Following ethical approval and the provision of written informed consent, serum samples were collected from three seizure patients for miRNA microarray analysis and from another 90 patients for confirmation of the miRNA expression. The patients were evaluated by an epileptologist and underwent a basal EEG and 3T MR brain imaging. Patients experiencing symptoms and EEG semiology were considered to be at seizure onset. Following treatment, when the symptoms and EEG semiology tended to be normal, patients were considered to be post-seizure. All samples were collected from consenting individuals from the Liaocheng People's Hospital, Taishan Medical College (Liaocheng, China) and Jinan University (Shenzhen, China). The sample collection was performed according to the protocols approved by the Ethics Committee of Jinan University. The blood samples were collected at seizure onset and post-seizure.

Venous blood (5 ml) was collected and stored at room temperature for 1 h, followed by centrifugation at 1,600 × g for 10 min at 4°C. The serum was gently collected and stored at −80°C or used in the following experiment. For all experiments, 500 µl of human serum was used and total RNA was extracted using TRIzol reagent (Thermo Fisher Scientific, Inc., Waltham, MA, USA). The concentration and the 260/280 nm absorbance ratio was detected using a Nanodrop 2000 (Thermo Fisher Scientific, Inc., Pittsburgh, PA, USA).

Total RNA (100 ng) was used to synthesize first strand cDNA at 42°C for 2 h using ArrayScript™ reverse transcriptase, second strand cDNA was at 16°C for 1 h, 65°C for 10 min using reverse transcriptase and DNA polymerase, respectively. T7 Biotin IVT Mix was used to generate multiple copies of biotin-modified aRNA from the double stranded cDNA templates, incubated for 2–14 h at 40°C (MessageAmp™ Premier RNA Amplification kit; Thermo Fisher Scientific, Inc.). miRNAs were individually detected using specific oligonucleotides (Takara Bio, Inc.). A single miRNA-specific oligonucleotide was designed against each mature miRNA sequence, and miRNA-specific primers were extended using DNA polymerase. Universal primers were used to amplify the cDNA templates and the primer complimentary to the array was fluorescently labeled using miRCURY Hy3/Hy5 Power Labeling kit (Takara Bio Inc.). The labeled, single-stranded PCR products were hybridized to a Human v2.0 miRNA Expression BeadChip (Illumina, Inc., San Diego, CA) with 1,146 human miRNAs (97% coverage in the miRBase 12.0 database) (14,15).

RT for the individual qPCR analyses was performed using 500 ng of total RNA and a Reverse Transcription kit (Promega Corporation, Madison, WI, USA). RT-specific primers for human miRNAs miR-378, miR-30a, miR-106b and miR-15a (Applied Biosystems; Thermo Fisher Scientific, Inc.) were used for all miRNA RT procedures. Individual qPCR analyses were performed on the 7900HT Fast Realtime system (Applied Biosystems; Thermo Fisher Scientific, Inc.) using miR-378-, miR-30a-, miR-106b- and miR-15a-specific Taqman miRNA assays (Applied Biosystems Thermo Fisher Scientific, Inc.). RNU6B was used for the normalization of miRNA expression levels. The mRNA levels of calcium/calmodulin-dependent protein kinase type IV (CAMK4), were determined using TaqMan Real-Time PCR analysis, according to manufacturer's protocols (Applied Biosystems; Thermo Fisher Scientific, Inc.). To determine mRNA expression levels, total RNA (10 ng) were reverse transcribed using iScript Reverse Transcription Supermix (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The cDNA templates were amplified with TaqMan PreAmp Master mix (Applied Biosystems; Thermo Fisher Scientific, Inc.). The PCR programs consisted in a hot start of 95°C for 10 min, followed by 40 cycles of 15 sec at 95°C and 1 min at 60°C. The primers used were: CAMK4, sense 5′-AGCACATTCAAACCACCACA-3′, antisense 5′-GGACTCAGAGATCCGTCTGC-3′; miR-30a, sense 5′-TGTAAACATCCTCGAC-3′, antisense 5′-ACATCCAGTGTAGCATA-3′; miR-378, sense 5′-ACACTCCAGCTGGGACTGGACTTGGAGTC-3′, antisense 5′-TGGTGTCGGAGTCG-3′; miR-106a, sence 5′-CGGAATTCATCTCGAGACGCCAACTTG-3′, antisense 5′-CGGGATCCCTTCATTCAAGGTCAATGAG-3′; miR-15a, sense 5′- GTCGTATCCAGTGCAGGGTCCGAGGTATTC-3′, antisense 5′-GCACTGGATACGACCACAAA-3′; RNU6B, sense 5′-CTCGCTTCGGCAGCACA-3′, antisense 5′-AACGCTTCACGAATTTGCGT-3′; and GAPDH, 5′-AAGGTCGGAGTCAACGGATT-3′, antisense 5′-CTGGAAGATGGTGATGGGATT-3′. Each sample was analyzed in triplicate, in three independent experiments. The level of each mRNA was measured using the quantification cycle (Cq) and the level of the target was calculated as described above for the miRNAs. The gene expression levels were normalized against the expression of GAPDH. The relative fold change in expression of the target gene transcript was determined using the comparative quantification method (2^−ΔΔCq) (16). High or low expression of miRNAs and mRNAs were defined on the basis of the median expression, determined separately for each cohort.

