Th eJournal of Brief Reviews Immunology

B10 Cells: A Functionally Defined Regulatory Subset Thomas F. Tedder B cells are commonly thought to enhance inflamma- particularly those that influence immune responses through tory immune responses. However, specific regulatory mechanistically distinct pathways. In many cases, regulatory B cell subsets recently were identified that downregu- B cells inhibit inflammation and disease through the pro- late adaptive and innate immunity, inflammation, and duction of either IL-10 alone or in combination with other autoimmunity through diverse molecular mechanisms. immunomodulatory cytokines or cell surface receptors. However, In both mice and humans, a rare, but specific, subset of IL-10 expression by B cells is not always necessary to suppress regulatory B cells is functionally characterized by its inflammation, and multiple regulatory B cell subsets and capacity to produce IL-10, a potent inhibitory cyto- mechanisms for reducing inflammation are likely to exist. The kine. For clarity, this regulatory B cell subset has been complexity of this issue was well demonstrated in studies of IBD. Mizoguchi et al. (24) first showed that regulatory B cells labeled as B10 cells, because their ability to downregu- 2/2 late immune responses and inflammatory disease is from the mesenteric lymph nodes of TCRa mice and their Ab products suppress colitis by enhancing the clearance of fully attributable to IL-10, and their absence or loss apoptotic cells. Ab blockade of the CD40 or B7-2 costimu- exacerbates disease symptoms in mouse models. This latory molecules on adoptively transferred B cells also elimi- review preferentially focuses on what is known about nates their suppressive regulatory effects on pathogenic T cells mouse B10 cell development, phenotype, and effector (1). This group subsequently showed that IL-10 production function, as well as on mechanistic studies that demon- by CD1dhigh B cells within the mesenteric lymph nodes of 2 2 strated their functional importance during inflamma- TCRa / mice inhibits chronic intestinal inflammation (2). tion, , and immune responses. The However, they also demonstrated a regulatory role for perito- 2 2 Journal of Immunology, 2015, 194: 1395–1401. neal cavity B1 cells from TCRa / mice during chronic colitis, possibly through natural Ab generation in response to micro- cells are generally thought to augment immune bial flora (25). In studies by other investigators, mesenteric responses. However, B cells also can suppress immune lymph node B cells in combination with CD8+ T cells protect 2 2 B responses in a variety of mouse models of autoim- mice from colitis induced by Gai2 / CD4+ T cells by in- munity and inflammation. Mizoguchi et al. (1, 2) first coined stigating the formation of immunosuppressive T cells (26). the term regulatory B cell to describe B cells that suppress Whether these regulatory activities and mechanisms of action disease in a mouse inflammatory bowel disease (IBD) model. are attributable to a single B cell subset or functionally diverse Similarly, the absence or loss of B cells with regulatory ac- B cell populations during IBD is unresolved. tivities exacerbates disease symptoms in models of experi- Wolf et al. (3) first demonstrated a requirement for B cells mental autoimmune encephalomyelitis (EAE) (3–7), type 1 during recovery from EAE, with mice rarely returning to diabetes (8), collagen-induced arthritis (9–11), contact hy- normal in the absence of B cells. Dittel and colleagues (27) persensitivity (12, 13), (14–16), and allergy (17). The subsequently demonstrated that B cell deficiency and the identification and evolving characterization of regulatory B cells absence of B7 expression delayed the emergence of Foxp3+ and their potential mechanisms of action were comprehensively regulatory T cells (Tregs) and IL-10 in the CNS during EAE. reviewed elsewhere (18–22). Therefore, this review specifically However, most recently, they demonstrated a novel IL-10–, focuses on mouse regulatory B cells that produce IL-10, a po- B7-, and MHC class II–independent regulatory role for tent inhibitory cytokine with pleiotropic activities in vitro and B cells in suppressing autoimmunity by the maintenance of in vivo, whereas it is appreciated that monocytes, subsets, Tregs via GITR ligand (28). In contrast, B cell IL-10 pro- and other cells also can produce IL-10 (23). duction dramatically modulated EAE resolution in a separate study (4). IL-10–competent B cells control EAE initiation (6), Regulatory B cells whereas Tregs and IL-10–competent B cells independently Because the existence of regulatory B cells is well documented, influence EAE resolution (7, 29, 30). Thus, functionally the current challenge is to comprehend the diversity of B cell distinct B cell subsets may regulate autoimmune responses at subsets with regulatory function that have been described, different times.

Department of Immunology, Duke University Medical Center, Durham, NC 27710 Abbreviations used in this article: B10pro, progenitor B10 cell; EAE, experimental autoimmune encephalomyelitis; IBD, inflammatory bowel disease; PIM, PMA, iono- Received for publication May 23, 2014. Accepted for publication November 19, 2014. mycin, and monensin; TIM-1, T cell Ig domain and mucin domain protein 1; Treg, This work was supported by the Lymphoma Research Foundation. Foxp3+ .

