In a recent study published in the journal Dr Proceedings of the National Academy of SciencesResearchers analyzed the inflammatory potential of fragments of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2).
Intensive research during the coronavirus disease 2019 (COVID-19) pandemic has helped to understand SARS-CoV-2 transmission. Still, what enables the virus to trigger a dangerous inflammatory response remains unclear. Studies have suggested that amphiphilic, cationic peptides from the innate immune system assemble amyloid-like with anionic nucleic acids and form proinflammatory complexes.
Research and results
The present study investigates whether fragmented SARS-CoV-2 peptides assemble into supramolecular complexes with anionic double-stranded RNA (dsRNA). The viral proteome was considered a reservoir of peptide fragments released after proteolytic destruction of virions. The researchers used a support vector machine (SVM) classifier to recognize antimicrobial peptide (AMP)-like sequences (xenoAMPs) in the SARS-CoV-2 proteome.
Viral protein sequences were scanned through a moving window of 24–34 amino acids to identify potential xenoAMPs and test whether they behaved like AMPs when cleaved at different positions. Sequences were selected based on the output provided by the classifier as a sigma (σ) score, where a strongly positive score indicates that the sequence is more likely to be an AMP.
Existence of exogenous mimics of pro-inflammatory host antimicrobial peptides (xenoAMPs) in SARS-CoV-2 protein. (A) SARS-CoV-2 protein scanned with a machine-learning AMP classifier. Each queried sequence is assigned a σ score that measures its AMP-ness. Three representative high-scoring sequences are studied: xenoAMP(ORF1ab), xenoAMP(S), and xenoAMP(M). Gray bars mark the location where corresponding sequences are selected (B) SARS-CoV-2 sequences are aligned and compared to their homologues in a common cold human coronavirus HCoV-OC43: control (ORF1ab), control(S), and control(M). Asterisks, colons, and periods indicate positions that are fully conserved residues, that have strongly similar properties, and those that have weakly similar properties, respectively. Colors are assigned to each residue using the ClustalX scheme. (C) σ score heatmaps comparing the distribution of high-scoring sequences in three proteins from SARS-CoV-2 and HCoV-OC43. The first amino acid of each sequence is colored according to its average σ score; Regions with negative mean σ scores (non-AMP) are colored white. The “hot spot” cluster of high-scoring sequences for SARS-CoV-2 (bright yellow regions bracketed in red boxes) has systematically higher scores and wider regions of sequence space than HCoV-OC43. This trend suggests that the hot spots of SARS-CoV-2 may generate higher scoring sequences for a greater diversity of enzymatic cleavage sites than those of HCoV-OC43.
Further, the team selected from this population (high-scoring) specific sequences with high cationic charge. Specifically, they focused on prototypical candidates from the membrane (M) protein, spike (S) protein, and open reading frame 1ab (ORF1ab) polyprotein. in silico Analyzes have shown that these xenoAMPs can be generated during proteasomal degradation, with matrix metalloproteinase 9 (MMP9) and neutrophil elastase (NE) able to generate them.
Next, the team compared the SARS-CoV-2 xenoAMP to the homologous sequences of SARS-CoV-1 and non-pandemic human CoVs. This showed that the sequences were partially conserved. A comparison of the σ score hit maps of ORF1ab, S, and M proteins between SARS-CoV-2 and HCoV-OC43 revealed that high-scoring sequences were clustered in hotspots, with SARS-CoV-2 hotspots having higher scores and spread over wider regions. than HCoV-OC43.
Further, mass spectrometry was performed on airway aspirate samples from severe COVID-19 patients. The team identified fragments of the host AMP, cathelicidin LL-37, in 20 samples (out of 29). In contrast, 28 samples contained viral peptide fragments, some of which had high enough σ scores to qualify as xenoAMPs.
Three xenoAMPs, xenoAMP(S), xenoAMP(M), and xenoAMP(ORF1ab), were experimentally observed to associate with chaperones and dsRNA in complexes similar to LL-37. Polyinosine:polycytidylic acid (poly(I:C)) was used as a synthetic analog to mimic viral dsRNA generated during replication. The structures of xenoAMPs-poly(I:C) complexes were intuitive to host AMPs-dsRNA complexes.
Next, the team investigated the robustness of these self-assembled proinflammatory complexes under non-optimal conditions. They found that the nanocrystalline structures were preserved when the participating xenoAMPs were shortened. In addition, SARS-CoV-2 xenoAMPs were found to co-crystallize with LL-37, suggesting that host AMPs and xenoAMPs may synergistically activate inflammatory responses.
The immunogenicity of xenoAMP from SARS-CoV-2 was compared with the homologous peptide from HCoV-OC43 using human monocytes. XenoAMP-poly(I:C)-treated monocytes expressed 1.7-fold more interleukin (IL)-8 than poly(I:C)-treated controls. In contrast, complexes formed with homologous peptides from HCoV-OC43 induced much lower IL-8 levels.
In addition, xenoAMP-poly(I:C) stimulation of primary human dermal microvascular endothelial cells (HDMVECs) triggered robust production of IL-6, which was not observed with complexes formed from the HCoV-OC43 peptide. Notably, xenoAMP-poly(I:C)-treated HDMVECs showed significant upregulation of several proinflammatory chemokine and cytokine genes.
Finally, the researchers measured the mice’s immune system. C57BL/6 mice not exposed to infection were treated with xenoAMP(ORF1ab)-poly(I:C) complex or poly(I:C)-alone (control). XenoAMP(ORF1ab)-poly(I:C) treatment increased plasma levels of IL-6 and CXC motif chemokine ligand 1 (CXCL1) by 1.6- and 2.2-fold, respectively, compared with poly(I:C)-alone. Furthermore, IL-6 and CXCL1 levels increased 1.2-fold in the lungs compared to control treatment.
In summary, the study illustrates an unexpected mechanism of inflammation propagated by uninfected cells of COVID-19, in which viral fragments mimic AMPs such as LL-37. This may be important for understanding why the host immune system of Covid-19 resembles that of those with autoimmune conditions such as rheumatoid arthritis and lupus.
The researchers found that host proteases can generate xenoAMP, suggesting that protease inhibitors that suppress xenoAMP generation may have clinical effects on viral-induced inflammation. Proteolytic degradation of SARS-CoV-2 may differ between host individuals, possibly explaining the variation in the outcome of infection, i.e., asymptomatic and fatal.
- Zhang Y, Bharti V, Dokoshi T, et al. Viral afterlife: SARS-CoV-2 as a reservoir of immunomimetic peptides that reassemble into proinflammatory supramolecular complexes. Proc Natl Acad Sci USA2024, DOI: 10.1073/pnas.2300644120, https://www.pnas.org/doi/10.1073/pnas.2300644120