Immunol Res (2008) 41:203–216 DOI 10.1007/s12026-008-8023-3

Anti-Sm B cell tolerance and tolerance loss in systemic lupus erythematosus Stephen H. Clarke

Published online: 1 July 2008 Ó Springer Science+Business Media, LLC 2008

Abstract Autoimmunity is a serious health problem and understanding the maintenance and loss of tolerance to self-antigens are key issues in developing new therapeutic strategies to treat these diseases. Despite considerable progress toward understanding B cell tolerance and tolerance loss, much remains unknown, particularly regarding B cells specific for antigens targeted in disease. Our interest in systemic lupus erythematosus (SLE), a B cell-mediated autoimmune disease characterized by the production of autoantibodies to numerous nuclear antigens, is focused on understanding B cell tolerance and tolerance loss to the SLE-specific autoantigen Sm in mice and humans. Our work aims to provide the cellular and molecular underpinnings for the development of rational therapies to target autoreactive B cells in human SLE. Keywords Systemic lupus erythematosus
B-lymphocytes
Tolerance
Sm
Dendritic cells
Anergy

Introduction Autoimmune diseases result from a failure of the immune system to discriminate between self and non-self. These diseases vary wildly in etiology with some mediated by T cells and others by B cells. We have had a long-standing interest in B cell-mediated autoimmunity and in systemic lupus erythematosus (SLE) in particular. Although immunologists have made great strides in understanding B cell tolerance and tolerance loss in autoimmunity, much remains unknown, especially regarding B cells specific for disease-associated autoantigens. Current treatment regimens for these diseases involve general immune suppression and, more recently B cell depletion therapies. However, significant side effects accompany each of these therapies and, therefore, there remains a strong need for therapies that specifically target autoreactive B cells. Development of these therapies requires a detailed understanding S. H. Clarke (&) Department of Microbiology and Immunology, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA e-mail: [email protected]

204

Immunol Res (2008) 41:203–216

of the cellular and molecular bases for B cell tolerance and tolerance loss. This review describes our work on B cell tolerance in mice and man to an SLE-associated autoantigen. SLE is an immune complex-mediated autoimmune disease, in which immune complex deposition in small blood vessels initiates an inflammatory reaction with subsequent tissue injury. For example, immune complex deposition in the kidney glomeruli can lead to glomerulonephritis, deposition in the joints can lead to arthritis, and deposition in small blood vessels lead to vasculitis. The autoantibodies produced in SLE are characteristically specific for nuclear antigens, such as single stranded and double stranded DNA, ribonucleoproteins (RNPs), histones, and to multiple other cellular and extracellular proteins [1, 2], although patients do not produce autoantibodies to all of these self-antigens. SLE can present with protean manifestations of inflammation involving multiple organs such as kidney, lungs, upper airways, skin, joints, and central and peripheral nervous systems. The acuity and rate of progression of disease varies from an insidious course to that of fulminant life threatening, multiorgan disease. The long-term course of disease varies from patient to patient with some attaining long-lasting remission and others suffering multiple relapses [3–5]. SLE is a multigenic disease [6, 7], although the identity of few of these genes is known. Linkage to defects in complement system genes is strong in SLE, particularly deficiencies in complement components C1q, C2, and C4 [8, 9]. Complement components have an important role in apoptotic cell clearance [10], which may explain this linkage. Apoptotic cells expose nuclear antigens, including DNA, histones, and RNPs, on the external leaflet of their cytoplasmic membranes and blebs, leading to the suggestion that apoptotic cells are a prominent source of antigen in the activation of autoreactive B cells in SLE [11, 12]. Indeed, immunization of mice with apoptotic cells induces transient anti-DNA and antiRNP responses [13, 14]. To understand how B cell tolerance is lost in SLE, we focused on B cells specific for the Smith (Sm) autoantigen; Sm is an RNP present in the nuclei of all cells and participates in RNA processing [15]. Because anti-Sm antibodies are unique to SLE, they are diagnostic [16]. Thus, understanding the normal cellular and molecular bases for anti-Sm B cell regulation and their activation in disease may provide insight into the etiology of SLE. To facilitate these studies, we generated IgH chain transgenic mice (2-12H) that have a high frequency of anti-Sm B cells [17]. The H chain VDJ rearrangement is from an anti-Sm B cell hybridoma of MRL/Faslpr origin and lacks amino acid replacement mutations [18]. Thus, the anti-Sm B cells generated by 2-12H mice likely represent the low-affinity anti-Sm B cell repertoire generated in normal mice. Anti-Sm B cells are detectable by flow cytometry using biotinylated Sm protein, and we initially reported that *30% of splenic B cells are anti-Sm based on flow cytometry, but subsequent functional analysis suggests that the majority of splenic B cells are low-affinity Sm-binders (Ramiro Diz and S. H. Clarke, manuscript submitted). 2-12H anti-Sm B cells vary in affinity due to the use of the endogenous L chain repertoire, but in some experiments, we combine the 2-12H transgene with a L chain transgene to generate anti-Sm B cells of uniform affinity [19]. Despite the high frequency of anti-Sm B cells in nonautoimmune 2-12H mice, the level of serum anti-Sm does not exceed that present in non-Tg mice. However, autoimmune 2-12H mice develop high serum levels of anti-Sm [20, 21]. Thus, non-autoimmune mice are tolerant to Sm and tolerance is lost in autoimmune mice.