Address correspondence and reprint requests to Dr. Thomas F. Tedder, Box 3010, Ó Department of Immunology, Room 353 Jones Building, Research Drive, Duke Univer- Copyright 2015 by The American Association of Immunologists, Inc. 0022-1767/15/$25.00 sity Medical Center, Durham, NC 27710. E-mail address: [email protected]

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1401329 1396 BRIEF REVIEWS: MOUSE REGULATORY B10 CELLS

The reality that diverse molecular mechanisms and cell plasmic IL-10 (31). Reliable B10pro cell numbers are difficult subsets contribute to regulatory B cell suppression of disease to extrapolate from .48-h cultures because some B cells has created confusion, which is exacerbated by the different proliferate, whereas dead cell frequencies also increase signifi- experimental approaches, phenotypes, and assays used to de- cantly. LPS similarly induces B10pro cell functional matura- scribe these cells. In particular, most studies fractionated B cells tion ex vivo, but it also induces B10 cell IL-10 secretion, which into subsets using phenotypic markers and then characterized reduces subsequent cytoplasmic IL-10 staining. B10+B10pro these B cell subsets based on their level of regulatory activity in cells normally represent 3–8% of spleen B cells (31). Al- different disease models (21). An alternative approach was though B10 cell numbers are dynamic during inflammation used that specifically characterizes regulatory B cell subsets and autoimmunity (6, 7, 13, 15, 31), B10pro cell numbers based on their functional capacity. In particular, we focused are relatively stable. on characterizing the B cells that selectively produce IL-10, B10 effector cells actively secrete IL-10 (Fig. 1A). However, regardless of their cell surface phenotype or tissue location. the intrinsic instability of il10 mRNA and transient low-level For clarity, this IL-10–producing regulatory B cell subset was IL-10 protein expression hinder their enumeration. To cir- labeled as B10 cells to highlight that IL-10 predominantly cumvent this, IL-10 reporter mice were engineered (36). Tiger accounts for their regulatory function during inflammation, mice have an IRES-GFP element inserted following the il10 autoimmune disease, and other adaptive and innate immune locus (37), allowing for enhanced IL-10 visualization due to responses (6, 7, 13, 15, 31–33), as well as to distinguish them the longer intracellular half-life of GFP versus IL-10 (38). from other regulatory B cell subsets that function through GFP+ B cells represent ,0.2% of total B cells in naive mice, different molecular pathways. This review highlights our but GFP expression mimics cytoplasmic IL-10 induction current understanding of mouse B10 cell function with following PIM stimulation ex vivo. 10BiT mice have multiple remarks on tolerance regulation and therapeutic potential. bacterial artificial chromosome transgene insertions that drive Thy1.1 expression under the control of il10 regulatory ele- B10 cell identification by function ments (39). Cell surface Thy1.1 expression is not measurable Currently, there are no phenotypic, transcription factor, or on B cells from naive mice and is delayed relative to IL-10 lineage markers that are unique to B10 cells. Therefore, B10 expression following ex vivo PIM stimulation, but it persists cells in mice and humans are functionally defined and enu- on the cell surface following the termination of IL-10 ex- merated by their competence to express measurable IL-10 pression, allowing for the surveillance of B10 effector cells following 5 h of ex vivo stimulation with PMA and ion- post–IL-10 production (38). B10 cell frequencies increase omycin (13, 32, 34). This combination of stimuli induces and 2–4-fold in Tiger and 10BiT mice following in vivo stimula- enhances translation of cytokine genes that are in tion (38). Only 1% of peritoneal cavity B cells from Tiger mice an accessible configuration for transcription, resulting in express GFP ex vivo, whereas up to 3% express GFP following measurable IL-10 protein production and secretion by B10 intestinal injury, demonstrating that gut inflammation induces effector cells (Fig. 1A). Monensin is included in these cultures B10 effector cells (35, 40). IL-10 reporter B-Green mice also to enhance cytoplasmic IL-10 staining by blocking protein have GFP sequences inserted into the il10 gene downstream of + transport. TLR agonists and other signals discussed later also the translation stop site (41). However, GFP B cells are only can induce il10 gene transcription. However, adding LPS or measurable in B-Green mice following Salmonella infection. CpG oligonucleotides to the 5-h PMA, ionomycin, and mon- Vert-X mice carry an IRES-GFP reporter inserted between the ensin (PIM) cultures only modestly enhances B10 cell enu- endogenous il10 gene stop codon and the poly(A) attachment meration (31). PIM cultures should be limited to 5 h because signal sequence (42). In naive Vert-X mice, 1.1% of spleen B cell viability decreases dramatically thereafter. Background B cells express GFP ex vivo, whereas 14% express GFP after staining of dead cells confounds accurate B10 cell enumera- LPS exposure. Thus, B10 effector cells are found at low fre- tion, so cell viability staining must be assessed in these func- quencies in mouse models of intestinal injury or infection 2 2 tional assays. B cells from IL-10 / mice are also necessary where B10 cells are known to regulate disease severity. controls to identify background IL-10 staining artifacts (34). B10 cells are predominantly found within the spleen at Phenotype/lineage/profession frequencies of 1–3% (Fig. 2) and also are found in low A variety of cell surface markers have been proposed for dif- numbers within blood, lymph nodes, Peyer’s patches, intes- ferent regulatory B cell subsets (21, 22). The generalized ex tinal tissues, and the CNS (7, 31, 35). B10 cells are prom- vivo phenotype of B10 cells from untreated mice is IgMhigh inent within the peritoneal cavity, but their numbers are low. IgDlowCD1dhighCD5+CD19highCD23lowB220high,with,10% B10 cells appear to express IL-10 exclusively, with the ex- coexpressing IgG or IgA (13, 15, 35, 38). Thereby, spleen ception of TNF-a, which is expressed at low levels by most B10 cells share surface markers with multiple phenotypically B cells following activation. Otherwise, B10 cells are a readily defined B cell