Early-forming aberrant aggregates in protein deposition diseases: structural characteristics, interaction with molecular chaperones, ability to trigger inflammation

Protein aggregation into amyloid fibrils is the hallmark of many human pathologies, including Alzheimer's disease and Parkinson's disease. The aberrant assembly of peptides and proteins into fibrillar aggregates proceeds through oligomeric intermediates thought to be the primary pathogenic species in many of these protein deposition diseases. Since the first year of my PhD, I have been studying protein aggregation, focusing on misfolded protein oligomers formed by the N-terminal domain of the bacterial HypF protein from E. coli (HypF-N). These oligomers are a useful model system in the context of disease-associated protein aggregation because they have the same properties as the disease-related ones and are highly stable, allowing a detailed structural and functional investigation. Moreover, under two different conditions, HypF-N aggregates into two types of oligomers characterized by an opposite biological activity, since only one is toxic whereas the other is benign to cultured cells, facilitating the utilization of appropriate control nontoxic oligomers in our experiments. In Chapter 2, the first one describing results in my thesis, the toxicity of HypF-N oligomers, as well as the aggregates formed by the amyloid-β peptide and the islet amyloid polypeptide, was tested in the presence of molecular chaperones, namely αB-crystallin, Hsp70, clusterin, α2-macroglobulin and haptoglobin. Molecular chaperones play a pivotal role in the regulation of the proteome homeostasis, as they facilitate protein folding, inhibit protein aggregation, disaggregate pre-formed assemblies and promote clearance of misfolded aggregates. In this chapter we show that molecular chaperones also affect the structure and toxicity of protein misfolded oligomers. Indeed, measures of the cell viability showed that all the five chaperones are effective in suppressing the toxicity of oligomers formed by all three proteins. Infrared spectroscopy and site-directed labeling experiments using pyrene ruled out a structural reorganization within the discrete HypF-N oligomers, even at the amino acid residue level. By contrast, analysis performed using confocal microscopy, SDS-PAGE and intrinsic fluorescence measurements revealed binding between the oligomers and the chaperones; atomic force microscopy (AFM) imaging indicated that large assemblies of oligomers are formed in the presence of the chaperones. This suggests that the chaperones bind to the oligomers and promote their assembly into larger species, with consequent decrease in their diffusional mobility and burial of hydrophobic surface. In Chapter 3, with the aim of identifying the structural determinants responsible for the toxicity of misfolded oligomers, we created a set of HypF-N oligomeric variants by replacing one or more charged aminoacids with apolar aminoacids into the sequence of the wt protein, and allowing the mutated proteins to aggregate under different conditions. The resulting oligomeric species were characterized by different levels of cytotoxicity, as assessed by measurements of MTT reduction, tests with the apoptotic marker Hoechst and measurements of Ca2+ influx. The structural properties of the oligomeric variants was performed by evaluating the exposure of their hydrophobic surfaces to the solvent with ANS binding and by measuring their size by means of turbidimetry and light scattering measurements. A significant correlation was found between ANS binding and size of the oligomers, indicating that an increase of surface hydrophobicity causes an increase of the size of the oligomers. Moreover, both superficial hydrophobicity and size were found to influence the oligomer biological activity, cooperating in determining the levels of toxicity of the aggregates. In Chapter 4, we used HypF-N toxic misfolded oligomers to investigate their interactions with transthyretin (TTR). This protein is a homotetrameric protein which can disassemble into its monomers, misfold and aggregate into fibrils whose growth is considered the cause of TTR amyloidoses. Nevertheless, an anti-amyloidogenic effect that prevents Aβ aggregation in vitro has recently been proposed for TTR. We have therefore explored the ability of three different types of TTR, human TTR (hTTR), mouse TTR (mTTR) and an engineered monomer of human TTR (M-TTR), to suppress the toxicity of HypF-N oligomers. Cell viability tests showed that hTTR, and to a greater extent M-TTR, can avoid the cell damage induced by protein oligomers, whereas mTTR does not show any protective effect. To shed light on the different behavior of the TTRs and on the molecular mechanism by which they can exert their potential protective ability, we have investigated the molecular structure of HypF-N oligomers after the incubation with TTRs. Thioflavin T assay and pyrene site-directed labeling showed that all the three types of TTR cannot structurally re-arrange toxic HypF-N oligomers into a nontoxic form. Intrinsic fluorescence measurements and SDS-PAGE indicated that TTRs are able to bind to the oligomers. Following this binding, hTTR, and to a greater extent M-TTR, induced the formation of larger species, as shown by AFM and turbidimetry measurements. By contrast, the interaction with mTTR does not induce such formation of clusters. These data indicate that TTR suppresses the toxicity of HypF-N oligomers similarly to well established chaperones with an efficacy that correlates with its ability to disassemble into monomers. Finally, in Chapter 5, we tested the ability of HypF-N oligomers to induce the inflammatory response. In fact, increasing evidence suggests that neurodegeneration associated to aggregation of proteins is the result of many causes. The uncontrolled immune response in the brain has recently been established to play a central role in the onset and progression of diseases, such as Alzheimer’s disease and Parkinson’s disease. For this reason we explored the inflammatory response to the injury caused by HypF-N toxic and not toxic misfolded oligomers, with particular attention to the role of Hsps, Hsp70 and αB-crystallin, as immune signals and potential suppressors of HypF-N oligomer-mediated inflammation. The results, obtained by the evaluation of microglia activation in terms of cytokine-release, showed that both the toxic and the nontoxic oligomers triggered a pro-inflammatory response, as assessed through ELISA measurements of a set of cytokines. Interestingly, at concentrations in which the two types of oligomers share the ability of leaving unaltered the cellular viability evaluated by MTT tests, the nontoxic species were found to be stronger inducers of inflammation with respect to the toxic oligomers. Such immune property of the nontoxic aggregates could be linked to their lower level of internalization in microglia cells and to the consequent maintenance of their stimulus from outside the cells. In addition, the nontoxic oligomers and the assemblies of toxic oligomers neutralized by chaperones were found to have the ability to induce inflammation without affecting cellular viability. In conclusion, the data presented in this thesis and collected entirely using misfolded protein oligomers by the model protein domain HypF-N have revealed (i) new structural determinants of protein oligomer toxicity, such as oligomer size and hydrophobic exposure and their interplay to determine toxicity, (ii) have revealed novel mechanisms by which molecular chaperones, including the emerging TTR, contribute to the maintenance of protein homeostasis and (iii) have shown how misfolded protein oligomers can be highly inflammatory, even in the absence of explicit toxicity.