ABCC8 mRNA expression is an independent prognostic factor for glioma and can predict chemosensitivity

Glioma is the most common primary intracranial tumor and is associated with very low survival rates. The development of reliable biomarkers can help to elucidate the molecular mechanisms involved in glioma development. Here the expression of ABCC8 mRNA, clinical characteristics, and survival information based on 1893 glioma samples from four independent databases were analyzed. The expression patterns of ABCC8 mRNA were compared by a Chi square test. The overall survival rate of gliomas was evaluated according to the expression level of ABCC8 mRNA. The prognostic value of this marker in gliomas was tested using Cox single factor and multi factor regression analyses. We found patients with low WHO grade, oligodendrocytoma, low molecular grade, IDH mutation, and 1p19q combined deletion had high ABCC8 mRNA expression. The patients with high expression of ABCC8 mRNA had longer survival. ABCC8 mRNA expression was a new independent prognostic index for glioma. Temozolomide chemotherapy was an independent index to prolong overall survival in high ABCC8 mRNA expression glioma patients, whereas in patients with low expression, there was no significant difference. So ABCC8 mRNA expression could be an independent prognostic indicator for glioma patients and could predict the sensitivity of glioma to temozolomide.

Glioma is the most common primary intracranial tumor, accounting for 81% of malignant brain tumors 1 . World Health Organization (WHO) grade 4 glioblastoma is the most malignant form of glioma with a 5-year relative survival rate of 5% 2 . In recent years, studies have elucidated some genetic changes in glioma, such as IDH1/2 3 , TP53 4 and ATRX 5 mutations, TERT promoter mutations 6 , MET-exon-14-skipping (METex14), PTPRZ1-MET (ZM) fusions, MET amplification 7 , MGMT promoter methylation 8 , and 1p/19q co-deletion 9 , which have been helpful in guiding the classification and treatment of glioma. As per the WHO's classification of central nervous system tumors in 2016, for the first time, diffuse gliomas were classified according to IDH1 or IDH2 mutations and the co-deletion of 1p and 19q chromosome arms 10 . However, the development of new and reliable biomarkers is necessary to further elucidate the molecular mechanism of glioma development.
ATP-binding cassette (ABC) transporters play a crucial role in the development of resistance by the efflux of anticancer agents outside of cancer cells. It was found that the expression of ABCC8 was down-regulated in pancreatic cancer 11 , lung adenocarcinoma 12 and triple negative breast cancer 13 . The low expression of ABCC8 was associated with poor prognosis in these tumors. Thompson et al. 14 . found ABCC8 expression was greater in supratentorial ependymoma compared to glioblastoma and metastases. However the expression and clinical significance of ABCC8 mRNA in gliomas are still unclear.
In this study, we collected the expression data and clinical information with respect to ABCC8 mRNA based on 656 glioma samples from CGGA database as testing set. The expression patterns of ABCC8 in different types of gliomas were compared and the overall survival (OS) rate of glioma patients was evaluated according to the expression level of ABCC8 mRNA, and the prognostic value of this marker in gliomas was tested. Meanwhile High ABCC8 mRNA expression in patients with better prognosis indicators. The ABCC8 mRNA expression was compared via box plots, among diferent clinical and molecular pathological characteristics, including gender, age, histopathology, WHO grade ( Fig. 1), IDH mutation and 1p/19q col-deletion status (Fig. 2). The patients were divided into two groups according to the median age of 42 years. The results showed that ABCC8 mRNA expression was higher in patients with better prognosis indicators, such as oligodendroglioma (p = 1.4e−32), low WHO grade (p = 2.5e−23), IDH mutation (p = 6.4e−20), 1p/19q col-deletion (p = 7.3e−23) and low who molecular grade (2016) (p = 2.8e−35). www.nature.com/scientificreports/ www.nature.com/scientificreports/ expression of ABCC8 mRNA was related to better OS than low expression (P = 0.000; Fig. 3), which was also confirmed by using the GSE16011 (p = 0.000), REMBRANDT (p = 0.000) and TCGA databases (p = 0.000) (Fig. 3).