The full-length CAMK4 3′-UTRs were cloned into the EcoRI and HindIII sites of the pMIR-REPORT luciferase vector (Ambion; Thermo Fisher Scientific, Inc.) using a PCR-generated fragment. A Luc-mut vector, in which the first seven nucleotides complementary to the miR-30a seed-region were mutated, served as a mutant control. The binding sites of CAMK4 were UGU UUA CA and these were replaced with the CAC CCG UG in the Luc-mut vector.

The Luc-wild-type (wt), Luc-mut and Luc-control vectors were co-transfected within miR-30a mimics into 293T cells (Cell bank of Chinese Academy of Sciences, Shanghai, China) using Lipofectamine 2000 (Thermo Fisher Scientific, Inc.). The pMIR-REPORT β-galactosidase control vector (Thermo Fisher Scientific, Inc.) was transfected into cells to serve as a control. Luciferase activity was measured in the cell lysates 48 h following transfection using a dual-light luminescent reporter gene assay kit (Applied Biosystems; Thermo Fisher Scientific, Inc.). The results were normalized against the activity of β-galactosidase (17).

Cells was lysed in ice-cold lysis buffer (1% NP-40; 50 mM Tris-HCl, pH 8.0; 100 mM sodium fluoride; 30 mM sodium pyrophosphate; 2 mM sodium molybdate; 5 mM EDTA and 2 mM sodium orthovanadate). Lysates were centrifuged at 10,000 × g for 15 min at 4°C. The concentration was determined by Pierce bicinchoninic acid Protein Assay kit (Thermo Fisher Scientific, Inc.). Protein lysates (50 µg) were separated on 10% SDS-polyacrylamide gels and electrophoretically transferred onto polyvinylidene difluoride membranes (EMD Millipore, Billerica, MA, USA). Membranes were blocked in 5% nonfat milk in TBS-Tween (TBST) for 1 h at room temperature. The primary rabbit anti-human CaMK4 antibody (cat. no. 4032; Cell Signaling Technology, Inc., Danvers, MA, USA) was diluted 1:1,000 in 5% milk with TBST and incubated with the membrane at 4°C overnight. Then blots were washed three times with TBST. Secondary goat anti-rabbit IgG (cat. no. ab6721; Abcam, Cambridge, UK) was diluted at 1:3,000 in milk with TBST and incubated with the membrane at 37°C for 1 h. The hybridization signals were detected by chemiluminescence (Invitrogen; Thermo Fisher Scientific, Inc.) and captured using a ChemiDOC (Bio-Rad Laboratories, Inc.).

All data are presented as the mean ± standard error of the mean. Analysis was performed using SPSS software (version 17; SPSS, Inc., Chicago, IL, USA). Comparisons between two groups were made using Student's t-test, whereas multigroup comparisons were made using analysis of variance followed by appropriate post-hoc testing. Pearson's correlation analysis was used to test the correlation between two groups. P<0.05 was considered to indicate a statistically significant difference.

Serum from three patients at seizure onset and post-seizure were used for microarray analysis. The patients included two men and one woman. Serum from an additional 90 patients at seizure onset and post-seizure were used for RT-qPCR analysis. The patients had a mean age of 39 years (range, 21–60), a mean number of years following epilepsy diagnosis of 16 years (range, 5–31) and a mean seizure frequency of 5 (range, 1–20) times per month.

The expression levels of 15 miRNAs were increased and 10 miRNAs were decreased at seizure onset, compared with post-seizure in the patients with epilepsy, as determined using miRNA microarray analysis (Fig. 1). The expression of miR-30a, miR-378, miR-106b and miR-15a was confirmed in the extended cohort of the three original patients and the additional 90 patients (indicated in Table I), using RT-qPCR analysis. The expression levels of miR-30a, miR-378, miR-106b and miR-15a, the top four overexpressed miRNAs identified, were enhanced at seizure onset, compared with post-seizure (Fig. 2).

The correlation between upregulated miRNAs and clinical parameters was statistically analyzed. The expression levels of miR-378, miR-106b and miR-15a were not associated with the age, gender, years following diagnosis or seizure frequency. The expression of miR-30a was positively associated with seizure frequency (Table I).

The online software program, Targetscan 6.0 (www.targetscan.org/vert_71/), was used to assist in identifying miR-30a targets. The miR-30a mimics decreased the luciferase activity of the CAMK4-3′-UTR-wt reporter (Fig. 3A). The results from the RT-qPCR and western blot analyses showed that the enhanced expression of miR-30a by the miR-30a mimics in the 293T cells led to a decrease in endogenous mRNA and protein levels of CaMK4 (Fig. 3B). The mRNA expression of CAMK4 was significantly reduced post-seizure, compared with at seizure onset (64/90; Fig. 3C). The expression of miR-30a and CAMK4 were demonstrated to be negatively correlated (Fig. 3D).