Positive and negative regulation of anti-Sm B cells Multiple mechanisms regulate autoreactive B cells by interfering with differentiation or activation. Tolerance checkpoints span the full length of the B cell differentiation and

Immunol Res (2008) 41:203–216

205 (-) selection

Bone Marrow

Spleen Tr

FO

MZ

PrePre-PC

PC

Peripheral deletion

Y Y Y YY Y Y YY Y

2-12H/V 4 high affinity

(+) selection

2-12H mod affinity

Y

2-12H/V 8 low affinity anergy

ILIL-6 CD40L TNF

PrePre-PC tolerance checkpoint

B-1 2-12H

Peritoneum Fig. 1 Positive and negative selection of anti-Sm B cells in non-autoimmune mice. The regulation of B cells is highly dependent on their affinity for Sm. Low-affinity anti-Sm B cells from 2-12H/Vj8 mice are able to leave the bone marrow, enter the spleen, and mature to the FO B cell stage, but selection to the MZ and B-1 cell subsets does not occur. B cell intrinsic mechanisms and cytokines (IL-6, CD40L, and TNFa) released by activated macrophages and DCs repress the activation of these B cells by TLR stimulation. L chain diverse 2-12H mice develop anti-Sm B cells of the FO, MZ, and B-1 cell subsets. The FO B cells are anergic regulated similarly to the low-affinity anti-Sm FO B cells of 2-12H/Vj8 mice. In addition, positive selection of anti-Sm B cells occurs in 2-12H mice to generate functional anti-Sm MZ and B-1 B cells. Constitutive anti-Sm MZ B cell activation drives their differentiation toward the PC stage, but differentiation arrests at a pre-PC tolerance checkpoint to prevent PC formation. Competition with nonautoimmune B cells for BAFF eliminates the high-affinity anti-Sm B cells of 2-12H/Vj4 mice at the transitional B cell stage. In the absence of competition from non-autoreactive B cells, high-affinity anti-Sm B cells differentiate to the FO, MZ, and B-1 B cell subsets, but are anergic. Analysis of these cells provides insight into the mechanisms of anergy

activation pathways, from the initial expression of an autoreactive B cell receptor (BCR) to the pre-plasma cell (PC) stage following activation [22–28]. Autoreactive B cells that bind self-antigen with high avidity (e.g., membrane-bound antigens) either delete upon initial BCR expression as immature B cells in the bone marrow, or edit their BCR by reinitiating V(D)J rearrangement [23, 24]. Autoreactive B cells that do not encounter self-antigen in the bone marrow or bind self with low avidity (e.g., soluble proteins) migrate to the spleen and either delete as immature transitional B cells [22, 25, 29] or become non-functional (anergy) mature follicular (FO) B cells [22]. Finally, some autoreactive B cells undergo cell death after antigen-induced activation, but before PC differentiation [27, 28]. Our analysis of anti-Sm B cells in 2-12H mice provides unique insights into the regulation of B cells specific for self-antigens targeted in SLE. While anti-Sm B cell deletion may occur in the bone marrow, many anti-Sm B cells reach the periphery and differentiate to FO, marginal zone (MZ), and B-1 B cells (Fig. 1) [13, 17, 30, 31]. Anti-Sm FO B cells are hyporesponsive to toll-like receptor (TLR) stimulation due to an inherent defect (Ramiro Diz and S. H. Clarke, manuscript submitted) and to repression by cytokines secreted by activated DCs and macrophages [30, 32]. Vilen and colleagues in this issue discuss these mechanisms. In contrast to FO B cell anergy, anti-Sm MZ B cells are responsive to TLR stimulation and signal normally in response to BCR crosslinking and

206

Immunol Res (2008) 41:203–216

therefore appear to be functional (Ramiro Diz and S. H. Clarke, manuscript submitted). This can explain how immunization of 2-12H mice with soluble Sm and apoptotic cells (which expose Sm on the external surface of their membranes) induces a transient anti-Sm response [17, 31]. Differentiation of MZ and B-1 B cells in general requires positive selection by antigen, and because they have a pre-activated phenotype, they can differentiate rapidly to antibody secreting cells (ASCs) upon antigen stimulation [33–35]. Thus, MZ and B-1 B cells provide an immediate response to foreign pathogens. This implies that anti-Sm antibodies produced by MZ and B-1 B cells have important survival value, possibly in defense against foreign pathogens or alternatively to aid in the clearance of apoptotic cells. Efficient uptake of apoptotic cells by professional phagocytic cells requires IgM [36], although the specificity of this IgM is not known. This raises the possibility that pathogenic autoantibodies arise by diversion of a normally protective response. That positive and negative selection of anti-Sm B cells occurs in non-autoimmune mice prompted us to examine the contributions of subset identity and affinity to anergy. Using monoclonal high- and low-affinity anti-Sm Tg models made by combining the 2-12H transgene with Vj4 or Vj8 L chain transgenes, respectively, we find that BCR affinity for Sm and subset identity integrate to determine anergy’s effectiveness, affinity threshold, and mechanism (Ramiro Diz and S. H. Clarke, manuscript submitted). Low-affinity anti-Sm B cells differentiate to only the FO B cell stage, but with increasing affinity for Sm, positive selection into the MZ and B-1 B cell subsets occurs. Increasing affinity also improves anergy’s effectiveness. High-affinity anti-Sm FO B cells are more hyporesponsive to TLR activation than those of low affinity. Anti-Sm MZ B cells show the same trend; low-affinity anti-Sm MZ B cells are functional and high-affinity anti-Sm MZ B cells are anergic. We conclude that subset differentiation and anergy are independently regulated and that BCR affinity for Sm controls anergy’s effectiveness. Subset identity has important control over anergy in two ways: setting the affinity threshold for anergy induction and determining mechanism (Ramiro Diz and S. H. Clarke, manuscript submitted). As is evident in 2-12H and 2-12H/Vj8 mice, low-affinity anti-Sm FO B cells are anergic, while anti-Sm MZ B cells of similar affinity are functional. In contrast, both the high-affinity anti-Sm FO and MZ B cells of 2-12H/Vj4 mice are anergic. Thus, we suggest that FO and MZ B cells have different affinity thresholds for anergy induction. 2-12H mice produce anti-Sm B cells that exceed the affinity threshold for FO B cell anergy induction, but not that for MZ B cell anergy induction, whereas 2-12H/Vj4 mice generate anti-Sm B cells that exceed the threshold for both FO and MZ B cell anergy induction. From a general perspective, differences between subsets in affinity thresholds for anergy induction can explain the higher degree of autoreactivity among MZ B cells than among FO B cells in normal mice [33]. This independent control may be vital to the differences in function between B cell subsets. Figure 2 illustrates the relationship between affinity, subset differentiation, and anergy. In addition to determining the affinity threshold for anergy induction, subset identity determines the mechanism of anergy. First, the mechanism of anergy in FO B cells changes during differentiation from a transitional B cell to a FO B cell. At the transitional B cell stage, hyporesponsiveness results from a failure to survive, but once these B cells reach the FO B cell stage, their ability to survive TLR activation improves markedly. At the FO B cell stage, TLR hyporesponsiveness is due primarily to an inability to activate the PC transcriptional program. BCR affinity alters the effectiveness of the transitional and FO B cell-specific mechanisms, but it does not change whether TLR hyporesponsiveness is dependent primarily on survival or the inability to activate the PC transcriptional program. Mechanistic differences are also evident between anti-Sm FO and MZ B cells. Both

