B cells and atherosclerosis

Abstract

B cells have emerged as important immune cells in cardiovascular disease. Initial studies have suggested that B cells protect against atherosclerosis development. However, subsequent studies demonstrating aggravation of atherosclerosis by B-2 cells have shed light on the subset-dependent effects of B cells. Here, we review the literature that has led to our current understanding of B cell regulation of atherosclerosis, touching on the importance of subsets, local regulation, human translation, and therapeutic potential.

atherosclerosis is a chronic inflammatory disease of major blood vessels and the primary underlying cause of cardiovascular disease (CVD). Circulating levels of LDL have been considered a major risk factor for atherosclerosis in humans. LDL accumulates in the artery wall, where it becomes oxidized (OxLDL), one of the major triggering events in the initiation of atherosclerotic lesion formation. OxLDL and other phospholipids (oxidative phospholipids) generate neoepitopes, termed oxidation-specific epitopes (10). These neoepitopes are recognized by both innate and adaptive immune cells, triggering a cascade of further events mediated by cytokines and chemokines that recruit more immune cells and further lipid accumulation. Together with immune cell recruitment, migration and proliferation of smooth muscle cells, apoptosis of foam cells, and the development of necrotic cores eventually lead to the formation of advanced lesions (36, 77, 101). Atherosclerotic lesions contain many immune cells, such as macrophages, mast cells, natural killer cells, T cells, and natural killer T cells, in the subendothelial space, i.e., the intima of the arterial wall (36, 102). In addition, immune cells, such as B cells, T cells, macrophages, and dendritic cells (DCs), occupy the vascular adventitia, perivascular adipose tissue (PVAT), and artery tertiary lymphoid organs (ATLOs) (30, 44, 69, 75, 91). The identification of how these cells regulate the response to oxidative phospholipids and other atherogenic stimuli is of key importance, and prior reviews have provided excellent overviews of this process (36, 37, 102). This review will focus on the role of B cells in atherosclerosis.

B Cells

B cells are lymphocytes that play important roles in both innate and adaptive immunity through both antibody (Ab) production and cytokine secretion. B cells are divided into two subtypes: B-1 and B-2 B cells.

B-1 B cells are innate-like B cells that produce natural Abs (NAbs). As B-1 cells show a rather restricted Ig variable region gene use, they undergo relatively limited or no affinity maturation, and their Abs show broad specificities with low-binding affinities. Interestingly, however, B-1 cells proliferate in response to self-antigens (Ags) and also form pools of short-lived, self-renewing B cells that produce most of the circulating NAbs of the IgM and IgA classes (5). Thus, B-1 cells are a subclass of B lymphocytes that are involved in innate humoral immune responses. B-1 cells arise predominantly from precursors in the fetal liver and constitute the earliest wave of mature peripheral B cells, undergoing self-renewal in the periphery. Recent studies have shown that B-1 cells can also arise from precursors in the bone marrow (BM) (Fig. 1) (23, 68). In adult mice, B-1 cells constitute a minor fraction (2–3%) of total B cells in secondary lymphoid organs (SLOs), such as the spleen, lymph nodes (LNs), and Peyer’s patches, but are abundant in the pleural and peritoneal cavities (50–70%) (38, 55). However, B-1 cells do not produce much Ab in the pleural and peritoneal cavities but migrate to the spleen for Ab production (Fig. 1) (15, 49, 104). Murine B-1 cells can be divided into B-1a (CD5⁺) and B-1b (CD5⁻) subtypes, based on CD5 expression (Fig. 1) (8). B-1a cells are important for the spontaneous production of IgM. Recent studies have shown that B-1b cells can confer T cell-independent, long-lasting, unmutated IgM memory to pathogens, such as Borrelia hermsii and Streptococcus pneumoniae (2, 33), and pathogen-associated polysaccharide Ag α(1,3)-dextran (26). The B-1b subtype shows IgA isotype switch capability and displays comparably high frequencies of somatic mutations in IgA-associated heavy-chain variable regions compared with B-1a cells (85, 86). These data suggest that B-1a and B-1b subtypes are not only phenotypically different but also have divergent functions.

