TDP-43 Inclusion Bodies Formed in Bacteria Are Structurally Amorphous, Non-Amyloid and Inherently Toxic to Neuroblastoma Cells

Abstract

Accumulation of ubiquitin-positive, tau- and α-synuclein-negative intracellular inclusions of TDP-43 in the central nervous system represents the major hallmark correlated to amyotrophic lateral sclerosis and frontotemporal lobar degeneration with ubiquitin-positive inclusions. Such inclusions have variably been described as amorphous aggregates or more structured deposits having an amyloid structure. Following the observations that bacterial inclusion bodies generally consist of amyloid aggregates, we have overexpressed full-length TDP-43 and C-terminal TDP-43 in E. coli, purified the resulting full-length and C-terminal TDP-43 containing inclusion bodies (FL and Ct TDP-43 IBs) and subjected them to biophysical analyses to assess their structure/morphology. We show that both FL and Ct TDP-43 aggregates contained in the bacterial IBs do not bind amyloid dyes such as thioflavin T and Congo red, possess a disordered secondary structure, as inferred using circular dichroism and infrared spectroscopies, and are susceptible to proteinase K digestion, thus possessing none of the hallmarks for amyloid. Moreover, atomic force microscopy revealed an irregular structure for both types of TDP-43 IBs and confirmed the absence of amyloid-like species after proteinase K treatment. Cell biology experiments showed that FL TDP-43 IBs were able to impair the viability of cultured neuroblastoma cells when added to their extracellular medium and, more markedly, when transfected into their cytosol, where they are at least in part ubiquitinated and phosphorylated. These data reveal an inherently high propensity of TDP-43 to form amorphous aggregates, which possess, however, an inherently high ability to cause cell dysfunction. This indicates that a gain of toxic function caused by TDP-43 deposits is effective in TDP-43 pathologies, in addition to possible loss of function mechanisms originating from the cellular mistrafficking of the protein.

Extracellular TDP‐43 aggregates target MAPK/MAK/MRK overlapping kinase (MOK) and trigger caspase‐3/IL‐18 signaling in microglia

“…Use of TDP43 (preformed TDP‐43 aggregate) preparations for in vitro studies has thus far been hampered because of the great difficulties in producing and purifying TDP‐43 protein. To study the impact of extracellular TDP‐43 aggregates on microglia, we employed previously characterized TDP‐43 aggregates produced from inclusion bodies (33, 36) that have been shown to recapitulate several features of ALS‐associated aggregates, including their size and amorphous structure, ubiquitination and phosphorylation susceptibility, and inherent neurotoxic capabilities (33). To determine the TDP‐43 content of aggregate preparations, SDS‐PAGE and protein quantification by band densitometry were used, which showed a TDP‐43 content of approximately 50% in the His‐tagged TDP43 vs. mock control aggregate preparations (Fig.…”

“…TDP‐43 and mock aggregates were obtained by purifying the inclusion body fraction after overexpression in Escherichia coli, as previously described (33). In brief, human full‐length TDP‐43 was overexpressed in E. coli BL21(DE3) cells by using a plasmid construct kindly provided by Yoshiaki Furukawa's laboratory (Keio University, Yokohama, Japan).…”

Structural, morphological, and functional diversity of amyloid oligomers

“…a-synuclein), a-helical intermediates may also be involved [80,81,107]. A number of stable, yet highly disordered oligomers have been observed as well [32,67,108]. Gradual increase in the oligomer size accompanied by conformational transition within the amyloid oligomers has been proposed to proceed by monomer dissociation from the less stable oligomers followed by association of this monomer with more stable aggregates [109][110][111].…”

An Amyloid-Like Pathological Conformation of TDP-43 Is Stabilized by Hypercooperative Hydrogen Bonds

“…Furthermore, when TDP-43 is expressed in bacteria and purified from inclusion bodies (Capitini et al, 2014) or expressed in yeast (Johnson et al, 2008) no evidence for amyloid formation was detected. Moreover, TDP-43 produced from human cells or E. coli did not bind Thioflavin T or Congo Red, although it did form highly cytotoxic oligomers (Fang et al, 2014).…”