subsets, including transitional, marginal zone, measurable subset of B cells that are functionally programmed marginal zone precursor, memory, and B1 B cells (6, 11, 13–15, to produce IL-10 in vivo. 38, 41, 43, 44). Spleen B10 cells are enriched within the B10 cells potentially originate from a progenitor B10 cell CD1dhighCD5+CD19high subpopulation (Fig. 2), where 15– subset (B10pro cells) that is not inherently competent to ex- 20% are B10 cells, and up to 50% are B10+B10pro cells (6, press IL-10 during 5-h assays (Fig. 1A). However, B10pro 13, 15, 29). Small numbers of B10 cells are also found within cells functionally mature into B10 cells following ex vivo other spleen B cell fractions. The phenotype of B10pro cells CD40 ligation during 48-h cultures. The addition of PIM after culture reflects their in vitro activation more than their during the last 5 h of these cultures induces both matured subset of origin. For example, most mouse and human B cells B10pro cells and B10 cells in the samples to express cyto- upregulate CD5 expression following CD40 stimulation in The Journal of Immunology 1397

FIGURE 1. Models for mouse B10 cell function. (A) The functional development of B10 and B10pro cells. B10 cells are functionally defined by their ability to express measurable cytoplasmic IL-10 after 5 h of PIM stimulation. Ex vivo stimulation induces B10pro cells to acquire IL-10 competence, while B10 effector cells (B10eff) cells actively secrete IL-10. B10pro and B10 cells within the spleen predominantly display a CD1dhighCD5+ cell surface phenotype. B10pro and B10 cells variably express CD5 and high CD1d depending on their tissue of origin. BCR signals are required for B10pro cell development; thereafter, CD40 signals, LPS or IL-21 can induce their functional maturation and ability to become B10eff cells. IL-10 reporter Tiger or 10BiT mice express GFP or Thy1.1 with delayed kinetics relative to IL-10, with the expression of GFP and Thy1.1 serving as more durable markers than IL-10 expression. After terminating IL-10 production, a small fraction of B10eff cells can differentiate into Ab-secreting plasma cells in vivo that secrete germline Abs with polyreactive, autoreactive or Ag-specific reactivity. (B) Model for B10 cell regulation of Ag-specific immunity. Some B cells capture select Ags that trigger appropriate BCR transmembrane signals (lightning bolt, step 1) and promote B10pro cell generation. During immune responses, B10pro cells present peptides to Ag-specific T cells through cognate MHC-class II (MHC II) interactions (step 2) that induce T cell activation and CD40/CD154 interactions, thereby inducing B10 cell IL-10 competence (step 3). Activated T cells then produce IL-21 locally, which binds to proximal B10 cell IL-21R (step 4). IL-21R signals induce B10 cell IL-10 production and effector function (step 5), which suppresses the activation of proximal Ag-specific T cells (step 6, ref. 29) and innate macrophage function (step 7, ref. 52). Absence of the correct BCR, CD40, MHC-class II, IL-21R, CD4+ T cells or IL-10 eliminates B10 effector cell function in vivo. vitro (31, 32). Spleen IL-10+ B cells are also enriched within IL-10 secretion (31). That B10 and B10pro cells preferentially the T cell Ig domain and mucin domain protein 1 (TIM-1)+ localize within the CD1dhighCD5+ B cell fraction also pro- compartment, and TIM-1+ B cells are enriched in the vides a convenient means to isolate a spleen population CD1dhighCD5+ compartment (45). However, IL-10+ B cells enriched in unstimulated B10 cells for functional and 2 are also present in the TIM-1 compartment, and TIM-1+ adoptive-transfer experiments. However, the CD1dhighCD5+ B cells are present in the non-CD1dhighCD5+ compartment. subset predominantly contains non-B10 cells that may have Thereby, intracellular IL-10 staining remains the only way to diverse functions. To control for this, it is critical to dem- visualize the entire B10 cell subset. onstrate that IL-10 production fully accounts for the regula- Although the developmental relationships between pheno- tory activities of the cells being studied by using comparable 2 2 2 2 typically defined B cell subsets are undoubtedly complex, B cells from IL-10 / mice, as one example. IL-10 / mice spleen B10+B10pro cells are found within both the marginal have normal numbers of spleen B cells and normal frequencies zone and follicular microenvironments in situ (46). Peritoneal of CD1dhigh and CD5+ B cells (7, 31), peritoneal cavity B1a, cavity B10 cells are also found in different proportions among B1b, and B2 cells (35), and IgM-secreting B cells (38), and the phenotypically defined B1a, B1b, and B2 cell sub- they develop normal T cell–dependent Ab responses (47) populations (Fig. 2), but they generally share similar pheno- prior to their development of inflammation and disease. types with their spleen counterparts (35). Thus, either B10 Moreover, autocrine IL-10 is not required for B10 cell de- cells traverse these different tissue microenvironments or cells velopment (38), although IL-10 is reported to have an auto- from each of these phenotypic subpopulations can become crine effect on mouse B1 cells (48). B10 cells. The transcription factors driving IL-10 production Using the size of the CD1dhighCD5+ B cell compartment or in B cells are unknown (23). Therefore, B10 cells do not other phenotypic subsets as surrogate markers for B10 cell appear to represent a lymphocyte lineage. expansion are not justifiable, because phenotype and IL-10 B10 cells can be isolated using commercial cytokine-capture competence are not synonymous. For example, most B10 assays following activation with PMA and ionomycin to induce cells are CD5+, but B10 cells only represent a subset of CD5+ 1398 BRIEF REVIEWS: MOUSE REGULATORY B10 CELLS

competence in vivo, because strong BCR signals may actu- ally divert intracellular signaling and induce B10pro cells to progress down a different functional pathway. In support of this, potent B cell stimulation with mitogenic anti-IgM Ab inhibits B10pro cell acquisition of IL-10 competence during in vitro cultures (31, 32, 35). Thus, BCR specificity and signaling intensity are likely developmental checkpoints that allow B10 cells to respond rapidly to self- or foreign Ags as a first line of defense to protect against vigorous immune responses that could lead to autoantibody production and tissue pathology.