Fig. 1.

Schematic of murine B cell origin, development, and their surface markers expression. B cells are divided into two major subsets: B-1 and B-2 cells. B-2 cells arise from precursors in the bone marrow (BM). Common lymphoid progenitors in the BM differentiate into pre-B cell (Pre B), pro-B cell (Pro B), and immature B cells (Imm B). Immature B cells leave the BM, enter the bloodstream, and then travel to the spleen. Immature B cells undergo further transitional stages (T-1 and T-2) and then differentiate into mature marginal zone B cells (MZ-B) and follicular B cells (Fol B). After antigenic stimulation, MZ-B cells differentiate into IgM-secreting plasma cells (PCs). After antigen stimulation or presentation from follicular dendritic cells (FDCs) and with the help of follicular helper T cell (T_Fh), Fol-B cells enter into germinal center reactions (GC-B) followed by differentiation into memory B cells (Mem B) and antigen-specific, antibody-secreting long-lived PCs (LPCs). These LPCs migrate to the BM and stay for longer periods. B-1 B cells develop from B-1 precursors in the fetal liver and adult BM and migrate to and reside in the peritoneal and pleural cavity. B-1 cells are divided into CD5⁺ B-1a and CD5⁻ B-1b cells. After activation, peritoneal B-1 cells lose surface CD11b expression and migrate to the spleen to secrete IgM.

B-2 cells are conventional B cells and participate in adaptive immune responses. B-2 cells arise from precursors in the BM. Common lymphoid progenitors in the BM differentiate into pre-B cell and pro-B cell stages by undergoing Ig heavy- and light-chain rearrangements and then develop into immature B cells. These immature B cells carry a mature B cell receptor (BCR) with unique specificity of IgM (6, 99). Immature B cells leave the BM, enter the bloodstream, and then travel to SLOs. Immature B cells undergo further transitional stages (T-1 and T-2) and then differentiate into mature B cells. These mature B cells differentiate into marginal zone (MZ)-B cells and follicular (Fol)-B cells (Fig. 1). MZ-B cells reside within the splenic marginal sinus. They mainly participate in the firstline defense against blood-borne pathogens and T cell-independent type II Ags, such as bacterial capsular polysaccharides, and differentiate into Ab-secreting plasma cells (PCs) (99). Due to participation in early immune responses, activation, and Ab production without T cell help, MZ-B cells are also considered innate immune cells. Fol-B cells are the major B-2 population in the periphery and differentiate into different B-2 cell subtypes during adaptive immune responses. Fol-B cells become activated by Ag stimulation and T cell help and then these activated B cells undergo germinal center (GC) reactions. B cells in GCs undergo class-switch recombination and somatic hypermutation steps and generate switched Ig and increased BCRs that are specific for Ag. These GC B cells undergo an affinity maturation and selection process for Ag with the help of Fol DCs and Fol helper T cells and then differentiate into Ag-specific, Ab-secreting, long-lived PCs or memory B cells (Fig. 1) (34). These PCs and memory B cells participate in long-lived, protective humoral immunity.

In addition to Ab-mediated immune responses, B cells regulate lymphoid tissue organization/development, modulate T cell and macrophage polarization, and regulate inflammatory reactions via secretion of discrete cytokines (57, 58, 89). Cytokine-producing B cells are referred to as “regulatory” B cells (B_regs) and “effector” B cells (25, 57, 58) and are not necessarily a unique subset but rather a functionally different phenotype derived from either B-1 or B-2 cells. IL-10-producing B_regs are called B-10 cells and have anti-inflammatory effects (103). IL-10, secreted by multiple cell types, including T cells, B cells, monocytes, macrophages, mast cells, eosinophils, and keratinocytes, is capable of suppressing both T cell helper types 1 and 2 polarized immune cells and inhibiting macrophage Ag presentation and proinflammatory cytokine production (4). Other studies have shown that B_regs can suppress inflammatory reactions by secreting other anti-inflammatory cytokines, such as transforming growth factor-β and IL-35 (74, 94, 100). Recently, Mauri and Menon (63) and Rosser and Mauri (83) reviewed B_reg subtypes identified in both mice and humans and how these B_reg subtypes can be induced in response to inflammation at different stages in development.

