T cell-B cell interactions in human autoimmune diseases

Abstract

Activation of autoreactive B cells and production of specific autoantibodies are hallmark features of many autoimmune diseases. B cell differentiation into antibody-secreting cells typically requires help from cognate T cells, which provide both cytokines and cell surface signals in an intricate intercellular interaction. A range of T cells can provide this help to B cells, including T follicular helper cells in follicles of secondary lymphoid organs, as well as T peripheral helper cells, which accumulate within inflamed target tissues in autoimmune diseases. Here we discuss recent observations about the phenotypes of B cell-helper T cells that accumulate in inflamed tissues and in circulation of patients with autoimmune diseases, the correlations between B cell-helper T cells and B cells in these tissues, and key mediators of productive T cell-B cell interactions, with a focus on mediators that are being targeted therapeutically. Understanding the scope of B cell-helper T cells and their functions will improve our ability to quantify and track pathologic T cell-B cell interactions in human autoimmune diseases and may highlight critical mediators that can be targeted to suppress these interactions therapeutically.

Introduction

Pathologic T cell-B cell interactions are a core feature of many autoimmune diseases. These interactions drive the production of class-switched, somatically hypermutated autoantibodies, which are specific to multiple autoimmune diseases including rheumatoid arthritis (RA), systemic lupus erythematosus (SLE), type I diabetes, and ANCA vasculitis. Furthermore, these autoantibodies can be detected years before the disease becomes clinically evident[1,2], suggesting breaks in tolerance for both B and T cells occurs long before clinical onset. Understanding these interactions is crucial for uncovering disease mechanisms and identifying early diagnostic markers.

T cells that help B cells

T cells that help B cells increase in number in many autoimmune diseases, especially those characterized by autoantibody production. T follicular helper (Tfh) cells, the prototype B cell-helper T cell population, are essential for robust generation of a high-affinity antibody response[3], and these cells expand in secondary lymphoid organs in many murine models of autoimmune disease. While lymph nodes and spleen are rarely sampled clinically in patients with autoimmune diseases, studies of blood and target tissues have demonstrated expansion of Tfh cells at both sites. For example, Tfh cells are increased in the circulation and in kidney of patients with SLE and lupus nephritis[4–6], and the lesions of patients with IgG4-related disease show prominent accumulation of prototypical CXCR5⁺ BCL6^hi Tfh cells[7].

However, unbiased analyses of T cells from blood or tissue of patients with the autoimmune diseases have highlighted B cell-helper T cells that differ from prototypical Tfh cells[8]. T peripheral helper (Tph) cells were first discovered in RA synovial tissue, where they share some characteristics with Tfh cells but also exhibit distinct features [9]. Tph cells resemble Tfh cells in their B cell helper function, expressing PD-1, ICOS, CD40 ligand, IL-21, and CXCL13. However, unlike Tfh cells, Tph cells have lower BCL6 expression, lack CXCR5, and instead express CCR2, CCR5, and CXCR6, enabling their migration into inflamed tissues. Phenotypically similar cells have been identified in the target tissues of patients with multiple other diseases, including celiac disease[10], ulcerative colitis[11], sarcoidosis[12], post-infectious Lyme arthritis[13], graft-versus-host disease[14], Sjogren’s disease[15], bullous pemphigoid[16], and also in tumors[17] (Figure 1). In addition, expansion of circulating Tph cells has been reported in over a dozen autoimmune diseases, with particularly prominent expansion in autoantibody-associated systemic rheumatic diseases including SLE and ANCA vasculitis[18,19]. While Tph cells in RA and SLE patients have a strong Th1 bias, Tph cells with a Th2 bias have now been reported in bullous pemphigoid[16] and in allergic bronchopulmonary aspergillus[20].

Figure 1. Tph cells in auto-immune disease.

Tph and Tph-like cells that have been observed in various disease contexts, particularly autoimmune disease. These cells typically share features with Tfh cells yet lack the expression of CXCR5.

