PLEK2, RRM2, GCSH: A Novel WWOX-Dependent Biomarker Triad of Glioblastoma at the Crossroads of Cytoskeleton Reorganization and Metabolism Alterations

Artificial Intelligence-Assisted Drug and Biomarker Discovery for Glioblastoma: A Scoping Review of the Literature

Background: Artificial intelligence (AI) has emerged as a transformative tool in healthcare, particularly in drug and biomarker discovery, where it can enhance precision, streamline discovery processes, and optimize treatment strategies. Despite its potential, the application of AI in glioblastoma (GB) research, especially in identifying novel biomarkers and therapeutic targets, remains underexplored. The aim of this review is to map the existing literature on AI-driven approaches for biomarker and drug discovery in GB, highlighting key trends and gaps in current research. Design: Following a PRISMA methodology, this scoping review examined studies published between 2012 and 2024. Searches were conducted across multiple databases, including MEDLINE (PubMed), Scopus, the Cochrane Library, and Web of Science (WOS). Eligible studies were screened, and relevant data were extracted and synthesized to provide a comprehensive overview of AI applications in GB research. Results: A total of 224 records were identified, including 210 from PubMed, 104 from Scopus, 4 from WOS, and 6 from the Cochrane Library. After screening and applying eligibility criteria, 33 studies were included in the final review. These studies showcased diverse AI methodologies applied to both drug discovery and biomarker identification, focusing on various aspects of GB biology and treatment. Conclusions: This scoping review reveals an increasing interest in AI-driven strategies for biomarker and drug discovery in GB, with promising initial results. However, further large-scale, rigorous studies are needed to validate real-world applications of AI and the development of standardized protocols to enhance reproducibility and clinical translation.

PLEK2 mediates metastasis and invasion via α5-nAChR activation in nicotine-induced lung adenocarcinoma

Evidence has shown a strong relationship between smoking and epithelial mesenchymal transition (EMT). α5-nicotinic acetylcholine receptor (α5-nAChR) contributes to nicotine-induced lung cancer cell EMT. The cytoskeleton-associated protein PLEK2 is mainly involved in cytoskeletal protein recombination and cell stretch migration regulation, which is closely related to EMT. However, little is known about the link between nicotine/α5-nAChR and PLEK2 in lung adenocarcinoma (LUAD). Here, we identified a link between α5-nAChR and PLEK2 in LUAD. α5-nAChR expression was correlated with PLEK2 expression, smoking status and lower survival in vivo. α5-nAChR mediated nicotine-induced PLEK2 expression via STAT3. α5-nAChR/PLEK2 signaling is involved in LUAD cell migration, invasion and stemness. Moreover, PLEK2 was found to interact with CFL1 in nicotine-induced EMT in LUAD cells. Furthermore, the functional link among α5-nAChR, PLEK2 and CFL1 was confirmed in mouse xenograft tissues and human LUAD tissues. These findings reveal a novel α5-nAChR/PLEK2/CFL1 pathway involved in nicotine-induced LUAD progression.

LINC01137/miR-186-5p/WWOX: a novel axis identified from WWOX-related RNA interactome in bladder cancer

