Transcriptional profiling of HOXA9-regulated genes in human glioblastoma cell models

The data here described pertain to the article by Pojo et al. (2015) [10] titled “A transcriptomic signature mediated by HOXA9 promotes human glioblastoma initiation, aggressiveness and resistance to temozolomide” (Pojo et al., 2015 [10]). HOX genes are part of the homeobox gene family, which encodes transcription factors crucial during embryonic development (Grier et al., 2005; Pearson et al., 2005 [6], [9]) and also in postdevelopmental regulation (Neville et al., 2002; Yamamoto et al., 2003; Takahashi et al., 2004; Morgan 2006 [8], [14], [13], [7]). Alterations interfering with the regulation of these genes may lead to tumorigenesis in adults. Due to their contributions in the control of important cellular processes, the deregulation of HOX genes is ultimately correlated with cancer treatment failure and patients' poor prognosis (Golub et al., 1999; Abdel-Fattah et al., 2006 [5], [1]; Costa et al., 2010 [4]; Pojo et al., 2015 [10]). Recently, our studies showed that HOXA9 overexpression is associated with poor prognosis in patients with glioblastoma (GBM), the most common and most malignant primary brain tumor. Mechanistically, HOXA9 is associated with resistance to chemotherapy and with pro-proliferative, pro-invasive and anti-apoptotic features (Costa et al., 2010 [4]; Pojo et al., 2015 [10]) in GBM in vitro models. Since HOXA9 is a transcription factor, its target genes can be the true biological effectors of its aggressiveness. In this context, whole genome Agilent's microarrays were used to obtain the full transcriptome of HOXA9 in a variety of GBM cell models, including human immortalized astrocytes, established GBM cell lines, and GBM patient-derived cell cultures. Here, we provide detailed methods, including experimental design and microarray data analyses, which can be accessed in Gene Expression Omnibus (GEO) under the accession number GSE56517. Additional interpretation of the data is included and supplemented in (Pojo et al., 2015 [10]).