Atlastin‐1 modulates seizure activity and neuronal excitability

Abstract Epilepsy is a neurological disease, and the main clinical manifestation is recurrent seizures. The exact etiology of epilepsy and the pathogenesis of the disorder are not yet fully understood. Atlastin‐1, a dynamin‐like GTPase, interacts with microtubules and is responsible for vesicle formation, both of which are highly associated with the development of epilepsy. Here, we reported that the expression level of atlastin‐1 protein was reduced in the temporal neocortex of patients with temporal lobe epilepsy and in the hippocampus and adjacent cortex of a pentylenetetrazol‐kindled epileptic mouse model. Cells expressing atlastin‐1 coexpressed the inhibitory synaptic marker GAD67 in the temporal cortex and hippocampus of patients with epilepsy and an epileptic mouse model. The lentivirus‐mediated overexpression of atlastin‐1 protein in the hippocampus of mice suppressed seizure activity in behavioral experiments. Patch‐clamp recordings in the Mg2+‐free epilepsy cell model showed that atlastin‐1 overexpression inhibited neuronal excitability by suppressing the discharge frequency of spontaneous action potentials rather than by changing the passive and active properties of action potentials. Inhibitory synaptic transmission, but not excitatory synaptic currents, increased after atlastin‐1 overexpression. These findings suggest that atlastin‐1 likely contributes to the occurrence and development of epilepsy through inhibitory synaptic transmission.


K E Y W O R D S
epilepsy, human patients, inhibitory synaptic transmission, neuronal excitability, seizure development. It is suggested that the dysregulation of microtubules and vesicle transport facilitates the occurrence and spread of epileptic discharges. [11][12][13] Nevertheless, the role of atlastin-1 in the development of epilepsy is largely unknown.
In the present study, we explored the expression of atlastin-1 protein in patients with epilepsy and a mouse model. Furthermore, we investigated behavioral and electrophysiological changes after the lentivirus (LV)-mediated overexpression of atlastin-1 in the hippocampus of mice to explore the possible roles of atlastin-1 in epilepsy.

| Ethics statement
All study protocols involving human subjects were approved by the Commission of Chongqing Medical University according to the guidelines of the Declaration of Helsinki. Informed consent for using human brain tissue was obtained from human subjects. All animal protocols for treatments were conducted according to the Commission of Chongqing Medical University for Ethics in Animal Experiments.

| Human tissues
Temporal cortical tissues from 18 patients with epilepsy and 12 control individuals were obtained randomly from our human brain tissue bank (Table 1-3). 11,14,15 Patients with epilepsy were diagnosed as temporal lobe epilepsy (TLE), as determined based on the 2001 seizures and were included in the number of seizures. The days until full kindling were recorded as the latency period.

| Western blot analysis
Frozen human brain tissues and the hippocampus of PTZ-kindled mice and control mice were collected and extracted for total protein analysis using a RIPA protein extraction kit (Beyotime Biotechnology). Protein was then boiled at 95°C for 5 minutes following the calculation of protein concentrations by an enhanced bicinchoninic acid (BCA) protein assay kit (Beyotime Institute of Biotechnology). Protein was separated on 10% SDS-PAGE gels, and the separated target proteins were immunoblotted with primary antibodies, atlastin-1 (rabbit, Abcam), and GADPH (rabbit, Proteintech), overnight at 4°C. The blots were then incubated with HRP-conjugated secondary antibodies. The blots were scanned and quantified using a fusion imaging system.

| Immunofluorescence labeling
Fresh human and animal brain tissues were postfixed for 24 hours at 4°C. Floating slices (10 μm in thickness) were permeabilized with 0.5% Triton X-100 for 30 minutes at room temperature and then incubated and blocked in goat serum. The slices were incubated with primary antibodies, atlastin-1 (rabbit, Abcam), GAD67 (mouse, SYSY), and MAP2 (guinea pig, SYSY) overnight at 4°C followed by incubation with secondary fluorescent-conjugated antibodies for 2 hours at room temperature. Images were captured by laser-scanning confocal microscopy (Nikon A1 + R Microsystems) on an Olympus IX 70 inverted microscope (Olympus). Areas of fluorescence intensity were analyzed using Image Pro Plus 6.0. The injected mice were allowed to fully recover for 2 weeks to prevent infection, and recovery was followed by PTZ injection for behavioral observation.

