Persistent antigen exposure during chronic viral infection and tumor development drives CD8 T cells into an exhausted, hypofunctional state. Understanding the molecular pathways that enforce T-cell exhaustion is critical for improving current immunotherapies. Previously, we have shown the bioactive lipid lysophosphatidic acid (LPA) regulates CD8 T-cell function through LPA receptor 5 (LPAR5) signaling, including demonstrating that Lpar5-/- CD8 T cells exhibit enhanced tumor clearance in murine models of melanoma. Importantly, significantly elevated levels of LPA have been identified in individuals with different cancers and persistent viral infections such as HIV, hepatitis C virus, and hepatitis B virus. To investigate the role of Lpar5 in the differentiation and maintenance of exhausted CD8 T cells, we utilized the lymphocytic choriomeningitis virus (LCMV) infection model. In response to infection with LCMV Clone 13, but not Armstrong, one-quarter of Lpar5-/- animals succumbed to infection, and this was accompanied by an increased frequency of LCMV-specific Lpar5-/- CD8 T cells maintained in a less terminally exhausted state. Using P14 transgenic mice, we demonstrate that Lpar5 acts in a cell-intrinsic and temporal manner to regulate CD8 T-cell accumulation and exhaustion programming during Clone 13 infection. The enhanced accumulation of Lpar5-/- P14 cells during the acute phase of Clone 13 infection appears to be regulated by Lpar5-mediated changes in T-cell survival and not through trafficking or proliferation. RNA sequencing analyses and surface phenotyping show that Lpar5 likely regulates CD8 T-cell exhaustion through modulation of NK receptor expression, including the CD94/NKG2A inhibitory axis.

As a metabolic center, the liver prevents inappropriate immune responses to abundant dietary antigens within the liver that could result in liver injury. This self-preservation mechanism can however decrease the efficiency of immunosurveillance of malignant cells by CD8 T cells. Hepatocellular carcinoma (HCC) is initiated by chronic viral infections, chronic alcohol consumption, and/or a fatty diet that leads to liver injury, fibrosis, and cirrhosis. HCC patients have high levels of dysfunctional and exhausted T cells, however, it is unclear which stage of HCC development contributes to T cell dysfunction. Repair of liver injury is initiated by interactions between injured hepatocytes and liver sinusoidal endothelial cells (LSEC), however, chronic injury can lead to fibrosis. Here, using a diethylnitrosamine/carbon tetrachloride (DEN/CCl4) mouse model of early HCC development, we demonstrate that chronic liver injury and fibrosis are sufficient to induce a CD8 T cell exhaustion signature with a corresponding increase in expression of immunosuppressive molecules on LSEC. We show that LSEC alter T cell function at various stages of T cell differentiation/activation. LSEC compete with dendritic cells presenting the same antigen to naïve CD8 T cells resulting in a unique T cell phenotype. Furthermore, LSEC abrogate killing of target cells, in an antigen-dependent manner, by previously activated effector CD8 T cells, and LSEC change the effector cell cytokine profile. Moreover, LSEC induce functional T cell exhaustion under low dose chronic stimulation conditions. Thus, LSEC critically regulate the balance between preventing/limiting liver injury and permitting sufficient tumor immunosurveillance with normal hepatic functions likely contributing to HCC development under conditions of chronic liver insult.

Resolving the molecular mechanisms of central B cell tolerance might unveil strategies that prevent autoimmunity. Here, using a mouse model of central B cell tolerance in which Forkhead box protein O1 (Foxo1) is either deleted or over-expressed in B cells, we show that deleting Foxo1 blocks receptor editing, curtails clonal deletion, and decreases CXCR4 expression, allowing high-avidity autoreactive B cells to emigrate to the periphery whereby they mature but remain anergic and short lived. Conversely, expression of degradation-resistant Foxo1 promotes receptor editing in the absence of self-antigen but leads to allelic inclusion. Foxo1 over-expression also restores tolerance in autoreactive B cells harboring active PI3K, revealing opposing roles of Foxo1 and PI3K in B cell selection. Overall, we show that the transcription factor Foxo1 is a major gatekeeper of central B cell tolerance and that PI3K drives positive selection of immature B cells and establishes allelic exclusion by suppressing Foxo1.

Precise regulation of B cell differentiation is essential for an effective adaptive immune response. Here, we show that B cell development in mice with B cell-specific Maf deletion is unaffected, but marginal zone B cells, germinal centre B cells, and plasmablasts are significantly more frequent in the spleen of naive Maf-deficient mice compared to wild type controls. In the context of a T cell-dependent immunization, Maf deletion causes increased proliferation of germinal centre B cells and extrafollicular plasmablasts. This is accompanied by higher production of antigen-specific IgG1 antibodies with minimal modification of early memory B cells, but a reduction in plasma cell numbers. Single-cell RNA sequencing shows upregulation of genes associated with DNA replication and cell cycle progression, confirming the role of Maf in cell proliferation. Subsequent pathway analysis reveals that Maf influences cellular metabolism, transporter activity, and mitochondrial proteins, which have been implicated in controlling the germinal centre reaction. In summary, our findings demonstrate that Maf acts intrinsically in B cells as a negative regulator of late B cell differentiation, plasmablast proliferation and germinal centre B cell formation. Maf is a transcription factor regulating pivotal biological processes in multiple immune cells, but its B-cell-intrinsic role is not fully known. Here authors show that genomic deletion of Maf in the B cell lineage does not disturb the sequence of developmental stages, however, removes an important inhibitory step to restrict the early steps of germinal centre B cell and extrafollicular plasmablast population expansion.

