Introduction

GBM ranks among the most aggressive malignancies in the adult central nervous system. It is prominently characterized by a trio of formidable challenges: an elevated recurrence rate, resistance to a wide array of therapeutic approaches, and a dismal prognosis1,2,3. Notably, around 30–35% of glioblastoma patients are afflicted with comorbid epilepsy4,5, a symptom that substantially deteriorates the quality of life6. The presence of seizures as the initial manifestation of GBM has been linked to extended survival, while the recurrence of seizures frequently shows a correlation with tumor progression7, suggesting epilepsy may influence GBM onset and progression. However, the importance of seizure management in GBM patients is often overlooked, as neuro-oncological studies typically prioritize tumor progression and overall survival8. Considering the multifactorial characteristics of glioma-related epileptogenic mechanisms, which have similarities with tumor growth processes and are currently not fully understood, grasping the underlying pathogenesis of tumor-related epilepsy is essential for devising effective treatment strategies9. Exploring these mechanisms has the potential to drive significant progress in both anti-epileptic and anti-tumor therapeutic approaches10.

In recent years, the dual functions of miRNAs in tumorigenesis and the regulation of epilepsy have attracted substantial attention. MiRNAs are single-stranded non-coding RNAs, renowned for their function in post-transcriptional gene regulation. They achieve this by mediating either mRNA degradation or translational inhibition11. MiRNAs regulate gene expression, affecting glioblastoma proliferation, invasion, and chemoresistance12. Aberrant expression of microRNAs (miRNAs) has been reported in human gliomas13, and their association with membrane channel AQP, widely expressed in human tissues, has been observed14. However, the body of research specifically focused on glioma-related epilepsy remains restricted15. Previous investigations conducted by our research team have underscored the dysregulation of miR-196b and miR-128 in predicting the incidence of epilepsy in low-grade glioma16,17. Intriguingly, miR-1290 demonstrates oncogenic characteristics in numerous solid tumors. However, its function in glioblastoma, especially within the context of epilepsy comorbidity, remains largely unclear18,19,20,21. Furthermore, the crosstalk between neurons and tumor cells within the glioblastoma microenvironment likely involves miRNA-mediated signaling networks that play a role in epileptogenesis. However, this remains a largely uncharted area of research.

Against this backdrop, the present study is designed to conduct a systematic exploration of the expression patterns and functional mechanisms of miR-1290 in glioblastoma patients with comorbid epilepsy. Through the integration of clinical cohort analyses and functional experiments, we endeavor to clarify whether miR-1290 simultaneously impacts tumor progression and epileptic seizures by modulating tumor-neuron interactions. This research not only offers fresh perspectives on the molecular subtyping of glioblastoma-associated epilepsy but also establishes a theoretical groundwork for the development of miR-1290-targeted combinatorial therapeutic strategies17.

Materials and methods

Patients and tissue samples

This study received approval from the Ethics Committee of Beijing Tiantan Hospital, and written informed consent was secured from all the enrolled patients (Ethics Approval Number: KY2019-042-02). From January 2006 to October 2013, a total of 1016 patients diagnosed with glioblastomas (GBMs) received treatment at the Glioma Treatment Center of Beijing Tiantan Hospital, China. Moreover, we had access to information that enabled the identification of individual participants both during and after data collection. Tissue samples were recovered immediately after surgery, snap-frozen using liquid nitrogen vapors and stored at -80℃ until needed. The pathological diagnosis of each patient was re-evaluated by integrating histopathology and molecular biomarkers according to the 2016 WHO classification of tumors of the central nervous system2 The transcriptomic data of the samples collected during the surgery were generated on the Agilent Whole Human Genome Array platform and were de-batched. For the purpose of creating more homogeneous sample conditions, only samples that strictly met the following criteria were selected: (1) age ≥ 18 years old age (2) no medication involving antiepileptic drugs (AEDs) before admission; (3) Clear preoperative medical history; and (4) pure WHO grade IV glioblastoma by histopathology. Finally, the study included 81 GBM patients (surgery year 2006–2008) with miRNA-array data from the CGGA dataset, of which 56 had available mRNA-array data. Additionally, 92 independent GBM samples (23 seizure patients and 69 non-seizure patients, randomly selected during surgery years 2009–2013) were used for validation through quantitative reverse-transcriptase polymerase chain reaction (qPCR) analysis. The data were accessed for research purposes on (23/05/2020).