Differentiation

207

Competition

Immunol Res (2008) 41:203–216

-

FO MZ

+ +

FO MZ FO

Function

MZ Affinity 2-12H/V 8

2-12H/V 4

Fig. 2 Affinity and subset identity integrate to determine the affinity threshold of B cell anergy. BCR affinity for Sm increases from left to right as indicated by the triangle at the bottom. Horizontal bars indicate the affinity range of differentiation and function. The absence of color in the bars indicates the lack of differentiation or function. Competition with non-autoreactive B cells affects differentiation and therefore we show differentiation in the presence and absence of competition. Brackets at the bottom indicate the affinity range for 2-12H/Vj8 and 2-12H/Vj4 anti-Sm B cells

anti-Sm FO and MZ B cells expressing the identical BCR undergo cell death in response to BCR crosslinking. However, BCR crosslinking induces a BCR signal in anti-Sm FO B cells, but not anti-Sm MZ B cells. Thus, the block in BCR-mediated activation of anergic FO and MZ B cells occurs at different levels in the BCR signaling cascade. For anti-Sm FO B cells, blockade affects a discrete survival pathway, whereas for anti-Sm MZ B cells blockade is due to uncoupling of the BCR from its signalsome. Thus, chronically stimulated autoreactive B cells of the FO and MZ subsets are inherently different in the mechanism of anergy. This may explain at least partially the mechanistic differences between tolerance models since, in some cases, autoreactive B cells reach only a transitional B cell stage, and in others they reach the FO B cell stage [19, 22, 29]. Competition for access to survival factors is also a mechanism of autoreactive B cell regulation. Cyster and colleagues have demonstrated that anergic B cells specific for hen egg lysozyme (HEL) have an increased dependence on B cell activating factor (BAFF) levels [37]. BAFF is a critical survival factor and is required for FO and MZ B cell differentiation. In the presence of large numbers of non-autoreactive B cells, anergic antiHEL B cells are unable to access sufficient BAFF and consequently undergo cell death at an immature stage. Competition plays an important role in anti-Sm B cell regulation. However, low-affinity anti-Sm FO and MZ B cells are competitive [19], but those of high affinity are not (Rhea Busick, Ramiro Dis, and S. H. Clarke, manuscript in preparation). We attribute this non-competitiveness to poor BAFF responsiveness, as BAFF-induced signaling in high-affinity anti-Sm B cells does not activate the Akt survival pathway. An understanding of how BCR signals in these cells control activation of the Akt pathway by BAFF may provide a new therapeutic target to control autoreactive B cell activation.

A tolerance checkpoint after antigen activation B cell tolerance mechanisms act at a number of differentiative stages or checkpoints. Since anti-Sm B cells are positively selected into the MZ and B-1 B cell subsets, but are not activated to become ASCs, we speculated that a tolerance checkpoint exists at a later stage, perhaps after activation by self-antigen. Evidence of this is our finding that anti-Sm B cells

208

Immunol Res (2008) 41:203–216

constitutively differentiate to an early pre-PC stage [27]. These early pre-PCs exhibit both B cell- and PC-specific features. They have normal or near normal levels of the B cell markers CD19, B220, and IgM, but express an intermediate level of the PC marker CD138 (CD138int) and are larger and more granular than splenic B cells. In addition, they are located predominantly along the B–T cell border and near the bridging channels, as expected of antigen-activated B cells and, consistent with this, they have increased the expression of the chemokine receptor CXCR5 and decreased the expression of CXCR4. Both FO and MZ B cells can become CD138int B cells (Kara L. Conway and S. H. Clarke, manuscript in preparation), but based on a functional analysis of pre-PCs (Ramiro Diz and S. H. Clarke, manuscript submitted), the majority in the steady state are probably of MZ B cell origin. They do not secrete antibody and have a high turnover rate (estimated half-life of *7 days). Thus, activation of anti-Sm MZ B cells occurs constitutively in non-autoimmune mice, but deletion before they reach the ASC stage prevents autoantibody production. This reveals the existence of a previously unidentified tolerance checkpoint at the pre-PC stage. In non-Tg mice, pre-PCs are enriched in anti-Sm B cells suggesting regulation of endogenous anti-Sm B cells occurs at this checkpoint. This is likely a significant checkpoint in autoimmunity, since, as discussed below, bypass of this checkpoint occurs in murine SLE [21].

Activation of anti-Sm B cells due to apoptotic cell clearance defects The long-term goal of this kind of research is to understand how tolerance is lost in autoimmunity. Because subset identity determines the mechanism of anergy, a critical first step is to identify the B cell subsets involved. To address this question, we examined the anti-Sm B cells of each subset in multiple mouse models of SLE. These include 2-12H mice of the autoimmune MRL genetic background (2-12H MRL) [20, 21], and 2-12H mice deficient in MerTK (2-12H/MerTKkd) [13], a receptor tyrosine kinase important for macrophage phagocytosis of apoptotic cells [38], and Fas (2-12H/Faslpr) [21], a pro-apoptotic receptor important to immune cell regulation [39]. MerTK and Fas deficiencies induce a mild SLE-like disease [38, 40], while the MRL background, which includes multiple unidentified SLE-susceptibility genes, induces a severe SLE-like disease, particularly when combined with Faslpr [40]. 2-12H mice of all three models develop high anti-Sm titers by 2–3 months of age with complete penetrance [21]. Coincident with the appearance of serum anti-Sm, anti-Sm MZ, and/or B-1 B cells are progressively lost due to activation and PC differentiation. Non-Sm binding MZ and B-1 B cells do not decrease in number suggesting that the loss of anti-Sm MZ and B-1 B cells is self-antigen driven. Although activation of anti-Sm MZ B cells does not require overcoming tolerance, differentiation to the PC stage requires bypass of the pre-PC checkpoint. This is demonstrated directly in 212H MRL/Faslpr mice by the more PC-like phenotype of anti-Sm CD138int pre-PCs compared to those in non-autoimmune mice and by the presence of a more activated antiSm CD138hi pre-PC subset [27]. Moreover, a high frequency of CD138int and CD138hi pre-PCs are ASCs. In addition, to MZ and B-1 B cell activation, Faslpr induces activation of FO B cells. Adoptive transfer experiments with anti-Sm FO B cells to autoimmune Faslpr mice indicates activation of anti-Sm FO B cells in Faslpr mice (Kara L. Conway and S. H. Clarke, manuscript in preparation). Since FO B cells are anergic, activation of these cells requires overcoming anergy. Thus, B cell tolerance to Sm is broken at two checkpoints in autoimmune mice, at the FO B cell stage to break anergy, and at the pre-PC stage to allow PC differentiation.

