Systemic Lupus Erythematosus (SLE) is a chronic inflammatory autoimmune disease which is more prevalent in females of childbearing age and is characterized by multi-organ involvement [1]. Self-tolerance and autoreactive B cells play important roles in the pathogenesis of SLE. The environmental risk factors connected to apoptosis or clearance of apoptosis debris lead to the breach of self-tolerance, causing pathogenic activation of B cells. As a result of this breach, autoantibodies are produced and accumulate in multiple end-organs, often in immune complex deposits, giving rise to a spectrum of symptoms which include proteinuria, hemocytopenia, joint pain, and swelling, and involvement of the nervous system manifesting as epilepsy or psychosis, for example.

2. B Cell Development and Subpopulations

Haematopoiesis is comprised by two major pathways for immune cell development, giving rise to myeloid and lymphoid lineages, respectively. B lymphocytes (B cells) constitute an important component of the lymphoid lineage; they develop in the foetal liver and then the bone marrow, where they assemble their antigen-binding B cell receptor (BCR) and go through negative selection. B cells then migrate to secondary lymphoid organs, where they go through positive selection and acquire affinity maturation of antibody responses. Negative selection depletes self-reactive and polyreactive B cells (whose BCRs bind self-antigens) from the repertoire, whereas positive selection ensures that the BCR is functional. It is, however, important to note that a negative selection of B cells is not always complete, which may result in self-reactive and polyreactive B cells that escape into the periphery, and some of those carry autoreactive properties that later can be found as disease-driving autoantibodies in patients with SLE [2]. Selection continues in secondary lymphoid organs, where B cells arriving from the bone marrow turn from transitional type 1 (T1) to transitional type 2 (T2) B cells. The T2 B cells are activated through BCRs and integrating signals, including those from NF-κB- and B cell-activating factors belonging to the TNF family (BAFF), which are needed for survival and, later, activation and selection. T2 B cells develop into naïve subpopulations that inhabit different microanatomical niches and have varying recirculation patterns ^[3][4][3,4]. The main division is between B1 and B2 B cells. B1 cells derive from foetal liver but can also develop from bone marrow in adults [5]. B1 cells can be further divided into B1a and B1b B cells depending on their phenotype. In mice, there are distinct B1 subsets, but an exact equivalent to these in humans remains to be determined and their existence is still debated [6]. B2 B cells can be divided into follicular B cells (FOBs) and marginal zone B cells (MZBs), and these exist in both mice and humans. Of the B2 B cell subsets, FOBs are the major source for B cells recruited into the germinal centre (GC) reaction during adaptive immune responses. The name of MZBs derives from their location in the marginal zone of the spleen [3]. The marginal zone of the spleen is a crucial region where blood-borne pathogens are sequestered by macro-phages with unique arrays of scavenger molecules. The MZBs reside predominantly in the marginal zone of the spleen and cannot be found as a distinct population in other tissues in mice. In humans, on the other hand, MZBs are more abundant and recirculate through lymphoid organs. The contrast is possibly partly due to the difference in the anatomical composition of the spleen between humans and mice, with the human spleen lacking marginal sinus [7]. Although MZBs belong to the B2 B cell population, they share some features with B1 B cells. The MZBs are larger and more metabolically active, and they can be more easily activated by both antigen specific and non-specific stimuli such as TLR activation. Just like B1 cells, MZBs express more IgM than IgD as their BCR and can produce natural antibodies without going through GC selection and activation [8]. They are also self-renewing and partially generated from naïve B cell precursors and unswitched memory B cells ^[9][10][9,10]. The polyreactive feature of the natural antibody repertoire, which primarily comprises IgM antibodies, is important for immune defence. It is thought to exert effects both in the early immune defence and clearance of apoptotic debris in the circulation and serves therefore as a buffer in the host peripheral immune system. However, as polyreactive antibodies are self-reactive by nature, B cells that produce polyreactive antibodies have been proposed to be a possible source of pathogenic self-reactive antibodies. For instance, the BCRs of MZBs are often more polyreactive or self-reactive in steady states than that of FOBs. However, if simultaneously stimulated with self-antigen and pathogen- and/or damage-associated molecular patterns (PAMPs/DAMPs), they become pathogenic and react to self-tissue or organs. Therefore, they are considered to participate in the initial breach of self-tolerance in autoimmune responses. For example, Mackay et al. found that even though MZBs and B1a B cells induce the production of proinflammatory autoantibodies in BAFF-transgenic (BAFF-Tg) mice that develop an SLE-like phenotype, the depletion of B1a B cells or MZBs does not protect BAFF-Tg mice against disease [11]. This would suggest that FOBs and GCs are required. In addition, the enhanced maturation of FOBs and decreased generation of MZBs by intrinsic factors such as TLR, NF-κB signals, or Y-linked autoimmune accelerator (Yaa) mutation are associated with the production of lupus autoantibodies in murine models [12].

