed significantly elevated c‐Fos induction after PTZ injection (Figure S2A–D). These findings indicate that PTZ‐induced seizures trigger widespread neuronal activation in the ACC.

Pentylenetetrazol (PTZ)‐induced seizures activate both excitatory neurons and γ‐aminobutyric acidergic (GABAergic) interneurons in the anterior cingulate cortex (ACC). (A) Representative immunofluorescence images showing c‐Fos expression (red) in the ACC of saline (upper row) and PTZ (lower row) mice. The boxed regions in the left panels are magnified in the subsequent panels. Scale bars: 1 mm (left panels), 100 μm (magnified panels). (B) Quantification of the total number of c‐Fos‐positive cells in the ACC. PTZ treatment significantly increased c‐Fos expression compared with saline controls (saline, n = 6; PTZ, n = 6; unpaired t‐test, p < .0001). (C) Representative immunofluorescence images showing colocalization of c‐Fos (red) with calcium/calmodulin‐dependent protein kinase II (CaMKII; green), a marker of excitatory neurons, in the ACC of saline (upper row) and PTZ (lower row) mice. Scale bars: 1 mm (left panels), 100 μm (magnified panels). (D) Quantification of the percentage of CaMKII‐positive neurons coexpressing c‐Fos. PTZ treatment significantly increased the proportion of c‐Fos‐positive excitatory neurons compared with saline controls (saline, n = 6; PTZ, n = 6; unpaired t‐test, p < .0001). (E) Representative immunofluorescence images showing colocalization of c‐Fos (red) with glutamate decarboxylase 67 (Gad67; green), a marker of GABAergic interneurons, in the ACC of saline (upper row) and PTZ (lower row) mice. Scale bars: 1 mm (left panels), 100 μm (magnified panels). (F) Quantification of the percentage of Gad67‐positive neurons coexpressing c‐Fos. PTZ treatment significantly increased the proportion of c‐Fos‐positive GABAergic interneurons compared with saline (saline, n = 6; PTZ, n = 6; unpaired t‐test, p < .0001). Data are presented as mean ± SEM, ****p < 0001. DAPI, 4,6‐diamidino‐2‐phenylindole.

3.3. ACC CaMKII neurons bidirectionally regulate seizurelike activity

To track the population dynamics of ACC excitatory neurons during seizures, we used fiber photometry to record calcium signals from CaMKII neurons (Figure 3A). Coimmunolabeling confirmed that 97.14% ± .85% of enhanced green fluorescent protein (EGFP)‐expressing cells were CaMKII‐positive, and 97.16% ± 1.48% of CaMKII‐positive cells expressed EGFP (Figure 3B). Heatmap visualization revealed notably enhanced calcium transients during PTZ‐induced seizures (Figure 3C). Signals surged within 0–.5 s of onset then dropped below baseline by .5–1 s (Figure 3D,E). We performed in vivo calcium imaging using head‐mounted miniaturized microscopes in freely behaving mice (Figure 3F). Whereas most neurons were relatively quiescent at baseline, PTZ‐induced seizures elicited synchronized, high‐amplitude calcium responses across the imaging field (Figure 3G). Representative single‐neuron traces confirmed substantial increases in both frequency and amplitude of calcium transients, with multiple neurons displaying near‐synchronous activity within the seizure window (Figure 3H). To quantify this observed synchronization, we performed pairwise correlation analysis across all recorded neurons. The synchrony index, calculated as the mean pairwise Pearson correlation coefficient, revealed a significant elevation during the ictal period relative to baseline (Figure S3A). Temporal analysis showed that synchrony peaked during the tonic–clonic phase and gradually declined following seizure termination, paralleling the behavioral and electrographic seizure progression (Figure S3B).

Excitatory neurons in the anterior cingulate cortex (ACC) are activated during pentylenetetrazol (PTZ)‐induced seizures, and bidirectional chemogenetic manipulation of these neurons modulates seizure severity. (A) Schematic of the fiber photometry recording system. DM, dichroic mirror; ND, neutral density filter. (B) Representative fluorescence images showing the optical fiber placement in the ACC (left) and coexpression of enhanced green fluorescent protein (EGFP; green) with calcium/calmodulin‐dependent protein kinase II (CaMKII; red) in ACC neurons. Scale bars: 1 mm (left), 10 μm (right). DAPI, 4,6‐diamidino‐2‐phenylindole. (C) Heatmaps of calcium transients (ΔF/F %) from individual excitatory neurons aligned to seizure onset in saline (left) and PTZ (right) mice. (D) Average calcium signal traces (ΔF/F %) of ACC excitatory neurons in saline and PTZ mice. (E) Quantification of the area under the curve of calcium signals across four time windows. PTZ induced significantly increased calcium activity during the 0–.5‐s window (saline, n = 136 neurons from n = 3 mice; PTZ, n = 238 neurons from n = 3 mice; Mann–Whitney U‐test, p < .0001) and the 1–1.5‐s window (Mann–Whitney U‐test, p < .0001), but not during the −.5 to 0‐s window (Mann–Whitney U‐test, p = .1734) or the .5–1 s window (Mann–Whitney U‐test, p = .0802). (F) Representative field of view from miniature microscope calcium imaging of ACC excitatory neurons expressing GCaMP. (G) Heatmap of calcium activity from simultaneously recorded ACC excitatory neurons during PTZ‐induced seizures. Neurons displayed synchronous calcium transients time‐locked to seizure events. (H) Representative raw calcium traces from individual neurons. Scale bars: 25 s (horizontal); Z‐score scale as indicated. (I) Experimental timelines for chemogenetic inhibition (upper) and activation (lower) experiments. CNO, clozapine‐N‐oxide; EEG, electroencephalographic. (J) Representative fluorescence images confirming viral expression (mCherry, red) in CaMKII‐positive (green) neurons in the ACC. Scale bars: 1 mm (left), 100 μm (right). (K) Schematic of the chemogenetic inhibition experiment. CaMKII‐Cre mice expressing human M4 muscarinic DREADD receptor [Gi‐coupled] (hM4D[Gi]) or mCherry control virus in the ACC received CNO injection followed by PTZ administration. (L) Representative EEG spectrograms and corresponding raw EEG traces from mCherry and hM4D(Gi) mice after PTZ injection. Color scale indicates power spectral density (μV²/Hz). Scale bars: 100 μV, 1 s. (M) Maximum behavioral seizure stage (mCherry, n = 6; hM4D[Gi], n = 8; unpaired t‐test, p = .0377). (N) Total number of generalized seizures (GSs; mCherry, n = 6; hM4D[Gi], n = 8; Mann–Whitney U‐test, p = .0013). (O) Latency to the first electrographic seizure onset (min; mCherry, n = 6; hM4D[Gi], n = 8; unpaired t‐test with Welch correction, p = .0057). (P) Latency to the first GS (mCherry, n = 6; hM4D[Gi], n = 8; Mann–Whitney U‐test, p = .0007). (Q) Schematic of the chemogenetic activation experiment. CaMKII‐Cre mice expressing human M3 muscarinic DREADD receptor [Gq‐coupled] (hM3D[Gq]) or mCherry control virus in the ACC received CNO injection without PTZ. (R) Representative EEG spectrograms and corresponding raw EEG traces from mCherry control (upper) and hM3D(Gq) (lower) mice after CNO injection alone. The boxed region is magnified below. Scale bars: 100 μV, 20 s. (S) Maximum behavioral seizure stage (mCherry, n = 7; hM3D[Gq], n = 8; Mann–Whitney U‐test, p = .0002). (T) Total number of GSs (mCherry, n = 7; hM3D[Gq], n = 7; Mann–Whitney U‐test, p = .0006). (U) Latency to the first electrographic seizure onset. mCherry control mice that did not exhibit electrographic seizures were assigned the maximum observation duration of 60 min (mCherry, n = 7; hM3D[Gq], n = 8; Mann–Whitney U‐test, p = .0003). (V) Latency to the first GS. mCherry control mice without GS were assigned the maximum observation duration of 60 min (mCherry, n = 7; hM3D[Gq], n = 8; Mann–Whitney U‐test, p = .0003). Data are presented as mean ± SEM, with individual data points representing single animals unless otherwise stated. *p < .05, **p < .01, ***p < .001, ****p < .0001.

