Congratulations to Xinghua Lu and Songjian Lu for their recent article published in Neuro Oncology
Tianzhi Huang, Angel A. Alvarez, Rajendra P. Pangeni, Craig M. Horbinski, Songjian Lu, Sung-Hak Kim, C. David James, Jeffery J. Raizer, John A. Kessler, Cameron W. Brenann, Erik P. Sulman, Gaetano Finocchiaro, Ming Tan, Ryo Nishikawa, Xinghua Lu, Ichiro Nakano, Bo Hu, and Shi-Yuan Chenga (2016). A regulatory circuit of miR-125b/miR-20b and Wnt signalling controls glioblastoma phenotypes through FZD6-modulated pathways. Neuro Oncol (2016) 18 (suppl 6): vi42. doi: 10.1093/neuonc/now212.169
Molecularly defined subclassification is associated with phenotypic malignancy of glioblastoma (GBM). However, current understanding of the molecular basis of subclass conversion that is often involved in GBM recurrence remain rudimentary at best. Here we report that canonical Wnt signalling that is active in proneural (PN) but inactive in mesenchymal (MES) GBM, along with miR-125b and miR-20b that are expressed at high levels in PN compared with MES GBM, comprise a regulatory circuit involving TCF4-miR-125b/miR-20b-FZD6. FZD6 acts as a negative regulator of this circuit by activating CaMKII–TAK1–NLK signalling, which, in turn, attenuates Wnt pathway activity while promoting STAT3 and NF-κB signalling that are important regulators of the MES-associated phenotype. These findings are confirmed by targeting differentially enriched pathways in PN versus MES GBM that results in inhibition of distinct GBM subtypes. Correlative expressions of the components of this circuit are prognostic relevant for clinical GBM. Our findings provide insights for understanding GBM pathogenesis and for improving treatment of GBM.
Glioblastoma (GBM) is the most common primary malignant brain tumour1. Integrated genomic analyses by The Cancer Genome Atlas (TCGA) Network revealed that GBM can be subclassified into three or four clinically relevant types: proneural (PN), neural, mesenchymal (MES) and classical GBM, each characterized by a distinct gene expression signature, as well as by specific genetic alterations2,3,4. Studies have shown that GBM with MES-associated gene expression signatures have worse prognosis compared with those with a PN subtype4,5. Glioma stem cells (GSCs) are tumour cell subpopulations in GBM that have stem cell-like properties, and are considered important contributors to GBM aggressiveness, recurrence and therapy resistance6. GSCs that grow as tumour spheres in vitro retain the characteristics of the original tumour and can be stratified according to tumour subtype7,8. PN and MES GSCs (henceforth referred as to glioma spheres) have distinct dysregulated signalling pathway ‘signatures' that contribute to their malignant phenotypes and response to radiation9. The conversion of PN to MES subtype in GBM has been reported by others, and usually in association with tumour recurrence following initial treatment10,11,12. Currently, there is substantial interest in determining whether GBM subtype-defining mutations, expression patterns, and signalling pathway dysregulation predict tumour response and adaptation to targeted therapies13,14.
MicroRNAs (miRs) are small noncoding RNAs that act as modulators of gene expression in all multicellular organisms. A unique feature of miRs is that a single miR can simultaneously regulate the expression of multiple target genes, thereby affecting numerous cellular behaviours including differentiation, proliferation, and survival15. Dysregulated miR expression plays an important role in cancer16,17. Distinct miR expression patterns have been described in cancer stem cells (CSCs) from individual tumours, including GBM, suggesting cancer cell type-specific miR functions18. Furthermore, for GBM, there are onco-miR clusters that influence survivorship19, underscoring the effects of miRs on tumour heterogeneity.
Wnt signalling is involved in embryonic development, adult tissue self-renewal, tissue repair, and cancer20. Canonical Wnt signalling is mediated from Frizzled (FZD) family receptors to β-catenin, which acts with transcription factors to induce the expression of genes that regulate differentiation and self-renewal, such as LEF1, SOX2, JAG1 and ID2 (ref. 21). Positive regulators of Wnt signalling include co-receptors low-density lipoprotein receptor-related protein 5 and 6 (LRP5/6), and Wnt ligand WNT3a, whereas negative regulators include FZD6, APC, AXIN and GSK3β. Canonical Wnt signalling regulates phenotypes of normal and CSCs, whereas non-canonical Wnt signalling controls cell movement and polarity22,23. In addition, several miRs have been shown to influence tumorigenesis by affecting the Wnt pathway activity24,25. In this study, we report that miR-125b and miR-20b are key mediators of the Wnt activity that are active in PN, but not in MES GBM. miR-125b, miR-20b and the Wnt pathway establish a regulatory circuit that includes FZD6 as the key negative regulator of the Wnt signalling. Due to the GBM subtype specificity of Wnt signalling, our results have important clinical implications for the development of targeted therapies against specific tumour subtypes.