3. Central and Peripheral B Cell Tolerance

The vast capacity of the adaptive immune system in recognizing different pathogens comes with a cost of potential autoreactivity. Several mechanisms and multilayers of tolerance have evolved to prevent self-reactivity and the occurrence of autoimmune disease. Central B cell tolerance occurs in the early stages of B cell development in the bone marrow when B cells are not fully matured and only express IgM [13]. B cells develop central tolerance through clonal depletion and receptor revision or receptor editing to avoid self-reactivity [14]. Receptor revision allows immunoglobulin gene recombination and a light chain switch in B cells carrying autoreactive antigen receptors. A large portion of immature B cells go through this process to become functionally unresponsive. At the same time, receptor revision contributes to immune diversity by promoting the use of antibody genes that initially rearrange inefficiently. This is the key mechanism for setting the repertoire of naïve and pre-immune B cells. In patients with primary immunodeficiencies, the mutations affecting central B cell tolerance are mostly coupled with BCR signalling, or with molecules that regulate BCR signalling [15]. Thus, if receptor revision is not regulated properly, more high-affinity autoreactive B cells can escape immune surveillance and be recruited to secondary lymphoid organs to subsequently drive autoimmune disease, including SLE. However, a central tolerance mechanism alone is not sufficient for proper B cell development; later B cells need to be fully activated and regulated in the periphery to acquire the capacity of secreting high-affinity antibodies. This is referred to as peripheral tolerance. The most important mechanism of peripheral tolerance is connected to modification by regulatory T cells. Thus, as is to be expected, primary immunodeficiencies affecting peripheral tolerance are mostly connected to T cell function and the capacity of B cells to interact with T cells. This includes signalling from the T cell receptor (TCR) and co-stimulatory molecules. Co-stimulation is important for B cell activation; in the absence of this, B cells can become anergic, meaning that they are more resistant to activation upon a secondary stimulus. The reason for keeping these potentially pathogenic B cell clones in the immune system remains elusive. One possibility could be that depletion of these B cell clones creates niches in the repertoire, giving rise to failed recognition of viruses and bacteria that may carry similar epitopes. Anergic B cells can indeed be useful; it has been shown that the antibody affinity of anergic B cells matures through somatic hypermutation away from autoreactivity [16]. This was proven in an in vivo model where responses to both self-antigen and related foreign antigen were examined. In this in vivo model, B cells directed towards self-antigens were anergic, but could be activated and mutate away from self-reactivity when exposed to the related foreign antigen. These experiments prove that anergic B cells can be used as a source for specific antibody responses to foreign antigens. They also support the current hypothesis that self-reactivity is allowed in the B cell repertoire because of the potential to fight infections. Still, the reservation of anergic B cells requires precise regulation to avoid the occurrence of human immunodeficiency or autoimmune disease, including SLE.

