The JI PBL Intereron Source
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Weih, D. S.
Right arrow Articles by Weih, F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Weih, D. S.
Right arrow Articles by Weih, F.
Right arrowPubmed/NCBI databases
*Gene*GEO Profiles
*HomoloGene*UniGene
*Substance via MeSH
The Journal of Immunology, 2001, 167: 1909-1919.
Copyright © 2001 by The American Association of Immunologists

Essential Role of RelB in Germinal Center and Marginal Zone Formation and Proper Expression of Homing Chemokines1

Debra S. Weih, Z. Buket Yilmaz and Falk Weih2

Forschungszentrum Karlsruhe, Institute of Toxicology and Genetics, Karlsruhe, Germany


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
High levels of the Rel/NF-{kappa}B family member RelB are restricted to specific regions of thymus, lymph nodes, and Peyer’s patches. In spleen, RelB is expressed in periarteriolar lymphatic sheaths, germinal centers (GCs), and the marginal zone (MZ). In this study, we report that RelB-deficient (relB-/-) mice, in contrast to nfkb1-/-, but similar to nfkb2-/- mice, are unable to form GCs and follicular dendritic cell networks upon Ag challenge in the spleen. RelB is also required for normal organization of the MZ and its population by macrophages and B cells. Reciprocal bone marrow transfers demonstrate that RelB expression in radiation-resistant stromal cells, but not in bone marrow-derived hemopoietic cells, is required for proper formation of GCs, follicular dendritic cell networks, and MZ structures. However, the generation of MZ B cells requires RelB in hemopoietic cells. Expression of TNF ligand/receptor family members is only moderately altered in relB-/- splenocytes. In contrast, expression of homing chemokines is strongly reduced in relB-/- spleen with particularly low mRNA levels of the chemokine B lymphocyte chemoattractant. Our data indicate that activation of p52-RelB heterodimers in stromal cells downstream of TNF/lymphotoxin is required for normal expression of homing chemokines and proper development of spleen microarchitecture.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
The adaptive immune system is characterized by the specific recognition of Ags, central and peripheral tolerance for self Ags, clonal expansion of Ag-specific lymphocytes, and immunological memory. In contrast to innate immune responses, the adaptive immune system requires highly organized and specialized lymphoid organs to exert its function. Development and maturation of B and T lymphocytes take place in primary lymphoid organs, such as bone marrow (BM)3 and thymus, whereas adaptive immune responses to pathogens are initiated in secondary lymphoid organs, such as the spleen and lymph nodes.

NF-{kappa}B plays an important role in immune, inflammatory, and stress responses (1). Five members of this transcription factor family have been identified in vertebrates: NF-{kappa}B1 (encoding the precursor molecule p105 and the processed form p50), NF-{kappa}B2 (encoding the precursor p100 and the processed form p52), RelA (p65), RelB, and c-Rel. The DNA-binding activity of Rel/NF-{kappa}B complexes is regulated by members of the I{kappa}B family, and several distinct I{kappa}B molecules with homologies to ankyrin repeats have been described. In most cell types, Rel/NF-{kappa}B proteins are trapped in the cytoplasm by the I{kappa}B inhibitors. A wide range of stimuli activates the I{kappa}B kinase complex, resulting in the phosphorylation, ubiquitination, and degradation of I{kappa}Bs. Consequently, the Rel/NF-{kappa}B proteins translocate to the nucleus and bind to so-called {kappa}B sequence motifs (2, 3, 4).

The classical NF-{kappa}B activity is a p50-RelA heterodimer, but most other possible homo- and heterodimeric complexes can occur depending on cell type and activation status. One exception is RelB, which only dimerizes with p50 or p52 forming potent transcriptional activators. In the mouse, high levels of RelB expression are restricted to specific regions of lymphoid organs, such as the thymic medulla, periarteriolar lymphatic sheaths (PALS) of the spleen, and the paracortex of lymph nodes. The basal {kappa}B-binding activity in thymus and spleen largely consists of p50-RelB and p52-RelB heterodimers, suggesting a role of RelB in the constitutive expression of {kappa}B-regulated genes in these tissues, whereas RelA and c-Rel complexes appear to be involved in the inducible {kappa}B-binding activity and gene activation (5).