Discussion
ABCC8 (ATP binding cassette subfamily C member 8), also known as ABC36 (member 36 of ATP binding cassette transporter superfamily), SUR1 (sulfonylurea receptor 1), or MRP8, is encoded by the ABCC8 gene located at 11p15.1. Gene Ontology (GO) notes related to the ABCC8 gene include potassium channel activity, ATP binding, sulfonylurea receptor activity, ATPase activity (GO molecular function for the ABCC8 gene). The ABCC8 gene encodes proteins that are members of the ATP binding cassette (ABC) transporter superfamily. ABC proteins transport various molecules (based on the Entrez gene summary for ABCC8 gene). The research findings of Martin et al. 15 show that the protein encoded by ABCC8 plays a role in regulating the ATP-sensitive potassium (K-ATP) channel and insulin release. The ATP sensitive potassium channel is a protein complex that can couple the cell energy level with cell excitability and control a wide range of physiological processes,  www.nature.com/scientificreports/ including hormone secretion, neuronal transmission, vasodilation, and cardiac and neuron pretreatment for ischemic injury. Studies of Zhang et al. 16 show that K-ATP channels are widely expressed in the central nervous system and are coupled with cell metabolism and electrical activity. The K-ATP channel in mature substantia nigra dopaminergic neurons consists of inward rectifying K-channel subunit 6.2 and SUR1, which regulates the K-ATP channel in a Parkinson's disease mouse model and has protective effects on dopamine neurons. ABCC8related diseases include diabetes, permanent neonatal and hyperinsulinemic hypoglycemia, and familial disease 1. The related pathways include inward rectifying K + channel and energy metabolism integration. Mutations and deletions in the protein (genecards summary for ABCC8 gene) were observed in infants with hyperinsulinemic hypoglycemia. ABCC8 is a transmembrane protein that regulates the activity of ion channels in neurons, glia, and endothelial cells. In neuroinflammatory diseases, SUR1 expression was found to be upregulated. In rodent models of subarachnoid hemorrhage 17 , ischemic stroke 18 , traumatic brain injury 19 , spinal cord injury 20 , and brain metastasis 21 , the inhibitory effect of glibenclamide on SUR1 was found to reduce edema and neuroinflammation by antagonizing cytotoxic edema and apoptosis. A retrospective study carried out to examine patients with diabetic acute ischemic stroke led to the finding that glibenclamide could improve neurological functions, reduce the conversion of bleeding, and reduce mortality 22,23 . The preclinical study of Thompson et al. 21 , using a brain metastasis mouse model, illustrated the potential benefits of glibenclamide for the prevention of edema. JHA Ruchira et al. 24 showed that SUR1 and its related transient receptor potential cation channel subfamily-m (Trpm4) channel are the key factors of brain edema and intracranial hypertension in traumatic brain injury and other neurological diseases. Glibenclamide reduces brain edema and intracranial hypertension by inhibiting this channel. The SNP encoding SUR1 can be used as a predictor of increases in intracranial pressure. Further, King et al. 25 demonstrated that sur1-trpm4 is a key target of stroke. After ischemia, the sur1-trpm4 channel is upregulated in all cells of the neurovascular unit (including neurons, astrocytes, microglia, oligodendrocytes, and microvascular endothelial cells). Moreover, the second-generation sulfonylurea glibenclamide can inhibit SUR1 at nanomolar concentration to reduce the brain swelling and death of patients.
At present, the expression and function of ABCC8 mRNA in brain tumors is less studied. Thompson et al. 14 analyzed six glioblastomas, 12 brain metastases, 11 medulloblastomas, nine supratentorial ependymomas, and eight posterior fossa ependymomas by immunofluorescence. The expression of ABCC8 and its co-localization with blood vessels, neurons, and glial cells was analyzed and compared by ANOVA. The results showed that the percentage of cells expressing ABCC8 in total tissue area (mean ± SD) was 3.9 ± 4 in glioblastoma, 4.1 ± 3.1 in brain metastasis, 8.2 ± 7.2 in medulloblastoma, 9.1 ± 7 in supratentorial ependymoma, and 8.1 ± 5.9 in the posterior cranial cavity. The expression of ABCC8 was higher in supratentorial ependymoma than in glioblastoma and metastasis (P < 0.05), and higher in medulloblastoma than in glioblastoma. .It is suggested that ABCC8 expression is lower in malignant brain tumors. This is consistent with our findings.
Our study found that the expression of ABCC8 mRNA was negatively correlated with the tumor tissue type, WHO grade, WHO molecular grade, IDH wildtype, and 1p/19q non-codel. The patients with oligodendroglioma, low WHO grade, low WHO molecular grade, IDH mutation, and 1p19q deletion had high ABCC8 mRNA expression. This is associated with their good prognosis, and a long survival period..ABCC8 mRNA was found to be an independent factor associated with good prognosis. However, the molecular mechanism underlying such effects has not been reported and requires further study. Mohelnikova et al. 11 found that the expression of ABCC8 in ductal adenocarcinoma of the pancreas was significantly downregulated compared to levels in adjacent non-tumor tissues. It was concluded that stem cells might play a role in the development and progression of ductal adenocarcinoma of the pancreas. Wang et al. 12 screened 6 genes (CLEC17A, TAGAP, ABCC8, BCAN, FLT3, and CCR2) related to immune and stromal cells in the tumor microenvironment through bioinformatics and constructed a risk assessment model to predict the prognosis of lung adenocarcinoma. They found patients with high expressions of CLEC17A, TAGAP, ABCC8, FLT3, and CCR2 had better prognosis and higher OS within 5 years. They demonstrated through a series of rigorous analyses that lung adenocarcinoma patients with high infiltration of immune cells (stromal cells) had better prognosis and earlier staging. The study of Hlavá et al. 13 showed that the expression level of ABCC8 in breast cancer is significantly correlated with the grade and expression of hormone receptors, which represents a potential modifying factor of chemotherapy progress and the response of breast cancer. ABCC8, C11, C12, C13, and A10 showed the highest level of downregulation within the cluster BC3, which significantly prevailed in ERnegative (p = 0.012), PRnegative (p = 0.009), and TNBC (p = 0.022) cases, that is, in patients with a generally worse prognosis. Zhou et al. 26 performed quantitative analysis of ABCC8 mRNA expression in breast cancer cells using real-time RT-PCR. The ABCC8 mRNA expression level was found to be related to the methylation state of the CpG island in the promoter region,that is, the expression of mRNA was higher in the low methylation state of the promoter region. Thus, the potential application of CpG island methylation in the promoter region of breast cancer cells to predict the chemosensitivity of breast cancer was indicated. In this study, we found that temozolomide chemotherapy was an independent index to prolong the survival of glioma patients with high expression of ABCC8 mRNA, but there was no significant difference in the survival of patients with low expression. This sugested that high ABCC8 mRNA expression could predict chemosensitivity of glioma. Our study found that in low ABCC8 mRNA expression glioma patients, radiotherapy was independent predictors of long OS. While in high ABCC8 mRNA expression group, there was no significant difference. This sugested that low ABCC8 mRNA expression could predict radiosensitivity of glioma. The mechanism needs further study.

