Molecular expression and functional analysis of genes in children with temporal lobe epilepsy

Temporal lobe epilepsy is the most common form of epilepsy. However, for this type of condition, antiseizure medication is not effective for children. As miRNAs are involved in the development of temporal lobe epilepsy in children, they may provide potential therapeutic ap-proaches for treatment. The primary aim of this study was to explore the expression and function of miR-135a-5p in children with temporal lobe epilepsy. Hippocampal slices from either normal (control) children or children with temporal lobe epilepsy were used to detect the expression of miR-135a-5p and its target gene caspase activity and apoptosis inhibitor 1. To further explore the role of miR-135a-5p in the development of temporal lobe epilepsy in children, primary hippocampal neurons from newborn rats were cultured in vitro in a magnesium-free medium to mimic the temporal lobe epilepsy condition in children. The effect of transfection of miR-135a-5p inhibitor into cells was also assessed. Apoptosis and proliferation of hippocampus cells was respectively assessed by flow cytome-try or 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay. The level of miR-135a-5p was significantly increased in both children with temporal lobe epilepsy and the epileptiform discharge model that employed newborn rat hippocampal neurons; whereas, the expression of caspase activity and apoptosis inhibitor 1 was down-regulated by overexpression of miR-135a-5p. Moreover, miR-135a-5p mediated the pro-apoptotic effect of temporal lobe epilepsy via repressing caspase activity and apoptosis inhibitor 1 expression. Additionally, miR-135a-5p reduced cell survival in the temporal lobe epilepsy condition. Overexpression of miR-135a-5p induced cell apoptosis through inhibition of caspase activity and apoptosis inhibitor 1 expression and suppressed cell survival in children with temporal lobe epilepsy.

Temporal lobe epilepsy is the most common form of epilepsy. However, for this type of condition, antiseizure medication is not effective for children. As miRNAs are involved in the development of temporal lobe epilepsy in children, they may provide potential therapeutic approaches for treatment. The primary aim of this study was to explore the expression and function of miR-135a-5p in children with temporal lobe epilepsy. Hippocampal slices from either normal (control) children or children with temporal lobe epilepsy were used to detect the expression of miR-135a-5p and its target gene caspase activity and apoptosis inhibitor 1. To further explore the role of miR-135a-5p in the development of temporal lobe epilepsy in children, primary hippocampal neurons from newborn rats were cultured in vitro in a magnesium-free medium to mimic the temporal lobe epilepsy condition in children. The effect of transfection of miR-135a-5p inhibitor into cells was also assessed. Apoptosis and proliferation of hippocampus cells was respectively assessed by flow cytometry or 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay. The level of miR-135a-5p was significantly increased in both children with temporal lobe epilepsy and the epileptiform discharge model that employed newborn rat hippocampal neurons; whereas, the expression of caspase activity and apoptosis inhibitor 1 was downregulated by overexpression of miR-135a-5p. Moreover, miR-135a-5p mediated the pro-apoptotic effect of temporal lobe epilepsy via repressing caspase activity and apoptosis inhibitor 1 expression. Additionally, miR-135a-5p reduced cell survival in the temporal lobe epilepsy condition. Overexpression of miR-135a-5p induced cell apoptosis through inhibition of caspase activity and apoptosis inhibitor 1 expression and suppressed cell survival in children with temporal lobe epilepsy.

