Follicular Helper T Cells in Systemic Lupus Erythematosus: Why Should They Be Considered as Interesting Therapeutic Targets?

Systemic lupus erythematosus (SLE) is a chronic autoimmune disease characterized by B cell hyperactivity leading to the production of autoantibodies, some of which having a deleterious effect. Reducing autoantibody production thus represents a way of controlling lupus pathogenesis, and a better understanding of the molecular and cellular factors involved in the differentiation of B cells into plasma cells could allow identifying new therapeutic targets. Follicular helper T cells (TFH) represent a distinct subset of CD4+ T cells specialized in providing help to B cells. They are required for the formation of germinal centers and the generation of long-lived serological memory and, as such, are suspected to play a central role in SLE. Recent advances in the field of TFH biology have allowed the identification of important molecular factors involved in TFH differentiation, regulation, and function. Interestingly, some of these TFH-related molecules have been described to be dysregulated in lupus patients. In the present review, we give an overview of the aberrant expression and/or function of such key players in lupus, and we highlight their potential as therapeutic targets.


Introduction
Systemic lupus erythematosus (SLE) is a severe systemic autoimmune disease and, as such, is characterized by a loss of self-tolerance. The etiology of SLE is not well defined, but genetic, hormonal, and environmental factors, as well as immune disorders, are likely implicated. During SLE, inflammation leads to damage of various tissues, including the joints, skin, kidneys, heart, lungs, blood vessels, and brain. Dysregulation of various components of the immune system can be observed at different stages of disease development, but hyperactivity of B cells, leading to excessive production of multiple autoantibodies (autoAb), is one of the major immunological stigmata of SLE. Indeed, SLE is characterized by the production of antinuclear autoAb (e.g., autoAb specific for chromatin) and by the formation of immune complexes, which contribute to tissue damage. Deposits of immune complexes in organs such as kidneys lead to subsequent inflammation through the activation of the complement system and the recruitment of inflammatory cells. The presence of autoAb is an absolute prerequisite for the development of lupus nephritis [1] and, interestingly, we demonstrated that pathogenic autoAb can be locally produced by plasma cells, which have homed to inflamed kidneys of lupus mice [2]. B cells and derivatives (plasma cells) are thus considered at the center of SLE pathogenesis and this is supported by the observation of a high frequency of plasma cell precursors in the blood of children with SLE [3]. Furthermore, an increase of circulating plasma cells in lupus patients is correlated with disease activity [4].
The generation of Ab can occur via the extrafollicular or the germinal center (GC) responses. The extrafollicular response leads to short-lived plasma cells, which do not go through the affinity maturation process. In contrast, the GC is the theater of intense cell collaboration between GC B cells and follicular helper T cells (T FH ) leading to the differentiation of long-lived plasma cells harboring high antigen-specificity. Interestingly, lupus autoAb are high affinity, somatically mutated, and class-switched immunoglobulin (Ig)G [5] indicating T and B cell collaboration [6] and intense GC activity. Therefore, it is likely that a dysfunction in B cell differentiation mechanisms occurs in lupus, leading to 2 Journal of Immunology Research excessive numbers of autoreactive plasma cells. It is particularly attracting and plausible to envisage that a dysregulation of T FH could be the underlying key factor.
In this review, we succinctly expose recent understanding in T FH biology (described in detail elsewhere; see [7] for review), in order to introduce important molecular factors involved in T FH differentiation, regulation, and function. We then give an overview of the aberrant expression and/or function of such key players in lupus patients, and we highlight their potential as therapeutic targets.