MicroRNA expression profiling

Total RNA was polyadenylated and then converted to cDNA using a biotin-labeled oligonucleotide dT primer with a universal PCR sequence. After cDNA synthesis, miRNA was individually detected using specific oligonucleotides. A single miRNA-specific oligonucleotide (MSO) 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. 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 of the miRBase 12.0 database).

Quantitative RT-PCR

Quantitative RT-PCR was conducted using a standard Taq-Man PCR kit on an ABI 7900 real-time PCR system. Real-time RT-PCR (qPCR) was used to investigate mRNA expression. PCR primers were designed using Primer Premier 5.0 based on the reported cDNA sequences (Table 1). All the primers and probes for has-miR-1290 and U6 rRNA endogenous controls for TaqMan microRNA assays were purchased from Applied Biosystems. Thermal cycling for the quantitative RT-PCR was carried out according to the manufacturer’s recommendation and the relative expression levels were calculated using the comparative Ct method.

Table 1 List for primers used for RT-PCR.
Full size table

Gene expression heatmap

To visualize the expression patterns of differentially expressed genes (DEGs), a heatmap was generated using the pheatmap package in R. Genes were clustered using hierarchical clustering with Euclidean distance as the metric and complete linkage for the dendrogram construction.

Target gene prediction and functional analysis

The target genes of miRNA 1290 were predicted by the TargetScan tool, with a screening criterion of context + + score ≥ -0.1, using a human genome-associated prediction model. Functional enrichment analyses were performed for the target gene list obtained by TargetScan using the DAVID tool, and the analyses included gene ontology (GO) functional annotation, KEGG pathway analysis, and other functional annotations22,23,24. The GO analyses included biological process (BP), molecular function (MF), and cellular component (CC), and significance tests were performed using Fisher ‘s Exact Test, P-values were corrected for multiple corrections by the Benjamini-Hochberg method, and the significance threshold was P-value < 0.05. and heatmap was plotted by https://www.bioinformatics.com.cn (last accessed on 10 Dec 2024), an online platform for data analysis and visualization25.

Statistical analysis

Statistical analyses were performed using GraphPad Prism 9.0. MiR-1290 expression, treated as a continuous variable, was analyzed for associations with baseline clinical characteristics using Student’s t-test. Chi-square tests were used for dichotomous variables, while the Mann-Whitney U test was employed for continuous nonparametric data.

Results

Clinical features of GBM-related seizures

This study included 173 patients who underwent primary resection of glioblastoma (GBM). Among these patients, 44 (25.4%) had preoperative seizures, whereas 129 (74.6%) had no seizure history. The patients with preoperative seizures exhibited a significantly longer survival time compared to those without seizures (Fig. 1).

Fig. 1
figure 1

Significant survival difference between epilepsy and non-epilepsy group in 173 GBM cases. (A) Patients with preoperative seizures exhibited longer Progression-free survival when compared with those without pre-operative seizures. (B) Patients with preoperative seizures exhibited longer overall survival when compared with those without pre-operative seizures.

Full size image

Patient characteristics

The average age of the patients during surgery was 48.3 years old, with an age range spanning from 17 to 72 years. Male glioblastoma (GBM) patients were significantly more prone to having preoperative seizures (p = 0.024). Regarding the seizure type, nearly half of the patients (47.7%) suffered from generalized seizures, while 52.3% experienced focal seizures. The factors predisposing patients to preoperative seizures are comprehensively presented in Table 2.

Table 2 Clinical and demographic characteristics of GBM patients.
Full size table

Differentially expressed MiRNAs and their impact in epilepsy

Genome-wide miRNA analysis was conducted on 81 samples to identify differentially expressed miRNAs between patients with and without preoperative seizures. Among these patients, 7 miRNAs were found to be significantly under-expressed in those with preoperative seizures compared to those without (p < 0.05 and at least 1.5-fold change, see Table 3). Notably, miR-1290 exhibited the highest fold change (> 2 fold). Given its significant differential expression, miR-1290 was selected for further investigation. We then used an independent cohort of 92 samples to validate the association between miR-1290 expression and the occurrence of preoperative seizures. As depicted in Fig. 2, patients experiencing preoperative seizures exhibited lower levels of miR-1290 compared to those without seizures (p = 0.018).

Table 3 Seven differently expressed MiRNAs between seizure and non-seizure group (fold change > 1.5, P < 0.05) *.
Full size table
Fig. 2
figure 2

Association of miR-1290 expression by qPCR with pre-operative seizures in GBM patients of validation cohort. The red area represents the epilepsy group, the green area represents the non-epilepsy group, and the expression of miR-1290 in the epilepsy group is lower than that in the non-epilepsy group.

