Progression of pathology in PINK1-deficient mouse brain from splicing via ubiquitination, ER stress, and mitophagy changes to neuroinflammation
Sylvia Torres-Odio,Jana Key,Hans-Hermann Hoepken,Júlia Canet-Pons,Lucie Valek,Bastian Roller,Michael Walter,Blas Morales-Gordo,David Meierhofer,Patrick N Harter,Michel Mittelbronn,Irmgard Tegeder,Suzana Gispert,Georg Auburger
DOI: https://doi.org/10.1186/s12974-017-0928-0
2017-08-02
Abstract:Background: PINK1 deficiency causes the autosomal recessive PARK6 variant of Parkinson's disease. PINK1 activates ubiquitin by phosphorylation and cooperates with the downstream ubiquitin ligase PARKIN, to exert quality control and control autophagic degradation of mitochondria and of misfolded proteins in all cell types. Methods: Global transcriptome profiling of mouse brain and neuron cultures were assessed in protein-protein interaction diagrams and by pathway enrichment algorithms. Validation by quantitative reverse transcriptase polymerase chain reaction and immunoblots was performed, including human neuroblastoma cells and patient primary skin fibroblasts. Results: In a first approach, we documented Pink1-deleted mice across the lifespan regarding brain mRNAs. The expression changes were always subtle, consistently affecting "intracellular membrane-bounded organelles". Significant anomalies involved about 250 factors at age 6 weeks, 1300 at 6 months, and more than 3500 at age 18 months in the cerebellar tissue, including Srsf10, Ube3a, Mapk8, Creb3, and Nfkbia. Initially, mildly significant pathway enrichment for the spliceosome was apparent. Later, highly significant networks of ubiquitin-mediated proteolysis and endoplasmic reticulum protein processing occurred. Finally, an enrichment of neuroinflammation factors appeared, together with profiles of bacterial invasion and MAPK signaling changes-while mitophagy had minor significance. Immunohistochemistry showed pronounced cellular response of Iba1-positive microglia and GFAP-positive astrocytes; brain lipidomics observed increases of ceramides as neuroinflammatory signs at old age. In a second approach, we assessed PINK1 deficiency in the presence of a stressor. Marked dysregulations of microbial defense factors Ifit3 and Rsad2 were consistently observed upon five analyses: (1) Pink1 -/- primary neurons in the first weeks after brain dissociation, (2) aged Pink1 -/- midbrain with transgenic A53T-alpha-synuclein overexpression, (3) human neuroblastoma cells with PINK1-knockdown and murine Pink1 -/- embryonal fibroblasts undergoing acute starvation, (4) triggering mitophagy in these cells with trifluoromethoxy carbonylcyanide phenylhydrazone (FCCP), and (5) subjecting them to pathogenic RNA-analogue poly(I:C). The stress regulation of MAVS, RSAD2, DDX58, IFIT3, IFIT1, and LRRK2 was PINK1 dependent. Dysregulation of some innate immunity genes was also found in skin fibroblast cells from PARK6 patients. Conclusions: Thus, an individual biomarker with expression correlating to progression was not identified. Instead, more advanced disease stages involved additional pathways. Hence, our results identify PINK1 deficiency as an early modulator of innate immunity in neurons, which precedes late stages of neuroinflammation during alpha-synuclein spreading.