Increased GluK1 Subunit Receptors in Corticostriatal Projection from the Anterior Cingulate Cortex Contributed to Seizure-Like Activities

Abstract

The corticostriatal connection plays a crucial role in cognitive, emotional, and motor control. However, the specific roles and synaptic transmissions of corticostriatal connection are less studied, especially the corticostriatal transmission from the anterior cingulate cortex (ACC). Here, a direct glutamatergic excitatory synaptic transmission in the corticostriatal projection from the ACC is found. Kainate receptors (KAR)-mediated synaptic transmission is increased in this corticostriatal connection both in vitro and in vivo seizure-like activities. GluK1 containing KARs and downstream calcium-stimulated adenylyl cyclase subtype 1 (AC1) are involved in the upregulation of KARs following seizure-like activities. Inhibiting the activities of ACC or its corticostriatal connection significantly attenuated pentylenetetrazole (PTZ)-induced seizure. Additionally, injection of GluK1 receptor antagonist UBP310 or the AC1 inhibitor NB001 both show antiepileptic effects. The studies provide direct evidence that KARs are involved in seizure activity in the corticostriatal connection and the KAR-AC1 signaling pathway is a potential novel antiepileptic strategy.

Keywords: AC1; anterior cingulate cortex; corticostriatal projection; kainate receptor; seizure; striatum.

Figure 1

The dorsal striatum neuron received glutamatergic direct descending projection from the ACC. A) The procedure for anterograde and retrograde tracing projections of the ACC using VISoR technology and 3D reconstruction. B) Upper: the lateral view of whole‐brain distribution from ACC inputs and outputs (red: DsRed⁺; green: EGFP⁺). Bottom: the horizontal, sagittal, and coronal views for one 3D‐reconstructed brain from the ACC to the striatum. scale bar, 500 µm. C) Fluorescence images of the brain slice containing ACC and striatum after injection of AAV helper and RV into the ACC, scale bar, 500 and 100 µm. D) The mCherry‐labeled retrograde neurons in the striatum, scale bar, 100 µm. E) Schematic of the viral injection site in the ACC for sparsely labeling experiment (left). Fluorescence images of the brain slice from the ACC‐striatum sparsely labeled (middle and right), scale bar: 200 µm (middle) and 100 µm (right). F) Retrograde adeno‐associated virus AAV2/R‐hSyn‐mCherry injected into the striatum in C57 mice. Immunostaining and statistics showing the colocalization of mCherry‐positive neurons in the ACC were mainly CaMKII‐positive neurons, n = 14 – 18 slices/4 mice, ***p < 0.001, unpaired t‐test; left scale bar, 100 µm; right scale bar, 20 µm. G) Schematic diagram and representative photomicrograph showing the viral injection site in the ACC, light stimulation site, and whole‐cell patch‐recording placement in the striatum, scale bar: 200 µm (left) and 20 µm (right). H) Blue light stimulation (473 nm) evoked action potentials in the ACC neuron at the I‐clamp configure and evoked EPSCs in the dorsal striatum neuron at the V‐clamp configure at single, 5, 10, and 20 Hz for 1 s. I) Blue light‐induced EPSCs were recorded in striatum neurons in ACSF and after the sequential application of TTX (1 µM) and 4‐AP (100 µM), n = 11 neurons/4 mice. J) Schematic diagram and representative recording diagram showing the placement of stimulating electrode in the ACC and recording electrode in the dorsal striatum (left and right upper). The morphological properties of neurons in the striatum were labeled by biocytin during recording (right bottom), scale bar: 20 µm. K) Sample traces and pooled data showed the input‐output relationship of basal EPSCs in the striatum neuron after stimulating the ACC (n = 7 neurons/3 mice). L) EPSCs blocked by NMDA receptor antagonist AP‐5 (50 µM) and AMPA/KA receptor antagonist CNQX (20 µM) together within the presence of GABA_A receptor antagonist picrotoxin (100 µM) in the ACC to striatum synaptic transmission. Sample time course points (left) and statistical results (right) showed the EPSCs in the presence of CNQX and AP‐5 (n = 7 neurons/4 mice, *p < 0.05 and ***p < 0.001, unpaired t‐test). error bars indicated SEM.