T FH Cells: From Their Generation to Their Regulation
The generation of high affinity Ab requires T/B interactions that mainly occur in GC. T FH cells represent a distinct subset of CD4 + T cells involved in GC formation and specialized in providing help to B cells to differentiate into plasma cells or memory B cells [8]. T FH express high levels of CXC chemokine receptor type 5 (CXCR5), PD-1 (Programmed Death-1), ICOS (Inducible T cell CO-Stimulator), and the regulator transcription factor Bcl6 (B cell lymphoma 6), which provide excellent markers for their identification. Moreover, secretion of high levels of IL-21 is a critical characteristic of T FH cells. T FH are generated after immunization or infection following the interaction of naive CD4 + T cells with dendritic cells (DC) within the T cell zone of secondary lymphoid organs (SLO). Signals provided by DC induce the expression of a myriad of proteins (transcription factors, surface molecules, and cytokines) that are essential for T FH generation, migration, and function. In fact, T FH differentiation is a multistage process (Figure 1), which can be sequentially defined as follows: (i) naive CD4 + T cells are activated by DC (thanks to the MHC-peptide complex/TCR interaction) in the T cell zone and become immature T FH (also called pre-T FH ) [9]; (ii) newly generated pre-T FH then migrate to the interfollicular zone, where cognate interactions with B cells allow the final maturation step; (iii) these mature T FH reach the GC in which T FH -GC B cell interactions will favor isotype class switch, somatic hypermutations, and affinity maturation.

Pre-T Generation: DC as the Stage Director.
The initial priming of CD4 + T cells requires cognate interactions and costimulatory signals delivered by DC through CD40, CD80/86, ICOSL, and OX40L (Table 1). CD28 (that binds CD80/86) was shown to be essential to T FH development as mice deficient for CD28 display CD4 + T cells that fail to upregulate CXCR5 and OX40, leading to disrupted GC formation [10]. In addition, upregulation of OX40L on DC following CD40-induced maturation allows CXCR5 expression by OX40 + T cells [11]. Moreover, ICOS signaling leads to an increased expression of the transcription factors Bcl6 and Ascl2 (achaete-scute homologue-2). The latter promote both the reciprocal CXCR5 upregulation and CCR7 downregulation on activated CD4 + T cells, which then become pre-T FH [12,13]. In turn, Bcl6 induces the expression of ICOS, PD1, CD40L, and SAP (SLAM-(Signaling Lymphocytic Activation Molecule-) Associated Protein; critical for T-B interaction).
Cytokines secreted by DC also play a pivotal role in pre-T FH development (Table 1). IL-6, a DC-derived proinflammatory cytokine, has been demonstrated to be the main soluble factor driving T FH differentiation in mice [14]. In humans, IL-12 has been shown to be the key cytokine that promotes T FH -like cell differentiation [15,16]. If, in the initial work, neither IL-6 nor IL-21 were described as being able to promote T FH differentiation [15], a recent study suggests that human plasmablasts produce IL-6, which is responsible for the subsequent differentiation of naive CD4 + T cells into B cell helpers CXCR5 + ICOS + Bcl6 + IL-21 + T cells [17]. IL-21 is required for T FH function but it is also an important factor for T FH generation [18] and, interestingly, both IL-6 and IL-12 are potent inducers of IL-21 expression in mice [19] and humans, respectively [15]. As IL-21 is an autocrine cytokine for pre-T FH generation, further studies are required to better clarify individual cytokine contributions. Cytokine signaling involves the subsequent activation of Janus kinase-STAT (Signal Transducer and Activator of Transcription) signaling pathway. STAT3 is a major signaling molecule for IL-6 and IL-21 [20,21], whereas IL-12 signaling occurs through STAT4 activation. However, IL-12-induced expression of IL-21 by human CD4 + T cells is compromised in patients with functional STAT3 deficiency, suggesting that IL-12 ability to promote IL-21-producing CD4 + T cells is predominantly STAT3 dependent [22]. Moreover, STAT3-deficient patients have reduced numbers of circulating T FH -like cells [23]. Altogether, these data suggest that the STAT3 signaling pathway plays an important role in T FH differentiation and subsequent B cell help.