| Patch-clamp recordings
Brain slices of adult C57/BL6 male mice were obtained as previously reported. 19,20 Briefly, 2 weeks after the intrahippocampal injection of LV, mice were anesthetized with pentobarbital (80 mg/kg, intraperitoneal). Brain slices (300 μm) were prepared with a Leica Glass microelectrodes (Sutter) were prepared by a pipette puller to record a final resistance of 3-5 MΩ when the pipette was filled with an internal solution. A multiclamp 700B amplifier (Axon) was used in whole-cell recording. 21 The obtained signals were sampled at 10 kHz and filtered at 2 kHz. A stable baseline was obtained for at least 5 minutes prior to recording. The access resistance was checked intermittently in current-clamp mode, and the data were abandoned if the access resistance (15)(16)(17)(18)(19)(20)   Paired-pulse ratios (PPRs) recordings were voltage-clamped at −70 mV using a paired-pulse protocol of two stimuli at an interpulse interval of 50 ms PPRs were calculated by the ratio of the second peak amplitude to the first peak amplitude.

| Statistical analysis
Student's t-test was used for independent-samples. Repeated measures ANOVA followed by post hoc t-tests were used to compare the differences in the daily seizure scores of PTZ-kindled mice ( Figure 2C).
The chi-square test was used to compare gender differences between TLE patients and controls. All data are presented as the mean ± SD P < .05 and P < .01 were considered statistically significance. n indicates the number of mice/slices or independent experiments. All samples included in each experiment were analyzed in triplicate. All graphs were prepared using GraphPad Prism 4 software.

| Reduced atlastin-1 protein in TLE patients
The expression patterns of atlastin-1 protein in the temporal lobe neocortex of patients with TLE and controls were detected by western blot and immunofluorescence labeling. The expression level of atlastin-1 protein in TLE patients was significantly lower than that in controls according to western blot ( Figure 1A). By immunofluorescence, atlastin-1 protein was mainly found in the neuronal cytoplasm. Cells in the temporal cortices of patients with epilepsy expressing atlastin-1 coexpressed the inhibitory synaptic marker GAD67 ( Figure 1B).
Compared with controls, atlastin-1 protein in TLE patients was found to have a low fluorescence intensity in quantitative immunofluorescence analysis ( Figure 1B).

| Decreased atlastin-1 protein in an epileptic mouse model
Atlastin-1 protein in the PTZ-kindled epileptic mouse model was further analyzed by Western blot and quantitative immunofluorescence. The expression results in the mouse model were similar to the expression results in humans. Atlastin-1 protein levels were significantly lower in both the neocortex ( Figure 1C) and hippocampus ( Figure 1E) of PTZ-kindled mice than in the corresponding regions in controls. Additionally, in quantitative immunofluorescence analysis, weak green staining was detected for atlastin-1 protein in the cortex ( Figure 1D) and hippocampus ( Figure 1F) of PTZ-kindled mice, whereas strong staining for atlastin-1 was detected in the corresponding regions of controls. In addition, atlastin-1 protein was mainly colocalized with the inhibitory synaptic marker GAD67 in the cortex and hippocampus of epileptic mice ( Figure 1D,F).

| Identify atlastin-1 overexpression in the mouse hippocampus
To

| Hippocampal overexpression of atlastin-1 protein suppressed seizure activity in a PTZ-induced epileptic mouse model
To further explore the potential relationship between atlastin-1 protein and the pathology of epilepsy, behavioral effects related to atlastin-1 protein were detected in the epileptic mouse model.
In PTZ-induced epileptic mice, behavioral observation showed that the daily seizure scores across 30 days were lower in atlastin-1 overexpressing mice than in controls ( Figure 2C). Compared with the controls, atlastin-1 overexpressing mice exhibited a lower total number of seizures ( Figure 2D) and a longer induced latency ( Figure 2E). These behavioral results suggested that atlastin-1 reduced epileptic seizure activity through the regulation of the severity, frequency, and latency of seizures.

| Atlastin-1 inhibited neural excitability in the Mg 2+ -free epilepsy cell model
To investigate the role of atlastin-1 in neural electrophysiology, wholecell patch clamp was performed in pyramidal neurons of the hippocampal CA3 area in Mg 2+ -free brain slices. We first studied the effect of atlastin-1 on the passive and active properties of action potentials (APs). 22,23 There were no significant differences in resting membrane potential, input resistance, AP duration, AP overshoot, or afterhyperpolarization between the two groups ( Figure 3A). However, the discharge frequency of spontaneous APs in the atlastin-1 overexpressing group was lower than that in the control group ( Figure 3B).
To further investigate whether atlastin-1-mediated neuronal hypoexcitability is an imbalance between the excitatory and (mIPSCs) were recorded. 24,25 In pyramidal neurons in the CA3 region of the hippocampus, the amplitude nor the frequency of mEPSCs was significantly changed between the atlastin-1 overexpressing group and the control group ( Figure C). However, the mIPSC amplitude, but not the frequency, was increased in atlastin-1 overexpressing mice compared with controls ( Figure 3D). We then analyzed PPRs to further explore the impact of atlastin-1 on presynaptic vesicle release. 26 There was no significant difference in the PPRs between atlastin-1 overexpressing mice and controls ( Figure 3E).