Memory B cells are comprised of unswitched (CD27+IgD+) and switched (CD27+IgD-) subsets. The origin and function of unswitched human memory B cells are debated in the literature, whereas switched memory B cells are primed to respond to recurrent infection. Unswitched memory B cells have been described to be reduced in frequency with severe SARS-CoV2 infection and here we characterize their activation status, BCR functionality, and contribution to virally-induced cytokine production. Analyses of whole blood from healthy individuals, people immunized against SARS-CoV2, and those who have had mild and severe SARS-CoV2 infection, confirm a reduction in the frequency of unswitched memory B cells during severe SARS-CoV2 infection and demonstrate this reduction is associated with increased levels of systemic TNFα. We further document how severe viral infection is associated with an increased frequency of 'IgD+' only memory B cells that correlate with increased IgG autoantibody levels. Unswitched and switched memory B cells from severe SARS-CoV2 infection displayed evidence of heightened activation with a concomitant reduction in the expression of the inhibitory receptor CD72. Functionally, both populations of memory B cells from severe SARS-COV2 infection harbored a signaling-competent BCR that displayed enhanced BCR signaling activity in the unswitched population. Finally, we demonstrate that B cells from mild SARS-CoV2 infection are poised to secrete pro-inflammatory cytokines IL-6 and TNFα. Importantly, unswitched memory B cells were a major producer of IL-6 and switched memory B cells were a major producer of TNFα in response to viral TLR ligands. Together these data indicate that B cells contribute to the inflammatory milieu during viral infection.

Lysophosphatidic acid (LPA) is an endogenous bioactive lipid that is produced extracellularly and signals to cells via cognate LPA receptors, which are G-protein coupled receptors (GPCRs). Mature lymphocytes in mice and humans express three LPA receptors, LPA2 , LPA5, and LPA6 , and work from our group has determined that LPA5 signaling by T lymphocytes inhibits specific antigen-receptor signaling pathways that ultimately impair lymphocyte activation, proliferation, and function. In this review, we discuss previous and ongoing work characterizing the ability of an LPA-LPA5 axis to serve as a peripheral immunological tolerance mechanism that restrains adaptive immunity but is subverted during settings of chronic inflammation. Specifically, LPA-LPA5 signaling is found to regulate effector cytotoxic CD8 T cells by (at least) two mechanisms: (i) regulating the actin-microtubule cytoskeleton in a manner that impairs immunological synapse formation between an effector CD8 T cell and antigen-specific target cell, thus directly impairing cytotoxic activity, and (ii) shifting T-cell metabolism to depend on fatty-acid oxidation for mitochondrial respiration and reducing metabolic efficiency. The in vivo outcome of LPA5 inhibitory activity impairs CD8 T-cell killing and tumor immunity in mouse models providing impetus to consider LPA5 antagonism for the treatment of malignancies and chronic infections.

Newly generated immature B cells that bind self-antigen with high avidity arrest in differentiation and undergo central tolerance via receptor editing and clonal deletion. These autoreactive immature B cells also express low surface levels of the coreceptor CD19, a key activator of the PI3K pathway. Signals emanating from both CD19 and PI3K are known to be critical for attenuating receptor editing and selecting immature B cells into the periphery. However, the mechanisms that modulate CD19 expression at this stage of B cell development have not yet been resolved. Using in vivo and in vitro models, we demonstrate that Cd19 de novo gene transcription and translation do not significantly contribute to the differences in CD19 surface expression in mouse autoreactive and nonautoreactive immature B cells. Instead, CD19 downregulation is induced by BCR stimulation in proportion to BCR engagement, and the remaining surface IgM and CD19 molecules promote intracellular PI3K-AKT activity in proportion to their level of expression. The internalized CD19 is degraded with IgM by the lysosome, but inhibiting lysosome-mediated protein degradation only slightly improves surface CD19. In fact, CD19 is restored only upon Ag removal. Our data also reveal that the PI3K-AKT pathway positively modulates CD19 surface expression in immature B cells via a mechanism that is independent of inhibition of FOXO1 and its role on Cd19 gene transcription while is dependent on mTORC1.

Lysophosphatidic acid (LPA) is a bioactive lipid which increases in concentration locally and systemically across different cancer types. Yet, the exact mechanism(s) of how LPA affects CD8 T cell immunosurveillance during tumor progression remain unknown. We show LPA receptor (LPAR) signaling by CD8 T cells promotes tolerogenic states via metabolic reprogramming and potentiating exhaustive-like differentiation to modulate anti-tumor immunity. We found LPA levels predict response to immunotherapy and Lpar5 signaling promotes cellular states associated with exhausted phenotypes on CD8 T cells. Importantly, we show that Lpar5 regulates CD8 T cell respiration, proton leak, and reactive oxygen species. Together, our findings reveal that LPA serves as a lipid-regulated immune checkpoint by modulating metabolic efficiency through LPAR5 signaling on CD8 T cells. Our study offers key insights into the mechanisms governing adaptive anti-tumor immunity and demonstrates LPA could be exploited as a T cell directed therapy to improve dysfunctional anti-tumor immunity.

Memory B cells are comprised of unswitched (CD27+IgD+) and switched (CD27+IgD-) subsets. The origin and function of unswitched human memory B cells remains controversial, whereas switched memory B cells are primed to respond to recurrent infection. Unswitched memory B cells were described to be reduced in frequency with severe SARS-CoV2 infection and here we report on their activation status, BCR functionality, and contribution to virally-induced cytokine production. Analyses of whole blood from healthy individuals, people immunized against SARS-CoV2, and those with mild or severe SARS-CoV2 infection, confirm a loss of unswitched memory B cells during severe SARS-CoV2 infection and demonstrate this loss is associated with increased levels of systemic TNFα. We further report an increased frequency of ‘IgD+’ only memory B cells that correlate with an increase in IgG autoantibody levels (anti-chromatin, anti-cardiolipin, anti-smith) in severe SARS-CoV2 infection. Unswitched and switched memory B cells from severe SARS-CoV2 infection displayed evidence of heightened activation with concomitant reduction in expression of the inhibitory receptor CD72. Functionally, both populations of memory B cells from severe SARS-COV2 infection harbored a signaling competent BCR that displayed enhanced activity in the unswitched population. Finally, we demonstrate that B cells from mild SARS-CoV2 infection secrete pro-inflammatory cytokines IL-6 and TNFα. Importantly, unswitched memory B cells were a major producer of IL-6 and switched memory B cells were a major producer of TNFα in response to viral TLR ligands. Together these data suggest that B cells contribute to the inflammatory milieu during viral infection. Supported by grants from NIH R01 AI124474 (RP) and R01 AI136534 (RMT).