Introduction
Epilepsy is a chronic disorder of the nervous system characterized by recurring seizures due to abnormal neuronal excitability (Henshall et al., 2014). It is estimated that the incidence of epilepsy in children ranges from 33 to 82 per 100,000 children per year (Nickels et al., 2011). Temporal lobe epilepsy (TLE) is the most common form of epilepsy with focal seizures originating in the temporal lobe (Thom, 2004). Unfortunately, children with TLE are often resistant to a single antiseizure medication and they need to take two or more appropriate medications (Nickels et al., 2011). However, many children still have seizures after treatment with a single antiseizure medication (Nickels et al., 2011). Given the unsuccessful medication trials, surgical treatment is an alternative. However, it is dependent on exact identification of the ictal-onset zone (Nickels et al., 2011;Gleissner et al., 2005). Thus, efforts to identify novel therapeutic targets for the treatment of TLE are currently ongoing.
miRNAs are a class of short non-coding RNAs that posttranscriptionally regulate gene expression by either mRNA degradation or translational inhibition of target mRNAs through imperfectly pairing of the 3'-untranslated regions (3'-UTR), with the latter being involved in various biological functions including cell proliferation, cell death, development, and metabolism (Bartel, 2009). Recently, numerous studies have indicated a potential role for miRNAs in the development and treatment of TLE both in animal models and in humans, including children. For instance, the implication of miR-146a, miR-221, and miR-222 in the inflammatory response occurring in TLE has been demonstrated (He et al., 2016;Aronica et al., 2010;Ashhab et al., 2013a,b). miRNAs can govern the process of TLE by negatively regulating TLE-related gene expression levels (Henshall et al., 2014;He et al., 2016;Kan et al., 2012;McKiernan et al., 2012).
A series of studies have investigated miRNA changes in human TLE McKiernan et al., 2012). A recent study revealed upregulation of brain-specific miR-135a-5p in adults with TLE . However, the expression of miR-135a-5p in children with TLE is unclear as TLE in children is different from the syndrome observed in adults (Nickels et al., 2011). Furthermore, the role of miR-135a-5p in TLE is still unknown. Previous studies have identified that TLE induces apopto-sis in neurons (Henshall and Simon, 2005;Henshall, 2007). More importantly, the miR-135a-5p-targeted gene CAAP1 predicted by TargetScan and miRanda is the inhibitor of a mitochondrial apoptosis pathway (Zhang et al., 2011). This suggests that miR-135a-5p may be involved in TLE-mediated apoptosis through regulating CAAP1. The primary aim of this study was to investigate the expression and function of miR-135a-5p in children with TLE.

Subjects
Frozen tissue samples from brain donors diagnosed with TLE (n = 15, age 11.2 ± 2.6 years, onset age 7.4 ± 1.3 years) and control individuals (n = 15, age 10.6 ± 3.7 years) were analyzed. Samples were stored at -80 • C. Three slices (400 µm thick) of hippocampus Area CA3 from TLE subjects or normal control subjects were collected by surgical excision of the epileptic focus using a manual tissue chopper. All samples were supplied by the Department of Neurosurgery, Beijing Tiantan Hospital, Capital Medical University. The study was approved by the Ethics Committee of the Capital Medical University (No.KY2015-006-01). Written consents were collected from all subjects before the study.

Primary culture of hippocampal neurons from newborn rats
Wistar rats were purchased from the animal center of Hubei College of Traditional Chinese Medicine and housed under standard laboratory condition. One male rat was housed with a female rat. Primary hippocampal cells from 100 P0 newborn rats were cultured as described previously (Nunez et al., 2008). Briefly, three slices of hippocampus were obtained from the brain and dissociated to single cells after digestion with 0.5% Trypsin at 37 • C for 15 minutes. Cells were then washed twice with Hank's buffered saline solution and treated with DNase I. Neurons were placed on glass coverslips with plating medium in a humidified 37 • C incubator (5% CO 2 ). After confirming cell viability and that they had attached, cells were cultured with Neurobasal medium. One-third of the medium was replaced with fresh Neurobasal medium once a week.

Establishment of epileptiform discharge model in rat hippocampal neurons
To mimic TLE condition in children, primary hippocampal neurons from newborn rats were cultured in magnesium-free medium (145 g NaCl, 2.5 g KCl, 10 g HEPES, 2 g CaCl 2 , 10 g glucose, 0.002 g glycine dissolved in 1000 mL distilled water, pH 7.2) at 37 • C for three hours, after which hippocampal epileptiform activity was induced (Lewis et al., 1990;Sombati and Delorenzo, 1995).

