key: cord-0320248-7i8r1ei0 authors: Sureshchandra, Suhas; Lewis, Sloan A.; Doratt, Brianna; Jankeel, Allen; Ibraim, Izabela; Messaoudi, Ilhem title: Single cell profiling of T and B cell repertoires following SARS-CoV-2 mRNA vaccine date: 2021-07-15 journal: bioRxiv DOI: 10.1101/2021.07.14.452381 sha: a9b4e02ff7364f4e9c78b41198f3daa2154847bf doc_id: 320248 cord_uid: 7i8r1ei0 mRNA based vaccines for SARS-CoV-2 have shown exceptional clinical efficacy providing robust protection against severe disease. However, our understanding of transcriptional and repertoire changes following full vaccination remains incomplete. We used single-cell RNA sequencing and functional assays to compare humoral and cellular responses to two doses of mRNA vaccine with responses observed in convalescent individuals with asymptomatic disease. Our analyses revealed enrichment of spike-specific B cells, activated CD4 T cells, and robust antigen-specific polyfunctional CD4 T cell responses in all vaccinees. On the other hand, CD8 T cell responses were both weak and variable. Interestingly, clonally expanded CD8 T cells were observed in every vaccinee, as observed following natural infection. TCR gene usage, however, was variable, reflecting the diversity of repertoires and MHC polymorphism in the human population. Natural infection induced expansion of larger CD8 T cell clones occupied distinct clusters, likely due to the recognition of a broader set of viral epitopes presented by the virus not seen in the mRNA vaccine. Our study highlights a coordinated adaptive immune response where early CD4 T cell responses facilitate the development of the B cell response and substantial expansion of effector CD8 T cells, together capable of contributing to future recall responses. The COVID-19 pandemic has spurred the rapid development of vaccines targeting SARS-CoV-2 that have garnered emergency use authorizations from the FDA and are being widely distributed (1, 2) . The Pfizer (BNT162b2) and Moderna (mRNA-1273) mRNA-based vaccines were the first to be approved and have proven to be safe and efficacious (94% effective) in adults and children over 12 years of age (3, 4) . However, the mechanisms by which these vaccines elicit long lasting cellular immune responses to SARS-CoV-2 and their mechanisms of action remain poorly understood. In this study, we aimed to address two while comparable humoral responses are generated with just one dose in convalescent subjects (5) . However, the ratio of binding to neutralizing antibodies after vaccination was greater than that after infection (6) . Additionally, most vaccinees had Th1-skewed T cell responses where early Tfh and Th1 CD4 responses correlate with effective neutralizing antibody responses after the first dose and CD8 effector responses after the second dose (7) . Furthermore, expanded T cell clones detected following vaccination were predominantly memory cells whereas those detected during infection were to effector cells with acute infection (8) . Collectively, these observations suggest distinct T and B cell responses following vaccination in comparison to natural infection. Additional studies that integrate functional, transcriptional, and repertoire analysis of the memory immune cell response to COVID-19 mRNA vaccination are needed (7) . In this study, we assayed humoral and cellular responses to two doses of mRNA vaccine (14 days post dose 2) in four individuals and compared parallel changes in their immune 4 repertoire with changes observed in three convalescent individuals who experienced asymptomatic/mild COVID-19 (~30 days after positive COVID test). Single dose of vaccine induced neutralizing titers comparable to levels seen following asymptomatic/mild SARS-CoV-2 infection. However, neutralizing titers in vaccinees increased several fold following second vaccine dose exceeding those detected following asymptomatic/mild SARS-CoV-2 infection. Antigen-specific B cells were detected after the second dose. Single cell analysis revealed an expansion of activated CD4 T cells but not CD8 T cells post vaccination. Robust antigen specific polyfunctional CD4 T cell responses were observed in all vaccinated individuals. Although CD8 T cell responses were weak and highly variable, effector memory CD8 T cell clones were expanded in every individual following vaccination. TCR gene usage was variable reflecting the diversity of T cell repertoires and MHC polymorphism in human population. Natural infection induced expansion of larger CD8 T cell clones, including distinct clusters likely due to the recognition of a broader set of epitopes presented by the virus not seen in the mRNA vaccine. To comprehensively assess the cellular and humoral immune response to COVID-19 vaccination, we collected blood from SARS-CoV-2 