PDF file - 75K, Supplementary Figure S1. Intracellular calcium mobilization was assessed in CD8+ T cells stimulated with 10 g/mL anti-CD3 with or without 20 M LPA (thin and bold lines, respectively). 20 M LPA treatment in the absence of TCR stimulation is shown in gray line. Supplementary Figure S2. LPA treatment inhibits CD8+ T cell activation with suboptimal TCR stimulation. C57BL/6 splenocytes were cultured for 24 hours with (center and right histograms) or without (left histogram) indicated plate-bound anti-CD3 and 1 g/mL soluble anti-CD28, in the presence (blue and green lines) or absence of LPA (red line, gray filled). Data shown is gated on CD8+ T cells. Representative of 2 independent experiments. Supplementary Figure S3. OTP pretreatment of APC does not affect antigen presentation. Erythrocyte-lysed splenocytes from a T cell-deficient mouse (TCR-/-) were pulsed with SIINFEKL peptide in the presence (black) or absence (white) of OTP, washed, and used to stimulate purified CD8+ OT-I T cells in the presence or absence of OTP. Supplementary Figure S4. Generation and characterization of LPA5-/- mice. A, Schematic diagram of Lpar5 targeted deletion. A portion of the Lpar5 gene was replaced with a LacZ/neo selection cassette. B, PCR genotyping displaying wild-type and mutant Lpar5. Primers are indicated in Supp. Fig. 4A. C, Southern hybridization indicating the proper targeting event in ESC clones. D, Proportions of T cells and CD4+ and CD8+ T cell subsets in spleens of LPA5+/+ and LPA5-/- mice.

The random recombination of immunoglobulin V(D)J gene segments produces unique IgM antibodies that serve as the antigen receptor for each developing B cell. Hence, the newly formed B cell repertoire is comprised of a variety of specificities that display a range of reactivity with self‐antigens. Newly generated IgM + immature B cells that are non‐autoreactive or that bind self‐antigen with low avidity are licensed to leave the bone marrow with their intact antigen receptor and to travel via the blood to the peripheral lymphoid tissue for further selection and maturation. In contrast, clones with medium to high avidity for self‐antigen remain within the marrow and undergo central tolerance, a process that revises their antigen receptor or eliminates the autoreactive B cell altogether. Thus, central B cell tolerance is critical for reducing the autoreactive capacity and avidity for self‐antigen of our circulating B cell repertoire. Bone marrow cultures and mouse models have been instrumental for understanding the mechanisms that regulate the selection of bone marrow B cells. Here, we review recent studies that have shed new light on the contribution of the ERK, PI3K, and CXCR4 signaling pathways in the selection of mouse and human immature B cells that either bind or do not bind self‐antigen.

Significance Cancers and chronic infectious pathogens often evade immune-mediated elimination by suppressing T cell function via engaging inhibitory receptors expressed on T cells. The phospholipid lysophosphatidic acid (LPA) is often increased systemically from basal concentrations upon development of cancer and chronic infections. We previously showed that, at these elevated concentrations, LPA suppresses the ability of CD8 T cells to kill malignant cells and to control tumor growth. Here, we demonstrate that LPA signaling suppresses T cell function via disrupting T cell receptor–induced cytoskeletal dynamics, immune synapse formation, signal transduction, and the tubulin code. This report identifies a targetable mechanism of receptor-mediated inhibition of T cell function that could be used in combination therapies to enhance antitumor and antiviral immunity.

