Multiple vaccines have been approved to control COVID-19 pandemic, with Pfizer/BioNTech (BNT162b2) being widely used. We conducted a longitudinal analysis of the immune response elicited after three doses of the BNT162b2 vaccine in individuals who have previously experienced SARS-CoV-2 infection and in unexperienced ones. We conducted immunological analyses and single-cell transcriptomics of circulating T and B lymphocytes, combined to CITE-seq or LIBRA-seq, and VDJ-seq. We found that antibody levels against SARS-CoV-2 Spike, NTD and RBD from wild-type, delta and omicron VoCs show comparable dynamics in both vaccination groups, with a peak after the second dose, a decline after six months and a restoration after the booster dose. The antibody neutralization activity was maintained, with lower titers against the omicron variant. Spike-specific memory B cell response was sustained over the vaccination schedule. Clonal analysis revealed that Spike-specific B cells were polyclonal, with a partial clone conservation from natural infection to vaccination. Spike-specific T cell responses were oriented towards effector and effector memory phenotypes, with similar trends in unexperienced and experienced individuals. The CD8 T cell compartment showed a higher clonal expansion and persistence than CD4 T cells. The first two vaccinations doses tended to induce new clones rather than promoting expansion of pre-existing clones. However, we identified a fraction of Spike-specific CD8 T cell clones persisting from natural infection that were boosted by vaccination and clones specifically induced by vaccination. Collectively, our observations revealed a moderate effect of the second dose in enhancing the immune responses elicited after the first vaccination. Differently, we found that a third dose was necessary to restore comparable levels of neutralizing antibodies and Spike-specific T and B cell responses in individuals who experienced a natural SARS-CoV-2 infection.

Interleukin-17 (IL-17)-producing helper T (TH17) cells are heterogenous and consist of nonpathogenic TH17 (npTH17) cells that contribute to tissue homeostasis and pathogenic TH17 (pTH17) cells that mediate tissue inflammation. Here, we characterize regulatory pathways underlying TH17 heterogeneity and discover substantial differences in the chromatin landscape of npTH17 and pTH17 cells both in vitro and in vivo. Compared to other CD4+ T cell subsets, npTH17 cells share accessible chromatin configurations with regulatory T cells, whereas pTH17 cells exhibit features of both npTH17 cells and type 1 helper T (TH1) cells. Integrating single-cell assay for transposase-accessible chromatin sequencing (scATAC-seq) and single-cell RNA sequencing (scRNA-seq), we infer self-reinforcing and mutually exclusive regulatory networks controlling different cell states and predicted transcription factors regulating TH17 cell pathogenicity. We validate that BACH2 promotes immunomodulatory npTH17 programs and restrains proinflammatory TH1-like programs in TH17 cells in vitro and in vivo. Furthermore, human genetics implicate BACH2 in multiple sclerosis. Overall, our work identifies regulators of TH17 heterogeneity as potential targets to mitigate autoimmunity.

Keywords: cross-reactive T cells, T cell cross-reactivity, T cell receptor (TCR), molecular mimicry, T cell repertoire, autoimmune diseases, heterologous immunity and protection, evolution of T cell immunity

Identifying defined T cell clones within a polyclonal population is key to clarifying their phenotype and function. Here, we present a protocol for detecting specified T cell clones in a heterogeneous cell population. We describe steps for stimulating human CD4+ T cells isolated from blood with a protein antigen, sorting antigen -specific cells by fluorescence-activated cell sorting, and detecting among these the presence of predefined T cell clones, based on their T cell receptor (TCR). TCR cDNA is amplified through 50-RACE (TCR-SMART) and detected by qPCR.For complete details on the use and execution of this protocol, please refer to Notarbartolo et al. (2021).1