To test causality, we bilaterally injected Cre‐dependent hM4D(Gi) vectors into the ACC of CaMKII‐Cre mice (Figure 3I–K). Coimmunolabeling confirmed that 96.70% ± 1.05% of mCherry‐expressing cells were CaMKII‐positive, and 96.04% ± 2.18% of CaMKII‐positive cells expressed mCherry (Figure 3J). Time–frequency EEG analysis showed that, whereas mCherry controls displayed high‐amplitude spike–wave complexes and broadband power increases after PTZ injection, chemogenetic inhibition markedly attenuated these pathological bursts (Figure 3L). Behavioral analysis confirmed that the hM4D(Gi) group showed reduced seizure stages (Figure 3M), fewer GS episodes (Figure 3N), and prolonged latencies to EEG seizure onset and GS (Figure 3O,P). Optogenetic inhibition via eNpHR3.0 produced similar antiseizure effects (Figure S3C–H).

Conversely, to test whether activation of ACC CaMKII neurons is sufficient to induce seizurelike activity in the absence of PTZ, we expressed hM3D(Gq) in this population (Figure 3Q). EEG recordings showed that CNO administration triggered paroxysmal spike discharges, and time–frequency analysis revealed intermittent high‐frequency power elevations coinciding with clonic behaviors (Figure 3R). Compared with mCherry controls, hM3D(Gq)‐expressing mice exhibited spontaneous seizures of varying severity (Figure 3S), more frequent GSs (Figure 3T), and shorter latencies to EEG seizure onset and GS (Figure 3U,V). This proepileptic effect was further validated by combined CNO and PTZ administration (Figure S3I–N), which significantly worsened seizure severity. Selective ablation of ACC CaMKII neurons using taCasp3 (Figure S4A–C) also reduced seizure stages (Figure S4D), decreased GS frequency (Figure S4E), and prolonged seizure latencies (Figure S4F,G). Collectively, these findings demonstrate that ACC CaMKII neurons bidirectionally modulate seizure susceptibility and severity.

3.4. ACC GABAergic neurons bidirectionally regulate seizurelike activity

We monitored ACC GABAergic neuron activity during seizures using fiber photometry in Vgat‐Cre mice (Figure 4A). Coimmunolabeling confirmed that 89.68% ± 5.20% of EGFP‐expressing cells were Vgat‐positive, and 87.26% ± 4.36% of Vgat‐positive cells expressed EGFP (Figure 4B). Recordings revealed significantly elevated calcium signals during PTZ‐induced seizures (Figure 4C,D), with a marked increase in AUC during the ictal phase (Figure 4E). To examine whether activation of ACC GABAergic neurons is sufficient to suppress PTZ‐induced acute seizure activity, we bilaterally injected Cre‐dependent hM3D(Gq) vectors into the ACC of Vgat‐Cre mice (Figure 4F–H). Coimmunolabeling confirmed that 81.40% ± 6.89% of mCherry‐expressing cells were Vgat‐positive, and 91.85% ± 4.12% of Vgat‐positive cells expressed mCherry. Following CNO activation, the hM3D(Gq) group showed significantly reduced PTZ‐induced discharge amplitudes and suppressed power across all frequency bands compared to controls (Figure 4I). Behavioral quantification indicated that GABAergic neuron activation significantly reduced both maximum seizure stage and GS frequency (Figure 4J,K), while substantially prolonging latencies to EEG seizure onset and GS (Figure 4L,M). Real‐time optogenetic activation using Channelrhodopsin‐2 (ChR2) during PTZ challenge produced consistent antiseizure effects (Figure S5A,B), including reduced seizure stages (Figure S5C), fewer GS episodes (Figure S5D), and prolonged latencies to EEG seizure onset (Figure S5E) and GS (Figure S5F).

γ‐Aminobutyric acidergic (GABAergic) interneurons in the anterior cingulate cortex (ACC) are recruited during pentylenetetrazol (PTZ)‐induced seizures, and bidirectional chemogenetic manipulation of these neurons modulates seizure severity. (A) Schematic of the fiber photometry recording system. DM, dichroic mirror; ND, neutral density filter. (B) Representative fluorescence images showing the optical fiber placement in the ACC (left) and coexpression of enhanced green fluorescent protein (EGFP; green) with glutamate decarboxylase 67 (Gad67; red) in ACC interneurons. Scale bars: 1 mm (left), 10 μm (right). DAPI, 4,6‐diamidino‐2‐phenylindole. (C) Heatmaps of calcium transients (ΔF/F %) from individual GABAergic interneurons aligned to seizure onset in saline (left) and PTZ (right) mice. (D) Average calcium signal traces (ΔF/F %) of ACC GABAergic interneurons in saline and PTZ mice. (E) Quantification of the area under the curve of calcium signals. PTZ treatment produced a significant increase in calcium activity compared with saline controls (saline, n = 219 neurons from n = 3 mice; PTZ, n = 239 neurons from n = 3 mice; Mann–Whitney U‐test, p < .0001). (F) Experimental timelines for chemogenetic activation (upper) and inhibition (lower) experiments. Vgat‐Cre mice received bilateral injection of rAAV‐DIO‐hM3D(Gq)‐mCherry or rAAV‐DIO‐hM4D(Gi)‐mCherry into the ACC, followed by electrode implantation after 21 days. CNO, clozapine‐N‐oxide; EEG, electroencephalographic. (G) Representative fluorescence images confirming viral expression (red) in Gad67‐positive (green) interneurons in the ACC. Scale bars: 1 mm (left), 10 μm (right). (H) Schematic of the chemogenetic activation experiment. Vgat‐Cre mice expressing hM3D(Gq) or mCherry control virus in the ACC received CNO injection followed by PTZ administration. (I) Representative EEG spectrograms and corresponding raw EEG traces from mCherry control (upper) and hM3D(Gq) (lower) mice after PTZ injection. Color scale indicates power spectral density (μV²/Hz). Scale bars: 100 μV, 1 s. (J) Maximum behavioral seizure stage (mCherry, n = 6; hM3D[Gq], n = 7; unpaired t‐test with Welch correction, p = .026). (K) Total number of generalized seizures (GSs; mCherry, n = 6; hM3D[Gq], n = 6; Mann–Whitney U‐test, p = .0065). (L) Latency to the first electrographic seizure onset (mCherry, n = 6; hM3D[Gq], n = 6; Mann–Whitney U‐test, p = .026). (M) Latency to the first GS (mCherry, n = 6; hM3D[Gq], n = 6; Mann–Whitney U‐test, p = .0022). (N) Schematic of the chemogenetic inhibition experiment. (O) Representative EEG spectrograms and corresponding raw EEG traces from mCherry control (upper) and hM4D(Gi) (lower) mice after CNO injection alone. Scale bars: 100 μV, 20 s. (P) Maximum behavioral seizure stage. mCherry control mice did not exhibit seizure behavior and were scored as stage 0 (mCherry, n = 6; hM4D[Gi], n = 6; Mann–Whitney U‐test, p = .0022). (Q) Total number of GSs (mCherry, n = 6; hM4D[Gi], n = 6; Mann–Whitney U‐test, p = .0152). (R) Latency to the first electrographic seizure onset. mCherry control mice that did not exhibit electrographic seizures were assigned the maximum observation duration of 60 min (mCherry, n = 6; hM4D[Gi], n = 6; Mann–Whitney U‐test, p = .0022). (S) Latency to the first GS. mCherry control mice without GS were assigned the maximum observation duration of 60 min (mCherry, n = 6; hM4D[Gi], n = 6; Mann–Whitney U‐test, p = .0152). Data are presented as mean ± SEM, with individual data points representing single animals unless otherwise stated. *p < .05, **p < .01, ****p < .0001.

To investigate the impact of silencing GABAergic neurons on seizure susceptibility, we selectively inhibited ACC Vgat neurons using hM4D(Gi) (Figure 4N). Following CNO administration alone, the EEG rapidly shifted into continuous high‐voltage tonic discharges, with spectrograms showing high‐intensity power bursts across all bands (Figure 4O). Compared to mCherry controls, hM4D(Gi)‐expressing mice exhibited spontaneous seizures (Figure 4P), increased GS frequency (Figure 4Q), and markedly reduced latencies to EEG seizure onset and GS (Figure 4R,S). This proepileptic effect was further validated under combined CNO and PTZ challenge (Figure S5G,H). Inhibiting GABAergic neurons significantly exacerbated seizure severity, with elevated maximum seizure stages (Figure S5I) and reduced latencies to EEG seizure onset (Figure S5K) and GS (Figure S5L), although total GS number was not significantly different (Figure S5J). These findings demonstrate that ACC GABAergic neurons exert bidirectional modulatory effects on acute seizure susceptibility and severity.

