Follicular helper T cells: potential therapeutic targets in rheumatoid arthritis

Abstract

Rheumatoid arthritis (RA) is a chronic autoimmune disease with joint and systemic inflammation that is accompanied by the production of autoantibodies, such as rheumatoid factor and anti-cyclic citrullinated peptide (anti-CCP) antibodies. Follicular helper T (Tfh) cells, which are a subset of CD4⁺ T cells, facilitate germinal center (GC) reactions by providing signals required for high-affinity antibody production and the generation of long-lived antibody-secreting plasma cells. Uncontrolled expansion of Tfh cells is observed in various systemic autoimmune diseases. Particularly, the frequencies of circulating Tfh-like (cTfh-like) cells, their subtypes and synovial-infiltrated T helper cells correlate with disease activity in RA patients. Therefore, reducing autoantibody production and restricting excessive Tfh cell responses are ideal ways to control RA pathogenesis. The present review summarizes current knowledge of the involvement of Tfh cells in RA pathogenesis and highlights the potential of these cells as therapeutic targets.

Introduction

Rheumatoid arthritis (RA) is an autoimmune disease that is characterized by chronic inflammation of the joint capsule and synovial membrane, which results in cartilage injury, bone erosion, joint destruction, and deformity [1]. RA is often accompanied by the involvement of extra-articular organs and serum rheumatoid factor, which ultimately result in the complete loss of normal joint function. Smoking, microbiota, sex, diet, and genetic factors are involved in RA development, but the precise underlying immunopathological mechanisms of RA are not completely defined [2]. Activated T cells, B cells, infiltrating macrophages, and inflammatory cytokines, such as interleukin-6 (IL-6) and tumor necrosis factor (TNF)-α, contribute to the pathogenesis of RA. Innate and adaptive immunity play roles in disease progression, particularly the production of autoantibodies, such as rheumatoid factor and anti-citrullinated protein antibodies (ACPAs) [3].

CD4⁺ T cells play a crucial role in B cell responses. T follicular helper (Tfh) cells were recently described as a distinct subset of CD4⁺ T cells with specialized B cell helper functions. Tfh cells express abundant C-X-chemokine receptor type 5 (CXCR5), which enables homing to B cell follicles. Tfh cells have an activated effector phenotype and express elevated levels of CD40 ligand (CD40L), inducible T cell costimulator (ICOS), programmed death-1 (PD-1), and B and T lymphocyte attenuator (BTLA) [4]. Tfh cells colocalize with antigen-specific B cells within germinal centers (GCs), which are transient structures located within B cell follicles in secondary lymphoid tissues in which somatic hypermutation of immunoglobulin (Ig) variable region genes and selection of high-affinity B cell clones occur. Although Tfh cells are essential for B cells to produce the high-affinity antibodies that play essential roles in protecting the body from infectious diseases, the overproduction of Tfh cells may also be associated with the development of autoimmune and inflammation diseases, such as RA, systemic lupus erythematosus (SLE), and Sjögren’s syndrome (SS) [5–8].

Numerous studies focused on Tfh cells in autoimmune diseases [9, 10], and evidence suggests a critical role of Tfh cells in the pathophysiology of RA because of the large number of Tfh cells in the synovial tissue of RA patients [11], the increased frequency of circulating Tfh-like cells (cTfh-like cells) among peripheral blood mononuclear cells (PBMCs) and increased levels of serum IL-21, IL-6, and IL-1β, which are positively correlated with the 28-joint count disease activity score (DAS28) and the levels of serum anti-cyclic peptide containing citrulline (anti-CCP) antibodies in RA patients [12, 13]. The present review discusses the biological functions of Tfh cells and the roles of these cells in RA to provide new insights into Tfh-targeted therapeutic strategies for the prevention and treatment of RA.

