Tweaking the B lymphocyte compartment in autoimmune diseases

Progress in understanding the multifaceted biology of B cells and their multiple contributions to autoimmunity is providing productive understanding of disease pathogenesis and is leading to novel strategies for therapeutic intervention.

The past years have brought numerous insights into the understanding of B cell biology. In turn, the discovery of previously unknown pathways is contributing to the delineation of the pathogenesis of autoimmune diseases. For its fifth meeting, the B cells and Autoimmunity Conference was held along the shores of the antibody-shaped Lake Como in northern Italy on 19–21 August 2013. This international conference brings together leading investigators in B cell biology and clinical immunologists interested in autoimmune diseases and provides a forum of discussion for assessing the effect of this expanding field on new approaches to disease intervention.

Store-operated Ca ²⁺ influx

The activation of lymphocytes requires Ca²⁺ influx, which results in expression of various genes encoding cytokines and chemokines and is mediated by specialized Ca²⁺ channels in the plasma membrane called 'Ca²⁺ release–activated Ca²⁺' (CRAC) channels¹. After binding of the antigen receptor, CRAC channels are activated by depletion of Ca²⁺ stores in the endoplasmic reticulum (ER) by a process called 'store-operated Ca²⁺ entry' (SOCE). The pore-forming subunit of the CRAC channel, Orai1, is activated by Stim1 and Stim2, which are single-transmembrane proteins located in the ER membrane. After calcium stores in the ER are depleted, Ca²⁺ dissociates from Stim1 and Stim2, which results in their translocation to the plasma membrane and the activation of Orai1 CRAC channels¹. Together, Orai1, Stim1 and Stim2 are critical for the function of CRAC channels and SOCE.

Tomohiro Kurosaki (Osaka University) generated mice with B cell–specific deletion of Stim1 and Stim2. The results showed that Stim1 and Stim2 are dispensable for B cell development, that cell proliferation mediated by the B cell antigen receptor (BCR) is dependent on SOCE, and that Stim-mediated SOCE is dispensable for antibody responses. Although B cell–specific deletion of Stim1 and Stim2 causes profound defects in BCR-induced SOCE and proliferation, B cell development and antibody responses are not affected². In addition, B cells lacking both Stim1 and Stim2 fail to produce the anti-inflammatory cytokine interleukin 10 (IL-10). Notably, the mutant mice exhibit exacerbation of the development and progression of experimental allergic encephalomyelitis (EAE), a mouse model of multiple sclerosis, and that severe EAE is ameliorated by the adoptive transfer of B cells transduced with Il10, which indicates that Stim-mediated production of IL-10 contributes to the function of regulatory B cells in vivo.

Reminiscent of those observations are the results of studies designed to probe the role of Ca²⁺ influx in T cell tolerance³. Combined T cell–specific deletion of Stim1 and Stim2 impairs the development and function of regulatory T cells and protects mice from EAE⁴. Mice with such T cell–specific deletion of Stim1 and Stim2 also develop spontaneous and progressive submandibular gland inflammation and damage comparable to that of primary Sjögren's syndrome. Notably, peripheral blood mononuclear cells from patients with that syndrome have a lower abundance of Stim1 and Stim2 and diminished SOCE, which suggests that deficiency in Stim1 and/or Stim2 may underlie the onset and progression of salivary gland lesions in Sjögren's syndrome.

Collectively, these studies suggest that Ca²⁺ signaling dependent on Orai and Stim proteins has a dominant role in immunotolerance. Specific cell functions regulated downstream of SOCE, such as cytokine production, are crucial for immunological homeostasis. Since most if not all influx of Ca²⁺ into lymphocytes is store operated and is dependent on the function of Stim1 and Stim2, delineating the mechanism(s) that modulate Stim expression may lead to the design of novel targeting strategies.