4. Regulation of B Cells by Innate Immune Cells

The different naïve B cell populations including both B1 and B2 populations can all be activated to produce antibodies and/or cytokines as a part of a natural immune response. Even though antigen reactivities are enriched in different naïve B cell populations, it is thought that all naïve B cells can be recruited to a specific response. The activation of B cells can either be T cell dependent or T cell independent. In GCs, follicular dendritic cells (FDCs) could present multimerized repetitive antigens, such as LPS or DNA fragments. B cell responses are induced in the absence of T cells when multimerized FDC antigen and antibody immune complexes (ICs) simultaneously cross-link multiple BCRs. This process also needs a second signal from TLRs to provoke a strong antibody response without producing memory B cells [17]. However, the most potent driving force for B cell activation is T cell dependent. The first step for T cell-dependent immune responses is that naïve B cells present antigenic peptides on MHC that have been taken up via the BCR. This presentation is important for the specificity of the immune response since it ensures that the T cells and B cells are specific for the same antigen, although they bind different epitopes on that antigen. Then, the activated B cells in B cell follicles move to the border of the T cell zone to interact with T cells. The binding of CD40 and CD40L through cell–cell contacts provides a secondary signal for the final activation of B cells. The interaction between T cells and B cells also stimulates the secretion of large amounts of cytokines by T helper (Th) cells, which mediates antibody subclass switching and supports the affinity maturation that occurs in the GC [18]. In SLE, autoreactive GCs are central for the autoantibody response and the generation of pathological B cell memory. It has been shown that even a small amount of autoreactive B cells can drive a GC response and work as an engine to support other immune reactivities and promote the epitope spreading that occurs in autoimmune diseases [19]. Several pre-GC checkpoints exist to ensure the correct GC reaction once encountered by antigens. One of these is governed by innate type natural killer T (NKT) cells that allow polyreactive B cells to be activated, while preventing them from entering the GC. NKT cells are T cells selected in the thymus for their recognition of the MHC class I-like molecule CD1d. CD1d is expressed by antigen-presenting cells, including B cells and some activated T cells. Instead of presenting peptides, it presents glycolipids that can be foreign or derived from the self. These self-lipids are produced from cell death and cellular stress responses that could potentially be connected to cell death and autoinflammation [20]. BAFF belongs to the tumour necrosis factor (TNF) family and is a survival factor for B cells that sets the threshold for B cell activation. Mice that are transgenic for BAFF have increased numbers of B cells (in particular, T2 and MZBs) and elevated levels of autoantibodies, and they progressively develop nephritis, as well as other autoimmune manifestations [21]. BAFF can be produced by several cell subtypes, including granulocytes such as neutrophils. Neutrophils can boost a natural antibody response through the production of BAFF and in response to inflammation [22]. In SLE patients, BAFF is commonly overexpressed and strongly involved in the pathogenesis of renal involvement ^[23][24][23,24]. Interestingly, BAFF and neutrophil gene signatures are correlated with SLE activity [25]. In addition, neutrophils upregulate CD1d and license NKT cells to regulate B cells and prevent those from entering the GC [26]. This mechanism allows for a boost of beneficial natural antibodies but blocks the entry of potentially dangerous autoreactive B cells to the GC. Thus, there is a newly discovered close connection between inflammation, activation of the innate immune system, and B cell activation [27].