Introduction: The discovery of non-coding RNA (ncRNA) dates back to the pre-genomics era, but the progress in this field is still dynamic and leverages current post-genomics solutions. WWOX is a global gene expression modulator that is scarcely investigated for its role in regulating cancer-related ncRNAs. In bladder cancer (BLCA), the link between WWOX and ncRNA remains unexplored. The description of AP-2α and AP-2γ transcription factors, known as WWOX-interacting proteins, is more commonplace regarding ncRNA but still merits investigation. Therefore, this in vitro and in silico study aimed to construct an ncRNA-containing network with WWOX/AP-2 and to investigate the most relevant observation in the context of BLCA cell lines and patients. Methods: RT-112, HT-1376, and CAL-29 cell lines were subjected to two stable lentiviral transductions. High-throughput sequencing of cellular variants (deposited in the Gene Expression Omnibus database under the GSE193659 record) enabled the investigation of WWOX/AP-2-dependent differences using various bioinformatics tools (e.g., limma-voom, FactoMineR, multiple Support Vector Machine Recursive Feature Elimination (mSVM-RFE), miRDB, Arena-Idb, ncFANs, RNAhybrid, TargetScan, Protein Annotation Through Evolutionary Relationships (PANTHER), Gene Transcription Regulation Database (GTRD), or Evaluate Cutpoints) and repositories such as The Cancer Genome Atlas (TCGA) and Cancer Cell Line Encyclopedia. The most relevant observations from cap analysis gene expression sequencing (CAGE-seq) were confirmed using real-time PCR, whereas TCGA data were validated using the GSE31684 cohort. Results: The first stage of the whole study justified focusing solely on WWOX rather than on WWOX combined with AP-2α/γ. The most relevant observation of the developed ncRNA-containing network was LINC01137, i.e., long non-coding RNAs (lncRNAs) that unraveled the core network containing UPF1, ZC3H12A, LINC01137, WWOX, and miR-186-5p, the last three being a novel lncRNA/miRNA/mRNA axis. Patients' data confirmed the LINC01137/miR-186-5p/WWOX relationship and provided a set of dependent genes (i.e., KRT18, HES1, VCP, FTH1, IFITM3, RAB34, and CLU). Together with the core network, the gene set was subjected to survival analysis for both TCGA-BLCA and GSE31684 patients, which indicated that the increased expression of WWOX or LINC01137 is favorable, similar to their combination with each other (WWOX↑ and LINC01137↑) or with MIR186 (WWOX↑/LINC01137↑ but MIR186↓). Conclusion: WWOX is implicated in the positive feedback loop with LINC01137 that sponges WWOX-targeting miR-186-5p. This novel WWOX-containing lncRNA/miRNA/mRNA axis should be further investigated to depict its relationships in a broader context, which could contribute to BLCA research and treatment.

Development and validation of a predictive model in diagnosis and prognosis of primary glioblastoma patients based on Homeobox A family

BACKGROUND

Homeobox A (HOXA) family is involved in the development of malignancies as either tumor suppressors or oncogenes. However, their roles in glioblastoma (GBM) and clinical significance have not been fully elucidated.

METHODS

HOXA mutation and expressions in pan-cancers were investigated using GSCA and Oncomine, which in GBM were validated by cBioPortal, Chinese Glioma Genome Atlas (CGGA), and The Cancer Genome Atlas (TCGA) datasets. Kaplan-Meier analyses were conducted to determine prognostic values of HOXAs at genetic and mRNA levels. Diagnostic roles of HOXAs in tumor classification were explored by GlioVis and R software. Independent prognostic HOXAs were identified using Cox survival analyses, the least absolute shrinkage and selection operator (LASSO) regression, quantitative real-time PCR, and immunohistochemical staining. A HOXAs-based nomogram survival prediction model was developed and evaluated using Kaplan-Meier analysis, time-dependent Area Under Curve, calibration plots, and Decision Curve Analysis in training and validation cohorts.

RESULTS

HOXAs were highly mutated and overexpressed in pan-cancers, especially in CGGA and TCGA GBM datasets. Genetic alteration and mRNA expression of HOXAs were both found to be prognostic. Specific HOXAs could distinguish IDH mutation (HOXA1-7, HOXA9, HOXA13) and molecular GBM subtypes (HOXA1-2, HOXA9-11, HOXA13). HOXA1/2/3/10 were confirmed to be independent prognostic members, with high expressions validated in clinical GBM tissues. The HOXAs-based nomogram model exhibited good prediction performance and net benefits for patients in training and validation cohorts.

CONCLUSION

HOXA family has diagnostic values, and the HOXAs-based nomogram model is effective in survival prediction, providing a novel approach to support the treatment of GBM patients.

Delineating the glioblastoma stemness by genes involved in cytoskeletal rearrangements and metabolic alterations

Literature data on glioblastoma ongoingly underline the link between metabolism and cancer stemness, the latter is one responsible for potentiating the resistance to treatment, inter alia due to increased invasiveness. In recent years, glioblastoma stemness research has bashfully introduced a key aspect of cytoskeletal rearrangements, whereas the impact of the cytoskeleton on invasiveness is well known. Although non-stem glioblastoma cells are less invasive than glioblastoma stem cells (GSCs), these cells also acquire stemness with greater ease if characterized as invasive cells and not tumor core cells. This suggests that glioblastoma stemness should be further investigated for any phenomena related to the cytoskeleton and metabolism, as they may provide new invasion-related insights. Previously, we proved that interplay between metabolism and cytoskeleton existed in glioblastoma. Despite searching for cytoskeleton-related processes in which the investigated genes might have been involved, not only did we stumble across the relation to metabolism but also reported genes that were found to be implicated in stemness. Thus, dedicated research on these genes in GSCs seems justifiable and might reveal novel directions and/or biomarkers that could be utilized in the future. Herein, we review the previously identified cytoskeleton/metabolism-related genes through the prism of glioblastoma stemness.