The analysis of Rel/NF-{kappa}B knockout mice revealed that these proteins have essential, but distinct roles in development and function of the immune system (6, 7). Mice with a targeted disruption of RelB display a complex phenotype, including multiorgan inflammation and multifocal defects in immune responses. RelB-deficient mice have thymic atrophy due to a reduced population of dendritic and medullary epithelial cells, lack clearly developed lymph nodes, and develop splenomegaly due to extramedullary hemopoiesis in the red pulp (8, 9, 10, 11). Humoral responses in RelB-deficient mice may be impaired due to an abnormal microarchitecture of the spleen, which does not support proper germinal center (GC) and marginal zone (MZ) formation. GCs are sites of intense B cell proliferation, selection, maturation, and death during Ab responses. Follicular dendritic cells (FDCs) are restricted to the light zones of GCs, and their ability to trap and retain immune complexes (ICs) on their surfaces for long periods of time may be important for the maintenance of immunological memory (12). The splenic MZ is the major route of entry of Ags, APCs, and lymphocytes into the white pulp. The flow of blood from terminating arterioles filters past macrophages, B cells, and dendritic cells (DCs) before reaching the red pulp and rejoining circulation via venous sinuses. Because asplenic people and animals are highly susceptible to encapsulated bacteria, it is thought that the MZ may have a critical role in alerting the immune system to these pathogens (13, 14).

In the present study, we focus our analysis on the phenotypical changes in the spleen microarchitecture in mice lacking p50/NF-{kappa}B1 or RelB. Several cell types could be identified that are affected by the lack of RelB, but that appear normal in nfkb1-/- mice. Adoptive transfer experiments revealed that RelB expression in radiation-resistant stromal cells, but not in BM-derived hemopoietic cells, is required for the establishment of GCs, FDC networks, and MZ structures. However, the generation of MZ B cells requires RelB in hemopoietic cells. We also demonstrate that expression of chemokines that play an important role in the organization of lymphoid organs is reduced in RelB-deficient mice.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Mice

Generation of nfkb1-/-, nfkb2-/-, and relB-/- mice has been described previously (10, 15, 16). Analyses were performed on mice with a mixed B6 x 129 genetic background (for adoptive BM transfer experiments, see below). All animals were housed and bred under standardized conditions with water and food ad libitum in a specific pathogen-free mouse facility at the Forschungszentrum Karlsruhe, Institute of Toxicology and Genetics.

Immunizations and IC trapping

SRBCs (ACILA GMN, Walldorf, Germany) were used to elicit T cell-dependent (TD) immune responses in mice. SRBCs in Alsevers were washed three times in PBS, and mice were injected i.p. with 100 µl 10% SRBC suspension in PBS and sacrificed 10 days later. For IC trapping, mice were immunized with SRBCs, and 6 days later injected i.v. with 200 µg preformed peroxidase antiperoxidase ICs (Dako Diagnostika, Hamburg, Germany) (17). Spleens were removed 24 h later, embedded in Polyfreeze (Polysciences, Warrington, PA), and stored at -80°C. Frozen sections were cut at 10 µm and acetone fixed, and the HRP Ag was detected by diaminobenzidine (DAB), followed by hematoxylin counterstaining and coverslipping.

Immunohistochemical analyses

Frozen blocks were cut at 8–10 µm, and after air drying, sections were fixed in cold acetone before immunohistochemical staining. Staining for GCs was performed with biotinylated peanut agglutinin (PNA; 1/100) and visualized with glucose oxidase reagents (Vector Laboratories, Burlingame, CA). All other immunohistochemical staining procedures were performed with standard avidin/biotin peroxidase complex procedures using products from Vector Laboratories. Endogenous peroxidase activity was quenched in 0.3% hydrogen peroxide, and sections were blocked with avidin D/biotin reagents, followed by 0.5% casein in PBS with 1.5% rat or rabbit serum, as required for appropriate blocking of nonspecific Ab binding. Primary Ab incubation was either at 4°C overnight or at room temperature for 2 h. Appropriate biotinylated secondary anti-rat or anti-rabbit Ab was diluted 1/100 and applied for 30 min. Either DAB or 3-amino-9-ethylcarbazole reagents were used for visualization of the immunostaining, followed by hematoxylin counterstaining and coverslipping. Frozen sections were stained for: IgD (clone 11-26, diluted 1/200; Southern Biotechnology Associates, Birmingham, AL); FDC-M1 (diluted 1/200; gift from M. Kosco-Vilbois, Sereno Pharmaceutical Research Institute, Geneva, Switzerland); CR1/CD35 (clone 8C12, diluted 1/100; PharMingen, San Diego, CA); mucosal addressin cellular adhesion molecule-1 (MAdCAM-1, clone MECA-367, diluted 1/25; PharMingen); ER-TR9 (diluted 1/100; Bachem, Heidelberg, Germany); MOMA-1 (diluted 1/50; Bachem); and ER-TR7 (diluted 1/100; Bachem). Paraffin sections were cut and stained with hematoxylin and eosin (H&E) or for RelB immunohistochemistry with polyclonal rabbit anti-RelB IgG (C-19, diluted 1/200; Santa Cruz Biotechnology, Santa Cruz, CA). For negative control slides, the primary Ab was substituted with normal mixed serum. All negative control slides were free of staining. Micrographs were taken with a Zeiss Axioskop and a Jenoptik ProgRes 3012 digital camera system.