The COVID-19 pandemic caused by Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) has led to almost seven million deaths worldwide. SARS-CoV-2 causes infection through respiratory transmission and can occur either without any symptoms or with clinical manifestations which can be mild, severe or, in some cases, even fatal. Innate immunity provides the initial defense against the virus by sensing pathogen-associated molecular patterns and triggering signaling pathways that activate the antiviral and inflammatory responses, which limit viral replication and help the identification and removal of infected cells. However, temporally dysregulated and excessive activation of the innate immune response is deleterious for the host and associates with severe COVID-19. In addition to its defensive role, innate immunity is pivotal in priming the adaptive immune response and polarizing its effector function. This capacity is relevant in the context of both SARS-CoV-2 natural infection and COVID-19 vaccination. Here, we provide an overview of the current knowledge of the innate immune responses to SARS-CoV-2 infection and vaccination.

Identifying activated T lymphocytes and differentiating antigen-specific from bystander T cells is crucial for understanding adaptive immune responses. This study investigates the efficacy of activation-induced markers (AIMs) in distinguishing these cell populations. We measured the expression of commonly used AIMs (CD25, CD38, CD40L, CD69, CD137, HLA-DR, ICOS, and OX40) in an in vitro T-cell activation system and evaluated their sensitivity, specificity, and positive predictive value. We demonstrated that individual AIMs, while specific in detecting activated CD4(+) T cells, poorly discriminate between antigen-specific and bystander activation, as assessed by a discriminative capacity (DC) score we developed. Our analysis revealed that dual AIM combinations significantly enhanced the ability to distinguish antigen-specific from bystander-activated T cells, achieving DC scores above 90%. These combinations also improved positive predictive value and specificity with a modest reduction in sensitivity. The CD25(hi)/ICOShi combination emerged as the most efficient, with an average sensitivity of 84.35%, specificity of 99.7%, and DC score of 90.12%. Validation through T-cell cloning and antigen re-stimulation confirmed the robustness of our predictions. This study provides a practical framework for researchers to optimize strategies for identifying and isolating antigen-specific human CD4(+) T lymphocytes and studying their phenotype, function, and T-cell receptor repertoire.

Single-cell RNA sequencing (scRNA-seq) has revolutionised our ability to explore the transcriptional landscape of complex tissues and uncover novel cell types and biological functions. However, the task of identifying and classifying cells from scRNA-seq datasets remains a major challenge. To address this issue, we developed a new computational tool called CIA (Cluster Independent Annotation) that can accurately identify cell types across different datasets without the need for a training dataset or complex machine learning processes. Based on predefined cell type signatures, CIA provides a highly user-friendly and practical solution to functional annotation of single cells. Our results demonstrate that CIA outperforms other state-of-the-art approaches, while also having significantly lower computational running times. Overall, CIA simplifies the process of obtaining graphical representations of signature enrichment scores and classification results, providing researchers with a powerful tool to explore the complex transcriptional landscape of single cells. For further details, see tutorials (). ### Competing Interest Statement The authors have declared no competing interest.

COVID-19 has proven to be particularly serious and life-threatening for patients presenting with pre-existing pathologies. Patients affected by rheumatic musculoskeletal disease (RMD) are likely to have impaired immune responses against SARS-CoV-2 infection due to their compromised immune system and the prolonged use of disease-modifying anti-rheumatic drugs (DMARDs), which include conventional synthetic (cs) DMARDs or biologic and targeted synthetic (b/ts) DMARDs. To provide an integrated analysis of the immune response following SARS-CoV-2 infection in RMD patients treated with different classes of DMARDs we carried out an immunological analysis of the antibody responses toward SARS-CoV-2 nucleocapsid and RBD proteins and an extensive immunophenotypic analysis of the major immune cell populations. We showed that RMD individuals under most DMARD treatments mount a sustained antibody response to the virus, with neutralizing activity. In addition, they displayed a sizable percentage of effector T and B lymphocytes. Among b-DMARDs, we found that anti-TNFα treatments are more favorable drugs to elicit humoral and cellular immune responses as compared to CTLA4-Ig and anti-IL6R inhibitors. This study provides a whole picture of the humoral and cellular immune responses in RMD patients by reassuring the use of DMARD treatments during COVID-19. The study points to TNF-α inhibitors as those DMARDs permitting elicitation of functional antibodies to SARS-CoV-2 and adaptive effector populations available to counteract possible re-infections.

