The SARS-CoV-2 virus, responsible for COVID-19, exhibits a wide range of symptoms, including significant neurological manifestations in approximately 30% of patients. These neurological symptoms, which can persist for months after infection (long COVID), include memory loss, sensory confusion, headaches, brain inflammation, and stroke. While SARS-CoV-2's neuroinvasiveness has been demonstrated, the precise molecular mechanisms underlying these neurological effects remain unclear. The observed symptoms share similarities with amyloid-related neurodegenerative diseases like Alzheimer's and Parkinson's, prompting the hypothesis that amyloid formation from SARS-CoV-2 proteins may play a role. The study explores this hypothesis by focusing on proteins with unclear roles in viral replication, particularly open reading frames (ORFs), which are often unstructured and therefore potential candidates for amyloid formation. Previous research showed that some proteins in related coronaviruses (like SARS-CoV-1) have amyloidogenic sequences. The authors propose that identifying amyloid-forming sequences within the SARS-CoV-2 proteome could shed light on the mechanisms of COVID-19-associated neurological symptoms, potentially leading to novel therapeutic strategies.

Existing research indicates that SARS-CoV-2 can affect various cell types in the brain, including glia, nerve cells, and neurons. Studies on brain organoids have shown SARS-CoV-2 infection and replication in human retinal cells, correlating with reported olfactory and visual symptoms in COVID-19 patients. Furthermore, proteins from SARS-CoV-2 and SARS-CoV-1 have been shown to possess sequences with a propensity to form amyloid assemblies. The similarity in proteomes between SARS-CoV-1 and SARS-CoV-2 suggests that amyloid formation from SARS-CoV-2 proteins could contribute to the neurological symptoms seen in COVID-19. The review also notes the potential roles of amyloid assemblies in pathogens, such as inflammatory stimulation, enhanced viral transmission, RNA binding and packaging during replication, and inhibition of host antiviral responses. However, the study notes the lack of empirical evidence directly linking SARS-CoV-2 amyloid formation to neurological symptoms, while highlighting the known role of inflammation in driving amyloid-related diseases.

The study employed a bioinformatic screen of SARS-CoV-2 ORF proteins to identify potential amyloidogenic peptide sequences. Two peptides, one from ORF6 and one from ORF10, were selected for synthesis based on predictions from the ZIPPER and TANGO algorithms. The amyloid-forming capacity of the synthesized peptides was investigated using atomic force microscopy (AFM), transmission electron microscopy (TEM), small-angle X-ray scattering (SAXS), wide-angle X-ray scattering (WAXS), circular dichroism (CD) spectroscopy, and thioflavin T (ThT) assays. AFM and TEM were used to visualize the morphology of the assembled peptides and quantify the distribution of assembly heights and contour lengths using FiberApp. SAXS/WAXS provided information on the structures in solution. CD spectroscopy and ThT assays were employed to confirm the amyloid nature of the assemblies and analyze the assembly kinetics. The cytotoxicity of the amyloid assemblies was assessed using a human neuroblastoma cell line (SH-SY5Y) through MTT assays, flow cytometry (using Annexin V and 7-AAD staining), and manual cell counting. Molecular dynamics simulations were used to create an atomistic model of the amyloid structure.

Both synthesized peptides (from ORF6 and ORF10) rapidly self-assembled into amyloid structures with diverse polymorphic morphologies. AFM and TEM imaging revealed the formation of polymorphic nanofibrous and crystalline assemblies, with ILIIM forming larger, multi-laminar crystalline structures and RNYIAQVD forming longer, needle-like structures. SAXS data indicated a 2D sheet-like structure for ILLIIM and a more complex structure for RNYIAQVD. WAXS confirmed the amyloid nature of both assemblies through the presence of Bragg peaks characteristic of β-sheet structures. CD spectroscopy supported the β-sheet-rich secondary structure of the assemblies, with ILLIIM displaying mostly left-handed β-sheets and RNYIAQVD primarily right-handed β-sheets. ThT assays showed rapid assembly kinetics without a lag phase, characteristic of short amyloidogenic peptides. Cytotoxicity assays revealed both peptide assemblies to be highly toxic to SH-SY5Y cells at low concentrations, inducing late-stage apoptosis without significant necrosis. The toxicity was more pronounced for ILLIIM, correlating with its reported higher cytotoxicity in previous research. Molecular dynamics simulations provided an atomistic model consistent with the observed WAXS data.

The findings demonstrate that two short peptides derived from SARS-CoV-2 ORF6 and ORF10 form amyloid assemblies with high toxicity towards neuronal cells. This supports the hypothesis that amyloidogenic peptides from SARS-CoV-2 might contribute to the neurological symptoms observed in COVID-19. The observed polymorphism and the correlation between morphology and cytotoxicity suggest that specific structural features of the amyloid assemblies may influence their neurotoxic potential. The study highlights the importance of considering the role of amyloid formation in viral pathogenesis and its implications for long COVID. The different morphologies of the amyloid assemblies from the two peptides might explain their differing toxicities, with the more crystalline structures of ILLIIM exhibiting greater toxicity. The rapid assembly kinetics of the peptides indicate a potential for rapid onset of toxicity in vivo. The observed toxicity mechanisms suggest the potential for programmed cell death (apoptosis) as a crucial factor in the neuronal damage associated with COVID-19.

This study provides compelling evidence that amyloidogenic peptides derived from SARS-CoV-2 ORF6 and ORF10 contribute to the neurotoxicity associated with COVID-19. The distinct polymorphic forms and their different toxicities highlight the need for further research into the structure-function relationship of these amyloid assemblies. Future studies should focus on investigating the in vivo formation of these amyloid structures, exploring the precise mechanisms of neurotoxicity, and evaluating the potential of targeting these amyloids for therapeutic intervention in COVID-19-related neurological complications.

The study primarily utilizes in vitro models. Further in vivo studies are needed to confirm the findings and to understand the precise role of these amyloidogenic peptides in the context of a whole organism. The study focuses on two specific peptides; additional investigation of other potential amyloidogenic sequences within the SARS-CoV-2 proteome is warranted. The exact mechanisms by which these amyloid assemblies induce apoptosis in neuronal cells require further investigation.