Proteomics identifies a type I IFN, prothrombotic hyperinflammatory circulating COVID-19 neutrophil signature distinct from non-COVID-19 ARDS
First Author: Leila Reyes
Journal/preprint name: medRxiv
Paper DOI: https://doi.org/10.1101/2020.09.15.20195305
Tags: Neutrophils, proteomics, LDN, ARDS, Type I IFN, platelets
This study reports on the proteomics and metabolomics analysis of peripheral blood neutrophils from patients with COVID-19 ARDS (CA; n=3), non-COVID-19 ARDS (NA; n=3-7), moderate COVID-19 (MC; n=3), and healthy controls (HC; n=4-7). For some analyses normal density neutrophils (NDN) and low density neutrophils (LDN; isolated from the PBMC layer) were analysed separately. LDN consisted of both mature (CD16+CD10+) and immature (CD16-CD10-) neutrophils. Due to the small samples sizes most conclusions will need further validation, but the data suggest that neutrophils in COVID-19 ARDS may have increased IFN I signalling, degranulation, and platelet binding compared to healthy controls.
Comparing the neutrophil proteome of COVID-19 ARDS patients to healthy controls highlights an increase in the type I interferon signalling pathway and in markers of platelet degranulation.
Mature LDN from moderate or ARDS COVID-19 patients show increased CD41 staining, suggesting increased binding to platelets, compared to NDN from healthy controls. Immature LDN from COVID-19 patients do not show increased CD41 staining.
NDN from ARDS COVID-19 patients have slightly decreased content of granule proteins compared to healthy controls.
NDN from ARDS COVID-19 patients have increased glucose content compared to healthy controls, the cause is unclear.
In vitro stimulation of healthy neutrophils with a TLR7/8 agonist under hypoxic conditions leads to an activated phenotype regarding surface receptors.
Impact for COVID-19 research:
No direct impact because the conclusions need further validation.
Study Type: case study, in vitro
Key Techniques: flow cytometry for activation markers, proteomics, metabolomics, extracellular flux analysis
Small group sizes; n=3 for COVID-19 ARDS and n=3 for moderate COVID-19.
p-hacking obvious: the grouping & subsetting of the samples is not consistent throughout the manuscript, and comparisons are often illogical (e.g. why not compare NA directly to CA in Fig. 3?).
Most statements in the text are based on comparisons that don’t show statistically significant differences.
Patient characteristics are only provided for ARDS patients not for moderate patients or HC. Unclear whether non-COVID-19 ARDS patients also had viral infection.