3.5. Activation of SST interneurons, but not PV interneurons, effectively attenuates seizures

Given the antiseizure effects of global GABAergic activation, we next investigated the specific contributions of PV and SST interneurons to acute seizure regulation within the ACC (Figure 5A,B). We first expressed hM3D(Gq) bilaterally in the ACC of PV‐Cre mice. Coimmunolabeling confirmed that 81.39% ± .71% of mCherry‐expressing cells were PV‐positive, and 84.17% ± 4.64% of PV‐positive cells expressed mCherry (Figure 5C). Following CNO activation, the hM3D(Gq) group still exhibited high‐amplitude continuous spike–wave discharges and power enhancement, comparable to mCherry controls, despite a trend toward attenuation (Figure 5E). Although PV activation showed a numerical trend toward seizure suppression, differences in maximum seizure stage, GS frequency, and seizure onset latencies were not statistically significant (Figure 5G–J).

Differential roles of parvalbumin (PV) and somatostatin (SST) interneurons in the anterior cingulate cortex (ACC) in regulating pentylenetetrazol (PTZ)‐induced seizures. (A) Experimental timeline. Cre‐dependent viral vectors were bilaterally injected into the ACC of PV‐Cre or SST‐Cre mice, followed by electrode implantation after 21 days. CNO, clozapine‐N‐oxide; EEG, electroencephalographic. (B) Schematic of the chemogenetic manipulation experiment in Cre mice. (C) Representative fluorescence images confirming viral expression (mCherry, red) in PV‐positive (green) interneurons in the ACC. Scale bars: 1 mm (left), 10 μm (right). DAPI, 4,6‐diamidino‐2‐phenylindole. (D) Representative fluorescence images confirming viral expression (mCherry, red) in SST‐positive (green) interneurons in the ACC. Scale bars: 1 mm (left), 10 μm (right). (E) Representative EEG spectrograms and corresponding raw EEG traces from PV‐Cre mice expressing mCherry control (upper) or hM3D(Gq) (lower) after CNO and PTZ administration. Color scale indicates power spectral density (μV²/Hz). Scale bars: 100 μV, 1 s. (F) Representative EEG spectrograms and corresponding raw EEG traces from SST‐Cre mice expressing mCherry control (upper) or hM3D(Gq) (lower) after CNO and PTZ administration. Scale bars: 100 μV, 1 s. (G) Maximum behavioral seizure stage in PV‐Cre mice (mCherry, n = 6; hM3D[Gq], n = 8; Mann–Whitney U‐test, p = .0636). (H) Total number of generalized seizures (GSs) in PV‐Cre mice (mCherry, n = 6; hM3D[Gq], n = 8; Mann–Whitney U‐test, p = .7802). (I) Latency to the first electrographic seizure onset in PV‐Cre mice (mCherry, n = 6; hM3D[Gq], n = 8; unpaired t‐test, p = .05). (J) Latency to the first GS in PV‐Cre mice (mCherry, n = 6; hM3D[Gq], n = 8; Mann–Whitney U‐test, p = .0813). (K) Maximum behavioral seizure stage in SST‐Cre mice (mCherry, n = 6; hM3D[Gq], n = 6; unpaired t‐test, p = .0145). (L) Total number of GSs in SST‐Cre mice (mCherry, n = 6; hM3D[Gq], n = 6; Mann–Whitney U‐test, p = .0108). (M) Latency to the first electrographic seizure onset (min) in SST‐Cre mice (mCherry, n = 6; hM3D[Gq], n = 6; Mann–Whitney U‐test, p = .0022). (N) Latency to the first GS (min) in SST‐Cre mice (mCherry, n = 6; hM3D[Gq], n = 6; Mann–Whitney U‐test, p = .0022). Data are presented as mean ± SEM, with individual data points representing single animals. *p < .05, **p < .01.

When we applied the same strategy to activate SST interneurons in SST‐Cre mice, coimmunolabeling confirmed that 82.58% ± 4.61% of mCherry‐expressing cells were SST‐positive (specificity), and 82.91 ± 3.81% of SST‐positive cells expressed mCherry (selectivity; Figure 5D). In contrast to the PV activation results, the SST‐hM3D(Gq) group showed reduced amplitudes, with spectrograms revealing attenuation of high‐frequency power (Figure 5F). Activating SST interneurons reduced maximum seizure stage (Figure 5K) and GS frequency (Figure 5L), while prolonging EEG seizure onset (Figure 5M) and GS latency (Figure 5N). Activation of SST interneurons is sufficient to suppress hyperexcitability; however, under PTZ‐induced disinhibition, activation of PV interneurons failed to produce comparable antiseizure effects.

3.6. Basal activity of both SST and PV neurons is essential for maintaining ACC network homeostasis

To investigate the roles of these subtypes, we performed selective chemogenetic inhibition by injecting Cre‐dependent hM4D(Gi) into the ACC of PV‐Cre and SST‐Cre mice (Figure 6A,B). Under CNO conditions, inhibiting either PV or SST neurons induced paroxysmal spike–wave and epileptiform discharges with high‐power fluctuations (Figure 6C,D). Compared to controls, PV and SST inhibition each significantly elevated seizure severity and reduced EEG seizure onset latency (Figure 6E–H). When combined with PTZ stimulation (Figure S6A,B), inhibition of neither PV nor SST interneurons significantly changed the maximum seizure grade or total GS frequency (Figure S6C,D,G,H). However, both procedures significantly reduced the time to onset of EEG seizures and the latency of GS (Figure S6E–J). Thus, despite differing effects upon activation, both interneuron populations are critically required to prevent network hyperexcitability and coregulate cortical inhibitory tone.

Chemogenetic inhibition of either parvalbumin (PV) or somatostatin (SST) interneurons in the anterior cingulate cortex (ACC) triggers spontaneous seizurelike activity. (A) Experimental timeline. Cre‐dependent viral vectors encoding hM4D(Gi) or mCherry control were bilaterally injected into the ACC of PV‐Cre or SST‐Cre mice, followed by electrode implantation after 21 days. CNO, clozapine‐N‐oxide; EEG, electroencephalographic. (B) Schematic of the chemogenetic inhibition experiment in Cre mice. (C) Representative EEG spectrograms and corresponding raw EEG traces from PV‐Cre mice expressing mCherry control (upper) or hM4D(Gi) (lower) after CNO injection alone. Color scale indicates power spectral density (μV²/Hz). Scale bars: 100 μV, 20 s. (D) Representative EEG spectrograms and corresponding raw EEG traces from SST‐Cre mice expressing mCherry control (upper) or hM4D(Gi) (lower) after CNO injection alone. Scale bars: 100 μV, 20 s. (E) Maximum behavioral seizure stage in PV‐Cre mice (mCherry, n = 7; hM4D[Gi], n = 7; Mann–Whitney U‐test, p = .0453). (F) Latency to the first electrographic seizure onset in PV‐Cre mice (mCherry, n = 7; hM4D[Gi], n = 7; Mann–Whitney U‐test, p = .0403). (G) Maximum behavioral seizure stage in SST‐Cre mice (mCherry, n = 7; hM4D[Gi], n = 7; Mann–Whitney U‐test, p = .0333). (H) Latency to the first electrographic seizure onset in SST‐Cre mice (mCherry, n = 7; hM4D[Gi], n = 7; Mann–Whitney U‐test, p = .0395). Data are presented as mean ± SEM, with individual data points representing single animals. *p < .05.