5. B Cell Functions in Autoimmunity and SLE

Autoimmune response is initiated when immunological tolerance to self-antigens cannot be maintained. As a result, naïve B cells are activated and differentiate into plasma cells. During this process, B cells switch survival factor dependence from BAFF to its closely related TNF family member, a proliferation-inducing ligand (APRIL) [28]. The activation of B cells leads to a state in which B cells produce specific regulatory cytokines including IL-10, IFNγ, TNF, GM-CSF, IL-6, IL-17, and IL-2 [29]. These cytokines have varying regulatory properties. For example, GM-CSF produced by so-called IRA B cells can promote the production of IL-12 from dendritic cells (DCs) in autoimmune diseases [30]. In SLE, the loss of B cell tolerance is controlled in a cell-intrinsic manner by Toll-like receptors (TLRs) that sense nucleic acids in endosomes. TLR7 drives the extrafollicular B cell response and the GC reaction that is involved in autoantibody production and acceleration of SLE, while TLR9 seems to protect against SLE [31]. SLE is characterized by immature B cells lacking IgD and CD27 (double negative B cells). Sanz et al. found a subset of CXCR5-CD11c+ cells in double-negative B cells representing pre-plasma cells. This group of effector B cells (pre-plasma cells) are predominant in SLE patients with active disease, lupus nephritis, or autoantibodies. The overproduction of pre-plasma cells is induced by unregulated TLR7 [32]. Interestingly, most of the excessively expanded pre-plasma cells also exhibit activation of a T-bet transcriptional network [33]. This connects to the newly discovered autoimmunity-associated B cells, also known as age-associated B cells (ABCs), where expression of the transcription factor T-bet has been implicated in autoimmunity and situations of chronic inflammation [34]. ABCs require TLR7 and TLR9 activation as well as Th1-type cytokines for their generation and have been found to produce a repertoire of antibodies that are autoreactive. Also, when depleting ABCs in murine models, the mice were protected from autoimmunity, suggesting a pathogenic role of ABCs in autoimmune diseases [35]. Of the cytokine-producing B cells, the most studied in SLE are the ones producing IL-10, oftentimes called regulatory B cells (Breg) [36]. Breg cells have been shown to be able to regulate autoimmunity and maintain immune homeostasis. Defects in the function or numbers of Bregs have been described in SLE and other autoimmune diseases ^[37][38][39][37,38,39]. This includes impaired production of IL-10 as well as recruitment of Breg cells in response to TLR9 stimulation of plasmacytoid (p)DCs. However, unlike Treg cells, B cells producing cytokines are thought to differentiate and eventually evolve to plasma cells. It is currently unknown what regulates their cytokine producing phase and how long the regulatory phase can be maintained. There are also no specific transcription factors that have been determined to keep B cells as cytokine producers or Bregs. Some Breg cells also express CD1d and interact with NKT cells. It has been shown that in SLE, NKT cells are dysfunctional and present in low numbers. However, the depletion of B cells with rituximab has been shown to reset the B cell populations to normal levels, including CD1d expression, and this, in turn, normalizes the NKT cell function and restores their numbers [40].

6. B Cell Directed Treatment in SLE

In recent years, research on the pathogenesis of SLE has made progress, advancing both diagnostics and investigations using histopathology to unravel cellular and molecular mechanisms underlying the disease. These findings have also resulted in progress regarding SLE drug therapeutics. Earlier SLE diagnosis has not only led to a better disease course in terms of disease activity, but also earlier and more suitable therapeutic interventions that limit organ damage. The systemic treatment for moderate to severe SLE includes steroids, antimalarial drugs, immuno-suppressants, and biological agents. Among new treatments, B cell-directed agents include biologics directly or indirectly targeting B cells, as well as immunomodulators such as synthetic dehydroepiandrosterone and hydroxychloroquine that suppress inflammation in a non-selective fashion.