Generation of enormous antigen receptor repertoire diversity through random VDJ recombination is the cornerstone of the adaptive immune system, but it comes at the cost of generating large numbers of self-reactive lymphocytes with potential for pathogenesis. It has long been appreciated that a series of elegant tolerance mechanisms are employed by lymphocytes to prevent autoimmune disease. Several of these are broadly analogous in T and B cells, including deletion in response to strong antigen signals delivered early during development, a requirement for co-stimulation in addition to antigen recognition to license immune responses by mature lymphocytes, and functional unresponsiveness of mature self-reactive lymphocytes (“anergy”). In addition, there are several unique mechanisms that distinguish T and B cell tolerance, such as receptor editing in central B cell tolerance and the generation of regulatory T cells. These mechanisms were discovered and molecularly dissected in large part through the development and study of BCR and TCR transgenic animals and other genetic mouse models. Newer insights are emerging through study of both mouse and human immunology and through cutting-edge analyses of unusual patient phenotypes. In this issue of Immunological Reviews, contributors describe work that has built on this classic literature to further define the biochemical, transcriptional, and epigenetic molecular mechanisms that mediate central and peripheral T and B cell tolerance (Kurosaki, Pelanda, Zikherman, Tsubata, Getahun). Selected reviews highlight how abnormalities in signals received by and propagated in lymphocytes can breach these tolerance checkpoints to produce autoimmune disease in mice and humans (Weiss, Kuchroo, Deenick, Acharya and Jackson). These include discussion of rare germline gain-of-function mutations in signaling molecules identified in familial autoimmune syndromes (Weiss, Deenick). Two other reviews describe the discovery of new pathogenic T and B cell populations in mouse models and human autoimmunity (Tph—Rao, ABCs—Pernis). Finally, Reed describes an exciting new role for somatic mutations in the evolution of pathogenic B cell clones in human autoimmunity. Together, the issue highlights the leading edges of our understanding of tolerance and autoimmunity both in model systems and patients. We hope that this will illuminate opportunities for further investigation of basic tolerance mechanisms and novel approaches to therapeutic intervention and reveal unexpected features of human autoimmune pathogenesis. Below, we provide an overview of the areas covered in this issue. During T cell development, recognition of either ubiquitous or tissue-specific antigens presented by specialized cell populations in the thymus can either drive self-reactive thymocytes into clonal deletion or divert them into the regulatory T cell (Treg) lineage. Zikherman and colleagues review work from many groups over three decades describing an essential role for the NR4A family of orphan nuclear receptors in these processes.1 The NR4A family members are immediate early genes that are rapidly induced by antigen receptor stimulation and their expression scales with both the intensity and duration of signaling. As a result, they are highly upregulated by strong TCR signaling in developing thymocytes. The NR4As play essential but redundant roles in both negative selection and in the generation and maintenance of Treg cells, linking TCR signal strength to both of these outcomes. These nuclear receptors reinforce the Treg transcriptional program, while suppressing inappropriate effector cytokine production. Uncoupling these roles from negative selection during thymic development has been technically challenging, and dissecting the mechanism(s) by which the NR4A family drives negative selection (both transcriptional and non-transcriptional) remains an important goal. How the NR4As toggle between promoting death and Treg fate is another key area for future investigation. Weiss and colleagues synthesize years of studies that have characterized loss-of-function and gain-of-function human genetic variants in ZAP70, along with genetically engineered and spontaneous ZAP70 mutations in mice.2 They demonstrate how modulation of TCR signal strength by this T cell-specific non-receptor tyrosine kinase impacts T cell repertoire selection during thymic development. They go on to show how such a skewed repertoire interacts with perturbed engagement of TCR-dependent peripheral tolerance mechanisms to drive autoimmunity both in mouse models of disease and patients (see below for further discussion). In now classic studies, it was shown that signal 1 (peptide-MHC) in the absence of signal 2 (co-stimulation via CD80-86/CD28 interactions) induces a state of functional unresponsiveness in mature T cells. Anjana Rao and colleagues subsequently showed that either ionomycin stimulation (which selectively drives NFAT into the nucleus) or forced expression of a constitutively active NFAT1 construct unable to bind AP-1, induced transcription of a limited set of genes with negative regulatory roles in T cells.3 These include genes that encode inhibitory co-receptors, ubiquitin ligases, and other negative regulators of TCR signaling, as well as a core set of transcription factors (TFs). Zikherman and colleagues review recent work from the labs of Chen Dong and Anjana Rao that reveal a key role for the NR4A family in regulation of the transcriptional and epigenetic landscape of tolerant T cells (both in exhausted and anergic T cells, collectively termed by some as “dysfunctional” T cells).1 These studies show that following chronic Ag stimulation, T cells alter their epigenetic landscape to increase the accessibility of NFAT and NR4A motifs, while NR4A-deficient T cells exhibit increased accessibility of NFAT-AP1 and NF-κB sites. Moreover, NR4A overexpression induces a tolerogenic transcriptional program, while NR4A-deficient T cells inappropriately express effector cytokines and evade tolerance and exhaustion. Weiss and colleagues describe a novel mouse model of T cell anergy driven by a gain-of-function mutation of ZAP70 (W131A) which triggers functional and transcriptional features of anergy in vivo in the absence of any model antigen.2 In these mutant animals, anergy is acquired after thymic development is complete and is imposed and maintained in part by the ubiquitin ligase C-CBL, while the ubiquitin ligase GRAIL and the inhibitory receptor PD-1 appear to be dispensable for functional anergy in this model. Conversely, a series of mice harboring a range of hypofunctional ZAP70 alleles not only exhibit a self-reactive repertoire owing to evasion of central tolerance mechanisms but also manifest defective engagement of peripheral tolerance mechanisms, including the anergy transcriptional program. These observations illustrate how T cells with severely impaired antigen receptor signal transduction can nevertheless differentiate into pathogenic effectors that drive in some cases, only autoantibody production, and—in the case of the SKG mouse model—a destructive inflammatory arthritis that recapitulates many features of human rheumatoid arthritis (RA). The waves of expression of activating and inhibitory molecules induced by normal T cell activation control the amplitude and duration of protective T cell responses. This so-called “tidal model of co-signaling” is described by Kuchroo and Krovi.4 Such activating and inhibitory co-stimulatory molecules include well-recognized players such as CD28 and CTLA-4, as well as PD-1, LAG-3, TIM-3, and other so-called checkpoint factors that are now targeted clinically via “checkpoint blockade” to unleash anti-tumor responses. Kuchroo and Krovi present a rigorous and systematic exploration of the roles played by these factors (so-called signal 2) in concert with TCR-peptide-MHC interactions (signal 1), and inflammatory cytokines (signal 3) to break tolerance and drive well-defined autoimmune diseases, such as multiple sclerosis (MS), type 1 diabetes (T1D), and inflammatory bowel disease/colitis (IBD), in both mouse models and patients. This extensive literature helps establish genetic risk factors for disease (most notably of course MHCII/HLA) and lays the groundwork for therapeutics in use and in development. By contrast to T cells, B cells employ a unique strategy—receptor editing—to salvage immature self-reactive B cells during their development in the bone marrow; by replacing light chains via additional rounds of recombination, strongly autoreactive B cells can be redeemed if the edited receptor sufficiently decreases self-reactivity. In their review, Pelanda and colleagues describe prior and recent work uncovering and testing pathways that operate to enforce or break tolerance in immature B cells.5 They describe how enforcing expression of an active form of PI3Kα or deleting the PI3K inhibitor PTEN results in the abrogation of receptor editing and the bone marrow egress and differentiation of high avidity autoreactive B cells. This establishes the fundamental importance of regulating the PI3K pathway during central B cell selection; in normal circumstances, PI3K activation terminates further VDJ recombination and promotes bone marrow egress and cell differentiation only in those immature B cells that display minimal or no self-reactivity. Conversely, BCR downregulation on self-reactive immature B cells short-circuits this process, leading to halted maturation and retention of self-reactive clones in the bone marrow. Pelanda and colleagues go on to describe recent studies with human immune system mice that tested pathways involved in the bone marrow retention or egress of human immature B cells. These studies demonstrate that the CXCR4 chemokine receptor, which is downregulated in non-autoreactive immature B cells following the activation of PI3K and upregulated in autoreactive cells following Ag binding, enforces the retention of autoreactive immature B cells within the bone marrow tissue. By contrast to CXCR4, receptors for the lipid S1P do not play a role in the bone marrow egress of immature B cells following the decrease of CXCR4 expression or activity. In their review, Bier and Deenick describe studies of both patients and mice harboring germline gain-of-function mutations in the PI3KCD gene.6 Defects in peripheral B cell tolerance are evident in this setting (described further below). However, defects in central B cell tolerance have only been investigated in a mouse model in which low-affinity self-reactive B cells leave the bone marrow in slightly lower numbers to become anergic in the periphery; in this model, expression of the gain-of-function PIK3CD mutation did not alter the proportion of immature B cells reaching the spleen. Further studies are needed to determine whether active PI3Kδ is sufficient or not to break central B cell tolerance in patients. Self-reactivity is, however, not fully purged by central tolerance in the bone marrow. About half of self-reactive bone marrow immature B cells exhibit a degree of self-reactivity below the threshold for central tolerance which, in combination with sufficient surface BCR and tonic BCR signaling, license them for bone marrow egress and cell differentiation. Of note, many patients with autoimmune disorders export from the bone marrow much larger numbers of autoreactive B cells.7 While some of these self-reactive B cells are generally arrested at the transitional cell stage, some are even able to enter the mature naïve compartment. Indeed, it is now well-appreciated that modest self-reactivity is a normal feature of mature naïve B cells even in healthy patients and mice. The retention and even selection of such specificities are thought to be important in order to fill holes in the repertoire that could be otherwise exploited by pathogens. In essence, self-reactive B cells are argued to harbor a reservoir of potentially protective BCR specificities available to be engaged in response to specific infectious agents. To meet this need, self-reactive B cells must be held in check to limit inappropriate responses to self-antigens, but the tolerance mechanism employed must be reversible, at least within a reasonable time. B cell anergy is the term coined to describe such non-deletional clonal tolerance and encompasses a range of inter-related mechanisms that both limit B cell responses to self-antigens and render them highly dependent upon a licensing second signal. As seen with receptor editing in the bone marrow, the absence of a “rescue signal” is followed by deletion; the half-life of some BCR Tg anergic B cell models can be as short as 1-3 days in competition with normal repertoire (in contrast to much longer lifespan of bulk repertoire), but the survival of anergic B cells can also be longer, depending on the degree of self-reactivity. In this context, given that NR4A gene expression scales with degree of self-reactivity, the NR4As (as discussed further below) could serve to link the degree of self-antigen recognition to half-life in vivo. Kurosaki and colleagues give an overview of historical studies of anergy utilizing BCR Tg models, as well as novel mechanistic insights.8 They describe key features of anergy originally revealed through the study of the IgHEL/sHEL model pioneered by Goodnow and colleagues,9 including IgM (but not IgD!) downregulation, follicular exclusion, shortened half-life, and impaired signal transduction. Engagement of inhibitory signaling circuits mediated by ITIM-containing inhibitory receptors, as well as constraint of second messenger PIP3 are highlighted as well (discussed further below). By contrast to T cell anergy, roles for transcriptional changes in B cell anergy are less well-understood. Kurosaki and colleagues review recent transcriptional and epigenetic mechanisms that help to enforce anergy. Among these is a role for NR4As in limiting the survival of self-reactive B cells, as also reviewed by Zikherman and colleagues.1 Other transcriptional roles for the NR4As in self-reactive B cells remain to be defined, and further exploration of both the transcriptional hierarchy at play in anergic B cells as well as epigenetic re-programming in these cells—while likely to vary across models—remains an exciting area for