In-depth proteomic analysis of Purkinje neurons reveals unique metabolic signatures caused by lack of mitochondrial fusion
Motori E1, Atanassov I1, Wendler S2, Folz-Donahue K1, Hinze Y1, Toni N3, Puyal JP3
1MPI for Biology of Ageing, Cologne, Germany; 2CECAD Cologne, Germany; 3UNIL Lausanne, Switzerland; 4Karolinska Institute Stockholm, Sweden
Mitochondrial dynamics are recognized for playing key roles in proper functioning of nerve cells. Accordingly, mutations in the Mfn2 gene, whose product mediates mitochondrial fusion, are associated with CMT2A disease in humans. Studies in mouse models have demonstrated that in the long term Mfn2 may be important for maintaining axonal as well as dendritic integrity, and in vitro work explained these phenotypes by suggesting that fusion preserves OXPHOS function by regulating mtDNA levels through organelle content mixing. Recently, we reported that Mfn2 is implicated in maintaining the terpenoid biosynthetic pathway in heart, challenging our knowledge of the role of mitochondrial fusion in differentiated tissues. Yet, it remains unknown if similar mechanisms apply for neurons in vivo and if these are the primary cause of neuronal degeneration. Here, we developed a FACS-based method to purify genetically labeled Purkinje neurons from Mfn2loxP::L7Cre mice (Mfn2cKO), which are characterized by loss of the Purkinje cell layer by 10-12 weeks of age. Time-course analysis showed a progressive depletion of mtDNA in vivo between 3 and 8 weeks, which ultimately resulted in OXPHOS dysfunction. To dissect the early cellular and metabolic alterations upon Mfn2 loss, we profiled the proteomic landscape of purified cells at 5 and 8 weeks, when neurons retained their electrophysiological properties, despite a fragmented mitochondrial network and aberrant cristae morphology. Analysis confirmed a progressive OXPHOS deficiency, emphasized by the decreased expression of respiratory chain subunits. Interestingly, the onset of OXPHOS failure was mirrored by a remodeling of the folate-dependent biosynthetic pathways, supporting the notion that this is a common signature in mitochondrial dysfunction. Surprisingly, Purkinje neurons also showed a profound metabolic rewiring at this early stage, and revealed the existence of new pathways linked to mitochondrial fusion in neurons. Taken together, our findings provide key insights into the tissue-specific metabolic alterations that loss of mitochondrial fusion induces in neurons in vivo.