Figure 2

KAR‐mediated EPSCs in the ACC‐dorsal striatum pathway. A and F) Schematic diagram showing the placement of stimulating electrode in layer V/VI of the ACC and recording electrode in the dorsal striatum A) and interior ACC F). B and G) In the presence of picrotoxin (100 µM) and AP‐5 (50 µM), KAR‐mediated EPSCs were observed after the application of a potent AMPA receptor antagonist GYKI 53655 (100 µM) and then blocked by AMPA/KA receptor antagonist CNQX (20 µM) in the ACC‐striatum (B) and interior ACC (G). Sample traces (top) and sample time course points (bottom) showed the EPSCs in the presence of GYKI 53655 and CNQX. C and H) Representative traces of KA receptor‐mediated EPSCs were obtained after the application of different numbers of stimulation (1, 5, 10, and 20 shocks) at 200 Hz in the presence of GYKI 53655. D) Summary results showed CNQX blocked KAR‐mediated EPSCs in the interior ACC but not in the ACC‐striatum (ACC‐striatum: n = 14 neurons/5 mice, p > 0.05, paired t‐test; ACC: n = 7 neurons/3 mice; *p < 0.05, paired t‐test). E) The percentage of KAR‐mediated EPSCs in basal synaptic transmission in the ACC‐striatum and interior ACC synapses. I) Statistical results showed that the peak amplitude of the KAR‐mediated EPSCs in the ACC‐striatum was smaller than those in the interior ACC by repetitive stimulations at 200 Hz (striatum: n = 13 neurons/5 mice; ACC: n = 10 neurons/4 mice; F _{(1, 92)} = 132.22, ***p < 0.001, two‐way ANOVA). J) Schematic diagram and representative photomicrograph showing the viral injection site in the ACC, light stimulation site, and whole‐cell patch‐recording placement in the striatum in CaMKIIα‐Cre mice, scale bar: 200 µm. K) Optical‐induced KAR‐mediated EPSCs in the ACC to striatum synapses in the presence of GYKI 53655 and CNQX (n = 9 neurons/3 mice, p > 0.05, paired t‐test). L) Sample traces (left) and statistical results (right) show that post‐synaptic currents evoked by puff‐application of glutamate (100 µM) were dramatically decreased by GYKI 53655 in the striatum neuron in the presence of picrotoxin and AP‐5, but still had residual currents. These residual currents can be completely blocked by CNQX (n = 6–8 neurons/4 mice, **p < 0.01, unpaired t‐test). M) Sample traces showed that no synaptic KAR‐mediated EPSCs in the striatum neuron were evoked by 10 shocks (200 Hz) of electrical stimulation in the presence of GYKI 53655 (top). Puff‐application of glutamate (100 µM) and GluK1 agonist ATPA (100 µM) both can induce KAR‐mediated currents. These KAR‐mediated currents were blocked by a specific GluK1 antagonist UBP310 (10 µM) and CNQX (bottom). N) Statistical results showing the amplitude of KAR‐mediated EPSCs evoked by 10 shocks (200 Hz) electrical stimulation, puff‐application of glutamate and ATPA in the presence of GYKI 53655, UBP310 and CNQX (glutamate: n = 7 neurons/3 mice, **p < 0.01, paired t‐test; ATPA: n = 5 neurons/3 mice, **p < 0.01, paired t‐test). ns. means no significant difference, error bars indicated SEM.