Recognition of Tph cells has helped to underscore a broader theme of a phenotypic variation in B cell-helper T cells. Though the nomenclature varies, a range of ‘Tfh-like’ T cell populations has been recognized in numerous contexts, generally unified by displaying features and functions associated with B cell help despite a low or absent expression of CXCR5. In murine models, CXCR4⁺ extrafollicular helper T cells provide help to B cells in extrafollicular foci via IL-21, analogous to Tfh cells in follicles[21]. Resident helper T cells, which display B cell-helper functions paired with CXCR6 expression, accumulate in lungs of mice after viral infection or chronic inflammation[22–24]. Recent studies in diverse murine models have further illustrated the presence of Tph cells in different contexts. Both Tph and Tfh cells infiltrate peripheral nerves in a murine model of autoimmune neuropathy and contribute to neuronal injury via IL-21 production[25]. Similarly, Tph cells accumulate in a murine model of central nervous system inflammation and promote B cell activation and formation of meningeal lymphoid aggregates[26]. A similar appearing Tph cell population accumulates in kidney after ischemia-reperfusion injury and contributes to accumulation of B cells and kidney lymphoid aggregates using a CD153-CD30 interaction with B cells[27]. Taken together, it appears that B cell-helper function, including production of IL-21 and CD40L, as well as CXCL13 production in primates, is separable from CXCR5 expression and follicular homing, such that a range of T cell populations exist that display B cell-helper function coupled with different migratory programs, including migratory programs that target inflamed tissues, barriers, and mucosal sites.

New humanized mouse models are advancing the ability to study human Tph cell differentiation, function, and dynamics. Human immune system mice, in which human T cells mature in an immunodeficient mouse, accumulate Tph and Tfh cells over time, which can promote B cell differentiation and antibody production, and also induce autoimmune tissue injury in an IL-21 dependent (though B cell-independent) manner[28]. Promising approaches such as these can be used to model pathology induced by ex vivo cells from patients, as demonstrated by adoptive transfer of T cells from lymph nodes of patients with idiopathic multicentric Castleman disease, which engrafted in immunodeficient mice and led to marked expansion of CXCL13⁺ Tph cells, tissue injury, and death[29]. Antibody neutralization of CXCL13 prolonged recipient survival in this model. In another application, melanoma patients treated with immune checkpoint inhibitor therapy were found to have a prominent expansion of an ICOS⁺ CD4 T cell population with Tph features, and transfer of these ICOS⁺ T cells, but not ICOS⁻ CD4 T cells, into MISTRG immunodeficient mice induced co-transferred B cells to differentiate into plasmablasts and produce human immunoglobulin[30].

T cell-B cell associations in rheumatoid arthritis

Detailed studies of the immune cell infiltrates in synovial tissue from joints of patients with RA have illustrated patterns of association between T cell subsets and B cells in a heterogenous clinical autoimmune disease. Patients with RA vary in the patterns of immune cell infiltration within the synovium, with many patients showing accumulation of both T cells and B cells in synovium, while other patients show infiltrates of T cells and myeloid cells, or primarily myeloid cells without lymphocytes[31]. Correlative studies across patient samples have demonstrated that the presence of B cell-helper T cells in synovium, including both Tph and Tfh cells, is significantly associated with the presence of B cells in the synovium, as well as the presence of lymphoid aggregates[32], consistent with the idea that Tph and Tfh cells promote B cell accumulation and activation in synovium and the formation of lymphoid aggregates. A finer resolution analysis of correlated cell populations across RA synovial tissue samples demonstrated a specific association between Tph cells and age-associated B cells (ABCs), a population of activated B cells associated with autoimmune diseases that is characterized by expression of Tbet and CD11c with low expression of CD21 and CXCR5[32–34]. Similarly, in analyses of T cell and B cell populations in the circulation of patients with SLE, Tph cells, but not Tfh cells, were strongly associated with circulating ABCs, perhaps both reflecting a prominent extrafollicular response[35]. Expansion of ABCs in vitro using B cells with monogenic defects has elegantly demonstrated the significant contribution of BCR and TLR signaling to ABC differentiation, as well as the other signals from, IL-21, IFN-γ and CD40[36]. Both Tfh and Tph cells could provide the source of IL-21, IFN-γ, and CD40L necessary for ABC differentiation. Within the same study ABCs were found to correlate with both circulating Tfh and Tph cells in various disease settings including RA. That both Tph cells and ABCs express low levels of CXCR5 seems unlikely to be a coincidence; it is possible that both of these cell types migrate to inflamed sites following inflammatory signals. It is currently unclear whether Tph cells help drive the development or accumulation of ABCs within a tissue, or whether ABCs induce T cell differentiation into ABCs, or both. In studies of Tph cells from joints of patients with juvenile idiopathic arthritis, synovial fluid Tph cells drove B cells towards an ABC phenotype more efficiently than did tonsil Tfh cells, suggesting a direct cellular interaction[37].