Immunol Res (2008) 41:203–216

209

Transfer of 2-12H B cells subsets to autoimmune Faslpr mice reveals that the B cells of each subset make different contributions to the anti-Sm response. Anti-Sm B-1 cells generate PCs of the lamina propria (LP) and mesenteric lymph node (MLN), but not the spleen, indicating that they contribute predominantly to a gut-associated response [21]. In contrast, anti-Sm FO, MZ, and pre-PCs generate splenic PCs, and not LP or MLN PCs (Kara L. Conway and S. H. Clarke, manuscript in preparation). The anti-Sm PCs of MZ B cell and pre-PC origin form rapidly after transfer; they are present by day 4 post-transfer, but gone by day 12. Anti-Sm FO B cells produce a delayed, but long-lived response; PCs are not detectable on day 4 post-transfer, but are detectable on day 12 and in reduced numbers on day 30 post-transfer. Thus, in autoimmunity, the contributions of anti-self B cells of each subset parallel their contributions to foreign antigen responses; MZ, pre-PCs, and B-1 B cells rapidly generate short-lived PCs, while FO B cells generate a delayed, but long-lived response, possibly involving germinal center formation. The autoantibodies produced by B cells of each subset are not likely to be equally pathogenic. Because MZ and B-1 B cells generally do not form germinal centers, as suggested by the rapid transient response after transfer to Faslpr mice, they are likely to produce mostly unmutated IgM autoantibodies similar to the non-pathogenic autoantibodies produced by apoptotic cell immunization of non-autoimmune mice. On the other hand, germinal center formation by FO B cells, as suggested by the delayed and prolonged response after transfer to Faslpr mice, may yield high-affinity IgG autoantibodies, some of which may be pathogenic. Thus, the FO B cell response is likely to be the more consequential to autoimmunity.

Dendritic cells regulate anti-Sm B cell activation Although the central role of DCs in T cell activation is well established, their direct role in B cell activation is less well established. DCs can induce Ig class-switch, B cell activation, and antibody production [41–43]. Moreover, using 2-photon intravital imaging, Germain and colleagues demonstrated that antigen-bearing DCs directly engage B cells resulting in B cell Ca2+ signaling, antigen acquisition, and extrafollicular accumulation [43]. Balazs et al. demonstrated a particular relevance of DCs to MZ B cell activation [44]. They showed that blood DCs can capture bacteria and migrate to the marginal zone where they activate MZ B cells specific for a bacteria associated antigen by a BAFF-dependent mechanism. Thus, DCs participate in B cell response indirectly through the activation of T cells, and directly through DC–B cell interaction. The activation of MZ B cells by DCs to foreign antigen demonstrated by Balazs et al. [44] may be relevant to the activation of anti-Sm MZ B cells, since DCs of the marginal zone can capture and phagocytize apoptotic cells. We tested this by both in vivo and in vitro experiments and find that DCs previously fed apoptotic cells induce anti-Sm ASC formation (Kara L. Conway and S. H. Clarke, manuscript in preparation). DCs not pulsed with apoptotic cells do not induce anti-Sm ASCs. The apoptotic cell-pulsed DCs are immature in phenotype, and TNFa-induced DC maturation completely abrogates their ability to induce anti-Sm ASCs. Activation is antigen specific, since activation of non-Sm binding B cells by apoptotic cell pulsed DCs does not occur, and activation is T cell dependent, since anti-Sm ASC formation does not occur in T cell-depleted mice. However, the T cells need not be antigen specific, which together with the inability of mature DCs to induce anti-Sm ASCs, argues against conventional T cell help in B cell activation. Thus, DCs can activate anti-Sm B cells, but this ability is restricted to those with an immature phenotype that have phagocytized apoptotic cells.

210

Immunol Res (2008) 41:203–216

Apoptotic cell-pulsed DCs activate only MZ B cells and induce their differentiation to the pre-PC stage. In the absence of T cells, pre-PC differentiation occurs normally, but these cells do not bypass the tolerance checkpoint at this stage indicating that signals provided by T cells are crucial to bypass of the pre-PC tolerance checkpoint. Optimal activation requires cell–cell contact, which may explain the requirement for CD40/CD40L and Fas/FasL interactions may explain. In addition, DCs have intact Sm on their surface (Diane Gutches and Barbara Vilen, pers. comm.), which would provide the self-antigen for B cell activation and account for the requirement for direct DC–B cell interaction. Moreover, it would explain the antigen specificity of this activation pathway. Activation also requires DC secretion of the cytokines IL-1b and BAFF. This requirement for BAFF parallels MZ B cell activation by DCs that have captured bacteria [44]. Our current efforts are to determine the mechanism of anti-Sm MZ B cell activation by DCs. These data suggest a model for anti-Sm B cell activation in non-autoimmune mice. Infection results in increased apoptosis and DCs in the marginal zone will capture apoptotic cells that enter the blood. DCs constitutively have intact Sm on their cell surface, which increases upon apoptotic cell phagocytosis (Diane Gutches and Barbara Vilen, pers. comm.), and thus DC–MZ B cell interaction delivers a B cell receptor (BCR) activation signal. DC binding of apoptotic cells delivers a BAFF secretion signal, since we find that optimal BAFF secretion requires apoptotic cell exposure. Activated MZ B cells differentiate to pre-PCs, which migrate to the B–T cell border of the follicles, possibly in tandem, as shown for other DC–B cell interactions [43]. Subsequent interaction with T cells induces bypass of the pre-PC tolerance checkpoint and PC differentiation, which migrate through bridging channels to the red pulp. We can only speculate as to the value of anti-Sm B cell activation by DCs. Activation in response to apoptotic cell phagocytosis suggests that anti-Sm and other apoptotic cellspecific antibodies aid in the clearance of excess apoptotic cells generated during infection. Increased apoptotic cell loads are associated with anti-Sm production, as suggested by our analysis of MerTK-deficient mice [13], and defects in apoptotic cell clearance are associated with murine and human SLE [38, 45, 46]. Restriction to immature DCs would prohibit concomitant activation of anti-Sm T cells or T cells specific for other apoptotic cell-associated self-antigens, thereby resulting in formation of only short-lived PCs and no memory B cells. Although T cells are required for anti-Sm MZ B cell activation, that they need not be antigen-specific suggests that the nature of their involvement is different from that required for germinal center formation. Thus, formation of only short-lived PCs ensures that the response will lapse once apoptotic cell availability decreases. In contrast to DCs from non-autoimmune mice, those from autoimmune Faslpr mice constitutively activate anti-Sm MZ B cells (Kara L. Conway and S. H. Clarke, manuscript submitted). Faslpr DCs activate anti-Sm MZ B cells independent of apoptotic cell phagocytosis and regardless of their maturation stage. Thus, Fas signals are vital to restricting the ability of DCs to activate MZ B cells. Fas is an activating receptor on DCs [47], and signaling through Fas induces cytokine secretion and co-stimulatory molecule expression. The mechanism for this constitutive ability is under investigation, but like activation by Faswt DCs, Faslpr DC-induced activation is CD40L, IL-1b, and BAFF dependent. DC activation of anti-Sm MZ B cells unconstrained by apoptotic cell exposure or maturation stage has important implications for autoantibody production in disease. Autoimmune DCs would be able to activate anti-Sm MZ B cells in the absence of increased apoptotic cell availability. Although immature DCs would generate only shortlived PCs, continual PC generation would result in constitutively high-serum anti-Sm