RNA analyses

RNA was extracted from spleen using peqGOLD TriFast reagent according to the manufacturer’s specifications (Peqlab Biotechnologie, Erlangen, Germany). For semiquantitative RT-PCR, 2 µg total RNA was oligo(dT) primed and reverse transcribed using SuperScript II from Life Technologies (Rockville, MD). The following PCR primers were used: TNF (5'-ATG AGC ACA GAA AGC ATG ATC-3' and 5'-TAC AGG CTT GTC ACT CGA ATT-3'); lymphotoxin (LT) {alpha} (5'-ATG ACA CTG CTC GGC CGT CT-3' and 5'-CTA CAG TGC AAA GGC TCC AAA-3'); LT{beta} (5'-TTG TTG GCA GTG CCT ATC ACT GTC C-3' and 5'-CTC GTG TAC CAT AAC GAC CCG TAC-3'); LIGHT (5'-AGA CTG CTG ACC TGC TTT G-3' and 5'-CCC TTC TTT CCT CCC TTT CC-3'); TNFR-I (5'-GAA CCT ACT TGG TGA GTG AC-3' and 5'-CAC AAC TTC ATA CAC TCC TC-3'); LT{beta}R (5'-TTA TCG CAT AGA AAA CCA GAC TTG C-3' and 5'-TCA AAG CCC AGC ACA ATG TC-3'); B lymphocyte chemoattractant (BLC) (5'-ATG AGG CTC AGC ACA GCA AC-3' and 5'-CCA TTT GGC ACG AGG ATT CAC-3'); EBV-induced molecule 1 ligand chemokine (ELC) (5'-GCC TCA GAT TAT CTG CCA T-3' and 5'-AGA CAC AGG GCT CCT TCT GGT-3'); secondary lymphoid organ chemokine (SLC) (5'-ATG ATG ACT CTG AGC CTC C-3' and 5'-GAG CCC TTT CCT TTC TTT CC-3'); CXCR5 (5'-ACT ACC CAC TAA CCC TGG AC-3' and 5'-AGG TGA TGT GGA TGG AGA GGA G-3'); CCR7 (5'-GAG AGA CAA GAA CCA AAA GCA C-3' and 5'-GGG AAG AAT TAG GAG GAA AAG G-3'); and {beta}-actin (5'-AGA GGT ATC CTG ACC CTG AAG TAC C-3' and 5'-CCA CCA GAC AAC ACT GTG TTG GCA T-3'). Amplification conditions using an MJ Research (Cambridge, MA) PTC-225 thermal cycler were 94°C for 1 min, 60°C for 1 min, and 72°C for 1 min for 25 cycles in the presence of 1 µCi [{alpha}-32P]dCTP. For negative controls, reverse transcriptase was omitted. Amplified products were separated in 6% polyacrylamide gels. Northern analysis using 15 µg total spleen RNA was performed as described previously (18). Equal loading was controlled by methylene blue staining of the membrane. Purified probes were labeled with [{alpha}-32P]dCTP using a random priming kit from Amersham Pharmacia Biotech (Piscataway, NJ). Quantifications were performed with a Fuji film FLA-3000 fluorescent image analyzer.