CD4 + and CD8 + T lymphocytes mediate most of the adaptive immune response against tumors. Naïve T lymphocytes specific for tumor antigens are primed in lymph nodes by dendritic cells. Upon activation, antigen-specific T cells proliferate and differentiate into effector cells that migrate out of peripheral blood into tumor sites in an attempt to eliminate cancer cells. After accomplishing their function, most effector T cells die in the tissue, while a small fraction of antigen-specific T cells persist as long-lived memory cells, circulating between peripheral blood and lymphoid tissues, to generate enhanced immune responses when re-encountering the same antigen. A subset of memory T cells, called resident memory T (T RM ) cells, stably resides in non-lymphoid peripheral tissues and may provide rapid immunity independently of T cells recruited from blood. Being adapted to the tissue microenvironment, T RM cells are potentially endowed with the best features to protect against the reemergence of cancer cells. However, when tumors give clinical manifestation, it means that tumor cells have evaded immune surveillance, including that of T RM cells. Here, we review the current knowledge as to how T RM cells are generated during an immune response and then maintained in non-lymphoid tissues. We then focus on what is known about the role of CD4 + and CD8 + T RM cells in antitumor immunity and their possible contribution to the efficacy of immunotherapy. Finally, we highlight some open questions in the field and discuss how new technologies may help in addressing them.

T follicular helper (T FH ) cells play an essential role in promoting B cell responses and antibody affinity maturation in germinal centers (GC). A subset of memory CD4 + T cells expressing the chemokine receptor CXCR5 has been described in human blood as phenotypically and clonally related to GC T FH cells. However, the antigen specificity and relationship of these circulating T FH (cT FH ) cells with other memory CD4 + T cells remain poorly defined. Combining antigenic stimulation and T cell receptor (TCR) Vβ sequencing, we found T cells specific to tetanus toxoid (TT), influenza vaccine (Flu), or Candida albicans ( C.alb ) in both cT FH and non‐cT FH subsets, although with different frequencies and effector functions. Interestingly, cT FH and non‐cT FH cells specific for C.alb or TT had a largely overlapping TCR Vβ repertoire while the repertoire of Flu‐specific cT FH and non‐cT FH cells was distinct. Furthermore, Flu‐specific but not C.alb ‐specific PD‐1 + cT FH cells had a “GC T FH ‐like” phenotype, with overexpression of IL21, CXCL13 , and BCL6 . Longitudinal analysis of serial blood donations showed that Flu‐specific cT FH and non‐cT FH cells persisted as stable repertoires for years. Collectively, our study provides insights on the relationship of cT FH with non‐cT FH cells and on the heterogeneity and persistence of antigen‐specific human cT FH cells.

How gene expression is controlled to preserve human T cell quiescence is poorly understood. Here we show that non-canonical splicing variants containing long interspersed nuclear element 1 (LINE1) enforce naive CD4 + T cell quiescence. LINE1-containing transcripts are derived from CD4 + T cell-specific genes upregulated during T cell activation. In naive CD4 + T cells, LINE1-containing transcripts are regulated by the transcription factor IRF4 and kept at chromatin by nucleolin; these transcripts act in cis , hampering levels of histone 3 (H3) lysine 36 trimethyl (H3K36me3) and stalling gene expression. T cell activation induces LINE1-containing transcript downregulation by the splicing suppressor PTBP1 and promotes expression of the corresponding protein-coding genes by the elongating factor GTF2F1 through mTORC1. Dysfunctional T cells, exhausted in vitro or tumor-infiltrating lymphocytes (TILs), accumulate LINE1-containing transcripts at chromatin. Remarkably, depletion of LINE1-containing transcripts restores TIL effector function. Our study identifies a role for LINE1 elements in maintaining T cell quiescence and suggests that an abundance of LINE1-containing transcripts is critical for T cell effector function and exhaustion.