6.1. Anti-Inflammatory and Chemotherapy Treatments and Their Impact on B Cells

Glucocorticoids have potent anti-inflammatory effects that also influence B cell activation. B cells express the glucocorticoid receptor and, thus, these treatments can have a cell intrinsic effect. Targeting this receptor affects several signalling molecules in B cells, including AP 1 and NF κB, which are transcription factors downstream the BCR [41]. Glucocorticoids can also directly mediate apoptosis of B cells. It has been shown that immature B cells are especially sensitive to glucocorticoids [42]. Another immunomodulatory drug that may be considered anti-inflammatory is hydroxychloroquine (HCQ), which inhibits the functions of nucleic acid-sensing TLRs. At the concentrations used in rheumatology, HCQ blocks TLR9 ligation, suppressing B cell differentiation into plasmablasts and, hence, IgG secretion [43]. Importantly, antimalarial agents are recommended for all patients with SLE, unless contraindicated, with HCQ being the drug of choice within this drug class, owing to, in relative terms, a more favourable safety profile [44]. Mycophenolate mofetil (MMF) is a derivative of mycophenolic acid, which is an inhibitor of inosine-5′-monophosphate dehydrogenase. Through depleting guanosine and deoxyguanosine nucleotides in T and B cells, MMF blocks lymphocyte proliferation/differentiation and the production of immunoglobins [45]. Cyclophosphamide (CYC) is an alkylating agent; its active metabolites are acrolein and phosphoramide mustard. The cytotoxicity of CYC is caused by DNA cross linking, ultimately resulting in apoptosis of cells undergoing division. When comparing the effects of CYC and MMF on treating SLE, it was found that both drugs influenced disease activity and B cell numbers, with CYC being more efficient in depleting naïve B cells and pre-switched memory B cells [46]. Moreover, drugs that primarily target T cells also exert indirect effects on B cell activation. One such drug is cyclosporine A (CsA), which is a calcineurin inhibitor and cytochrome P450 inhibitor. While the mechanism of action of CsA is mostly related to T cell suppression, e.g., by inhibiting IL-2 production by DCs, impeding T cell proliferation [47], CsA has also been shown to have effects on B cells and B cell migration ^[48][49][48,49]. CsA inhibits IL-2, IL-6, and IFNγ production by CD4+ T cells; especially through the inhibitory effects of IL-4 and IL-6, CsA indirectly inhibits the growth and differentiation of B cells, and the production of immunoglobulins: Refs. ^[50][51][50,51]. Another calcineurin inhibitor that is used in SLE, mainly in the context of lupus nephritis, is tacrolimus, which has tested in multiple trials with encouraging results, especially in Asian LN populations [52]. A reduction in the percentage of immature B cells was observed when PBMCs were incubated with tacrolimus in vitro [53]. The influence of tacrolimus to B cell maturation and antibody response was indirectly assisted by T cells ^[54][55][54,55]. While CsA and tacrolimus may be considered established drugs for the treatment of SLE and LN, voclosporin was recently approved for the treatment of active lupus nephritis, in combination with MMF, after a successful phase III randomized clinical trial. Voclosporin is a new generation, more potent, and more stable calcineurin inhibitor, whose metabolites are more quickly eliminated, adding ease of surveillance to the advantages of voclosporin over CsA and tacrolimus. Another drug with immunosuppressive effects is methotrexate (MTX), an antifolate antimetabolite that inhibits DNA synthesis. The resulting anti-inflammatory effect is mediated by an accumulation of adenosine that inhibits T cell activation and, in turn, B cells. In connection to autoimmunity, MTX and CYC combined as a treatment of arthritis in mice lead to changes in cellular numbers in lymph nodes and the spleen, including a decrease in Breg and DC numbers [56]. Lastly, azathioprine also constitutes a traditional non-targeted immunosuppressant that is widely used in SLE [44], acting after conversion to 6-mercaptopurine (6-MP), an immunosuppressant prodrug. Among multiple uses in SLE, azathioprine is used as a remission maintenance treatment for LN and is considered a safe drug during pregnancy.