B Cells in Atherosclerosis

Evidence for a protective role.

B cells have emerged as important immune cells in the regulation of atherosclerosis. Caligiuri et al. (12) originally observed that after splenectomy, there was a reduction in B cells and aggravation of atherosclerosis in apolipoprotein E-deficient (ApoE^−/−) mice. Immunomagnetic separation of spleen-derived T and B cells from diseased ApoE^−/− mice and their adoptive transfer into splenectomized, young, nonatherosclerotic ApoE^−/− mice, which were maintained on a high-fat diet (HFD) for 12 wk, have demonstrated that the atheroprotective effect was conferred by B cells. Consistent with these findings, LDL receptor-deficient (LDLR^−/−) mice transplanted with BM from mice with a disrupted BCR gene [B cell deficient (µMT^−/−)] exhibited a 30−40% increase in atherosclerotic lesions in the proximal and distal aortas compared with control mice [wild-type (WT) BM transplanted into LDLR^−/−] (59). LDLR^−/− mice reconstituted with BM cells from µMT^−/− mice possess ≤1% of their normal B cell population, providing further evidence that B cells may protect from atherosclerosis development. Recent work from our group (18) demonstrated that adoptive transfer of splenic B cells reduced diet-induced atherosclerosis in µMT^−/− (Table 1). Prior passive and active immunization studies (3, 9, 11, 24, 27, 28, 72, 73, 106) have supported an atheroprotective role for B cell-derived Abs, particularly IgM. Exogenous immunization of rabbits with malondialdehyde-modified LDL (MDA-LDL) (73) or other OxLDL (3) reduced atherosclerosis. Immunization studies in murine models have confirmed Ag-specific Ab responses to immunization and attenuation of atherosclerosis (28, 106).

Table 1.

Summary of different experimental studies that have demonstrated the role of B cell subtypes in murine atherosclerosis

Experiment (Reference)	Atherosclerosis
B cell transfers
Total splenic B cells (12)	↓↓
B-2 cells (5 × 106) (52)	↑↑
B-1 cells (5 × 106) (52)	↓↓
B-2 cells (30 and 60 × 106) (18)	↓↓
B-1a cells (1 × 105) (53)	↓↓
B-1b cells (1 × 105) (82)	↓↓
Bregs (CD21hiCD23hiCD24hi) (93)	Neointima formation ↓↓
Bone marrow transplantation
µMT−/− compared with WT → LDLR−/− (59)	↑↑
BAFFR−/− compared with WT → LDLR−/− (88)	↓↓
B cell depletion therapy
Anti-CD20 treatment (1, 52)	↓↓
Anti-BAFFR treatment (50)	↓↓
BAFFR deletion (51)	↓↓

B_regs, regulatory B cells; µMT^−/−, B cell-deficient mice; WT, wild-type; LDLR^−/−, LDL receptor knockout; BAFFR, B cell-activating factor receptor.

Evidence for an atherogenic role.

B cell activating factor (BAFF) is important for B cell survival, activation, and differentiation by binding to its receptors, such as BAFF receptor (BAFFR), B cell maturation Ag, and transmembrane activator and calcium-modulating ligand interactor (TACI). Studies that use approaches that deplete predominantly B-2 cells, such as anti-CD20 Ab, anti-BAFFR mAb, and BAFFR deficiency (BAFFR^−/−), attenuated atherosclerosis development in ApoE^−/− and LDLR^−/− mice (1, 50–52, 88). Kyaw et al. (52) demonstrated that B cell depletion by anti-CD20 Ab attenuated atherosclerosis development in ApoE^−/− mice maintained on an HFD and that adoptive transfer of 5 × 10⁶ B-2, but not B-1, B cells into total lymphocyte-deficient ApoE^−/− mice (ApoE^−/−, recombination activating gene 2 deficient, common cytokine receptor γ-chain deficient) and ApoE^−/−, µMT^−/− mice aggravated atherosclerotic lesion size and macrophage accumulation in young mice maintained on a HFD. Similar results were reported at the same time by Ait-Oufella et al. (1), who observed that B cell depletion by anti-CD20 Ab reduced atherosclerosis in young ApoE^−/− and LDLR^−/− mice maintained on a HFD (Fig. 2 and Table 1).