Flow cytometric analyses

Flow cytometry was performed using a BD Biosciences (Mountain View, CA) FACStarPlus flow cytometer and cell sorter. Splenocytes were isolated and RBCs were lysed according to standard procedures (19). For analysis of MZ B cells, splenocytes were labeled with anti-CD23 PE (clone B3B4, 1/200 dilution; PharMingen) and anti-CD21/CD35 FITC (clone 7G6, 1/100 dilution; PharMingen) mAbs in FACS buffer (Ca2+/Mg2+-free PBS, 0.5% BSA). For analysis of surface expression of LT{beta}R ligands, splenocytes were cultured overnight (2 x 106/ml) in RPMI 1640 supplemented with 10% heat-inactivated FCS, penicillin (100 U/ml), streptomycin (100 µg/ml), L-glutamine (2 mM), and 2-ME (50 µM), and either induced with 80 nM PMA and 0.5 µM ionomycin or treated with DMSO as a solvent control. For FACS staining, cells were treated with Fc Block (clone 2.4G2, 1/200 dilution; PharMingen) and then incubated with anti-CD4 FITC (clone RM4-5, 1/100 dilution; PharMingen) or anti-IgD FITC (clone 11-26c.2a, 1/100 dilution; PharMingen) mAbs in FACS buffer. Ligand binding to the LT{beta}R was detected with a rLT{beta}R human IgG1 fusion protein (20) (1/200 dilution), followed by biotinylated mouse-absorbed goat F(ab')2 anti-human IgG (1/100 dilution; Southern Biotechnology Associates) and streptavidin-PE (1/200 dilution; PharMingen). All incubations were for 30 min on ice, followed by two washes with FACS buffer. Analysis was restricted to small cells with a low sideward scatter. An average of 104 cells was recorded in each case.

Adoptive BM transfers and analysis of chimerism

BM cells were isolated from femora of 2- to 3-mo-old wild-type (wt) or relB-/- mice and injected (4–6 x 106 cells i.v. per mouse) into either relB-/- or wt controls (2–3 months old). Before injection, recipient mice had been irradiated with 2 x 550 rad (3-h interval) and rested for 4–6 h after the second irradiation. The following transfers were performed: wt->wt (B6->B6), relB-/-->wt (B6->B6), and wt->relB-/- (B6 x 129->B6 x 129). Six to 8 wk later, recipient mice were injected i.p. with SRBCs. For immunohistochemical analysis, spleens were collected 10 days after immunization. For the analysis of chimerism, peripheral blood was collected by cardiac puncture at necropsy, leukocytes were prepared using ACK lysis buffer, and DNA was extracted for PCR genotyping, as previously described (21). Only mice in which the genotype of PBLs was completely of donor origin were further analyzed.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Immunohistochemical analysis of RelB expression in the spleen

In situ hybridization and immunohistochemistry experiments showed that RelB expression in spleen of naive wt mice is restricted to the white pulp, with high levels in DCs (22, 23). To examine whether RelB is expressed in GCs, we immunized wt mice with SRBCs and stained spleen sections with RelB-specific Abs (Fig. 1Go). White pulp, red pulp, PALS, GCs, and MZ could clearly be identified by H&E staining (Fig. 1GoA). Anti-RelB immunohistochemistry demonstrated very strong expression in PALS of the T cell area (Fig. 1Go, B and C). Lower levels of RelB expression could also be detected in GCs (Fig. 1Go, B and D) and in the MZ (Fig. 1Go, B and E). The specificity of the Ab was demonstrated by the lack of labeling in sections from RelB-deficient mice (data not shown).



View larger version (154K):
[in this window]
[in a new window]
  		FIGURE 1. Immunohistochemical detection of RelB in spleen from SRBC-stimulated wt mice. A, Staining of spleen section with H&E showing white pulp (WP), red pulp (RP), PALS, GC, and MZ. B, Staining with anti-RelB Ab reveals expression in PALS, GCs, and the MZ of the white pulp. C–E, High power views of RelB-positive cells in the white pulp shown in B. C, High expression of RelB in PALS of the T cell area. Lower levels of RelB expression can also be detected in GCs (D) and the MZ (E). Sections in B–E were counterstained with hematoxylin.

