Faculty Bibliography

Both SARS-CoV-2 infection and vaccination elicit potent immune responses. A number of studies have described immune responses to SARS-CoV-2 infection. However, beyond antibody production, immune responses to COVID-19 vaccines remain largely uncharacterized. Here, we performed multimodal single-cell sequencing on peripheral blood of patients with acute COVID-19 and healthy volunteers before and after receiving the SARS-CoV-2 BNT162b2 mRNA vaccine to compare the immune responses elicited by the virus and by this vaccine. Phenotypic and transcriptional profiling of immune cells, coupled with reconstruction of the B and T cell antigen receptor rearrangement of individual lymphocytes, enabled us to characterize and compare the host responses to the virus and to defined viral antigens. While both infection and vaccination induced robust innate and adaptive immune responses, our analysis revealed significant qualitative differences between the two types of immune challenges. In COVID-19 patients, immune responses were characterized by a highly augmented interferon response which was largely absent in vaccine recipients. Increased interferon signaling likely contributed to the observed dramatic upregulation of cytotoxic genes in the peripheral T cells and innate-like lymphocytes in patients but not in immunized subjects. Analysis of B and T cell receptor repertoires revealed that while the majority of clonal B and T cells in COVID-19 patients were effector cells, in vaccine recipients clonally expanded cells were primarily circulating memory cells. Importantly, the divergence in immune subsets engaged, the transcriptional differences in key immune populations, and the differences in maturation of adaptive immune cells revealed by our analysis have far-ranging implications for immunity to this novel pathogen.

UNLABELLED:Adaptive immune responses are induced by vaccination and infection, yet little is known about how CD4+ T cell memory differs between these two contexts. Notable differences in humoral and cellular immune responses to primary mRNA vaccination were observed and associated with prior COVID-19 history, including in the establishment and recall of Spike-specific CD4+ T cells. It was unclear whether CD4+ T cell memory established by infection or mRNA vaccination as the first exposure to Spike was qualitatively similar. To assess whether the mechanism of initial memory T cell priming affected subsequent responses to Spike protein, 14 people who were receiving a third mRNA vaccination, referenced here as the booster, were stratified based on whether the first exposure to Spike protein was by viral infection or immunization (infection-primed or vaccine-primed). Using multimodal scRNA-seq of activation-induced marker (AIM)-reactive Spike-specific CD4+ T cells, we identified 220 differentially expressed genes between infection- and vaccine-primed patients at the post-booster time point. Infection-primed participants had greater expression of genes related to cytotoxicity and interferon signaling. Gene set enrichment analysis (GSEA) revealed enrichment for Interferon Alpha, Interferon Gamma, and Inflammatory response gene sets in Spike-specific CD4+ T cells from infection-primed individuals, whereas Spike-specific CD4+ T cells from vaccine-primed individuals had strong enrichment for proliferative pathways by GSEA. Finally, SARS-CoV-2 breakthrough infection in vaccine-primed participants resulted in subtle changes in the transcriptional landscape of Spike-specific memory CD4+ T cells relative to pre-breakthrough samples but did not recapitulate the transcriptional profile of infection-primed Spike-specific CD4+ T cells. Together, these data suggest that CD4+ T cell memory is durably imprinted by the inflammatory context of SARS-CoV-2 infection, which has implications for personalization of vaccination based on prior infection history. ONE SENTENCE SUMMARY/UNASSIGNED:SARS-CoV-2 infection and mRNA vaccination prime transcriptionally distinct CD4+ T cell memory landscapes which are sustained with subsequent doses of vaccine.

The current abundance of immunotherapy clinical trials presents an opportunity to learn about the underlying mechanisms and pharmacodynamic effects of novel drugs on the human immune system. Here, we present a protocol to study how these immune responses impact clinical outcomes using large-scale high-throughput immune profiling of clinical cohorts. We describe the Human Immune Profiling Pipeline, which comprises an end-to-end solution from flow cytometry results to computational approaches and unsupervised patient clustering based on lymphocyte landscape. For complete details on the use and execution of this protocol, please refer to Lyudovyk et al. (2022).¹.