The differentiation of Tfh cells in the GC

Interactions between T cells and B cells are critical for the induction of humoral immune responses that protect against infection. However, these interactions may also be deleterious and drive autoimmunity. Long-lived T cell-dependent humoral immunity is derived from specialized microanatomical structures known as GCs that form in secondary lymphoid organs, such as the spleen and lymph nodes, upon infection or immunization with a T cell-dependent antigen [14]. Tfh cells play vital roles in GC responses. Briefly, naïve CD4⁺ T cells are localized in the T-cell zone and initially primed by dendritic cells (DCs) via major histocompatibility complex (MHC) II-peptide-T cell receptor (TCR) interactions. Once activated, CD4⁺ T cells upregulate many costimulatory molecules, such as CD40L, ICOS, and OX40, which favor crosstalk with DCs. IL-6 (in mice) and IL-12 (in humans) secreted by DCs further induce these activated T cells to become pre-Tfh cells [15, 16]. The upregulation of CXCR5 and downregulation of chemokine receptor 7 (CCR7) attracts pre-Tfh cells to the T–B border via a C–X–C motif chemokine 13 (CXCL13) gradient. Pre-Tfh and B cells stably interact in this region via SLAM-associated protein (SAP)/signaling lymphocyte activation molecule (SLAM) and ICOS/ICOS ligand (ICOSL), which marks the late event in Tfh differentiation and stable Tfh cell generation [4]. Notably, PD-1 is a conventional inhibitory molecule that is highly expressed by Tfh cells. Recent research showed that PD-1 was essential for positioning Tfh cells in the T-B cell border and increasing the stringency of GC affinity selection [17].

Molecules involved in Tfh cell function: conventional and novel

Mature Tfh cells (CXCR5⁺⁺PD–1⁺⁺CCR7⁻) migrate toward GCs, where GC Tfh cells provide help via various mechanisms, such as cell surface CD40L/CD40 interactions or the production of cytokines, such as IL-21 or IL-4, to promote B cell proliferation and somatic hypermutation, which allows B cells to differentiate into high-affinity plasma cells or memory B cells [18]. Recent studies showed that IL-9 also played an important role in GC development because IL-9 drives the development of GC memory B cells. The deletion of IL-9 from GC Tfh cells led to impaired memory B cell formation [19]. High PD-1 expression accompanies the activation of Tfh cells. Although several studies suggested that PD-1 deficiency promoted the expansion of Tfh cells, reduced expression of IL-21 and IL-4 and defects in GC output were also observed, including decreased GC B cell survival and impairment in memory B cell and plasma cell generation [20, 21]. The inhibitory molecule PD-1 may also play an important role in GC responses. Jingwen et al. showed that PD-1 and PD-L1 interactions between Tfh and GC B cells were required to maintain the stringency of affinity-based selection [17]. However, several studies suggested that PD-1/PD-L1 interaction restrained the normal GC response. The role of PD-1 on Tfh cells and the molecular mechanisms for providing help to GC B cells are not clear.

In addition to the costimulatory signals and cytokines driven by Tfh cells, Tfh cells produce high amounts of dopamine, which causes the rapid translocation of intracellular ICOSL to the B cell surface. The ICOSL/ICOS interaction enhances the accumulation of CD40L and chromogranin B granules at the Tfh cell synapse and increases the synapse area. The resulting feed-forward loop allows maximal T–B cell entanglement and the CD40L expression threshold required for GC B cell survival and/or selection, which further increases total GC output [22].