Real-Time PCR
miRNA and mRNA were isolated from human brain tissue and cultured rat hippocampal neurons, respectively, by using Trizol reagent (Invitrogen, Carlsbad, CA, USA) and respectively reverse transcribed by either the qScript TM microRNA cDNA Synthesis Kit for miRNA (Quanta BioSciences, Gaithersburg, MD, USA) or the QuantiTect Reverse Transcription Kit for mRNA (Qiagen, Valencia, CA, USA). Real-time PCR was performed using the Maxima SYBR Green/ROX qPCR Master Mix assay (Thermo Fisher Scientific, Rockford, IL, USA) with GAPDH or U6 as internal controls in the StepOnePlus system (Applied Biosystem, Foster City, CA, USA). Following amplification in triplicate for each sample, Ct values were acquired with manual thresholds by 7500 System SDS software (Applied Biosystem). Relative expression levels were normalized to the expression level of GAPDH or U6 according to the ∆∆CT method. The expression levels between different groups were compared using 2 −∆∆Ct , while p-values ≤ 0.05 were considered statistically significant. Primer sequences were: miR-135a-5p: 5'-TCACATAGGAATAAAAAGCCATA-3'; Caap1 F: 5'-CTGTTCCTCCACCAGAAGATG-3'; Caap1 R: 5'-ACCAGCTTTCATTAGGGCTTTA-3'.

Western blotting
Western blotting was performed as previously described (Zhong et al., 2014). Human brain tissues and cultured rat primary hippocampal neurons were sonicated in ice-cold lysis buffer according to the manufacturer's instruction (Cell Signaling Technology, Danvers, MA, USA) with a protease inhibitor cocktail (Sigma-Aldrich, St. Louis, MO, USA) to extract protein. Equal amounts of protein (30 µg) from each group or the Precision Plus Protein Standard used for marker (Bio-Rad Laboratories, Hercules, CA, USA) were resolved by SDS-PAGE, transferred onto PVDF membranes and immunoblotted with primary antibodies for CAAP1 (human)/Caap1 (rat) and GAPDH at 1:1000 dilutions in 5% nonfat milk. CAAP1 antibody (NBP1-86644) was obtained from Novus Biologicals (Littleton, CO, USA). HRP-conjugated goat anti-rabbit (Jackson ImmunoResearch Laboratories, West Grove, PA, USA) secondary antibody was used to detect primary antibody. The intensities of target protein bands were assessed by densitometry and normalized to those of GAPDH (Abcam). The signals were detected by the SuperSignal West Femto Maximum Sensitivity Substrate kits (Thermo Fisher Scientific) and the densitometric intensities were determined using Image J.

Transfection
miR-135a-5p inhibitor was synthesized by Life Technologies (Grand Island, NY, USA). Cultured rat primary hippocampal neurons were transfected at a confluency of 85-90% on a 96 well plate. Prior to transfection, regular medium was replaced with serum-free Opti-MEM ® I medium. 5 µL of 100 µM miR-135a-5p inhibitor and 5 µL Lipofectamine T M RNAiMAX (Invitrogen, Carlsbad, CA, USA) were added into 50 µL Opti-MEM ® I medium and incubated for five minutes at room temperature. Cells were then transfected with miR-135a-5p inhibitor mixed with Lipofectamine T M RNAiMAX and incubated at 37 • C for six hours. After transfection, Opti-MEM ® I medium was replaced with regular medium. After 24 hours cells were cultured in a magnesiumfree medium at 37 • C for three hours. The sequence of miR-135a-5p inhibitor was: 5'-UCACAUAGGAAUAAAAAGCCAUA-3'.

Flow cytometric apoptosis assay
1 × 10 6 cultured rat primary hippocampal neurons were harvested from each group and washed by incubation buffer (10 mmol/L HEPES/NaOH, pH 7.4, 140 mmol/L NaCl, 5 mmol/L CaCl 2 ). Hippocampal neurons were then suspended in 100 µL PBS with 1 µg Annexin V-FITC (Becton Dickinson, Franklin Lakes, NJ, USA) and Propidium iodide (PI) (Thermo Fisher Scientific), followed by dark incubation at room temperature for 10-15 minutes. After a single wash with incubation buffer, SA-FLOUS was added to each sample and dark incubated at 4 • C for 20 minutes before analysis by flow cytometry.

MTT assay
A MTT assay kit (Abcam, Cambridge, MA, USA) was used to detect the survival of cultured rat primary hippocampal neurons . Medium was discarded from the cell cultures in the 96 well plate. Then 50 µL of serum-free media and 50 µL of MTT solution were added into each well and plates were incubated at 37 • C for three hours. After incubation, 150 µL of MTT solvent was added into each well. The plate was shaken on an orbital shaker for 15 minutes and absorbance was measured at OD = 590 nm within one hour after shaking.