Both Tph and Tfh cells accumulate in RA synovial tissue, and they are strongly co-correlated across patient samples[32]. While the two populations share substantial transcriptomic and epigenetic similarities, they differ in several key features. Their expression of migratory receptors differs starkly, which provides a clear, distinguishing cytometric handle. In addition, synovial Tph cells show a much stronger signature of recent TCR activation than do Tfh cells, as well as more substantial clonal expansion[33,38]. One appealing model is that Tph cells accumulate within an inflamed tissue, migrating to the tissue following signals produced by tissue inflammation, including CCL2 and CCL5, and then activated Tph cells produce CXCL13 to promote recruitment of Tfh cells. The two populations also appear clonally related. Analyses of T cell receptor sequences of Tph and Tfh cells from RA synovium show shared TCRs between CXCR5⁻ Tph cells and CXCR5⁺ Tfh cells, indicating that they can share a common progenitor and perhaps may cross-differentiate, though the direction of differentiation is unclear[33]. Murine studies have also indicated clonal sharing between Tph and Tfh cells, with trajectory analyses suggesting the Tph state as an endpoint of differentiation[25]; thus, it is possible that Tfh cells, either in SLOs or in synovial aggregates, can change their phenotype to convert into Tph cells. Supporting this idea, Tph and Tfh cells from lymph nodes of individuals with HIV share TCR repertoires, and the CXCR5⁻ Tph cells show open regions of chromatin at CXCR5, suggesting that CXCR5 may have been previously expressed in these cells[39]. The signals that may induce Tfh conversion into Tph cells are not known.

Cell surface mediators of T cell-B cell communication

Tph and Tfh cells express a variety of costimulatory and coinhibitory receptors sharing characteristics of an activated T cell phenotype. These receptors enable T cells to interact dynamically with their environment—often in response to antigen exposure—and to modulate their effector responses based on these external cues. Similarly, B cells exhibit distinct patterns of costimulatory and inhibitory receptors that vary with their differentiation and activation states, mirroring the adaptability seen in T cells. The receptor-ligand interactions at the cell surface initiate downstream signaling pathways that lead T and B cells to release soluble mediators, which provide critical guidance for effector cell states (Figure 2). Identifying the essential physical mediators of T-B cell interactions, especially under conditions of dysregulated inflammation, is crucial for the rational design of targeted therapies.

Figure 2. Schematic of Tph and Tfh cell interactions with B cells.

T cell-B cell interactions in the secondary lymphoid organs or in the periphery are regulated by both cell surface (costimulatory and inhibitory receptors) and soluble mediators (chemokines and cytokines). Cognate antigen interaction of Tph or Tfh cells with B cells, is further shaped by the CD28 and TNF superfamilies, whereby the balance of costimulatory (green) and coinhibitory (red) interactions impacts several functions. Downstream of these are soluble mediators of communication, directing B cell differentiation (i.e. ABC differentiation vs plasmablast generation), as well as the migration to, and position of cells within inflamed tissues. The net result in the context of autoimmunity is production of autoantibody (both local and systemic) as well as maintenance/expansion of auto-reactive effector cells. IC, immune complex; PC plasma cell; GC germinal center; Ab, antibody; Ag, antigen.