CD40L is critical for the development and maintenance of the GC and the processes associated with B cell affinity maturation, immunoglobulin class switching, and the differentiation of long-lived plasma cells [23–25]. Imaging studies showed that most cognate interactions between effector T cells and antigen-presenting cells (APCs) in vivo are too short for de novo CD40L synthesis [26]. These CD4⁺ helper T cells provide the necessary help for B cells via preformed CD40L (pCD40L). Notably, pCD40L is present in Th1 and Tfh cells, which may function in helper T cell-mediated adaptive immune responses [27]. Gardell and Parker demonstrated that CD40L was transferred from helper T cells to B cells, which allowed sustained CD40 signaling and enabled the survival and proliferation of antigen-presenting B cells in the GC [28]. Based on this phenomenon, two possible mechanisms were proposed for CD40L transfer from T cells to B cells: proteolytic shedding or membrane transfer. CD40L is a homotrimeric type II transmembrane protein that is a member of the TNF superfamily of molecules. CD40L is released from cells as an 18-kDa fragment that has full biological activity, and it is received by B cells [29]. Notably, Gardell and Parker also suggested another model [28] that CD40L may be released in exosomes that carry related effector molecules to exert biological effects via membrane fusion or internalization by target cells [30]. Exosomes are nanosized extracellular vesicles (EVs) (30–150 nm in diameter) that are formed by inward budding of the endosomal membrane within multivesicular bodies. Full-length CD40L was observed in exosomes derived from platelets, mast cells, and lymphoblasts [31–33], and T cells release CD40L-containing exosomes [34]. Mouse experiments showed that the transfer of T cell-derived EVs induced GC reactions and antibody production that depended on a restricted set of exosomal microRNA molecules, including miR-20a-5p, miR-25-3p, and miR-155-3p [35, 36]. Therefore, it is very likely that Tfh cells help B cells via the release of exosomes containing CD40L or other functional molecules (Fig. 1).

Fig. 1.

Molecular mechanisms of germinal center Tfh cells helping B cells (see text for details)

Circulating Tfh-like cells and subsets in RA patients

A small analogous subset of circulating CD4⁺CXCR5⁺ T cells was described recently in human peripheral blood that share functional properties with secondary lymphoid organ (SLO) Tfh cells. Although circulating CD4⁺CXCR5⁺ and CD4⁺CXCR5⁻ T cells express minimal Bcl-6, CD4⁺CXCR5⁺ T cells exhibit a greater capacity for IL-21 production and superior B cell helper capacity, and are known as circulating Tfh-like cells (cTfh-like cells) [37]. The cTfh-like cells highly express CCR7, which allows these cells to migrate to lymph nodes, compared to GC Tfh cells, which downregulate CCR7 expression for B cell follicle entry [38]. Although there are several differences between cTfh-like and GC Tfh cells, a growing number of studies demonstrated a strong correlation between cTfh-like cells and plasmablasts, autoantibodies, and disease activity in several autoimmune diseases, such as myasthenia gravis (MG), systemic sclerosis (SSc), and SLE [39–41].

An excessive generation of CXCR5⁺PD-1⁺ Tfh-like cells was observed in RA patients, and the frequency of these cells correlated with the disease activity score for RA (DAS28) [5]. Our previous study also showed that the frequency of cTfh-like cells was associated with the level of anti-CCP antibodies in RA patients [42]. RA patients with active disease showed increased frequencies of circulating plasmablasts compared to patients in remission [12], which suggested that the excessive generation of circulating Tfh-like cells and plasmablasts plays a role in the pathogenesis of RA. CD4⁺CXCR5⁻PD-1^hi cells were also observed in the peripheral blood of RA patients and termed circulating T peripheral helper (cTph) cells. Similar to the CD4⁺CXCR5⁺PD-1^hi Tfh-like cells, both populations of PD-1^hiCD4⁺ T cells possessed the capacity to help B cells and were increased in the peripheral blood of RA patients. However, cTph cells were present only in patients with active disease, and cTfh-like cells were constitutively elevated. Methotrexate (MTX) treatment reduced cTph cells and RF levels but not cTfh-like cells, which suggests that the increased frequency of cTph cells is associated with the generation of pathogenic autoantibodies and may be a promising marker of RA clinical activity [43].

