We obtained surgical specimens of 98 gliomas and 16 nontumoral brain tissues (control) from the Pathology Department of the Shenzhen Second People’s Hospital, after obtaining written informed consent from the patients. The formalin-fixed, paraffin-embedded (FFPE) tissue sections with a 5-μm thickness were cut for subsequent miR-637 in situ hybridization (ISH). This study complied with the principles of the Helsinki Declaration and was approved by the Ethics Committee of the Shenzhen Second People’s Hospital.

FFPE tissue sections, adhered to glass slides, were deparaffinized in xylene and rehydrated using gradient ethanol. We have previously described the details of this procedure [4]. The hybridization images were acquired under a fluorescent microscope.

Human glioma cell lines, U373 and U251 cells, were purchased from the China Academia Sinica Cell Repository (Shanghai, China) and cultured in Dulbecco’s Modified Eagle Medium (Gibco) that contained 10% fetal bovine serum (Gibco). All cell lines were incubated at 37˚C in a humidified incubator with 5% CO₂/95% air.

We purchased the dsRNA oligonucleotides of miR-637 mimics and the scrambled sequence (Scr) of Ribobio. U373 and U251 cells in the miR-637 group and the corresponding Scr group were transfected with the corresponding dsRNA oligonucleotides using X-tremeGENE siRNA Transfection Reagent (Roche).

The total RNA from each group was extracted using TRIzol reagent (Invitrogen). miR-637 was quantified using the Bulge-loop miRNA quantitative reverse transcription polymerase chain reaction (qRT-PCR) Detection Kit (Ribobio). U6 was used as the internal miR-637 control. All reactions were performed on a QuantStudio^TM Dx Real-Time PCR Instrument (Thermo Fisher). The fold changes of miR-637 were calculated via the 2^−ΔΔCt method.

We used transwell inserts (24-well, 8-μm pore size; Millipore) with or without a 1:4 diluted Matrigel (BD Bioscience) coating layer to assess the invasive and migratory abilities of the cells. The cells that reached the lower surface were fixed with methanol and stained with 0.1% crystal violet. We have previously described the details of this procedure [5]. Lastly, cells from each chamber were counted under a microscope (×400).

To identify significant candidate target genes of miR-637, we employed four different types of online tools (TargetScan, http://www.targetscan.org/mamm_31/; miRDB, http://www.mirdb.org/; miRTarBase, http://mirtarbase.mbc.nctu.edu.tw/php/index.php; miRWalk, http://zmf.umm.uni-heidelberg.de/apps/zmf/mirwalk2/). Next, we searched for candidate target genes that were identified by at least three tools using the VennDiagram R package.

The Kyoto Encyclopedia of Genes and Genomes (KEGG) (https://www.kegg.jp/) is a database that allows the systematic analysis of gene functions and the linking of genomic information with higher-order functional information. To analyze the candidate target genes at the functional level, we performed KEGG pathway analysis with the Database for Annotation, Visualization, and Integrated Discovery software (DAVID version 6.7; http://www.david.niaid.nih.gov) online tool. Significant enrichment was set at P < 0.05.

All statistical analyses were performed using SPSS 24.0 software. Results are presented as the mean ± standard deviation. The normality of distribution was estimated using the Kolmogorov-Smirnov test. Between-group differences were analyzed using a one-way analysis of variance or t-test. To determine the optimal cutoff level of miR-637 expression for glioma diagnosis, we performed a receiver operating characteristic (ROC) curve analysis. Youden’s index (J = sensitivity + specificity − 1) was used to determine the optimal cutoff level [10]. Specifically, the cutoff value for obtaining the highest Youden’s index was considered the optimal cutoff level. The level of statistical significance was set at P < 0.05. All cell line experiments were performed at least thrice with triplicate samples.

To determine miR-637 expression in gliomas and its association with the WHO grades of gliomas, we used ISH to determine endogenous miR-637 levels in the FFPE specimens. Glioma tissue had significantly lower miR-637 expression than did control tissue (P < 0.001). Further, the expression level was lower when the glioma grade was higher, with the expression in glioblastoma being the lowest (P < 0.001; Figure 1). ROC curve analysis of miR-637 expression in both tissue groups revealed an area under the ROC curve of 0.9254 (95% confidence interval, 0.8698 - 0.9818; P < 0.0001; Figure 2(A) and Figure 2(B)). This demonstrated that miR-637 expression in glioma tissues could differentiate between gliomas and nontumoral brain tissues. The optimal cutoff value for glioma diagnosis using the miR-637 labeling index was ≤ 6.015% (sensitivity, 77.55%; specificity,

93.75%). ROC curve analysis of miR-637 expression for differentiating between low-grade and high-grade gliomas showed that the area under the ROC curve was 0.8046 (95% confidence interval, 0.7190 - 0.8902; P < 0.0001; Figure 2(C) and Figure 2(D)). This indicates that miR-637 expression in glioma tissues could differentiate between glioma grades. The optimal cutoff value for glioma diagnosis using the miR-637 labeling index was ≤ 0.7050% (sensitivity, 86.84%; specificity, 71.67%).

The transfection efficiency of miR-637 mimics was detected through qRT-PCR. We observed a significant increase in miR-637 levels in U373 and U251 cells after the transfection of miR-637 mimics (P < 0.001; Figure 3(E)).

We used the transwell-chamber assay to assess the effects of miR-637 on glioma

cell migration and invasion. Compared with the Scr group, there was a sharp reduction in the number of migrating cells in the miR-637 mimics transfected group (P < 0.01; Figure 3(A) and Figure 3(B)). The transwell-chamber invasion assay showed that miR-637 could effectively inhibit U373 and U251 cell invasion (P < 0.001; Figure 3(C) and Figure 3(D)). Our findings indicate that miR-637 downregulation could promote U373 and U251 cell migration and invasion.

As shown in Table 1, 136 common candidate target genes were identified from all four tools (Figure 4). The candidate target genes were enriched in KEGG pathways including pathways in cancer and the Rap1 signaling pathway, focal adhesion pathway, and cell adhesion pathway (Table 1). These pathways are involved in cell proliferation, migration, and invasion, which are closely associated with tumor progression and metastasis.

KEGG, Kyoto Encyclopedia of Genes and Genomes.