Immunol Res (2008) 41:203–216

211

titers. The absence of affinity maturation of this response means that the antibodies produced by constitutive MZ B cell activation are unlikely to be pathogenic. However, the ability of mature Faslpr DCs to activate MZ B cells would allow concomitant anti-Sm B and T cell activation, resulting in germinal center formation and the production of longlived PCs and memory B cells. Germinal centers are sites of affinity maturation and highaffinity autoantibodies produced by long-lived PCs likely have a greater probability of being pathogenic. This predicts that autoimmune MRL/Faslpr and Faslpr mice have two anti-Sm B cell populations: one producing predominantly unmutated, IgM antibodies of low affinity, and the second producing mutated, IgG antibodies of high affinity. Our previous analysis of anti-Sm hybridomas from MRL/Faslpr mice is consistent with this prediction [18]. As is typical of secondary responses to foreign antigen in non-autoimmune mice, multiple sets of clonally related B cells compose the anti-Sm hybridoma panel derived from individual MRL/Faslpr mice, and many clonal sets were mutated IgG producing hybridomas. However, unlike responses to foreign antigen, the anti-Sm clonal sets consisted of many that were unmutated IgM producing clones indicating extensive clonal expansion with little or no somatic mutation. We propose that these IgM clones derive from non-germinal center responses, while the mutated IgG clones derive from the germinal center responses. Anti-DNA and RF hybridomas from autoimmune mice also exhibit evidence of these parallel responses [48, 49] suggesting dual DC-mediated activation pathways are not unique to anti-Sm B cells. These data demonstrate that DCs have a central role in regulating the anti-Sm response in non-autoimmune and autoimmune settings. In addition, they reveal that Fas signaling restricts the ability to activate anti-Sm MZ B cells to immature DCs that have phagocytized apoptotic cells. In the absence of this control, constitutive activation of anti-Sm MZ B cells occurs. These findings identify a new mechanism for autoreactive B cell activation in disease resulting from defective DC–B cell interactions, and reveal a new function for Fas in preventing pathogenic autoantibody production.

Autoreactive B cells in human SLE: a translational study An inability to discriminate between autoreactive and non-autoreactive B cells impedes our understanding of B cells in human SLE. Multiple studies have documented differences in the B cell subsets that are present in peripheral blood of SLE patients. In comparison to healthy control B cells, SLE patient B cells are hyperactive, susceptible to antigen-induced cell death, and show elevated Ca++ responses and hyperphosphorylation of cytosolic proteins [50–54]. SLE patients also have higher numbers of memory B cells, activated B cells, and B-1 B cells [55–59]. However, the relevance of any of these differences to disease is not established. We have undertaken an analysis of B cells in human SLE, and identified a unique population of B cells in a subset of SLE patients and in a subset of patients with antineutrophil cytoplasmic autoantibody (ANCA)-vasculitis [60]. ANCA-vaculitis is similar to SLE in that it too is characterized by pathogenic autoantibodies (specific for neutrophil intracellular proteins myeloperoxidase and proteinase 3) [61, 62], has a relapsing and remitting course, and is treated with immunosuppressive regimes [63]. The defining characteristic of this unique population is the expression of high levels of CD19 (CD19hi), which is 2- to 3-fold higher than on healthy control B cells. In both diseases, 20–30% of patients have elevated numbers of CD19hi B cells, and among this group, the percentage of peripheral blood B cells that are CD19hi ranges from 7% to 50%.