Based on the differential expression of C–X–C-chemokine receptor 3 (CXCR3) and chemokine receptor 6 (CCR6), three phenotypic and functionally distinct subsets of cTfh-like cells in blood were identified: Tfh-like Th1 (Tfh1), Tfh-like Th2 (Tfh2), and Tfh-like Th17 (Tfh17) cells. The imbalance in these cTfh-like subsets is likely involved in the pathogenesis of some autoimmune diseases [6, 37, 44]. These cells were first identified in blood, and patients with juvenile dermatomyositis, a systemic autoimmune disease, exhibited profoundly skewed levels of Tfh2 and Tfh17 cells. Tfh2 and Tfh17 cells induce naïve B cells to secrete IgG via IL-21 but differentially modulate isotype switching. Tfh2 cells promoted IgG and IgE secretion, and Tfh17 cells promoted IgG and IgA secretion. In contrast, Tfh1 cells lacked the capacity to help B cells. These findings suggested that different Tfh-like cell subsets selectively regulate humoral immune responses [37]. Sun et al. analyzed the PBMCs of RA patients and reported that the frequencies of Tfh2 and Tfh17 cells were significantly higher in the high activity group than the low activity group [45]. However, other studies also showed that only the Tfh2 cell frequency was increased in the RA group, and the Tfh17 cell frequency was unchanged [46]. Although changes in the ratios of Tfh cell subsets in different RA studies are inconsistent (Table 1), evidence suggests that alterations in the (Tfh2 + Tfh17)/Tfh1 ratios are good indicators that reflect the development of RA [37].

Table 1.

cTfh-like cell subsets in RA

Dysfunction of Tfh subsets	Gate in	Correlation	References
Tfh1-Tfh2↑Tfh17↑	CD3+CD4+CXCR5+	Positive correlation: Tfh2/Tfh17—IL-6/IL-17/MMP-3 Negative correlation: Tfh2/Tfh17—TGF-β	[47]
Tfh1-Tfh2↑Tfh17-	CD3+CD4+CXCR5+CD45RA−	–	[46]
Tfh1-Tfh2↑Tfh17-	CD3+CD4+CXCR5+	–	[48]
Tfh17↑	CD4+CXCR5+CD45RA−	Positive correlation: Tfh17—ACPA Negative correlation: Tfh17—St6gal1 expression in plasmablasts	[49]
Tfh17↑	CD4+CXCR5+CD45RO+	Positive correlation: CXCR5+ Th17 cells—baseline disease activity (low CDAI scores)	[50]
Tfh1-Tfh2-Tfh17-	CD3+CD4+CXCR5+	Positive correlation: (Tfh2 + Tfh17)/Tfh1 ratios—DAS28 (Tfh2 + Tfh17)/Tfh1 ratios—Plasmablasts	[12, 37]
Tfh1-Tfh2-Tfh17-	CD4+CXCR5+	–	[51]

Peripheral T helper cells in synovial tissues

Sufficient interactions between antigen-specific T and B cells are essential for the generation of high-affinity antibody-producing plasma cells and long-term memory B cells. This process normally occurs in SLOs. However, in pathological conditions involving chronic inflammation, especially autoimmune diseases, T and B cells are also frequently found in inflamed nonlymphoid tissues, where the anatomical distinction between inflamed peripheral tissues and lymph node follicles begins to blur. Transferring CD4⁺ T cells specific for citrullinated fibrinogen (CF) into mice with collagen-induced arthritis (CIA) enhanced the severity of autoimmune arthritis, which was reflected by an increase in the levels of pathogenic IgG2a against mouse type II collagen and the infiltration of these CD4⁺ T cells into the synovium. CF-specific T cells may have systemic effects on B cell responses and local effects on the inflammatory environment [52]. Notably, immunohistochemistry first showed that CXCR5 was present on B cells and T cells in the synovium of RA patients but not in non-RA patients, which suggested that the noninflammatory receptor CXCR5 played an important role in the pathogenesis of RA [53]. An enrichment of CD4⁺CXCR5⁺ICOS⁺ Tfh cells was observed in synovial tissues of patients with RA, and these cells were absent in osteoarthritis and normal synovial tissues [11, 54].