Antineoplastic Nature of WWOX in Glioblastoma Is Mainly a Consequence of Reduced Cell Viability and Invasion

Following the discovery of WWOX, research has moved in many directions, including the role of this putative tumor suppressor in the central nervous system and related diseases. The task of determining the nature of WWOX in glioblastoma (GBM) is still considered to be at the initial stage; however, the influence of this gene on the GBM malignant phenotype has already been reported. Because most of the available in vitro research does not consider several cellular GBM models or a wide range of investigated biological assays, the present study aimed to determine the main processes by which WWOX exhibits anticancer properties in GBM, while taking into account the phenotypic heterogeneity between cell lines. Ectopic WWOX overexpression was studied in T98G, DBTRG-05MG, U251MG, and U87MG cell lines that were compared with the use of assays investigating cell viability, proliferation, apoptosis, adhesion, clonogenicity, three-dimensional and anchorage-independent growth, and invasiveness. Observations presenting the antineoplastic properties of WWOX were consistent for T98G, U251MG, and U87MG. Increased proliferation and tumor growth were noted in WWOX-overexpressing DBTRG-05MG cells. A possible explanation for this, arrived at via bioinformatics tools, was linked to the TARDBP transcription factor and expression differences of USP25 and CPNE2 that regulate EGFR surface abundance. Collectively, and despite various cell line-specific circumstances, WWOX exhibits its anticancer nature mainly via a reduction of cell viability and invasiveness of glioblastoma.

Assessing Metabolic Markers in Glioblastoma Using Machine Learning: A Systematic Review

Glioblastoma (GBM) is a common and deadly brain tumor with late diagnoses and poor prognoses. Machine learning (ML) is an emerging tool that can create highly accurate diagnostic and prognostic prediction models. This paper aimed to systematically search the literature on ML for GBM metabolism and assess recent advancements. A literature search was performed using predetermined search terms. Articles describing the use of an ML algorithm for GBM metabolism were included. Ten studies met the inclusion criteria for analysis: diagnostic (n = 3, 30%), prognostic (n = 6, 60%), or both (n = 1, 10%). Most studies analyzed data from multiple databases, while 50% (n = 5) included additional original samples. At least 2536 data samples were run through an ML algorithm. Twenty-seven ML algorithms were recorded with a mean of 2.8 algorithms per study. Algorithms were supervised (n = 24, 89%), unsupervised (n = 3, 11%), continuous (n = 19, 70%), or categorical (n = 8, 30%). The mean reported accuracy and AUC of ROC were 95.63% and 0.779, respectively. One hundred six metabolic markers were identified, but only EMP3 was reported in multiple studies. Many studies have identified potential biomarkers for GBM diagnosis and prognostication. These algorithms show promise; however, a consensus on even a handful of biomarkers has not yet been made.

MiR-202-3p inhibits the proliferation and metastasis of lung adenocarcinoma cells by targeting RRM2

Background

Lung adenocarcinoma (LUAD) is the most common type of lung cancer, and its pathogenesis is still unclear. The present study aimed to investigate the role of miR-202-3p and its downstream target gene, ribonucleotide reductase regulatory subunit M2 (RRM2), in the occurrence and development of LUAD and elucidate the correlation between RRM2 and the clinicopathological stage and prognosis of LUAD.

Methods

The expression of miR-202-3p was analyzed using the CancerMIRNome database and quantitative polymerase chain reaction (qPCR). The effects of miR-202-3p and RRM2 on the proliferation, migration, and invasion of A549 cells were analyzed. A dual luciferase reporter assay was used to verify the targeting of miR-202-3p and RRM2. Additionally, the correlation between RRM2 expression and clinicopathology was analyzed.

Results

(I) MiR-202-3p was lowly expressed in LUAD and the LUAD cell lines. qPCR confirmed that microRNA (miRNA) transfection was effective and sufficient for subsequent experiments. (II) MiR-202-3p inhibited the proliferation, invasion, and migration of LUAD cells. (III) There was a targeting relationship between miR-202-3p and RRM2, and miR-202-3p affected the expression of the RRM2 protein. RRM2 was highly expressed in lung cancer tissue. (IV) RRM2 was associated with the clinicopathological staging of lung cancer. The prognosis of patients with low RRM2 expression was better, and the prognostic sensitivity of RRM2 to lung cancer was high. RRM2 may exert its effects via the Notch pathway. (V) Si-RRM2 inhibited the expression of the RRM2 protein. RRM2 promoted the proliferation, migration, and invasion of LUAD cells. A miR-202-3p inhibitor restored the inhibitory effect of si-RRM2 on LUAD cells.