The significance of these interactions is underscored by findings from numerous genome-wide association studies (GWAS). The latest GWAS catalog (www.ebi.ac.uk/gwas) highlights multiple costimulatory, inhibitory, and cytokine ligand/receptor genes associated with various autoimmune diseases. Many of these genes belong to the B7/CD28 and TNF receptor superfamilies and are prominently expressed on activated T and B cells. While several of these genes are recognized for their key roles in mediating T-B cell interactions, our understanding of how they integrate and coordinate complex immune responses is still expanding. In this context, particular attention will be given to the roles of CD40, OX40, and ICOS, as they represent key modulators of T-B cell communication and immune regulation, with emerging implications for therapeutic intervention in autoimmune diseases. Additionally, we will touch on the broader roles of inhibitory receptors, with a focus on the PD-1/PD-L1/2 axis, and highlight emerging therapeutic strategies targeting these pathways.

The interaction between ICOS and its ligand ICOSL is critical for Tfh cell differentiation and effective germinal center (GC) reactions[40]. Induced by TCR triggering, ICOS signals through a pathway distinct from CD28, playing a central role in T-B cell interactions. Recent studies have revealed that ICOS expression is regulated by a risk allele near the CD28/ICOS/CTLA4 region, which is linked to autoimmune diseases such as RA and type 1 diabetes[41]. ICOS activation enhances T cell activation, promotes glycolysis, and drives T cells towards an IL-21+ and/or CXCL13+ Tph cell phenotype[41,42]. It also plays a role in the accumulation of follicular regulatory T cells in murine models, which regulate humoral immune responses[43]. However, therapeutic blockade of ICOS-ICOSL interactions, sometimes combined with BAFF blockade, has shown mixed results in clinical trials for SLE[44,45].

CD40 on resting B cells and its ligand CD40LG on activated T cells form another key receptor-ligand pair involved in T-B cell interactions with important implications for autoimmunity[46]. Known for driving somatic hypermutation and class switch recombination[47], recent research has deepened our understanding of the upstream processes triggered by CD40 engagement. The strength of the TCR-pMHC interaction influences the duration of CD40-CD40LG interactions, which regulate B cell fate decisions, such as plasma cell differentiation or recycling back into the germinal center[48–50]. Therapeutics targeting CD40-CD40LG interactions have been in development since the 1990s, with several progressing past Phase 2 trials for use in RA, Sjogren’s disease, and SLE[51–54]. In Sjogren’s patients, treatment with the CD40L antagonist dazodalibep reduced disease activity, CXCL13 levels, and rheumatoid factor autoantibody levels [54].

OX40, a receptor highly expressed on B cell-helper T cells, plays a significant role in T-B cell interactions [3,55]. While its role in Tfh cell differentiation remains debated[56,57], its regulation alongside ICOS and its localization at T/B border regions suggest it supports B cell interactions[58]. OX40L expression on activated B cells has been implicated in murine lupus models, where it contributes to plasmablast differentiation, autoantibody production, and kidney immunoglobulin deposition[59]. Whether OX40L expression on B cells influences other autoimmune diseases remains unclear, as OX40-OX40L signaling impacts T cell effector responses across various cell types[60]. Despite uncertainties in its precise role in T-B cell crosstalk, recent OX40-targeting therapies have shown promise in reducing adaptive immune responses in healthy volunteers and have been effective in atopic dermatitis[44,61,62].