Full size image

Biological insights of miR-1290

To elucidate the biological implications of miR-1290, we analyzed gene expression patterns in 56 patients with overlapping mRNA array data. Utilizing a top 100-gene signature predicted by miR-1290, we observed distinct clustering of genes according to seizure symptoms (Fig. 3).

Fig. 3
figure 3

Gene clustering result according to seizure symptom by using a top 100 genes signature predicted by miR-1290.

Full size image

To further explore the potential role of miR-1290 in GBM, we conducted GO analysis. The most relevant genes of miR-1290, or a characteristic gene list of the cell cluster, were uploaded to the https://www.bioinformatics.com.cn. The official gene symbol was selected as an identifier, and Homo sapiens was selected as species. Finally, Gene Ontology (GO) analysis (Fig. 4) and Kyoto Encyclopedia of Genes and Genomes (KEGG) (Fig. 5) pathway analysis enrichment results were obtained. The top five results in ascending order of P-value (P < 0.05) were displayed in this study. From the results, we were able to see that biological processes most related to miR-1290 include regulation of transcription by RNA polymerase II, regulation of DNA-templated transcription, and positive regulation of transcription by RNA polymerase II. Moreover, miR-1290’s most related cellular components were nucleus. The molecular functions were RNA polymerase II cis-regulatory region sequence-specific DNA binding.

Fig. 4
figure 4

Results of GO analysis by using target genes of miR-1290. miR-1290 is closely associated with transcriptional regulation in glioblastoma. Biological processes (BP), cellular components (CC), and molecular functions are mostly related to miR-1290. Biological processes most related to miR-1290 include regulation of transcription by RNA polymerase II, regulation of DNA-templated transcription, and positive regulation of transcription by RNA polymerase II. Moreover, miR-1290’s most related cellular components were nucleus. The molecular functions were RNA polymerase II cis-regulatory region sequence-specific DNA binding .

Full size image
Fig. 5
figure 5

Results of KEGG analysis by using target genes of miR-1290. miR-1290 is most relevant to the function of tight junctions.

Full size image

The relationship between miR-1290 and survival

In our study, miR-1290 was the only miRNA associated with patient survival. Survival analysis (such as Kaplan-Meier survival curves and Cox regression analysis) revealed that the expression level of miR-1290 significantly impacted patient survival time. Compared to the other six miRNAs (with a cutoff percentile of 50%), miR-1290 showed a distinct survival prognostic effect (Fig. 6). Specifically, the high-expression group of miR-1290 exhibited a significantly different survival rate compared to the low-expression group, suggesting that miR-1290 plays an important role in the biological process of glioma.

Fig. 6
figure 6

Survival plots of miR-1290 (at 0.5 percentiles). Red represents the high-expression group, green represents the low-expression group, and the low-expression group survived longer than the high-expression group.

Full size image

Discussion

Glioma-related epilepsy stands as one of the most common manifestations of symptomatic epilepsy26. Clinical observations indicate that, in contrast to larger and fast-growing lesions, smaller and slower-growing tumors are correlated with higher seizure frequencies27. Proposed explanations encompass the proclivity of high-grade brain tumors to localize in white matter regions, the potential disruption of epilepsy-related conduction pathways caused by rapid tumor growth, and the unfavorable situation in which some GBM patients do not survive long enough to develop epilepsy28.

Numerous theories have been put forward to elucidate the molecular mechanisms underlying this phenomenon. Peritumoral alterations, including changes in pH, hypoxia, fluctuations in neurotransmitter levels, the participation of astrocytes, the modulation of ion channels and receptors, metabolic transitions, the disruption of the blood-brain barrier, and immune responses, might jointly contribute to the epileptogenicity detected in glioma patients29. These microenvironmental alterations mainly emerge as secondary consequences of the tumor itself, with the tumor acting as the primary etiological factor for tumor-related epilepsy30. Consequently, a more in-depth elucidation of the inherent molecular genetic factors underlying tumor-induced epilepsy is of utmost importance for optimizing patient evaluation and treatment strategies.

Accumulating evidence suggests that the expression patterns of specific miRNAs might assume crucial roles in the pathogenesis of epilepsy and hold the potential to be targeted for innovative therapeutic interventions31. Our research constitutes the inaugural quantitative analysis employing miRNA arrays in clinical GBM samples. This analysis not only offers invaluable insights into seizure-related factors but also uncovers abnormal miRNA expression patterns.