Conclusions

MiR-202-3p was lowly expressed in lung cancer tissue. MiR-202-3p overexpression inhibited the proliferation and metastasis of lung cancer cells. RRM2 was highly expressed in lung cancer tissue and promoted the proliferation and metastasis of lung cancer cells. MiR-202-3p targeted and inhibited RRM2, thereby reducing the proliferation and metastasis of LUAD cells. LUAD patients with low RRM2 expression had a better prognosis, and the expression level of RRM2 was correlated with the clinical characteristics of lung cancer patients.

Determination of WWOX Function in Modulating Cellular Pathways Activated by AP-2α and AP-2γ Transcription Factors in Bladder Cancer

Following the invention of high-throughput sequencing, cancer research focused on investigating disease-related alterations, often inadvertently omitting tumor heterogeneity. This research was intended to limit the impact of heterogeneity on conclusions related to WWOX/AP-2α/AP-2γ in bladder cancer which differently influenced carcinogenesis. The study examined the signaling pathways regulated by WWOX-dependent AP-2 targets in cell lines as biological replicates using high-throughput sequencing. RT-112, HT-1376 and CAL-29 cell lines were subjected to two stable lentiviral transductions. Following CAGE-seq and differential expression analysis, the most important genes were identified and functionally annotated. Western blot was performed to validate the selected observations. The role of genes in biological processes was assessed and networks were visualized. Ultimately, principal component analysis was performed. The studied genes were found to be implicated in MAPK, Wnt, Ras, PI3K-Akt or Rap1 signaling. Data from pathways were collected, explaining the differences/similarities between phenotypes. FGFR3, STAT6, EFNA1, GSK3B, PIK3CB and SOS1 were successfully validated at the protein level. Afterwards, a definitive network was built using 173 genes. Principal component analysis revealed that the various expression of these genes explains the phenotypes. In conclusion, the current study certified that the signaling pathways regulated by WWOX and AP-2α have more in common than that regulated by AP-2γ. This is because WWOX acts as an EMT inhibitor, AP-2γ as an EMT enhancer while AP-2α as a MET inducer. Therefore, the relevance of AP-2γ in targeted therapy is now more evident. Some of the differently regulated genes can find application in bladder cancer treatment.

Prognostic significance of AP-2α/γ targets as cancer therapeutics

Identifying genes with prognostic importance could improve cancer treatment. An increasing number of reports suggest the existence of successful strategies based on seemingly "untargetable" transcription factors. In addition to embryogenesis, AP-2 transcription factors are known to play crucial roles in cancer development. Members of this family can be used as prognostic factors in oncological patients, and AP-2α/γ transcription factors were previously investigated in our pan-cancer comparative study using their target genes. The present study investigates tumors that were previously found similar with an emphasis on the possible role of AP-2 factors in specific cancer types. The RData workspace was loaded back to R environment and 3D trajectories were built via Monocle3. The genes that met the requirement of specificity were listed using top_markers(), separately for mutual and unique targets. Furthermore, the candidate genes had to meet the following requirements: correlation with AP-2 factor (through Correlation AnalyzeR) and validated prognostic importance (using GEPIA2 and subsequently KM-plotter or LOGpc). Eventually, the ROC analysis was applied to confirm their predictive value; co-dependence of expression was visualized via BoxPlotR. Some similar tumors were differentiated by AP-2α/γ targets with prognostic value. Requirements were met by only fifteen genes (EMX2, COL7A1, GRIA1, KRT1, KRT14, SLC12A5, SEZ6L, PTPRN, SCG5, DPP6, NTSR1, ARX, COL4A3, PPEF1 and TMEM59L); of these, the last four were excluded based on ROC curves. All the above genes were confronted with the literature, with an emphasis on the possible role played by AP-2 factors in specific cancers. Following ROC analysis, the genes were verified using immunohistochemistry data and progression-related signatures. Staining differences were observed, as well as co-dependence on the expression of e.g. CTNNB1, ERBB2, KRAS, SMAD4, EGFR or MKI67. In conclusion, prognostic value of targets suggested AP-2α/γ as candidates for novel cancer treatment. It was also revealed that AP-2 targets are related to tumor progression and that some mutual target genes could be inversely regulated.