Studies also revealed a strikingly expanded population of PD-1⁺ CXCR5⁻CD4⁺ T cells in the RA synovium [55–57]. These cells express factors associated with the capacity to help B cells, including IL-21, CXCL13, ICOS, and macrophage activating factor (MAF) and were named T peripheral helper (Tph) cells. Similar to Tfh cells in lymphoid tissues, Tph cells induced plasma cell differentiation in vitro via IL-21 and SLAM family member 5 (SLAMF5). Tfh and Tph cells are present in the same tissues [55] (Table 2). However, the transcriptome of Tph cells is different from conventional Tfh cells. The Bcl6/Blimp-1 ratio of synovial Tph cells is uniquely decreased compared to Tfh cells [55]. Tph cells uniquely express chemokine receptors that allow these cells to directly migrate to inflamed sites, such as chemokine receptor 2 (CCR2), CX3C chemokine receptor 1 (CX3CR1), and chemokine receptor 5 (CCR5), which are not expressed on conventional CXCR5⁺ Tfh cells [58]. This expression may partially explain why these T cells migrate to inflamed tissues and interact with B cells. Further studies are needed to assess the generation of Tph cells and elucidate their precise functions in inflamed tissues to provide evidence for the treatment and study of RA and other autoimmune diseases.

Table 2.

Synovial T helper cells in RA

Tissue	Detection method	Phenotype	Function	References
Synovial fluid	Flow cytometry	CD4+CXCR5+ICOS+	–	[11]
Synovial tissue	Immunofluorescence	CD4+CXCR5+ICOS+	–	[54]
Synovial tissue	Flow cytometry	CD4+CXCR5+PD-1high /CD4+CXCR5−PD-1high	–	[55]
Synovial fluid	Flow cytometry	CD4+CXCR5−PD-1high ICOS+IL-21+BCL-6low Blimp-1highCXCL13+	Induce plasma cell differentiation in vitro through IL-21 secretion and SLAMF5 interaction	[55]
Synovial fluid	Flow cytometry/immunofluorescence	CD3+CXCR5−BCL-6− CXCL13+	CXCL13 production and secretion	[57]
Synovial tissue	Immunofluorescence	CD3+CD4+CXCR5− SOX4+CXCL13+	Correlate with ELS formation	[59]
Synovial tissue	Flow cytometry	CD3+CD4+PD-1highICOS+	–	[60]

Pathogenic role of IL-21 in RA

Investigations of the implication of Tfh cells in disease pathogenesis led to the evaluation of the impact of their signature cytokine, IL-21. IL-21R^−/− C57BL/6 mice were resistant to the development of CIA [61]. IL-21R deficiency in K/BxN mice also completely blocked the spontaneous development of arthritis. Notably, in a K/BxN serum-transferred arthritis model, all C57BL/6 mice developed arthritis of similar severity regardless of the presence or absence of IL-21R. This result indicated that IL-21 was required for arthritis development during the adaptive autoimmune responses instead of the stage of innate effector function [62]. Th17 cells are critical for autoimmune arthritis development [63]. IL-21 and IL-6 have emerged as key for the cytokine-promoting differentiation of Th17 cells, and Tfh and Th17 cells secrete large amounts of IL-21 [64]. However, the proportion of Tfh cells was significantly reduced, and Th17 cells were even higher in IL-21R^−/− K/BxN mice, which completely lose the spontaneous development of arthritis. Katharine E et al. performed an adoptive transfer experiment and showed that RORγt^−/− and RORγt^+/+ CD4⁺ T cells induced arthritis with similar kinetics and severity, and no difference was observed in serum anti-GPI IgG titers [65]. All these results demonstrated that IL-21 plays a pathogenic role in the development of autoimmune arthritis via mechanisms that dependent on Tfh cells, but may not be dependent on Th17 cells [62].