Tolerance 'decisions' in the periphery

In addition to signaling via the BCR, signaling through BAFF-R, receptor for B cell–activating factor (BAFF; also know as BLyS), is important in determining B cell fate. Mutually dependent relationships between the BCR and BAFF-R coordinate the conflicting needs of maintaining a balance between generating a repertoire devoid of potentially dangerous autoreactive B cells and preserving a repertoire of potentially protective B cells⁵. Signals from the BCR generate a rate-limiting substrate required for transcriptional regulatory systems that are activated by BAFF-R. In addition to responding to BAFF, B cells typically respond to antigens in the context of other factors, including ligands of Toll-like receptors (TLRs). In studies in collaboration with Ann Marshak-Rothstein (University of Massachusetts), Michael P. Cancro (University of Pennsylvania) described a previously unknown post-proliferative death response that occurs in B cells following stimulation with BCR-delivered TLR9 ligands⁶. This self-limiting activation is unique to physically linked ligands of the BCR and TLR9, as it is not observed following stimulation via the BCR or TLR9 alone or simultaneous but independent ligation of the BCR and TLR9. The death observed involves cleavage of caspase-9 and caspase-3 and is 'rescued' by overexpression of the antiapoptotic factor Bcl-x_L, consistent with intrinsic mitochondrial apoptosis. Moreover, it is mediated by the kinase p38, occurs after arrest in the G1 phase of the cell cycle and is common to preimmune mouse B cell subsets and naive human B cells. Survival signals delivered via BAFF or costimulation via the receptor CD40 rescue cells from death and afford survival and differentiation⁶. This unappreciated crosstalk between the signaling pathways of the BCR and those of TLR9 limits responses to antigens containing TLR9 ligands and may help to explain why circumventing this response-limiting system can lead to sustained autoantibody production.

Studies by Michael C. Carroll (Harvard Medical School) of the role of the complement system in B cell activation also shed light on self-tolerance mechanisms. Deficiency in the complement component C4 is a major susceptibility factor in human lupus and, on certain backgrounds, predisposes mice to lupus-like disease. To investigate the role of complement in B cell tolerance, C4-deficient mice were crossed with a BCR-knock-in mouse line in which B cells are specific for a lupus autoantigen. The mice develop a break of self-tolerance at the transitional stage, a loss of B cell anergy and a greater propensity to form self-reactive germinal centers. Since C4 deficiency results in impaired clearance of apoptotic cells, the phenotype observed could be due to type I interferon produced in response to apoptotic cells via the TLR7 pathway and increased BAFF production⁷. Consistently, blockade of the receptor for type I interferon restores anergy and B cell tolerance.

Pathways of autophagy

Autophagy is a degradation pathway that involves the engulfment of cytoplasmic contents and their delivery and lysosomal degradation. Basal levels of autophagy contribute to the physiological turnover of proteins and to the removal of altered and/or damaged organelles⁸. In addition, many immunological processes are very dependent on autophagy; these include the recognition and destruction of pathogens, antigen presentation, the development and function of lymphocytes, and inflammation. In B lymphocytes, autophagy promotes survival and development. Mice with B cell–specific deletion of Atg5, a protein with which cells degrade cytoplasmic constituents and organelles, have fewer peritoneal B-1a cells and a defect in early B cell development⁹. In addition, beclin-1, another key protein in autophagy, is required for the maintenance of undifferentiated-early lymphocyte progenitor populations¹⁰.

Because autophagy is a central integrator of stress responses, and plasma cells are subject to severe ER, oxidative and proteasomal stress, Simone Cenci (San Raffaele Scientific Institute) addressed the role of autophagy in the differentiation of plasma cells and humoral responses. He found higher expression of autophagy-related transcripts and considerable flux in autophagy during the differentiation of plasma cells, in support of the proposal of a role for autophagy in the development of plasma cells¹¹. Remarkably, autophagy exerts an unexpected negative control on the transcriptional repressor Blimp-1 and immunoglobulin production by directly restricting ER size and stress signaling and thereby supports energy metabolism and the viability of plasma cells. In vivo, autophagy sustains both T cell–dependent antibody responses and T cell–independent antibody responses and is needed to maintain the pool of long-lived plasma cells in the bone marrow. Moreover, whereas autophagy mediates the cross-presentation of phagocytosed antigens by dendritic cells to CD4⁺ T cells, Cenci has found that autophagy is dispensable for the presentation of soluble antigen by B cells to cognate T cells in the germinal center reaction in vivo. Overall, these studies show that autophagy serves a crucial adaptive role in plasma cells and is required for short-term and long-term humoral immunity.

