RESUMO
The p90 ribosomal S6 kinase (RSK) family, a downstream target of Ras/extracellular signal-regulated kinase signaling, can mediate cross-talk with the mammalian target of rapamycin complex 1 pathway. As RSK connects two oncogenic pathways in gliomas, we investigated the protein levels of the RSK isoforms RSK1-4 in nontumoral brain (NB) and grade I-IV gliomas. When compared to NB or low-grade gliomas (LGG), a group of glioblastomas (GBMs) that excluded long-survivor cases expressed higher levels of RSK1 (RSK1hi ). No difference was observed in RSK2 median-expression levels among NB and gliomas; however, high levels of RSK2 in GBM (RSK2hi ) were associated with worse survival. RSK4 expression was not detected in any brain tissues, whereas RSK3 expression was very low, with GBM demonstrating the lowest RSK3 protein levels. RSK1hi and, to a lesser extent, RSK2hi GBMs showed higher levels of phosphorylated RSK, which reveals RSK activation. Transcriptome analysis indicated that most RSK1hi GBMs belonged to the mesenchymal subtype, and RSK1 expression strongly correlated with gene expression signature of immune infiltrates, in particular of activated natural killer cells and M2 macrophages. In an independent cohort, we confirmed that RSK1hi GBMs exclude long survivors, and RSK1 expression was associated with high protein levels of the mesenchymal subtype marker lysosomal protein transmembrane 5, as well as with high expression of CD68, which indicated the presence of infiltrating immune cells. An RSK1 signature was obtained based on differentially expressed mRNAs and validated in public glioma datasets. Enrichment of RSK1 signature followed glioma progression, recapitulating RSK1 protein expression, and was associated with worse survival not only in GBM but also in LGG. In conclusion, both RSK1 and RSK2 associate with glioma malignity, but displaying isoform-specific peculiarities. The progression-dependent expression and association with immune infiltration suggest RSK1 as a potential progression marker and therapeutic target for gliomas.
Assuntos
Neoplasias Encefálicas/metabolismo , Glioma/metabolismo , Linfócitos do Interstício Tumoral/imunologia , Proteínas de Membrana/metabolismo , Proteínas Quinases S6 Ribossômicas 90-kDa/metabolismo , Transcriptoma/imunologia , Antígenos CD/metabolismo , Antígenos de Diferenciação Mielomonocítica/metabolismo , Neoplasias Encefálicas/genética , Neoplasias Encefálicas/imunologia , Neoplasias Encefálicas/mortalidade , Bases de Dados Genéticas , Perfilação da Expressão Gênica , Regulação Neoplásica da Expressão Gênica/genética , Regulação Neoplásica da Expressão Gênica/imunologia , Glioblastoma/genética , Glioblastoma/metabolismo , Glioma/genética , Glioma/imunologia , Glioma/secundário , Humanos , Imuno-Histoquímica , Células Matadoras Naturais/metabolismo , Macrófagos/metabolismo , Proteínas de Membrana/genética , Gradação de Tumores , Fosforilação , Isoformas de Proteínas , Proteínas Quinases S6 Ribossômicas 90-kDa/genética , Transdução de Sinais/genética , Transdução de Sinais/imunologia , Transcriptoma/genéticaRESUMO
Background: Human biological material has become an important resource for biomedical research. Tumor Biobanks are facilities that collect, store and distribute samples of tumor and normal tissue for further use in basic and translational cancer research. mRNA-translation has been demonstrated to modulate protein levels and is considered a fundamental post-transcriptional mechanism of gene expression regulation. Thus, determining translation efficiencies of individual mRNAs in human tumors may add another layer of information that contributes to the understanding of tumorigenic pathways. To analyze the RNAs actively engaged in translation, RNAs associated with ribosomes (polysomes) are isolated, identified and compared to total RNA. However, the application of this technique in human tumors depends on the stability of the polysomal structure under Biobank storage conditions that usually consists of ultra-low temperature. Since the effect of freezing on the stability of the polysomal structure in stored tumor samples is not known, it is essential to evaluate this factor in the frozen samples, validating the use of biobank samples in studies of translational efficiency. Methods: Xenograft tumors were divided in two parts, half was subject to immediate processing, and half was frozen for posterior analysis. Both parts were subject to polysomal separation, RNA extraction and identification through RNAseq. Results: It was possible to successfully extract and identify total and polysomal RNA from both fresh and frozen tumoral tissue. The quantification of the polysome profile indicated no difference in the translational efficiency estimated in fresh versus frozen tissue. Gene expression data from the fresh versus frozen tissues were compared and the correlation between the polysome associated fresh x frozen (R = 0,89) and total fresh x frozen (0,90) mRNAs was calculated. No difference was identified between the two conditions. Conclusions: We demonstrated that tissue freezing does not affect the polysomal structure, consequently validating the viability of the use of biobank stored tissue for polysome associated RNA analysis (AU)