Fig. 2.

Simplified schematic of murine B cell subsets and their role in atherosclerosis. B cells are divided into two major subsets: B-1 and B-2 cells. B-1 cells (B-1a and B-1b) secrete natural IgM and attenuate diet-induced atherosclerosis. B-1a-derived innate response activator (IRA) B cells secrete granulocyte-macrophage colony-stimulating factor, which is important for Ly6Chi monocyte development, and promote atherosclerosis. B-2 cells are divided into MZ-B cells and Fol-B cells. Fol-B cells secrete IgG, stimulate inflammatory T cells (T), and aggravate atherosclerosis. The role of MZ-B cells and regulatory B cells (B_regs) in atherosclerosis is unclear. APC, antigen-presenting cells; INFγ, interferon-γ.

These two groups further demonstrated the impact of B-2 cells in promoting atherosclerosis by manipulating the BAFFR pathway. ApoE^−/−/BAFFR^−/− double-knockout mice showed reduced numbers of conventional B-2, but not B-1a, cells and attenuated atherosclerosis compared with control mice (ApoE^−/−) maintained on a HFD (51). Treatment of ApoE^−/− mice with anti-BAFFR mAb produced a similar effect (50). Sage et al. (88) observed reduced atherosclerosis in the aortic root in LDLR^−/− mice reconstituted with BAFFR^−/− BM compared with control mice (LDLR^−/− mice reconstituted with WT BM) maintained on a HFD (Table 1).

B Cell Effects on Atherosclerosis Are Subset Dependent

Studies demonstrating that the reduction of B-2 cells attenuated atherosclerosis initially seemed at odds with prior immunization studies (3, 28, 73, 106) and the earlier studies of Caligiuri et al. (12), Doran et al. (18), and Major et al. (59), demonstrating that B cells attenuated atherosclerosis. Yet Ait-Oufella et al. (1), Kyaw et al. (50–52), and Sage et al. (88) demonstrated that anti-CD20 Ab and loss of BAFFR activity predominantly affected the B-2 and not B-1 cell population, shedding light on the important aspect of B cell subset-dependent effects on atherosclerosis. In contrast to B-2 cells, followup studies (42, 53, 82) revealed that B-1 cells are atheroprotective. B-1-derived natural IgM Abs can bind to the oxidation motifs in LDLs, the phosphocholine head group on cell wall polysaccharides of pathogens, such as S. pneumoniae and apoptotic cells (95). In vitro experiments have demonstrated that natural IgM can block the uptake of OxLDL by macrophages (Fig. 2) (41), a key pathogenic step in atherosclerosis formation. Lewis et al. (56) demonstrated that secreted IgM is important for atheroprotection. Increased atherosclerosis was observed in female secreted IgM-deficient LDLR^−/− mice compared with control mice irrespective of type of diet (low fat or high fat). Kyaw et al. (53) demonstrated that adoptive transfer of peritoneal B-1a cells increased IgM in atherosclerotic lesions and attenuated disease in splenectomized ApoE^−/− mice compared with control ApoE^−/− mice. More recently, Kyaw and coworkers (42) demonstrated that Toll-like receptor 4 and myeloid differentiating factor 88 are essential for B-1a-derived, IgM-mediated atheroprotection in splenectomized ApoE^−/− mice. Whereas B-1a cells had been thought to be the major source of IgM NAb, we (82) have recently shown that B-1b cells produce substantial IgM NAb in vivo, and adoptive transfer of B-1b cells attenuated diet-induced atherosclerosis in recombination activating gene 1-deficient ApoE^−/− mice compared with PBS-injected control mice (Fig. 2 and Table 1).