naive volunteers prior to mRNA vaccination (baseline), and 2 weeks following prime-boost vaccination (post-vaccination dose 2; n=4) ( Figure 1A) . These responses were compared to those generated by individuals who experienced asymptomatic SARS-CoV-2 infection using longitudinal samples collected before (baseline) and ~30 days after exposure (convalescent, n=3). The demographics and vaccine information are provided in Supplementary Table 1 . Both infection and vaccination induced binding (Supp Figure 1A ) and neutralizing antibodies (Supp Figure 1B) , as early as 2 weeks following the first dose of vaccination that increased several folds following booster vaccination ( Figure 1B ) and reached slightly higher levels than those achieved following asymptomatic/mild infection (p=0.09). Given that full protection against the virus is achieved two weeks after the booster, we chose this time point for the additional downstream analyses. To specifically assess distinct memory responses, we sorted memory T and B cells, and circulating plasmablasts from PBMC before and 2 weeks after booster vaccination or ~30 days after SARS-CoV-2 infection (Supp Figure 1C) (Figures 1B and 1C) . We next examined the response to vaccination within the B cell compartment. Both vaccination and asymptomatic infection resulted in the reduction of naïve and expansion of memory B cells, with these changes being more prominent with natural infection (Supp Figure 2A ). Importantly, antigen (spike) specific B cells were detected in circulation two weeks after prime-boost vaccination (Supp Figure 2B and Figure 2A ). Examination of memory B cells using single cell RNA sequencing revealed four major clusters ( Figure 2B ) exhibiting distinct patterns of immunoglobulin genes (Supp Figure 2C ): 1) a less mature cluster B1 expressing high IGHD and IGHM; 2) cluster B2 expressing lower IGHM but higher IGHA1; 3) cluster B3 sharing features with B2 but also expressing IGHG1 and IGHG2; and cluster B4 expressing were up-regulated with vaccination and not convalescence ( Figure 3E ). Within the CD8 EM subset, both vaccination and infection up-regulated BCL3 ( Figure 3F ), which is essential for maximum IFN secretion following secondary antigen stimulation. However, tissue homing 8 factor SELPG (encoding P-selectin) and cytotoxic gene GNLY (encoding Granulysin) were induced only with vaccination ( Figure 3F ). We next interrogated antigen specific CD4 and CD8 T cell responses with vaccination. Total PBMC were stimulated with an overlapping peptides library covering the entire sequence of the spike protein for 24 hours, surface stained, fixed, and analyzed for cytokine production using flow cytometry (Supp Figure 3C ). Spike specific polyfunctional IFN +IL-2+ and IFN +TNF + CD4, but not CD8, T cells were evident 2 weeks post prime-boost vaccination Figure 3F) . Finally, enhanced secreted levels of apoptotic factor sFas was observed in both CD4 and CD8 T cells following vaccination (Supp Figure 3G ). Next, we compared the changes in T cell clonal dynamics with infection or vaccination. Vaccination was associated with a shift towards increased CDR3 lengths ( Figure 4A) . Infection was associated with expansion of large clones (>100 cells), while vaccination induced expansion of primarily small sized clones (2-3 cells) (Figures 4B, Supp Figure 4A , B); however these expansions were highly variable among different vaccinated individuals (Supp Figure 4B ). Finally, both vaccination and infection were associated with a drop in clonal diversity; however, this drop was more dramatic following infection ( Figure 4C ). Figure 4G) . We observed diverse patterns of clonal expansion with few clones that expanded dramatically in Vac 1-3 while several smaller clones with limited expansion were detected in Vac4 ( Figure 4D ). Finally, the T cell clones that expanded with vaccination or convalescence occupied distinct space within the UMAP (Figures 4E and 4F) . Interestingly, the top expanded clones following vaccination were mostly CD8 EM with smaller involvement of CD8 CM and activated CD8 T cells (Figures 1B and 4E) . However, expanded clones within convalescent individuals included both CD8 EM and CD8 CM (Figures 1B and 4F ). The establishment of immunity against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has become a central focus of current research efforts. Natural immunity following infection and vaccine-generated immunity provide two different pathways to immunity against the disease. mRNA vaccines have demonstrated significant protection against severe COVID-19 disease. Findings from human trials of Pfizer/BioNTech and Moderna suggest 95% maximal protection within 1-2 months after the second vaccine dose, including against several circulating variants of concern (9, 