RA patients have higher serum levels of IL-21 than healthy individuals, which correlate with the disease activity score, the frequency of cTfh-like cells, and serum levels of autoantibodies (e.g., RF, anti-CCP, and anti-mutated citrullinated vimentin (anti-MCV)) [66–68]. IL-21R is highly expressed in memory B cells, macrophages, and fibroblasts in inflamed synovial tissues of RA patients [69, 70]. IL-21/IL-21R is involved in B cell development and function and induces B cell activation, expansion, and the induction of plasma cells [71] to play an important role in B cell responses in the joint and sites of inflammation. However, IL-21 also directly promotes the aggressive proliferation, migration, invasion, and matrix metalloproteinase (MMP) secretion of fibroblast-like synoviocytes (FLSs) in RA patients [72, 73], which indicates a pathological role in inflamed synovial tissues.

Potential strategies for the targeting of follicular helper T cells to treat RA

Targeted Tfh cell-related surface molecules

CD40L: The CD40/CD40L pathway plays an essential role in the initial phase of Tfh cell development and function [74, 75]. The frequency of Tfh cells was significantly reduced in patients with CD40L deficiency [76]. CD40/CD40L signaling in RA promotes autoantibody production and class switching and the expression of proinflammatory cytokines, chemokines, adhesion molecules, and other costimulatory molecules. This pathway is directly involved in tissue damage via stimulation of the expression of MMP and receptor activator of NF-kB ligand (RANKL) [77]. The expansion of a CD4⁺ T cell subpopulation that highly expresses CD40L in the peripheral blood was identified in RA patients. Increased levels of sCD40L and the overexpression of CD40L mRNA were also observed in these patients. There was a gradual increase in CD40L when the data were stratified according to DAS28 [78]. Notably, the upregulated expression of CD40 and CD40L was also observed in undifferentiated arthritis (UA), early RA, and established RA synovial tissues [79]. These findings suggest CD40L as an ideal marker of the clinical activity of RA.

CD40L blockade antibodies reduced the number of Tfh cells in draining lymph nodes (LNs) and nondraining LNs and prevented allogeneic humoral immune responses in mice [80]. The CD40L-binding protein VIB4920, which reduces T cell-dependent antibody responses [81], dose-dependently suppressed antigen-specific IgG in healthy volunteers after priming and boosting with KLH. VIB4920 also significantly reduced circulating Ki67⁺ dividing B cells, class-switched memory B cells, and plasma cell gene signatures after immunization [82]. VIB4920 treatment in RA patients significantly reduced disease activity and immunological and inflammatory biomarkers [82]. Notably, studies revealed unexpected anabolic bone activity related to pharmacological CD40L suppression, which promoted bone formation and the accretion of vertebral bone mass in mice. This finding may provide a novel strategy to reduce RA activity and promote bone formation to ameliorate osteoporotic bone loss [83]. Another CD40 ligand antibody, BG9588, also reduced anti-dsDNA antibodies and decreased hematuria in patients with proliferative lupus nephritis [84]., The antagonistic anti-CD40 monoclonal antibody BI655064 reduced activated B cell numbers, autoantibody production, and inflammatory and bone resorption markers in RA patients with an inadequate response to methotrexate (MTX-IR) without serious adverse events [85].

ICOS: Tfh cells highly express ICOS. ICOS is a member of the CD28 superfamily that is induced on T cells upon activation. ICOSL (also known as B7RP-1) is the sole ligand of ICOS and is primarily present on the surface of B cells or other antigen-presenting cells [86]. Primary immunodeficiency in patients lacking ICOS activity further supports the requirement for ICOS/ICOSL in humoral immune responses [87]. In established allergic asthmatic mice, blocking antibodies targeting the ICOS/ICOS-L pathway substantially reduced LNs and lung Tfh cells but had no significant effect on the differentiation of other CD4⁺ T cell subsets, which reduced GC responses, allergen-specific IgE production, and airway hyperresponsiveness [88]. Therefore, inhibition of ICOS/ICOSL activity is a novel strategy with great prospects to interfere with Tfh–B cell interactions and modulate autoantibody production in autoimmune diseases, such as RA. Notably, a randomized, double-blind, parallel-group, placebo-controlled, multiple-dose study showed that a fully human antibody against ICOSL, AMG557, showed safety and potential efficacy in the treatment of patients with SLE and active lupus arthritis, which supports further evaluation of the clinical efficacy of ICOSL blockade in patients with RA[89].