Th17 cells are a heterogenous cell population consisting of non-pathogenic Th17 cells (npTh17) that contribute to tissue homeostasis and pathogenic Th17 cells (pTh17) that are potent mediators of tissue inflammation. To reveal regulatory mechanisms underlying Th17 heterogeneity, we performed combined ATAC-seq and RNA-seq and discovered substantial differences in the chromatin landscape of npTh17 and pTh17 cells both in vitro and in vivo . Compared to other CD4 + T cell subsets, npTh17 cells share accessible chromatin programs with T regs , and pTh17 cells have an intermediate profile spanning features of npTh17 cells and Th1 cells. Integrating single-cell ATAC-seq and single-cell RNA-seq, we inferred self-reinforcing and mutually exclusive regulatory networks controlling the different cell states and predicted transcription factors (TFs) shaping the chromatin landscape of Th17 cell pathogenicity. We validated one novel TF, BACH2, which promotes immunomodulatory npTh17 programs and restrains pro-inflammatory Th1-like programs in Th17 cells and showed genetic evidence for protective variants in the human BACH2 locus associated with multiple sclerosis. Our work uncovered mechanisms that regulate Th17 heterogeneity, revealed shared regulatory programs with other CD4 + T cell subsets, and identified novel drivers of Th17 pathogenicity as potential targets to mitigate autoimmunity.