further exploration. Among epigenetic mechanisms regulating anergy, Kurosaki and colleagues summarize their findings related to the demethylases Tet2/Tet3. When these demethylases are deleted, anergic B cells upregulate CD86 and become competent to recruit T cell help and produce autoantibodies, although hyperactivation of T cells in Tet2/Tet3 double-deficient mice also contributes to autoimmunity. The critical role for an inhibitory circuit operating in B cells was dramatically revealed by development of lupus-like disease in mice harboring genetic lesions affecting this pathway. These include mice deficient for the Src-family kinase (SFK) Lyn, which not only mediates BCR signaling by phosphorylating ITAM tyrosines (along with other SFKs) but also plays an essential non-redundant role in engaging ITIM-containing inhibitory receptors like CD22, CD72, and FcγRIIb. These ITIM co-receptors in turn recruit the tyrosine phosphatase SHP-1 or the lipid phosphatase SHIP1. In this issue, Getahun reviews how this inhibitory pathway operates and discusses work from his group and others elucidating which arms of the pathway are required to maintain anergy in mature B cells.10 The PTPase SHP-1 dephosphorylates key signaling molecules that operate proximally in the BCR signal transduction cascade. By contrast, SHIP1 targets membrane-associated PIP3 which is a key second messenger that facilitates assembly of a B cell signalosome at the membrane and is essential for signal amplification. Moreover, the anergic state is characterized by diminished levels of PIP3 in response to antigen stimulation, which is likely related to the elevated levels of both SHIP1 and PTEN observed in anergic B cells both in mice and humans. Excessive PIP3 supply can breach anergy (perhaps contributing to disease in patients with PIK3CD gain-of-function mutations).6 Lyn-dependent engagement of ITIM-PTPase inhibitory circuits is clearly important to regulate B cell responses to self-antigens. Interestingly, individual inhibitory co-receptors, by virtue of their binding specificities, confer tolerance to specific classes of autoantigens. CD22 is generally thought to suppress immune responses to self-sialic acids. Tsubata discusses two additional examples, CD72 and Siglec10/SiglecG (human/mouse).11 CD72 binds Sm/RNP (RNA-associated) autoantigens and, when co-ligated with the BCR, serves to suppress B cell responses directed against these antigens; in the absence of CD72, such dangerous nucleic-acid (NA) containing autoantigens can be delivered to the NA-sensor TLR7 in endosomes to produce autoimmunity (see additional discussion below). Indeed, deletion of CD72 in mice results in lupus-like disease characterized by anti-RNA antibodies. Siglec10/G recognizes various gangliosides and is thought to specifically suppress B cell responses to these T-independent antigens. Moreover, a rare genetic variant in humans encodes a hypomorphic allele of Siglec10 that predisposes to the development of Guillain-Barre syndrome and will be discussed further below when we address the genetic architecture of human autoimmunity. B cells (similar to T cells) rely on receiving not only just antigen receptor stimulation (signal 1), but also a licensing second signal that is classically delivered in the form of T cell help. However, other danger signals delivered by TLR ligands can substitute for T cell help. Therefore, inappropriate provision of such a second signal represents a key mechanism to breach B cell tolerance independent of cognate T cell help. B cells have evolved a unique viral-sensing mechanism that can be coopted to break tolerance and produce nucleic acid reactive autoantibodies that are the central feature of SLE. Virus-specific BCRs can deliver viral particles to endosomes that harbor NA-sensing TLRs which in turn deliver a second signal that synergizes with BCR stimulation to license humoral immune responses. Because NA-sensing endosomal TLRs (particularly TLRs 7/8 and 9) are capable of driving T-independent B cell proliferation and differentiation, this virus-sensing pathway represents a major vulnerability in B cell tolerance. Indeed, classic work from Ann Marshak-Rothstein and from Silvia Bolland showed how this pathway can be coopted to produce DNA-reactive and RNA-reactive autoantibodies in mouse models, and likely account for the pathognomonic anti-nuclear antibodies (ANAs) that characterize SLE.12, 13 Therefore, elaborate and layered mechanisms have evolved that limit TLR7/9-mediated B cell responses to self-NA. In addition to suppression of BCR signaling by CD72 co-engagement as discussed by Tsubata, Acharya and Jackson discuss other B cell-intrinsic mechanisms to suppress inappropriate TLR7/9 responses to RNA and DNA-containing autoantigens.14 These include nucleases that degrade self-NA, tightly regulated subcellular endosomal localization of NA-TLRs, and endolysosomal trafficking of these NA-TLRs via non-canonical autophagy flux that terminates their signaling. Defects in any of these pathways are sufficient to unleash ANAs and lupus-like disease. Although the requirement for a second signal for B cell responses has been long appreciated, as has the molecular basis of such signals, it has been less clear what factors constrain B cells that receive signal 1 in the absence of signal 2. Zikherman and colleagues recently described a role for the NR4A family that helps to enforce B cell dependence upon second signals, thus limiting inappropriate responses to self-antigens.1 They uncovered a transcriptional negative feedback loop triggered by BCR stimulation and mediated by the NR4A family that restrains upregulation of a small subset of key primary response genes, including Batf and Myc. Provision of a second signal in excess completely bypasses this negative feedback loop. Unexpectedly, under conditions of limiting T cell help, the NR4As restrict the response of immunodominant clones by restraining expression of multiple targets, including Cd86, that serve to engage T cell help. Exploring this pathway in human B cells and disease will be of great interest in future work. Pernis and colleagues describe the discovery of so-called age-associated B cells or ABCs, initially in aged mice, then in autoimmune mouse models, and more recently in the context of certain infections and humans. These B cells are characterized by expression of Tbet transcription factor as well as CD11c surface expression and have been shown to represent an atypical memory B cell population generated in response to a combination of signals, including IFNγ, IL-21, BCR, and TLR stimuli. ABCs are of great significance in a range of contexts because they are thought to harbor pathogenic self-reactive BCRs as well as anti-viral BCRs, and their expansion, reactivation, and differentiation may make an important contribution to autoimmunity and anti-viral responses. Recent studies suggest that ABCs can be generated via both T-dependent and T-independent pathways. Pernis and colleagues systematically review the transcriptional machinery of ABCs and present evidence indicating that the family of SWEF proteins operate in both B cells and T cells to regulate the function of ABCs.15 By systematically reviewing the basic biology of this increasingly appreciated B cell population, Pernis and colleagues lay the groundwork for future studies to understand complex interactions that generate these cells, and whether they are viable targets for the treatment of human SLE. The discovery and characterization of Tph cells—reviewed in this issue by Marks and Rao—represents an extremely elegant example of basic discovery that starts with patients.16 Rao and colleagues initially identified a population of T cells with high PD-1 expression but lacking CXCR5 that is remarkably and uniquely enriched in the joints of patients with seropositive (but not seronegative) rheumatoid arthritis (RA). They went on to show that these cells, also detectable in peripheral blood, express a transcriptional program and harbor functional features of B cell helper cells with the capacity to secrete IL-21 and promote B cell differentiation; they are therefore much like Tfh cells but are localized outside of secondary lymphoid organ germinal centers and instead home to peripheral sites. Since their initial discovery, Tph-like cells have been described in other human patient populations, in multiple organs, and in certain mouse models. It is highly likely that Tph cells contribute to disease in seropositive RA patients and further efforts to understand the origin, activity, and maintenance of these cells, including how they interact reciprocally with B cells, and how to target them selectively, will be of both basic and translational importance. Even as human genome-wide association studies (GWAS) sought to uncover the genetic architecture of common polygenic autoimmune diseases (revealing associations with variants impacting type I IFN and TLR7 signaling pathways in SLE, for example), it was becoming apparent that some heritability was not fully accounted for by common SNPs (>5% prevalence in the population). Indeed, rare variants make an important contribution to this missing heritability. These are particularly challenging to identify in part because such variants tend to vary significantly across human populations. One example described by Tsubata and alluded to earlier is a loss-of-function variant in Siglec10 in which a single mis-sense mutation results in loss of binding to ganglioside ligands. A small genetic case-control study suggests that this variant may confer a substantial risk for development of Guillain-Barre syndrome (GBS), which is classically triggered by infection (eg, Campylobacter) but also exhibits host genetic contribution. This finding suggests that by binding gangliosides, Siglec10 suppresses B cell responses to these self-antigens and prevents development of anti-ganglioside antibodies (found in GBS patients) that can in turn interact with neurons. Other examples of rare variants that confer high risk for disease that are not covered in this volume of Immunological Reviews include mutations in the nuclease Trex1 that exposes cytosolic NA-sensors to retroelements and drives cGAS-STING activation and type I interferon release, and haploinsufficiency for TNFAIP3 which encodes A20, a vital negative regulator of NF-κB signaling.17, 18 Genetic variants with high relative risk for disease really fall on a spectrum with classical Mendelian diseases of immune dysfunction. Indeed, familial Mendelian diseases characterized by both immunodeficiency and autoimmunity are becoming increasingly recognized, in some cases rare, and in others spanning multiple families with tens and even hundreds of affected patients. These disorders of immune dysregulation have served to shed light on basic human immunology and also on the mechanisms at play in complex and much more common polygenic disorders. An example is described by Weiss and colleagues of a unique single family with early-onset multiple autoimmune disorders including pemphigus, glomerulonephritis, and colitis.2 This family was found to have a weak gain-of-function mutation in ZAP70 thought to alter T cell sensitivity to self-antigens and resulting in increased T cell help for B cells. Although discovered less than 10 years ago, one of the most well-studied Mendelian diseases of immune dysregulation is activated PI3K delta syndrome (APDS1), which is caused by gain-of-function mutations in the catalytic subunit of PI3K encoded by PIK3CD and that now includes more than 200 affected individuals worldwide. Bier and Deenick describe patients with this syndrome, and findings from studies that span basic mouse models and translational investigations. This represents a case study of dysregulation of one signaling pathway that breaches tolerance at multiple checkpoints and with likely contributions from multiple cell types, including B cells and T cells.6 Finally, Joanne Reed describes an exciting new twist on germline genetic variants.19 She reviews what is known about the origin of the unique specificities of autoantibodies found in human rheumatic diseases such as lupus, including mechanisms by which self-reactive B cells are normally “redeemed” in the germinal center, and conversely how—in some cases—they emerge de novo from the germinal center resulting in hard to eradicate long-lived plasma cells that secrete somatically mutated high affinity autoantibodies for years and sometimes for a lifetime. Of particular interest is the unexpected recent discovery by Reed and colleagues that some pathogenic plasma cell clones in the context of Sjogren's-associated cryoglobulinemia, can evolve over years by acquiring a somatic lymphoma driver mutation that affords the clone a growth advantage. Not only does this set the stage for evolution to lymphoma, but also it facilitates the development of antibodies with unique properties (in this case capacity to form a cryoprecipitate and cause associated vasculitis). It remains to be determined whether this is the tip of the proverbial iceberg in evolution of human autoantibodies, or rather a special case relating to this specific syndrome. We anticipate that the systematic search for these types of somatic mutations impacting B cells will continue to capitalize on cutting-edge technologies to capture, sequence, and clone antigen-specific human B cells and BCRs. Discovery can start in mice with a model system or with the unique features of human autoimmune disease as epitomized by the range of reviews covered in this issue. But their capacity to converge and cross boundaries from basic to translational is the hallmark of great immunology. The reviews in this collection highlight new advances in understanding a classic problem—perhaps the essential challenge faced by the adaptive immune system—tolerance and autoimmunity. In doing so, they illuminate the boundaries of current knowledge and the opportunity for future discovery and translation to benefit human health (Figure 1). This article was supported by NIAID R01 AI124474, NIAID R01 AI152535, and NIAID R21 AI156232 to RP, and NIAID R01 AI148487, NIAID R01 AI155653, and Rheumatology Research Foundation grant to JZ. JZ is a scientific advisor for Walking Fish Therapeutics and for Cimeio.

