Acute seizures induces the decreased level of total m6A in the hippocampus of mice
To determine whether m6A could participate in seizures, we used m6A-ELISA assay to assess the hippocampal m6A in PTZ treated mice and the result showed that hippocampal m6A content was significantly decreased in mice with seizures when compared with the normal saline (NS) mice (t = 4.717, p = 0.0001) (Fig. 1, A). Next, we verified the m6A level by immunohistochemistry assay, and the results showed that PTZ-treated mice exhibited significantly decreased level of m6A compared to normal saline (NS) mice (t = 5.135, p = 0.0021) (Fig. 1, B). In addition, m6A-dot blot further showed that hippocampal m6A level was decreased in PTZ-treated mice when compared with the NS group (t = 3.420, p = 0.0091) (Fig. 1, C). These results suggested that acute seizures induced by PTZ injection could down-regulate the hippocampal m6A level.
Betaine treatment mitigates seizure discharge and neuron overactivity
Betaine, considered as a m6A agonist, could significantly elevate the systemic m6A levels [21]. We investigated the effect of betaine on acute seizures and hippocampal m6A level in PTZ treated mice. One-way ANOVA results showed that the m6A level (F(2,12) = 20.54, p = 0.0001) was significant different in NS, PTZ and Betaine-PTZ group. Bonferroni post-test analysis revealed that betaine (600mg/kg) treatment increased the level of m6A compared to mice with seizures (t = 6.408, p = 0.0001) (Fig. 2, A). Then behavioral tests showed that 600 mg/kg, but not 200 mg/kg, dosage suppressed seizure-like behaviors, as reflected by reduced seizure latency (t = 3.037, p < 0.05) and seizure grade (t = 2.637, p < 0.05), and alleviation of susceptibility score (t = 2.754, p < 0.05) (Fig. 2, B). Moreover, transcription of immediate early genes (IEGs) in neurons is highly sensitive to neuronal activity induced by sensory and behavioral stimuli [22]. Epileptic seizures can induce rapid and dramatic changes in IEGs expression [23], which have been used to estimate the degree of seizure-like activity [24]. We next examined the effect of betaine on the expression of IEGs including c-Fos, Egr1, Arc and NPAS4, at 1 h after PTZ injection. RT-qPCR analysis of hippocampal tissues showed remarkably increasing mRNA levels of c-Fos (t = 6.013, p < 0.0001), Egr1 (t = 5.512, p = 0.0001), Arc (t = 5.168, p = 0.0003) and NPAS4 (t = 5.732, p < 0.0001) in mice injected with PTZ compared to mice injected with NS group, while betaine treatment remarkably decreased the expression of c-Fos (t = 6.742, p < 0.0001), Egr1 (t = 5.052, p = 0.0004), Arc (t = 5.199, p = 0.0003) and NPAS4 (t = 4.931, p = 0.0005) (Fig. 6B, C) (Fig. 2, C). All these results indicated that systematic m6A agonist treatment mitigated PTZ-induced seizure-like behaviors and inhibited neuron overactivity, which might be due to the enhanced hippocampal m6A level.
METTL14-dependent m6A modification participates in acute seizures
As m6A modification is installed by m6A methyltransferases (METTL3, METTL14 and WTAP), and removed by m6A demethylases (FTO and ALKBH5) [25], we used RT-qPCR to detect the relative expression levels of m6A methyltransferases and dimethyl transferases that might lead to altered hippocampal m6A level. We found that only the METTL14 mRNA level was significantly lower in the hippocampus of PTZ-treated mice (t = 2.646, p = 0.0151) (Fig. 3, A). Western blot confirmed a reduced level of METTL14 in the hippocampus of PTZ mice (t = 5.205, p = 0.0004) (Fig. 3, B). Furthermore, we investigated the expression of METTL14 by using immunohistochemistry, and representative images of METTL14-positive cells are shown in Fig. 3, C. The results showed that PTZ stimulation decreased the expression of METTL14 (t = 4.265, p = 0.0053) (Fig. 3, C). Thus, the decreased hippocampal m6A level in mice with acute seizures might be due to the down-regulated METTL14.