Cancer patients have greater chances of being cured if diagnosed at early stages. Integrated information on TCR repertoire and transcriptional signatures of circulating tumour-associated T cells promise the future development of antigen-agnostic minimally invasive tests able to diagnose early-stage cancer. Cancer patients have greater chances of being cured if diagnosed at early stages (Figure 1). Early diagnosis results not only in higher survival rates but also in a better quality of life and lower healthcare costs, thus representing a strategic asset for public healthcare. Indeed, the introduction of screening programs has substantially contributed to reducing cancer mortality rates.1, 2 Currently, organ-specific imaging techniques, such as colonoscopy, mammography, and low-dose computerized tomography scan, are used to detect cancer3 but their sensitivity at early stages can be limited, they can be invasive and possibly have side effects. The identification of circulating cancer-associated biomarkers and cell-free DNA in blood has allowed the development of minimally invasive tests for cancer detection based on liquid biopsies. However, such tests require the pre-selection of cancer-specific panels which limits their availability to selected cancer types. Moreover, the specificity of these markers has been questioned4 and hindered their application in the clinical routine. CD4+ and CD8+ T lymphocytes are pivotal in anti-tumour immunity. Tumour-specific effector T cells are generated as soon as the first immunogenic tumour cells appear, can kill them, and contribute to cancer elimination. Upon clearance of tumour cells, a subset of effector T cells differentiates into tumour-specific resident and recirculating memory T cells, which provide local and systemic protection against the reemergence of the same antigen. T cells exert their immune function also during the equilibrium phase of tumour development before cancer manifests clinically due to the evolution of immune escape mechanisms. Therefore, the detection of tumour-specific T cells may represent a very efficient tool for tumour diagnosis even at stages when the cancer is invisible by imaging. One of the main challenges in detecting tumour-specific T cells is that often the tumour-derived antigens are not known. Strategies based on whole-exome sequencing have been used to identify mutated proteins in tumour cells, but they are relatively inefficient in predicting the peptides recognized by the immune system of the host.5 However, cancer and T cell repertoire coevolve,6 thus changes in the latter may inform about cancer development. The T cell receptor (TCR) defines the antigen specificity of T cells and marks the identity of individual T cell clones. The advent of high-throughput TCR-sequencing technologies, first in bulk and then at the single-cell resolution, has enabled the possibility to monitor changes in the T cell repertoire from limited biological material. Recently, different studies have focused on developing computational tools to identify cancer-associated T cells for the noninvasive early diagnosis of cancer. In this frame, Beshnova et al. developed a deep convolutional neural network model for cancer-associated TCRs (DeepCAT) recognition, based on the complementarity-determining region 3 (CDR3) sequence of the TCRs of T cells infiltrating 4200 tumour samples, covering 32 cancer types, compared to those of T cells from non-cancer controls.7 In this study, CDR3 sequences were reconstructed from bulk RNA-seq data, and their biochemical features were used to construct the computational model. The model was then integrated with TCR-seq data generated from circulating T cells of cancer patients, virus-infected individuals, and healthy controls. The algorithm efficiently predicted TCRs associated with cancer antigens not present in the training dataset and could discern TCRs of tumour-infiltrating T cells from TCRs of circulating ones. Along the same line, Li et al. developed a relatively simple TCR motif-based algorithm to specifically detect early-stage lung cancer, which was able to predict the disease in a validation cohort with a sensitivity of 72% and a specificity of 91%.8 More recently, Ji and colleagues implemented an integrated framework for noninvasive tumour screening starting from a TCR motif-based model. Using both bulk and single-cell TCR-seq data of T cells isolated from patient-matched tumours and peripheral blood mononuclear cells, they first defined a TCR repertoire risk score based on the relative frequency of circulating tumour-associated TCRs compared with that of non-tumour-associated TCRs identified in healthy individuals. Then, since T cell activation is usually accompanied by proliferation, they integrated information about the clonal diversity of the identified tumour-associated TCRs as a measure of T cell clonal expansion. Finally, they identified a tumour-associated T cell transcriptional signature that combined with the TCR repertoire risk score generated a Cancer Risk Score.9 These studies paved the way for developing new diagnostic tools for the detection of early-stage cancer (Figure 2). However, there are still some major issues to tackle and solve before these tools can reach applicability in the clinical routine. First, while the mentioned tools were quite efficient in distinguishing between TCRs associated with cancer and those from healthy individuals, their performance was not as brilliant when tested for discriminating between TCRs associated with cancer and viral infections.7 This observation may raise concerns about the specificity of the algorithms and calls for further improvement and more thorough testing in prospective studies in a real-world clinical setting. The increasing number of single-cell sequencing datasets generated from patients with microbial infections (e.g. by SARS-CoV-2) may provide the data to be integrated during model construction to improve algorithm performances in this direction. Second, it will be necessary to define the ability of the mentioned computational tools to discriminate the type of cancer eventually detected. The detection of early-stage cancer, invisible to other diagnostic techniques, without additional information about its origin or anatomic location, may be of limited utility. It would warn patients about the need for follow-up screenings but may induce physicians to expose patients to an excess of (possibly inconclusive) diagnostic tests. To overcome this issue, a T cell-based diagnostic algorithm should be implemented to recognize tumour type-specific features for instance, based on tissue-specific transcriptional signatures that are maintained by tumour-specific T cells upon tissue egress and entry into the circulation. Finally, whether these algorithms will be able to detect any cancer type or if they will require a minimal tumour mutational load or T cell infiltration has to be determined yet. In conclusion, T cell-based methods, integrating information on TCR repertoire and transcriptional signatures, upon overcoming current limitations, promise the future development of antigen-agnostic minimally invasive tests able to diagnose early-stage cancer with a higher and broader specificity than current biomarker-based tests and superior sensitivity to existing imaging-based techniques. Dr. S. Notarbartolo contributed to the preparation and collection of original literatures and figures and the writing and editing of manuscript. Dr. S. Notarbartolo was responsible for the structural designs, scientific quality and writing. Not applicable. Not applicable. The authors declare no conflict of interest. Not applicable. Data sharing is not applicable to this article as no new data were created or analyzed in this study.