Cognate interactions regulate B10 cell effector function in vivo At the cellular level, B10 cell–derived IL-10 inhibits T cell activation, as well as IFN-g and Th17 cell cytokine responses FIGURE 2. Phenotypic distribution of mouse B10 cells. Most B10 cells localize within the spleen and peritoneal cavity (red wedges in top circles). B10 in vivo (29). B10 cells are not able to inhibit polyclonal T cell cells are predominantly enriched within the spleen CD1dhighCD5+ B cell activation or proliferation following CD3 and CD28 ligation + subset and peritoneal cavity B1a subset (red wedges in histogram quadrants), in vitro, but they do suppress Ag-specific CD4 T cell IFN-g but are also found within other phenotypically defined subsets. Circle and and TNF-a responses and Th17 cell differentiation (7, 51). wedge sizes denote relative B cell numbers of each subset. Nonetheless, B10 cells are present in T cell–deficient nude mice (31). Likewise, B cell MHC class I, MHC class II, B cells (Fig. 2). Additionally, CD5+ or CD1dhighCD5+ B cell CD1d, CD40, or intrinsic IL-10 expression is not required frequencies do not predict B10 cell frequencies in different for B10 cell development (31, 38). However, MHC class models of disease, strains of mice, or lines of transgenic mice II– and CD40-dependent cognate interactions between B10 (31). Although most B cells in NOD and SJL mice are CD5+, cells and CD4+ T cells are essential for B10 effector cell only a few have the capacity to express IL-10 ex vivo or fol- function during T cell–driven autoimmunity (29). The re- lowing in vitro stimulation. Thus, B10 cells represent a quirement for Ag-specific interactions between CD4+ T cells functionally programmed B cell subset that pursues a regula- and B10 cells to generate B10 effector cells suggests a regula- tory “profession” through their capacity to express IL-10. tory-feedback loop whereby localized B10 effector cell IL-10 secretion preferentially downregulates Ag-specific T cell responses Ag regulates B10 cell development during cognate B10:T cell interactions, although IL-10 B10pro and B10 cell development is not stochastic, because suppression of Ag presentation by dendritic cells and mac- their numbers are reduced dramatically in transgenic mice with rophages is also possible (Fig. 1B). Either way, Ag specificity a fixed BCR (31) and in mice with decreased BCR signaling, ultimately regulates B10 cell effector function. 2 2 such as CD19-deficient (CD19 / ) mice, whereas CD19 IL-21 is a key cytokine for regulating B cell effector functions. overexpression expands B10 cell numbers (13, 31). The ab- Nonetheless, IL-21 is not required for B10 cell generation, and 2 2 sence of B10 cells in CD19 / mice leads to exacerbated IL-21 exposure does not induce most B cells to express IL-10, inflammation and disease symptoms during contact hyper- despite their IL-21R expression (29). However, IL-21 and its sensitivity, EAE, and IBD (6, 13, 40). In contrast, increasing receptor are necessary for B10 cell expansion and the induction CD40 signaling by CD22 deficiency and ectopic B cell of IL-10–secreting B10 effector cells during the induction of CD154 expression in transgenic mice increases B10+B10pro autoimmunity. In contrast, TGF-b and IFN-g inhibit B10pro cell numbers 10-fold in vivo (46). The calcium sensors cell maturation in vitro (29). Thus, different cytokines may STIM1 and STIM2 are also required for B cell IL-10 pro- elicit or inhibit B10 cell expansion and IL-10 secretion in vivo. duction after BCR stimulation (49). Thus, BCR signaling B10 cells suppress dendritic cell activation, cytokine pro- pathways appear to drive B cell acquisition of IL-10 compe- duction, and their ability to present Ag (7). B10 cells also tence, thereby selecting for Ag-specific B10 cells in vivo. suppress macrophage activation, cytokine and NO production, Ag-driven BCR signaling during B10 cell development likely and phagocytosis in an IL-10–dependent manner (6, 33). Al- explains why B10 cell regulation of inflammation and auto- though serum IL-10 increases with inflammation, cognate B10 immunity appears to be Ag specific under physiologic in vivo cell:T cell interactions are required to regulate monocyte acti- conditions (4, 6, 13, 29). In vivo selection by Ag also explains vation during Listeria infection (Fig. 1B) (52). This reveals the why B10 cells are primed to proliferate rapidly in response to potent effects of IL-10 in vivo and reinforces the need for mitogenic stimulation in vitro compared with other B cells multiple checkpoints in the control of B10 effector cell function (31, 35). Human B10 and B10pro cells are found within the during both innate and adaptive immune responses. These CD27+ memory compartment, consistent with the concept of checkpoints are likely to restrict B10 cell IL-10 production in vivo Ag experience (32). Like B10 cells, spleen marginal within the local microenvironment and selectively inhibit local zone and B1 B cells are more responsive to mitogenic signals macrophage and Ag-specific T cell function during inflamma- than other B cells (50), but it is not possible to delineate these tion and autoimmunity without inducing . three subsets because of their overlapping phenotypes. Despite Ag simulation in vivo, B10 cells express diverse germline B10 cells respond to innate signals BCRs with only rare mutations (35, 38). Therefore, low- LPS induces mouse B10pro cells to mature and become B10 affinity or chronic Ag–BCR stimulation may drive IL-10 effector cells, whereas CpG oligonucleotides induce B10 cells The Journal of Immunology 1399 to produce IL-10 (Fig. 1A) (31). IL-33 and B cell–produced with lupus and (61, 62). Tregs with the IL-35 also were reported to enhance B10 cell numbers (53– capacity to limit collateral tissue damage are also present at 57). TLR signals, infection, and apoptotic cells can also sig- sites of inflammation in active