Statistical Analysis
Data is reported as mean ± standard error (±SE). Statistical differences were determined by use of the Student's t-test between two groups and two-way ANOVA between multiple groups. After the two-way ANOVA analysis, a Tukey's multiple-comparison test was conducted to estimate the significance of the differences. All statistical analyses were conducted by the SigmaStat 3.5 software. A p-value < 0.05 was considered statistically significant.

miR-135a-5p expression is upregulated in the hippocampus of children with TLE
The clinical characteristics of brain donors are summarized in Table. 1. Compared to the control group (n = 15), miR-135a-5p expression was significantly increased in the hippocampus of 80% of the children diagnosed with TLE (n = 15) detected by qPCR (Fig. 1A). In contrast, as the predicted target gene of miR-135a-5p (Fig. 1B), CAAP1 gene expression in hippocampal neurons from 87% of the children with TLE (n = 15) was decreased compared to that in control group ( Fig. 1C and Fig. 1D). These data suggested that miR-135a-5p was also upregulated in children with TLE, as previously reported for adult cases . Moreover, the upregulation of miR-135a-5p suppressed the expression of its target gene CAAP1 in TLE. However, the expression of miR-135a-5p and CAAP1 were not significantly changed in all children with TLE, suggesting that other genes may be involved in TLE formation.

miR-135a-5p expression is induced in cultured rat hippocampal neurons that displayed epileptiform activity
To mimic the TLE condition in children and to further investigate the expression of miR-135a-5p and Caap1 in vitro, primary hippocampal cells from newborn rats were cultured. After three hours of treatment in magnesium-free medium, the expression of miR-135a-5p was increased significantly compared to that of control hippocampal cells ( Fig. 2A). Moreover, the overexpression of miR-135a-5p repressed the expression of Caap1 in hip- pocampal neurons cultured in magnesium-free medium ( Fig. 2B and Fig. 2C). These data suggested that miR-135a-5p expression was upregulated, whereas, Caap1 expression was decreased both in vivo and in vitro under the TLE condition.

miR-135a-5p induces TLE-like hippocampal cell apoptosis
To determine whether miR-135a-5p mediates the pro-apoptotic effect of hippocampal cells in children with TLE, the miR-135a-5p inhibitor was employed. In cultured hippocampal neurons apoptosis was detected by flow cytometry. Treatment of magnesiumfree medium dramatically triggered cell apoptosis compared to that in the control group (Fig. 3Aand Fig. 3B). However, transfection of miR-135a-5p inhibitor into cells cultured in magnesiumfree medium resulted in a significantly lower number of apoptotic cells (Fig. 3A and Fig. 3B). Additionally, treatment by miR-135a-5p inhibitor did not affect apoptosis of control cells (Fig. 3A and Fig. 3B). Moreover, the addition of miR-135a-5p inhibitor to cells treated with magnesium-free medium prevented the expression of Caap1 inhibited by the overexpression of miR-135a-5p ( Fig. 4A and Fig. 4B). These data collectively suggest that miR-135a-5p mediates TLE-induced apoptosis of hippocampal cells through suppression of Caap1 expression.

miR-135a-5p inhibits TLE-like hippocampal cell survival
To further explore the role of miR-135a-5p in children with TLE, a MTT assay was employed to detect the proliferation of hippocampal neurons. After culture in magnesium-free medium, cell survival assessed by cell metabolic activity was reduced compared to that of control cells. Conversely, transfection by the miR-135a-5p inhibitor significantly increased cell survival (Fig. 5). Moreover, the addition of miR-135a-5p inhibitor did not promote cell survival under normal condition (Fig. 5). The above data suggests that miR-135a-5p was also involved in the regulation of neuron survival in TLE.