Targeted Tfh cell-related cytokines

IL-21: IL-21 is a signature cytokine produced by Tfh cells that plays a central role in the activation and proliferation of various immune cells involved in RA pathogenesis, including Th17 cells, Tfh cells, B cells, FLSs, and macrophages [90]. Mouse experiments showed that an anti-IL-21 blocking antibody significantly reduced the titers of autoantibodies, diminished tissue-infiltrating Tfh cells, and delayed the progression of autoimmune disease [91]. CIA mice are a recognized RA model, and treatment with the IL-21R-Fc fusion protein significantly decreased the histological signs and clinical symptoms in a dose–response manner [92]. Blockade of IL-21 was also tested in randomized, double-blind, placebo-controlled trials. The human recombinant anti-IL-21 monoclonal antibodies, NNC0114-0005 and NNC0114-0006, were assessed for safety, pharmacodynamics, and efficacy in RA patients and healthy subjects. A significant reduction in DAS28 and C-reactive protein (CRP) was found in the anti-IL-21 antibody-treated group with large improvements in swollen and painful joints [93, 94].

IL-6: As an important cytokine involved in Tfh cell differentiation, elevated levels of IL-6 were found in the serum and synovial fluid of affected joints in RA patients [95, 96] and correlated with disease activity [97]. An increase in serum IL-6 was also associated with Tfh cell generation, B cell activation, and radiographic progression in RA patients [98, 99]. Several recent IL-6 signal blockers were widely used in RA treatment, with favorable efficacy and safety profiles. The IL‑6R blockers, tocilizumab [100, 101], and sarilumab [102, 103], which bind mIL-6R and sIL-6R, showed great clinical efficacy according to several clinical trials. Another bispecific anti-IL-6R nanobody with low toxicity and immunogenicity in vivo, ALX-0061 (vobarilizumab) [104], is currently in clinical development, with promising results from a phase I/II trial in RA [105, 106]. Humanized mAbs for IL‑6 blockade, such as olokizumab and sirukumab, were investigated in phase III trials. Olokizumab treatment induced sustained improvements across a range of patient-related outcomes in MTX-IR RA patients [107, 108]. Sirukumab also induced significant improvements in RA symptoms with inhibition of structural damage progression and improvements in quality of life [109, 110].

Reducing the frequency of Tfh cells

IL-6 and IL-21 support Tfh cell differentiation in mice, and IL-2, IL-10, and IL-12 suppress Tfh cell fate [111, 112]. Low-dose IL-2 treatment inhibited the differentiation and/or function of Tfh cells in SLE patients because IL-2 signaling and downstream STAT5 activation inhibited the differentiation of Tfh cells, at least in mice [113, 114]. Exogenous IL-2 also controlled the balance of Th1, Th17, regulatory T cells (Treg), and follicular regulatory T cell (Tfr) distribution in RA patients treated with TNF-α inhibitors, which supports its application as a combined strategy to treat RA [115, 116]. Recent studies showed that IL-12 induced high T-bet expression in T cells and suppressed Tfh cell differentiation, which provides another viable strategy to control excessive immune responses [117]. Based on the high expression of PD-1 on Tfh cells, Reighard et al. designed PD-L1 chimeric antigen receptor (CAR) NK cells that robustly and discriminately eliminated Tfh cells in vitro and in a humanized mouse model of lupus-like disease [118, 119]. This research established a novel strategy for the selective depletion of pathogenic Tfh cells or other PD-1-positive cells in RA.