human autoimmune diseases, nificantly expand B10 and B10 effector cell numbers in vivo but they are too few in number to adequately control disease (38, 42, 52, 58). Nonetheless, MyD88, environmental patho- (63). Alternatively, autoimmune disease and inflammatory gens, and gut-associated flora are not required for normal B10 cytokine production may impact B10 cell maturation and/or cell development in vivo (31). Therefore, BCR-associated functional capacity in vivo. Therefore, the ex vivo expansion signals appear to preferentially program B10 cell IL-10 com- of a patient’s B10 cells in the absence of inflammatory factors petence, whereas overlapping adaptive and innate signaling before reinfusion may offer great promise for treating auto- pathways can induce B10 cell activation, proliferation, and immunity, allergy, and transplant rejection. effector function. Undoubtedly, IFNs, ILs, and cytokines that The in vivo depletion of all mature B cells, which includes B10 generally regulate B cell function are also likely to influence cells, can have therapeutic, as well as detrimental, effects in B10pro and B10 cells. Thus, there is no reason to partition B10 mouse models of human disease. Specifically, disease symptoms cells into adaptive or innate subsets because they respond like during contact hypersensitivity, EAE, and skin transplant re- most other B cells to both signals. jection can be more severe in B cell–depleted mice (6, 13, 64). Likewise, the depletion of B10 cells using a mAb that pre- B10 effector cell function and therapeutic potential dominantly spares other B cell subsets results in enhanced cel- The rapid in vivo expansion of B10 cells in disease models lular, innate, and humoral immune responses and exacerbates suggests acute Ag-specific B10 cell expansion and/or B10pro autoimmune disease (7, 46, 52). Therefore, future interventions cell maturation into B10 cells during disease initiation. In that decrease B10 cell numbers may prove beneficial in cancer either case, overall B10 cell numbers remain relatively low. therapy, infectious disease, and vaccine responses. Nonetheless, even small numbers of B10 cells among purified CD1dhighCD5+ B cells, peritoneal cavity B cells, or other B10 effector cell differentiation subsets can significantly inhibit disease initiation and subse- IL-10 is not required for B cell or B10 cell development, quent pathology in adoptive-transfer models of contact hy- differentiation, or Ab production in mice (38). However, persensitivity (13), EAE (4, 6, 7), lupus (15, 16), IBD (35, a small fraction of B10 effector cells can differentiate into Ab- 40), and graft-versus-host disease (59), whereas comparable producing cells in vivo and in vitro as they lose the capacity to 2 2 B cells isolated from IL-10 / mice are without therapeutic express IL-10 (Fig. 1A). B10 cells express the – effect. B10 cells isolated from Ag-primed mice or mice with associated transcription factors blimp1, xbp1,andirf4, with inflammation or autoimmunity are more potent than com- downregulated pax5 and bcl6 transcripts relative to follicular parable naive cells, consistent with disease driving the clonal B cells, likely due to chronic BCR selection and activation in expansion of Ag-specific B10 cells before their adoptive vivo (38). Even so, B10 effector cells from naive or LPS- transfer. Ex vivo B10 cell activation before adoptive transfers treated mice are not themselves plasmablasts/plasma cells, further enhances their regulatory efficacy in vivo (7). There- but they can differentiate into Ab-secreting cells at low fre- fore, B10 cells in relatively small numbers can have potent quencies. IL-10 reporter mice provide a means to identify B10 suppressive effects in vivo, particularly during Ag-specific effector cells after transient IL-10 expression, where the phe- responses. notype of spleen B10 cells and IL-10 reporter+ B cells remains The therapeutic potential of B cells induced to express IL-10 predominantly IgMhighCD1dhighCD5+CD19highCD23low was first demonstrated in landmark studies by the Londei B220high, even after in vivo LPS treatment (35). Variable low laboratory, where agonistic CD40 mAbs effectively treated frequencies of B10 effector cells express the CD43, GL7, and mice with collagen-induced arthritis (9, 60). Total spleen CD138 activation markers, and rare CD138highB220int/low cells B cell populations from in vitro culture systems that stimulate have been identified among IL-10 reporter+ B cells in naive, through CD40 or Ag sensitization also were shown to be LPS-treated, and infected mice (38, 41, 42). However, ,0.1% therapeutic during disease initiation and progression (10, 11, of reporter+ B cells isolated from LPS-treated 10BiT mice are 51). The therapeutic potential of purified B10 cells also was Ab-secreting cells (38). Thus, like most B cells, some B10 validated in studies in which B10 effector cells were expanded effector cells can differentiate into plasma cells. In contrast, ex vivo by a million-fold in a CD40- and IL-21–dependent other investigators report that Ab-secreting cells, at least in culture system (29). Remarkably, B10 effector cells purified some cases, are the major source of B cell–derived regulatory from these cultures dramatically suppressed both EAE initi- IL-10 and IL-35 in vivo (53, 65). The fate of most other B10 ation and established disease. Thus, B10 cells are regulatory effector cells and whether they become memory cells or regain during early inflammatory responses but also are functional the capacity to express IL-10 are unknown. during the course of disease in concert with other regulatory Plasma cells that derive from B10 cells secrete autoreactive, cells (7). Translation of these studies into humans may pro- polyreactive, and Ag-specific IgM, as well as IgG reactive with vide a novel immunotherapy for patients with complex disease T cell–dependent Ags (Fig. 1), reflecting their diverse BCR that does not respond to current treatments. sequences (35, 38). Despite their ability to switch isotypes, A better quantification of human Ag–specific B10 cells is B10 cell BCRs are predominantly germline and without so- needed to understand their relative contributions during hu- matic hypermutation, arguing against a germinal center man disease. B10 and B10pro cell numbers are generally origin. However, connectivity maps from multiparameter normal or elevated in patients with autoimmune disease (32). analyses of human B cell phenotypes, function, and gene However, disease may progress when IL-10–producing regu- expression profiles suggest that IL-10–secreting B cells may latory B cells are inadequate, as suggested for some patients undergo specific differentiation toward a germinal center fate 1400 BRIEF REVIEWS: MOUSE REGULATORY B10 CELLS