10). Recommendations from CDC indicate that individuals are not fully protected until 2 weeks after the second dose of vaccine (11). In this study, we (13, 19, 20) . Despite the significant expansion of CD8 T cell clone, frequency of S-specific CD8 T cells was small and variable. This discrepancy could be due to bystander activation. Expanded T cell clones with vaccination and infection occupied distinct space on single cell maps, highlighting differences in the breadth of the epitopes recognized in vaccinated compared to infected individuals (18) . Interestingly, early post boost induction of spike specific memory B cells has been shown to correlate negatively with age following mRNA vaccination (5) . Incidentally, the subject with the lowest CD8 T cell expansion and RBD binding antibodies in this study is also the oldest subject in our cohort . This is in line with data from clinical trials showing lower neutralizing responses after 100 ug dose and faster waning of the response following low dose (25 ug) of mRNA-1273 (12, 21) in the elderly. Furthermore, natural infection has been shown to impair SARS-CoV-2 specific priming of CD8 T cells in the elderly (22, 23) . Whether that defect extends to vaccine induced early CD4 T cell responses or subsequent CD8 T cell expansion remains to be seen. Our study, however, was limited by sample size to draw definitive conclusions on weakening of vaccine responses in the aged. Moreover, it is still unclear what magnitude of neutralizing response confers protective immunity to the virus. Limitations of our study include small sample size, and restriction to participants receiving mRNA vaccine. Due to limited blood samples collected, we were unable to perform additional analysis such as phenotyping of circulating T follicular helper cells (cTfh) following vaccination. Given such few memory B cells from each subject, we were unable to perform rigorous somatic hypermutation analysis at the single cell level. Finally, future studies will have to focus on long-term protection (both cellular and humoral) of two doses of mRNA vaccine against the numerous variants of SARS-CoV-2 and mechanisms of decline in quality of protection (if any) in the elderly. This study was approved by the University of California Irvine Institutional Review Boards. Informed consent was obtained from all enrolled subjects. All participants in this study were healthy and did not report any comorbidities. All vaccines For convalescent subjects (CONV group), blood samples collected pre-exposure to SARS-CoV-2 (Baseline) and ~30 days post convalescence were included in the analysis. These subjects experienced asymptomatic/mild COVID-19. Detailed characteristics of participants and experimental breakdown by sample is provided in Supp. Table 1 . Whole blood samples were collected in EDTA vacutainer tubes. PBMC and plasma samples were isolated after whole blood centrifugation 1200 g for 10 minutes at room temperature in SepMate tubes (STEMCELL Technologies). Plasma was stored at -80 0 C until analysis. PBMC were cryo-preserved using 10% DMSO/FBS and Mr. Frosty Freezing containers (Thermo Fisher Scientific) at -80C then transferred to a cryogenic unit 24 hours later until analysis. dying cells (cells with more than 20% total mitochondrial gene expression), and cells expressing both a TCR and BCR clonotype were excluded during initial QC. Data normalization and variance stabilization was performed on the integrated object using the NormalizeData and ScaleData functions where a regularized negative binomial regression corrected for differential effects of mitochondrial gene expression levels. The HTODemux function was then used to demultiplex donors and further to identify doublets, which were then removed from the analysis. Dimension reduction was performed using RunPCA function to obtain the first 30 principal components followed by integration using Harmony. Clusters were visualized using the UMAP algorithm as implemented by Seurat's RunUMAP function. Cell types were assigned to individual clusters using FindMarkers function with a fold change cutoff of at least 0.4. List of cluster specific markers identified from this study are cataloged in Supp Table 2 . Differential expression analysis was performed using MAST using default settings in Seurat. All disease comparisons were performed relative to healthy donors from corresponding age groups. Only statistically significant genes (Log 10 (fold change) cutoff ≥ 0.25; adjusted p-value ≤ 0.05) were included in downstream analysis TCR and BCR reads were aligned to VDJ-GRCh38 ensembl reference using Cell Ranger 4.0 (10X Genomics) generating sequences and annotations such as gene usage, clonotype frequency, and cell specific barcode information. As an additional QC, only cells