Pre-T Migration to the T-B Border and T Maturation: B Cells
Enter the Scene. Thanks to CXCR5 expression enhancement and CCR7 downregulation ( Table 2), pre-T FH cells migrate to the B cell follicle in response to a CXCL13 gradient and their interaction with antigen-specific B cells at the T-B border contributes to final T FH differentiation. Indeed, the lower frequency of T FH cells in B celldeficient mice suggests that B cells are also important for the generation of T FH cells [24]. At this stage, B cells act as the major antigen-presenting cells (APC) for primed-T FH that will then fully differentiate into GC T FH cells. Mature T FH and B cells that have formed stable T-B conjugates move together into the follicle to form GC [31]. Stable T-B conjugate formation requires interaction between ICOS on T FH and ICOSL expressed by B cells, as well as SLAM interactions (Table 2). SLAM are transmembrane receptors expressed on both T FH and B cells. SAP, which is the adaptor signaling protein downstream of SLAM, was demonstrated to be important for stabilizing cognate T-B interactions. Indeed, SAP-deficient CD4 + T cells have an impaired capacity to stably interact with cognate B cells, resulting in a failure to induce B cell clonal expansion [32]. Moreover, patients with X-linked Lymphoproliferative disease (XLP), an immunodeficiency resulting from mutations in the SH2D1A gene which encodes SAP, harbor humoral defects characterized by hypogammaglobulinemia and reduced numbers of T FH [33]. B cells thus play a key role in the T FH maturation step by both acting as APC and stabilizing T FH -GC B cell interactions through ICOSL and SLAM.

T Function: The Final
Act of the Story. The major function of T FH is to enhance high affinity memory Ab responses following migration to GC. In the follicles, T FH -GC B cell crosstalk involves CD40L, IL-21, PD-1, and BAFF (B cell Activating Factor) ( Table 3). The signal delivered through interaction between PD-1 on T FH and PD-L1 expressed by GC B cells is crucial for GC B cell survival [34]. IL-21 production by T FH directly regulates B cell proliferation and classswitch, and the IL-21 pathway has been identified as a critical   [36]. Moreover, BAFF production by T FH is critical for the survival of high affinity B cell clones [37].
In summary, molecules that have been described to play a key role in T FH biology do not display equivalent functions. Some are necessary for T FH migration from the T cell zone to the GC, others are absolutely required for their development or function, and finally some of them are essential for T FH maintenance and survival (Tables 1-3).

T Regulation.
Considering the important role of T FH cells in humoral immunity, a balance between stimulatory and inhibitory mechanisms regulating their function is required for immune homeostasis. However, while signals important for T FH development are clearly defined nowadays, little is known about mechanisms involved in their regulation. The coinhibitory PD-1/PD-L1 pathway can limit T FH expansion and consequently the humoral Ig response [38]. Similarly, it was demonstrated that the inhibitory receptor B and T Lymphocyte Attenuator (BTLA) suppresses GC B cell development and subsequent IgG responses by inhibiting IL-21 production by T FH cells [39] (Table 3). Recently, the existence of regulatory T cells (Treg) able to inhibit GC responses was described. This subset of regulatory T cells of thymic origin was first identified in mice [40] and named T FR (follicular regulatory T cells). They express typical markers of both T FH cells (Bcl6, CXCR5, PD-1, and ICOS) and classical Treg (Foxp3); they localize in the GC and possess suppressive activity. A CD4 + T cell population coexpressing Foxp3, Bcl6, and CXCR5 was also visualized in human tonsils [41]. Moreover, microRNA have recently emerged as potent regulators of T FH differentiation. Indeed, the miR-17∼92 cluster was shown to promote T FH differentiation by repressing PTEN (Phosphatase and TEnsin homolg), PHLPP2 (Pleckstrin Homology domain and Leucine-rich repeat Protein Phosphatase) (phosphatases that inhibit Bcl6 expression through interfering with PI3K signaling), and ROR (Retinoic acid-related Orphan Receptor ) expression [42,43]. On the other hand, miR-10a negatively regulates T FH differentiation by directly inhibiting Bcl6 expression [44]. Similarly, miR-146a, a microRNA that is highly expressed in T FH cells, was recently described as a negative regulator of T FH cell numbers [45]. miR-146a deficiency leads to accumulation of both T FH and GC B cells, likely due to enhanced ICOSL and ICOS expression on GC B cells and T FH cells, respectively [45].