The mechanisms by which FOXP3⁺ T follicular regulatory (Tfr) cells simultaneously steer antibody formation toward microbe or vaccine recognition and away from self-reactivity remain incompletely understood. To explore underappreciated heterogeneity in human Tfr cell development, function, and localization, we used paired TCRVA/TCRVB sequencing to distinguish tonsillar Tfr cells that are clonally related to natural regulatory T cells (nTfr) from those likely induced from T follicular helper (Tfh) cells (iTfr). The proteins iTfr and nTfr cells differentially expressed were used to pinpoint their in situ locations via multiplex microscopy and establish their divergent functional roles. In silico analyses and in vitro tonsil organoid tracking models corroborated the existence of separate T_reg-to-nTfr and Tfh-to-iTfr developmental trajectories. Our results identify human iTfr cells as a distinct CD38⁺, germinal center-resident, Tfh-descended subset that gains suppressive function while retaining the capacity to help B cells, whereas CD38^- nTfr cells are elite suppressors primarily localized in follicular mantles. Interventions differentially targeting specific Tfr cell subsets may provide therapeutic opportunities to boost immunity or more precisely treat autoimmune diseases.

The human immune response to SARS-CoV-2 antigen after infection or vaccination is defined by the durable production of antibodies and T cells. Population-based monitoring typically focuses on antibody titer, but there is a need for improved characterization and quantification of T cell responses. Here, we utilize multimodal sequencing technologies to perform a longitudinal analysis of circulating human leukocytes collected before and after BNT162b2 immunization. Our data reveal distinct subpopulations of CD8 ⁺ T cells which reliably appear 28 days after prime vaccination (7 days post boost). Using a suite of cross-modality integration tools, we define their transcriptome, accessible chromatin landscape, and immunophenotype, and identify unique biomarkers within each modality. By leveraging DNA-oligo-tagged peptide-MHC multimers and T cell receptor sequencing, we demonstrate that this vaccine-induced population is SARS-CoV-2 antigen-specific and capable of rapid clonal expansion. Moreover, we also identify these CD8 ⁺ populations in scRNA-seq datasets from COVID-19 patients and find that their relative frequency and differentiation outcomes are predictive of subsequent clinical outcomes. Our work contributes to our understanding of T cell immunity, and highlights the potential for integrative and multimodal analysis to characterize rare cell populations.

Around the world, rollout of COVID-19 vaccines has been used as a strategy to end COVID-19-related restrictions and the pandemic. Several COVID-19 vaccine platforms have successfully protected against severe SARS-CoV-2 infection and subsequent deaths. Here, we compared humoral and cellular immunity in response to either infection or vaccination. We examined SARS-CoV-2 spike-specific immune responses from Pfizer/BioNTech BNT162b2, Moderna mRNA-1273, Janssen Ad26.COV2.S, and SARS-CoV-2 infection approximately 4 months post-exposure or vaccination. We found that these three vaccines all generate relatively similar immune responses and elicit a stronger response than natural infection. However, antibody responses to recent viral variants are diminished across all groups. The similarity of immune responses from the three vaccines studied here is an important finding in maximizing global protection as vaccination campaigns continue.