Inhibitory receptors play crucial roles in T-B cell interactions, being upregulated on activated T cells and expressed on various effector subsets in autoimmune models[63]. Both Tph and Tfh cells express several inhibitory receptors, but with distinct patterns in RA synovium. Tph cells in RA synovium tend to express higher levels of LAG3 and HAVCR2 than Tfh cells, but lower levels of TIGIT[32]. High PD-1 expression is a key feature of both Tph cells in RA synovium and Tfh cells in germinal centers, often used to identify these cells. Among T cells in RA synovium, high PD-1 expression uniquely marks Tph and Tfh cells, distinguishing them from other activated T cells with intermediate PD-1 levels. The PD-1/PD-L1/PD-L2 axis is well studied in germinal center responses and regulates Tfh cell positioning[64,65]. Beyond the germinal center, this pathway also influences extrafollicular B cell responses and autoantibody production, limiting T cell helper expansion[66]. As such, an agonistic anti-PD-1 antibody reduced disease activity in a Phase 2 clinical trial, supporting its potential in suppressing pathologic immune responses in RA[67].

The ligands for many additional inhibitory receptors are expressed on activated B cells, expanding our understanding of their role in T-B cell communication. For example, the interaction between BTLA on T cells and HVEM on GC B cells regulates B cell help by modulating CD40L expression on T cells[68]. Other inhibitory receptors, such as TIGIT and LAG-3, have distinct roles, though their specific contributions to B cell responses are not fully understood[69]. CD155 expression on B cells may bind to TIGIT on activated Tfh cells to control Tfh expansion[69]. While these inhibitory receptors are characteristic of B cell-helper T cells, their specific roles in T-B cell interactions require further research.

Secreted factors that mediate T cell-B cell interactions

Soluble cytokines exchanged between T cells and B cells are crucial mediators of their interactions. Several T cell-derived cytokines have well-described roles in B cell survival and differentiation, including IL-21, CXCL13, IL-4, and IL-10[70]. Tph and Tfh cells from humans and non-human primates produce high levels of CXCL13, a potent B cell chemoattractant that is the only known ligand for CXCR5. Overexpression of CXCL13 is sufficient to induce B cell recruitment and formation of lymphoid aggregates[71], and blockade of CXCL13 reduces Tph cell-induced disease in humanized mice[29]. Regulation of T cell production of CXCL13 is an area of active research, with type I interferon, aryl hydrocarbon receptor, and SOX4 identified as key regulators[72,73]. Given the strong association between Tph cells and ABCs, it is possible that Th1-skewed Tph cells, as found in patients with RA and SLE, may preferentially skew B cells towards an ABC phenotype, which can be induced by a combination of factors including IL-21, CD40, and IFN-γ[74]. The pivotal role of IL-21 in driving multiple pathogenic B cell populations makes it an attractive target for therapeutic intervention. Several clinical trials have been pursued evaluating IL-21 blockade in autoimmunity diseases, though with few successes described thus far[75,76]. B cell-derived cytokines such as IL-6, TNF, and IL-10 also significantly influence T-B cell interactions after T cell activation[77]. IL-6, which has been linked to Tfh differentiation, was recently shown to inhibit IL-2 signaling, promoting germinal center Tfh formation [63]. As therapeutics blocking IL-6 and TNF are widely used in the treatment of patients with RA, detailed analyses of the effects of these agents on T cell-B cell interactions in treated patients may provide an opportunity better understand the specific roles of these factors in influencing pathologic adaptive immune responses in patients with autoimmune diseases.

Conclusion

A range of T cell populations with the ability to help B cells expand in patients with autoimmune diseases, contributing to pathologic T cell-B cell interactions both within secondary lymphoid organs and in extrafollicular foci and target tissues. Understanding the nuanced roles of costimulatory and coinhibitory receptors in these T-B cell interactions remains a critical area for advancing therapeutic strategies in autoimmune diseases. While the synergistic actions of costimulatory pathways, such as ICOS-ICOSL, CD40-CD40L, and OX40-OX40L, have been extensively studied, the interplay between these and inhibitory receptors like PD-1 and BTLA requires further exploration. Identifying how these pathways integrate to modulate immune responses across specific contexts—such as germinal center reactions, extrafollicular responses, or chronic inflammation—is essential. Moreover, a deeper understanding of the temporal and spatial dynamics of receptor-ligand interactions will allow for the development of more targeted therapies, enabling precise modulation of immune responses tailored to the underlying pathophysiology.