Reversing the immunosuppressive nature of the tumor microenvironment is critical for the successful treatment of cancers with immunotherapy drugs. Murine cancer models are extremely limited in their diversity and suffer from poor translation to the clinic. To serve as a more physiological preclinical model for immunotherapy studies, this protocol has been developed to evaluate the treatment of human tumors in a mouse reconstituted with a human immune system. This unique protocol demonstrates the development of human immune system (HIS, "humanized") mice, followed by implantation of a human tumor, either a cell-line derived xenograft (CDX) or a patient derived xenograft (PDX). HIS mice are generated by injecting CD34+ human hematopoietic stem cells isolated from umbilical cord blood into neonatal BRGS (BALB/c Rag2-/- IL2RγC-/- NODSIRPα) highly immunodeficient mice that are also capable of accepting a xenogeneic tumor. The importance of the kinetics and characteristics of the human immune system development and tumor implantation is emphasized. Finally, an in-depth evaluation of the tumor microenvironment using flow cytometry is described. In numerous studies using this protocol, it was found that the tumor microenvironment of individual tumors is recapitulated in HIS-PDX mice; "hot" tumors exhibit large immune infiltration while "cold" tumors do not. This model serves as a testing ground for combination immunotherapies for a wide range of human tumors and represents an important tool in the quest for personalized medicine.