Figure 3

KAR‐mediated EPSCs increased after Mg²⁺‐free ACSF‐induced in vitro seizure‐like activity. A)The sample traces showing seizure‐like activity were induced by perfusion of Mg²⁺‐free ACSF in the striatum neuron at the I‐clamp. B) Sample traces (upper) and sample time course points (bottom) show the EPSCs in the presence of GYKI 53655, UBP310, and CNQX in the striatum neuron by stimulating the ACC after 2 h seizure‐like activity. C) Statistical results show the percentage of KAR‐mediated EPSCs amplitude after the perfusion of Mg²⁺‐free ACSF for 0.5, 1, 2, and 2 h with in presence of UBP310 and CNQX in the striatum neuron by stimulating the ACC. Perfusion with Mg²⁺‐free ACSF was time‐dependently increased KAR‐mediated EPSCs and blocked by UBP310 and CNQX (ACSF: n = 10 neurons/4 mice; Mg²⁺‐free ACSF for 0.5 h, n = 9 neurons/4 mice; 1 h, n = 10 neurons/4 mice; 2 h, n = 10 neurons/4 mice; 2 h + UBP310, n = 5 neurons/3 mice; 2 h + CNQX, n = 8 neurons/3 mice; *p < 0.05 and ***p < 0.001, * means compared with ACSF; ^# p < 0.05, ^# means compared with Mg²⁺‐free ACSF 2 h). D) Representative traces of KAR‐mediated EPSCs were obtained after the application of different numbers of stimuli (1, 5, 10, and 20 shocks) at 200 Hz after 2 h of seizure‐like activity. E) Summary results showing the peak amplitude of the KAR‐mediated EPSCs by repetitive stimulations (200 Hz) after perfusing Mg²⁺‐free ACSF for 0.5, 1, and 2 h in the ACC‐striatum synapses, n = 6–7 neurons/3‐4 mice. F) Pooled data showing the input‐output relationship of KAR‐mediated EPSCs after perfusing Mg²⁺‐free ACSF for 0.5, 1, and 2 h. G) Statistical results show the percentage of KAR‐mediated EPSCs in wild‐type (WT) mice and GluK1 ^−/‐ mice. KAR‐mediated EPSCs were no increased after perfusing Mg²⁺‐free ACSF for 2 h in the GluK1 ^−/− mice (WT: n = 9 neurons/3 mice, ***p < 0.001, unpaired t‐test; GluK1 ^−/−: n = 6–7 neurons/3 mice, p > 0.05, unpaired t‐test). H) Statistical results show the amplitude of the KAR‐mediated EPSCs by repetitive stimulations (200 Hz) after perfusing Mg²⁺‐free ACSF for 2 h in the GluK1 ^−/− mice (n = 8–13 neurons/3‐4 mice, F _{(1, 76)} = 0.49, p > 0.05, two‐way ANOVA). I) Statistical results show the input‐output relationship after perfusing Mg²⁺‐free ACSF for 2 h in the GluK1 ^−/− mice (F _{(1, 95)} = 0.34, p > 0.05, two‐way ANOVA). Error bars indicated SEM.

Figure 4

The roles of GluK1 upregulated in the ACC‐striatum pathway after PTZ‐induced in vivo seizure activity. A, B) Schematic diagram and sample trance of EEG recording after intraperitoneal (i.p.) injection of PTZ (50 mg kg⁻¹) induced seizure‐like activity in mice. C, D) Representative and statistics western blots of KAR subunits GluK1, GluK2/3, GluK4, and GluK5 expression in total homogenates of the ACC, striatum, and hippocampus, n = 5–6. E, F) Representative and statistics western blots of AMPAR subunits GluA1, GluA1‐Ser 831, GluA1‐Ser 845, and NMDAR subunits GluN2A, GluN2B expression in total homogenates of the ACC, striatum, and hippocampus, n = 5–6. G) Upper: schematic diagram showing the placement of stimulating electrode in the ACC and recording electrode in the dorsal striatum in PTZ model slice. Bottom: representative trace of action potential firing rate in striatum neuron in saline and PTZ groups. H) The sample traces and summary time course points show the amplitude percentage of EPSCs in the PTZ‐induced seizure and saline mice in the ACC‐striatum (Saline: n = 6 neurons/3 mice; PTZ: n = 9 neurons/5 mice). I) Statistical results show that the percentage of KAR‐mediated EPSCs increased in the PTZ group (Saline: n = 8 neurons/3 mice; PTZ: n = 13 neurons/5 mice; ***p < 0.001, paired t‐test). J) Statistical results showing the amplitude of the KAR‐mediated EPSCs evoked by repetitive stimulations (200 Hz) were significantly increased in the PTZ group compared with the saline group (Saline: n = 8 neurons/3 mice; PTZ: n = 12 neurons/4 mice; F _{(1, 72)} = 28.66, P < 0.001, two‐way ANOVA). K) Statistical results showing the input‐output curves of the KAR‐mediated EPSCs were shifted to the left in the PTZ group compared with the saline group (Saline: n = 8 neurons/3 mice; PTZ: n = 9 neurons/4 mice; F _(1,75) = 11.37, p < 0.01, two‐way ANOVA). *p < 0.05, **p < 0.01, and ***p < 0.001, error bars indicated SEM.