Background Checkpoint inhibitor (CPI) antibodies induce blockade of CTLA-4 and PD-1 to reinvigorate the immune system to induce tumor cell clearance. Despite the success of CPI in cancer treatment, CPIs often result in toxicity due to immune related adverse events (irAEs).1 Both mechanistic understanding and predictive biomarkers for irAEs is lacking. It is necessary to characterize the effect of CPI on the immune system to reduce therapeutic toxicity and improve patient outcomes. CD4 T follicular helper cells (Tfh) strongly express PD-1 and CTLA-4, yet little is known about the effects of CPI on these cells. Tfh provide help to antibody producing B cells within germinal centers in lymphoid tissue.2 Although this is an essential mechanism of adaptive immunity, several groups have shown hyperactivity in Tfh drives germinal center activity, resulting in increased help to autoreactive B cells and the generation of disease-causing autoantibodies.3, 4, 5 In the context of CPI, Tfh dysfunction may also drive the induction of autoantibodies and associated irAEs. Methods Using flow cytometry and scRNAseq, we are analyzing PBMC samples from the Checkmate-238 clinical trial in which patients with resected stage III or IV melanoma received either aPD-1 or aCTLA-4 monotherapy.6Patients had blood drawn on the same day as the first immunotherapy infusion (baseline) and two weeks post-baseline. Tfh may be key to understanding irAE, but the study of Tfh in humans is challenging as these cells are commonly found in germinal centers of lymph nodes. To overcome this, our group has shown that circulating Tfh cells (cTfh) are recent emigrants from the lymph node and can be used to study cellular dynamics.7 Results In this study, aCTLA-4-treated patients developed irAE at higher rates than aPD1-treated patients, with 46% of aCTLA-4 patients developing Grade 3 or 4 irAE compared to 14% of aPD-1 patients.6 We are interested in determining if the higher rate of irAE associated with aCTLA-4 was correlated to cTfh responses. Indeed, in flow cytometry studies, our lab identified a dramatic induction of activated cTfh following aCTLA-4 but not aPD-1 therapy, with a median foldchange of 7.5 within two weeks after baseline in aCTLA-4 patients but no difference in aPD-1 patients (figure 1). Conclusions The greater influx of cTfh and higher incidence of severe irAEs suggest that aCTLA-4 therapy may predispose patients to dysregulated Tfh proliferation and induction of autoantibody responses.

Anti-programmed death-1 (anti-PD-1) immunotherapy reinvigorates CD8 T cell responses in patients with cancer but PD-1 is also expressed by other immune cells, including follicular helper CD4 T cells (Tfh) which are involved in germinal centre responses. Little is known, however, about the effects of anti-PD-1 immunotherapy on noncancer immune responses in humans. To investigate this question, we examined the impact of anti-PD-1 immunotherapy on the Tfh-B cell axis responding to unrelated viral antigens. Following influenza vaccination, a subset of adults receiving anti-PD-1 had more robust circulating Tfh responses than adults not receiving immunotherapy. PD-1 pathway blockade resulted in transcriptional signatures of increased cellular proliferation in circulating Tfh and responding B cells compared with controls. These latter observations suggest an underlying change in the Tfh-B cell and germinal centre axis in a subset of immunotherapy patients. Together, these results demonstrate dynamic effects of anti-PD-1 therapy on influenza vaccine responses and highlight analytical vaccination as an approach that may reveal underlying immune predisposition to adverse events.

BACKGROUND:SARS-CoV-2 vaccines currently authorized for emergency use have been highly successful in preventing infection and lessening disease severity. The vaccines maintain effectiveness against earlier SARS-CoV-2 Variants of Concern but the heavily mutated, highly transmissible Omicron variant presents an obstacle both to vaccine protection and monoclonal antibody therapies. METHODS:Pseudotyped lentiviruses were incubated with serum from vaccinated and boosted donors or therapeutic monoclonal antibody and then applied to target cells. After 2 days, luciferase activity was measured in a microplate luminometer. Resistance mutations of the Omicron spike were identified using point-mutated spike protein pseudotypes and mapped onto the three-dimensional spike protein structure. FINDINGS/RESULTS:Virus with the Omicron spike protein was 26-fold resistant to neutralization by recovered donor sera and 26-34-fold resistance to Pfizer BNT162b2 and Moderna vaccine-elicited antibodies following two immunizations. A booster immunization increased neutralizing titres against Omicron. Neutralizing titres against Omicron were increased in the sera with a history of prior SARS-CoV-2 infection. Analysis of the therapeutic monoclonal antibodies showed that the Regeneron and Eli Lilly monoclonal antibodies were ineffective against the Omicron pseudotype while Sotrovimab and Evusheld were partially effective. INTERPRETATION/CONCLUSIONS:The results highlight the benefit of a booster immunization to protect against the Omicron variant and demonstrate the challenge to monoclonal antibody therapy. The decrease in neutralizing titres against Omicron suggest that much of the vaccine efficacy may rely on T cells. FUNDING/BACKGROUND:The work was funded by grants from the NIH to N.R.L. (DA046100, AI122390 and AI120898) and 55 to M.J.M. (UM1AI148574).