Enhancing Tfr cell differentiation and function

As a suppressor in the GC response, Tfr cells limit the size of GCs and act as a counterbalance to Tfh cells [120]. Increasing the frequency of Tfr cells may be a useful approach to inhibit excessive GC responses in autoimmune diseases. Li and colleagues demonstrated that TGF-β and IL-2 promoted the development of Tfr cells and played pivotal roles in inhibiting GC responses [121]. IL-21 is one of the most important cytokines associated with Tfh cell differentiation because it promoted Tfh cell fate and inhibited Tfr cell development in BXD2 mice [122]. Therefore, blockade of IL-21 signaling may be a great strategy to inhibit Tfh cells and induce Tfr cell differentiation to further eliminate excessive GC responses and prevent the development of autoimmune diseases. There are few studies on Tfr cell-related therapeutic strategies for the treatment of RA, but it may be promising to target Tfr and balance GC responses in future studies (Fig. 2).

Fig. 2.

Therapeutic Tfh-related targets in RA: present and future. Several approved RA therapies or currently in clinical trials affect the differentiation and function of Tfh cells. NNC0114-0005/NNC0114-0006 and olokizumab and sirukumab are mAbs that target the Tfh-related cytokines IL-21 and IL-6, respectively. Tocilizumab, sarilumab, and ALX-0061 target IL-6R and may be used to treat RA. Blockade of costimulatory molecules with AMG557 (ICOSL), abatacept (CD28) [123], BI655064 (CD40), and VIB4920/BG9588 (CD40L) may also modulate Tfh cell differentiation by decreasing the strength of T-B interactions. Reducing the frequency of Tfh cells or enhancing Tfr cell differentiation and function may also be a feasible strategy to treat RA. IL-2, IL-10, and IL-12 directly suppress Tfh cell generation, and IL-2 and TGF-β also regulate Tfr cells, which may further restrain the excessive Tfh cell response

Conclusion

Although the prognosis of RA has improved markedly in the past few decades, RA remains and incurable disease. Improved knowledge of this disease is critical in the development of more potent and specific treatments. Tfh cells are specialized helpers of B cells that are involved in B cell activation and differentiation and play a critical role in GC responses, which is essential for the generation of autoantibody-producing plasma cells after antigen exposure. These roles make Tfh cells promising therapeutic targets in the treatment of autoantibody-mediated autoimmune diseases, such as RA. Therefore, the signals required for Tfh cell development or function, such as CD40L, IL-21, and IL-6, may be interesting targets to reduce Tfh cell numbers (and/or the proportion of Tfh cells) or modulate their function and reduce GC responses to inhibit RA development. Another therapeutic option involves enhancing the molecular and cellular regulators of Tfh cells, such as IL-2, IL-10, and Tfr cells, which negatively regulate Tfh cell differentiation or function. A recent study also showed that moderate alcohol consumption had immune modulatory effects that inhibited the development of RA via modulation of the Bcl6/PD-1/IL-21 axis in Tfh cells. This study showed great prospects for the use of nutrition to prevent and treat autoimmune diseases [124]. In summary, RA is a complex autoimmune disease with multiple regulatory mechanisms, and traditional monotherapy may have limitations. Combinations of strategies could pave the way for the development of immunotherapies for RA. It is important to systematically assess the correlation between the number of circulating Tfh-like cells and markers of RA diagnosis, severity, treatment efficacy, and prognosis. Although studies are currently underway to elucidate the detailed mechanisms involved in RA, the targeting of Tfh cells, their cytokines and functional molecules may be a promising therapeutic strategy in the treatment of RA.

Author contributions

JL drafted the manuscript. JW, XX, and HP discussed and revised the manuscript. SW designed the study and revised the manuscript. All authors read and approved the final manuscript.

Funding

This work was supported by the National Natural Science Foundation of China (Grant Nos. 81771759 and 82071835), Postgraduate Research & Practice Innovation Program of Jiangsu Province (Grant NO. KYCX20_3050), and Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions.

Declarations

Conflict of interest

The authors declare that they have no competing interests.