| D ISCUSS I ON
Our study first identified that the expression level of atlastin-1 protein was decreased in brain tissues from patients with epilepsy and Epilepsy is a chronic brain disease, and the mechanisms are mainly focused in the brain, which then triggers a secondary cascade of various pathology events. 1,2 Thus, brain tissue from humans is the most valuable and reliable material with which to explore clinical-pathological alterations in epilepsy. In the present study, we first collected the temporal lobe cortex of patients with TLE to verify that atlastin-1 protein was significantly reduced in TLE patients. Thus, atlastin-1 protein may be closely associated with epilepsy.
Due to ethical requirements, we could not acquire normal hippocampal tissue from humans; thus, relatively normal brain tissue from patients with increased intracranial pressure due to head trauma was obtained as the control. Given that we could not equally compare the expression level of atlastin-1 between the normal hippocampus and the pathological hippocampus of TLE, we detected the atlastin-1 expression level in an epileptic mouse model to provide direct evidence that the atlastin-1 levels in the hippocampus decreased during seizure activity. Notably, atlastin-1 was expressed at significantly lower levels in the hippocampus and adjacent temporal cortex of the epileptic mice than in controls. These results further indicated that the abnormal expression of atlastin-1 protein in the brain may be involved in epileptogenesis.
The expression results suggested that decreased atlastin-1 protein levels might play a protective role against seizure activity. For a more in-depth analysis, the potential causal relationship between atlastin-1 proteins and epilepsy was determined. We used an in vivo intervention utilizing lentivirus to overexpress atlastin-1 protein in the may not be involved in the atlastin-1-mediated changes in neuronal excitability. 22,23 mIPSCs are supposed to indicate gamma-aminobutyric acid (GABA)-mediated inhibitory synaptic transmission; in contrast, mEPSCs reflect glutamate-mediated excitatory synaptic transmission. 32 We further investigated the role of atlastin-1 on excitability and inhibitory balance in pyramidal neurons. Without influencing mEPSCs, atlastin-1 highly reduced neural excitability due to increased mIPSC currents. In addition, atlastin-1 only reduced mIPSC amplitudes, but not frequencies, and it had no effect on the PPR.
The frequency of postsynaptic currents reflects the number of presynaptic neurotransmitter vesicles released in a unit of time, and the amplitude of postsynaptic currents indicates the effect on the postsynaptic receptor. 33 The PPR is also a reliable index of the contribution of presynaptic neurotransmitter vesicles release to postsynaptic currents. 26 There was no significant difference in mIPSC frequency and PPR after atlastin-1 overexpression, indicating that presynaptic vesicle release might not be responsible for the atlastin-1-mediated changes in mIPSCs. Atlastin-1-mediated changes in synaptic currents might be controlled by postsynaptic GABA receptors rather than presynaptic GABA neurotransmitter release. In addition, atlastin-1 colocalized with the inhibitory synaptic marker GAD67 in the temporal cortex of TLE patients and PTZ-kindled mice, further indicating its role in inhibitory synaptic transmission.
The atlastin-1 protein is encoded by the SPG3A gene which is often found to be mutated in hereditary spastic paraplegia (HSP). HSP is a group of inherited neurodegenerative disorder characterized by progressive spasticity and weakness in the lower limbs. Complex forms of HSP exhibit additional symptoms such as seizures. 34 Atlastin-1 is enriched in the adult brain, but is also present at much lower levels in some other human tissues such as smooth muscle, adrenal gland, testis, lung, and kidney. In the brain, atlastin-1 is widely expressed in the pyramidal neurons in lamina V of the cerebral cortex and in the CA1 and CA3 areas of the hippocampus, 4 which are the most important brain regions associated with the pathogenesis of experimental epilepsy. 35 As a dynamin-like GTPase, atlastin regulates the stability of microtubules in Drosophila. 5 Recent studies have identified that atlastin-1 is implicated in vesicle budding from ER-derived microsomes and in the vesicle transport step of the ER-Golgi interface. 10 Notably, as shown in previous studies, microtubule and vesicle formation are essential for the transportation of vesicles containing GABA receptors from ER-Golgi to the plasma membrane in epilepsy. 36,37 Thus, we speculate that atlastin-1 might be implicated in GABA receptor-related vesicle transport in epilepsy, thereby increasing GABA receptor-mediated inhibitory synaptic transmission and causing neural hypoexcitation.
In conclusion, we reported for the first time that atlastin-1 plays an important role in epileptogenesis by modulating seizure activity and neuronal excitability. However, the mechanism underlying the atlastin-1 change in inhibitory synaptic input should be further investigated.

ACK N OWLED G M ENT
This work was supported by grants from the National Natural Science Foundation of China (No. 81301110).

CO N FLI C T S O F I NTE R E S T
The authors declare no competing interests.