Results achieved and their exploitation and dissemination
1. Discovery of repurposed drugs targeting UPR. Discovery that trazodone modulates the UPR in a safe and effective way preventing neurodegeneration in two models of disease. (Halliday M. et al., Brain 2017). Scientific Commentary: Brain (doi:10.1093/brain/awx107); Research Highlights: Nature Reviews Neurology; Featured in 62 news outlets, including
http://www.bbc.co.uk/news/health-39641123?yptr=yahoo(se abrirá en una nueva ventana);
http://www.bbc.co.uk/news/health-39641123(se abrirá en una nueva ventana))
2. Role of non-cell autonomous UPR signalling in neurodegeneration
Our previous work established the cell-autonomous role of UPR over-activation in neurons (Moreno et al., 2012), but the role of this signalling in astrocytes and its contribution to disease was unknown. Recent interest in astrocyte activation states has raised the fundamental question of how these cells, normally essential for synapse and neuronal maintenance, become pathogenic. We found P-PERK signalling generates a distinct reactivity state in astrocytes that alters the astrocytic secretome, leading to loss of synaptogenic function in vitro. We found that the same P-PERK-dependent astrocyte reactivity state is harmful to neurons in vivo, in mice with prion neurodegeneration. Targeting this signalling exclusively in astrocytes in prion-diseased animals was sufficient to prevent neuronal loss and significantly prolonged survival. It was as effective as targeting neuronal UPR dysregulation (Fig.1). Thus, the astrocyte reactivity state resulting from UPR over-activation represents a distinct, non-cell-autonomous, pathogenic mechanism that potentially provides multiple new targets for restoring synapses and neuroprotection in neurodegenerative disease (Smith et al., 2020) (2 F1000 prime recommendations; Research highlights: Nature Reviews Neuroscience 21, 119 (2020) and Research highlights: Science Translational Medicine: DOI: 10.1126/scitranslmed.aba2916.
3. Mechanisms of non-canonical activation/regulation of PERK signalling
How PERK signalling is affected by proteins aggregating in the cytoplasm rather than the ER lumen is currently unknown. We propose that in an additional level of control of the production of eIF2α-P by PERK activation exists, regulated through a non-canonical mechanism driven by interactions within PERK’s cytoplasmic domain.
We have made the following progress:
3.1 Selective modulation of PERK signalling via residue T799 in kinase domain insert loop. Using genetic and pharmacological approaches in vitro and in vivo, we discovered a new mechanism for selectively attenuating PERK downstream signalling by targeting a specific phosphorylation site, residue T799 (a target of Akt), within PERK’s kinase insert loop. Phosphorylation at T799 reduces PERK signalling and eIF2α-P levels, independently of PERK kinase activity and of wider ISR activation. Induction of T799 phosphorylation with a small molecule activator of Akt reduced PERK signalling in prion-diseased mice preventing neuronal loss (Fig.2) and significantly increasing survival, importantly without pancreatic toxicity that plagues direct PERK kinase inhibition (Hughes et al., 2020) (cover story Science Signaling; top story on Science website 08/20).
3.2. discovery that disturbing neuronal mitochondrial function activates the UPR to induce cytokine Fgf21 (Restelli LM. et al., Cell Rep 2018)
3.3 Identified, using targeted proteomics, a number of proteins which interact with PERK, providing new therapeutic target for modulating PERK/eIF2α signalling (manuscript in preparation)
4. Defining nature of translatome changes in health and diseaese (manuscript in preparation)
5. Translational studies: detecting changes in global protein synthesis rates in human disease We are running an experimental medicines study to measure cerebral protein synthesis rates (rCPS) in the brains of individuals with AD and age-matched controls. This is to test the hypothesis of whether or not UPR-mediated translational failure is relevant in human disease, as in mouse models. If so, we predict UPR/ISR modulators would be therapeutic in humans as in mice. (In progress).
6. Several invited review articles published