212

Immunol Res (2008) 41:203–216

CD19hi B cells are somatically mutated memory B cells that are predominantly IgG. Biases in the distribution of mutations and in VH gene use by CD19hi B cells of one patient suggest that these cells have undergone antigen selection [60]. CD19hi B cells appear to be functional; they become ASCs in response to BCR crosslinking and cytokine stimulation, phosphorylate multiple signaling intermediates (CD19, p38, JNK, and Akt) in response to BCR crosslinking, and exhibit high basal levels of phosphorylated Syk (pSyk) and ERK (pERK) [64]. The high basal levels of pSyk and pERK also suggest recent activation by antigen. Consistent with this, CD19hi B cells express low levels of the B cell-specific transcription factor PAX-5 and high levels of the PC transcription factor XBP-1 [64], and are somewhat larger and more granular than naı¨ve B cells [60]. Thus, we propose that CD19hi B cells are antigen selected memory B cells of germinal center origin that have just begun PC differentiation in response to antigen encounter. To investigate the significance of CD19hi B cells to disease, we tested for associations between their presence in circulation and clinical findings. We find no evidence of an association with disease severity in a cross-sectional study of 41 SLE patients [60]. However, there is a significant correlation with end stage renal disease (ESRD) (P = 0.04) and severe neurological manifestations of seizures and psychoses (P = 0.01) [64]. In addition, we found significant associations with the presence of certain autoantibodies. Proteome array analysis with 67 autoantigens indicates a significant correlation between the presence of a circulating CD19hi population and high levels of autoantibodies specific for small nuclear RNPs, particularly Sm [60]. In addition, they correlate negatively with high levels of anti-glomerular autoantibodies. Altogether, the CD19hi population identifies a subset of SLE patients likely to experience more severe disease outcomes, suggesting that they play a role in disease pathogenesis. A role in disease pathogenesis predicts an enrichment of autoreactive B cells in the CD19hi population. Since high anti-Sm titers correlate strongly with the presence of CD19hi B cells in circulation, we polyclonally activated CD19hi and non-CD19hi B cells from four patients in vitro and determined the frequency of anti-Sm B cells in each population by ELISpot. We find an enrichment of anti-Sm B cells in the CD19hi B cell subset [64]. Moreover, the frequency of anti-Sm B cells in this subset correlates with the log of the serum anti-Sm titer, such that a doubling of the frequency of anti-Sm B cells in the CD19hi population results in a nearly 100-fold increase in serum anti-Sm titer. This indicates that CD19hi B cells undergo extensive clonal expansion before becoming PCs. Thus, these data establish a link between CD19hi B cells and serum autoantibody production, indicating that CD19hi B cells contribute to autoimmunity. Our working model, illustrated in Fig. 3, is that CD19hi B cells are autoreactive memory B cells that have recently exited lymphoid tissues and entered the circulation due to antigen activation. They may be newly formed from germinal centers or be reactivated memory B cells generated earlier. Based on [10-fold increase in expression of the chemokine receptor CXCR3 and their responsiveness to its ligand CXCL9 [64], we suggest that CD19hi B cells are homing to sites of inflammation. They express little CXCR4, the chemokine receptor required for bone marrow homing, and do not respond to its ligand CXCL12, indicating that they are not likely to be in transit to the bone marrow [64]. The CXCR3 ligand CXCL9 is present in the kidney and CNS of SLE patients [65, 66], and thus, CD19hi B cells may be in transit to these sites, explaining their association with ESRD and neurological manifestations. At these sites, the CD19hi B cells undergo significant clonal expansion before becoming PCs. The pathology associated with these cells may be due to localized high autoantibody levels or to the production of cytokines and activation of other inflammatory cells.

Immunol Res (2008) 41:203–216

213

Lymphoid tissue CD19hi

CD19lo naive

Self antigen

Inflammatory site

Self antigen

Circulation

reactivated memory

new memory

Germinal center CD19hi

size, granularity act marker CD21 CD20 pSyk, pERK CXCR3 CXCR4, CXCR5

T T T T T

Clonal expansion

Plasma cells

CD19lo

Fig. 3 Recently activated CD19hi B cells enter the circulation and traffic to sites of inflammation. CD19hi B cells are newly generated memory B cells that arise from a germinal center in response to self-antigen or self-antigen reactivated memory B cells. They differ as indicated from circulating naı¨ve B cells. Based on chemokine receptor expression, we predict that these cells enter sites of inflammation rather than the bone marrow where they undergo clonal expansion before becoming PCs

In summary, we have established a link between CD19hi B cells and autoantibody production in autoimmunity. These findings point to CD19hi B cells as a therapeutic target in SLE, ANCA vasculitis, and possible other autoimmune diseases. In the meantime, they may serve as a marker to identify patients that are at increased risk of severe clinical outcomes and that may benefit from treatment that is more aggressive. Acknowledgements I am indebted to all of the past and present members of my laboratory for their dedication and perseverance that has made this progress possible. I am also indebted to Dr. Barbara Vilen and the members of her laboratory for their advice and many helpful discussions. NIAID, the Arthritis Foundation, and the Lupus Foundation of America have funded this work.

References 1. Hahn BH. Antibodies to DNA. N Engl J Med. 1998;338(19):1359–68. 2. Su W, Madaio MP. Recent advances in the pathogenesis of lupus nephritis: autoantibodies and B cells. Semin Nephrol. 2003;23(6):564–8. 3. Ciruelo E, de la Cruz J, Lopez I, Gomez-Reino JJ. Cumulative rate of relapse of lupus nephritis after successful treatment with cyclophosphamide. Arthritis Rheum. 1996;39(12):2028–34. 4. Illei GG, Takada K, Parkin D, Austin HA, Crane M, Yarboro CH, et al. Renal flares are common in patients with severe proliferative lupus nephritis treated with pulse immunosuppressive therapy: longterm followup of a cohort of 145 patients participating in randomized controlled studies. Arthritis Rheum. 2002;46(4):995–1002. 5. Nachman PH, Hogan SL, Jennette JC, Falk RJ. Treatment response and relapse in antineutrophil cytoplasmic autoantibody-associated microscopic polyangiitis and glomerulonephritis. J Am Soc Nephrol. 1996;7(1):33–9. 6. Harley JB, Alarcon-Riquelme ME, Criswell LA, Jacob CO, Kimberly RP, Moser KL, et al. Genomewide association scan in women with systemic lupus erythematosus identifies susceptibility variants in ITGAM, PXK, KIAA1542 and other loci. Nat Genet. 2008;40(2):204–10. 7. Hom G, Graham RR, Modrek B, Taylor KE, Ortmann W, Garnier S, et al. Association of systemic lupus erythematosus with C8orf13-BLK and ITGAM-ITGAX. N Engl J Med. 2008;358(9):900–9. 8. Truedsson L, Bengtsson AA, Sturfelt G. Complement deficiencies and systemic lupus erythematosus. Autoimmunity. 2007;40(8):560–6. 9. Manderson AP, Botto M, Walport MJ. The role of complement in the development of systemic lupus erythematosus. Annu Rev Immunol. 2004;22:431–56.