4. DISCUSSION

Generalized seizures arise from aberrant reconfiguration across large‐scale brain networks. ¹ , ¹⁶ The anterior cingulate cortex is extensively connected with limbic structures, thalamic nuclei, and neocortical regions. ¹² , ¹⁷ Despite this anatomical prominence, the specific gating mechanisms by which the ACC participates in seizure propagation have remained poorly understood. Here, we employed multimodal recordings coupled with cell type‐specific manipulations to systematically dissect the contributions of distinct ACC neuronal populations to acute seizure activity in a PTZ‐induced mouse model. Our results demonstrate opposing functional roles for CaMKII neurons and GABAergic interneurons. Suppressing CaMKII neuronal activity or enhancing GABAergic transmission mitigated PTZ‐induced seizures, whereas the converse manipulations provoked epileptiform activity in otherwise naïve mice. Furthermore, we revealed functional heterogeneity within GABAergic subtypes; chemogenetic activation of SST‐positive interneurons produced a significant anticonvulsant effect, whereas activation of PV‐positive interneurons did not yield a significant effect; however, inhibition of either subtype exacerbated seizure severity. These data establish that the stability of the ACC network does not rely on a simple linear excitation–inhibition balance, but rather on an asymmetric, subtype‐specific regulatory mechanism in which SST and PV interneurons exert distinct yet complementary roles.

By specifically inhibiting CaMKII neurons, we significantly reduced PTZ‐induced acute seizures. This aligns with the classical hypothesis that attenuating major projection neuron output limits seizure propagation. ¹ , ¹⁸ Similarly, specific activation of Vgat‐positive GABAergic neurons resulted in marked suppression of seizure stage and duration. Enhanced inhibition raises the threshold for paroxysmal depolarizing shifts by increasing shunting inhibition and hyperpolarizing currents. ⁷ , ¹⁹ This bidirectional validation indicates that either directly weakening excitatory drive or enhancing inhibitory tone exerts antiseizure effects by remodeling the E/I setpoint of the ACC. It should be noted that, because PTZ induces seizures via systemic disinhibition rather than through focal initiation, this study demonstrates that ACC neuronal activity participates in modulating the expression of acute seizures but does not identify the ACC as a primary seizure onset zone.

Notably, we uncovered a marked asymmetry in this seizure activity triggered by different manipulations. GABAergic inhibition exhibited shorter latencies and more severe tonic phases compared to those induced by direct CaMKII neuronal activation. This disparity likely reflects differential engagement of homeostatic feedback circuits. Under baseline conditions, cortical networks rely heavily on negative feedback to maintain stability. ²⁰ When excitatory output is artificially elevated, local microcircuits rapidly mobilize compensatory inhibition to partially offset the excitatory perturbation. ²¹ By contrast, targeted ablation of GABAergic function dismantles the core feedback architecture and leaves networks profoundly disinhibited, allowing even physiological inputs to trigger runaway excitation. ⁷ , ²² This network‐level asymmetry is further supported biophysically by the single‐cell calcium dynamics we observed. Unlike CaMKII neurons, which exhibit a pronounced postseizure calcium undershoot (Figure 3D), GABAergic neurons do not show suppression below baseline following the seizure‐associated rise in calcium signals (Figure 4D). In excitatory neurons, the postseizure calcium trough may reflect the summation of multiple inhibitory processes, including activity‐dependent hyperpolarization mediated by small‐conductance (SK) and large‐conductance (BK) calcium‐activated potassium channels, ²³ hyperpolarizing currents generated by Na⁺/K⁺‐adenosine triphosphatase (ATPase) activation, ²⁴ firing suppression mediated by adenosine via A1 receptors, ²⁵ and GABAergic inhibition provided by persistently active interneurons in the late seizure phase. ²⁶ In contrast, GABAergic neurons recover to baseline levels more rapidly, possibly reflecting the enriched mitochondrial distribution characteristic of GABAergic interneurons, which provides enhanced capacity for both ATP‐dependent calcium extrusion and direct mitochondrial calcium sequestration. ²⁷

Regarding distinct interneuron subtypes, chemogenetic activation of SST neurons significantly suppressed seizures, whereas comparable activation of PV interneurons did not produce a significant antiseizure effect. However, in the absence of PTZ, inhibiting either PV or SST interneurons alone was sufficient to trigger spontaneous epileptiform activity. Together, these findings support a synergistic model in which PV and SST interneurons cooperatively regulate ACC network excitability through complementary inhibitory functions, consistent with their distinct computational roles of division versus subtraction in cortical circuits. ²⁸ Several factors may underlie the differential effects of SST and PV neuron activation. First, these interneuron subtypes differ in synaptic targeting; SST neurons primarily innervate distal dendrites of pyramidal neurons, whereas PV neurons mainly target the soma and axon initial segment. ²⁹ , ³⁰ Under the intense and spatially widespread excitatory drive induced by systemic PTZ, dendritic inhibition by SST neurons may more effectively limit excitatory input integration and thereby raise the threshold for spike generation. Second, SST neurons exhibit weak frequency adaptation, and their output synapses have short‐term facilitation properties, ³¹ whereas PV interneurons are susceptible to depolarization block under extreme pathological conditions. ³² , ³³ Furthermore, high loads of postsynaptic chloride influx can overload KCC2 cotransporters and cause a depolarizing shift in the GABA reversal potential. ³⁴ , ³⁵ Third, in networks strongly driven by PTZ, profound somatic saturation might mask PV neuron activation's antiseizure efficacy via a ceiling effect. Functionally, PV‐mediated perisomatic inhibition restricts initial pyramidal cells bursting, ³⁴ , ³⁶ whereas SST‐mediated dendritic inhibition limits downstream excitation spread. ³⁷ Thus, our activation results likely reflect differing baseline recruitment thresholds during acute global disinhibition, rather than proving SST neurons are uniquely superior in controlling seizures.

Our findings reinforce that cooperative interneuron networks are central to maintaining homeostasis, as inhibiting GABAergic cells drove more severe acute epileptiform activity than activating excitatory neurons. Although selectively activating SST neurons suppressed PTZ‐evoked seizures, both PV and SST populations are required to regulate cortical excitability. Consequently, modulating these subtypes provides rationale for exploring ACC circuits in broader neuromodulation research. ³⁸ , ³⁹

Several limitations remain. Crucially, the acute PTZ model relies on systemic disinhibition and does not mimic spontaneous focal seizures or chronic epileptogenesis. ⁴⁰ , ⁴¹ Additionally, although various ACC populations participate in modulating these acute events, our findings do not establish the region as a primary seizure initiation hub. Delineating the precise functional hierarchy of the ACC will require future chronic epilepsy models paired with simultaneous multiregion recordings.

5. CONCLUSIONS

Our study demonstrates that seizures can be suppressed by either activating GABAergic interneurons or inhibiting CaMKII neurons in the ACC. Although activating SST rather than PV interneurons sufficiently inhibits seizures, inhibiting either subtype triggers epileptiform discharges, suggesting both contribute to seizure control through complementary roles. These results reveal cell type‐specific contributions of ACC neurons to acute seizure regulation.

AUTHOR CONTRIBUTIONS

Ziqian Yan: Conceptualization; data curation; formal analysis; writing—original draft. Ting Tang: Formal analysis; methodology; data curation. Kaishan Wang: Formal analysis; methodology. Bin Fu: Formal analysis; methodology. Zesheng Li: Formal analysis; writing—original draft. Mingshan Liu: Data curation; formal analysis. Hongyi Huang: Data curation; formal analysis. Yanfeng Yang: Formal analysis. Qiang Zheng: Conceptualization; supervision. Yangmin Zheng: Supervision; writing—review and editing. Yongzhi Shan: Methodology; writing—review and editing. Yumin Luo: Conceptualization; supervision; formal analysis; writing—review and editing. Guoguang Zhao: Conceptualization; data curation; formal analysis; methodology; writing—review and editing.

FUNDING INFORMATION

The study was supported by Beijing Municipal Science & Technology Commission (grant No. Z221100007422016 to G.Z.), Beijing Hospitals Authority Clinical Medicine Development of Special Funding Support, Yangfan Project (grant No. ZLRK202319 to G.Z.), and the project Implantable Closed‐Loop Neural Decoding System Based on Brain‐Computer Interface Technology (grant No. Z241100009024058 to G Z).

CONFLICT OF INTEREST STATEMENT

None of the authors has any conflict of interest to disclose.