6.2. B Cell Suppressing/Depleting Therapies

6.2.1. Anti-BAFF/APRIL

B lymphocyte stimulator (BLyS, also known as BAFF) is produced by antigen-presenting cells, neutrophils, activated T cells, and endothelial cells. BAFF has important properties for normal B cell differentiation, maturation, and antibody production ^[57][58][57,58]. BAFF provides essential signals for B cell activation and survival via the NF-κB and MAPK pathways, mediated by three receptors: the BAFF receptor (BAFF-R), transmembrane activator and calcium modulator cyclophilin ligand interactor (TACI), and B cell maturation antigen (BCMA). BAFF-R ligation provides key signals, while TACI and BCMA also bind APRIL [28]. In 2000, it was found that overactivation of BAFF may bypass T cell surveillance in negative selection. BAFF transgenic mice exerted lupus-like features, including proteinuria, nephritis, and high levels of anti-dsDNA autoantibodies, making this cytokine an attractive target for the treatment of SLE ^[59][60][59,60]. Blocking the BAFF/APRIL signal with a recombinant fusion protein harbouring TACI in SLE mouse models (MRL/lpr and NZB×NZW) ameliorated lupus-like phenotypes [61]. Also, in patients with SLE, there are higher BAFF and APRIL levels and the BAFF/BAFFR is correlated with disease activity [62]. All these findings have encouraged further investigations into the BAFF and APRIL antagonists seeking to treat SLE. In early phase I and II clinical trials, belimumab, a fully human monoclonal antibody targeting BAFF, was proven to be effective and displayed a favourable safety profile [63]. In two subsequent phase III, multicentre, randomized, placebo-controlled clinical trials (RCTs) of active SLE, a dosage of 10 mg/kg belimumab administered intravenously every fourth week plus standard therapy was superior to standard therapy in inducing responses, according to the SLE Responder Index 4 (SRI-4) criteria at week 52 in both trials (Table 1) ^[64][65][64,65]. A decrease of more than 50% and 43% circulating CD20+ B cells and plasma cells was observed in the belimumab-treated groups in these studies [66]. Upon proven efficacy in these two pivotal phase III RCTs, belimumab became the first biological drug to be approved by the EMA and FDA for SLE, after more than 50 years with no trial successes before. Pooled data analysis from the BLISS trials showed that hypocomplementemia and anti-dsDNA positivity, as well as high disease activity, were baseline predictors of response to belimumab [67]. Lupus nephritis (LN) affects up to 60% of SLE patients and is, along with cardiovascular disease, the most common reason for mortality in SLE. LN most commonly occurs early during the course of SLE and is often the manifestation that results in diagnosis [68]. The combination of belimumab and other immunosuppressants has shown promising results in treating some severe subtypes of LN [69]. In addition, the sub-group analysis of the BLISS trial showed improvements of LN, leading to BLISS-LN, a phase III multinational RCT that enrolled 448 patients with active LN (Table 1) [70]. At week 104, 43% of the patients who received belimumab (10 mg/kg) on top of standard induction therapy with glucocorticoids and MMF or CYC met the trial primary endpoint (primary efficacy renal response), compared with 32% in the group of patients who received standard induction therapy along with placebo, thus indicating the efficacy of belimumab in treating patients with active LN and resulting in its official approval by regulatory agencies also for this indication. In parallel with the development of the BAFF antagonist, a soluble BAFF receptor fusion protein, TACI-Ig, was also developed which neutralizes both BAFF and APRIL, known as atacicept. As BAFF and APRIL share two receptors, TACI and BCMA, and tend to form heterotrimers in circulation, both cytokines are targeted with this approach. As APRIL provides the signal for the survival of plasmablasts and plasma cells, atacicept exerts substantial neutralizing effects. In the 24-week multicentre, randomized, double-blinded, placebo-controlled phase IIb study of atacicept (ADDRESS II trial) with a total of 306 patients included, significantly more patients receiving atacicept 75 mg or 150 mg subcutaneously achieved an SRI-4 response at the endpoint compared with the placebo group (Table 1) [71]. The trial, however, did not meet the primary outcome. Recently, telitacicept, or RC18, which is another fusion protein comprising a recombinant TACI receptor, fused to the Fc domain of human IgG [72]. Patients in the telitacicept group exhibited a greater SRI-4 frequency at week 48 as a measure of improved disease. Telitacicept has now been approved by the National Medical Products Administration (MNPA) for the treatment of patients with SLE in China (Table 1) [73]. The Phase II clinical trial of telitacicept in the United States is ongoing.