RNA sequence quantification.
After mapping reads to the human genome (GENCODE v19 and HG19), we used the RSEM software (v1.2.31) to quantify the expression level of genes/transcripts from RNA sequence data. The preparation, sequencing and data analysis of the RNAseq library are the same as our previous studies 28 .
Statistical analysis. According to the median value of ABCC8 mRNA expression, glioma patients were separated into low and high expression groups in the CGGA, GSE16011, REMBRANDT and TCGA databases. Box plots were used to evaluate the ABCC8 mRNA expression in the subgroups according to clinical and molecular characteristics such as gender, age, histopathology, WHO grade, IDH mutation, 1p/19q codel status, and WHO molecular grade (2016) by R sofware (version 3.6.1) (https ://www.r-proje ct.org/) and related packages. Chi-square tests were utilized to evaluate correlations between the expression of ABCC8 mRNA and clinical information by SPSS 25.0 software package. The OS was calculated from the date of histological diagnosis until death or the last follow-up. Kaplan-Meier survival analysis and log-rank tests were used to evaluate the statistical significance associated with stratified survival groups by SPSS 25.0 software package. The potential prognostic factors were selected by Cox analysis. Correlations between ABCC8 mRNA expression, survival and other clinical characteristics of glioma patients were confrmed by using multifactor Cox analysis. P values less than 0.05 were considered statistically significant.

Conclusions
Our study preliminarily confirmed the expression of ABCC8 mRNA in glioma and its clinical significance. ABCC8 mRNA expression can be used as an independent prognostic indicator for glioma patients. Its high expression can be used to predict the chemosensitivity of glioma. Meanwhile low ABCC8 mRNA expression can be used to predict the sensitivity of glioma to radiotherapy. However, the molecular mechanism of its action needs to be elucidated by further studies.