ETHICAL PUBLICATION STATEMENT

I confirm that I have read the journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

Supporting information

Figure S1. Characterization of γ‐aminobutyric acidergic (GABAergic) interneuron subtypes in the anterior cingulate cortex (ACC). (A) Representative immunofluorescence images showing the distribution of parvalbumin (PV; red, upper row) and somatostatin (SST; red, lower row) interneurons in the ACC. The boxed regions in the left panels are magnified in the subsequent panels to show colocalization with glutamate decarboxylase 67 (Gad67; green), a marker of GABAergic neurons. Scale bars: 1 mm (left panels), 100 μm (magnified panels). (B) Quantification of the relative proportions of PV‐positive (43.8%), SST‐positive (38.3%), and other GABAergic interneuron subtypes (17.9%) among total Gad67‐positive neurons in the ACC (n = 3 mice).

EPI-67-5531-s005.tif^{(23.4MB, tif)}

Figure S2. Pentylenetetrazol (PTZ)‐induced seizures increase c‐Fos expression in both parvalbumin (PV) and somatostatin (SST) interneurons in the anterior cingulate cortex (ACC). (A) Representative immunofluorescence images showing c‐Fos expression (red) in PV‐positive interneurons (green) in the ACC of saline (upper row) and PTZ (lower row) mice. The boxed regions in the left panels are magnified in the subsequent panels. Scale bars: 1 mm (left panels), 100 μm (magnified panels). (B) Quantification of the percentage of PV‐positive neurons coexpressing c‐Fos. (saline, n = 6; PTZ, n = 6; unpaired t‐test with Welch correction, p < .0001). (C) Representative immunofluorescence images showing c‐Fos expression (red) in SST‐positive interneurons (green) in the ACC of saline (upper row) and PTZ (lower row) mice. The boxed regions in the left panels are magnified in the subsequent panels. Scale bars: 1 mm (left panels), 100 μm (magnified panels). (D) Quantification of the percentage of SST‐positive neurons coexpressing c‐Fos (saline, n = 6; PTZ, n = 6; unpaired t‐test, p < .0001). Data are presented as mean ± SEM, with individual data points representing single animals. ****p < .0001.

EPI-67-5531-s008.tif^{(39.9MB, tif)}

Figure S3. Quantification of neuronal synchrony during pentylenetetrazol (PTZ)‐induced seizures and validation of chemogenetic and optogenetic manipulation of excitatory neurons in anterior cingulate cortex. (A) Pairwise Pearson correlation coefficients of calcium signals from simultaneously recorded excitatory neurons during baseline and seizure periods (baseline, n = 2080 neurons; seizure, n = 2080 neurons; Mann–Whitney U‐test, p = .0207). (B) Representative trace showing the synchrony index (mean pairwise correlation) over time. The synchrony index increased sharply during seizure episodes (red dashed line, seizure mean = .018) compared to baseline period (blue dashed line, baseline mean = .007). (C) Schematic of the chemogenetic manipulation experiment in calcium/calmodulin‐dependent protein kinase II (CaMKII)‐Cre mice. (D) Schematic illustration of the experimental design. (E) Maximum behavioral seizure stage in CaMKII‐Cre mice expressing mCherry control or hM3D(Gq) after clozapine‐N‐oxide injection followed by PTZ administration (mCherry, n = 6; hM3D[Gq], n = 6; Mann–Whitney U‐test, p = .0152). (F) Total number of generalized seizures (GSs; mCherry, n = 6; hM3D(Gq), n = 6; Mann–Whitney U‐test, p = .2965). (G) Latency to the first electrographic seizure onset (mCherry, n = 6; hM3D[Gq], n = 6; unpaired t‐test with Welch correction, p = .0026). (H) Latency to the first GS (mCherry, n = 6; hM3D[Gq], n = 6; Mann–Whitney U‐test, p = .026). (I) Schematic of the optogenetic inhibition experiment in CaMKII‐Cre mice. (J) Schematic illustration of the experimental design. (K) Maximum behavioral seizure stage in CaMKII‐Cre mice expressing mCherry control or enhanced Natronomonas pharaonis halorhodopsin (eNpHR) after optogenetic inhibition followed by PTZ administration (mCherry, n = 6; eNpHR, n = 6; Mann–Whitney U‐test, p = .0346). (L) Total number of GSs (mCherry, n = 6; eNpHR, n = 6; Mann–Whitney U‐test, p = .0087). (M) Latency to the first electrographic seizure onset (min; mCherry, n = 6; eNpHR, n = 6; unpaired t‐test with Welch correction, t = 4.805, df = 5.321, p = .0041). (N) Latency to the first GS (min; mCherry, n = 6; eNpHR, n = 6; Mann–Whitney U‐test, p = .0043). Data are presented as mean ± SEM, with individual data points representing single animals unless otherwise stated. *p < .05, **p < .01, ***p < .001.

EPI-67-5531-s001.tif^{(29.9MB, tif)}

Figure S4. Caspase‐3 (Casp3)‐mediated silencing of excitatory neurons in the anterior cingulate cortex (ACC) suppresses pentylenetetrazol (PTZ)‐induced seizures. (A) Schematic of the experimental timeline. (B) Schematic illustration of the experimental design. (C) Representative fluorescence images confirming viral expression in the ACC. (D) Maximum behavioral seizure stage in mice expressing enhanced green fluorescent protein (EGFP) control or Casp3 after PTZ administration (EGFP, n = 7; Casp3, n = 7; Mann–Whitney U‐test, p = .0006). (E) Total number of generalized seizures (GSs; EGFP, n = 7; Casp3, n = 7; Mann–Whitney U‐test, p = .0006). (F) Latency to the first electrographic seizure onset (EGFP, n = 7; Casp3, n = 7; Mann–Whitney U‐test, p = .0484). (G) Latency to the first GS (EGFP, n = 7; Casp3, n = 7; Mann–Whitney U‐test, p = .0006). Data are presented as mean ± SEM, with individual data points representing single animals. *p < .05, ***p < .001.

EPI-67-5531-s004.tif^{(11MB, tif)}

Figure S5. Additional validation of γ‐aminobutyric acidergic interneuron contribution to seizure regulation using optogenetic and chemogenetic approaches in the anterior cingulate cortex. (A) Schematic of the optogenetic activation experiment in Vgat‐Cre mice. (B) Schematic illustration of the experimental design. (C) Maximum behavioral seizure stage in Vgat‐Cre mice expressing mCherry control or Channelrhodopsin‐2 (ChR2) after optogenetic activation followed by pentylenetetrazol (PTZ) administration (mCherry, n = 7; ChR2, n = 6; Mann–Whitney U‐test, p = .0146). (D) Total number of generalized seizures (GSs; mCherry, n = 7; ChR2, n = 6; Mann–Whitney U‐test, p = .0204). (E) Latency to the first electrographic seizure onset (mCherry, n = 7; ChR2, n = 6; unpaired t‐test with Welch correction, p = .0458). (F) Latency to the first GS (mCherry, n = 7; ChR2, n = 6; unpaired t‐test with Welch correction, p = .0297). (G) Schematic of the chemogenetic inhibition experiment in Vgat‐Cre mice. (H) Schematic illustration of the experimental design. (I) Maximum behavioral seizure stage in Vgat‐Cre mice expressing mCherry control or hM4D(Gi) after clozapine‐N‐oxide injection followed by PTZ administration (mCherry, n = 6; hM4D[Gi], n = 6; Mann–Whitney U‐test, p = .0022). (J) Total number of GSs (mCherry, n = 6; hM4D[Gi], n = 6; Mann–Whitney U‐test, p > .9999). (K) Latency to the first electrographic seizure onset (mCherry, n = 6; hM4D[Gi], n = 6; unpaired t‐test, p = .0002). (L) Latency to the first GS (mCherry, n = 6; hM4D[Gi], n = 6; unpaired t‐test with Welch correction, p = .0028). Data are presented as mean ± SEM, with individual data points representing single animals. *p < .05, **p < .01, ***p < .001.