(66), whereas autocrine IL-10 promotes human B cell dif- 8. Tian, J., D. Zekzer, L. Hanssen, Y. Lu, A. Olcott, and D. L. Kaufman. 2001. -activated B cells down-regulate Th1 immunity and prevent ferentiation into plasmablasts (67). Nevertheless, the con- autoimmune diabetes in nonobese diabetic mice. J. Immunol. 167: 1081–1089. tributions of B10 cell–derived Abs to their in vivo regulatory 9. Mauri, C., D. Gray, N. Mushtaq, and M. Londei. 2003. Prevention of arthritis by -producing B cells. J. Exp. Med. 197: 489–501. effects appear modest, because B10 cell regulatory effects are 10. Gray, M., K. Miles, D. Salter, D. Gray, and J. Savill. 2007. Apoptotic cells protect dependent on IL-10 production. Nonetheless, Ags that elicit mice from autoimmune inflammation by the induction of regulatory B cells. Proc. B10pro and B10 cell development are likely to be cleared by Natl. Acad. Sci. USA 104: 14080–14085. 11. Evans, J. G., K. A. Chavez-Rueda, A. Eddaoudi, A. Meyer-Bahlburg, Abs coming from B10 cell–derived plasma cells, potentially D. J. Rawlings, M. R. Ehrenstein, and C. Mauri. 2007. Novel suppressive function providing a regulatory checkpoint that meters B10 cell de- of transitional 2 B cells in experimental arthritis. J. Immunol. 178: 7868–7878. 12.Watanabe,R.,M.Fujimoto,N.Ishiura,Y.Kuwano,H.Nakashima,N.Yazawa, velopment and peripheral tolerance. H.Okochi,S.Sato,T.F.Tedder,andK.Tamaki.2007.CD19expressionin B cells is important for suppression of contact hypersensitivity. Am. J. Pathol. 171: 560–570. Conclusions 13. Yanaba, K., J.-D. Bouaziz, K. M. Haas, J. C. Poe, M. Fujimoto, and T. F. Tedder. Historically, although B cell subsets have been studied based 2008. A regulatory B cell subset with a unique CD1dhiCD5+ phenotype controls on their phenotype, B10 cells demonstrate a clear case where T cell-dependent inflammatory responses. Immunity 28: 639–650. 14. Blair, P. A., K. A. Chavez-Rueda, J. G. Evans, M. J. Shlomchik, A. Eddaoudi, function trumps phenotype, because their only unique phe- D. A. Isenberg, M. R. Ehrenstein, and C. Mauri. 2009. Selective targeting of B cells notypic characteristic is their capacity to express IL-10. There is with agonistic anti-CD40 is an efficacious strategy for the generation of induced no transcription factor signature unique to B10 cells, likely regulatory T2-like B cells and for the suppression of lupus in MRL/lpr mice. J. Immunol. 182: 3492–3502. reflecting the fact that B10 cells represent a transient functional 15. Haas, K. M., R. Watanabe, T. Matsushita, H. Nakashima, N. Ishiura, H. Okochi, program rather than reflecting a cell lineage or stage of dif- M. Fujimoto, and T. F. Tedder. 2010. Protective and pathogenic roles for B cells during systemic autoimmunity in NZB/W F1 mice. J. Immunol. 184: 4789–4800. ferentiation. Because BCR specificity and signaling pathways 16. Watanabe, R., N. Ishiura, H. Nakashima, Y. Kuwano, H. Okochi, K. Tamaki, are critical checkpoints in the generation of B10 cells, the S. Sato, T. F. Tedder, and M. Fujimoto. 2010. Regulatory B cells (B10 cells) have a suppressive role in murine lupus: CD19 and B10 cell deficiency exacerbates sys- finding that some B10 cells express autoreactive BCRs suggests temic autoimmunity. J. Immunol. 184: 4801–4809. that autoantigens may be of particular importance in their 17. Amu, S., S. P. Saunders, M. Kronenberg, N. E. Mangan, A. Atzberger, and development. It is also likely that most Ags elicit balanced P. G. Fallon. 2010. Regulatory B cells prevent and reverse allergic airway inflam- mation via FoxP3-positive T regulatory cells in a murine model. J. Allergy Clin. responses between conventional B cells and B10 cells, although Immunol. 125: 1114–1124, e8. the factors that influence this balance are undefined. 18. Mizoguchi, A., and A. K. Bhan. 2006. A case for regulatory B cells. J. Immunol. 176: 705–710. Historically, B cell diversity has focused on the spectra of 19. Fillatreau, S., D. Gray, and S. M. Anderton. 2008. Not always the bad guys: B cells expressed BCRs. However, the functional characterization of as regulators of autoimmune pathology. Nat. Rev. Immunol. 8: 391–397. 20. Rieger, A., and A. Bar-Or. 2008. B-cell-derived interleukin-10 in autoimmune B10 cells and other regulatory B cell subsets highlighted the disease: regulating the regulators. Nat. Rev. Immunol. 8: 486–487. unanticipated functional diversity within the B cell pool, which 21. DiLillo, D. J., T. Matsushita, and T. F. Tedder. 2010. B10 cells and regulatory has yet to be fully explored. It is likely that additional regu- B cells balance immune responses during inflammation, autoimmunity, and cancer. Ann. N. Y. Acad. Sci. 1183: 38–57. latory B cell subsets will be identified. Thus, the extraordinary 22. Mauri, C., and A. Bosma. 2012. Immune regulatory function of B cells. Annu. Rev. combination of B cell functional diversity along with BCR Immunol. 30: 221–241. 23. Saraiva, M., and A. O’Garra. 2010. The regulation of IL-10 production by immune diversity will provide an unanticipated regulatory web that cells. Nat. Rev. Immunol. 10: 170–181. undoubtedly will cover the spectrum of immune system 24. Mizoguchi, A., E. Mizoguchi, R. N. Smith, F. I. Preffer, and A. K. Bhan. 1997. Suppressive role of B cells in chronic colitis of T cell receptor a mutant mice. J. Exp. function beyond what is currently known for B10 cell regu- Med. 186: 1749–1756. lation of innate, adaptive, cellular, and humoral immune 25. Shimomura, Y., E. Mizoguchi, K. Sugimoto, R. Kibe, Y. Benno, A. Mizoguchi, and responses. A. K. Bhan. 2008. Regulatory role of B-1 B cells in chronic colitis. Int. Immunol. 20: 729–737. 26. Wei, B., P. Velazquez, O. Turovskaya, K. Spricher, R. Aranda, M. Kronenberg, Acknowledgments L. Birnbaumer, and J. Braun. 2005. Mesenteric B cells centrally inhibit CD4+ T cell I thank Jacquelyn Lykken, Kathleen Candando, John Sleasman, Guglielmo colitis through interaction with regulatory T cell subsets. Proc. Natl. Acad. Sci. USA Venturi, and many collaborators for their countless contributions. 102: 2010–2015. 27. Mann, M. K., K. Maresz, L. P. Shriver, Y. Tan, and B. N. Dittel. 2007. B cell reg- ulation of CD4+CD25+ T regulatory cells and IL-10 via B7 is essential for recovery from experimental autoimmune encephalomyelitis. J. Immunol. 178: 3447–3456. Disclosures 28. Ray, A., S. Basu, C. B. Williams, N. H. Salzman, and B. N. Dittel. 2012. A novel The author has no financial conflicts of interest. IL-10-independent regulatory role for B cells in suppressing autoimmunity by maintenance of regulatory T cells via GITR ligand. J. Immunol. 188: 3188–3198. 29. Yoshizaki, A., T. Miyagaki, D. J. DiLillo, T. Matsushita, M. Horikawa, References E. I. Kountikov, R. Spolski, J. C. Poe, W. J. Leonard, and T. F. Tedder. 2012. 1. Mizoguchi, E., A. Mizoguchi, F. I. Preffer, and A. K. Bhan. 2000. Regulatory role of Regulatory B cells control T-cell autoimmunity through IL-21-dependent cognate mature B cells in a murine model of inflammatory bowel disease. Int. Immunol. 12: interactions. Nature 491: 264–268. 597–605. 30. Hoehlig, K., P. Shen, V. Lampropoulou, T. Roch, B. Malissen, R. O’Connor, + 2. Mizoguchi, A., E. Mizoguchi, H. Takedatsu, R. S. Blumberg, and A. K. Bhan. S. Ries, E. Hilgenberg, S. M. Anderton, and S. Fillatreau. 2012. Activation of CD4 + 2002. Chronic intestinal inflammatory condition generates IL-10-producing regu- Foxp3 regulatory T cells proceeds normally in the absence of B cells during EAE. latory B cell subset characterized by CD1d upregulation. Immunity 16: 219–230. Eur. J. Immunol. 42: 1164–1173. 3. Wolf, S. D., B. N. Dittel, F. Hardardottir, and C. A. Janeway, Jr. 1996. Experi- 31. Yanaba, K., J. D. Bouaziz, T. Matsushita, T. Tsubata, and T. F. Tedder. 2009. The mental autoimmune encephalomyelitis induction in genetically B cell-deficient development and function of regulatory B cells expressing IL-10 (B10 cells) requires mice. J. Exp. Med. 184: 2271–2278. antigen receptor diversity and TLR signals. J. Immunol. 182: 7459–7472. 4. Fillatreau, S., C. H. Sweenie, M. J. McGeachy, D. Gray, and S. M. Anderton. 2002. 32. Iwata, Y., T. Matsushita, M. Horikawa, D. J. Dilillo, K. Yanaba, G. M. Venturi, B cells regulate autoimmunity by provision of IL-10. Nat. Immunol. 3: 944–950. P. M. Szabolcs, S. H. Bernstein, C. M. Magro, A. D. Williams, et al. 2011. 5. Matsushita, T., M. Fujimoto, M. Hasegawa, K. Komura, K. Takehara, Characterization of a rare IL-10-competent B-cell subset in humans that parallels T. F. Tedder, and S. Sato. 2006. Inhibitory role of CD19 in the progression of mouse regulatory B10 cells. Blood 117: 530–541. experimental autoimmune encephalomyelitis by regulating cytokine response. Am. 33. Horikawa, M., V. Minard-Colin, T. Matsushita, and T. F. Tedder. 2011. Regu- J. Pathol. 168: 812–821. latory B cell production of IL-10 inhibits lymphoma depletion during CD20 im- 6. Matsushita, T., K. Yanaba, J.-D. Bouaziz, M. Fujimoto, and T. F. Tedder. 2008. munotherapy in mice. J. Clin. Invest. 121: 4268–4280. Regulatory B cells inhibit EAE initiation in mice while other B cells promote disease 34. Matsushita, T., and T. F. Tedder. 2011. Identifying regulatory B cells (B10 cells) progression. J. Clin. Invest. 118: 3420–3430. that produce IL-10 in mice. Methods Mol. Biol. 677: 99–111. 7. Matsushita, T., M. Horikawa, Y. Iwata, and T. F. Tedder. 2010. Regulatory B cells 35. Maseda, D., K. M. Candando, S. H. Smith, I. Kalampokis, C. T. Weaver, (B10 cells) and regulatory T cells have independent roles in controlling experimental S. E. Plevy, J. C. Poe, and T. F. Tedder. 2013. Peritoneal cavity regulatory B cells autoimmune encephalomyelitis initiation and late-phase immunopathogenesis. (B10 cells) modulate IFN-g+CD4+ T cell numbers during colitis development in J. Immunol. 185: 2240–2252. mice. J. Immunol. 191: 2780–2795. The Journal of Immunology 1401

36. Bouabe, H. 2012. Cytokine reporter mice: the special case of IL-10. Scand. expansion during Listeria infection governs innate and cellular immune responses in J. Immunol. 75: 553–567. mice. J. Immunol. 190: 1158–1168. 37. Kamanaka, M., S. T. Kim, Y. Y. Wan, F. S. Sutterwala, M. Lara-Tejero, J. E. Gala´n, 53. Shen, P., T. Roch, V. Lampropoulou, R. A. O’Connor, U. Stervbo, E. Hilgenberg, E. Harhaj, and R. A. Flavell. 2006. Expression of interleukin-10 in intestinal S. Ries, V. D. Dang, Y. Jaimes, C. Daridon, et al. 2014. IL-35-producing B cells are detected by an interleukin-10 reporter knockin tiger mouse. Immunity critical regulators of immunity during autoimmune and infectious diseases. Nature 25: 941–952. 507: 366–370. 38. Maseda, D., S. H. Smith, D. J. DiLillo, J. M. Bryant, K. M. Candando, C. T. Weaver, 54. Wang, R.-X., C.-R. Yu, I. M. Dambuza, R. M. Mahdi, M. B. Dolinska, Y. V. Sergeev, and T. F. Tedder. 2012. Regulatory B10 cells differentiate into -secreting cells P. T. Wingfield, S.-H. Kim, and C. E. Egwuagu. 2014. Interleukin-35 induces reg- after transient IL-10 production in vivo. J. Immunol. 188: 1036–1048. ulatory B cells that suppress autoimmune disease. Nat. Med. 20: 633–641. 39. Maynard, C. L., L. E. Harrington, K. M. Janowski, J. R. Oliver, C. L. Zindl, 55. Mauri, C., and K. Nistala. 2014. Interleukin-35 takes the ‘B’ line. Nat. Med. 20: A. Y. Rudensky, and C. T. Weaver. 2007. Regulatory T cells expressing interleukin 580–581. 2 10 develop from Foxp3+ and Foxp3 precursor cells in the absence of interleukin 56. Sattler, S., G. S. Ling, D. Xu, L. Hussaarts, A. Romaine, H. Zhao, L. Fossati- 10. Nat. Immunol. 8: 931–941. Jimack, T. Malik, H. T. Cook, M. Botto, et al. 2014. IL-10-producing regulatory 40. Yanaba, K., A. Yoshizaki, Y. Asano, T. Kadono, T. F. Tedder, and S. Sato. 2011. B cells induced by IL-33 (BregIL-33) effectively attenuate mucosal inflammatory IL-10-producing regulatory B10 cells inhibit intestinal injury in a mouse model. responses in the gut. J. Autoimmun. 50: 107–122. Am. J. Pathol. 178: 735–743. 57. Tedder, T. F., and W. J. Leonard. 2014. Autoimmunity: regulatory B cells—IL-35 41. Neves, P., V. Lampropoulou, E. Calderon-Gomez, T. Roch, U. Stervbo, P. Shen, and IL-21 regulate the regulators. Nat Rev Rheumatol 10: 452–453. A. A. Kuhl,€ C. Loddenkemper, M. Haury, S. A. Nedospasov, et al. 2010. Signaling 58.Miles,K.,J.Heaney,Z.Sibinska,D.Salter,J.Savill,D.Gray,andM.Gray. via the MyD88 adaptor protein in B cells suppresses protective immunity during 2012. A tolerogenic role for Toll-like receptor 9 is revealed by B-cell interaction Salmonella typhimurium infection. Immunity 33: 777–790. with DNA complexes expressed on apoptotic cells. Proc.Natl.Acad.Sci.USA 42. Madan, R., F. Demircik, S. Surianarayanan, J. L. Allen, S. Divanovic, A. Trompette, 109: 887–892. N. Yogev, Y. Gu, M. Khodoun, D. Hildeman, et al. 2009. Nonredundant roles for 59. Le Huu, D., T. Matsushita, G. Jin, Y. Hamaguchi, M. Hasegawa, K. Takehara, B cell-derived IL-10 in immune counter-regulation. J. Immunol. 183: 2312–2320. T. F. Tedder, and M. Fujimoto. 2013. Donor-derived regulatory B cells are im- 43. Spencer, N. F., and R. A. Daynes. 1997. IL-12 directly stimulates expression of portant for suppression of murine sclerodermatous chronic graft-versus-host disease. IL-10 by CD5+ B cells and IL-6 by both CD5+ and CD5- B cells: possible in- Blood 121: 3274–3283. volvement in age-associated cytokine dysregulation. Int. Immunol. 9: 745–754. 60. Mauri, C., L. T. Mars, and M. Londei. 2000. Therapeutic activity of agonistic 44. Brummel, R., and P. Lenert. 2005. Activation of marginal zone B cells from lupus monoclonal against CD40 in a chronic autoimmune inflammatory mice with type A(D) CpG-oligodeoxynucleotides. J. Immunol. 174: 2429–2434. process. Nat. Med. 6: 673–679. 45. Ding, Q., M. Yeung, G. Camirand, Q. Zeng, H. Akiba, H. Yagita, G. Chalasani, 61. Blair, P. A., L. Y. Noren˜a, F. Flores-Borja, D. J. Rawlings, D. A. Isenberg, M. H. Sayegh, N. Najafian, and D. M. Rothstein. 2011. Regulatory B cells are M. R. Ehrenstein, and C. Mauri. 2010. CD19+CD24hiCD38hi B cells exhibit identified by expression of TIM-1 and can be induced through TIM-1 ligation to regulatory capacity in healthy individuals but are functionally impaired in systemic promote tolerance in mice. J. Clin. Invest. 121: 3645–3656. Lupus Erythematosus patients. Immunity 32: 129–140. 46. Poe, J. C., S.-H. Smith, K. M. Haas, K. Yanaba, T. Tsubata, T. Matsushita, and 62. Flores-Borja, F., A. Bosma, D. Ng, V. Reddy, M. R. Ehrenstein, D. A. Isenberg, and T. F. Tedder. 2011. Amplified B lymphocyte CD40 signaling drives regulatory B10 C. Mauri. 2013. CD19+CD24hiCD38hi B cells maintain regulatory T cells while cell expansion in mice. PLoS ONE 6: e22464. limiting TH1andTH17 differentiation. Sci. Transl. Med. 5: 173ra23. 47. Kuhn,€ R., J. Lo¨hler, D. Rennick, K. Rajewsky, and W. Muller.€ 1993. Interleukin- 63. Valencia, X., and P. E. Lipsky. 2007. CD4+CD25+FoxP3+ regulatory T cells in 10-deficient mice develop chronic enterocolitis. Cell 75: 263–274. autoimmune diseases. Nat. Clin. Pract. Rheumatol. 3: 619–626. 48. Sindhava, V., M. E. Woodman, B. Stevenson, and S. Bondada. 2010. Interleukin- 64. DiLillo, D. J., R. Griffiths, S. V. Seshan, C. M. Magro, P. Ruiz, T. M. Coffman, 10 mediated autoregulation of murine B-1 B-cells and its role in Borrelia hermsii and T. F. Tedder. 2011. B lymphocytes differentially influence acute and chronic infection. PLoS ONE 5: e11445. allograft rejection in mice. J. Immunol. 186: 2643–2654. 49. Matsumoto, M., Y. Fujii, A. Baba, M. Hikida, T. Kurosaki, and Y. Baba. 2011. The 65. Ries, S., E. Hilgenberg, V. Lampropoulou, P. Shen, V. D. Dang, S. Wilantri, calcium sensors STIM1 and STIM2 control B cell regulatory function through I. Sakwa, and S. Fillatreau. 2014. B-type suppression: a role played by “regulatory interleukin-10 production. Immunity 34: 703–714. B cells” or “regulatory plasma cells”? Eur. J. Immunol. 44: 1251–1257. 50. Lopes-Carvalho, T., and J. F. Kearney. 2004. Development and selection of mar- 66. Lin, W., D. Cerny, E. Chua, K. Duan, J. T. Yi, N. B. Shadan, J. Lum, M. Maho- ginal zone B cells. Immunol. Rev. 197: 192–205. Vaillant, F. Zolezzi, S. C. Wong, et al. 2014. Human regulatory B cells combine 51. Yang, M., J. Deng, Y. Liu, K. H. Ko, X. Wang, Z. Jiao, S. Wang, Z. Hua, L. Sun, phenotypic and genetic hallmarks with a distinct differentiation fate. J. Immunol. G. Srivastava, et al. 2012. IL-10-producing regulatory B10 cells ameliorate collagen- 193: 2258–2266. induced arthritis via suppressing Th17 cell generation. Am. J. Pathol. 180: 2375–2385. 67. Heine, G., G. Drozdenko, J. R. Grun,€ H. D. Chang, A. Radbruch, and M. Worm. 52. Horikawa, M., E. T. Weimer, D. J. DiLillo, G. M. Venturi, R. Spolski, 2014. Autocrine IL-10 promotes human B-cell differentiation into IgM- or IgG- W. J. Leonard, M. T. Heise, and T. F. Tedder. 2013. Regulatory B cell (B10 cell) secreting plasmablasts. Eur. J. Immunol. 44: 1615–1621.