METTL14 over-expression in hippocampus of mice attenuates seizure discharge and neuron overactivity
To further explore the role of METTL14 in acute seizures induced by PTZ stimulation, METTL14 over-expression lentivirus was injected into dentate gyrus (DG) using stereotaxic apparatus (Fig. 4, A). RT-qPCR and western blot demonstrated that both the mRNA level (t = 4.971, p < 0.0001) (Fig. 4, B) and the protein level (t = 3.485, p = 0.0059) (Fig. 4, C, D) of METTL14 were up-regulated significantly in METTL14 over-expression lentivirus injection group compared with the control GFP-labeled lentivirus injection group. Furthermore, m6A dot-blot verified METTL14 over-expression could significantly increase the levels of m6A (t = 2.515, p = 0.0361) (Fig. 4, E). Importantly, seizure latency (t = 2.186, p = 0.0388), seizure grade (t = 2.237, p = 0.0358), and susceptibility score (t = 2.550, p = 0.0179) were also ameliorated in acute seizure condition after METTL14 up-regulation (Fig. 4, F, G and H). Similarly, we also examined the effect of METTL14 over-expression on the expression of IEGs. RT-qPCR analysis revealed that METTL14 over-expression could remarkably decrease mRNA levels of c-Fos (t = 5.232, p = 0.0001), Egr1 (t = 3.142, p = 0.0161), Arc (t = 5.007, p = 0.0002) and NPAS4 (t = 4.166, p = 0.0014) induced by PTZ compared to mice injected with control lentivirus (Fig. 4, I). These findings demonstrated that METTL14 over-expression could attenuate PTZ-induced seizure-like behaviors and inhibited neuron overactivity.
YTHDC1-dependent m6A modification participates in acute seizures
m6A modification has been shown affected by m6A reader proteins (YTHDC1, YTHDC2, YTHDF1, YTHDF2 and YTHDF3), which bind to the m6A methylation site to alter gene regulation, such as mRNA splicing, nuclear exporting, translation and decay [26]. We used RT-qPCR to detect mRNA levels of these “readers”, which showed only the mRNA level of YTHDC1 was decreased in the hippocampus of PTZ treated mice (t = 2.693, p = 0.0136) (Fig. 5, A). The protein variation of YTHDC1 was also demonstrated by western blot analysis (t = 13.29, p < 0.0001) (Fig. 5, B and C). In addition, the YTHDC1 protein level was also up-regulated by METTL14 over-expression (t = 6.182, p = 0.0001) (Fig. 5, D). Furthermore, immunohistochemistry assay demonstrated the expression of YTHDC1 was decreased induced by PTZ administration (t = 7.822, p = 0.0002). These results indicated that YTHDC1 may be a key “readers” of m6A in acute seizures.
YTHDC1 over-expression in hippocampus of mice alleviates seizure discharge and neuron overactivity
We further determined the potential role of YTHDC1 in acute seizures. YTHDC1 over-expression lentivirus was injected into dentate gyrus (DG) to promote YTHDC1 expression (Fig. 6, A). RT-qPCR and western blot demonstrated that YTHDC1 over-expression lentivirus injection could up-regulate both the mRNA level (t = 4.971, p < 0.0001) (Fig. 6, B) and the protein level (t = 3.485, p = 0.0059) (Fig. 6, C) of YTHDC1 increased significantly when compared with the control lentivirus. Then behaviour tests revealed that YTHDC1 over-expression could significantly ameliorate seizure latency (t = 2.119, p = 0.0456) (Fig. 6, D), seizure grade (t = 2.853, p = 0.0095) (Fig. 6, E), and susceptibility score (t = 5.097, p < 0.0001) (Fig. 6, F) in acute seizure condition. Additionally, RT-qPCR analysis revealed that the mRNA levels of c-Fos (t = 4.409, p = 0.0006), Egr1 (t = 2.731, p = 0.0357), Arc (t = 4.762, p = 0.0002) and NPAS4 (t = 3.084, p = 0.0137) were significantly decreased in YTHDC1 OE-PTZ mice compared to GFP-PTZ group (Fig. 6, G). Therefore, our data indicated that YTHDC1 over-expression could attenuate PTZ-induced seizure-like behaviors and suppress neuron overactivity.