214

Immunol Res (2008) 41:203–216

10. Hart SP, Smith JR, Dransfield I. Phagocytosis of opsonized apoptotic cells: roles for ‘old-fashioned’ receptors for antibody and complement. Clin Exp Immunol. 2004;135(2):181–5. 11. Casciola-Rosen LA, Anhalt G, Rosen A. Autoantigens targeted in systemic lupus erythematosus are clustered in two populations of surface structures on apoptotic keratinocytes. J Exp Med. 1994; 179(4):1317–30. 12. Rosen A, Casciola-Rosen L. Autoantigens as substrates for apoptotic proteases: implications for the pathogenesis of systemic autoimmune disease. Cell Death Differ. 1999;6(1):6–12. 13. Qian Y, Wang H, Clarke SH. Impaired clearance of apoptotic cells induces the activation of autoreactive anti-Sm marginal zone and B-1 B cells. J Immunol. 2004;172(1):625–35. 14. Mevorach D, Zhou JL, Song X, Elkon KB. Systemic exposure to irradiated apoptotic cells induces autoantibody production. J Exp Med. 1998;188(2):387–92. 15. Yong J, Wan L, Dreyfuss G. Why do cells need an assembly machine for RNA-protein complexes? Trends Cell Biol. 2004;14(5):226–32. 16. Tan EM. Antinuclear antibodies: diagnostic markers for autoimmune diseases and probes for cell biology. Adv Immunol. 1989;44:93–151. 17. Santulli-Marotto S, Retter MW, Gee R, Mamula MJ, Clarke SH. Autoreactive B cell regulation: peripheral induction of developmental arrest by lupus-associated autoantigens. Immunity. 1998; 8(2):209–19. 18. Bloom DD, Davignon JL, Retter MW, Shlomchik MJ, Pisetsky DS, Cohen PL, et al. V region gene analysis of anti-Sm hybridomas from MRL/Mp-lpr/lpr mice. J Immunol. 1993;150(4):1591–610. 19. Borrero M, Clarke SH. Low-affinity anti-Smith antigen B cells are regulated by anergy as opposed to developmental arrest or differentiation to B-1. J Immunol. 2002;168(1):13–21. 20. Santulli-Marotto S, Qian Y, Ferguson S, Clarke SH. Anti-Sm B cell differentiation in Ig transgenic MRL/Mp-lpr/lpr mice: altered differentiation and an accelerated response. J Immunol. 2001;166(8): 5292–9. 21. Qian Y, Conway KL, Lu X, Seitz HM, Matsushima GK, Clarke SH. Autoreactive MZ and B-1 B-cell activation by Faslpr is coincident with an increased frequency of apoptotic lymphocytes and a defect in macrophage clearance. Blood. 2006;108(3):974–82. 22. Goodnow CC, Cyster JG, Hartley SB, Bell SE, Cooke MP, Healy JI, et al. Self-tolerance checkpoints in B lymphocyte development. Adv Immunol. 1995;59:279–368. 23. Gay D, Saunders T, Camper S, Weigert M. Receptor editing: an approach by autoreactive B cells to escape tolerance. J Exp Med. 1993;177(4):999–1008. 24. Nemazee D, Weigert M. Revising B cell receptors. J Exp Med. 2000;191(11):1813–7. 25. Mandik-Nayak L, Bui A, Noorchashm H, Eaton A, Erikson J. Regulation of anti-double-stranded DNA B cells in nonautoimmune mice: localization to the T-B interface of the splenic follicle. J Exp Med. 1997;186(8):1257–67. 26. Cyster JG, Hartley SB, Goodnow CC. Competition for follicular niches excludes self-reactive cells from the recirculating B-cell repertoire. Nature. 1994;371(6496):389–95. 27. Culton DA, O’Conner BP, Conway KL, Diz R, Rutan J, Vilen BJ, et al. Early preplasma cells define a tolerance checkpoint for autoreactive B cells. J Immunol. 2006;176(2):790–802. 28. William J, Euler C, Primarolo N, Shlomchik MJ. B cell tolerance checkpoints that restrict pathways of antigen-driven differentiation. J Immunol. 2006;176(4):2142–51. 29. Merrell KT, Benschop RJ, Gauld SB, Aviszus K, Decote-Ricardo D, Wysocki LJ, et al. Identification of anergic B cells within a wild-type repertoire. Immunity. 2006;25(6):953–62. 30. Kilmon MA, Rutan JA, Clarke SH, Vilen BJ. Low-affinity, Smith antigen-specific B cells are tolerized by dendritic cells and macrophages. J Immunol. 2005;175(1):37–41. 31. Qian Y, Santiago C, Borrero M, Tedder TF, Clarke SH. Lupus-specific antiribonucleoprotein B cell tolerance in nonautoimmune mice is maintained by differentiation to B-1 and governed by B cell receptor signaling thresholds. J Immunol. 2001;166(4):2412–9. 32. Kilmon MA, Wagner NJ, Garland AL, Lin L, Aviszus K, Wysocki LJ, et al. Macrophages prevent the differentiation of autoreactive B cells by secreting CD40 ligand and IL-6. Blood. 2007;110:1595–602. 33. Lopes-Carvalho T, Kearney JF. Development and selection of marginal zone B cells. Immunol Rev. 2004;197:192–205. 34. Lopes-Carvalho T, Kearney JF. Marginal zone B cell physiology and disease. Curr Dir Autoimmun. 2005;8:91–123. 35. Oliver AM, Martin F, Gartland GL, Carter RH, Kearney JF. Marginal zone B cells exhibit unique activation, proliferative and immunoglobulin secretory responses. Eur J Immunol. 1997;27(9):2366–74. 36. Quartier P, Potter PK, Ehrenstein MR, Walport MJ, Botto M. Predominant role of IgM-dependent activation of the classical pathway in the clearance of dying cells by murine bone marrow-derived macrophages in vitro. Eur J Immunol. 2005;35(1):252–60.

