Facultad de Medicina

Transcription factors are common targets of epigenetic inactivation in human cancer. Promoter hypermethylation and subsequent silencing of transcription factors can lead to further deregulation of their targets. In this study, we explored the potential epigenetic deregulation in cancer of Ikaros family genes, which code for essential transcription factors in cell differentiation and exhibit genetic defects in hematologic neoplasias. Unexpectedly, our analysis revealed that Ikaros undergoes very specific promoter hypermethylation in colorectal cancer, including in all the cell lines studied and around 64% of primary colorectal adenocarcinomas, with increasing proportions in advanced Duke's stages. Ikaros hypermethylation occurred in the context of a novel long-range epigenetic silencing (LRES) region. Reintroduction of Ikaros in colorectal cancer cells, ChIP-chip analysis, and validation in primary samples led us to identify a number of direct targets that are possibly related with colorectal cancer progression. Our results not only provide the first evidence that LRES can have functional specific effects in cancer but also identify several deregulated Ikaros targets that may contribute to progression in colorectal adenocarcinoma

Dictyostelium and human MidA are homologous proteins that belong to a family of proteins of unknown function called DUF185. Using yeast two-hybrid screening and pull-down experiments, we showed that both proteins interact with the mitochondrial complex I subunit NDUFS2. Consistent with this, Dictyostelium cells lacking MidA showed a specific defect in complex I activity, and knockdown of human MidA in HEK293T cells resulted in reduced levels of assembled complex I. These results indicate a role for MidA in complex I assembly or stability. A structural bioinformatics analysis suggested the presence of a methyltransferase domain; this was further supported by site-directed mutagenesis of specific residues from the putative catalytic site. Interestingly, this complex I deficiency in a Dictyostelium midA− mutant causes a complex phenotypic outcome, which includes phototaxis and thermotaxis defects. We found that these aspects of the phenotype are mediated by a chronic activation of AMPK, revealing a possible role of AMPK signaling in complex I cytopathology

The natural history of cancers associated with virus exposure is intriguing, since only a minority of human tissues infected with these viruses inevitably progress to cancer. However, the molecular reasons why the infection is controlled or instead progresses to subsequent stages of tumorigenesis are largely unknown. In this article, we provide the first complete DNA methylomes of double-stranded DNA viruses associated with human cancer that might provide important clues to help us understand the described process. Using bisulfite genomic sequencing of multiple clones, we have obtained the DNA methylation status of every CpG dinucleotide in the genome of the Human Papilloma Viruses 16 and 18 and Human Hepatitis B Virus, and in all the transcription start sites of the Epstein-Barr Virus. These viruses are associated with infectious diseases (such as hepatitis B and infectious mononucleosis) and the development of human tumors (cervical, hepatic, and nasopharyngeal cancers, and lymphoma), and are responsible for 1 million deaths worldwide every year. The DNA methylomes presented provide evidence of the dynamic nature of the epigenome in contrast to the genome. We observed that the DNA methylome of these viruses evolves from an unmethylated to a highly methylated genome in association with the progression of the disease, from asymptomatic healthy carriers, through chronically infected tissues and pre-malignant lesions, to the full-blown invasive tumor. The observed DNA methylation changes have a major functional impact on the biological behavior of the viruses.

Predictions of intramolecular residue–residue contacts were assessed as part of the seventh community-wide Critical Assessment of Structure Prediction experiment (CASP7). As in past assessments, we focused on contacts that lie far apart in sequence as these are likely to be more informative in predicting protein structure. One lab did somewhat better than others according to our assessment, and there is some reason to think that this lab's results represent progress over CASP6. In general, contacts inferred from 3D structural predictions are similar in accuracy to those predicted by contact prediction methods. However, contact prediction methods were more accurate for some targets.