Regulation of Follicular Dendritic Cell Networks by Activated T Cells: The Role of CD137 Signaling1
http://www.100md.com
免疫学杂志 2005年第14期
Abstract
B cells, but not T cells, are considered to be important for the formation of follicular dendritic cell (FDC) clusters. Stimulation with agonist mAbs against CD137 (4-1BB), a TNFR family member primarily expressed on activated T cells, was effective in promoting T cell responses, but paradoxically suppressed T-dependent humoral immunity and autoantibody production in autoimmune disease models. Our present study shows that agonistic anti-CD137 treatment activates T cells, resulting in diminished FDC networks in B cell follicles, which are important components in T-dependent humoral immune responses both before and after the initiation of an immune response. Pretreatment with anti-CD137 before the secondary immunization inhibited memory Ab responses. Interestingly, CD137 costimulation-induced diminishment of FDC is T cell dependent. In addition, both CD4+ and CD8+ T cells are recruited into FDC area and are able to regulate FDCs by CD137 costimulation through a direct or indirect mechanism. These studies have revealed a previously unappreciated role of T cells in the regulation of FDC networks.
Introduction
Germinal centers (GC)3 form around specialized accessory follicular dendritic cells (FDCs) in lymphoid follicles, which provide a critical microenvironment for T cell-dependent humoral immune responses. Within GC, Ag-specific B cells efficiently undergo clonal expansion, isotype switching, somatic mutation, affinity maturation, and the generation of plasma and memory cells in response to T-dependent Ags. The formation and maintenance of GC depend on the dynamic interactions among B cells, T cells, and FDCs (1, 2, 3). FDCs play important roles in GC B cell proliferation, survival, and differentiation in both Ag-dependent and -independent manners (4, 5), whereas activated T cells induce the differentiation of GC B cells by providing CD40L and cytokines (6). When Abs are produced during a primary immune response, Ab-Ag complexes are formed and trapped for long periods of time on FDCs due to abundant Ig Fc and complement receptors (CR), among which CR1 and CR2 are the key immune complex-trapping molecules within primary follicles (2, 7). FDCs then present intact Ag for recognition by B cells, which is believed to be crucial for the development of high affinity, isotype-switched, and memory B cell responses (4, 5, 8, 9). In addition to delivering Ag, FDCs seem to supply numerous nonspecific stimuli during GC responses for the generation of an optimal B cell response (10). FDCs express adhesion molecules, including ICAM-1, VCAM-1, and mucosal addressin-cell adhesion molecule-1 (MAdCAM-1), which provide GC B cells with potent accessory signals to prevent apoptosis and aid in B cell selection during receptor maturation processes (11, 12, 13, 14, 15).
Multiple studies suggest that the development of FDCs in lymphoid organs critically depends on the presence of B cells and the cytokines produced by them, but does not depend on T cells (16, 17, 18, 19, 20). Furthermore, TCR-deficient mice show FDC structures similar to those of wild-type (wt) mice (18). Taken together, it appears that B cells are the primary cell type required for the development of FDC, and T cells are not required for the formation of FDC.
CD137 (4-1BB) is a member of the TNFR superfamily (TNFRSF9) and a well-defined T cell costimulatory receptor induced by TCR activation (21). Agonistic mAbs against CD137 have been shown to be effective in promoting T cell-mediated immune responses in vivo (22, 23, 24), but they paradoxically also inhibit T-dependent humoral immunity (25) and prevent autoantibody production in various autoimmune disease models (26, 27, 28). Previous studies attributed CD137-mediated suppression of T-dependent Ab responses to the inhibition of Th cells due to the induction of their anergy, activation-induced cell death, or induction of CD11c+CD8+ suppressor T cells and to the depletion of B cells (25, 26, 27, 28). Our current study unexpectedly found that CD137 costimulation dramatically down-regulates FDC networks in a T cell-dependent manner. Such treatment is able to diminish FDC even after GC formation. These studies clearly demonstrate that activated T cells have a role in regulating FDC.
Materials and Methods
Mice
C57BL/6 wt (B6/wt) mice were purchased from the National Cancer Institute. B6.129S2-Cd4tm1Mak (B6/CD4–/–), B6.129S2-Cd8atm1Mak (B6/CD8–/–), and B6.129P2-Tcrbtm1Mom/J (B6/TCR–/–) mice were purchased from The Jackson Laboratory. Animals were housed in a specific pathogen-free facility maintained by the University of Chicago Animal Resources Center. All animal protocols were approved by the institutional animal care and use committee at University of Chicago.
In vivo immunization and treatment with Abs
The mice were immunized i.p. with 1 x 108 SRBC (Colorado Serum) and/or 100 μg of keyhole limpet hemocyanin (KLH) Ag (Sigma-Aldrich) and treated with PBS, rat IgG, or agonistic anti-CD137 mAb (2A). For secondary immune responses, mice were rechallenged 6 wk later with the same doses of Ag and treated with 2A 8 days before or on the days of rechallenge. 2A was generated as previously described (29), and ascites was produced in SCID mice and purified by passage over a protein G-coupled Sepharose column. Rat IgG was purchased from Sigma-Aldrich and served as a control Ab. Spleens and sera were collected 2 wk after the first immunization and 1 wk after the second immunization.
Measurement of Ag-specific IgG
IgG anti-SRBC Abs were measured as previously described (30). In brief, 96-well Falcon plates (BD Biosciences) were coated with 150 μl of 0.1% SRBC diluted in PBS. Plates were incubated at room temperature for 50–60 min, followed by 0.25% glutaraldehyde (Sigma-Aldrich) in PBS for 5 min. For anti-KLH Abs, 96-well Immulon-4 plates (Dynatech Laboratories) were coated with KLH (10 μg/ml) at 37°C for 1 h. Unbound Ags were washed away with PBS, and plates were then blocked with 0.1% BSA in PBS blocking buffer at 37°C for 1–2 h. Diluted mouse sera were added and incubated at 37°C for 1 h. Pooled sera from C57BL/6 wt mice that were immunized with SRBC and KLH for 2 wk were used as the standard and arbitrarily defined as 100 U. Bound Abs were detected using 100 μl of 1/2000 diluted alkaline phosphatase-conjugated goat anti-mouse IgG-Ab (Southern Biotechnology Associates), followed by addition of the alkaline phosphatase substrate p-nitrophenyl phosphate (Sigma-Aldrich) at 1 mg/ml.
Evaluation of spleen follicle structure by immunohistochemical staining
Spleens were harvested, treated, and stained as previously described (30, 31). In brief, spleens were harvested, embedded in OCT compound (Miles), and frozen at –80°C. Frozen sections (6–8 μm thick) were fixed in cold acetone. Endogenous peroxidase was quenched with 0.2% H2O2 in methanol. After washing, the sections were stained by first incubating them with FITC-conjugated B220 and biotinylated anti-CR (CD35; 8C12; BD Pharmingen) or biotinylated peanut agglutinin (PNA; Vector Laboratories) all at 1/75 to 1/100 dilutions. For MAdCAM-1 and B220 double staining, the sections were first stained by purified anti-MAdCAM-1 (BD Pharmingen), followed by biotinylated goat anti-rat IgG(H+L) (Southern Biotechnology Associates). The sections were then stained with FITC-conjugated B220. HRP-conjugated rabbit anti-FITC (diluted 1/30; DakoCytomation) and alkaline phosphatase-conjugated streptavidin (diluted 1/20; Zymed Laboratories) were added 1 h later. Color development for bound alkaline phosphatase and HRP was performed with an alkaline phosphatase reaction kit (Vector Laboratories) and a Sigma Fast 3,3'-diaminobenzidine tablet set (Sigma-Aldrich). For FDC and T cell subset marker double staining, the tissue sections were first incubated with anti-mouse CD4 or anti-mouse CD8 (BD Pharmingen), followed by biotinylated secondary Ab (Vector Laboratories) and then avidin-biotin-peroxidase complex-alkaline phosphatase (AK-5000; Vector Laboratories). Color was developed with Alkaline Phosphatase Substrate kit III (Vector Laboratories), then the slides were stained with biotinylated CD35, and 3% H2O2 was used to block the endogenous peroxidase activity. The tissues were incubated with avidin-biotin-peroxidase complex-HRP (PK-6100; Vector Laboratories), and color was developed with 3,3'-diaminobenzidine (DakoCytomation). The blocking steps were added to avoid the cross-reaction between the two immunoreactions.
Adoptive transfer
B6/wt mice were immunized with 1 x 108 SRBC plus 100 μg of KLH and treated with either 150 μg of control rat IgG or 2A. Five days later, T cells were purified from these mice using the Pan T cell Isolation kit (Miltenyi Biotec) and were adoptively transferred into B6/TCR–/– mice. One day later, reconstituted B6/TCR–/– recipients were immunized with SRBC plus KLH. Another 10 days later, spleens were harvested, embedded in OCT for tissue staining, and sera were collected for Ab detection by ELISA.
Statistical analysis
All statistics were determined using unpaired Student’s two-tailed t test.
Results
Anti-CD137 treatment inhibits T-dependent IgG response and prevents GC formation
Agonistic mAbs against CD137 were effective in enhancing T cell-mediated responses, butparadoxically inhibited T-dependent humoral immunity. The mechanisms involved in this inhibition have not been well elucidated. SRBC and KLH are commonly used T cell-dependent Ags. Thus, we chose these Ags to examine why CD137 costimulation results in diminished T cell-dependent humoral immune responses. B6/wt mice were immunized with 1 x 108 SRBC plus 100 μg of KLH i.p. and treated with either 100 μg of control Ig or agonistic mAb against CD137 (2A) on the day of immunization. Two weeks later, the amounts of IgG anti-SRBC and anti-KLH detected in the sera of control Ig-treated mice were much higher than those in 2A-treated mice as previously reported (25) (Fig. 1A). To explore the mechanisms of 2A-mediated suppression of the humoral response, we examined GC formation, a key event in maturation of the humoral response. As expected, both PBS-treated and control rat IgG-treated mice showed fully developed GCs in the lymphoid follicles of the spleen (Fig. 1, B and C). On the contrary, 2A treatment completely blocked GC formation in spleen follicles (Fig. 1D).
CD137 costimulation down-regulates FDC
GC formation (cluster of activated B and T cells) inside B cell follicles in vivo requires support from FDCs. Therefore, we examined the effects of CD137 cross-linking on the formation of FDC networks. Two weeks after immunization and Ab treatment, the control mice showed massive GC and FDC network in spleen (Fig. 2, A and C). Surprisingly, we found that 2A treatment on the day of immunization with SRBC and KLH dramatically down-regulated FDC networks and eliminated GC in the spleen, as demonstrated by the disappearance of FDC-specific markers such as CR1 (CD35; Fig. 2, B and D) and FDC-specific mAb (data not shown). The diminished FDC networks were also confirmed by the disappearance of MAdCAM-1 staining within the follicles (Fig. 2F) compared with control spleen (Fig. 2E). Injection of control rat IgG in SRBC- and KLH-immunized mice had no effect on the down-regulation of FDC compared with that in mice immunized with SRBC and KLH alone (data not shown). The reduction of FDC was observed within 7–10 days after 2A treatment, and FDCs recovered 4 wk after treatment. Therefore, it is possible that anti-CD137 treatment inhibits the T cell-mediated humoral immune response by preventing the formation of FDC networks and GC.
CD137 costimulation interferes with FDC and GC after GC formation
To determine whether CD137 engagement could regulate FDC and GC structures after the establishment of a humoral immune response, we immunized B6 mice with 1 x 108 SRBC i.p.; 10 days later, when GCs had formed around FDC in the spleen (Fig. 3, A and B), the mice were treated with either 100 μg of 2A or control rat IgG. Unexpectedly, 10 days after treatment, although the control mice maintained abundant FDCs and GCs (Fig. 3, C and D), 2A-treated mice displayed dramatically reduced FDC networks and were completely lacking in GC (Fig. 3, E and F). Therefore, CD137 signaling can eliminate existing GC after the initiation of a humoral immune response.
The requirement for GC in the memory phase of the humoral immune responses is not clear. Due to the fact that CD137 stimulation can abolish FDC and GC even after initiation of the immune response and that it takes 7–10 days for the dramatic down-regulation of FDC by 2A treatment, we tested whether CD137 ligation inhibits the memory immune response against T-dependent Ags by treating mice before or on the day of the second immunization. B6 mice were immunized with 1 x 108 SRBC with 100 μg of KLH i.p.; 6 wk later, mice were rechallenged with the same Ags. The mice were divided into three groups: 1) mice treated with 200 μg of 2A 8 days before the second immunization, 2) mice treated with 200 μg of 2A on the day of the second immunization, and 3) control mice treated with control rat IgG 8 days before the second immunization. Additionally, mice that were immunized on the day of the second immunization without a first immunization were used as a primary immunization control. As shown in Fig. 4, when 2A treatment began on the day of the second challenge, the IgG recall response against SRBC (Fig. 4A) and KLH (Fig. 4B) was not changed compared with the memory response in the IgG control group. This is in accordance with the previous report that anti-CD137 mAb does not inhibit the secondary IgG anti-pneumococcal surface protein A response in R36A-primed mice (32). However, when 2A administration began 8 days before rechallenge, the memory IgG response against both SRBC and KLH was dramatically reduced compared with that in rat IgG-treated mice. We found that when 2A was administered 8 days before rechallenge, analysis of the spleen follicles on the day of rechallenge showed dramatically reduced FDCs and completely absent GC. When 2A treatment began on the day of the second immunization, reduced FDC and elimination of GC were delayed to 1 wk later (data not shown) compared with the typical recall IgG response time of 2–3 days after rechallenge. These studies imply the potential roles for FDCs and GC in the initiation of the memory phase of humoral responses.
Down-regulation of FDC by CD137 engagement is T cell dependent and can be mediated by both CD4+ and CD8+ T cells
CD137 is primarily expressed on activated T cells, but not on B cells. To test the cellular target of 2A for down-regulating FDC networks during a humoral response, we analyzed the effects of CD137 costimulation on FDC networks in T cell-deficient mice. To our surprise, 2A administration diminishes the formation of FDC networks in wt (Fig. 5, A, B, and I), but not TCR-deficient (Fig. 5, C, D, and I) mice, suggesting that 2A-mediated suppression of FDC clustering is T cell dependent. Therefore, we examined which T cell population is involved. Interestingly, 2A treatment dramatically reduced FDCs in both CD8-deficient and CD4-deficient mice (Fig. 5, E–I), suggesting that both CD4+ and CD8+ T cells are able to down-regulate FDC networks after CD137 cross-linking.
To examine whether CD137 signaling on activated CD4+ and CD8+ T cells can directly interact with FDC, we tested the presence of T cell subsets within the FDC area 5 days after Ag immunization and Ab treatment, a time point before the significant diminishment of FDC. In control mice, we observed that there are few CD4+ cells (Fig. 6A), but no CD8+ T cells (Fig. 6B), present in the FDC area. However, in anti-CD137-treated mice, a significant number of CD8+ T cells were found within the FDC area (Fig. 6D), and CD4+ T cells were also dramatically increased (Fig. 6C). These data suggest that CD137 costimulation promotes the recruitment of activated CD4+ and CD8+ T cells into the FDC network, and they also induce the down-regulation of FDC.
Adoptive transfer of T cells from anti-CD137-treated mice into TCR–/– mice does not diminish FDC in recipients
As well as its expression on T cells, CD137 is detected on FDC (33, 34). To test whether CD137-activated T cells are sufficient to down-regulate FDC in the absence of CD137 engagement on other cell types, we examined whether adoptive transfer of T cells from Ag-immunized and 2A-treated mice could interfere with the FDC network and humoral responses in TCR–/– recipients. B6/wt mice were immunized with SRBC and KLH and treated with either control rat IgG or 2A. Five days later, T cells were purified from these mice and adoptively transferred into TCR–/– mice. The reconstituted recipients were then challenged with SRBC plus KLH the next day. Ten days later, spleens and serum were collected and analyzed. We observed that both groups showed similar FDC (Fig. 7, A and C) and GC (Fig. 7, B and D), and there was no increased CD4 or CD8+ T cell staining within the FDC area in 2A-treated T cell recipients (Fig. 7, G and H) compared with control recipients (Fig. 7, E and F). Accordingly, both groups of mice produced similar level of IgG anti-SRBC and anti-KLH (Fig. 7I). These data imply that CD137 costimulation-activated T cells without continuing CD137 engagement are not capable of down-regulating the FDC network, or that direct CD137 signaling on T cells and other cell types, such as FDC, is essential.
CD137 engagement down-regulates FDC in naive mice
Previous studies have focused on the role of CD137 engagement in an Ag-dependent system; however, baseline T cell activation is sufficient for the CD137 agonist to inhibit FDC, because anti-CD137 treatment without active immunization could down-regulate FDC networks in wt, CD4–/–, and CD8–/– mice (Table I). These studies imply that CD137 costimulation is able to activate T cells in vivo with or without active immunization.
Discussion
Administration of agonistic anti-CD137 Ab can potently inhibit T-dependent humoral immune responses and limit autoantibody production, but the mechanisms are rather unclear. Our current study shows that 2A treatment can diminish the formation of FDC clusters, a critical component in GC formation and T-dependent humoral immune responses. Therefore, it is possible that such treatment could interfere with the T cell-mediated humoral immune response by preventing the formation of FDC and GC. Furthermore, our study shows that CD137 costimulation is not only able to down-regulate FDCs and prevent GC formation, it also can diminish FDCs and eliminate existing GC after the initiation of a humoral immune response. This is consistent with the idea that anti-CD137 treatment interferes with autoantibody production even after autoimmune disease establishment (26, 27). Similarly, 2A treatment can abolish the formation of FDC networks in Fas-deficient mice in the absence of exogenous Ags, followed by a diminished production of autoantibodies (data not shown). Therefore, such treatment may be useful for reducing pre-existing humoral autoimmune responses and blocking the continuation of autoimmune responses.
Our study has revealed an unappreciated role of T cells in the regulation of FDC formation. Although CD137 is detected on FDC (33, 34), the down-regulation of FDC by CD137 engagement seems to be indirect, because anti-CD137 treatment is not able to down-regulate FDC in the absence of TCR+ cells. Interestingly, GC formation is believed to be dependent on the interactions among B cells, CD4+ T cells, and FDCs. CD8+ T cells are generally not involved in the GC response. We show CD137 signaling could recruit both CD4+ and CD8+ T cells into FDC network and induces its diminishment by unknown mechanisms. One possible explanation for these observations is that CD137 engagement down-regulates FDC networks through the activation of CD4+ or CD8+ T cell. In support of this, Wu et al. (32) reported that anti-4-1BB-mediated suppression of the anti-pneumococcal surface protein A response is associated with splenic CD4+ and CD8+ T cell expansion and activation.
Although the diminishment of FDC by CD137 engagement is T cell dependent, whether CD137 signaling on other cell types is also required remains to be determined. When purified T cells from mice immunized with Ag and treated with anti-CD137 are transferred into TCR–/– mice, they failed to interfere with FDC and GC formation, and Ag-specific Ab production was not impaired. One possible explanation is that the feature of anti-CD137-activated T cells requires CD137 signaling for their maintenance. In TCR knockout recipients, without continuing CD137 signaling, control and 2A-treated T cells underwent similar homeostatic and Ag-specific proliferation. The difference between control and anti-CD137-treated T cells was diluted. The evidence that no increased CD4 and CD8+ T cell recruitment into the FDC area in spleens from recipients of anti-CD137-treated T cells compared with control treatment supports this explanation. It is also possible that except for its engagement on T cells, direct CD137 signaling on FDC is required as well. Our preliminary data showed when B6/wt mice were treated with multiple low doses of anti-CD3 (50 μg) or Con A (7.5 mg/kg), which induces massive T cell activation, the number of FDC clusters in spleen was diminished by both treatments (data not shown). CD137 has been shown to be a potent T cell costimulatory molecule, and agonistic anti-CD137 treatment dramatically promotes T cell activation, especially CD8+ T cells in vivo. Together, these studies imply that activated T cells could play a role in regulating FDC networks.
It seems that baseline T cell activation is sufficient for the CD137 agonist to diminish FDC, because anti-CD137 treatment alone, without immunization with exogenous Ag, could down-regulate the FDC network (Table I). Concordantly, Foell et al. (35) reported an increase in spleen weight and total cell number after anti-CD137 mAb treatment. We also observed an enlarged spleen and especially enlarged lymph nodes with increased percentages of activated CD69+ T cells in naive mice treated with agonistic anti-CD137 (data not shown). These studies imply that CD137 costimulation is able to activate T cells in vivo without additional immunization. Therefore, CD137 costimulation could down-regulate FDCs through T cells in an Ag-independent manner.
Our study does not exclude other mechanisms by which agonistic anti-CD137 Ab regulates humoral immune response. Mittler et al. (25) first reported that CD137 costimulation inhibited T cell-dependent humoral immune responses via the induction of Th cell anergy. They found that adoptive transfer of T cells from SRBC-immunized and anti-CD137-treated mice along with B cells from untreated mice into C.B-17 SCID mice failed to generate an anti-SRBC IgG response, whereas the opposite combination produced normal primary and secondary humoral responses to SRBC, isolating the defect to T cells. However, the lack of a CD4+ T cell response can also be explained by the diminished CD4+ T cells after initial activation, because our earlier study showed that such treatment can induce an initial activation, followed by activation-induced cell death (36). Using TCR transgenic DO11.10 mice, we found that administration of agonistic anti-CD137 initially promoted Ag-specific CD4+ T cell proliferation, but subsequently accelerated their depletion by inducing apoptosis. In accordance, in an experimental autoimmune encephalomyelitis model, such treatment initially up-regulated autoreactive Th1 cell functions before down-regulating the overall response (36). These studies suggest the diametric effect of CD137 costimulation on CD4+ T cells. Our previous study showed that agonistic anti-CD137 treatment greatly reduced autoantibody production in Fas-deficient mice accompanied by a diminishment of the number of CD4+ T cells and autoantibody-producing B cells (26). It is possible that the lack of FDC may reduce the survival of autoantibody-producing B cells, because FDC could provide a survival signal for activated B cells. Conversely, in lupus-prone NZB x NZW F1 female mice, Mittler et al. (27) found that agonistic anti-CD137 treatment rapidly suppressed autoantibody production regardless of the age or disease status of the animal without dramatic depletion of either B cells or CD4+ T cells. Reduced IL-2 and IL-4 production by CD4+ T cells was detected in treated mice, and the adoptive transfer of Ag-primed CD4+ T cells or DCs could override anti-CD137-mediated protection. The authors hypothesized that CD137-mediated signaling anergized CD4+ T cells during priming at the DC interface. Interestingly, Seo et al. (28) recently reported that agonistic anti-CD137 mAb treatment inhibited rheumatoid arthritis through induction of CD11c-expressing CD8+ T regulatory cells, which produce IFN- to stimulate dendritic cells and monocytes to produce the enzyme indolamine-2,3-dioxygenase, which further inhibits CD4+ T cell function and autoantibody production.
In summary, these data show that CD137 engagement reduces FDC networks in a T cell-dependent manner, and either CD4+ or CD8+ T cells are able to regulate FDC through CD137 costimulation by direct or indirect mechanisms. This study suggests that T cells could play an active role in regulating FDC networks within a humoral response. It is likely that multiple mechanisms are involved in the suppression of T cell-dependent humoral immune responses by CD137 costimulation. First, it inhibits Th cell function through numerous mechanisms (25, 26, 27, 28). Secondly, it promotes B cell reduction (26). Thirdly, CD137 costimulation induces a T cell-dependent down-regulation of FDC networks, which, as Ag-trapping accessory cells within GCs, are essential for the development of humoral immune responses and memory. Taken together, our study reveals the unappreciated role of T cells in modulating FDC and supports the idea that agonist anti-CD137 can be a powerful reagent in treating autoantibody-mediated autoimmune diseases.
Acknowledgments
We thank Lieping Chen for the generous gift of agonistic anti-CD137 mAb.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported in part by National Institutes of Health Research Grants R01AI062026, R01DK58897, and P01CA09296-01 (to Y.X.F.) and National Institutes of Health Training Grant 5T32HL07237 (to Y.S.).
2 Address correspondence and reprint requests to Dr. Yonglian Sun or Dr, Yang-Xin Fu, Department of Pathology, University of Chicago, MC3083, 5841 South Maryland Avenue, J541, Chicago, IL 60637. E-mail address: ysun1@bsd.uchicago.edu or yfu@uchicago.edu
3 Abbreviations used in this paper: GC, germinal center; CR, complement receptor; FDC, follicular dendritic cell; KLH, keyhole limpet hemocyanin; MAdCAM, mucosal addressin-cell adhesion molecule; PNA, peanut agglutinin; wt, wild type.
Received for publication October 26, 2004. Accepted for publication May 3, 2005.
References
Fu, Y. X., D. D. Chaplin. 1999. Development and maturation of secondary lymphoid tissues. Annu. Rev. Immunol. 17: 399-433.
Kosco-Vilbois, M. H., J. Y. Bonnefoy, Y. Chvatchko. 1997. The physiology of murine germinal center reactions. Immunol. Rev. 156: 127-136.
MacLennan, I. C.. 1994. Germinal centers. Annu. Rev. Immunol. 12: 117-139.
Burton, G. F., D. H. Conrad, A. K. Szakal, J. G. Tew. 1993. Follicular dendritic cells and B cell costimulation. J. Immunol. 150: 31-38.
Tew, J. G., J. Wu, D. Qin, S. Helm, G. F. Burton, A. K. Szakal. 1997. Follicular dendritic cells and presentation of antigen and costimulatory signals to B cells. Immunol. Rev. 156: 39-52.
Li, L., Y. S. Choi. 2002. Follicular dendritic cell-signaling molecules required for proliferation and differentiation of GC-B cells. Semin. Immunol. 14: 259-266.(Yonglian Sun2, Sarah E. B)
B cells, but not T cells, are considered to be important for the formation of follicular dendritic cell (FDC) clusters. Stimulation with agonist mAbs against CD137 (4-1BB), a TNFR family member primarily expressed on activated T cells, was effective in promoting T cell responses, but paradoxically suppressed T-dependent humoral immunity and autoantibody production in autoimmune disease models. Our present study shows that agonistic anti-CD137 treatment activates T cells, resulting in diminished FDC networks in B cell follicles, which are important components in T-dependent humoral immune responses both before and after the initiation of an immune response. Pretreatment with anti-CD137 before the secondary immunization inhibited memory Ab responses. Interestingly, CD137 costimulation-induced diminishment of FDC is T cell dependent. In addition, both CD4+ and CD8+ T cells are recruited into FDC area and are able to regulate FDCs by CD137 costimulation through a direct or indirect mechanism. These studies have revealed a previously unappreciated role of T cells in the regulation of FDC networks.
Introduction
Germinal centers (GC)3 form around specialized accessory follicular dendritic cells (FDCs) in lymphoid follicles, which provide a critical microenvironment for T cell-dependent humoral immune responses. Within GC, Ag-specific B cells efficiently undergo clonal expansion, isotype switching, somatic mutation, affinity maturation, and the generation of plasma and memory cells in response to T-dependent Ags. The formation and maintenance of GC depend on the dynamic interactions among B cells, T cells, and FDCs (1, 2, 3). FDCs play important roles in GC B cell proliferation, survival, and differentiation in both Ag-dependent and -independent manners (4, 5), whereas activated T cells induce the differentiation of GC B cells by providing CD40L and cytokines (6). When Abs are produced during a primary immune response, Ab-Ag complexes are formed and trapped for long periods of time on FDCs due to abundant Ig Fc and complement receptors (CR), among which CR1 and CR2 are the key immune complex-trapping molecules within primary follicles (2, 7). FDCs then present intact Ag for recognition by B cells, which is believed to be crucial for the development of high affinity, isotype-switched, and memory B cell responses (4, 5, 8, 9). In addition to delivering Ag, FDCs seem to supply numerous nonspecific stimuli during GC responses for the generation of an optimal B cell response (10). FDCs express adhesion molecules, including ICAM-1, VCAM-1, and mucosal addressin-cell adhesion molecule-1 (MAdCAM-1), which provide GC B cells with potent accessory signals to prevent apoptosis and aid in B cell selection during receptor maturation processes (11, 12, 13, 14, 15).
Multiple studies suggest that the development of FDCs in lymphoid organs critically depends on the presence of B cells and the cytokines produced by them, but does not depend on T cells (16, 17, 18, 19, 20). Furthermore, TCR-deficient mice show FDC structures similar to those of wild-type (wt) mice (18). Taken together, it appears that B cells are the primary cell type required for the development of FDC, and T cells are not required for the formation of FDC.
CD137 (4-1BB) is a member of the TNFR superfamily (TNFRSF9) and a well-defined T cell costimulatory receptor induced by TCR activation (21). Agonistic mAbs against CD137 have been shown to be effective in promoting T cell-mediated immune responses in vivo (22, 23, 24), but they paradoxically also inhibit T-dependent humoral immunity (25) and prevent autoantibody production in various autoimmune disease models (26, 27, 28). Previous studies attributed CD137-mediated suppression of T-dependent Ab responses to the inhibition of Th cells due to the induction of their anergy, activation-induced cell death, or induction of CD11c+CD8+ suppressor T cells and to the depletion of B cells (25, 26, 27, 28). Our current study unexpectedly found that CD137 costimulation dramatically down-regulates FDC networks in a T cell-dependent manner. Such treatment is able to diminish FDC even after GC formation. These studies clearly demonstrate that activated T cells have a role in regulating FDC.
Materials and Methods
Mice
C57BL/6 wt (B6/wt) mice were purchased from the National Cancer Institute. B6.129S2-Cd4tm1Mak (B6/CD4–/–), B6.129S2-Cd8atm1Mak (B6/CD8–/–), and B6.129P2-Tcrbtm1Mom/J (B6/TCR–/–) mice were purchased from The Jackson Laboratory. Animals were housed in a specific pathogen-free facility maintained by the University of Chicago Animal Resources Center. All animal protocols were approved by the institutional animal care and use committee at University of Chicago.
In vivo immunization and treatment with Abs
The mice were immunized i.p. with 1 x 108 SRBC (Colorado Serum) and/or 100 μg of keyhole limpet hemocyanin (KLH) Ag (Sigma-Aldrich) and treated with PBS, rat IgG, or agonistic anti-CD137 mAb (2A). For secondary immune responses, mice were rechallenged 6 wk later with the same doses of Ag and treated with 2A 8 days before or on the days of rechallenge. 2A was generated as previously described (29), and ascites was produced in SCID mice and purified by passage over a protein G-coupled Sepharose column. Rat IgG was purchased from Sigma-Aldrich and served as a control Ab. Spleens and sera were collected 2 wk after the first immunization and 1 wk after the second immunization.
Measurement of Ag-specific IgG
IgG anti-SRBC Abs were measured as previously described (30). In brief, 96-well Falcon plates (BD Biosciences) were coated with 150 μl of 0.1% SRBC diluted in PBS. Plates were incubated at room temperature for 50–60 min, followed by 0.25% glutaraldehyde (Sigma-Aldrich) in PBS for 5 min. For anti-KLH Abs, 96-well Immulon-4 plates (Dynatech Laboratories) were coated with KLH (10 μg/ml) at 37°C for 1 h. Unbound Ags were washed away with PBS, and plates were then blocked with 0.1% BSA in PBS blocking buffer at 37°C for 1–2 h. Diluted mouse sera were added and incubated at 37°C for 1 h. Pooled sera from C57BL/6 wt mice that were immunized with SRBC and KLH for 2 wk were used as the standard and arbitrarily defined as 100 U. Bound Abs were detected using 100 μl of 1/2000 diluted alkaline phosphatase-conjugated goat anti-mouse IgG-Ab (Southern Biotechnology Associates), followed by addition of the alkaline phosphatase substrate p-nitrophenyl phosphate (Sigma-Aldrich) at 1 mg/ml.
Evaluation of spleen follicle structure by immunohistochemical staining
Spleens were harvested, treated, and stained as previously described (30, 31). In brief, spleens were harvested, embedded in OCT compound (Miles), and frozen at –80°C. Frozen sections (6–8 μm thick) were fixed in cold acetone. Endogenous peroxidase was quenched with 0.2% H2O2 in methanol. After washing, the sections were stained by first incubating them with FITC-conjugated B220 and biotinylated anti-CR (CD35; 8C12; BD Pharmingen) or biotinylated peanut agglutinin (PNA; Vector Laboratories) all at 1/75 to 1/100 dilutions. For MAdCAM-1 and B220 double staining, the sections were first stained by purified anti-MAdCAM-1 (BD Pharmingen), followed by biotinylated goat anti-rat IgG(H+L) (Southern Biotechnology Associates). The sections were then stained with FITC-conjugated B220. HRP-conjugated rabbit anti-FITC (diluted 1/30; DakoCytomation) and alkaline phosphatase-conjugated streptavidin (diluted 1/20; Zymed Laboratories) were added 1 h later. Color development for bound alkaline phosphatase and HRP was performed with an alkaline phosphatase reaction kit (Vector Laboratories) and a Sigma Fast 3,3'-diaminobenzidine tablet set (Sigma-Aldrich). For FDC and T cell subset marker double staining, the tissue sections were first incubated with anti-mouse CD4 or anti-mouse CD8 (BD Pharmingen), followed by biotinylated secondary Ab (Vector Laboratories) and then avidin-biotin-peroxidase complex-alkaline phosphatase (AK-5000; Vector Laboratories). Color was developed with Alkaline Phosphatase Substrate kit III (Vector Laboratories), then the slides were stained with biotinylated CD35, and 3% H2O2 was used to block the endogenous peroxidase activity. The tissues were incubated with avidin-biotin-peroxidase complex-HRP (PK-6100; Vector Laboratories), and color was developed with 3,3'-diaminobenzidine (DakoCytomation). The blocking steps were added to avoid the cross-reaction between the two immunoreactions.
Adoptive transfer
B6/wt mice were immunized with 1 x 108 SRBC plus 100 μg of KLH and treated with either 150 μg of control rat IgG or 2A. Five days later, T cells were purified from these mice using the Pan T cell Isolation kit (Miltenyi Biotec) and were adoptively transferred into B6/TCR–/– mice. One day later, reconstituted B6/TCR–/– recipients were immunized with SRBC plus KLH. Another 10 days later, spleens were harvested, embedded in OCT for tissue staining, and sera were collected for Ab detection by ELISA.
Statistical analysis
All statistics were determined using unpaired Student’s two-tailed t test.
Results
Anti-CD137 treatment inhibits T-dependent IgG response and prevents GC formation
Agonistic mAbs against CD137 were effective in enhancing T cell-mediated responses, butparadoxically inhibited T-dependent humoral immunity. The mechanisms involved in this inhibition have not been well elucidated. SRBC and KLH are commonly used T cell-dependent Ags. Thus, we chose these Ags to examine why CD137 costimulation results in diminished T cell-dependent humoral immune responses. B6/wt mice were immunized with 1 x 108 SRBC plus 100 μg of KLH i.p. and treated with either 100 μg of control Ig or agonistic mAb against CD137 (2A) on the day of immunization. Two weeks later, the amounts of IgG anti-SRBC and anti-KLH detected in the sera of control Ig-treated mice were much higher than those in 2A-treated mice as previously reported (25) (Fig. 1A). To explore the mechanisms of 2A-mediated suppression of the humoral response, we examined GC formation, a key event in maturation of the humoral response. As expected, both PBS-treated and control rat IgG-treated mice showed fully developed GCs in the lymphoid follicles of the spleen (Fig. 1, B and C). On the contrary, 2A treatment completely blocked GC formation in spleen follicles (Fig. 1D).
CD137 costimulation down-regulates FDC
GC formation (cluster of activated B and T cells) inside B cell follicles in vivo requires support from FDCs. Therefore, we examined the effects of CD137 cross-linking on the formation of FDC networks. Two weeks after immunization and Ab treatment, the control mice showed massive GC and FDC network in spleen (Fig. 2, A and C). Surprisingly, we found that 2A treatment on the day of immunization with SRBC and KLH dramatically down-regulated FDC networks and eliminated GC in the spleen, as demonstrated by the disappearance of FDC-specific markers such as CR1 (CD35; Fig. 2, B and D) and FDC-specific mAb (data not shown). The diminished FDC networks were also confirmed by the disappearance of MAdCAM-1 staining within the follicles (Fig. 2F) compared with control spleen (Fig. 2E). Injection of control rat IgG in SRBC- and KLH-immunized mice had no effect on the down-regulation of FDC compared with that in mice immunized with SRBC and KLH alone (data not shown). The reduction of FDC was observed within 7–10 days after 2A treatment, and FDCs recovered 4 wk after treatment. Therefore, it is possible that anti-CD137 treatment inhibits the T cell-mediated humoral immune response by preventing the formation of FDC networks and GC.
CD137 costimulation interferes with FDC and GC after GC formation
To determine whether CD137 engagement could regulate FDC and GC structures after the establishment of a humoral immune response, we immunized B6 mice with 1 x 108 SRBC i.p.; 10 days later, when GCs had formed around FDC in the spleen (Fig. 3, A and B), the mice were treated with either 100 μg of 2A or control rat IgG. Unexpectedly, 10 days after treatment, although the control mice maintained abundant FDCs and GCs (Fig. 3, C and D), 2A-treated mice displayed dramatically reduced FDC networks and were completely lacking in GC (Fig. 3, E and F). Therefore, CD137 signaling can eliminate existing GC after the initiation of a humoral immune response.
The requirement for GC in the memory phase of the humoral immune responses is not clear. Due to the fact that CD137 stimulation can abolish FDC and GC even after initiation of the immune response and that it takes 7–10 days for the dramatic down-regulation of FDC by 2A treatment, we tested whether CD137 ligation inhibits the memory immune response against T-dependent Ags by treating mice before or on the day of the second immunization. B6 mice were immunized with 1 x 108 SRBC with 100 μg of KLH i.p.; 6 wk later, mice were rechallenged with the same Ags. The mice were divided into three groups: 1) mice treated with 200 μg of 2A 8 days before the second immunization, 2) mice treated with 200 μg of 2A on the day of the second immunization, and 3) control mice treated with control rat IgG 8 days before the second immunization. Additionally, mice that were immunized on the day of the second immunization without a first immunization were used as a primary immunization control. As shown in Fig. 4, when 2A treatment began on the day of the second challenge, the IgG recall response against SRBC (Fig. 4A) and KLH (Fig. 4B) was not changed compared with the memory response in the IgG control group. This is in accordance with the previous report that anti-CD137 mAb does not inhibit the secondary IgG anti-pneumococcal surface protein A response in R36A-primed mice (32). However, when 2A administration began 8 days before rechallenge, the memory IgG response against both SRBC and KLH was dramatically reduced compared with that in rat IgG-treated mice. We found that when 2A was administered 8 days before rechallenge, analysis of the spleen follicles on the day of rechallenge showed dramatically reduced FDCs and completely absent GC. When 2A treatment began on the day of the second immunization, reduced FDC and elimination of GC were delayed to 1 wk later (data not shown) compared with the typical recall IgG response time of 2–3 days after rechallenge. These studies imply the potential roles for FDCs and GC in the initiation of the memory phase of humoral responses.
Down-regulation of FDC by CD137 engagement is T cell dependent and can be mediated by both CD4+ and CD8+ T cells
CD137 is primarily expressed on activated T cells, but not on B cells. To test the cellular target of 2A for down-regulating FDC networks during a humoral response, we analyzed the effects of CD137 costimulation on FDC networks in T cell-deficient mice. To our surprise, 2A administration diminishes the formation of FDC networks in wt (Fig. 5, A, B, and I), but not TCR-deficient (Fig. 5, C, D, and I) mice, suggesting that 2A-mediated suppression of FDC clustering is T cell dependent. Therefore, we examined which T cell population is involved. Interestingly, 2A treatment dramatically reduced FDCs in both CD8-deficient and CD4-deficient mice (Fig. 5, E–I), suggesting that both CD4+ and CD8+ T cells are able to down-regulate FDC networks after CD137 cross-linking.
To examine whether CD137 signaling on activated CD4+ and CD8+ T cells can directly interact with FDC, we tested the presence of T cell subsets within the FDC area 5 days after Ag immunization and Ab treatment, a time point before the significant diminishment of FDC. In control mice, we observed that there are few CD4+ cells (Fig. 6A), but no CD8+ T cells (Fig. 6B), present in the FDC area. However, in anti-CD137-treated mice, a significant number of CD8+ T cells were found within the FDC area (Fig. 6D), and CD4+ T cells were also dramatically increased (Fig. 6C). These data suggest that CD137 costimulation promotes the recruitment of activated CD4+ and CD8+ T cells into the FDC network, and they also induce the down-regulation of FDC.
Adoptive transfer of T cells from anti-CD137-treated mice into TCR–/– mice does not diminish FDC in recipients
As well as its expression on T cells, CD137 is detected on FDC (33, 34). To test whether CD137-activated T cells are sufficient to down-regulate FDC in the absence of CD137 engagement on other cell types, we examined whether adoptive transfer of T cells from Ag-immunized and 2A-treated mice could interfere with the FDC network and humoral responses in TCR–/– recipients. B6/wt mice were immunized with SRBC and KLH and treated with either control rat IgG or 2A. Five days later, T cells were purified from these mice and adoptively transferred into TCR–/– mice. The reconstituted recipients were then challenged with SRBC plus KLH the next day. Ten days later, spleens and serum were collected and analyzed. We observed that both groups showed similar FDC (Fig. 7, A and C) and GC (Fig. 7, B and D), and there was no increased CD4 or CD8+ T cell staining within the FDC area in 2A-treated T cell recipients (Fig. 7, G and H) compared with control recipients (Fig. 7, E and F). Accordingly, both groups of mice produced similar level of IgG anti-SRBC and anti-KLH (Fig. 7I). These data imply that CD137 costimulation-activated T cells without continuing CD137 engagement are not capable of down-regulating the FDC network, or that direct CD137 signaling on T cells and other cell types, such as FDC, is essential.
CD137 engagement down-regulates FDC in naive mice
Previous studies have focused on the role of CD137 engagement in an Ag-dependent system; however, baseline T cell activation is sufficient for the CD137 agonist to inhibit FDC, because anti-CD137 treatment without active immunization could down-regulate FDC networks in wt, CD4–/–, and CD8–/– mice (Table I). These studies imply that CD137 costimulation is able to activate T cells in vivo with or without active immunization.
Discussion
Administration of agonistic anti-CD137 Ab can potently inhibit T-dependent humoral immune responses and limit autoantibody production, but the mechanisms are rather unclear. Our current study shows that 2A treatment can diminish the formation of FDC clusters, a critical component in GC formation and T-dependent humoral immune responses. Therefore, it is possible that such treatment could interfere with the T cell-mediated humoral immune response by preventing the formation of FDC and GC. Furthermore, our study shows that CD137 costimulation is not only able to down-regulate FDCs and prevent GC formation, it also can diminish FDCs and eliminate existing GC after the initiation of a humoral immune response. This is consistent with the idea that anti-CD137 treatment interferes with autoantibody production even after autoimmune disease establishment (26, 27). Similarly, 2A treatment can abolish the formation of FDC networks in Fas-deficient mice in the absence of exogenous Ags, followed by a diminished production of autoantibodies (data not shown). Therefore, such treatment may be useful for reducing pre-existing humoral autoimmune responses and blocking the continuation of autoimmune responses.
Our study has revealed an unappreciated role of T cells in the regulation of FDC formation. Although CD137 is detected on FDC (33, 34), the down-regulation of FDC by CD137 engagement seems to be indirect, because anti-CD137 treatment is not able to down-regulate FDC in the absence of TCR+ cells. Interestingly, GC formation is believed to be dependent on the interactions among B cells, CD4+ T cells, and FDCs. CD8+ T cells are generally not involved in the GC response. We show CD137 signaling could recruit both CD4+ and CD8+ T cells into FDC network and induces its diminishment by unknown mechanisms. One possible explanation for these observations is that CD137 engagement down-regulates FDC networks through the activation of CD4+ or CD8+ T cell. In support of this, Wu et al. (32) reported that anti-4-1BB-mediated suppression of the anti-pneumococcal surface protein A response is associated with splenic CD4+ and CD8+ T cell expansion and activation.
Although the diminishment of FDC by CD137 engagement is T cell dependent, whether CD137 signaling on other cell types is also required remains to be determined. When purified T cells from mice immunized with Ag and treated with anti-CD137 are transferred into TCR–/– mice, they failed to interfere with FDC and GC formation, and Ag-specific Ab production was not impaired. One possible explanation is that the feature of anti-CD137-activated T cells requires CD137 signaling for their maintenance. In TCR knockout recipients, without continuing CD137 signaling, control and 2A-treated T cells underwent similar homeostatic and Ag-specific proliferation. The difference between control and anti-CD137-treated T cells was diluted. The evidence that no increased CD4 and CD8+ T cell recruitment into the FDC area in spleens from recipients of anti-CD137-treated T cells compared with control treatment supports this explanation. It is also possible that except for its engagement on T cells, direct CD137 signaling on FDC is required as well. Our preliminary data showed when B6/wt mice were treated with multiple low doses of anti-CD3 (50 μg) or Con A (7.5 mg/kg), which induces massive T cell activation, the number of FDC clusters in spleen was diminished by both treatments (data not shown). CD137 has been shown to be a potent T cell costimulatory molecule, and agonistic anti-CD137 treatment dramatically promotes T cell activation, especially CD8+ T cells in vivo. Together, these studies imply that activated T cells could play a role in regulating FDC networks.
It seems that baseline T cell activation is sufficient for the CD137 agonist to diminish FDC, because anti-CD137 treatment alone, without immunization with exogenous Ag, could down-regulate the FDC network (Table I). Concordantly, Foell et al. (35) reported an increase in spleen weight and total cell number after anti-CD137 mAb treatment. We also observed an enlarged spleen and especially enlarged lymph nodes with increased percentages of activated CD69+ T cells in naive mice treated with agonistic anti-CD137 (data not shown). These studies imply that CD137 costimulation is able to activate T cells in vivo without additional immunization. Therefore, CD137 costimulation could down-regulate FDCs through T cells in an Ag-independent manner.
Our study does not exclude other mechanisms by which agonistic anti-CD137 Ab regulates humoral immune response. Mittler et al. (25) first reported that CD137 costimulation inhibited T cell-dependent humoral immune responses via the induction of Th cell anergy. They found that adoptive transfer of T cells from SRBC-immunized and anti-CD137-treated mice along with B cells from untreated mice into C.B-17 SCID mice failed to generate an anti-SRBC IgG response, whereas the opposite combination produced normal primary and secondary humoral responses to SRBC, isolating the defect to T cells. However, the lack of a CD4+ T cell response can also be explained by the diminished CD4+ T cells after initial activation, because our earlier study showed that such treatment can induce an initial activation, followed by activation-induced cell death (36). Using TCR transgenic DO11.10 mice, we found that administration of agonistic anti-CD137 initially promoted Ag-specific CD4+ T cell proliferation, but subsequently accelerated their depletion by inducing apoptosis. In accordance, in an experimental autoimmune encephalomyelitis model, such treatment initially up-regulated autoreactive Th1 cell functions before down-regulating the overall response (36). These studies suggest the diametric effect of CD137 costimulation on CD4+ T cells. Our previous study showed that agonistic anti-CD137 treatment greatly reduced autoantibody production in Fas-deficient mice accompanied by a diminishment of the number of CD4+ T cells and autoantibody-producing B cells (26). It is possible that the lack of FDC may reduce the survival of autoantibody-producing B cells, because FDC could provide a survival signal for activated B cells. Conversely, in lupus-prone NZB x NZW F1 female mice, Mittler et al. (27) found that agonistic anti-CD137 treatment rapidly suppressed autoantibody production regardless of the age or disease status of the animal without dramatic depletion of either B cells or CD4+ T cells. Reduced IL-2 and IL-4 production by CD4+ T cells was detected in treated mice, and the adoptive transfer of Ag-primed CD4+ T cells or DCs could override anti-CD137-mediated protection. The authors hypothesized that CD137-mediated signaling anergized CD4+ T cells during priming at the DC interface. Interestingly, Seo et al. (28) recently reported that agonistic anti-CD137 mAb treatment inhibited rheumatoid arthritis through induction of CD11c-expressing CD8+ T regulatory cells, which produce IFN- to stimulate dendritic cells and monocytes to produce the enzyme indolamine-2,3-dioxygenase, which further inhibits CD4+ T cell function and autoantibody production.
In summary, these data show that CD137 engagement reduces FDC networks in a T cell-dependent manner, and either CD4+ or CD8+ T cells are able to regulate FDC through CD137 costimulation by direct or indirect mechanisms. This study suggests that T cells could play an active role in regulating FDC networks within a humoral response. It is likely that multiple mechanisms are involved in the suppression of T cell-dependent humoral immune responses by CD137 costimulation. First, it inhibits Th cell function through numerous mechanisms (25, 26, 27, 28). Secondly, it promotes B cell reduction (26). Thirdly, CD137 costimulation induces a T cell-dependent down-regulation of FDC networks, which, as Ag-trapping accessory cells within GCs, are essential for the development of humoral immune responses and memory. Taken together, our study reveals the unappreciated role of T cells in modulating FDC and supports the idea that agonist anti-CD137 can be a powerful reagent in treating autoantibody-mediated autoimmune diseases.
Acknowledgments
We thank Lieping Chen for the generous gift of agonistic anti-CD137 mAb.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported in part by National Institutes of Health Research Grants R01AI062026, R01DK58897, and P01CA09296-01 (to Y.X.F.) and National Institutes of Health Training Grant 5T32HL07237 (to Y.S.).
2 Address correspondence and reprint requests to Dr. Yonglian Sun or Dr, Yang-Xin Fu, Department of Pathology, University of Chicago, MC3083, 5841 South Maryland Avenue, J541, Chicago, IL 60637. E-mail address: ysun1@bsd.uchicago.edu or yfu@uchicago.edu
3 Abbreviations used in this paper: GC, germinal center; CR, complement receptor; FDC, follicular dendritic cell; KLH, keyhole limpet hemocyanin; MAdCAM, mucosal addressin-cell adhesion molecule; PNA, peanut agglutinin; wt, wild type.
Received for publication October 26, 2004. Accepted for publication May 3, 2005.
References
Fu, Y. X., D. D. Chaplin. 1999. Development and maturation of secondary lymphoid tissues. Annu. Rev. Immunol. 17: 399-433.
Kosco-Vilbois, M. H., J. Y. Bonnefoy, Y. Chvatchko. 1997. The physiology of murine germinal center reactions. Immunol. Rev. 156: 127-136.
MacLennan, I. C.. 1994. Germinal centers. Annu. Rev. Immunol. 12: 117-139.
Burton, G. F., D. H. Conrad, A. K. Szakal, J. G. Tew. 1993. Follicular dendritic cells and B cell costimulation. J. Immunol. 150: 31-38.
Tew, J. G., J. Wu, D. Qin, S. Helm, G. F. Burton, A. K. Szakal. 1997. Follicular dendritic cells and presentation of antigen and costimulatory signals to B cells. Immunol. Rev. 156: 39-52.
Li, L., Y. S. Choi. 2002. Follicular dendritic cell-signaling molecules required for proliferation and differentiation of GC-B cells. Semin. Immunol. 14: 259-266.(Yonglian Sun2, Sarah E. B)