Living repository of millions of T and B cell receptor sequences from patients with COVID-19
bioinformatics immunology/immunity inflammation
Authors:Schultheiß C. et al.
Link to paper: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3624431
Journal/ Pre-Print:Immunity (Sneak Peak; under review)
Tags: Bioinformatics, Immunology/Immunity, Inflammation
Generation of a living repository of TCR/BCR sequences (available at gateway.ireceptor.org). TCR analysis show clusters exclusively enriched in the cohort of “active disease” containing public clonotypes, and others exclusively enriched in mild recovered patients, including low pGEN clusters.
Anti-SARS-CoV2 antibodies are present in most patients. BCR sequencing revealed converging IGHV3-driven BCR clusters (with restricted JH usage) in individual producing anti SARS-CoV2 antibodies. Some BCR clones had sequence similarity with a single chain antibody selected (by others) for anti-spike reactivity.
Schulteiss et al. combined next generation sequencing (NGS), ELISA and flow cytometry to analyse and compare the adaptive immune response from blood of patients recovered from mild COVID19, hospitalised with severe disease, and healthy controls. As previously reported, they observed an increase of IL6, IL8, IL-10, IP-10, as well as BAFF in severe cases and APRIL, sCD40l in the recovered cohort. While T lymphopenia was observed, activation was present with an increased expression of several inhibitory molecules. Moreover, the T cell response was slightly different from classical anti-viral response, with an increase of Treg, TH17, TFH and a trend of unconventional TH1*. All patients generated antibodies against SARS-CoV2. Interestingly, NGS analysis of TCR/BCR sequences was used to extract clusters of clonotypes that could be linked to a better outcome of the response. The repertoire analyses are very interesting and although but further analysis is required.
Impact for SARS-CoV2/COVID19 research efforts
Understand the immune response to SARS-CoV2/COVID19. A better understanding of the overall adaptive immune response and BCR/TCR repertoire analysis from mild and severe patients might help for the design of vaccine strategy.
In silico study / bioinformatics study
Clinical Cohort study (19 recovered patients, 20 patients with severe disease)
Strengths and limitations of the paper
Novelty: This paper provides insights into the cellular and humoral features of a COVID-19 response, and develops a new online resource for NGS analysis of TCR/BCR sequencing. The repertoire analysis is novel and some of the repertoire patterns look very interesting.
Standing in the field:The phenotypic data corroborate data from others such as a predominant T lymphopenia in comparison to a less affected B cell population, the shift of CD4+/CD8+ ratio towards CD4+, the increase of cytokines such as IL-6, IL-8, IP-10, and the presence of neutralising antibodies in most of the patients. The repertoire analysis presents some associations that had not been described elsewhere.
Appropriate statistics: Statistics are described. For multiple groups comparison, the ANOVA tests are not combined with multiple tests comparisons that would have been great. No normality tests have been performed while parametric tests were used.
Viral model used:SARS-CoV2 (direct study on infected patients blood).
Translatability:Not directly translatable. Far from bedside.
- They propose the analysis of their cohorts to look at efficient (recovered mild) versus less efficient (acute severe disease) response, but even if it is interesting to look at the repertoire on these two cohorts, the time after infection is different and might affect the results.
- The description of panels used for cell population analysis by flow cytometry are missing.
- In figure 4, definition of the samples used for the T cell population analysis is missing. It is not clear which patient samples they are using.
- Although the repertoire analysis looks good it is difficult to follow/define the analyses as the figure legend and text do not provide adequate explanation or justified rationale for the pipelines. It looks like there is a lot more that could be done with this data. The discussion acknowledges that this is a starting point and much more could be done.
- For the BCR clonotypes that relate to a known scAb should be tested for spike reactivity to demonstrate the observation is founded in real specificity (this may require also isolating the light chains).