 
RelB-deficient mice lack GCs

Previous studies revealed that RelB is required for normal Ab responses after immunization with TD Ags. In particular, RelB-deficient mice show reduced isotype switching, whereas IgM responses are higher than in control animals (11). To analyze whether the impaired Ab response to TD Ags relates to an impaired microarchitecture in secondary lymphoid organs that cannot support proper GC formation, we immunized mice with SRBCs, isolated the spleens 10 days after challenge, and prepared cryosections for immunohistochemical evaluation. Control mice developed typical GCs with PNA+ clusters surrounded by IgD+ B cells (Fig. 2GoA). The plant lectin PNA binds to centroblasts/centrocytes, whereas IgD is highly expressed on mature follicular B cells and down-regulated on most GC B cells. Confirming previous reports (24, 25), nfkb1-/- mice also developed GCs, although PNA staining was overall decreased compared with wt mice (Fig. 2GoB). Similar staining in spleens of RelB-deficient mice failed to detect any GCs and PNA+ clusters. In addition, primary B cell follicles were disorganized and IgD+ B cells were scattered in the T cell area of the splenic white pulp (Fig. 2GoC). Lack of B and T cell segregation in relB-/- spleen was also observed in sections stained with B220 and CD4/CD8 mAbs (data not shown).



View larger version (188K):
[in this window]
[in a new window]
  		FIGURE 2. RelB-deficient mice lack GCs and FDC networks that trap ICs. Splenic microarchitecture and IC trapping of wt (A, D, G, and J), nfkb1-/- (B, E, H, and K), and relB-/- mice (C, F, I, and L). Mice were immunized with SRBCs, and 10 days later splenic cryosections were stained with both anti-IgD (red) and PNA (dark blue, A–C), or with anti-CR1/CD35 (brown, D–F), or with FDC-M1 (brown, G–I). Trapping of preformed ICs in spleens from wt (J), nfkb1-/- (K), and relB-/- mice (L). Extracellular staining (brown) within GCs stems from ICs trapped on FDCs, whereas the intracellular staining stems from ICs phagocytosed by macrophages. Sections in D–L were counterstained with hematoxylin. WP, White pulp; RP, red pulp.

 
Lack of FDC networks and IC trapping in RelB-deficient mice

FDC networks trap and retain ICs on their surfaces and play a crucial role in selecting Ag-specific B cells during Ab responses. To analyze the FDC network in detail, spleen sections from immunized mice were stained with anti-mouse CR1/CD35 (clone 8C12) and FDC-M1 mAbs, both recognizing FDCs. Immunohistochemical analysis revealed a normal pattern of CR1/CD35 staining within splenic B cell follicles in both wt and nfkb1-/- mice (Fig. 2Go, D and E). In contrast, CR1/CD35+ cells were absent from the spleen of relB-/- mice (Fig. 2GoF). A similar result was observed in spleen sections stained with the FDC-M1 mAb. Typical FDC networks were detected within B cell follicles of wt mice, whereas relB-/- animals did not show any FDC-M1 staining. FDC-M1 staining could also be detected in spleens from nfkb1-/- mice, although fewer cells were stained compared with wt controls (Fig. 2GoG–I).

FDCs in GCs retain Ag in an unprocessed form on their surface for the selection of B cells (12). To address to which extent Ag retention occurred in spleens from control and mutant mice, we performed IC trapping experiments. Animals were immunized with SRBCs, and 6 days later injected with preformed peroxidase antiperoxidase ICs. Spleens were removed next day, and the HRP Ag was detected by DAB histochemistry on cryosections. At this time point, FDCs in the GC region of the spleen are the only cell types with extracellular enzyme (17). ICs were readily trapped on FDCs in control mice (Fig. 2GoJ). Whereas reduced IC trapping was observed in nfkb1-/- mice (Fig. 2GoK), relB-/- mice showed no IC trapping in the splenic white pulp (Fig. 2GoL). The strong intracellular staining stems from ICs phagocytosed by macrophages, which in wt and nfkb1-/- mice were largely restricted to the red pulp and the MZ, but which were scattered within the white pulp of relB-/- mice. In summary, RelB is essential for the formation of GCs and FDC networks and for the retention of native Ags in the spleen, whereas the p50 subunit of NF-{kappa}B contributes only minimally to these phenomena.

