Identification and Characterization of FoxO-target genes by Chromatin Immunoprecipitation
Year: 2009
Type: dissertation
Abstract: SUMMARY: Loss of muscle mass occurs in many debilitating diseases including sepsis, burn injury, cancer, diabetes, heart and renal failure, but also in other catabolic conditions like denervation, immobilization, glucocorticoid treatment, or fasting. The reduction in muscle cell size is caused by an increase of the activity of proteolytic systems, such as the ubiquitin-proteasome system and the lysosome autophagic system. Gene expression studies revealed a set of genes which are commonly up- and down-regulated in different atrophying conditions. These genes are called atrophic related genes or atrogenes (Jagoe and Goldberg, 2001; LECKER et al., 2004). Two of the most induced genes, whose activation precedes muscle loss, are muscle-specific and encode for ubiquitin ligase. These two enzymes, named MURF1 and Atrogin-1, are rate-limiting in the ubiquitination process and therefore, regulate the proteasomal-dependent protein breakdown (Gomes et al., 2001). Atrogin-1 is controlled by FoxO transcription factors, which are in turn inhibited by the IGF1-PI3K-Akt pathway. In the absence of growth factors, like IGF1 or insulin, AKT is not activated and does not block FoxOs which can translocate into the nuclei and interact with promoters of their target genes (Sandri et al., 2004). In mammalian muscle, there are three FoxO transcription factors: FoxO1, FoxO3 and FoxO4. All FoxOs share the same “forkhead box” which contains the DNA-binding domain. Previous studies showed FoxO1 and FoxO3 to be involved in muscle atrophy (Kamei et al., 2004). However, it is completely unknown how many genes are required for FoxO-mediated muscle loss and whether these genes are or are not commonly regulated by both FoxO1 and FoxO3. In this study we proprosed to understand which are the atrophy-related genes target of FoxO1 and FoxO3. We set up ChIP technology in order to dissect in vivo, the interaction between FoxOs and promoters of FoxO’s target genes. This technique combines the specificity of immunoprecipitation in purifying specific proteins with the power of PCR to amplify specific sequences of DNA. First, we identified the potential FoxO1 and FoxO3 potential binding sites on the atrogin-1 promoter. Bioinformatic analysis of atrogin1 promoter showed that there are 14 potential FoxO binding sites. Second, we used a cell culture system to establish different parameters for a successful ChIP . We then moved to in vivo studies of both overexpressed FoxOs and endogenous FoxO proteins. Tibialis Anterior (TA) of adult CD1 mice were transfected with constitutively active HA tagged: ca-FoxO3 or ca-.FoxO1. After seven days muscles were collected, lysated and analysed for ChIP. ChIP studies demonstrated that FoxO3 strongly binds many sites of atrogin-1 promoter, while FoxO1 weakly interacts with only one site. To better understand which sites are physiologically relevant for atrogin-1 regulation endogenous FoxO1 and FoxO3 were immunoprecipitated. Fasting was used as a model of atrophy, since this model strongly activates FoxOs increasing atrogin-1 levels to more than 10 fold. ChIP experiments were performed on TA muscles of fed and starved mice. Interestingly, we could demonstrate by ChIP that FoxO3 preferentially binds chromatin at three sites of atrogin-1 promoter. These sites showed also an increase of histone acetylation during fasting. FoxO1 displayed a weak interaction with atrogin-1 promoter. To understand if both FoxOs are able to regulate atrogin-1 expression, we used a functional assay in which the atrogin-1 promoter was cloned upstream the luciferase gene. This reporter was transfected together with FoxO1 or FoxO3 in the tibialis anterior muscle and seven days later the assay was performed. This approach confirmed that FoxO3, but not FoxO1, could transactivate the atrogin-1 promoter. Since both FoxOs can induce muscle loss, the probability that other genes are required for muscle atrophy was seriously considered. Indeed overexpression of atrogin-1 per sè is not sufficient to induce muscle wasting. To understand how FoxOs cause muscle loss we focused on the atrophy-related genes which belong to other proteolytic systems and especially to autophagy-lysosome system. In fact LC3 and Binp3 are among the upregulated atrogenes and are rate limiting proteins in autophagosome formation. Their promoters contain several potential FoxO binding sites and ChIP experiments showed that FoxO3 binds chromatin of LC3 and Bnip3 promoters in specific sites during starvation. Functional promoter assay confirmed that FoxO3 binding on LC3 promoter is essential to activate transcription. Moreover, gain and loss of function experiments showed that FoxO3 regulates Binp3 protein levels. Altogether these findings show that FoxO3, but not FoxO1, coordinates the two major proteolytic systems of the cell by regulating the promoter activity of atrogin-1, LC3 and Binp3 genes. However these results have not clarified yet which atrogenes are regulated by FoxO1 and if there are other atrogenes under FoxO3 control. For this reason, we extend the ChIP assay to a genome wide analysis using a ChIP-on-Chip approach. This technique combines the Chromatin immunoprecipitation with hybridization on genomic microarray to allow a global analysis of protein–DNA interaction. The technique was set up using a cell culture system. Differentiated C2C12 myotubes were infected with adenovirus expressing HA tagged c.a.FoxO3 and 48 hours later a ChIP was performed using anti-HA antibody. The results of ChIP on Chip identified several binding sites on genome. Many of these regions are promoters for transcription factors, enzymes of different metabolic pathways and different kinases involved in regulation of different pathways. Importantly a FoxO3 binding site was detected in regulatory regions of other eight atrophy-related genes. When the function of this binding sites was validated it was found that 5 of these atrogenes were modulated by FoxO3. In conclusion, this PhD project establish the technical basis for the identification of the fundamental genes, under the control of a certain transcription factor, involved in the atrophy program. All these data will contribute to the development of new pharmacological strategy against muscle loss.
Author: Giulia Milan
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W2281379115