Contributing to research themes:
Background:
Quantitative proteomics is able to provide a comprehensive, unbiased description of changes to cells caused by viral infection, but interpretation may be complicated by differential changes in infected and uninfected ‘bystander’ cells, or the use of non-physiological cellular models.
Methods:
In this paper, we use fluorescence-activated cell sorting (FACS) and quantitative proteomics to analyse cell-autonomous changes caused by authentic SARS-CoV-2 infection of respiratory epithelial cells, the main target of viral infection in vivo. First, we determine the relative abundance of proteins in primary human airway epithelial cells differentiated at the air-liquid interface (basal, secretory and ciliated cells). Next, we specifically characterise changes caused by SARS-CoV-2 infection of ciliated cells. Finally, we compare temporal proteomic changes in infected and uninfected ‘bystander’ Calu-3 lung epithelial cells and compare infection with B.29 and B.1.1.7 (Alpha) variants.
Results:
Amongst 5,709 quantified proteins in primary human airway ciliated cells, the abundance of 226 changed significantly in the presence of SARS-CoV-2 infection (q <0.05 and >1.5-fold). Notably, viral replication proceeded without inducing a type-I interferon response. Amongst 6,996 quantified proteins in Calu-3 cells, the abundance of 645 proteins changed significantly in the presence of SARS-CoV-2 infection (q < 0.05 and > 1.5-fold). In contrast to the primary cell model, a clear type I interferon (IFN) response was observed. Nonetheless, induction of IFN-inducible proteins was markedly attenuated in infected cells, compared with uninfected ‘bystander’ cells. Infection with B.29 and B.1.1.7 (Alpha) variants gave similar results.
Conclusions:
Taken together, our data provide a detailed proteomic map of changes in SARS-CoV-2-infected respiratory epithelial cells in two widely used, physiologically relevant models of infection. As well as identifying dysregulated cellular proteins and processes, the effectiveness of strategies employed by SARS-CoV-2 to avoid the type I IFN response is illustrated in both models.