However, many questions remain about the context-dependent nature of B cell subset functions. Rauch et al. (79) identified a unique B-1a cell-derived subtype, called innate response activator (IRA) B cells, in a sepsis model. They proposed that these IRA B cells are phenotypically and functionally distinct, develop in the spleen after LPS stimulation, and produce granulocyte-macrophage colony-stimulating factor (GM-CSF). GM-CSF is an atherogenic growth factor that promotes the differentiation of inflammatory Ly-6C^high monocytes (Fig. 2) (79, 80). Mice deficient in IRA B cells were protected from atherosclerosis (39). These data provide the first evidence for an atherogenic role for B-1a-derived IRA B cells.

Interestingly, Jackson et al. (46) recently demonstrated that overexpression of BAFF attenuates atherosclerosis via TACI-mediated B cell activation in ApoE^−/− mice. Transgenic BAFF ApoE^−/− mice had higher levels of atheroprotective IgM, such as anti-PC IgM and anti-MDA-LDL IgM, and reduced cholesterol levels in serum compared with ApoE^−/− and transgenic BAFF ApoE^−/− TACI-deficient mice, suggesting that BAFF-dependent regulation of B cell survival and atheroprotection is linked to subset-dependent, specific BAFFR activity.

In addition to B-1 and B-2 cells, B_regs have been implicated in modulating atherosclerosis. B_regs produce immune-suppressive cytokines, such as IL-10 and transforming growth factor-β, and suppress other immune-mediated conditions, such as experimental autoimmune encephalomyelitis, collagen-induced arthritis, and colitis (25, 62, 66). Therefore, it is natural to speculate that they could be atheroprotective. Indeed, it is well established that depletion of IL-10 increased infiltration of inflammatory cells, production of inflammatory cytokines, and aggravated atherosclerosis in mice (13, 60). IL-10 levels within the aorta have been linked to B_regs accumulation and reduced overall aortic leukocyte content (29). Yet there are conflicting results as to the role of IL-10-producing B_regs in atherosclerosis. Strom et al. (93) found that the CD21^hiCD23^hiCD24^hi B_reg subset was increased in the draining LNs of ApoE^−/− mice. Adoptive transfer of these cells into female ApoE^−/− mice attenuated neointima formation in response to perivascular collar-induced carotid artery injury. Inhibition of IL-10 using a neutralizing Ab or adoptive transfer of B cells from IL-10-deficient mice prevented LN-derived B cell atheroprotection (Table 1). In contrast, Sage et al. (87) demonstrated that male LDLR^−/− mice irradiated and reconstituted with 80% µMT^−/− BM and 20% BM from IL-10-deficient mice had no difference in the size and cellular contents of the lesion compared with 20% WT BM, despite marked reductions of IL-10 in B_regs. As such, the role of IL-10-producing B_regs in atherosclerosis remains unclear (Fig. 2).

Local B Cell Immune Responses in Atherosclerosis

Ab production and disease-specific immune responses are thought to occur in SLOs. However, lymphocyte infiltration, followed by TLO formation in the adventitia adjacent to atherosclerotic plaque, has been observed in humans and aged mice (30, 43, 69). Moreover, Ig repertoire analysis of human arterial-wall lymphocytes demonstrated that resident B cells expressed switched isotypes of Ig heavy chains with hypermutated variable regions and inversion of the λ-to-κ ratio of L-chain use and activation-induced cytidine deaminase, suggesting that B cell recruitment and differentiation were taking place in the arterial wall (35). Recently, we have published (91) that ATLOs harbor both B-1 (B-1a and B-1b) and B-2 B cell subtypes. The B-2 B cell subtypes in ATLOs include transitional, Fol, GC, class-switched memory B cells, and PCs. In addition, B-1 cells and PCs in ATLOs can secrete IgM and IgG locally. These data, together with previous observations that B cells are important in Ag presentation in ATLOs, reveal atherosclerosis-specific B cell immunity, which includes effector B cells, PCs, and several immunosuppressive B cell subtypes (44, 91). These findings suggest that (auto) Ag-dependent hypermutation, proliferation, affinity maturation, Ig class switching, memory B cell generation, and differentiation into long-lived PCs may be carried out in the arterial adventitia. It has been suggested that ATLOs provide a new paradigm of atherosclerosis-specific B cell immunity and are the principal lymphoid tissue that orchestrates atherosclerosis-specific B cell immunity in the abdominal aorta of ApoE^−/− mice during aging (67, 91, 105). All of these data suggest that disease-specific immune reactions may occur locally.