In addition to its role in the differentiation of plasma cells, autophagy exerts beneficial effects on immunity and inflammation, as the following observations attest. In B cells, receptor engagement induces the rapid and extensive formation of autophagosomes, in which antigens are efficiently processed. In addition, ligation of the BCR recruits TLR9 to autophagosomes for further interaction with its target ligand¹². Since stimulation of the BCR in the absence of costimulation induces potent cell death by autophagy, autophagy seems to be linked to the establishment and maintenance of B cell tolerance to self, and autophagy induced by ligation of the BCR may be involved in the negative regulation of autoimmunity. That view indicates that impairment in the function of autophagy may contribute to the induction and/or perpetuation of autoimmunity (Fig. 1). Indeed, mounting evidence indicates that distorted functions of autophagy are linked to the pathogenesis of autoimmune disease. In genome-wide association studies, genetic variants in or near ATG5 are associated with systemic lupus erythematosus (SLE) in both Caucasian populations and Chinese populations. Direct investigation of autophagy in human lupus and mouse models has demonstrated enhanced autophagy in both, which supports the view that autophagy could promote the survival of autoreactive cells¹³.

Figure 1: Potential consequences of impairment in the function of autophagy.

Autophagy is important for several aspects of the development and differentiation of B cells. Impairment in the functions of autophagy could contribute to the induction and/or perpetuation of autoimmunity by various means.

Conceivably, the absence of proteins involved in autophagy could result in the defective clearance of apoptotic cells. Deregulation of autophagy might modulate susceptibility to autoimmunity and influence pathogenesis through several potential mechanisms. A defect in Atg5 could result in the abnormal clearance of dying cells, a hallmark of lupus, or in the increased secretion of proinflammatory cytokines. Serum obtained from patients with SLE, followed by complement inactivation, can substantially activate autophagy; moreover, lymphocytes from patients with SLE are resistant to the induction of autophagy and display upregulation of genes encoding molecules that negatively regulate autophagy, such as α-synuclein¹⁴. Those observations further support the proposal that unbalanced autophagy is linked to autoimmune disease. Along those lines, antibodies to citrullinated proteins (specific markers of rheumatoid arthritis present before the clinical onset of this disease) have been linked to autophagy¹⁵. Of interest, a large-scale replication study of rheumatoid arthritis has also associated susceptibility to this disease with the intergenic region between the genes encoding Blimp-1 and Atg5, which further suggests the involvement of Atg5 in autoimmunity.

To develop effective and more-specific therapeutic strategies, it will be essential to understand the complex interplay between autophagy and autoimmunity. In autoimmune diseases such as SLE, the elimination of long-lived autoreactive plasma cells, which are protected in survival niches and are refractory to immunosuppression and the depletion of B cells, remains an important therapeutic challenge. Delineating the pathways that ensure protein homeostasis and underlie the survival of plasma cells may offer new strategies for inducing the depletion of plasma cells. Another avenue would be the development of therapeutic strategies aimed at recovering lymphocyte homeostasis by restoring susceptibility to autophagy. Drugs able to induce autophagy (for example, rapamycin and its analogs, such as everolimus) could lead in the long run to new therapeutic avenues for autoimmune diseases.

Therapeutic perspectives

Multiple sclerosis is a demyelinating autoimmune disease of the central nervous system with heterogeneous pathophysiological and clinical manifestations and an unclear etiology. Despite strong evidence for the involvement of T cells in multiple sclerosis in humans and in EAE, the only cell-specific antibody that has proven to be highly specific in relapsing-remitting multiple sclerosis acts on B cells. The identification of a major immunogenic region in myelin basic protein (MBP) that can induce tolerance in B cells to this neuroantigen would offer the prospect of progress in the development of novel approaches for the diagnosis and treatment of multiple sclerosis. Alexander G. Gabibov (Russian Academy of Sciences) has constructed an MBP-derived recombinant epitope library that covers the entire molecule and has identified two MBP fragments recognized 'preferentially' by serum from patients with progressive or relapsing-remitting multiple sclerosis. To determine the immunomodulatory potential of the epitopes selected and to identify appropriate biomarkers for subsequent analysis, Gabibov assessed the spectrum of anti-MBP autoantibodies in EAE models¹⁶. Remarkably, selected immunodominant MBP peptides encapsulated in mannosylated small unilamellar vesicles reduced the maximal disease score during the first attack of EAE and even prevented disease exacerbation. In addition to reducing the median cumulative disease score, the encapsulated peptides prevented the production of circulating autoantibodies, downregulated the synthesis of T helper type 1 cytokines and induced the production, in the central nervous system, of a brain-derived neurotrophic factor that promotes the survival of existing neurons and the growth and differentiation of new neurons and synapses.

