Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Oct 12.
Published in final edited form as: Immunity. 2008 Oct 2;29(4):615–627. doi: 10.1016/j.immuni.2008.07.016

Fas Receptor Expression in Germinal-Center B Cells Is Essential for T and B Lymphocyte Homeostasis

Zhenyue Hao 1,*, Gordon S Duncan 1,6, Jane Seagal 2,6, Yu-Wen Su 1, Claire Hong 1, Jillian Haight 1, Nien-Jung Chen 1, Andrew Elia 1, Andrew Wakeham 1, Wanda Y Li 1, Jennifer Liepa 1, Geoffrey A Wood 3, Stefano Casola 4, Klaus Rajewsky 2,7, Tak W Mak 1,5,7,*
PMCID: PMC3470429  NIHMSID: NIHMS196485  PMID: 18835195

SUMMARY

Fas is highly expressed in activated and germinal center (GC) B cells but can potentially be inactivated by misguided somatic hypermutation. We employed conditional Fas-deficient mice to investigate the physiological functions of Fas in various B cell subsets. B cell-specific Fas-deficient mice developed fatal lymphoproliferation due to activation of B cells and T cells. Ablation of Fas specifically in GC B cells reproduced the phenotype, indicating that the lymphoproliferation initiates in the GC environment. B cell-specific Fas-deficient mice also showed an accumulation of IgG1+ memory B cells expressing high amounts of CD80 and the expansion of CD28-expressing CD4+ Th cells. Blocking T cell-B cell interaction and GC formation completely prevented the fatal lymphoproliferation. Thus, Fas-mediated selection of GC B cells and the resulting memory B cell compartment is essential for maintaining the homeostasis of both T and B lymphocytes.

INTRODUCTION

The Fas (CD95, Apo-1) receptor induces apoptosis upon engagement by Fas ligand (FasL) and is critical for maintaining homeostasis in the peripheral lymphoid organs (Krammer, 1999, 2000). Fas-deficient lpr mice and FasL-deficient gld mice develop lymphadenopathy, splenomegaly, and a lupus-like autoimmune disease characterized by autoantibody production and the accumulation of abnormal Thy1+B220+CD4−CD8− double negative (DN) T cells (Takahashi et al., 1994; Watanabe-Fukunaga et al., 1992). In humans, Fas mutations result in an autoimmune lymphoproliferative syndrome (ALPS; Fisher et al.,1995; Rieux-Laucat et al., 1995) that resembles the disease in lpr and gld mice. B cells are thought to be critical drivers of ALPS because, in lpr mice, ablation of B cells or restoration of B cell Fas expression mostly prevents the autoimmune disease (Shlomchik et al., 1994) (Fukuyama et al., 2002). Thus, Fas deficiency in T cells is not sufficient for ALPS, and the interaction between T and B cells is required for the full-blown disorder.

A T cell stimulated by the binding of antigen to its T cell receptor (TCR) cannot become fully activated until it interacts with costimulatory molecules expressed by the antigen-presenting cell (APC) (Carreno and Collins, 2002). When this APC is a B cell, T cell-B cell interaction occurs during which intercellular contacts are formed to promote and regulate the activation, differentiation, and effector functions of T cells. The binding of CD28 expressed on activated T cells to CD80 and/or CD86 on activated B cells delivers a powerful costimulatory signal to the T cells (Lenschow et al., 1996; Linsley and Ledbetter, 1993). Additional T cell costimulation is mediated by the binding of inducible costimulator (ICOS), which is induced on T cells after activation (Hutloff et al., 1999). ICOS binds to ICOS ligand (ICOSL) constitutively expressed on B cells (Yoshinaga et al., 1999). During a primary humoral response, T cell-B cell interactions through costimulatory molecules are also required for germinal center (GC) formation and the GC reaction (Carreno and Collins, 2002). T cell dependent (Td) antigen-activated B cells enter the primary follicle and form a GC where B cell expansion, somatic hypermutation, affinity maturation, and differentiation into antibody-secreting plasma cells and memory B cells take place (Rajewsky, 1996). Within the GC, only B cells with high affinity for the cognate antigen interact with CD4+ T helper (Th) cells specific for the same antigen and thus are selected for differentiation into either memory B cells or plasma cells. Memory B cells become potent APCs with upregulated CD80 and CD86 expression and thus induce rapid activation and expansion of the Th cell population (Liu et al., 1995; Rajewsky, 1996). These activated Th cells in turn drive the accelerated generation of plasma cells and memory B cells that mediate a rapid and robust secondary antibody response.

The GC reaction also involves the negative selection of GC B cells with either suboptimal affinity for the cognate antigen or affinity for a self-antigen (autoreactive). Fas-FasL interaction has been implicated in this negative selection because, in transgenic mice, Fas-expressing autoreactive B cells are eliminated upon interaction with FasL+ CD4+ T cells (Rathmell et al., 1995). However, although GC B cells express high amounts of Fas and are susceptible to Fas-mediated apoptosis in vitro Hennino et al., 2001), the role of Fas-mediated apoptosis of GC B cells in immune responses to Td antigen is controversial. Smith et al. (1995) showed that the GC reaction and memory B cell generation were normal in immunized lpr mice. However, Takahashi et al. (2001) demonstrated that the selection of GC B cells and the resulting memory B cell repertoire were abnormal in immunized lpr mice. These and other studies show that much remains unclear about the role of Fas-mediated cell death in controlling the memory B cell repertoire in vivo.

It has not been possible to use lpr mice to clarify Fas functions in B cells because these mutants lack Fas in all tissues and have a distorted immune system dominated by the peculiar DN T cells (Cohen and Eisenberg, 1991). It is not known whether the drastic expansion of DN T cells affects the GC reaction in lpr mice. We decided to build on our previous studies of conditional Fas-deficient mice (Hao et al., 2004) and analyzed two strains of B cell-specific Fas-deficient mice: one in which the Fas gene was ablated in all B cell developmental stages and another in which Fas was deleted only in GC B cells and in the memory B cells and plasma cells derived from them. Of note, Fas expression was intact in T cells of both strains. We show that Fas-mediated apoptosis in GC B cells and the resulting memory B cell population were essential for the homeostasis of both T and B cells.

RESULTS

Ablation of Fas Specifically in B Cells Leads to Fatal Lymphocyte Infiltration

To ablate Fas specifically in B cells, we crossed Fasfl/fl mice, in which the Fas gene is flanked by two loxP sites (Hao et al., 2004), to Cd19-Cre mice to yield Fasfl/flCd19-Cre mice on the C57Bl/6 genetic background. At both the DNA and protein levels, the efficiency of Fas deletion in B cells of Fasfl/flCd19-Cre mice was >80% (Figure 1A). Monitoring of a cohort of aging Fasfl/flCd19-Cre mice together with their control Fasfl/fl and Fasfl/+Cd19-Cre littermates revealed that mutant mice of age >6 months were often smaller than controls and appeared to be weak. A few mutants had skin lesions around the neck or ears (data not shown). Approximately 80% of Fasfl/flCd19-Cre mice died between 7 and 17 months of age, whereas all control mice survived the 18 month observation period (Figure 1B). The death of the mutants most likely arose from a lymphoproliferative disorder that usually became obvious at age >10 months and manifested as splenomegaly and lymphoadenopathy (data not shown). However, unlike lpr mice, there was no expansion of the unusual DN T cell subset in Fasfl/flCd19-Cre mice (Figure S1A available online).

Figure 1. Fasfl/flCd19-Cre Mice Die Prematurely Because of Lymphocyte Infiltration in Multiple Organs.

Figure 1

Open in a new tab

(A) Confirmation of Fas deletion. The left side shows a representative Southern blot of naive FASFL/fl and Fasfl/flCd19-Cre B cells showing deletion of the Fas gene in 83% of the latter (87.3% ± 4%, n = 4). The right side shows a representative flow-cytometric analysis of Fasfl/+Cd19-Cre and Fasfl/flCd19-Cre B cells (stimulated for 2–4 days with LPS) showing that Fas protein expression was lacking in 81% of the mutant cells (85.4% ± 4.8%, n = 4).

(B) Reduced life span. Kaplan-Meier survival curves of Fasfl/flCd19-Cre mice compared to controls.

(C) Organ infiltration. Focal interstitial lymphocyte accumulation is shown in H&E-stained sections of livers, lungs, and kidneys of Fasfl/fl and Fasfl/flCd19-Cre mice at the ages of 6 and 14 months. Scale bars apply to all panels in a row.

Lpr mice die of kidney failure caused by massive autoantibody production leading to immune-complex deposition in the kidney (Cohen and Eisenberg, 1991). The amounts of serum IgM and IgG autoantibodies were greater in Fasfl/flCd19-Cre mice compared to those in Fasfl/fl controls but were still several folds lower than amounts in Fasdel/del mice in which Fas is inactivated ubiquitously (Figure S1B). This elevation in autoantibody titers paralleled a general trend in the mutants toward increased serum antibody concentrations (Figure S1C). To ascertain the cause of death of Fasfl/flCd19-Cre mice, we performed histopathologic analyses of the liver, lungs, and kidneys. No abnormalities were observed in these organs in 2-month-old Fasfl/flCd19-Cre mice (data not shown). However, in 6- and 14-month-old mutants, there were dramatic infiltrations of mononuclear cells in the liver and lungs (Figure 1C). Although there was negligible infiltration in the mutant kidney at age 6 months, infiltrates were clearly present at 14 months. Immunohistochemical staining showed that these infiltrates were composed of a mixture of T and B cells (data not shown). Further analysis of Fasfl/flCd19-Cre liver revealed that the accumulating lymphocytes took the form of multifocal periportal infiltrates that invaded and destroyed the liver parenchyma (Figure 1C). Similarly, Fasfl/flCd19-Cre lung displayed interstitial perivascular T and B cell infiltrates, and Fasfl/flCd19-Cre kidney showed focal interstitial and periglomerular infiltrates (Figure 1C). Abnormal lymphocyte infiltrates were not seen in the pancreas, stomach, brain, or heart (data not shown). Lymphocytes were rare in the nonlymphoid organs of control mice (Figure 1C). Thus, Fasfl/flCd19-Cre mice die of multiple organ failure caused by massive lymphocyte infiltration and inflammatory tissue destruction.

Activation and Hyperproliferation of Fas-Deficient B Cells and Fas-Sufficient T Cells

B cell development in the bone marrow (BM) and spleen of 2- to 3-month-old Fasfl/flCd19-Cre mice was comparable to that in Fasfl/fl controls (Supplemental Results and Figure S2). However, flow-cytometric analysis of the cells present in the enlarged lymphoid organs of Fasfl/flCd19-Cre mice showed that the total lymphocyte number in Fasfl/flCd19-Cre spleen (187.3 ± 74.7 × 106) was increased by 2-fold compared that in control spleen (91.1 ± 14.4 × 106) at 6–7 months (Figure 2A; p < 0.05). As expected, B cells were increased in number in Fasfl/flCd19-Cre spleen compared to controls (95 ± 37.2 × 106 versus 53.7 ± 5.47 × 106), although this difference was not statistically significant (p = 0.065). Unexpectedly, we also observed a 3-fold expansion in the total number of T cells of Fasfl/flCd19-Cre spleen compared to that of controls (62.3 ± 20.9 × 106 versus 22.2 ± 4.07 × 106; p < 0.05).

Figure 2. Hyperproliferation and Activation of T and B Cells and Disrupted Lymphoid-Organ Architecture in Fasfl/flCd19-Cre Mice.

Figure 2

Open in a new tab

(A) Increased cell numbers. Absolute numbers of total lymphocytes, B cells, and T cells in spleens of Fasfl/fl and Fasfl/flCd19-Cre mice at age 6–7 months (n = 3/genotype). For all figures, *p < 0.05 and **p < 0.005.

(B) Altered surface marker expression. CD4+ T cells, total T cells, and total B cells from Fasfl/fl and Fasfl/flCd19-Cre mice at the ages of 2, 6, or 14 months were examined by flow cytometry for expression of CD28, ICOS, ICOSL, and MHCII as indicated.

(C) Increased proliferation. BrdU incorporation by B220+IgM+B cells and CD4+ and CD8+ T cells was measured by flow cytometry. The percentage of BrdU+ cells is indicated.

(D and E) Sections of spleen (D) and LN (E) from 14-month-old Fasfl/fl and Fasfl/flCd19-Cre mice were stained with H&E (top and middle) or with immunohistochemical staining for detection of T and B cells (bottom). T cells that were stained positively with anti-CD3 were dark blue in color, whereas B cells that were stained positively with B220 antibody were brown. Scale bars apply to control and mutant samples in a pair.

To determine the activation status of the accumulating lymphocytes, we examined various molecules known to be important for B or T cell activation. Unstimulated B cells and T cells from Fasfl/flCd19-Cre mice showed control amounts of the expression of CD40, and of CTLA-4 and FasL, respectively (Figure S3 and data not shown). Moreover, the level of CD40 up-regulation on Fas-deficient B cells in response to anti-IgM stimulation was comparable to that of the control (Figure S3A). Similarly, anti-CD3-stimulated T cells from Fasfl/flCd19-Cre mice upregulated CD40L, CTLA-4, and FasL to amounts seen in controls (Figures S3B–S3D). However, we found that CD28, a molecule that is constitutively expressed on unstimulated CD4+ T cells appeared to be slightly upregulated on CD4+ T cells from 6- and 14-month-old Fasfl/flCd19-Cre mice compared to controls (Figure 2Bi). ICOS was already slightly upregulated on unstimulated T cells of 2-month-old mutant mice and was increased significantly compared to controls at 6 and 14 months (Figure 2Bii; p < 0.005). There was no difference in the amounts of ICOSL expressed on B cells from 2-, 6-, or 14-month-old Fasfl/fl or Fasfl/flCd19-Cre mice (Figure 2Biii). Intriguingly, B cells from 6- and 14-month-old Fasfl/flCd19-Cre mice expressed MHC class II more abundantly than did control B cells (Figure 2Biv; p < 0.005), implying that Fas-deficient B cells have increased APC capacity. Thus, the lymphocytes that accumulate in aged Fasfl/flCd19-Cre mice are activated.

To determine whether the lymphocyte accumulation in the mutants was due at least in part to hyperproliferation, we fed aged Fasfl/fl and Fasfl/flCd19-Cre mice BrdU-containing drinking water and assessed BrdU incorporation in vivo. BrdU incorporation was observed in 3–4-fold more B cells and CD4+ and CD8+ T cells from the mutant mice compared to controls (Figure 2C). Thus, the fatal lymphoproliferation in Fasfl/flCd19-Cre mice is due to the activation and hyperproliferation of both T and B cells.

Disrupted Lymphoid Architecture in Aged Fasfl/flCd19-Cre Mice

We next analyzed lymphoid-organ architecture in Fasfl/flCd19-Cre mice. At age 2 or 6 months, the mutants showed normal spleen and LN structures (data not shown). At age 14 months, hematoxylin-eosin (H&E) and immunohistochemical staining of control spleens confirmed that both T and B cells were normally localized in the periarteriolar lymphoid sheath (PALS) and the B cell corona, respectively (Figures 2Di, 2Diii, and 2Dv). In contrast, T and B cells were mixed together in the mutant white pulp (Figures 2Dii, 2Div, and 2Dvi). Similarly, T and B cells in control LNs were normally localized in the paracortical area and lymphoid follicles, respectively (Figures 2Ei, 2Eiii, and 2Ev), whereas T and B cells were totally mixed in the mutant LN (Figures 2Eii, 2Eiv, and 2Evi). Higher magnification of these stained spleen and LN sections showed that, compared to controls, lymphocytes in the mutants were large and had abundant cytoplasm (Figures 2Diii, 2Div, 2Eiii, and 2Eiv). This age-dependent disruption of lymphoid architecture is consistent with the abnormal lymphocyte activation and expansion observed in these mutants.

Cytokine Elevation and Abnormal Differentiation from Fasfl/flCd19-Cre Mice

The elevated expression of ICOS on T cells from Fasfl/flCd19-Cre mice indicated that these cells were activated. To determine whether cytokine production by these cells might be contributing to the death of the mutant mice, we examined serum cytokines in Fasfl/fl and Fasfl/flCd19-Cre mice. Amounts of IFN-γ, TNF-α, IL-10, and IL-6 were elevated in the mutant serum by an estimated 3-fold (2783 ± 439 ng/ml versus 1043 ± 565 ng/ml; p < 0.005), 7-fold (53.5 ± 42.5 ng/ml versus 7 ± 5.1 ng/ml; p < 0.05), 70-fold (1099 ± 720 ng/ml versus 16.5 ± 17 ng/ml; p < 0.005), and 20-fold (1099 ± 719.5 ng/ml versus 46.3 ± 35.8 ng/ml; p < 0.05), respectively (Figure 3A). No differences in serum amounts of IL-1α, IL-1β, IL-2, IL-4, IL-12, IL-13, IL-17, MIP-2, or TGF-β were detected (data not shown). Consistent with these findings, more IFN-γ-producing CD4+ T cells (1.1% ± 0.44% versus 0.19% ± 0.08%; p < 0.05) and IL-6-producing B220+ cells (4.2% ± 2.1% versus 0.32% ± 0.38%; p < 0.05) were found in the lymphoid organs of Fasfl/flCd19-Cre mice compared to those of controls (Figure 3B). IL-10-producing CD4+ T cells were also slightly (but not significantly) increased (2.85% ± 1.97% versus 0.59% ± 0.18%; p > 0.05). Numbers of TNF-α-producing T and B cells were not increased at all (data not shown), suggesting that other cell types are responsible for the elevated TNF-α production in Fasfl/flCd19-Cre mice.

Figure 3. Increased Cytokine Production in Fasfl/flCd19-Cre Mice.

Figure 3

Open in a new tab

(A) Elevated serum cytokines. Serum amounts of IFN-γ, TNF-α, IL-10, and IL-6 were determined in Fasfl/fl and Fasfl/flCd19-Cre mice of age ≥ 11 months. Results shown represent the mean cytokine level ± SD (n = 4–6 mice/genotype).

(B) Increased numbers of cytokine-producing cells. Cells from LNs of aged Fasfl/fl and Fasfl/flCd19-Cre mice were fixed with 1.6% PFA, subjected to intracellular staining with antibodies recognizing the indicated cytokine, and analyzed by flow cytometry. Numbers are the percentage of cells in the gated lymphocyte population that produced the indicated cytokine (n = 3–4).

To investigate whether the altered cytokine production in the mutant correlated with abnormal Th1 and Th2 cell differentiation, we tested whether Fasfl/flCd19-Cre CD4+ T cells could differentiate normally in vitro. Under Th1 cell-differentiation conditions, approximately 3-fold more IFN-γ- or TNF-α-producing cells arose from Fasfl/flCd19-Cre CD4+ T cells than from controls (Figure S4A, left). Under Th2 cell differentiation conditions, a similar percentage of control and Fasfl/flCd19-Cre CD4+ T cells produced IL-4, whereas an increased percentage of Fasfl/flCd19-Cre CD4+ T cells produced IL-10 compared to controls (51% versus 30%) (Figure S4A, middle). More IFN-γ-, TNF-α-, and IL-10-producing cells were also observed when T cells from Fasfl/flCd19-Cre mice were stimulated with PMA plus ionomycin (data not shown). However, there was no increase in the percentage of IL-17-producing cells generated when Fasfl/flCd19-Cre T cells were cultured under Th17 cell-differentiation conditions (Figure S4A, right).

Taken together, these results show that a deficiency of Fas in B cells has profound effects not only on serum-cytokine amounts but also on T cell differentiation.

Abnormal Signaling of T Cells from Fasfl/flCd19-Cre Mice

We next examined intracellular signaling in T cells from Fasfl/flCd19-Cre mice. STAT1 is phosphorylated in wild-type (WT) T and B cells in response to IFN-γ (Pestka et al., 2004). Upon in vitro IFN-γ treatment, CD4+ and CD8+ T cells of Fasfl/flCd19-Cre mice showed an age-dependent decrease in STAT1 (pY701) phosphorylation (Figure S4B). In contrast, IFN-γ-induced STAT1 phosphorylation was normal in B cells of Fasfl/flCd19-Cre mice. STAT3 is phosphorylated in WT T and B cells in response to IL-10 (Riley et al., 1999). However, CD4+ and CD8+ T cells from Fasfl/fl mice did not respond well to IL-10 as shown by their poor phosphorylation of STAT3 (pY705) (Figure S4C, top). CD4+ and CD8+ T cells from Fasfl/flCd19-Cre mice showed an age-dependent increase in STAT3 phosphorylation in response to IL-10 (Figure S4C, bottom). In contrast, STAT3 phosphorylation in Fasfl/flCd19-Cre B cells in response to IL-10 was comparable to that in controls at all three ages examined.

IL-10 is known to inhibit IFN-γ production by Th1 cells (Moore et al., 1990). To determine whether the elevated serum IL-10 in Fasfl/flCd19-Cre mice might be inhibiting the function of their Th1 cells, we cultured control and mutant in vitro-differentiated Th1 cells in mrIL-10 for 40 hr, restimulated them for 16 hr with anti-CD3, and identified IFN-γ-producing cells by intracellular staining. Flow-cytometric analyses showed that a culture in IL-10 reduced the percentage of mutant Th1 cells that were able to produce IFN-γ (56.1% ± 3.8% → 39.5% ± 3.8%; p < 0.005), whereas the ability of control Th1 cells to produce IFN-γ was not significantly affected (Figure S5A). Thus, elevated IL-10 inhibits the functionality of in vitro-differentiated Th1 cells from Fasfl/flCd19-Cre mice. Consistent with these results and those shown in Figure S4C, purified T cells, from Fasfl/flCd19-Cre mice, that were cultured in IL-10 showed more pSTAT3 in the nuclear fraction compared to controls (Figure S5B). These data suggest that IL-10 triggers STAT3-mediated signaling in the mutant Th1 cells that leads to a block in IFN-γ production. Thus, the elevated serum IL-10 apparent in Fasfl/flCd19-Cre mice may help to terminate the inflammatory response, consistent with a report that IL-10 cooperates with FasL to terminate antigen-specific immune responses (Barreiro et al., 2004).

Taken together, these results show that a deficiency of Fas in B cells has profound effects on T cell signaling and function.

Lymphoproliferation Associated with B Cell-Specific Fas Deficiency Initiates in the GC

The abnormalities displayed by the Fas-sufficient T cells in Fasfl/flCd19-Cre mice suggested that these T cells are affected by their interactions with the Fas-deficient B cells present in these mutants. Fas is highly expressed in GC B cells (Hennino et al., 2001), and the GC is a site where critical T cell-B cell interactions occur. To understand the physiological role of Fas in GC B cells and to pinpoint why Fasfl/flCd19-Cre mice fail to maintain lymphocyte homeostasis, we generated mutants in which Fas was ablated only in GC B cells. We crossed Fasfl/fl mice with Ighg1-Cre mice (Casola et al., 2006) and subsequently with lpr mice on the MRL background (Vidal et al., 1998) to yield Fasfl/lprIghg1-Cre mice on the F1 (C57BL/6 × MRL) background. Flow-cytometric analysis confirmed the efficient deletion of the Fas gene in GC B cells (PNAhiCD38lo) of our Fasfl/lprIghg1-Cre mice (Figure 4A, right). The Fas gene was not deleted in T cells, follicular B cells (FB), or marginal zone B cells (MZ) of the mutant (Figure 4A, bottom left).

Figure 4. Ablation of Fas in GC B Cells Reproduces the Lymphoproliferative Disorder in Fasfl/flCd19-Cre Mice.

Figure 4

Open in a new tab

(A) Confirmation of Fas deletion in GC B cells. The top portion shows flow-cytometric analysis of GC B cells (PNAhiCD38lo) among gated CD19+ B cells of Fasfl/lpr and Fasfl/lprIghg1-Cre mice. Numbers are the percentage of GC B cells among total B cells. The bottom-right portion shows the absence of Fas expression in GC B cells from mLN of Fasfl/lprIghg1-Cre mice. The bottom-left portion shows a Southern blot of T cells (T), follicular B cells (FB), and marginal zone B cells (MZ) from Fasfl/+Ighg1-Cre mice confirming the absence of Cre-mediated recombination in cells other than GC B cells.

(B) Increased weight of spleens from 3.5- and 8-month-old Fasfl/lpr and Fasfl/lprIghg1-Cre mice. Each symbol represents an individual mouse.

(C) Increased absolute numbers of total lymphocytes and B and T cells in spleens of 8-month-old Fasfl/lpr and Fasfl/lprIghg1-Cre mice. Results shown represent the mean ± SD (n = 3/genotype).

(D) Disrupted splenic architecture. We stained spleen sections from 8-month-old Fasfl/lpr (left) and Fasfl/lprIghg1-Cre (right) mice with H&E to reveal the splenic architecture. Scale bars apply to control and mutant samples in a pair.

(E) Accumulation of GC B cells. Representative flow-cytometric analyses of GC B cells (PNAhiCD38lo) within gated B cells in spleen (SP) and mLN of 3- to 3.5-month-old Fasfl/lpr and Fasfl/lpr Ighg1-Cre mice before (no imm) and 30 days after immunization with SRBC.

(F) Accumulation of GC B cells in Fasfl/flCd19-Cre mice. Representative flow cytometric analyses of GC B cells among B220+ B cells in spleen (SP) and mLN of unimmunized Fasfl/fl and Fasfl/flCd19-Cre mice. Numbers in (F) and (G) are the percentage of GC B cells (PNAhiCD38lo) among gated B cells.

Like Fasfl/flCd19-Cre mice, Fasfl/lprIghg1-Cre mice showed a defect in lymphocyte homeostasis as evidenced by their enlarged spleens and LNs (Figure 4B and data not shown). Compared to controls, the total lymphocyte number in the spleen of 8-month-old Fasfl/lprIghg1-Cre mice was increased by approximately 2-fold and both the B cell and T cell compartments had expanded (Figure 4C). The increase of these cell numbers was all statistically significant (p < 0.005). H&E staining of spleen sections from control Fasfl/lpr mice showed well-defined red pulp and white pulp (Figure 4D, top left), whereas the Fasfl/lprIghg1-Cre spleen exhibited disrupted architecture with no clear boundary between the red and white pulp (Figure 4D, top right). Higher magnification of these spleen sections revealed that cells from the Fasfl/lprIghg1-Cre spleen were larger and had more abundant cytoplasm (Figure 4D, bottom). The abnormal spleen architecture in Fasfl/lprIghg1-Cre mice was also virtually identical to that in Fasfl/flCd19-Cre mice (Figure 2D). Although the Fasfl/lprIghg1-Cre mice were on the F1 (C57BL/6 × MRL) background, their phenotype was almost identical to that of Fasfl/flCd19-Cre mice on a C57BL/6 background, with respect to development of splenomegaly and lymphoadenopathy characterized by both B cell and T cell expansion, distorted lymphoid-organ architecture, and increased GC B cell numbers. Thus, ablation of Fas in GC B cells alone reproduces the entire lymphoproliferative disorder observed in Fasfl/flCd19-Cre mice.

We hypothesized that it was the loss of Fas-mediated apoptosis that initiated the immune disorder in GC B cells. In support of this idea, although the percentage of GC B cells in the spleen of an unimmunized Fasfl/lprIghg1-Cre mouse was similar to that in controls (Figure 4E, top left), the percentage of GC B cells in the mesenteric LN (mLN) of the mutant was increased by approximately 2-fold (Figure 4E, top right). When we immunized mutant and control mice with sheep red blood cells (SRBCs; a Td antigen), the proportion of GC B cells in the spleen and mLN of control mice returned to baseline by 30 days postimmunization (Figure 4E, bottom and data not shown). In contrast, Fasfl/lprIghg1-Cre mice immunized with SRBCs exhibited a 3–5-fold increase in GC B cells over controls and maintained this increase until at least day 30 postimmunization. Interestingly, unimmunized aged Fasfl/flCd19-Cre mice showed a 2- to 4-fold increase over controls in the number of spontaneous GCs present in the lymphoid organs (Figure 4F). Taken together, these data suggest that the immune disorder resulting from B cell-specific Fas deficiency initiates in the GC.

The Lymphoproliferative Disorder in Fasfl/flCd19-Cre Mice Is T Cell Dependent

GC formation is known to be critically dependent on T cell-B cell interaction (Kalia et al., 2006). To test whether the immune disorder in Fasfl/flCd19-Cre mice was T cell-dependent, we removed the T cells from these mutants by crossing Fasfl/flCd19-Cre mice with Tcrb−/− mice in which T cell development is blocked in the thymus as a result of the absence of the TCRβ chain (Mombaerts et al., 1992). The resulting Fasfl/flCd19-CreTcrb−/− mice possess normal numbers of Fas-deficient B cells but show only a few CD4+ and CD8+ T cells in the peripheral lymphoid organs (Figure S6A). Strikingly, the upregulated MHC class II present on Fasfl/flCd19-Cre B cells had returned to normal on Fasfl/flCd19-CreTcrb−/− B cells (Figure 5A). Furthermore, the proportion of BrdU+ cells in the B cell compartment dropped from 28.6% in Fasfl/flCd19-Cre mice to 10.6% in Fasfl/flCd19-CreTcrb−/− mice, a level comparable to that in Fasfl/fl controls (Figure 5B). As a consequence, the size of the Fasfl/flCd19-CreTcrb−/− spleen was close to normal (data not shown). The total B cell cellularity in LNs of Fasfl/flCd19-CreTcrb−/− mice was also similar to that of Fasfl/fl controls (p = 0.47), whereas the LNs of Fasfl/flCd19-Cre mice had approximately 5-fold more B cells than controls (p < 0.005) (Figure 5C). In the spleen, H&E staining revealed that the disrupted architecture prominent in Fasfl/flCd19-Cre mice (Figure 5Dii) was restored to normal in Fasfl/flCd19-CreTcrb−/− mice (Figure 5Diii). Finally, the lymphocyte infiltration in Fasfl/flCd19-Cre liver (Figure 5Dv) was absent in Fasfl/flCd19-CreTcrb−/− liver (Figure 5Dvi). Thus, T cells are required for the B cell lymphoproliferation observed in Fasfl/flCd19-Cre mice. On the basis of these results and our finding that mice lacking Fas specifically in GC B cells reproduce the phenotype of Fasfl/flCd19-Cre mice, we theorize that interactions between the Fas-sufficient T cells and Fas-deficient B cells within the GCs of these mutants drive the immune disorder.

Figure 5. Hyperproliferation and Activation of Fas-Deficient B Cells Is T Cell Dependent.

Figure 5

Open in a new tab

Fasfl/flCd19-CreTcrb−/− mice lack T cells and have Fas-deficient B cells. B cells from Fasfl/fl, Fasfl/flCd19-Cre, and Fasfl/flCd19-CreTcrb−/− mice were examined for (A) MHC class II expression as in Figure 2B; (B) BrdU incorporation as in Figure 2C; and (C) B cell cellularity in the LN as in Figure 2A. The mean cellularity ± SD is shown (n = 3–9 mice/genotype).

(D) H&E stained sections of spleen and liver from mice of the indicated genotypes. Scale bars apply to an entire row.

Expansion of Memory B Cell Compartment in Fasfl/flCd19-Cre Mice

The defect in the negative selection of GC B cells in the absence of Fas (Figure 4) prompted us to investigate whether the plasma cells and memory B cells that differentiated from these GC B cells exhibited abnormalities as well. The number of CD138+ (syndecan-1) cells in the spleen and LNs of aged Fasfl/flCd19-Cre mice was greater by 2- to 3-fold than controls (data not shown). When we examined the B220+IgM−IgD− B cell population (which contains memory B cells and GC B cells), we found an approximately 3-fold increase in the frequency and a 10-fold increase in the absolute numbers of these cells in the spleens of Fasfl/flCd19-Cre mice compared to in controls (Figure 6A, middle; p < 0.005). Consistent with this observation, the proportion and total numbers of splenic B220+IgG1+ B cells that were IgM−IgD− were increased approximately 10-fold and 30-fold, respectively, over controls (Figure 6B, top, middle; p < 0.05). Approximately 98% of B220+IgG1+ B cells in the mutant (90% in controls) were CD38+ cells (data not shown), indicating that these cells were predominantly memory B cells rather than GC B cells (Takahashi et al., 2001). WT memory B cells typically are larger than naive B cells and show a body-wide circulation pattern (Liu et al., 1995) (Linton and Klinman, 1992). The B220+IgG1+ memory B cells of Fasfl/fl and Fasfl/flCd19-Cre mice were indeed larger than naive B cells and substantial numbers of them were found both in the circulation and mLN (data not shown).

Figure 6. The Lymphoproliferative Disorder in Fasfl/flCd19-Cre Mice Is Caused by Memory B Cells and Requires CD28-CD80 Signaling.

Figure 6

Open in a new tab

Fasfl/flCd19-CreCd28−/− mice lack CD28 in all cells and Fas in B cells only. Fasfl/fl, Fasfl/flCd19-Cre, and Fasfl/flCd19-CreCd28−/− mice were examined for (A) percentage of B220+IgM−IgD− B cells. Mean numbers ± SD of B220+IgM−IgD− B cells in spleens of Fasfl/fl, Fasfl/flCd19-Cre, and Fasfl/flCd19-CreCd28−/− mice (n = 3–4/genotype) were 7.6 ± 1.78 × 105, 37.8 ± 5.1 × 105 and 4.8 ± 0.96 × 105, respectively. Fasfl/fl, Fasfl/flCd19-Cre, and Fasfl/flCd19-CreCd28−/− mice were examined for (B) percentage of B220+IgG1+ memory B cells (top) and CD80 and CD86 expression (bottom). Mean numbers ± SD of B220+IgG1+ memory B cells in spleens of Fasfl/fl, Fasfl/flCd19-Cre and Fasfl/flCd19-CreCd28−/− mice (n = 3–4/genotype) were 2.9 ± 1.3 × 104, 85.2 ± 58.6 × 104 and 1.73 ± 1.17 × 104, respectively. (C) shows the expression of the indicated markers on T and B cells as in Figure 2B; (D) shows STAT1 phosphorylation in IFN-γ-stimulated T cells as in Figure S4B; (E) shows BrdU incorporation as in Figure 2C; and (F) shows LN cellularity (right) as in Figure 2A; the mean cellularity ± SD is shown (n = 3–7 mice/genotype). *(*)p < 0.05 for Fasfl/fl versus Fasfl/flCd19-Cre mice but p < 0.005 for Fasfl/flCd19-Cre versus Fasfl/flCd19-CreCd28−/− mice. Mice were ≥11 months old (A, B, C, E, and F) or 5 to 6 months old (D).

CD28-CD80 Interaction Is Required for the Lymphoproliferative Disorder in Fasfl/flCd19-Cre Mice

WT memory B cells are potent activators of T cells because of their enhanced upregulation of CD80 and CD86 (Liu et al., 1995). Memory B cells from Fasfl/flCd19-Cre mice showed upregulated CD80 but control amounts of CD86 (Figure 6B, bottom). To investigate whether CD28-CD80 signaling was important for the accumulation of memory B cells and the abnormal T and B cell phenotypes of Fasfl/flCd19-Cre mice, we crossed these mutants with Cd28−/− mice (Suh et al., 2004) to generate Fasfl/flCd19-CreCd28−/− mice. Analysis by flow cytometry confirmed that CD28 expression was absent on T cells of the compound mutants (Figure S6B). The expansion of the B220+IgM−IgD− B cell population observed in Fasfl/flCd19-Cre mice did not occur when CD28 was deleted (Figure 6A, right), and the percentage of B220+IgG1+ B cells (of which approximately 90% were CD38+) in Fasfl/flCd19-CreCd28−/− mice was reduced to 30% compared to that in Fasfl/fl controls (Figure 6B, top right). Thus, the accumulation of IgG1+ memory B cells in Fasfl/flCd19-Cre mice is CD28 dependent, and the reduction of IgG1+ memory B cells in Fasfl/flCd19-CreCd28−/− mice is likely to be due to the impairment of GC formation that occurs in the absence of CD28 (Ferguson et al., 1996). The lack of CD28 in these mutants also disrupted the excessive upregulation of ICOS and MHC class II observed in Fasfl/flCd19-Cre B cells (Figure 6C). Thus, the abnormal activation of both T and B cells in Fasfl/flCd19-Cre mice depends on CD28.

The hyporesponsiveness to IFN-γ of T cells from Fasfl/flCd19-Cre mice was restored to normal in the absence of CD28, as evidenced by the reappearance of control amounts of STAT1 phosphorylation in T cells of Fasfl/flCd19-CreCd28−/− mice (Figure 6D). Analysis of in vivo BrdU incorporation showed that CD28 ablation not only prevented hyperproliferation but also reduced the proliferation of CD4+ and CD8+ T cells and B220+ B cells to 30%–50% compared to in Fasfl/fl controls (Figure 6E). The enlarged spleen present in Fasfl/flCd19-Cre mice became normal in size in nine of ten Fasfl/flCd19-CreCd28−/− mice, and none of the compound mutants had enlarged LNs (data not shown). Indeed, LN cellularity in Fasfl/flCd19-CreCd28−/− mice was lower than in the Fasfl/fl controls (15.4 ± 8.6 × 106 versus 22.6 ± 8.64 × 106), and this reduction held true when B and T cell numbers were examined separately (Figure 6F). Lastly, lymphoid-organ architecture was restored to normal and lymphocyte infiltration in the liver was prevented in Fasfl/flCd19-CreCd28−/− mice (Figure S7).

These results reinforce the conclusion drawn from our analyses of Fasfl/lprCγ1Cre and Fasfl/flCd19-CreTcrb−/− mice: that the lymphoproliferative disorder in B cell-specific Fas KO mice initiates in the GC where negative selection of GC B cells is impaired in the absence of Fas-mediated apoptosis (Takahashi et al., 2001). Memory B cells that would otherwise have been deleted instead expand and interact with CD4+ Th cells, resulting in T cell expansion that is amplified by a positive-feedback loop operating through the CD28-CD80 pathway. These accumulating lymphocytes infiltrate the organs and induce a potentially fatal inflammatory cytokine storm.

DISCUSSION

Our Fasfl/flCd19-Cre and Fasfl/lprIghg1-Cre mice are unique in that Fas is deleted only in B lineage cells and all other cells are Fas sufficient. The Fas-deficient B cells in both of our mutant strains were activated and hyperproliferative. Unexpectedly, the Fas-sufficient T cells in Fasfl/flCd19-Cre mice were also activated and hyperproliferative. Our analyses suggest that Fas expression by B cells has an important role in the maintenance of T cell homeostasis via T cell-B cell interaction and that this interaction occurs in the GC. On the basis of the literature and our in vivo analyses of Fasfl/lprIghg1-Cre and Fasfl/flCd19-Cre mice, we propose the following model of how Fas-mediated negative selection of GC and memory B cells regulates T and B lymphocyte homeostasis. Antigen (Ag)-activated B cells interact with Ag-activated Th cells and then initiate GC formation. Some progeny GC B cells through somatic hypermutation may express BCRs with improved binding affinity for the cognate Ag (high affinity), whereas others will bear BCRs with suboptimal affinity for the Ag or a BCR that recognizes self-Ag (autoreactive). With the help of the activated Th cells, the high affinity GC B cells are positively selected to differentiate into either plasma cells or memory B cells. In the case of Fas-competent B cells, B cells with suboptimal affinity or autoreactivity are eliminated through Fas-dependent apoptosis and lymphocyte homeostasis is maintained. In the absence of functional Fas, GC B cells expressing BCRs with suboptimal affinity or autoreactivity do not undergo apoptosis and are able to differentiate into plasma cells or memory B cells. Memory B cells (including those derived from naive B cells with suboptimal affinity) are potent APCs, and the greatly increased numbers of these cells in turn activate large numbers of Th cells, including rare autoreactive Th cells. The lymphoproliferative disorder is thus initiated. Upon interaction with an activated Th cell, a memory B cell further upregulates CD80 and becomes an even more potent APC. Additional Th cells that receive CD28-mediated costimulation from these memory B cells undergo rapid expansion and return to the GC to select more suboptimal affinity and autoreactive GC B cells into the memory B cell pool. This positive-feedback loop drives the proliferation of both T and B cells such that a fatal loss of homeostasis occurs. We recognize that further investigation is needed to clarify the extent to which interference with GC formation contributes to the prevention of lymphoproliferation in Fasfl/flCd19-Cre mice upon removal of T cells or CD28. However, the recapitulation of the disease phenotype in Fasfl/lprIghg1-Cre mice argues in favor of the interpretation that GC and post-GC B cells play a dominant pathogenic role.

Most studies agree that Fas is highly expressed in GC B cells and has an essential role in GC B cell apoptosis both in vitro (Hennino et al., 2001) and in vivo (Takahashi et al., 2001). During the GC reaction, antigen-activated B cells undergo somatic Ig-gene hypermutation that may increase antibody affinity. However, some of these alterations may change antibody specificity and/or result in the acquisition of autoreactivity (Hande et al., 1998; Rajewsky, 1996). Several studies have demonstrated that Fas-mediated apoptosis is essential for eliminating autoreactive B cells (Fukuyama et al., 2002; Rathmell et al., 1995; William et al., 2002). When somatic hypermutation mistakenly introduces deleterious alterations into the Fas gene in GC B cells (Muschen et al., 2000), impaired Fas-mediated negative selection of GC B cells could potentially result in autoreactive B cell generation as proposed by our model.

The ablation of T cells or CD28 in Fasfl/flCd19-Cre mice totally abolished lymphocyte hyperproliferation and infiltration. In the absence of T cells or CD28, B cell activation was also decreased as judged by reductions in MHC class II expression and hyper-proliferation and by the restoration of baseline B cell numbers. Because CD4+ Th cells and their expression of CD28 are essential for GC formation (Kalia et al., 2006), and the lymphoproliferative disorder of Fasfl/flCd19-Cre mice initiates in the GC, the disease was prevented in animals in which GC formation was abolished by the genetic removal of T cells or CD28. These considerations may partially explain why the Fas-deficient B cells that resisted homeostatic control in the presence of T cells became sensitive to such control when T cells or CD28 were absent. However, it is also possible that when T cell-B cell interaction and GC formation are disrupted, a Fas-independent mechanism is able to eliminate excess Fas-deficient B cells and restore B cell homeostasis. It may be that, like the survival of WT B cells, the survival of Fas-deficient B cells is controlled by BAFF in the local environment (Mackay et al., 2003), but this remains to be investigated.

Lpr and Fasfl/flCd19-Cre mice both develop autoimmune lymphoproliferation but the peculiar DN T cells predominant in the former are not present in the latter. In addition, although CD28 ablation in Fasfl/flCd19-Cre mice completely prevents the immune disorder, CD28 ablation in lpr mice blocks only the development of autoimmune symptoms; lymphoproliferation is in fact accelerated in the spleen of these mutants (Tada et al., 1999). Similarly, ICOS ablation in lpr mice that should abolish GC formation fails to prevent splenomegaly and lymphadenopathy (Zeller et al., 2006). Thus, the CD28-B7 and ICOS-ICOSL pathways do not appear to play major roles in the expansion of DN T cells in lpr mice. In contrast, CD28-CD80 interaction either between activated B cells and CD4+ Th cells during initial GC formation or between memory B cells and CD4+ Th cells in an established GC is critical for the lymphoproliferative disorder observed in Fasfl/flCd19-Cre mice.

Fas-mediated apoptosis is involved in the negative selection of autoreactive B cells in the bone marrow; such a selection establishes central B cell tolerance (Lamoureux et al., 2007) (Seagal et al., 2003). In this study, we have demonstrated that the Fas-mediated apoptosis that is in GC B cells and that establishes peripheral tolerance also functions as an essential checkpoint governing T cell and memory B cell homeostasis. Defects such as Fas deficiency that result in the unintended positive selection of “inappropriate” memory B cells can lead to amplification of the Th cell population and highly destructive lymphoproliferation.

EXPERIMENTAL PROCEDURES

Mice

Fasfl/fl mice and Fasfl/flCd19-Cre mice were as described (Hao et al., 2004). To delete Fas specifically in GC B cells, we bred Fasfl/fl mice with Ighg1-Cre mice (Casola et al., 2006). We backcrossed the resulting Fasfl/flIghg1-Cre mice for eight generations to C57BL/6 mice and then bred them to lpr mice on the MRL background (Vidal et al., 1998) to yield Fasfl/lprIghg1-Cre mice on the F1 (C57BL/6 × MRL) background. T cells were deleted in Fasfl/flCd19-Cre mice by crossing to Tcrb−/− mice (Mombaerts et al., 1992). The resulting Fasfl/+Tcrb+/− F1 mice, with or without Cd19-Cre, were intercrossed for generation of Fasfl/flCd19-CreTcrb−/− compound mutant mice. Cd28 was deleted in Fasfl/flCd19-Cre mice by crossing to Cd28−/− mice (Suh et al., 2004). The resulting Fasfl/+Cd28+/− F1 mice, with or without Cd19-Cre, were intercrossed for generation of Fasfl/flCd19-CreCd28−/− compound mutant mice. Fasfl/flCd19-Cre, Fasfl/flCd19-CreTcrb−/− and Fasfl/+Cd19-CreCd28+/− strains were all backcrossed to C57BL/6 for more than eight generations. Because Cre expression can be toxic to cells (Schmidt-Supprian and Rajewsky, 2007), we included Fasfl/+Cd19-Cre, Fasfl/+Ighg1-Cre, Faslpr/+Cγ1Cre, Fasfl/+Cd19-CreTcrb−/−, and/or Fasfl/+Cd19-CreCd28−/− mice in all initial analyses of the corresponding mutants. These control mice were phenotypically indistinguishable from the control animals used for the analyses presented in this paper, as judged by cellularity and size of spleen and LNs. “Aged” mice were ≥6 months old unless otherwise specified. All animal experiments were approved by the University Health Network Animal Care Committee.

Flow Cytometry

Spleen and LN cells (1 × 106) were stained with antibodies recognizing the following: IgM (II/41), IgD(11-26c.2a), I-Ab (AF6-120.1), IgG1 (X56), NK cells (DX5), Pan-TER-119 (TCR-119), Thy1.2 (Cfo 1), CD3 (145-2C-11), CD4 (Gk1.5 or RM4.5), CD8 (53-6.7), CD11b (M1/70), CD11c (HL3), CD16/CD32 (2.4G2), CD19 (1D3), CD21 (7G6), CD23 (B3B4), CD25 (PC61), CD28 (37.51), CD38 (90), CD40 (3/23), CD43 (S7), CD45R/B220 (RA3-6B2), CD69 (H1.2F3), CD80 (16-10A1), CD86 (GL1), CD95 (Jo-2), CD152 (CTLA-4; UC10-4F10-11), CD154 (CD40 ligand; MR1), CD275 (ICOSL; HK5.3), and CD278 (ICOS; 7E.17G9). All antibodies were from BD Biosciences or eBioscience. Data were acquired by either BD FACSCalibur or BD FACSCanto flow cytometer and analyzed with either BD CellQuest or the FlowJo analysis program (Tree Star).

BrdU Incorporation

Mice were fed BrdU-containing drinking water (1mg/ml) for 3 days. Spleen and LN cells were immunostained as described above. BrdU was detected with a BrdU-Flow kit (BD Biosciences).

STAT Phosphorylation

Cultured LN cells (1 × 106) were stimulated for 15 min with either IFN-γ(100U/ml; Biosource) or mrIL-10 (100 ng/ml; Biosource), fixed with 1.6% paraformaldehyde, and incubated for 30 min at room temperature. After one wash in PBS, ice-cold methanol (100%) was added dropwise and cells were incubated on ice for 30 min. Washed cells were incubated with antibodies recognizing B220 (RA3-6B2), CD4 (RM4.5), CD8 (53-6.7), phospho-STAT1 (pY701), or phospho-STAT-3 (pY705) (all fromBDBiosciences) and then analyzed by flow cytometry.

Immunohistology

Mouse organs were fixed either in 10% formalin or 4% PFA at 4°C for 20–24 hr, embedded in paraffin, sectioned, and stained with H&E according to standard procedures. Immunostaining of T and B cells in lymphoid and nonlymphoid organs was as previously described (Hao et al., 2004) with slight modifications.

Statistical Analyses

The Student’s t test was employed for statistical analyses. Values are expressed as the mean ± SD. A p value of <0.05 was considered statistically significant. *p < 0.05; **p < 0.005.

Supplementary Material

1

ACKNOWLEDGMENTS

We thank various Mak lab members for technical assistance, T. Calzascia for help with Th17 cell differentiation, Toronto Centre for Phenogenomics for help with histology, and M. Saunders for scientific editing. This work was supported by grants to T.W.M. from the Terry Fox Cancer Foundation, Canadian Institutes of Health Research, and Leukemia and Lymphoma Society; to Z.H. from Canadian Institutes of Health Research; and to K.R. from the Deutsche Forschungsgemeinschaft through SFB 243 and the National Institutes of Health (USA).

Footnotes

SUPPLEMENTAL DATA

Supplemental Data include Supplemental Results, Supplemental Experimental Procedures, and seven figures and can be found with this article online at http://www.immunity.com/cgi/content/full/29/4/615/DC1/.

REFERENCES

  1. Barreiro R, Luker G, Herndon J, Ferguson TA. Termination of antigen-specific immunity by CD95 ligand (Fas ligand) and IL-10. J. Immunol. 2004;173:1519–1525. doi: 10.4049/jimmunol.173.3.1519. [DOI] [PubMed] [Google Scholar]
  2. Carreno BM, Collins M. The B7 family of ligands and its receptors: New pathways for costimulation and inhibition of immune responses. Annu. Rev. Immunol. 2002;20:29–53. doi: 10.1146/annurev.immunol.20.091101.091806. [DOI] [PubMed] [Google Scholar]
  3. Casola S, Cattoretti G, Uyttersprot N, Koralov SB, Segal J, Hao Z, Waisman A, Egert A, Ghitza D, Rajewsky K. Tracking germinal center B cells expressing germ-line immunoglobulin gamma1 transcripts by conditional gene targeting. Proc. Natl. Acad. Sci. USA. 2006;103:7396–7401. doi: 10.1073/pnas.0602353103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Cohen PL, Eisenberg RA. Lpr and gld: Single gene models of systemic autoimmunity and lymphoproliferative disease. Annu. Rev. Immunol. 1991;9:243–269. doi: 10.1146/annurev.iy.09.040191.001331. [DOI] [PubMed] [Google Scholar]
  5. Ferguson SE, Han S, Kelsoe G, Thompson CB. CD28 is required for germinal center formation. J. Immunol. 1996;156:4576–4581. [PubMed] [Google Scholar]
  6. Fisher GH, Rosenberg FJ, Straus SE, Dale JK, Middleton LA, Lin AY, Strober W, Lenardo MJ, Puck JM. Dominant interfering Fas gene mutations impair apoptosis in a human autoimmune lymphoproliferative syndrome. Cell. 1995;81:935–946. doi: 10.1016/0092-8674(95)90013-6. [DOI] [PubMed] [Google Scholar]
  7. Fukuyama H, Adachi M, Suematsu S, Miwa K, Suda T, Yoshida N, Nagata S. Requirement of Fas expression in B cells for tolerance induction. Eur. J. Immunol. 2002;32:223–230. doi: 10.1002/1521-4141(200201)32:1<223::AID-IMMU223>3.0.CO;2-4. [DOI] [PubMed] [Google Scholar]
  8. Hande S, Notidis E, Manser T. Bcl-2 obstructs negative selection of autoreactive, hypermutated antibody V regions during memory B cell development. Immunity. 1998;8:189–198. doi: 10.1016/s1074-7613(00)80471-9. [DOI] [PubMed] [Google Scholar]
  9. Hao Z, Hampel B, Yagita H, Rajewsky K. T cell-specific ablation of Fas leads to Fas ligand-mediated lymphocyte depletion and inflammatory pulmonary fibrosis. J. Exp. Med. 2004;199:1355–1365. doi: 10.1084/jem.20032196. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Hennino A, Berard M, Krammer PH, Defrance T. FLICE-inhibitory protein is a key regulator of germinal center B cell apoptosis. J. Exp. Med. 2001;193:447–458. doi: 10.1084/jem.193.4.447. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Hutloff A, Dittrich AM, Beier KC, Eljaschewitsch B, Kraft R, Anagnostopoulos I, Kroczek RA. ICOS is an inducible T-cell co-stimulator structurally and functionally related to CD28. Nature. 1999;397:263–266. doi: 10.1038/16717. [DOI] [PubMed] [Google Scholar]
  12. Kalia V, Sarkar S, Gourley TS, Rouse BT, Ahmed R. Differentiation of memory B and T cells. Curr. Opin. Immunol. 2006;18:255–264. doi: 10.1016/j.coi.2006.03.020. [DOI] [PubMed] [Google Scholar]
  13. Krammer PH. CD95(APO-1/Fas)-mediated apoptosis: Live and let die. Adv. Immunol. 1999;71:163–210. doi: 10.1016/s0065-2776(08)60402-2. [DOI] [PubMed] [Google Scholar]
  14. Krammer PH. CD95’s deadly mission in the immune system. Nature. 2000;407:789–795. doi: 10.1038/35037728. [DOI] [PubMed] [Google Scholar]
  15. Lamoureux JL, Watson LC, Cherrier M, Skog P, Nemazee D, Feeney AJ. Reduced receptor editing in lupus-prone MRL/lpr mice. J. Exp. Med. 2007;204:2853–2864. doi: 10.1084/jem.20071268. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Lenschow DJ, Walunas TL, Bluestone JA. CD28/B7 system of T cell costimulation. Annu. Rev. Immunol. 1996;14:233–258. doi: 10.1146/annurev.immunol.14.1.233. [DOI] [PubMed] [Google Scholar]
  17. Linsley PS, Ledbetter JA. The role of the CD28 receptor during T cell responses to antigen. Annu. Rev. Immunol. 1993;11:191–212. doi: 10.1146/annurev.iy.11.040193.001203. [DOI] [PubMed] [Google Scholar]
  18. Linton PJ, Klinman NR. The generation of memory B cells. Semin. Immunol. 1992;4:3–9. [PubMed] [Google Scholar]
  19. Liu YJ, Barthelemy C, de Bouteiller O, Arpin C, Durand I, Banchereau J. Memory B cells from human tonsils colonize mucosal epithelium and directly present antigen to T cells by rapid up-regulation of B7-1 and B7-2. Immunity. 1995;2:239–248. doi: 10.1016/1074-7613(95)90048-9. [DOI] [PubMed] [Google Scholar]
  20. Mackay F, Schneider P, Rennert P, Browning J. BAFF AND APRIL: A tutorial on B cell survival. Annu. Rev. Immunol. 2003;21:231–264. doi: 10.1146/annurev.immunol.21.120601.141152. [DOI] [PubMed] [Google Scholar]
  21. Mombaerts P, Clarke AR, Rudnicki MA, Iacomini J, Itohara S, Lafaille JJ, Wang L, Ichikawa Y, Jaenisch R, Hooper ML, et al. Mutations in T-cell antigen receptor genes alpha and beta block thymocyte development at different stages. Nature. 1992;360:225–231. doi: 10.1038/360225a0. [DOI] [PubMed] [Google Scholar]
  22. Moore KW, Vieira P, Fiorentino DF, Trounstine ML, Khan TA, Mosmann TR. Homology of cytokine synthesis inhibitory factor (IL-10) to the Epstein-Barr virus gene BCRFI. Science. 1990;248:1230–1234. doi: 10.1126/science.2161559. [DOI] [PubMed] [Google Scholar]
  23. Muschen M, Re D, Jungnickel B, Diehl V, Rajewsky K, Kuppers R. Somatic mutation of the CD95 gene in human B cells as a side-effect of the germinal center reaction. J. Exp. Med. 2000;192:1833–1840. doi: 10.1084/jem.192.12.1833. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Pestka S, Krause CD, Walter MR. Interferons, interferon-like cytokines, and their receptors. Immunol. Rev. 2004;202:8–32. doi: 10.1111/j.0105-2896.2004.00204.x. [DOI] [PubMed] [Google Scholar]
  25. Rajewsky K. Clonal selection and learning in the antibody system. Nature. 1996;381:751–758. doi: 10.1038/381751a0. [DOI] [PubMed] [Google Scholar]
  26. Rathmell JC, Cooke MP, Ho WY, Grein J, Townsend SE, Davis MM, Goodnow CC. CD95 (Fas)-dependent elimination of self-reactive B cells upon interaction with CD4+ T cells. Nature. 1995;376:181–184. doi: 10.1038/376181a0. [DOI] [PubMed] [Google Scholar]
  27. Rieux-Laucat F, Le Deist F, Hivroz C, Roberts IA, Debatin KM, Fischer A, de Villartay JP. Mutations in Fas associated with human lymphoproliferative syndrome and autoimmunity. Science. 1995;268:1347–1349. doi: 10.1126/science.7539157. [DOI] [PubMed] [Google Scholar]
  28. Riley JK, Takeda K, Akira S, Schreiber RD. Interleukin-10 receptor signaling through the JAK-STAT pathway. Requirement for two distinct receptor-derived signals for anti-inflammatory action. J. Biol. Chem. 1999;274:16513–16521. doi: 10.1074/jbc.274.23.16513. [DOI] [PubMed] [Google Scholar]
  29. Schmidt-Supprian M, Rajewsky K. Vagaries of conditional gene targeting. Nat. Immunol. 2007;8:665–668. doi: 10.1038/ni0707-665. [DOI] [PubMed] [Google Scholar]
  30. Seagal J, Edry E, Keren Z, Leider N, Benny O, Machluf M, Melamed D. A fail-safe mechanism for negative selection of isotype-switched B cell precursors is regulated by the Fas/FasL pathway. J. Exp. Med. 2003;198:1609–1619. doi: 10.1084/jem.20030357. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Shlomchik MJ, Madaio MP, Ni D, Trounstein M, Huszar D. The role of B cells in lpr/lpr-induced autoimmunity. J. Exp. Med. 1994;180:1295–1306. doi: 10.1084/jem.180.4.1295. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Smith KG, Nossal GJ, Tarlinton DM. FAS is highly expressed in the germinal center but is not required for regulation of the B-cell response to antigen. Proc. Natl. Acad. Sci. USA. 1995;92:11628–11632. doi: 10.1073/pnas.92.25.11628. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Suh WK, Tafuri A, Berg-Brown NN, Shahinian A, Plyte S, Duncan GS, Okada H, Wakeham A, Odermatt B, Ohashi PS, Mak TW. The inducible costimulator plays the major costimulatory role in humoral immune responses in the absence of CD28. J. Immunol. 2004;172:5917–5923. doi: 10.4049/jimmunol.172.10.5917. [DOI] [PubMed] [Google Scholar]
  34. Tada Y, Nagasawa K, Ho A, Morito F, Koarada S, Ushiyama O, Suzuki N, Ohta A, Mak TW. Role of the costimulatory molecule CD28 in the development of lupus in MRL/lpr mice. J. Immunol. 1999;163:3153–3159. [PubMed] [Google Scholar]
  35. Takahashi T, Tanaka M, Brannan CI, Jenkins NA, Copeland NG, Suda T, Nagata S. Generalized lymphoproliferative disease in mice, caused by a point mutation in the Fas ligand. Cell. 1994;76:969–976. doi: 10.1016/0092-8674(94)90375-1. [DOI] [PubMed] [Google Scholar]
  36. Takahashi Y, Ohta H, Takemori T. Fas is required for clonal selection in germinal centers and the subsequent establishment of the memory B cell repertoire. Immunity. 2001;14:181–192. doi: 10.1016/s1074-7613(01)00100-5. [DOI] [PubMed] [Google Scholar]
  37. Vidal S, Kono DH, Theofilopoulos AN. Loci predisposing to autoimmunity in MRL-Fas lpr and C57BL/6-Faslpr mice. J. Clin. Invest. 1998;101:696–702. doi: 10.1172/JCI1817. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. 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;356:314–317. doi: 10.1038/356314a0. [DOI] [PubMed] [Google Scholar]
  39. William J, Euler C, Christensen S, Shlomchik MJ. Evolution of autoantibody responses via somatic hypermutation outside of germinal centers. Science. 2002;297:2066–2070. doi: 10.1126/science.1073924. [DOI] [PubMed] [Google Scholar]
  40. Yoshinaga SK, Whoriskey JS, Khare SD, Sarmiento U, Guo J, Horan T, Shih G, Zhang M, Coccia MA, Kohno T, et al. T-cell co-stimulation through B7RP-1 and ICOS. Nature. 1999;402:827–832. doi: 10.1038/45582. [DOI] [PubMed] [Google Scholar]
  41. Zeller GC, Hirahashi J, Schwarting A, Sharpe AH, Kelley VR. Inducible co-stimulator null MRL-Faslpr mice: Uncoupling of autoantibodies and T cell responses in lupus. J. Am. Soc. Nephrol. 2006;17:122–130. doi: 10.1681/ASN.2005080802. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS196485-supplement-1.pdf (932.4KB, pdf)

RESOURCES