EPI-67-5531-s002.tif^{(9.9MB, tif)}

Figure S6. Additional validation of parvalbumin and somatostatin interneuron contribution to seizure regulation using chemogenetic inhibition in the anterior cingulate cortex. (A) Schematic of the experimental timeline. (B) Schematic illustration of the experimental design. (C) Maximum behavioral seizure stage in Vgat‐Cre mice expressing mCherry control or hM4D(Gi) after clozapine‐N‐oxide (CNO) injection without pentylenetetrazol (PTZ; mCherry, n = 6; hM4D[Gi], n = 7; Mann–Whitney U‐test, p = .3963). (D) Total number of generalized seizures (GSs; mCherry, n = 6; hM4D[Gi], n = 7; Mann–Whitney U‐test, p > .9999). (E) Latency to the first electrographic seizure onset (mCherry, n = 7; hM4D[Gi], n = 6; unpaired t‐test, p = .0054). (F) Latency to the first GS (mCherry, n = 7; hM4D[Gi], n = 6; unpaired t‐test, p = .02). (G) Maximum behavioral seizure stage in Vgat‐Cre mice expressing mCherry control or hM4D(Gi) after CNO injection followed by PTZ administration (mCherry, n = 6; hM4D[Gi], n = 6; Mann–Whitney U‐test, p = .1667). (H) Total number of GSs (mCherry, n = 6; hM4D[Gi], n = 6; unpaired t‐test, p = .5995). (I) Latency to the first electrographic seizure onset (mCherry, n = 6; hM4D[Gi], n = 6; Mann–Whitney U‐test, p = .0022). (J) Latency to the first GS (mCherry, n = 6; hM4D[Gi], n = 6; unpaired t‐test, p = .0344). Data are presented as mean ± SEM, with individual data points representing single animals. *p < .05, **p < .01.

EPI-67-5531-s003.tif^{(15.3MB, tif)}

Table S1.

EPI-67-5531-s007.xls^{(51KB, xls)}

Data S1.

EPI-67-5531-s006.docx^{(17.5KB, docx)}

ACKNOWLEDGMENTS

The authors have nothing to report.

DATA AVAILABILITY STATEMENT

The research data are available from the corresponding authors upon request.

REFERENCES

1. Fisher RS, Acevedo C, Arzimanoglou A, Bogacz A, Cross JH, Elger CE, et al. ILAE official report: a practical clinical definition of epilepsy. Epilepsia. 2014;55(4):475–482. [DOI] [PubMed] [Google Scholar]
2. Thijs RD, Surges R, O'Brien TJ, Sander JW. Epilepsy in adults. The Lancet. 2019;393(10172):689–701. [DOI] [PubMed] [Google Scholar]
3. Scheffer IE, Berkovic S, Capovilla G, Connolly MB, French J, Guilhoto L, et al. ILAE classification of the epilepsies: position paper of the ILAE Commission for Classification and Terminology. Epilepsia. 2017;58(4):512–521. [DOI] [PMC free article] [PubMed] [Google Scholar]
4. Kwan P, Brodie MJ. Early identification of refractory epilepsy. N Engl J Med. 2000;342(5):314–319. [DOI] [PubMed] [Google Scholar]
5. Chen Z, Brodie MJ, Liew D, Kwan P. Treatment outcomes in patients with newly diagnosed epilepsy treated with established and new antiepileptic drugs: a 30‐year longitudinal cohort study. JAMA Neurol. 2018;75(3):279–286. [DOI] [PMC free article] [PubMed] [Google Scholar]
6. Feigin VL, Vos T, Nair BS, Hay SI, Abate YH, Abd Al Magied AH, et al. Global, regional, and national burden of epilepsy, 1990–2021: A systematic analysis for the global burden of disease study 2021. Lancet Public Health. 2025;10(3):e203–e227. [DOI] [PMC free article] [PubMed] [Google Scholar]
7. Sohal VS, Rubenstein JLR. Excitation‐inhibition balance as a framework for investigating mechanisms in neuropsychiatric disorders. Mol Psychiatry. 2019;24(9):1248–1257. [DOI] [PMC free article] [PubMed] [Google Scholar]
8. Wang Y, Wang Y, Xu C, Wang S, Tan N, Chen C, et al. Direct septum‐hippocampus cholinergic circuit attenuates seizure through driving somatostatin inhibition. Biol Psychiatry. 2020;87(9):843–856. [DOI] [PubMed] [Google Scholar]
9. Wicker E, Forcelli PA. Optogenetic activation of the reticular nucleus of the thalamus attenuates limbic seizures via inhibition of the midline thalamus. Epilepsia. 2021;62(9):2283–2296. [DOI] [PMC free article] [PubMed] [Google Scholar]
10. Bush G, Luu P, Posner MI. Cognitive and emotional influences in anterior cingulate cortex. Trends Cogn Sci. 2000;4(6):215–222. [DOI] [PubMed] [Google Scholar]
11. Braga AM, Fujisao EK, Verdade RC, Paschoalato RP, Paschoalato RP, Yamashita S, et al. Investigation of the cingulate cortex in idiopathic generalized epilepsy. Epilepsia. 2015;56(11):1803–1811. [DOI] [PubMed] [Google Scholar]
12. Blumenfeld H. Impaired consciousness in epilepsy. The Lancet Neurology. 2012;11(9):814–826. [DOI] [PMC free article] [PubMed] [Google Scholar]
13. Li XH, Shi W, Zhao ZX, Matsuura T, Lu JS, Che J, et al. Increased GluK1 subunit receptors in Corticostriatal projection from the anterior cingulate cortex contributed to seizure‐like activities. Adv Sci. 2024;11(39):2308444. [DOI] [PMC free article] [PubMed] [Google Scholar]
14. Etkin A, Büchel C, Gross JJ. The neural bases of emotion regulation. Nat Rev Neurosci. 2015;16(11):693–700. [DOI] [PubMed] [Google Scholar]
15. Kepecs A, Fishell G. Interneuron cell types are fit to function. Nature. 2014;505(7483):318–326. [DOI] [PMC free article] [PubMed] [Google Scholar]
16. Isaacson JS, Scanziani M. How inhibition shapes cortical activity. Neuron. 2011;72(2):231–243. [DOI] [PMC free article] [PubMed] [Google Scholar]
17. Devinsky O, Morrell MJ, Vogt BA. Contributions of anterior cingulate cortex to behaviour. Brain. 1995;118:279–306. [DOI] [PubMed] [Google Scholar]
18. Trevelyan AJ, Sussillo D, Watson BO, Yuste R. Modular propagation of epileptiform activity: evidence for an inhibitory veto in neocortex. J Neurosci. 2006;26(48):12447–12455. [DOI] [PMC free article] [PubMed] [Google Scholar]
19. de Curtis M, Avoli M. GABAergic networks jump‐start focal seizures. Epilepsia. 2016;57(5):679–687. [DOI] [PMC free article] [PubMed] [Google Scholar]
20. Ilie A, Raimondo JV, Akerman CJ. Adenosine release during seizures attenuates GABAA receptor‐mediated depolarization. J Neurosci. 2012;32(15):5321–5332. [DOI] [PMC free article] [PubMed] [Google Scholar]
21. Sohal VS, Zhang F, Yizhar O, Deisseroth K. Parvalbumin neurons and gamma rhythms enhance cortical circuit performance. Nature. 2009;459(7247):698–702. [DOI] [PMC free article] [PubMed] [Google Scholar]
22. Sloviter RS. Decreased hippocampal inhibition and a selective loss of interneurons in experimental epilepsy. Science. 1987;235(4784):73–76. [DOI] [PubMed] [Google Scholar]
23. Soboleva EB, Amakhin DV, Sinyak DS, Zaitsev AV. Modulation of seizure‐like events by the small conductance and ATP‐sensitive potassium ion channels. Biochem Biophys Res Commun. 2022;623:74–80. [DOI] [PubMed] [Google Scholar]
24. Amakhin DV, Sinyak DS, Soboleva EB, Zaitsev AV. HCN channels promote Na/K‐ATPase activity during slow afterhyperpolarization after seizure‐like events in vitro. J Physiol. 2025;603(5):1197–1223. [DOI] [PubMed] [Google Scholar]
25. Breton VL, Dufour S, Chinvarun Y, Del Campo JM, Bardakjian BL, Carlen PL. Transitions between neocortical seizure and non‐seizure‐like states and their association with presynaptic glutamate release. Neurobiol Dis. 2020;146:105124. [DOI] [PubMed] [Google Scholar]
26. Aeed F, Shnitzer T, Talmon R, Schiller Y. Layer‐ and cell‐specific recruitment dynamics during epileptic seizures in vivo. Ann Neurol. 2020;87(1):97–115. [DOI] [PubMed] [Google Scholar]
27. Kontou G, Antonoudiou P, Podpolny M, Szulc BR, Arancibia‐Carcamo IL, Higgs NF, et al. Miro1‐dependent mitochondrial dynamics in parvalbumin interneurons. Elife. 2021;10:e65215. [DOI] [PMC free article] [PubMed] [Google Scholar]
28. Wilson NR, Runyan CA, Wang FL, Sur M. Division and subtraction by distinct cortical inhibitory networks in vivo. Nature. 2012;488(7411):343–348. [DOI] [PMC free article] [PubMed] [Google Scholar]
29. Lovett‐Barron M, Turi GF, Kaifosh P, Lee PH, Bolze F, Sun X‐H, et al. Regulation of neuronal input transformations by tunable dendritic inhibition. Nat Neurosci. 2012;15(3):423–430. [DOI] [PubMed] [Google Scholar]
30. Urban‐Ciecko J, Barth AL. Somatostatin‐expressing neurons in cortical networks. Nat Rev Neurosci. 2016;17(7):401–409. [DOI] [PMC free article] [PubMed] [Google Scholar]
31. Magloire V, Cornford J, Lieb A, Kullmann DM, Pavlov I. KCC2 overexpression prevents the paradoxical seizure‐promoting action of somatic inhibition. Nat Commun. 2019;10(1):1225. [DOI] [PMC free article] [PubMed] [Google Scholar]
32. Cammarota M, Losi G, Chiavegato A, Zonta M, Carmignoto G. Fast spiking interneuron control of seizure propagation in a cortical slice model of focal epilepsy. J Physiol. 2013;591(4):807–822. [DOI] [PMC free article] [PubMed] [Google Scholar]
33. Toyoda I, Fujita S, Thamattoor AK, Buckmaster PS. Unit activity of hippocampal interneurons before spontaneous seizures in an animal model of temporal lobe epilepsy. J Neurosci. 2015;35(16):6600–6618. [DOI] [PMC free article] [PubMed] [Google Scholar]
34. Drexel M, Romanov RA, Wood J, Weger S, Heilbronn R, Wulff P, et al. Selective silencing of hippocampal parvalbumin interneurons induces development of recurrent spontaneous limbic seizures in mice. J Neurosci. 2017;37(34):8166–8179. [DOI] [PMC free article] [PubMed] [Google Scholar]
35. Kaila K, Price TJ, Payne JA, Puskarjov M, Voipio J. Cation‐chloride cotransporters in neuronal development, plasticity and disease. Nat Rev Neurosci. 2014;15(10):637–654. [DOI] [PMC free article] [PubMed] [Google Scholar]
36. Roth FC, Hu H. An axon‐specific expression of HCN channels catalyzes fast action potential signaling in GABAergic interneurons. Nat Commun. 2020;11(1):2248. [DOI] [PMC free article] [PubMed] [Google Scholar]
37. Headley DB, Latimer B, Aberbach A, Nair SS. Spatially targeted inhibitory rhythms differentially affect neuronal integration. Elife. 2026;13:RP95562. [DOI] [PMC free article] [PubMed] [Google Scholar]
38. Paz JT, Davidson TJ, Frechette ES, Delord B, Parada I, Peng K, et al. Closed‐loop optogenetic control of thalamus as a tool for interrupting seizures after cortical injury. Nat Neurosci. 2013;16(1):64–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
39. Krook‐Magnuson E, Armstrong C, Oijala M, Soltesz I. On‐demand optogenetic control of spontaneous seizures in temporal lobe epilepsy. Nat Commun. 2013;4(1):1376. [DOI] [PMC free article] [PubMed] [Google Scholar]
40. Meeren HK, Pijn JPM, Van Luijtelaar EL, Coenen AM, da Silva FHL. Cortical focus drives widespread corticothalamic networks during spontaneous absence seizures in rats. J Neurosci. 2002;22(4):1480–1495. [DOI] [PMC free article] [PubMed] [Google Scholar]
41. Löscher W. Animal models of seizures and epilepsy: past, present, and future role for the discovery of antiseizure drugs. Neurochem Res. 2017;42(7):1873–1888. [DOI] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