Discussion
Epilepsy, including TLE, is characterized by unprovoked seizures (Henshall et al., 2014;Thom, 2004). Mounting evidence from animal models and human patients including children, have shown that seizures trigger apoptosis in neurons and lead to conse-quent brain injury and epileptogenesis (Henshall and Simon, 2005;Henshall, 2007;Yamamoto et al., 2006). Defining the mechanisms that underly seizure-induced neuronal apoptosis in TLE may be helpful for identifying novel strategies to protect the brain against seizures.
In this study, miR-135a-5p was found to be upregulated in children with TLE. Recent studies have indicated that miR-135a-5p is a novel apoptosis inducer. For example, in malignant glioma, miR-135a-5p is capable of inducing apoptosis (Wu et al., 2012;Zhang et al., 2016). Similarly, miR-135a-5p plays a key role on apoptosis in ovarian cancer cells (Tang et al., 2013). Furthermore, miR-135a-5p enhances hydrogen peroxide-induced apoptosis in rat cardiomyoblast (Liu et al., 2017). Additionally, inhibition of miR-135a-5p protects A549 cells from apoptosis (Zhao et al., 2014). In the aforementioned studies, overexpression of miR-135a-5p is essential to trigger apoptosis and they all indicate that miR-135a-5p mediates neuronal cell apoptosis that could cause TLE in children. Indeed, results from this study provide further evidence in support of this hypothesis by mimicking the TLE condition seen in children in an in vitro hippocampal cell model based on newborn rats. Apoptosis can be activated by intrinsic pathways through disruption of intracellular organelle function, such as by mitochondrial dysfunction, or by an extrinsic pathway via activation of death receptors (Danial and Korsmeyer, 2004). Both of these pathways have been found to be triggered in TLE (Henshall, 2007). However, several studies have demonstrated that miR-135a-5p is solely involved in a mitochondria-dependent apoptosis pathway (Wu et al., 2012;Liu et al., 2017;Zhao et al., 2014), which suggests miR-135a-5p could regulate the intrinsic pathway of apoptosis in children with TLE. Although previous studies have shown that miR-135a-5p initiates apoptosis by downregulating expression of pro-apoptotic Bcl-2 family members, including Bcl-2 and Bcl-xL (Zhao et al., 2014;Navarro et al., 2009), a novel target gene of miR-135a-5p was identified in this study and its expressed protein is also an anti-apoptotic protein. A previous report has suggested that Caap1 modulates a mitochondrial apoptosis pathway in a caspase-10 dependent manner (Zhang et al., 2011). The present study indicates that CAAP1 expression was decreased in children with TLE and inhibition of miR-135a-5p restores the mRNA level of Caap1 in cultured rat hippocampal neurons that displayed epileptiform activity while the miR-135a-5p-induced apoptosis was suppressed.   Experiments repeated three times (n = 3). No Mg 2+ : magnesium-free medium. **p < 0.05 vs. Control group; ## p < 0.05 vs Control+miR-135a-5p inhibitor group; ∆∆ p < 0.05 vs. No Mg 2+ group.
This suggests that CAAP1 may be involved in cell death occurring in children with TLE.
Although CAAP1 is described as a novel apoptosis inhibitor, it may play other key roles in children with TLE. Apart from reducing apoptosis, inhibition of miR-135a-5p also improves cultured rat hippocampal neuron survival after treatment in a magnesiumfree medium while the expression of Caap1 is restored. Previous study has indicated that CAAP1 is able to interact with acetylase KAT7 (Stelzl et al., 2005), which is crucial for DNA replication (Iizuka et al., 2006). Several studies have shown that KAT7 interacts with proteins to positively regulate cell survival (Pardo et al., 2017;Santos et al., 2018;Newman et al., 2017). For example, KAT7 is essential for T-cell survival through maintaining global histone-H3 lysine 14 acetylation (H3K14ac) (Newman et al., 2017). Moreover, KAT7 promotes embryonic stem cell survival via interaction with a novel binding partner Niam (Pardo et al., 2017). In children with TLE, Caap1 may act as a chaperon of KAT7 to modulate neuron cell survival.
To the best of the authors knowledge, this is the first investigation of the role of miR-135a-5p in TLE, which could be significant for the development of novel therapeutic treatments for those with this condition. However, pre-clinical animal experiments are required to further reveal the underlying mechanism of how CAAP1 may suppress TLE-induced neuronal cell apoptosis.

Summary
In summary, the expression of miR-135a-5p is increased in both children with TLE and TLE-like rat hippocampus neurons. Moreover, upregulation of miR-135a-5p initiates TLE-like hippocampus rat neuron cell apoptosis through suppression of Caap1 expression in vitro, which implies it may have a similar effect on the survival of hippocampal neurons in children with TLE.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Authors' contributions
XJW and BQG designed the study. YJW, ZRS and SCR collected the data, WLY and YXD analyzed the data, CXT and YZY analyzed the results and drafted the manuscript.