Papel del factor de transcripción ATF5 en la Enfermedad de Huntington

Abstract

Activating transcription factor-5 (ATF5) is a stress-response transcription factor induced upon different cell stressors like fasting, amino-acid limitation, cadmium or arsenite. ATF5 is also an effector of the unfolded protein response (UPR), a cellular reaction to endoplasmic reticulum stress (ERS). As a mediator of the UPR, ATF5 induces transcription of proteostatic and anti-apoptotic factors, like MCL1. In the brain, high ATF5 levels are found in gliomas and in neural progenitor cells, which need to decrease their ATF5 levels to differentiate into mature neurons or glia. This initially led to believe that ATF5 is not expressed in adult neurons. More recently, we reported basal neuronal ATF5 expression in adult mouse brain and its neuroprotective induction during UPR in a mouse model of status epilepticus. Here we aimed to explore whether ATF5 is also expressed by neurons in human brain both in basal conditions and in Huntington's disease (HD), where UPR has been described to be partially impaired due to defective ATF6 processing. Apart from confirming that ATF5 is present in human adult neurons, here we report accumulation of ATF5 within the characteristic polyglutamine-containing neuronal nuclear inclusions in brains of HD patients and R6/1 mice. This correlates with decreased levels of soluble ATF5 and of its anti-apoptotic target MCL1. We then confirmed the deleterious effect of ATF5 deficiency in a Caenorhabditis elegans model of polyglutamine-induced toxicity. We also report the generation of two transgenic mouse lines for the conditional overexpression of ATF5 in adult neurons or astrocytes. However, despite a visible induction of transgenic mRNA we failed to observe an accumulation of the recombinant protein in any of the structures or times analyzed. Induction of ERS in transgenic primary neurons did not induce accumulation of the exogenous protein. Additionally, crossbreeding of transgenic animals and R6/1 mice was not effective in the induction of recombinant ATF5 overexpression nor the attenuation of pathologic phenotype. Finally, both the transfection of ATF5 and the transfection of MCL1 separately reduced apoptosis in a cellular model of HD. These results reflect that decreased ATF5 in HD, probably secondary to sequestration into inclusions, renders neurons more vulnerable to mutant huntingtin-induced apoptosis and that ATF5-increasing interventions might have therapeutic potential for HD.