EPI-67-5531-s005.tif^{(23.4MB, tif)}

EPI-67-5531-s008.tif^{(39.9MB, tif)}

EPI-67-5531-s001.tif^{(29.9MB, tif)}

EPI-67-5531-s004.tif^{(11MB, tif)}

EPI-67-5531-s002.tif^{(9.9MB, tif)}

EPI-67-5531-s003.tif^{(15.3MB, tif)}

Table S1.

EPI-67-5531-s007.xls^{(51KB, xls)}

Data S1.

EPI-67-5531-s006.docx^{(17.5KB, docx)}

Data Availability Statement

The research data are available from the corresponding authors upon request.

[epi70365-bib-0001] 1. Fisher RS, Acevedo C, Arzimanoglou A, Bogacz A, Cross JH, Elger CE, et al. ILAE official report: a practical clinical definition of epilepsy. Epilepsia. 2014;55(4):475–482. [DOI] [PubMed] [Google Scholar]

[epi70365-bib-0002] 2. Thijs RD, Surges R, O'Brien TJ, Sander JW. Epilepsy in adults. The Lancet. 2019;393(10172):689–701. [DOI] [PubMed] [Google Scholar]

[epi70365-bib-0003] 3. Scheffer IE, Berkovic S, Capovilla G, Connolly MB, French J, Guilhoto L, et al. ILAE classification of the epilepsies: position paper of the ILAE Commission for Classification and Terminology. Epilepsia. 2017;58(4):512–521. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0004] 4. Kwan P, Brodie MJ. Early identification of refractory epilepsy. N Engl J Med. 2000;342(5):314–319. [DOI] [PubMed] [Google Scholar]

[epi70365-bib-0005] 5. Chen Z, Brodie MJ, Liew D, Kwan P. Treatment outcomes in patients with newly diagnosed epilepsy treated with established and new antiepileptic drugs: a 30‐year longitudinal cohort study. JAMA Neurol. 2018;75(3):279–286. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0006] 6. Feigin VL, Vos T, Nair BS, Hay SI, Abate YH, Abd Al Magied AH, et al. Global, regional, and national burden of epilepsy, 1990–2021: A systematic analysis for the global burden of disease study 2021. Lancet Public Health. 2025;10(3):e203–e227. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0007] 7. Sohal VS, Rubenstein JLR. Excitation‐inhibition balance as a framework for investigating mechanisms in neuropsychiatric disorders. Mol Psychiatry. 2019;24(9):1248–1257. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0008] 8. Wang Y, Wang Y, Xu C, Wang S, Tan N, Chen C, et al. Direct septum‐hippocampus cholinergic circuit attenuates seizure through driving somatostatin inhibition. Biol Psychiatry. 2020;87(9):843–856. [DOI] [PubMed] [Google Scholar]

[epi70365-bib-0009] 9. Wicker E, Forcelli PA. Optogenetic activation of the reticular nucleus of the thalamus attenuates limbic seizures via inhibition of the midline thalamus. Epilepsia. 2021;62(9):2283–2296. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0010] 10. Bush G, Luu P, Posner MI. Cognitive and emotional influences in anterior cingulate cortex. Trends Cogn Sci. 2000;4(6):215–222. [DOI] [PubMed] [Google Scholar]

[epi70365-bib-0011] 11. Braga AM, Fujisao EK, Verdade RC, Paschoalato RP, Paschoalato RP, Yamashita S, et al. Investigation of the cingulate cortex in idiopathic generalized epilepsy. Epilepsia. 2015;56(11):1803–1811. [DOI] [PubMed] [Google Scholar]

[epi70365-bib-0012] 12. Blumenfeld H. Impaired consciousness in epilepsy. The Lancet Neurology. 2012;11(9):814–826. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0013] 13. Li XH, Shi W, Zhao ZX, Matsuura T, Lu JS, Che J, et al. Increased GluK1 subunit receptors in Corticostriatal projection from the anterior cingulate cortex contributed to seizure‐like activities. Adv Sci. 2024;11(39):2308444. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0014] 14. Etkin A, Büchel C, Gross JJ. The neural bases of emotion regulation. Nat Rev Neurosci. 2015;16(11):693–700. [DOI] [PubMed] [Google Scholar]

