Role of acetylation in TDP-43 pathophysiology

Abstract

The TAR DNA binding protein 43kDa (TDP-43) is a predominantly nuclear protein that is involved mainly in mRNA splicing, processing and transport. TDP-43 is also the main component of the insoluble inclusions found in neurons in both in amyotrophic lateral sclerosis (ALS) and a subset of fronto-temporal lobar degeneration cases (FTLD-TDP). Within these inclusions, TDP-43 is hyperphosphorylated, ubiquitinated, acetylated and fragmented. The regulation of these modifications and their impact in the physiology and aggregation process of TDP-43 is not fully understood yet. The objective of the following study was to identify new acetylation sites in TDP-43 and characterize their impact in TDP-43’s pathophysiology. In order to do this, wild type TDP-43 was expressed in HEK293E cells, purified and analysed via mass spectrometry. Four novel acetylation targets were identified, two of which were around the nuclear localisation signal while the other two were at the RNA recognition domain. The impact of these acetylation sites was assessed by generating lysine-to-arginine and lysine-to-glutamine substitutions, which showed the impact of acetylation-mimic mutations at lysines 84 and 136 in nuclear-cytoplasmic trafficking and RNA-binding of TDP-43, respectively. The effect of K136Q substitution in RNA binding was examined via RNA-protein pulldown, RNA-binding filter assays and by looking at its effect on CFTR splicing. K136Q causes a reduction in RNA-binding and a subsequent loss in RNA-splicing capabilities. In addition, K136Q TDP-43 is found in nuclear phase-separated bodies, with reduced recovery rate after photobleaching. K136Q TDP-43 eventually loses solubility and coalescences into solid, phosphorylated aggregates, similar to those found in ALS and FTLD patients. To validate all these results, an amber suppressed TDP-43 model which introduces acetylated lysine at either K84 or K136, was devised. Acetylated TDP-43 behaved in a similar manner to the K-to-Q mutants characterised, validating the previous results. To further characterise TDP-43 acetylation, antibodies against acetylated K84 TDP-43 and K136 TDP-43 were developed and tested against full length, amber suppressed TDP-43. With this new tool, Sirt1 was identified as the responsible enzyme deacetylating K136, disengaging the phase-separated TDP-43 and preventing its further aggregation. The identification of acetylation as a modulator of TDP-43 cytoplasmic mislocalisation and RNA-binding offers a powerful tool to modulate TDP-43 activity. Furthermore, this study describes how acetylation at K136 can trigger TDP-43 phase separation and eventual pathological aggregation, identifying the responsible enzymes that can mitigate this process. Future research into this specific mechanism could be pivotal for ALS and FTLD early diagnosis and treatment.