Evidences Supporting the Involvement of T FH in Systemic Lupus Erythematosus (SLE)
The main function of T FH cells consists in regulating the clonal selection of GC B cells and providing B cells with signals for Ig production, isotype switching, and somatic hypermutations. As abnormal activation of B cells and autoAb production are central to autoimmune diseases, such as lupus, altered T FH differentiation, function, and regulation were suspected to play a role in lupus pathogenesis. First hypotheses regarding the role of T FH cells in SLE development are based on studies using mice deficient for Roquin1 (a negative regulator of ICOS mRNA stability) in which an excessive number of T FH cells and GC reactions and high levels of IL-21 are associated with a lupus-like phenotype [50,51]. Other evidences come from studies on IL-21, the main cytokine produced by T FH , in lupus mice. High IL-21mRNA as well as elevated IL-21 serum levels were described in BXSB.Yaa mice, which develop an SLE-like disease [52]. The use of a fusion protein consisting in the IL-21R linked to the Fc domain of a mouse IgG2a (IL-21R.Fc, which therefore binds to IL-21 and prevents activation of its receptor) revealed a complex biphasic role of IL-21 in this mouse model as it increases or diminishes the disease severity depending of the stage of the disease at the time of IL-21 neutralization (at early or late stages). This could be related to the action of IL-21 on B cells but also on T cell responses [53]. In lupus MRL/lpr mice, activated CD4 + T cells secrete 10 times more IL-21 than control mice [54] and IL-21R deficiency leads to reduced numbers of T FH cells [55]. In addition, abundant T FH -like cells are located outside the GC where they support extrafollicular B cell differentiation and plasmablast maturation in BXSB-Yaa and MRL-Fas lpr lupus mice [56,57]. In the latter and contrary to what was expected, the extrafollicular pathway was shown to be the most important way to generate hypermutated autoAbs [58]. However, there is no evidence to date supporting the involvement of such extrafollicular response in human SLE. T FH cells are located in SLO; therefore the major problem encountered in studies of human T FH is that lymphoid tissues of lupus patients cannot be easily accessed, making it difficult to identify T FH cells and to determine whether the generation or function of these cells is dysregulated. First studies were based on the enumeration of CD4 + CXCR5 + in peripheral blood as GC T FH counterparts. Using this strategy, it was shown in human SLE that circulating T FH cells (cT FH ) defined as CD4 + CXCR5 + PD-1 +/high and/or ICOS + T lymphocytes are expanded in lupus patients and their presence correlates with a more severe disease phenotype [59][60][61][62][63][64]. Recent studies have more rigorously characterized peripheral CXCR5 + CD4 + T cells. Morita et al. have described a circulating population in healthy donors that shares common phenotypic and functional characteristics with T FH cells from GC [65]. The authors named it T FH -like cells. Moreover, they distinguished three subclasses, that is, T FH 17, T FH 2, and T FH 1, defined according to the expression of the CCR6 and CXCR3 chemokine receptors: T FH 17 cells are CXCR3 − CCR6 + cells whereas T FH 2 cells are CXCR3 − CCR6 − cells and T FH 1 cells are CXCR3 + CCR6 − cells. T FH 17 and T FH 2 cells were identified as able to provide help to B cells via IL-21 production, resulting in IgM and IgG secretion, whereas T FH 1 have limited helper functions. However, ICOS expressing T FH 1 are able to help memory B cells (but not naive B cells) to produce Ab following influenza vaccination [66]. Moreover, Morita and colleagues showed that patients with juvenile dermatomyositis displayed a profound skewing of cT FH cells towards T FH 2 and T FH 17 cells that correlated with disease activity, suggesting that an altered balance of T FH subtypes contributes to human autoimmunity [65]. Recently, the differential expression of ICOS, PD-1, and CCR7 interestingly allowed distinguishing three memory cT FH subsets defined as activated cells (ICOS + PD1 hi CCR7 lo ) or quiescent cells (ICOS − PD1 + CCR7 int and ICOS − PD1 − CCR7 hi ) [67,68]. In SLE patients, the frequency of CCR7 lo PD1 hi CXCR5 + CD4 + T cells is significantly higher than in healthy individuals [67]. The CCR7 lo PD1 hi subset is indicative of active T FH differentiation and its overrepresentation is associated with elevated autoAb titers and high disease activity [67]. By analyzing CXCR3 and CCR6 expression, we also interestingly described an altered phenotype of cT FH cells characterized by the enhanced frequency of B cell helper T FH 2like CXCR3 − CCR6 − cells and a decreased frequency of CXCR3 + CCR6 − T FH 1-like cells (not able to provide B cell help) in lupus patients with an active disease [69].

Molecules and/or Cytokines Involved in T FH Generation/Regulation Are Associated with Lupus Pathogenesis
Aberrant expression and/or function of T FH -related molecules are associated with lupus-like disease in mice [54,70]. Similarly, in lupus patients, numbers of molecules involved in T FH generation and/or regulation have been described to be dysregulated.

Surface
Molecules. CD40/CD40L pathway plays an essential role in the initial phase of T FH development (T-DC interaction in the T cell zone; Figure 1, [11]) and function (T FH -GC B cell crosstalk in the GC; [71]). Interestingly, CD40L was found to be constitutively expressed at abnormally high levels on T cells (but also on B cells and monocytes) from lupus patients [72,73]. Furthermore, CD4 + T cells from female lupus patients, which overexpressed CD40L mRNA, were able to promote autologous B cell stimulation and autoAb production [74]. ICOS-mediated PI3K signaling is absolutely required for T FH differentiation, for T FH migration into the follicle [75], and also for T FH maintenance [76]. PTEN acts as a negative regulator of the PI3K signaling pathway, leading to the inhibition of Bcl6 expression and T FH differentiation. Interestingly, PTEN expression is significantly decreased in SLE B cells [77]; however, to the best of our knowledge, its expression in lupus CD4 + T cells (especially T FH ) has not been investigated yet. ICOS expression has been found to be enhanced in CD4 + T cells from lupus patients compared to healthy donors [78,79] and ICOS levels were higher in patients with nephritis than in those without nephritis [80]. Moreover, infiltrated ICOS + T cells were shown to be in close contact with B cells in lupus kidneys [79].
Interaction between OX40L (on DC) and OX40 (on activated CD4 + T cells) is also important for T FH development. OX40 expression by lupus peripheral blood cells was found to be predominantly restricted to memory CD45RO + CD4 + T cells and its levels correlated with disease activity [81]. Moreover, OX40 has also been found to be highly expressed in kidneys of patients with lupus nephritis [82]. Importantly, the upstream region of the OX40 gene contains a single risk haplotype for SLE, which is correlated with increased expression of OX40 mRNA and protein [83]. Finally, it was Journal of Immunology Research 7 recently shown that OX40 signal promotes, ex vivo, the generation of T FH -like cells that are functional B cell helpers [84].