[epi70365-bib-0015] 15. Kepecs A, Fishell G. Interneuron cell types are fit to function. Nature. 2014;505(7483):318–326. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0016] 16. Isaacson JS, Scanziani M. How inhibition shapes cortical activity. Neuron. 2011;72(2):231–243. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0017] 17. Devinsky O, Morrell MJ, Vogt BA. Contributions of anterior cingulate cortex to behaviour. Brain. 1995;118:279–306. [DOI] [PubMed] [Google Scholar]

[epi70365-bib-0018] 18. Trevelyan AJ, Sussillo D, Watson BO, Yuste R. Modular propagation of epileptiform activity: evidence for an inhibitory veto in neocortex. J Neurosci. 2006;26(48):12447–12455. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0019] 19. de Curtis M, Avoli M. GABAergic networks jump‐start focal seizures. Epilepsia. 2016;57(5):679–687. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0020] 20. Ilie A, Raimondo JV, Akerman CJ. Adenosine release during seizures attenuates GABAA receptor‐mediated depolarization. J Neurosci. 2012;32(15):5321–5332. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0021] 21. Sohal VS, Zhang F, Yizhar O, Deisseroth K. Parvalbumin neurons and gamma rhythms enhance cortical circuit performance. Nature. 2009;459(7247):698–702. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0022] 22. Sloviter RS. Decreased hippocampal inhibition and a selective loss of interneurons in experimental epilepsy. Science. 1987;235(4784):73–76. [DOI] [PubMed] [Google Scholar]

[epi70365-bib-0023] 23. Soboleva EB, Amakhin DV, Sinyak DS, Zaitsev AV. Modulation of seizure‐like events by the small conductance and ATP‐sensitive potassium ion channels. Biochem Biophys Res Commun. 2022;623:74–80. [DOI] [PubMed] [Google Scholar]

[epi70365-bib-0024] 24. Amakhin DV, Sinyak DS, Soboleva EB, Zaitsev AV. HCN channels promote Na/K‐ATPase activity during slow afterhyperpolarization after seizure‐like events in vitro. J Physiol. 2025;603(5):1197–1223. [DOI] [PubMed] [Google Scholar]

[epi70365-bib-0025] 25. Breton VL, Dufour S, Chinvarun Y, Del Campo JM, Bardakjian BL, Carlen PL. Transitions between neocortical seizure and non‐seizure‐like states and their association with presynaptic glutamate release. Neurobiol Dis. 2020;146:105124. [DOI] [PubMed] [Google Scholar]

[epi70365-bib-0026] 26. Aeed F, Shnitzer T, Talmon R, Schiller Y. Layer‐ and cell‐specific recruitment dynamics during epileptic seizures in vivo. Ann Neurol. 2020;87(1):97–115. [DOI] [PubMed] [Google Scholar]

[epi70365-bib-0027] 27. Kontou G, Antonoudiou P, Podpolny M, Szulc BR, Arancibia‐Carcamo IL, Higgs NF, et al. Miro1‐dependent mitochondrial dynamics in parvalbumin interneurons. Elife. 2021;10:e65215. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0028] 28. Wilson NR, Runyan CA, Wang FL, Sur M. Division and subtraction by distinct cortical inhibitory networks in vivo. Nature. 2012;488(7411):343–348. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0029] 29. Lovett‐Barron M, Turi GF, Kaifosh P, Lee PH, Bolze F, Sun X‐H, et al. Regulation of neuronal input transformations by tunable dendritic inhibition. Nat Neurosci. 2012;15(3):423–430. [DOI] [PubMed] [Google Scholar]

[epi70365-bib-0030] 30. Urban‐Ciecko J, Barth AL. Somatostatin‐expressing neurons in cortical networks. Nat Rev Neurosci. 2016;17(7):401–409. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0031] 31. Magloire V, Cornford J, Lieb A, Kullmann DM, Pavlov I. KCC2 overexpression prevents the paradoxical seizure‐promoting action of somatic inhibition. Nat Commun. 2019;10(1):1225. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0032] 32. Cammarota M, Losi G, Chiavegato A, Zonta M, Carmignoto G. Fast spiking interneuron control of seizure propagation in a cortical slice model of focal epilepsy. J Physiol. 2013;591(4):807–822. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0033] 33. Toyoda I, Fujita S, Thamattoor AK, Buckmaster PS. Unit activity of hippocampal interneurons before spontaneous seizures in an animal model of temporal lobe epilepsy. J Neurosci. 2015;35(16):6600–6618. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0034] 34. Drexel M, Romanov RA, Wood J, Weger S, Heilbronn R, Wulff P, et al. Selective silencing of hippocampal parvalbumin interneurons induces development of recurrent spontaneous limbic seizures in mice. J Neurosci. 2017;37(34):8166–8179. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0035] 35. Kaila K, Price TJ, Payne JA, Puskarjov M, Voipio J. Cation‐chloride cotransporters in neuronal development, plasticity and disease. Nat Rev Neurosci. 2014;15(10):637–654. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0036] 36. Roth FC, Hu H. An axon‐specific expression of HCN channels catalyzes fast action potential signaling in GABAergic interneurons. Nat Commun. 2020;11(1):2248. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0037] 37. Headley DB, Latimer B, Aberbach A, Nair SS. Spatially targeted inhibitory rhythms differentially affect neuronal integration. Elife. 2026;13:RP95562. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0038] 38. Paz JT, Davidson TJ, Frechette ES, Delord B, Parada I, Peng K, et al. Closed‐loop optogenetic control of thalamus as a tool for interrupting seizures after cortical injury. Nat Neurosci. 2013;16(1):64–70. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0039] 39. Krook‐Magnuson E, Armstrong C, Oijala M, Soltesz I. On‐demand optogenetic control of spontaneous seizures in temporal lobe epilepsy. Nat Commun. 2013;4(1):1376. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0040] 40. Meeren HK, Pijn JPM, Van Luijtelaar EL, Coenen AM, da Silva FHL. Cortical focus drives widespread corticothalamic networks during spontaneous absence seizures in rats. J Neurosci. 2002;22(4):1480–1495. [DOI] [PMC free article] [PubMed] [Google Scholar]

[epi70365-bib-0041] 41. Löscher W. Animal models of seizures and epilepsy: past, present, and future role for the discovery of antiseizure drugs. Neurochem Res. 2017;42(7):1873–1888. [DOI] [PubMed] [Google Scholar]

PERMALINK

Anterior cingulate cortex neuron subtypes differentially regulate seizures

Ziqian Yan

Ting Tang

Kaishan Wang

Bin Fu

Zesheng Li

Mingshan Liu

Hongyi Huang

Yanfeng Yang

Qiang Zheng

Yangmin Zheng

Yongzhi Shan

Yumin Luo

Guoguang Zhao

Abstract

Objective

Methods

Results

Significance

Key points.

1. INTRODUCTION

2. MATERIALS AND METHODS

2.1. Animals

2.2. Acute seizure model and behavioral assessment

2.3. EEG recording and analysis

2.4. Stereotaxic surgery

2.5. In vivo fiber photometry

2.6. In vivo miniscope calcium imaging

2.7. Chemogenetic manipulation

2.8. Optogenetic manipulation

2.9. Immunofluorescence staining

2.10. Statistical analysis

3. RESULTS

3.1. PTZ induces robust behavioral and electrophysiological seizures in mice

FIGURE 1.

3.2. Widespread activation of excitatory and inhibitory neurons in the ACC during seizures

FIGURE 2.

3.3. ACC CaMKII neurons bidirectionally regulate seizurelike activity

FIGURE 3.

3.4. ACC GABAergic neurons bidirectionally regulate seizurelike activity

FIGURE 4.

3.5. Activation of SST interneurons, but not PV interneurons, effectively attenuates seizures

FIGURE 5.

3.6. Basal activity of both SST and PV neurons is essential for maintaining ACC network homeostasis

FIGURE 6.

4. DISCUSSION

5. CONCLUSIONS

AUTHOR CONTRIBUTIONS

FUNDING INFORMATION

CONFLICT OF INTEREST STATEMENT

ETHICAL PUBLICATION STATEMENT

Supporting information

ACKNOWLEDGMENTS

DATA AVAILABILITY STATEMENT

REFERENCES

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases