IL-6 is the prototypical member of the IL-6 family, characterized by common usage of the signal transducing transmembrane glycoprotein 130 (gp130 or IL-6Rβ) (12, 13). The family also includes IL-11, ciliary neurotrophic factor (CNTF), oncostatin M (OSM), leukemia inhibitory factor (LIF), cardiotrophin 1 (CT-1) and cardiotrophin-like cytokine factor 1 (CLCF1) (14). Members of the IL-6 family are pleiotropic cytokines produced by a wide variety of immune as well as non-immune cells and they play crucial roles in a variety of autoimmune diseases. In addition to their roles in hematopoiesis and maintenance of immune homeostasis, they regulate diverse cellular process that affect embryonic development, cognition, intermediary metabolism and ageing. They signal through a non-signaling alpha (α) receptor subunit (IL-6R, CNTFR, LIFR, OSMR or IL-11R) which binds the cognate cytokine and the cytokine:α-receptor complex then interacts with gp130 (IL-6Rβ) to form the high-affinity homodimer receptor complex (15). Signaling through gp130 homodimer results in the phosphorylation of receptor-associated Janus kinases (JAK1, JAK2) that preferentially recruit STAT3 to the cytokine receptor complex for activation via tyrosine phosphorylation (16) ( Figure 1 ). Another important feature of IL-6 family cytokines is their propensity to generate soluble receptors that derive from shedding extracellular portion of their α or β receptors through limited proteolysis by metalloproteinases such as ADAM17 (a disintegrin and metalloproteinase domain-containing protein 17) and ADAM10 (17, 18). The soluble α receptor (e.g. sIL6R) retains capacity to bind cognate cytokine (e.g. IL-6) in vivo to form receptor-ligand complexes that interact with membrane-bound gp130 to activate cells. Thus far three distinct forms of IL-6-mediated signaling have been described: (i) The classic IL-6 signaling where IL-6 binds membrane-bound IL-6R (mIL-6R), initiates signaling by interacting with gp130 homodimers. This results in activation of JAK1/JAK2, followed by tyrosine-phosphorylation of STAT3 and subsequent translocation of pSTAT3 homodimers into the nucleus; (ii) IL-6 trans-signaling is suggested to broaden the variety of cells responsive to IL-6, IL-11 or possibly CNTF. In IL-6 trans-signaling, IL-6 and sIL-6R complexes initiate signal transduction by recruiting and binding membrane-bound gp130 (mgp130); (iii) IL-6 trans-presentation or cluster signaling is another receptor engagement mechanism where IL-6-mIL6R complexes on a cell activate gp130-gp130 homodimers on a neighboring cell, resulting in enhanced lymphocyte activation or activation of target cells that lack specific receptors for IL-6 cytokines. Suppressor of cytokine signaling-1 (SOCS1) and SOCS3 are important negative-feedback regulators of IL-6 family cytokines that prevent excessive cytokine activation that can lead to tissue damage and disease. SOCS3 functions by targeting gp130 receptor, JAK homology domains and STATs for ubiquitin-mediated degradation in the proteasome (19, 20) ( Figures 1 , 2 ).

With regards to the role of IL-6 in human diseases, it is notable that blood levels of IL-6 in a healthy human is 1-5 pg/ml and can increase to 1000 folds in some inflammatory conditions and up to several µg/ml during sepsis (21). Pre-clinical studies have also implicated IL-6 in impaired immune responses, arthritis, colitis, and multiple sclerosis and activation of mucosal T cells induced by IL-6-sIL-6R exacerbates Crohn disease, a chronic inflammatory disease of the gastrointestinal tract chronic intestinal inflammation (22–26). Thus, elevated level of IL-6 is pathognomonic sign of inflammation and drugs like Tocilizumab (Actemra) that target IL-6 and/or IL-6R are effective and approved for the treatment of autoimmune diseases including rheumatoid arthritis (27). Therapeutic targeting of components of the cytokine receptor signaling complex utilized by IL-6-like cytokines including gp130 or the proinflammatory IL-6 trans-signaling pathway are also effective and represent new avenues for drug discovery. Thus, treatments approved or undergoing clinical trials in the USA include cytokine-blocking antibodies (Abs), receptor-blocking antibodies, agonist cytokine-antibody complexes, small-molecule Jak inhibitors (Jakinibs), STAT-binding inhibitory peptides (8), small-molecule STAT inhibitors (28). Although JAK inhibitors are in use for the treatment of rheumatoid arthritis and other inflammatory diseases (29, 30), it must be emphasized that JAKs do not selectively inhibit IL-6 signaling pathway and off-target effects may interfere with other cytokines or growth factors that utilize JAK/STAT pathway (28). Nonetheless, because IL-6 cytokines predominantly utilize STAT3 pathway and the pathogenic Th17 cells that mediate CNS autoimmune diseases require STAT3 (31, 32), there is significant interest in developing therapies that target STAT3 pathway. These emerging therapies include SOCS mimetics and nanobodies that suppress disease in mouse models of uveitis and multiple sclerosis (33–35).

Similar to the relatively low amount of IL-6 in human blood that increases in disease state (1-5 pg/ml), the level of soluble IL-6 receptor (sIL-6R) is around 40-75 ng/ml and can increase by 2-10 folds during inflammatory diseases. In contrast, sgp130 levels in healthy individuals is high (~250-400 ng/ml), remains unaltered during inflammation and this has led to the suggestion that endogenous sgp130 serves to restrain or buffer systemic IL-6 trans-signaling activities (21, 36). These and other experimental findings indicate that IL-6 trans-signaling is responsible for the proinflammatory activities of IL-6 while the classic signaling mechanism mediated by direct interaction of IL-6 and mIL6R promotes anti-inflammatory activities of IL-6. In this context, the monoclonal antibody, Olamkicept (soluble gp130Fc or sgp130Fc), that specifically inhibits IL-6 trans-signaling without interfering with classic IL-6 signaling, is effective treatment for inflammatory bowel disease (IBD). Its clinical effectiveness correlates with suppression of STAT3 phosphorylation and transcriptional changes in the inflamed mucosa (37, 38). Compared to other antibody treatments, very low amount of sgp130 is required to prevent fatal sepsis in mice (39), underscoring the merit of blocking of IL-6 trans-signaling. Olamkicept is well tolerated and clinically desirable because of practicality of limiting the pharmacological dosing regimen.

IL-12 was identified and purified from supernatant fluid of EBV transformed B lymphoblastoid cell line in 1989 (40, 41). The 70kDa heterodimeric glycoprotein, comprised of IL-12p40 and IL-12p35, was initially designated as NK cell stimulatory factor (NKDSF). In 2000, the IL-12p19 protein was identified on the basis of its homology with IL−6 and was demonstrated to dimerized with the IL-12p40 subunit to form IL-23 (48). Both cytokines were subsequently found to exhibit similar as well as distinct biological activities (49). However, mice deficient in IL-12p40 deficient mice were subsequently shown to be resistant to several mouse autoimmune diseases including experimental autoimmune encephalomyelitis (EAE), experimental autoimmune uveitis (EAU) and T cell mediated colitis (48). These findings prompted re-evaluation of functions previously attributed to IL−12 based on IL-12 neutralization experiments, deletion of IL-12p40 or its receptor (IL−12Rβ1). These investigations revealed that IL−23-driven Th17 cells, rather than the Th1 subset, mediates autoimmune or infectious diseases previously attributed to IL-12-induced Th1 subset (48). Although EAU was also thought to be a Th1-mediated disease (50), mice with targeted deletion of STAT3 were found to be resistant to this CNS autoimmune disease of the retina. Although these mice with loss of STAT3 in CD4⁺ T cells also lack Th17 cells they exhibit exaggerate expansion of Th1 cells, providing direct evidence that IFN-γ produced by Th1 cells are protective while the IL-17 producing Th17 cells are responsible for inducing ocular pathology in the EAU model (31, 51).

Ebi3 was identified in 1996 from a subtractive hybridization screen of genes expressed in Epstein-Barr virus (EBV) transformed B cell lines (52). It is structurally homologous to IL-12p40 and contains no membrane anchoring motifs, suggesting that it is a secreted protein (52). Given the homology of Ebi3 to IL-12p40, IL-12p35 was posited as a potential pairing partner for Ebi3 and this was confirmed by co-expression studies identifying the IL-12p35:Ebi3 heterodimer as IL-35 (53, 54). It is also of note that the Ebi3 protein has 3 methionine and 4 cysteine residues while IL-12p35 has 10 methionine and 7 cysteine residues and this may account for the interaction between the two subunits. Similar to IL-35, discovery of IL-27 derived from in silico and biochemical approaches using genomic databases of α-helical cytokines of the IL-6 family. These studies led to the identification of IL-27p28 (also called IL-30) and its pairing with EbI3 to form the heterodimeric IL-27 cytokine (55). The preferential expression of IL-27 and IL-35 by activated lymphocytes and myeloid cells (DC, monocytes and macrophages), underscore the broad roles these cytokines play in regulating immunity. Despite the significant therapeutic interest in IL-27 and IL-35, mechanisms that regulate production of IL-27 or IL-35 cytokine is not well understood. Unlike proinflammatory members of the IL-12 family cytokines (IL-12 and IL-23) that are secreted and function as disulfide-linked heterodimers, IL-27 and IL-35 are non-covalently linked IL-27p28:Ebi3:IL-12p35:Ebi3 heterodimers, respectively. Important unresolved issues relating to the immunobiology of IL-27 and IL-35 include physiological cues that initiate the association of their two subunit proteins, how stability of the heterodimeric cytokine is maintained and mechanism that regulate bioavailability of the native IL-27 or IL-35 under physiological conditions. Both cytokines share the common Ebi3 subunit which encodes a single signal-peptide that is cleaved concomitant with translocation into the lumen of the endoplasmic reticulum, suggesting that it is a secreted protein. In contrast, IL-12p35, in the mouse species, contains two tandem signal-peptides and do not conform to the classic co-translational model of signal-peptide removal. Thus, it has been predicted that IL-12p35 is retained in the endoplasmic reticulum or distal Golgi as a transmembrane protein (56, 57).

The central nervous system (CNS) comprises of immune privilege tissues that include the brain, spinal cord and retina and these CNS tissues are sequestrated from the peripheral immune system by the blood-brain-barrier (BBB), blood-retina barrier (BRB) and the neurovascular unit (NVU) (10, 11). The immune system of the CNS is unique and exquisitely regulated to confer adequate protection from pathogens while calibrating intensity of the response to avoid collateral damage of the terminally differentiated neurons required for cognition and vision (58). While much is known about the roles played by Th1 and Th17 cells in fueling the vicious cycles of relapsing-remitting inflammation that characterizes multiple sclerosis (59) and the sight-threatening intraocular inflammatory diseases or uveitis (51, 60), less is known about the protective roles B cells play. Here we discuss the roles B cells play in suppressing CNS autoimmune diseases with a particular focus on IL-27-producing and IL-35-producing regulatory B cells (Bregs) and IL-35 containing exosomes (i35-exosomes) and IL-27-exosomes (i27-exosomes) that they secrete.

Studying the role lymphocytes play in autoimmune diseases led to discovery of IL-35-producing Breg cells (i35-Bregs) that suppress neuroinflammatory diseases and also IL-35-producing Treg cells that maintain immune tolerance and promote tumor T cell exhaustion (47, 61–64). These observations spurred interest in developing i35-Breg immunotherapy for CNS autoimmune diseases (47). Despite the role of IL-35 in suppressing inflammatory diseases and inhibiting cytotoxic CD8⁺ anti-tumor T cells that prevent tumor metastasis (47, 61, 64), gaps in mechanistic understanding of how IL-35 mediates immunosuppression have prevented its therapeutic use. On the other hand, recent reports have suggested that IL-12p35 and Ebi3 are not necessarily secreted as heterodimeric IL-35 cytokine, but that these IL-12p35 and Ebi3 subunits are independent anti-inflammatory cytokines (45, 46, 65). This has raised interest in understanding the immunobiology of the i35-Breg cell and the mechanisms that underlie its immunosuppressive activities.

Thus far, there is no unique marker or set of markers that exclusively identify the i35-Breg cell and designation of a i35-Breg cell is mainly based on production of IL-35 as well as the capacity to suppress inflammation either in vivo or in vitro. Although the capacity to produce immune-suppressive cytokines by i35-Bregs is firmly established, mechanisms that mediate IL-35 production by B cells is not well understood. Interferon regulatory factor-4 (IRF-4) and IRF-8 are expressed in developing B cells and play critical roles in B cell development and effector functions (66, 67). Neither IRF-4 nor IRF-8 binds DNA strongly, and they depend on the basic leucine zipper transcription factor ATF-like (BATF) to recruit them to promoter sites of genes they regulate (66). These crippled transcription factors bind competitively to BATF and concentration-dependent competition between them control cell-fate decisions of the differentiating B cell (67, 68). In activated B cells, BATF dimerizes with IRF-4, resulting in the recruitment of BATF-IRF-4 complex to AP1-IRF-composite elements (AICEs) of the Il12a locus, suggesting the involvement of BATF-IRF-4-AICE complex in regulating the IL-35 transcriptional programs of Breg cells (69). In fact, Irf4 ^fl/flCd19^+/Cre B cells with loss of IRF-4 in B cells are defective in generating i35-Bregs, underscoring the role of IRF-4 in skewing of B cells toward the i35-Breg phenotype. Therapeutic importance of i35-Bregs is suggested by pre-clinical studies showing that 500,000 ex-vivo generated i35-Bregs is sufficient to suppress uveitis or encephalomyelitis in the EAU or EAE mouse model, respectively (47, 70). Most importantly, the i35-Breg cells were administered 4 days after EAU or EAE induction. Mitigation of disease in mice with ongoing disease indicates that i35-Breg immunotherapy can be used to treat individuals with established autoimmune disease.

An innate-like IL-27 producing population of natural regulatory B-1a cell (i27-Breg) was recently identified in the peritoneal cavity and human umbilical cord blood (8). Most Breg cells, including the i35-Breg cells, derive from the B-2 lymphocyte lineage which are the most abundant B cells in mammals that develop in the bone marrow and reside mainly in the spleen. In contrast, B-1 lymphocytes are the earliest B cells to arise during development from the fetal yoke sac and they reside mainly in the peritoneal cavity and cord blood (71). Similar to the role played by BATF and IRF transcription factors in the transcriptional program of i35-Breg cells, BATF dimerizes with IRF-8, resulting in the recruitment of BATF-IRF-8 complex to AICEs of il27p28 locus, underscoring involvement of BATF-IRF-8-AICE complex in regulating i27-Breg transcriptional program (8). Irf8 ^fl/flCd19^+/Cre mice that do not express IRF-8 are defective in generating i27-Bregs, indicating that IRF-8 may skew B cells toward the i27-Breg phenotype. Similar to i35-Breg cells, adoptive transfer of 500,000 i27-Breg cells to mice with EAU or EAE suppressed and ameliorated uveitis and encephalomyelitis respectively, in pre-clinical studies (8).

The i27-Bregs and i35-Bregs secrete exosomes that contain IL-27 (i27-Exosomes) or IL-35 (i35-Exosomes), respectively (70, 72). These nanosized extracellular vesicles range from 50 nm to 150 nm as determined by electron microscopy. Importantly, their small sizes allow them to cross the blood-brain-barrier or blood-retina-barrier, making them ideal vehicles for therapeutic delivery of immunosuppressive cytokines like IL-27 or IL-35 into the CNS. Besides B cell-derived exosomes, Treg cells that produce IL-35 (iT_R35) also secrete IL-35-coated extracellular vesicles and they promote infectious tolerance (73). In preclinical studies, ex-vivo generated i27-Exosomes or i35 Exosomes were evaluated as potential stand-alone therapy for uveitis in the EAU model. The i27-Exosomes or i35-Exosomes ameliorated uveitis by suppressing expansion of pathogenic Th17 cells while inducing expansion regulatory lymphocytes that produce immunosuppressive cytokines including IL-10, IL-27 and IL-35 (70). Importantly, these exosomes also propagate infectious-tolerance mechanisms that suppressed uveitis by upregulating the expression of checkpoint proteins on the plasma membrane of autoreactive T-cells. These proteins that are associated with T-cell exhaustion or anergy include PD-1 (programmed cell death protein 1), LAG-3 (lymphocyte-activation gene 3), CTLA-4 (cytotoxic T-lymphocyte associated protein).

CNS diseases result from dysregulation of intricate biochemical pathways activated and utilized by the myriads of neurons, microglia and other myeloid cell types that mediate critical functions of the brain, spinal cord and neuroretina (58). Cytokines, growth factors, hormones and neurotransmitters that regulate and coordinate the activities of these diverse cell types in the CNS are in turn under stringent regulation by transcription factors that control the activation or repression of genes involved physiological processes of CNS tissues. Thus, a major challenge in the treatment of CNS diseases derives from the fact that transcription factors are undruggable because they are intracellular proteins, and they are not amenable to antibody therapy. This is particularly relevant in context of CNS autoimmune diseases such as uveitis and multiple sclerosis. The transcription factor STAT3 is implicated in uveitis because mice with loss of STAT3 in CD4⁺ T cells do not develop EAU (31). In multiple sclerosis, IFN-γ-producing Th1 cells induce ascending paralysis and promote recruitment of inflammatory cells into the thoracolumbar spinal cord. On the other hand, IL-17-producing Th17 cells mediate atypical MS, characterized by cerebellar ataxia (74–76). Although Jak inhibitors are effective and approved by the FDA, toxicity considerations limit their use (28–30).

Camelids (alpaca, llama, camel, vicuna and guanaco) produce non-canonical immunoglobulins comprised of a single variable heavy chain (VHH) and no light chain (77). In contrast to conventional antibodies which are 150kDa (~15nm), VHH miniature antibodies (also called Nanobodies) are only 15kDa (~2.5nm). Therapeutic interest in Nanobodies in CNS autoimmune diseases derives from the fact that they readily enter cells by transcytosis and into the CNS after inflammation-induced break-down of blood-brain-barrier or blood-retinal-barrier (BRB) (77). A novel Nanobody (SBT-100) was recently produced by immunization of Dromedary camels (Arabian camel) with human STAT3. Extensive characterization of SBT-100 revealed that it is specific to the highly conserved SH2 domain shared by all STAT proteins. In the EAU model SBT-100 inhibited the expansion of pathogenic Th17 cells and suppresses uveitis in mice (35). Because STAT1 and STAT3 are required for the development of naïve T cells into Th1 and Th17 lymphocyte, pre-clinical studies examined whether simultaneous inhibition of STAT1 and STAT3 with SBT-100 would be effective therapy for multiple sclerosis. The SBT-100 Nanobody prevented development of encephalomyelitis by inhibiting the recruitment of CD4⁺ T cells into the brain and spinal cord and diminished the capacity of encephalitogenic T cells to transfer EAE to naïve syngeneic mice (78). The success of SBT-100 in suppressing uveitis or encephalomyelitis in mice provides a template for use of SBT-100 Nanobody in modulating undruggable transcription factors such as STAT1 and STAT3 (35, 78). Camelid-derived Nanobodies are not immunogenic in humans because of their high sequence homology with human antibodies and the small size facilitates transport across the BBB and BRB during inflammation providing important advantages for therapeutic use in CNS diseases.