Mechanisms of aggregation of FUS and TDP-43 in a cellular model of amyotrophic lateral sclerosis

Amyotrophic Lateral Sclerosis (ALS) is one of the most common adult-onset motor neuron disease (MND) and is characterized by the selective loss of upper and lower motor neurons. The accumulation of aggregates containing wild type and mutant proteins (SOD1, TDP-43 and FUS/TLS) is a pathological hallmark of post-mortem motor neurons from ALS patients. It is still not clear whether these large protein aggregates play a central role in ALS initiation and progression, or if they represent a defensive response aimed at protecting cells. Aggregates may result toxic for motor neurons because they entrap proteins critical for their viability, or because they cause a mechanical hindrance and impairment of axonal transport, or because they affect specific organelles such as mitochondria. The formation of such intracellular aggregates may depend on the accumulation of misfolded proteins generated either as a direct consequence of mutation, or as a consequence of oxidative stress (a common feature of many neurodegenerative diseases, including ALS). Until now mechanisms of aggregation are understood only for mutant SOD1 that is the first ALS-related protein that has been discovered. There is now general consensus that the toxic function gained by mutant SOD1 is to be related by its propensity to aggregate and to mislocalize, thus jeopardizing neuronal viability. Moreover, it is established that i. the formation of SOD1 aggregates is mediated by cysteine residues and is the consequence of covalent disulfide cross-linking and non-covalent interactions; ii. the alteration of the GSH/GSSG ratio and downstream modulation of protein cysteines redox state is a crucial event in aggregation of mutant SOD1 and iii. the removal of SOD1 aggregates in motoneuronal cell cultures is possible through the modulation of the redox cellular state. Similarly to SOD1, also FUS and TDP-43 are found aggregated in cytoplasmic inclusions in tissues from ALS patients and animal models. They are both nuclear RNA-binding proteins structurally similar to heterogeneous ribonucleoproteins (hnRNPs), and are involved in RNA processing. The general aim of this work was to understand mechanisms of aggregation of FUS and TDP-43 and to analyze whether removal of their aggregates through the redox modulation of protein thiols may represent a useful therapeutic option. To this aim we engineered mouse motoneuron-like NSC34 cells for the inducible expression of 5Flag-FUS (wild-type and ALS-mutant) or 3Myc TDP-43 (wild-type or ALS-mutant or the truncated isoform of 35 kDa) fused proteins. Through standard immunochemical techniques we confirmed proteins mislocalization and aggregation that well recapitulate the ALS phenotype of our cellular models. We then analyzed the determinants leading aggregation of these proteins. Results obtained in this work indicate for the first time that, despite their many common structural features (e.g. the presence of RNA Recognition Motifs and Nuclear localizations signals), FUS and TDP-43 differ in their aggregation mechanism. We demonstrated that FUS aggregation occurs without a cysteine-dependent mechanism, while TDP-43 aggregation is in part mediated by cysteine residues. Moreover, we highlight the relative importance of the Nterminal and C-terminal moieties of TDP-43 in the aggregation process and the weight of each of the six cysteine residues in determining unfolding and aggregation of these two different domains. In particular, we have shown that the two N-term cysteine residues contribute to the seeding for oligomerization, which is then accomplished by mechanisms depending on the four cysteines in the RNA-recognition motifs of the full length isoform. Moreover, endogenous pathological C-terminus fragments of TDP-43 seems to be completely recruited in oligomers. Ab initio modelling of TDP-43 structure, molecular dynamics and molecular docking analysis support our in vitro results showing a differential accessibility of cysteine residues that contributes to aggregation propensity. Besides to ability to form inclusions in ALS patients, FUS and TDP-43 share induction of mitochondrial dysfunction as a common pathological feature. In our cellular models both proteins colocalize with mitochondrial markers, but they have a different impact on the mitochondrial physiology. Further investigations are needed to understand if there is a correlation between the different mechanisms of aggregation and different mitochondrial alterations in cells.