Severe SARS-CoV-2 infection is associated with strong inflammation and autoantibody production against diverse self-antigens, suggesting a system-wide defect in B cell tolerance. BND cells are a B cell subset in healthy individuals harboring autoreactive but anergic B lymphocytes. In vitro evidence suggests inflammatory stimuli can breach peripheral B cell tolerance in this subset. We asked whether SARS-CoV-2–associated inflammation impairs BND cell peripheral tolerance. To address this, PBMCs and plasma were collected from healthy controls, individuals immunized against SARS-CoV-2, or subjects with convalescent or severe SARS-CoV-2 infection. We demonstrate that BND cells from severely infected individuals are significantly activated, display reduced inhibitory receptor expression, and restored BCR signaling, indicative of a breach in anergy during viral infection, supported by increased levels of autoreactive antibodies. The phenotypic and functional BND cell alterations significantly correlate with increased inflammation in severe SARS-CoV-2 infection. Thus, autoreactive BND cells are released from peripheral tolerance with SARS-CoV-2 infection, likely as a consequence of robust systemic inflammation.

Double negative (DN) B cells (CD27-IgD-) comprise a heterogenous population of DN1, DN2, and the recently described DN3 and DN4 subsets. In autoimmune disease, DN2 cells are reported to be precursors to autoreactive antibody secreting cells and expansion of DN2 cells is linked to elevated interferon levels. Severe SARS-CoV-2 infection is characterized by elevated systemic levels of pro-inflammatory cytokines and serum autoantibodies and expansion of the DN2 subset in severe SARS-CoV-2 infection has been reported. However, the activation status, functional capacity and contribution to virally-induced autoantibody production by DN subsets is not established. Here, we validate the finding that severe SARS-CoV-2 infection is associated with a reduction in the frequency of DN1 cells coinciding with an increase in the frequency of DN2 and DN3 cells. We further demonstrate that with severe viral infection DN subsets are at a heightened level of activation, display changes in immunoglobulin class isotype frequency and have functional BCR signaling. Increases in overall systemic inflammation (CRP), as well as specific pro-inflammatory cytokines (TNFα, IL-6, IFNγ, IL-1β), significantly correlate with the skewing of DN1, DN2 and DN3 subsets during severe SARS-CoV-2 infection. Importantly, the reduction in DN1 cell frequency and expansion of the DN3 population during severe infection significantly correlates with increased levels of serum autoantibodies. Thus, systemic inflammation during SARS-CoV-2 infection drives changes in Double Negative subset frequency, likely impacting their contribution to generation of autoreactive antibodies.