Immunol Res (2008) 41:203–216

215

37. Lesley R, Xu Y, Kalled SL, Hess DM, Schwab SR, Shu HB, Cyster JG. Reduced competitiveness of autoantigen-engaged B cells due to increased dependence on BAFF. Immunity. 2004;20(4):441–53. 38. Scott RS, McMahon EJ, Pop SM, Reap EA, Caricchio R, Cohen PL, et al. Phagocytosis and clearance of apoptotic cells is mediated by MER. Nature. 2001;411(6834):207–11. 39. Watanabe-Fukunaga R, Brannan CI, Copeland NG, Jenkins NA, Nagata S. Lymphoproliferation disorder in mice explained by defects in Fas antigen that mediates apoptosis. Nature. 1992;356(6367):314–7. 40. Theofilopoulos AN, Dixon FJ. Murine models of systemic lupus erythematosus. Adv Immunol. 1985;37:269–390. 41. Gerloni M, Lo D, Zanetti M. DNA immunization in relB-deficient mice discloses a role for dendritic cells in IgM?IgG1 switch in vivo. Eur J Immunol. 1998;28(2):516–24. 42. MacPherson G, Kushnir N, Wykes M. Dendritic cells, B cells and the regulation of antibody synthesis. Immunol Rev. 1999;172:325–34. 43. Qi H, Egen JG, Huang AY, Germain RN. Extrafollicular activation of lymph node B cells by antigenbearing dendritic cells. Science. 2006;312(5780):1672–6. 44. Balazs M, Martin F, Zhou T, Kearney J. Blood dendritic cells interact with splenic marginal zone B cells to initiate T-independent immune responses. Immunity. 2002;17(3):341–52. 45. Baumann I, Kolowos W, Voll RE, Manger B, Gaipl U, Neuhuber WL, et al. Impaired uptake of apoptotic cells into tingible body macrophages in germinal centers of patients with systemic lupus erythematosus. Arthritis Rheum. 2002;46(1):191–201. 46. Lu Q, Lemke G. Homeostatic regulation of the immune system by receptor tyrosine kinases of the Tyro 3 family. Science. 2001;293(5528):306–11. 47. Guo Z, Zhang M, An H, Chen W, Liu S, Guo J, et al. Fas ligation induces IL-1beta-dependent maturation and IL-1beta-independent survival of dendritic cells: different roles of ERK and NF-kappaB signaling pathways. Blood. 2003;102(13):4441–7. 48. Shlomchik MJ, Aucoin AH, Pisetsky DS, Weigert MG. Structure and function of anti-DNA autoantibodies derived from a single autoimmune mouse. Proc Natl Acad Sci USA. 1987;84(24):9150–4. 49. Shlomchik MJ, Marshak-Rothstein A, Wolfowicz CB, Rothstein TL, Weigert MG. The role of clonal selection and somatic mutation in autoimmunity. Nature. 1987;328(6133):805–11. 50. Blaese RM, Grayson J, Steinberg AD. Increased immunoglobulin-secreting cells in the blood of patients with active systemic lupus erythematosus. Am J Med. 1980;69(3):345–50. 51. Budman DR, Merchant EB, Steinberg AD, Doft B, Gershwin ME, Lizzio E, et al. Increased spontaneous activity of antibody-forming cells in the peripheral blood of patients with active SLE. Arthritis Rheum. 1977;20(3):829–33. 52. Suzuki H, Sakurami T, Imura H. Relationship between reduced B cell susceptibility to IgM antibodies and reduced IgD-bearing B cells in patients with systemic lupus erythematosus. Arthritis Rheum. 1982;25(12):1451–9. 53. Liossis SN, Kovacs B, Dennis G, Kammer GM, Tsokos GC. B cells from patients with systemic lupus erythematosus display abnormal antigen receptor-mediated early signal transduction events. J Clin Invest. 1996;98(11):2549–57. 54. Tsokos GC, Wong HK, Enyedy EJ, Nambiar MP. Immune cell signaling in lupus. Curr Opin Rheumatol. 2000;12(5):355–63. 55. Bijl M, Horst G, Limburg PC, Kallenberg CG. Expression of costimulatory molecules on peripheral blood lymphocytes of patients with systemic lupus erythematosus. Ann Rheum Dis. 2001;60(5):523–6. 56. Folzenlogen D, Hofer MF, Leung DY, Freed JH, Newell MK. Analysis of CD80 and CD86 expression on peripheral blood B lymphocytes reveals increased expression of CD86 in lupus patients. Clin Immunol Immunopathol. 1997;83(3):199–204. 57. Odendahl M, Jacobi A, Hansen A, Feist E, Hiepe F, Burmester GR, et al. Disturbed peripheral B lymphocyte homeostasis in systemic lupus erythematosus. J Immunol. 2000;165(10):5970–9. 58. Jacobi AM, Odendahl M, Reiter K, Bruns A, Burmester GR, Radbruch A, et al. Correlation between circulating CD27 high plasma cells and disease activity in patients with systemic lupus erythematosus. Arthritis Rheum. 2003;48(5):1332–42. 59. Smith HR, Olson RR. CD5+ B lymphocytes in systemic lupus erythematosus and rheumatoid arthritis. J Rheumatol. 1990;17(6):833–5. 60. Culton DA, Nicholas MW, Bunch DO, Zhen QL, Kepler TB, Dooley MA, et al. Similar CD19 dysregulation in two autoantibody-associated autoimmune diseases suggests a shared mechanism of B-cell tolerance loss. J Clin Immunol. 2007;27(1):53–68. 61. Falk RJ, Jennette JC. Anti-neutrophil cytoplasmic autoantibodies with specificity for myeloperoxidase in patients with systemic vasculitis and idiopathic necrotizing and crescentic glomerulonephritis. N Engl J Med. 1988;318(25):1651–7.

216

Immunol Res (2008) 41:203–216

62. Jennette JC, Hoidal JR, Falk RJ. Specificity of anti-neutrophil cytoplasmic autoantibodies for proteinase 3. Blood. 1990;75(11):2263–4. 63. Jennette JC, Falk RJ. Small-vessel vasculitis. N Engl J Med. 1997;337(21):1512–23. 64. Nicholas MW, Dooley MA, Hogan SL, Anolik J, Looney J, Sanz I, et al. A novel subset of memory B cells is enriched in autoreactivity and correlates with adverse outcomes in SLE. Clin Immunol. 2008;126(2):189–201. 65. Narumi S, Takeuchi T, Kobayashi Y, Konishi K. Serum levels of ifn-inducible PROTEIN-10 relating to the activity of systemic lupus erythematosus. Cytokine. 2000;12(10):1561–5. 66. Okamoto H, Katsumata Y, Nishimura K, Kamatani N. Interferon-inducible protein 10/CXCL10 is increased in the cerebrospinal fluid of patients with central nervous system lupus. Arthritis Rheum. 2004;50(11):3731–2.