PVAT has also been implicated in the local regulation of atherosclerosis. Adipocytes in PVAT secrete both proinflammatory and anti-inflammatory cytokines (78), and PVAT-derived monocyte chemoattractant protein-1 increased carotid artery neointimal formation in response to wire injury in LDLR^−/− mice (61). PVAT, adjacent to atherosclerotic plaque in humans, is more inflammatory than PVAT adjacent to nondiseased vessels, suggesting that PVAT may participate in local immune reactions that could modulate atherogenesis (14). Fat-associated lymphoid clusters (FALCs) have been found in adipose tissue, and these FALCs regulate local inflammation by secreting Ag-specific Abs (7, 47, 70). B-1 cells, B-2 cells, and FALCs are present in the PVAT of young and aged ApoE^−/− mice (75, 105), suggesting that B cells and FALCs in PVAT may participate in local immune responses in atherosclerosis.

B Cells in Human Atherosclerosis

How these B cell findings in animal models will apply to understand the pathogenesis of human atherosclerosis or how this may impact on therapy remain unclear. An unbiased systems biology approach using the Framingham Heart Study participants identified that genes associated with B cell activation were most strongly enriched in control subjects but not in coronary heart disease patients, suggesting that B cells may play a protective role in human atherosclerosis (45). Notably, plasma levels of IgM to MDA-LDL in humans are associated with less coronary artery disease and fewer cardiovascular events (97, 98) and can predict 15-yr CVD outcomes (98), suggesting that IgM to MDA-LDL-producing B cells in humans may also be atheroprotective. However, which B cell fraction was responsible for this IgM production was unknown. Recently, Griffin et al. (32) identified a circulating human B cell subset with functional properties similar to those associated with murine B-1 cells. This human B-1 cell subset (CD19⁺ CD20⁺ CD27⁺ CD43⁺) secreted IgM spontaneously, stimulated T cells to proliferate, and demonstrated tonic intracellular signaling. Engelbertsen et al. (22) found that these human B-1 cells produce IgM to a modified epitope on LDL ApoB, a marker inversely associated with CVD events. Further studies are needed to characterize fully human B-1 cells producing IgM specific to an oxidation-specific epitope.

Biologicals that target B cells have entered the clinical arena for the treatment of autoimmune diseases, such as rheumatoid arthritis and systemic lupus erythematosus (16, 17, 19–21, 31, 48, 54, 64, 65, 76, 84, 90, 92). Interestingly, patients with rheumatoid arthritis and systemic lupus erythematosus are at increased risk of CVD (40, 81), yet the impact of these therapies on CVD risk in humans remains unclear (71, 96). We have recently reviewed this literature and performed a meta-analysis of existing studies with rituximab that included CVD end points; however, larger and longer studies are needed to draw any conclusions (71).

Conclusions

Significant advances have been made in our understanding of the role of B cells in murine atherosclerosis over the last two decades. Yet many unanswered questions remain, such as the impact of local versus systemic effects, identification of all of the molecular and cellular regulators of B cell functions linked to atherosclerosis, and the impact of biological agents that deplete B cells on human CVD. The answers to these questions have promising potential to provide for novel strategies for CVD prevention.

GRANTS

Funding for this work was provided by National Heart, Lung, and Blood Institute Grants R01-HL-136098, P01-HL-055798, and R01-HL-107490 (to C. A. McNamara).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

P.S. and C.A.M. conceived and designed the research; P.S. prepared figures; P.S. drafted manuscript; P.S. and C.A.M. edited and revised manuscript; P.S. and C.A.M. approved final version of manuscript.