Deep Sequencing of B Cell Receptor Repertoires From COVID-19 Patients Reveals Strong Convergent Immune Signatures

Abstract

Deep sequencing of B cell receptor (BCR) heavy chains from a cohort of 31 COVID-19 patients from the UK reveals a stereotypical naive immune response to SARS-CoV-2 which is consistent across patients. Clonal expansion of the B cell population is also observed and may be the result of memory bystander effects. There was a strong convergent sequence signature across patients, and we identified 1,254 clonotypes convergent between at least four of the COVID-19 patients, but not present in healthy controls or individuals following seasonal influenza vaccination. A subset of the convergent clonotypes were homologous to known SARS and SARS-CoV-2 spike protein neutralizing antibodies. Convergence was also demonstrated across wide geographies by comparison of data sets between patients from UK, USA, and China, further validating the disease association and consistency of the stereotypical immune response even at the sequence level. These convergent clonotypes provide a resource to identify potential therapeutic and prophylactic antibodies and demonstrate the potential of BCR profiling as a tool to help understand patient responses.

Keywords: B-cell repertoire; BCR; COVID-19; SARS-CoV-2; antibody; convergence.

Figure 1

B cell responses to SARS-COV-2 infection. (A) IGHV gene segment usage distribution per isotype. (B) Isotype subclass distribution between IGHA and IGHG subclasses, and (C) mean BCR CDRH3 lengths from COVID-19 patients compared to healthy controls. For (A–C), bars show mean values +/− standard error of the mean. Comparisons performed using t-tests, with adjusted p values using Bonferroni correction for multiple comparisons; * p < 0.05, ** p < 0.005, *** p < 0.0005.

Figure 2

Response characteristics of SARS-CoV-2 infection. (A) Distribution of sequences with different numbers of mutations from germline. (B) Relationship between the proportion of the repertoire comprised by unmutated sequences, and the disease state (C) Individual sequences were clustered together into related groups to identify clonal expansions (clonotypes). Diversity of all clonotypes in the repertoire calculated using the Shannon diversity index. To normalize for different sequence numbers for each sample, a random subsample of 1,000 sequences was taken. (D) Network graphs of a representative repertoire from a single COVID-19 patient, and a single HC sample giving a graphical representation of the diversity. Each point represents one of the subsampled sequences, and sequences within the same clonotype are linked together. (E) Relationship between repertoire diversity and disease state.

Figure 3

Convergent BCR sequence signature within individuals infected with SARS-CoV-2. (A) Data from all patients and healthy controls were clustered together to identify convergent clonotypes. Shown is the number of clonotypes shared by different numbers of participants, grouped by whether the clonotypes are also present in the healthy control dataset. Of the convergent clonotypes, (B) the mean mutation count, and (C) the CDRH3 AA sequence length was compared between those that were convergent only within the SARS-CoV-2 patients, and those that were also convergent with the healthy control dataset. (D) Heatmap of the 1,254 convergent COVID-19-associated clonotypes (observed between 4 or more COVID-19 participants) with the 1,180 convergent clonotypes from six influenza vaccination (FLU) samples, and 351 convergent clonotypes from six metastatic breast cancer (BC) patient biopsy samples, demonstrating that the convergent signatures are unique to each cohort.

Figure 4

Matches of the 1,254 convergent clonotypes identified in the present study to the BCR sequence data from the six COVID-19 patients from Stanford, USA (15). (A) Plotted along the x-axis are the 463 convergent clonotypes represented in at least one sample from the Stanford study. Each row represents a separate BCR repertoire from the Stanford study; pink shading indicates that the convergent clonotype has a match. (B) Number of matches that the 1,254 convergent clonotypes have to the Stanford samples in comparison to the random selection of 1,254 clonotypes from the healthy controls.