Seizure characteristics of GBM patients

The prevalence of seizures (25.9%) as the initial symptom in our patient cohort aligns with previously reported rates (23-52%). Notably, we found no significant differences in seizure occurrence among different tumor locations. Interestingly, our study identified a correlation between gender and seizure incidence in GBM patients, with males exhibiting a higher likelihood of seizures compared to females. This suggests a potential gender-related pathogenesis of seizures in GBM, consistent with recent literature, which underscores the existence of gender-specific clinical and molecular genetic features in GBM. These findings emphasize the importance of considering gender-related mechanisms in understanding GBM tumorigenesis and malignant progression.

MiR-1290 and tumor malignancy

Increasing evidence identifies miRNAs as pivotal regulators of glioma phenotype, targeting critical genes and signaling pathways involved in tumorigenesis and malignant progression32. In our study, we identified 7 miRNAs significantly down-regulated in GBM patients with seizures, some of which also correlated with survival, suggesting a shared role of miRNAs in both epileptogenesis and tumorigenesis.

MiR-1290 dysregulation has been observed in various human cancers, including lung adenocarcinoma, colorectal cancer, and esophageal cancer. Recent studies have identified miR-1290 as an oncogene in glioma, supporting its relevance in gliomas33. Yan et al. demonstrated that miR-1290 promotes glioma development by enhancing cell proliferation and metastasis, consistent with our findings linking miR-1290 expression to survival outcomes34. These findings suggest that up-regulation of miR-1290 may contribute to tumorigenesis and malignant progression, and could potentially serve as a diagnostic and prognostic marker for gliomas.

MiR-1290 and cerebrospinal fluid (CSF)

GBM has insidious early-stage symptoms and is difficult to detect. However, CSF samples can reflect the physiological and pathological states of the central nervous system. If the level of miR-1290 can be accurately detected in CSF, it will be of great significance for early-stage diagnosis and treatment.

In the future, we plan to conduct in-depth research. On the one hand, we will actively collect more CSF samples from patients, especially those in the prodromal or early-stage of GBM, analyze the expression changes of miR-1290 at different stages, and establish its association with disease progression. On the other hand, we will establish a strict healthy control cohort to compare the expression differences of miR-1290 in the CSF of healthy people and GBM patients, and clarify its specificity and sensitivity as a disease biomarker. By measuring the level of miR-1290 in CSF before surgery, we can provide more accurate prognostic assessments for patients and assist in the formulation of personalized treatment and surgical plans.

MiR-1290 and seizure occurrence

The epileptic mechanisms involving miR-1290 remain unclear and likely depend on its biological processes. In our study, we observed a correlation between miR-1290 expression and seizure occurrence. Our previous research indicated that overexpression of miRNA-196b in low-grade gliomas promoted seizure susceptibility by increasing cell proliferation, which appears contradictory to our current findings16. One plausible explanation is that the expansive growth of low-grade gliomas typically does not disrupt epileptic pathways, whereas the aggressive behavior of high-grade tumors like GBMs may disrupt neural circuits due to invasive growth, resulting in lower seizure risk35,36. Thus, GBM patients with fewer seizures often exhibit poorer survival due to higher tumor malignancy, suggesting distinct mechanisms between low and high-grade brain tumors.

Given that neurons play a crucial role in initiating synchronous neuronal activity, it is plausible that miR-1290 influences seizures by regulating neuronal excitability. This could represent another critical mechanism underlying tumor-related epilepsy, warranting further investigation.

Conclusions and perspectives

Our study provides pioneering evidence that miR-1290 serves as a novel and valuable biomarker for predicting seizure susceptibility and overall survival in GBM patients. Prospective controlled studies are warranted to validate these findings rigorously. Further investigation of miR-1290 target genes and downstream pathways will lead to the identification of drug targets that can advance the prognosis of GBM patients through the molecular mechanisms involved in epileptogenesis and tumor progression.

Limitations

There are certain limitations in the current study. On the one hand, the postoperative MRI data of some patients are missing, resulting in incomplete data on the extent of tumor resection during surgery. This may affect the accurate analysis of the relationship among surgical outcomes, the degree of tumor resection, and other research factors. On the other hand, this study only compared glioblastoma patients with or without preoperative epilepsy, without including healthy individuals as control. This makes it difficult to clearly define the expression differences of miR-1290 between the healthy state and the diseased state.