with one productive alpha and one productive beta chain were retained for downstream analyses. CDR3 sequences were required to have length between 5 and 27 amino acids, start with a C, and not contain a stop codon. Cells with both TCR and BCR (<0.1%) assignments were excluded from the analysis and all downstream analysis performed using the R package immunarch. Data were first parsed through repLoad function in immunarch, and clonality examined using repExplore function. Family and allele level distributions of TRA and TRB genes were computed using geneUsage function. Diversity estimates (Hill numbers) were calculated using repDiversity function and tracking of abundant clonotypes was performed using trackClonotype function. Clonal assignments based on heavy and light chains were determined using change-o package in Immcantation portal. Briefly, heavy chain file was clonally clustered separately correct clonal groups assigned based on light chain data, removing cells associated with more than one heavy chain. Germline sequences were reconstructed using IgBlast. Gene usage, isotype abundance, and clonotype abundance were calculated using Alakazam package in Immcantation portal. Data sets were first tested for normality. All pairwise comparisons for readouts before/after vaccine and infection were tested using paired t-test. For comparisons involving multiple groups, differences were tested using one-way ANOVA followed by Holm Sidak's multiple comparisons tests. P-values less than or equal to 0.05 were considered statistically significant. Values Dead cells were excluded using the Ghost Dye Red 710 (Tonbo). T cell phenotyping was conducted using an additional panel of antibodies -CD4 CD8b (2ST8.5H7, Beckman Coulter), CCR7 (G043H7, Biolegend) All samples were acquired on the Attune NxT acoustic focusing cytometer (Life Technologies). Data were analyzed using FlowJo v10 (TreeStar, Ashland, OR, USA). and 5% CO 2 . Plates were spun, surface stained using an antibody cocktail containing CD4 (OKT4, BioLegend), CD8b (2ST8.5H7 Approximately 5x10e4 CD4 and CD8 T cells were sorted and stimulated with 1 ug of the SARS-CoV-2 peptide pools Pool 5 (S protein) or anti CD3 (positive control) for 16h at 37C and 5% MIP-1ɑ, MIP-1β, TNFɑ, and Perforin per manufacturer's instructions and run on Magpix The first 12 months of COVID-19: a timeline of immunological insights Nucleosidemodified mRNA vaccines induce potent T follicular helper and germinal center B cell responses Safety and Efficacy of the BNT162b2 mRNA Covid-19 Vaccine Durability of Responses after SARS-CoV-2 mRNA-1273 Vaccination Distinct antibody and memory B cell responses in SARS-CoV-2 naive and recovered individuals following mRNA vaccination mRNA vaccination induces functionally diverse antibodies to NTD, RBD, and S2. Cell vaccine BNT162b1 elicits human antibody and TH1 T cell responses Rapid induction of antigen-specific CD4+ T cells guides coordinated humoral and cellular immune responses to SARS-CoV-2 mRNA vaccination Safety and Efficacy of the BNT162b2 mRNA Covid-19 Vaccine Efficacy and Safety of the mRNA-1273 SARS-CoV-2 Vaccine Low dose mRNA-1273 COVID-19 vaccine generates durable T cell memory and antibodies enhanced by pre-existing crossreactive T cell memory. medRxiv Single-cell multi-omics analysis of the immune response in COVID-19 Evolution of antibody immunity to SARS-CoV-2 Prolonged evolution of the human B cell response to SARS-CoV-2 infection mRNA vaccines induce persistent human germinal centre responses. Nature. 2021. 17. Painter MM. Rapid induction of antigen-specific CD4+ T cells guides coordinated humoral and cellular immune responses to SARS-CoV-2 mRNA vaccination SARS-CoV-2 genome-wide T cell epitope mapping reveals immunodominance and substantial CD8(+) T cell activation in COVID-19 patients Single-cell landscape of immunological responses in patients with COVID-19 Identification of potential vaccine targets for COVID-19 by combining single-cell and bulk TCR sequencing Safety and Immunogenicity of SARS-CoV-2 mRNA-1273 Vaccine in Older Adults Cytotoxic CD8(+) T Cell Response in Elderly COVID-19 Patients Impaired Priming of SARS-CoV-2-Specific Naive CD8+ T Cells in Older Subjects We are grateful to all participants in this study. We thank Dr. Jennifer Atwood for assistance with sorting and imaging flow cytometry in the flow cytometry core at the Institute for Immunology, UCI. We thank Dr. Melanie Oakes from UCI Genomics and High-Throughput Facility for assistance with 10X library preparation and sequencing. Aspects of experimental design figures were generated using graphics from Biorender.com. The authors declare no competing interests. The datasets supporting the conclusions of this article are available on NCBI's Sequence Read Archive (SRA# pending).