Immune checkpoint inhibitors have been found to be effective in metastatic MSI-high colorectal cancers (CRC), however, have no efficacy in microsatellite stable (MSS) cancers, which comprise the majority of mCRC cases. Cabozantinib is a small molecule multi-tyrosine kinase inhibitor that is FDA approved in advanced renal cell, medullary thyroid, and hepatocellular carcinoma. Using Human Immune System (HIS) mice, we tested the ability of cabozantinib to prime MSS-CRC tumors to enhance the potency of immune checkpoint inhibitor nivolumab. In four independent experiments, we implanted distinct MSS-CRC patient-derived xenografts (PDXs) into the flanks of humanized BALB/c-Rag2nullIl2rγnullSirpαNOD (BRGS) mice that had been engrafted with human hematopoietic stem cells at birth. For each PDX, HIS-mice cohorts were treated with vehicle, nivolumab, cabozantinib, or the combination. In three out of the four models, the combination had a lower tumor growth rate compared to vehicle or nivolumab-treated groups. Furthermore, interrogation of the HIS in immune organs and tumors by flow cytometry revealed increased Granzyme B+, TNFα+ and IFNγ+ CD4+ T cells among the human tumor infiltrating leukocytes (TIL) that correlated with reduced tumor growth in the combination-treated HIS-mice. Notably, slower growth correlated with increased expression of the CD4+ T cell ligand, HLA-DR, on the tumor cells themselves. Finally, the cabozantinib/nivolumab combination was tested in comparison to cobimetinib/atezolizumab. Although both combinations showed tumor growth inhibition, cabozantinib/nivolumab had enhanced cytotoxic IFNγ and TNFα+ T cells. This pre-clinical in vivo data warrants testing the combination in clinical trials for patients with MSS-CRC.

About 5% of B cells in healthy mice and humans are allelically or isotypically included and hence co-express two different antibodies. In mice, dual antibody B cells (B 2R ) expand with systemic autoimmunity, co-express autoreactive and non-autoreactive antibodies, and participate in immune responses, but this phenomenon is strain dependent. This study was developed with two goals: 1) to establish the contribution of TLR and IFN receptor signaling to the development of germinal center B cells that express two antibodies in MRL/ lpr mice; and 2) to determine whether B 2R B cells are increased and particularly activated in a subset of adult patients diagnosed with systemic lupus erythematosus (SLE). Results from the MRL/ lpr studies indicate that the enhanced differentiation of dual-κ B cells into germinal center B cells is due to a heightened response to TLR7 and TLR9 signaling, further fueled by an increased response to type II IFN. To understand the clinical and translational implications of our observations in mouse B 2R B cells, cohorts of SLE patients and healthy controls were recruited and evaluated for expression of dual BCRs. Results from flow cytometry and microscopy revealed supraphysiological frequencies of κ + λ + B 2R cells in one fourth of the SLE patients. Abnormal numbers of κ + λ + B cells correlated with higher frequencies of activated naïve B cells and age-associated B cells, and a lower proportion of “B cells that are naïve IgD + ” (BND). However, results from single cell V(D)J sequencing demonstrated that these high κ + λ + SLE patients harbored normal frequencies of κ + λ + and other B 2R B cells. and we further show that their B cells were instead decorated by κ and λ VH4-34 autoantibodies. Thus, our findings indicate that elevated flow cytometric detection of isotypically-included B cells can identify patients with high titers of B cell-reactive VH4-34 autoantibodies and abnormal distribution of B cell subsets relevant to autoimmunity.