Figure 5

Inhibition of the corticostriatal projection from the ACC attenuated seizure behavior. A) The ΔF/F of representative trace (upper) and heatmaps (bottom) show that Ca²⁺ signals are recorded by fiber‐photometry of the ACC (left) and striatum (right) in mice injected with saline (n = 7 mice). B) The ΔF/F of representative trace (upper) and heatmaps (bottom) show that Ca²⁺ signals of the ACC (left) and striatum (right) in mice injected with PTZ (n = 6 mice). The colored bar on the right of heatmaps indicates ΔF/F (%). C) Schematic diagram and representative photomicrograph showing the chemical genetics viral injection site in the ACC, and intraperitoneal injection of CNO (1 mg kg⁻¹), scale bar: 100 µm. D) Chemical inhibition of the activity of the ACC reduced PTZ‐induced seizure severity scores (Saline, n = 12 mice; hM3Dq, n = 10 mice, p = 0.8576, unpaired t‐test; hM4Di, n = 10 mice, **p < 0.01, unpaired t‐test). E) Chemical inhibition of the activity of the ACC reduced the PTZ‐induced seizure susceptibilities (Saline vs hM3Dq, p = 0.3111, unpaired t‐test; Saline vs hM4Di, **p < 0.01, unpaired t‐test). F) Chemical inhibition of the activity of the ACC increased the latency time to the partial clonus (PC) (Saline vs hM4Di, *p < 0.05, unpaired t‐test) and generalized tonic‐clonic (GTC) (Saline vs hM4Di, ***p < 0.001, unpaired t‐test) stage of PTZ‐induced seizures. (G) The rate of mortality of mice injected with saline or CNO after PTZ injection. H) Schematic diagram and representative photomicrograph showing the viral injection site in the ACC, and light stimulation site in the striatum, scale bar: 100 µm. I) Yellow light (593 nm) inhibition of the corticostriatal projection from the ACC reduced the PTZ‐induced seizure severity scores (EYFP, n = 12 mice; ChR2, n = 11 mice, p = 0.4044, unpaired t‐test; eNpHR, n = 11 mice, **p < 0.01, unpaired t‐test). J) Seizure susceptibilities were reduced by light inhibition of the corticostriatal projection (EYFP vs ChR2, p = 0.7134, unpaired t‐test; EYFP vs eNpHR, ***p < 0.001, unpaired t‐test). K) Light inhibition of the corticostriatal projection from the ACC increased the latency time to the PC (EYFP vs eNpHR, ***p < 0.001, unpaired t‐test) and generalized clonus (GC) stages (EYFP vs eNpHR, **p < 0.01, unpaired t‐test). L) The rate of mortality of mice in EYFP, ChR2, and eNpHR groups after blue or yellow light stimulation. n.s means no significant difference, error bars indicated SEM.

