Search

Diffuse gliomas are the commonest of malignant brain tumors with high-grade tumors carrying dismal prognosis. Preclinical models have proven themselves as poor predictors of clinical efficacy, attributed to the lack of a comparable tumor microenvironment. Comparative genomics of canine and human gliomas provide an attractive alternative modality to identify conserved drivers and underlying mutational processes of glioma as well as evaluate life history tradeoffs related to tissue context and aging that can shape type and relative timing of drivers in gliomagenesis. We performed a comparative genomics analysis between whole genome-, exome-, transcriptome- and methylation-sequencing of 77 canine gliomas, and human pediatric (n=217) and adult gliomas (n=822). We found alterations in common with those in human pediatric and adult gliomas in the Tp53 and cell cycle pathways, receptor tyrosine kinase and Idh1 R132 mutations. Canine gliomas showed similarity to human pediatric gliomas in terms of lower mutational rates (0.2–0.29 per megabase), hotspot mutations and amplifications of Pdgfrα, and robust aneuploidy constrained within the syntenic regions of Pdgfrα and Myc, but also in the known pediatric drivers, Hist1 and Acvr1 genes. A mutational signature reflecting homologous repair defect was detected in canine and pediatric but not adult gliomas, potentially resulting in the observed higher rates of genomic instability. Canine gliomas in majority classified as pediatric tumors using a commonly used human brain methylation classifier (Capper et al. 2018) and cell-of-origin analysis by deconvoluting canine transcriptome using lineage-specific gene signatures. By providing a large canine glioma genomic-sequencing dataset and comparing it with human glioma, our study provides unique insights into glioma etiology and the chronology of glioma-causing somatic alterations. Further, our results effectively position preclinical models of spontaneous canine glioma for use in understanding glioma drivers, and evaluate novel therapies targeting aneuploidy, especially for pediatric brain tumors.

Calretinin (CALB2) is a diagnostic marker for epithelioid mesothelioma. It is also a prognostic marker since patients with tumors expressing high calretinin levels have better overall survival. Silencing of calretinin decreases viability of epithelioid mesothelioma cells. Our aim was to elucidate mechanisms regulating calretinin expression in mesothelioma. Analysis of calretinin transcript and protein suggested a control at the mRNA level. Treatment with 5-aza-2′-deoxycytidine and analysis of TCGA data indicated that promoter methylation is not likely to be involved. Therefore, we investigated CALB2 promoter by analyzing ~1kb of genomic sequence surrounding the transcription start site (TSS) + 1 using promoter reporter assay. Deletion analysis of CALB2 proximal promoter showed that sequence spanning the –161/+80bp region sustained transcriptional activity. Site-directed analysis identified important cis- regulatory elements within this –161/+80bp CALB2 promoter. EMSA and ChIP assays confirmed binding of NRF-1 and E2F2 to the CALB2 promoter and siRNA knockdown of NRF-1 led to decreased expression of calretinin. Cell synchronization experiment showed that calretinin expression was cell cycle regulated with a peak of expression at G1/S phase. This study provides the first insight in the regulation of CALB2 expression in mesothelioma cells.