Cytokines.
Cytokine signals are absolutely required for T FH differentiation. Elevated levels of IL-6 have been found in the serum and in the urine of active SLE patients [85][86][87]. The increased frequency of IL-6-producing peripheral blood mononuclear cells (PBMC) correlates with disease severity/activity and treatment response [88]. Raised expression of gp130 (one of the two subunits of the IL-6 receptor) has been found on CD4 + T cells and B cells from patients with active SLE, while an important reduction in the gp130 expression by B lymphocytes was observed upon immunosuppressive treatment leading to milder disease activity [89]. Factors responsible for the constitutive expression of IL-6 in SLE have not been elucidated yet.
Serum IL-21 levels were found to be elevated in patients with SLE [69,90], especially in patients with lupus nephritis, and to correlate with disease severity [90]. The real-time PCR analysis of skin biopsies taken from 3 lupus patients also revealed that IL-21 transcripts were significantly increased compared to control individuals [91]. Furthermore, the percentages of CD4 + T cells producing IL-21 are significantly enhanced in lupus patients [92]. Finally, polymorphisms within the IL-21R and the IL-21 genes have been reported and may confer risk for SLE: a polymorphism in IL-21R (namely, rs3093301) was found to associate with lupus in 2 independent cohorts [93], a genetic association of two SNPs located in intronic regions of the IL-21 gene (rs2221903 and rs907715) was described [94], and the variant allele rs2055979A of the IL-21 gene was recently found to be associated with increased IL-21 levels [95].
Regarding BAFF, lupus sera have been shown to contain elevated levels of this cytokine and those levels correlate with both anti-dsDNA titers [96][97][98] and disease activity [99]. Finally, it has been reported that IL-2 production (which inhibits T FH differentiation) upon TCR stimulation is impaired in SLE T lymphocytes [100,101]. This lower IL-2 production could be explained by imbalanced expression between the transcription factors cAMP response element (CRE) binding protein (CREB) and the CRE-modulator (CREM), which, respectively, enhance and suppress the IL-2 gene transcription [102].

Transcription Factors, miRNA, and Regulatory T Cells.
STAT3, which is activated by cytokines such as IL-6 and IL-21, binds to the Bcl6 promoter leading to high levels of Bcl6 expression and is thus important for T FH differentiation. T cells from patients with SLE display increased levels of total and phosphorylated STAT3 [103,104].
Reduced expression of miR-146a (a negative regulator of T FH development) has been reported in PBMC from SLE patients [105] and seems to correlate with disease activity [105]. Moreover, a genome-wide association study has highlighted a variant, that is, rs2431697, in an intergenic region between PTTG1 (Pituitary Tumor-Transforming 1) and miR-146a, associated with lupus susceptibility [106]. Interestingly, the risk allele of this SNP correlates with a diminution of miR-146a levels [107].
To date, the analysis of frequency and/or functionality of T FR cells in an autoimmune context has not been reported. However, although there may be some discrepancies due to variations in phenotype analysis, peripheral regulatory T cells (CD4 + CD25 + T cells) seem to play a role in human lupus pathogenesis. Several studies reported that a decreased number of Treg might contribute to the pathogenesis [108][109][110][111], but there were conflicting data regarding Treg function in lupus patients. The in vitro suppressive activity of these cells was found to be defective in some reports [111,112] but other studies showed that the suppressive activity of highly purified Treg from lupus patients is not altered. It has been proposed that defective suppression in lupus could be attributed either to a higher sensitivity of Treg to Fas-mediated apoptosis in an SLE context [108] or to a lower susceptibility of effector T cells to Treg suppression [113]. Finally, it has been shown that IFNproduction by lupus APCs might be responsible for altered Treg functionality [114].

