ScRNA-seq going viral: using sequencing technologies to understand the severity of COVID-19 patients

Since the beginning of the SARS-CoV-2 pandemic, determining which cell types can be infected has been a major focus of research efforts. While the virus clearly infects airway cells, how the virus interacts with other cell types and tissues is not well understood. This information is important because if we know the range of infectivity of the virus, we can better understand the specific signs and symptoms of SARS-CoV-2 in patients, as well as design treatment targeting specific tissues. 

Early efforts to address this question have focused, understandably, on mining existing datasets, including single cell atlases published prior to the pandemic. Single cell sequencing (scRNA-seq) is a relatively new technique which can look at the genetic material at a single cell resolution. It’s very powerful for understanding diverse populations of cells because it can capture important details of individual cells rather than just the population average. The assumption has been that if cells express the surface proteins ACE2 and TMPRSS2, they should be permissive of infection with SARS-CoV-2. ACE2 is a receptor that is recognized by the receptor binding domain of the SARS Spike protein, and TMPRSS2 is a cell surface protease that supports ACE2 entry into cells. Analysis of scRNA-seq atlases has confirmed that lung epithelial cells, particularly in the airway, are susceptible to SARS-CoV-2 infection based on expression of ACE2 and TMPRSS2. A range of other cells are also likely targets, including intestinal epithelial cells.

However, what we’ve all been waiting for is new data scRNA-seq data acquired with cells from patients infected with SARS-CoV-2, in order to determine which cell types contain viral RNA and are therefore infected. To this end, Ido Amit’s group at Weizmann Institute of Science in Rehovot, Israel re-analyzed data generated by the 10x Genomics platform for single cell RNA sequencing (Liao et al). Samples were collected by bronchoalveolar lavage from 9 COVID19 patients hospitalized in Shenzhen, China in early 2020. Using a computational technique to map sequencing results to the appropriate species, including both human and viral genomes, Bost et al. were able to identify cell types that contained SARS-CoV-2 RNA.

Comparing severe with milder cases of COVID19 by this analysis, Bost et al. found a number of differences in multiple compartments, also reported by Liao et al. in their paper. For example, severe patients had a higher proportion of SPP1+ monocyte-derived macrophages bearing an inflammatory signature consistent with the cytokine storm reported in severely ill patients. Importantly, the detection of viral RNA by Bost et al. also permitted identification of specific cell types infected by SARS-CoV-2. They found that none of the 3 mild patients’ samples had detectable virus by scRNAseq, whereas viral RNA was found in all 6 severe samples, reflecting poor virologic control. Moreover, while unsurprisingly epithelial cells contained viral RNA, SPP1+ monocyte-derived macrophages also contained viral reads, demonstrating the role of these inflammatory myeloid cells in the encounter with (and likely the inflammatory reaction to) SARS-CoV-2. Interestingly, infection of alveolar macrophages was not detected. The therapeutic significance of this finding is not yet obvious. However, it is tempting to speculate that since alveolar macrophages did not seem to be infected, they could be harnessed early in the infection to attack the virus. If these cells could clear the virus or hamper its spread, they could prevent or lessen the subsequent immune cell infiltration, which could be causing additional damage.

Applied so far only to lavage cells, this meta-RNAseq approach will be useful for future datasets, as they become available. For example, SARS-CoV-2 patients can present with neurological or kidney problems, and the ways that the virus can infect these organs are not well understood. The extent of SARS-CoV-2 infection of multiple compartments in multiple tissues can now be better defined, with relevance to understanding the reach and consequence of the infection throughout the body.

By Mallar Bhattacharya

https://bhattacharyalab.ucsf.edu

References cited:

Read the original article at: https://www.cell.com/cell/pdf/S0092-8674(20)30568-7.pdf

This paper builds off of the results from: https://doi.org/10.1038/s41591-020-0901-9

The ACE2 receptor on lung epithelial cells, from Anthony Venida’s earlier post.