Cell invasive amyloid assemblies from SARS-CoV-2 peptides can form multiple polymorphs with varying neurotoxicity, 2024, Sanislav, Annesley et al.

[SNT Gatchaman]

Cell invasive amyloid assemblies from SARS-CoV-2 peptides can form multiple polymorphs with varying neurotoxicity
Oana Sanislav; Rina Tetaj; Metali; Julian Ratcliffe; William Phillips; Annaleise R. Klein; Ashish Sethi; Jiangtao Zhou; Raffaele Mezzenga; Sina S. Saxer; Mirren Charnley; Sarah J. Annesley; Nicholas P. Reynolds

The neurological symptoms of COVID-19, often referred to as neuro-COVID include neurological pain, memory loss, cognitive and sensory disruption. These neurological symptoms can persist for months and are known as Post-Acute Sequalae of COVID-19 (PASC). The molecular origins of neuro-COVID, and how it contributes to PASC are unknown, however a growing body of research highlights that the self-assembly of protein fragments from SARS-CoV-2 into amyloid nanofibrils may play a causative role.

Previously, we identified two fragments from the SARS-CoV-2 proteins, Open Reading Frame (ORF) 6 and ORF10, that self-assemble into neurotoxic amyloid assemblies. Here we further our understanding of the self-assembly mechanisms and nano-architectures formed by these fragments and their biological responses. By solubilising the peptides in a fluorinated solvent, we eliminate insoluble aggregates in the starting materials (seeds) that change the polymorphic landscape of the assemblies. The resultant assemblies are dominated by structures with higher free energies (e.g. ribbons and amorphous aggregates) that are less toxic to cultured neurons but do affect their mitochondrial respiration.

We also show the first direct evidence of cellular uptake of viral amyloids. This work highlights the importance of understanding the polymorphic behaviour of amyloids and the correlation to neurotoxicity, particularly in the context of neuro-COVID and PASC.

Link | PDF (Nanoscale)

 

[SNT Gatchaman]

Some summary quotes from intro and results/discussion —

Previous works have shown that Amyloid-β (Aβ) aggregates are pathological to neuronal cells triggering cell death by a number of different mechanisms, including DNA damage, inhibition of ion-motive ATPases, inflammation and loss of Ca2+ homoeostasis. Further Aβ has been shown to cause impaired mitochondrial respiration in neuronal cells, and AD patients have significantly reduced mitochondrial respiration. Conversely, work from Annesley and co-workers has shown that α-synuclein fibrils (the molecular hallmark of Parkinson’s Disease) result in hyperactive mitochondrial respiration in neuronal cell lines.

In our previous work […] we identified peptide fragments from the viral proteins ORF6 (ILLIIM) and ORF10 (RNYIAQVD) that assembled into a polymorphic array of amyloid architectures. ILLIIM almost exclusively formed crystalline structures with a well-defined cross-β structure and RNYIAQVD formed a mixture of amyloid crystals and twisted ribbons. It is well established different amyloid polymorphs possess different free energies with crystalline polymorphs being more stable (lower free energy) than fibrils and ribbons. Therefore, in Charnley et al. we hypothesised that the added stability of the ILLIIM assemblies is connected to their increased cytotoxicity (i.e. slower cellular clearance of more stable amyloid crystals leads to increased toxicity).

We believe this is the first direct microscopic evidence of amyloid fibrils from fragments of SARS-CoV-2 viral proteins crossing the cell membrane. This is of interest as amyloid fibrils are known to have an almost ubiquitous ability to interact with and disrupt cell membranes and consequently they are also highly effcient transfection agents. Our results suggest that viral amyloid fibrils could be aiding viral infection and/or cellular escape during the viral replication cycle, suggesting a previously unconsidered role of amyloids in SARS-CoV-2 infection.

As it is known that amyloid assemblies can have complex effects on mitochondrial function, thus we performed a series of experiments to quantify changes in mitochondrial respiration in the viral amyloids studied here. In Fig. 6e we show changes in the basal oxygen consumption rate (OCR) in the presence of the viral amyloids.

The effects of amyloid assemblies on mitochondrial respiration can vary and previous investigations have found evidence of both increased and decreased OCR. A speculative explanation as to the lack of significant change in OCR at low (but still cytotoxic, Fig. 6d) concentrations is that we are seeing the results of two populations of cells within the same sample. One population is undergoing hyperactive mitochondrial respiration in the presence of the amyloids […] and a second population consists of dead or dying cells […]. At low concentrations the changes to the overall OCR are almost completely cancelled out by the two populations but at higher peptide concentrations when the population of dead or dying cells becomes larger, we see a significant drop in basal OCR (albeit still partially masked by a small population of hyperactive cells).

Eliminating the amyloid seeds from the starting materials results in the formation of higher energy polymorphs which, whilst able to cross the cell membrane have less cytotoxic effects on cultured neurons, however we still do observe significant reductions in overall metabolic activity and mitochondrial respiration. This data adds to the growing body of research suggesting amyloid aggregation may drive a number of cellular pathologies in COVID-19 and PASC.

Concluding —

We show, for the first time, direct evidence of cellular uptake of viral amyloids into neuronal cells, further supporting the hypothesis that some of the neurological symptoms of COVID-19 and PASC may possess a neurotoxic amyloid aetiology. By adjusting the assembly protocol we were able to eliminate preformed amyloid seeds, enabling us to generate higher energy polymorphs. This allowed us to explore the effect of these polymorphs on cells and reveal that they are less cytotoxic. Whilst this data is confined to in vitro studies it may offer promise that the neurological effects of COVID-19 and PASC are transient, due to the apparent relative instability of viral amyloids compared to amyloids from neurodegenerative diseases like Alzheimer’s disease.

 

[SNT Gatchaman]

Follow-on from Neurotoxic amyloidogenic peptides in the proteome of SARS-COV2: potential implications for neurological symptoms in COVID-19 (2022, Nature Communications)