CELF2 Sustains a Proliferating/OLIG2+ Glioblastoma Cell Phenotype via the Epigenetic Repression of SOX3.

Glioblastomas (GBs) are incurable brain tumors. The persistence of aggressive stem-like tumor cells after cytotoxic treatments compromises therapeutic efficacy, leading to GBM recurrence. Forcing the GBM cells to irreversibly abandon their aggressive stem-like phenotype may offer an alternative to conventional cytotoxic treatments. Here, we show that the RNA binding protein CELF2 is strongly expressed in mitotic and OLIG2-positive GBM cells, while it is downregulated in differentiated and non-mitotic cells by miR-199a-3p, exemplifying GBM intra-tumor heterogeneity. Using patient-derived cells and human GBM samples, we demonstrate that CELF2 plays a key role in maintaining the proliferative/OLIG2 cell phenotype with clonal and tumorigenic properties. Indeed, we show that CELF2 deficiency in patient-derived GSCs drastically reduced tumor growth in the brains of nude mice. We further show that CELF2 promotes TRIM28 and G9a expression, which drive a H3K9me3 epigenetic profile responsible for the silencing of the SOX3 gene. Thus, CELF2, which is positively correlated with OLIG2 and Ki67 expression in human GBM samples, is inversely correlated with SOX3 and miR-199a-3p. Accordingly, the invalidation of SOX3 in CELF2-deficient patient-derived cells rescued proliferation and OLIG2 expression. Finally, patients expressing SOX3 above the median level of expression tend to have a longer life expectancy. CELF2 is therefore a crucial target for the malignant potential of GBM and warrants attention when developing novel anticancer strategies.

A network embedding approach to identify active modules in biological interaction networks.

The identification of condition-specific gene sets from transcriptomic experiments is important to reveal regulatory and signaling mechanisms associated with a given cellular response. Statistical methods of differential expression analysis, designed to assess individual gene variations, have trouble highlighting modules of small varying genes whose interaction is essential to characterize phenotypic changes. To identify these highly informative gene modules, several methods have been proposed in recent years, but they have many limitations that make them of little use to biologists. Here, we propose an efficient method for identifying these active modules that operates on a data embedding combining gene expressions and interaction data. Applications carried out on real datasets show that our method can identify new groups of genes of high interest corresponding to functions not revealed by traditional approaches. Software is available at https://github.com/claudepasquier/amine.

Naegleria genus pangenome reveals new structural and functional insights into the versatility of these free-living amoebae.

Introduction: Free-living amoebae of the Naegleria genus belong to the major protist clade Heterolobosea and are ubiquitously distributed in soil and freshwater habitats. Of the 47 Naegleria species described, N. fowleri is the only one being pathogenic to humans, causing a rare but fulminant primary amoebic meningoencephalitis. Some Naegleria genome sequences are publicly available, but the genetic basis for Naegleria diversity and ability to thrive in diverse environments (including human brain) remains unclear. Methods: Herein, we constructed a high-quality Naegleria genus pangenome to obtain a comprehensive catalog of genes encoded by these amoebae. For this, we first sequenced, assembled, and annotated six new Naegleria genomes. Results and Discussion: Genome architecture analyses revealed that Naegleria may use genome plasticity features such as ploidy/aneuploidy to modulate their behavior in different environments. When comparing 14 near-to-complete genome sequences, our results estimated the theoretical Naegleria pangenome as a closed genome, with 13,943 genes, including 3,563 core and 10,380 accessory genes. The functional annotations revealed that a large fraction of Naegleria genes show significant sequence similarity with those already described in other kingdoms, namely Animalia and Plantae. Comparative analyses highlighted a remarkable genomic heterogeneity, even for closely related strains and demonstrate that Naegleria harbors extensive genome variability, reflected in different metabolic repertoires. If Naegleria core genome was enriched in conserved genes essential for metabolic, regulatory and survival processes, the accessory genome revealed the presence of genes involved in stress response, macromolecule modifications, cell signaling and immune response. Commonly reported N. fowleri virulence-associated genes were present in both core and accessory genomes, suggesting that N. fowleri's ability to infect human brain could be related to its unique species-specific genes (mostly of unknown function) and/or to differential gene expression. The construction of Naegleria first pangenome allowed us to move away from a single reference genome (that does not necessarily represent each species as a whole) and to identify essential and dispensable genes in Naegleria evolution, diversity and biology, paving the way for further genomic and post-genomic studies.

NvERTx: a gene expression database to compare embryogenesis and regeneration in the sea anemone Nematostella vectensis.

For over a century, researchers have been comparing embryogenesis and regeneration hoping that lessons learned from embryonic development will unlock hidden regenerative potential. This problem has historically been a difficult one to investigate because the best regenerative model systems are poor embryonic models and vice versa. Recently, however, there has been renewed interest in this question, as emerging models have allowed researchers to investigate these processes in the same organism. This interest has been further fueled by the advent of high-throughput transcriptomic analyses that provide virtual mountains of data. Here, we present Nematostella vectensis Embryogenesis and Regeneration Transcriptomics (NvERTx), a platform for comparing gene expression during embryogenesis and regeneration. NvERTx consists of close to 50 transcriptomic data sets spanning embryogenesis and regeneration in Nematostella These data were used to perform a robust de novo transcriptome assembly, with which users can search, conduct BLAST analyses, and plot the expression of multiple genes during these two developmental processes. The site is also home to the results of gene clustering analyses, to further mine the data and identify groups of co-expressed genes. The site can be accessed at http://nvertx.kahikai.org.