Figure 6

PTZ‐induced seizure behaviors were attenuated in GluK1 ^−/− mice or in mice with an injection of GluK1 antagonist UBP310. A) The PTZ‐induced seizure severity scores were reduced in the GluK1^−/− mice (WT and GluK1^−/− , each n = 15 mice, *p < 0.05, unpaired t‐test). B) The PTZ‐induced seizure susceptibilities were decreased in the GluK1^−/− mice (*p< 0.05, unpaired t‐test). C) The latency time to the PC stage was delayed in GluK1^−/− mice after PTZ‐induced seizure (***p < 0.001, unpaired t‐test). D) The rate of mortality induced by PTZ in the GluK1^−/− mice. E) Schematic diagram showing the drug cannula embedded site in the striatum, and intraperitoneal injection of PTZ. F) UBP310 injected into the striatum reduced the PTZ‐induced seizure severity scores (Vehicle, n = 9 mice; UBP310, n = 10 mice; ***p < 0.001, unpaired t‐test). G) UBP310 injected into the striatum reduced the PTZ‐induced seizure susceptibilities (*p < 0.05, unpaired t‐test). H) The latency time to the stage of seizures was not changed in mice with striatum‐injected UBP310. I) Schematic diagram showing the drug cannula embedded site in the ACC, and intraperitoneal injection of PTZ. J) UBP310 injected into the ACC reduced the PTZ‐induced seizure severity scores (Vehicle, n = 9 mice; UBP310, n = 11 mice; *p < 0.05, unpaired t‐test). K) UBP310 injected into the ACC did not affect the PTZ‐induced seizure susceptibilities. L) The latency time to the stage of seizures was not changed in mice with ACC‐injected UBP310. Error bars indicated SEM.

Figure 7

Calcium‐stimulated adenylyl cyclase subtype 1 (AC1) was involved in seizure‐like activity. A) Upper: schematic diagram showing the placement of stimulating electrode in the ACC and recording electrode in the dorsal striatum in Mg²⁺‐ACSF culture slice. Bottom: statistical results showing that KAR‐mediated EPSCs were not increased after perfusing Mg²⁺‐free ACSF for 2 h in AC1 ^−/− mice, but increased in WT and AC8 ^−/− mice (WT: n = 9 neurons/3 mice; AC1 ^−/−, n = 7 neurons/3 mice; AC8 ^−/−, n = 6–8 neurons/3 mice; **p < 0.01 and ***p < 0.001, unpaired t‐test). B) Statistical results showed the amplitude of the KAR‐mediated EPSCs by repetitive stimulations (200 Hz) (F _{(1, 48)} = 1.78, P > 0.05, two‐way ANOVA) and the input‐output relationship (F _{(1, 60)} = 1.81, p > 0.05, two‐way ANOVA) after perfusing Mg²⁺‐free ACSF for 2 h in the AC1 ^−/− mice. C) Statistical results show the amplitude of the KAR‐mediated EPSCs by repetitive stimulations (200 Hz) (F _{(1, 48)} = 13.53, p < 0.001, two‐way ANOVA) and the input‐output relationship (F _{(1, 55)} = 4.53, p < 0.05, two‐way ANOVA) after perfusing Mg²⁺‐free ACSF for 2 h in the AC8 ^−/− mice. D) Representative and statistics western blots of AC1 expression in total homogenates of the ACC, striatum, and hippocampus, n = 5–6. E) The PTZ‐induced seizure severity scores, seizure susceptibilities, the latency time to the stage of seizures, and the rate of mortality in the WT (n = 15 mice), AC1 ^−/− (n = 16 mice), and AC8 ^−/− (n = 10 mice). F) EEG recording showed that NB001 reduced the amplitude and firing rate of EEG compared with the saline group. G) The ΔF/F of representative trace (upper) and heatmaps (bottom) show that Ca²⁺ signals of the ACC (left) and striatum (right) in mice injected with NB001 + PTZ. The colored bar on the right of heatmaps indicates ΔF/F(%), n = 4 mice, scale bar: 200 µm. H) The PTZ‐induced seizure severity scores (F _{(4, 62)} = 3.018.53, p < 0.05, one‐way ANOVA) and seizure susceptibilities (F _{(4, 64)} = 8.050, p < 0.0001, one‐way ANOVA) were significantly decreased in the i.p. injected AC1 inhibitor NB001 (Saline, n = 12 mice; 2 mg/kg, n = 12 mice; 10 mg kg⁻¹, n = 12 mice; 20 mg kg⁻¹, n = 15 mice; 50 mg kg⁻¹, n = 16 mice). The latency time to the stage of PC, GC, and GTC seizures induced by PTZ was increased in NB001‐injected mice (F _{(4, 169)} = 10.10, p < 0.0001, two‐way ANOVA). n.s means no significant difference, *p < 0.05, **p < 0.01, and ***p < 0.001, error bars indicated SEM.