We have described a child suffering from Mendelian susceptibility to mycobacterial disease (MSMD) due to autosomal recessive, complete T-bet deficiency, which impairs IFN-γ production by innate and innate-like adaptive, but not mycobacterial-reactive purely adaptive, lymphocytes. Here, we explore the persistent upper airway inflammation (UAI) and blood eosinophilia of this patient. Unlike wild-type (WT) T-bet, the mutant form of T-bet from this patient did not inhibit the production of Th2 cytokines, including IL-4, IL-5, IL-9, and IL-13, when overexpressed in T helper 2 (Th2) cells. Moreover, Herpesvirus saimiri–immortalized T cells from the patient produced abnormally large amounts of Th2 cytokines, and the patient had markedly high plasma IL-5 and IL-13 concentrations. Finally, the patient’s CD4+ αβ T cells produced most of the Th2 cytokines in response to chronic stimulation, regardless of their antigen specificities, a phenotype reversed by the expression of WT T-bet. T-bet deficiency thus underlies the excessive production of Th2 cytokines, particularly IL-5 and IL-13, by CD4+ αβ T cells, causing blood eosinophilia and UAI. The MSMD of this patient results from defective IFN-γ production by innate and innate-like adaptive lymphocytes, whereas the UAI and eosinophilia result from excessive Th2 cytokine production by adaptive CD4+ αβ T lymphocytes.

To understand how a protective immune response against SARS-CoV-2 develops over time, we integrated phenotypic, transcriptional and repertoire analyses on PBMCs from mild and severe COVID-19 patients during and after infection, and compared them to healthy donors (HD). A type I IFN-response signature marked all the immune populations from severe patients during the infection. Humoral immunity was dominated by IgG production primarily against the RBD and N proteins, with neutralizing antibody titers increasing post infection and with disease severity. Memory B cells, including an atypical FCRL5+ T-BET+ memory subset, increased during the infection, especially in patients with mild disease. A significant reduction of effector memory, CD8+ T cells frequency characterized patients with severe disease. Despite such impairment, we observed robust clonal expansion of CD8+ T lymphocytes, while CD4+ T cells were less expanded and skewed toward TCM and TH2-like phenotypes. MAIT cells were also expanded, but only in patients with mild disease. Terminally differentiated CD8+ GZMB+ effector cells were clonally expanded both during the infection and post-infection, while CD8+ GZMK+ lymphocytes were more expanded post-infection and represented bona fide memory precursor effector cells. TCR repertoire analysis revealed that only highly proliferating T cell clonotypes, which included SARS-CoV-2-specific cells, were maintained post-infection and shared between the CD8+ GZMB+ and GZMK+ subsets. Overall, this study describes the development of immunity against SARS-CoV-2 and identifies an effector CD8+ T cell population with memory precursor-like features.

In response to pathogenic threats, naive T cells rapidly transition from a quiescent to an activated state, yet the underlying mechanisms are incompletely understood. Using a pulsed SILAC approach, we investigated the dynamics of mRNA translation kinetics and protein turnover in human naive and activated T cells. Our datasets uncovered that transcription factors maintaining T cell quiescence had constitutively high turnover, which facilitated their depletion following activation. Furthermore, naive T cells maintained a surprisingly large number of idling ribosomes as well as 242 repressed mRNA species and a reservoir of glycolytic enzymes. These components were rapidly engaged following stimulation, promoting an immediate translational and glycolytic switch to ramp up the T cell activation program. Our data elucidate new insights into how T cells maintain a prepared state to mount a rapid immune response, and provide a resource of protein turnover, absolute translation kinetics and protein synthesis rates in T cells (https://www.immunomics.ch).

In the version of this article initially published, in the legend to Fig. 1b, the description of the frequency of TH17-IL-10+ clones was incomplete for the first group; this should read as follows: "...13 experiments with clones isolated from CCR6+CCR4+CXCR3- T cells...". Also, the label along the vertical axis of the bottom right plot in Figure 5b was incomplete; the correct label is 'IFN-γ+ cells (%)'. Finally, in the first sentence of the final paragraph of the final Results subsection, the description of the regions analyzed was incorrect; that sentence should begin: "DNA motif-enrichment analysis of the subset-specific H3K27ac-positive regions...". The errors have been corrected in the HTML and PDF versions of the article.