In healthy subjects, insulin-reactive B cells are 'censored' in the bone marrow. In type 1 diabetes, they are critical for disease development but must compete with non-autoreactive B cells for survival factors and entry into follicular niches. Studies of mice of the nonobese diabetic strain and clinical trials based on the depletion of B cells have identified B cells as a promising therapeutic target for the prevention and reversal of type 1 diabetes. To develop a strategy that enables restoration of tolerance to islet antigens in the pancreas, James W. Thomas (Vanderbilt University) is using transgenic mice that express a polyclonal repertoire in which only 1–2% of mature B cells recognize insulin, which thereby models the selection events of polyclonal conditions¹⁷. The administration of an antibody to insulin selectively eliminates insulin-reactive B cells in vivo and prevents disease in mice. The fact that substantial protection from disease is achieved by a monoclonal antibody to insulin in a polyclonal repertoire in which the frequency of anti-insulin B cells in the developing repertoire is very low adds to the growing body of evidence suggesting that insulin is a critical autoantigen in type 1 diabetes and represents a potential target for the restoration of immunotolerance.

Predicting the success of therapeutics

Autoimmune diseases arise in people genetically predisposed to such diseases and require contributions from epigenetic and environmental factors. Since the underlying molecular mechanisms are not completely understood, current therapies are directed to target pathways, but the beneficial effects remain unsatisfactory. In addition, markers that can be used to select patients who will respond to the treatment are lacking. Lupus, for example, exhibits a high degree of heterogeneity. An antibody to BAFF is the only biological treatment for lupus approved by the US Centers for Disease Control at present, but baseline concentrations of BAFF in the serum cannot be used to predict which patients would respond to the treatment¹⁸. Similarly, in rheumatoid arthritis, Crohn's disease and psoriasis, blockade of TNF is beneficial in only approximately two thirds of patients¹⁹. Therefore, tailoring the treatment course to a patient's own underlying autoimmune physiopathology would be effective for predicting the success or toxicity of therapeutics. In a strategy based on the fact that the receptor for the B cell–specific surface antigen CD20 is expressed by B cells but absent on plasmablasts, a biomarker is showing promise in B cell–depletion therapy in rheumatoid arthritis. Thus, high expression of IgJ, a marker for antibody-secreting plasmablasts, is predictive of unresponsiveness to antibodies that effect their depletion via CD20 (ref. 20).

Another drawback to biotherapies is that the administration of a bioproduct can trigger an immune response to the injected protein. Such anti-drug immune responses may lead to inflammatory side effects or to neutralization of the drug. A major goal in the development of biopharmaceutical drugs is therefore to predict which patients are likely to develop anti-drug responses. William Sanderson and Claudia Mauri (University College London) are using a new immunophenotyping platform that allows screening of the expression of 332 cell-surface markers in parallel by flow cytometry. This has enabled the generation of a cell-surface-marker 'signature' for different cell subsets of the immune system and has provided a phenotypic counterpart to gene-array data. Comparison of B cell–subset signatures in a cross-sectional cohort of patients with rheumatoid arthritis treated with antibody to TNF could lead to the identification of predictive markers for those patients who will develop a response to the injected biodrug. Since autoimmune diseases involve pathological alterations in several molecular pathways, it will also be important to explore potentially synergistic combinations of therapies. This would enable the blockade of key pathways involved in the pathogenesis of these complex diseases. Here too, the selection of effective combination therapies should optimally be based on profiling of individual patients.

As briefly summarized above, the inspiring atmosphere of Como, the birthplace of the father of the voltaic battery (Alessandro Volta), catalyzed vibrant discussions that further revealed that the multifaceted biology and physiology of B cells is much more complex than initially conceived. In parallel, the wealth of emerging strategies that target B cells now sets the stage for an exciting new era in the treatment and prevention of a variety of autoimmune diseases.