Targeting T FH : From Lupus Mice to Lupus Patients
Data obtained from various lupus mouse models have already highlighted how blockade of signaling pathways involved in T FH generation could lead to disease improvement. The administration of a blocking ICOS-L specific monoclonal Ab (mAb) to lupus NZB/W mice interrupted T FH cell development leading to a decrease of autoAb levels and glomerulonephritis [115,116]. Similar results were obtained in MRL/lpr lupus mice displaying a genetic deletion of ICOS [57]. Blockade of the CD40L-CD40 signaling pathway also led to the reduction of lupus symptoms in different mouse models [117,118]. Treatment of MRL/lpr lupus mice with a neutralizing anti-IL-6R mAb has favorable effects on renal function and leads to a reduction of anti-dsDNA Ab levels [119]. In NZB/W mice, chronic administration of anti-IL-6 or anti-IL-6R mAb improves survival and reduces the progression of proteinuria and anti-dsDNA levels [120,121]. In lupus-prone NZB/W and MRL/lpr mice, raised levels of BAFF are detected at the onset of the disease [122] and treatment with either TACI-Ig or BR3-Ig is effective at preventing clinical disease and ameliorating renal injury [123]. Regarding IL-21, its neutralization using IL-21R.Fc showed an improvement of biological and clinical signs of the disease in MRL/lpr lupus mice and BXSB-Yaa mice [53,54]. Moreover, the administration of Ab specific for the IL-21R to MRL/lpr mice significantly reduced anti-dsDNA Ab titers and IgG deposits in the kidneys when compared to control mice [124]. In NZB/W mice, such IL-21R blocking even allowed reversing nephritis and halting disease progression in mice with preexisting lupus [125]. By using a miRNA-delivery approach via bacteriophage MS2 virus-like particles, Pan and colleagues recently showed that restoring the loss of miR-146a was effective in abolishing autoAb production and delaying SLE progression in lupus-prone mice [126]. Interestingly  also, treatment with the small molecule called Stattic (an inhibitor initially reported to block the phosphorylation, dimerization, and nuclear translocation of STAT3 in tumor cells) delayed the onset of proteinuria and reduced both anti-dsDNA autoAb and inflammatory cytokine levels in MRL/lpr lupus mice [127].
There is growing evidence of T FH involvement in the pathogenesis of human SLE. Several therapeutic tools targeting T FH biology already exist and even if their direct effect on T FH development has not been evaluated, some of them were shown to improve the disease. Tocilizumab, a humanized mAb specific for the -chain of the IL-6 receptor (which prevents IL-6 from binding to membrane bound and soluble IL-6 receptors), has been recently tested in SLE patients with promising results [128]. Interestingly, Tocilizumab therapy in rheumatoid arthritis patients leads to a significant reduction in circulating T FH cell numbers and IL-21 production [17]. Belimumab, a human mAb that binds soluble BAFF, therefore inhibiting recognition by BAFF specific receptors has been tested in patients and results from phase III clinical trials have demonstrated the safety profile and efficacy in controlling lupus in a broad range of patients [129]. Belimumab is the first biologic to meet its primary endpoint in a phase III clinical trial for lupus patients and it was approved by the US Food and Drug Administration in 2011. Among other potential therapeutic candidates, are those targeting T-B interactions, such as IDEC-131 (anti-CD40L Ab), AMG 557 (anti-ICOSL Ab), Abatacept (CTLA4-Ig), or targeting cytokines such as ATR-07 (anti-IL-21R Ab), NNC0114-0006 (anti-IL-21 Ab), Atacicept (TACI-Ig), and small molecules inhibiting cytokine signaling pathways (Tofacitinib, a Jak-STAT inhibitor) (Figure 2).

Concluding Remarks
Although prognosis in SLE has improved markedly in the last 40 years, a better knowledge of the disease remains of prime importance to develop more potent and specific treatments. New targeted therapies designed to block pathways involved in disease pathogenesis are on the horizon. One promising option could be to specifically target factors involved in the generation of plasma cells responsible for the production of pathogenic autoAb in lupus. T FH play a critical role in B cell activation and differentiation, and recent data have evidenced their involvement in lupus pathogenesis. Signals required for T FH development may thus represent interesting targets in order to reduce T FH numbers (and/or to correct the altered proportion of T FH subsets) or to qualitatively and/or quantitatively modulate their function. Another exciting therapeutic option consists in enhancing the negative molecular and cellular regulators of T FH , such as miRNA or T FR .