Impaired formation of splenic MZ macrophage (MZM) populations in RelB-deficient mice

To investigate the role of p50/NF-{kappa}B1 and RelB in development and organization of the splenic MZ, we performed a comparative immunohistochemical analysis of spleen sections from wt, nfkb1-/-, and relB-/- mice. The MZ is characterized by the presence of two distinct macrophage populations, MZM and metallophilic marginal macrophages (MMM), separating red and white pulp in the spleen. These specialized macrophages can be discriminated by the ER-TR9 and MOMA-1 mAbs, respectively (26, 27). Staining of spleen sections from wt and nfkb1-/- mice revealed the two distinct macrophage populations with MZM outside of the marginal sinus (Fig. 3Go, A and B) and MMM, forming a layer outlining the white pulp (Fig. 3Go, D and E). In contrast, formation of these splenic MZM populations was severely impaired in relB-/- mice. The number of ER-TR9-positive MZM was markedly reduced, and they did not form the characteristic ring-like structure (Fig. 3GoC). MOMA-1-positive MMM were present, but scattered throughout the red pulp (Fig. 3GoF). Interestingly, nfkb1-/- mice also had MMM in their red pulp, and the number of MMM in the MZ was slightly reduced compared with wt controls (Fig. 3GoE).



View larger version (184K):
[in this window]
[in a new window]
  		FIGURE 3. Defective organization of the splenic MZ and the marginal sinus in mice lacking RelB. MZ microarchitecture of wt (A, D, G, and J), nfkb1-/- (B, E, H, and K), and relB-/- mice (C, F, I, and L). For splenic MZM populations, cryosections were stained with the ER-TR9 mAb to detect MZM (red, A–C) or with the MOMA-1 mAb to detect MMM (red, D–F). Marginal sinus-lining cells were stained with the MECA-367 mAb directed against MAdCAM-1 (brown, G–I). The reticular fibroblast network in the splenic red pulp was stained with the ER-TR7 mAb (red, J–L). All sections were counterstained with hematoxylin. WP, White pulp; RP, red pulp; CA, central arteriole.

 
Deficient marginal sinus development in relB-/- spleen

The splenic MZ consists of a stromal cell framework in which specialized macrophages, APCs, and a special subset of B cells can move and interact. The MAdCAM-1 is expressed on stromal sinus-lining cells in the MZ. The MECA-367 mAb specifically binds to MAdCAM-1 and can be used as a marker to distinguish the MZ from the white pulp (28, 29). As shown in Fig. 3Go, G and H, both wt and nfkb1-/- mice had a very similar expression pattern of MAdCAM-1 in spleen. The MECA-367 mAb clearly separated red and white pulp, and also stained some FDCs within the B cell follicles of these animals. In marked contrast, MAdCAM-1 was not expressed in the spleen of relB-/- mice, indicating deficient marginal sinus development in the absence of RelB (Fig. 3GoI).

The reticular fibroblast network in the splenic red pulp can be visualized with the ER-TR7 mAb (30). In wt and nfkb1-/- mice, ER-TR7 stained the stromal cell network demarcating the marginal sinus at the border between the MZ and the white pulp. Only few ER-TR7-positive cells were detected in the PALS, while follicular areas were negative (Fig. 3Go, J and K). This organization was defective in spleens from relB-/- mice. The structure of the marginal sinus appeared disrupted, and the reticular fibroblast network was dispersed throughout the white pulp (Fig. 3GoL). Together, these data demonstrate that RelB is required for normal organization of the MZ and its population by macrophages.

RelB-deficient mice lack MZ B cells

MZ B cells represent another population of leukocytes that plays an important role in humoral immunity to T cell-independent (TI) Ags (14, 31). This specialized B cell population is characterized by high surface expression of CD21/CD35 and negative/low levels of CD23, whereas follicular B cells are CD21/CD35intermediateCD23high (32). Flow cytometric analysis of splenocytes from RelB-deficient mice revealed a relative reduction of follicular B cells and a complete lack of CD21/CD35highCD23negative-low MZ B cells (Fig. 4Go), in agreement with the result from the anti-CR1/CD35 immunohistochemistry shown in Fig. 2GoF.



View larger version (34K):
[in this window]
[in a new window]
  		FIGURE 4. RelB is required for the development of MZ B cells. Splenocytes were isolated from wt and relB-/- mice as well as from lethally irradiated wt mice that were reconstituted with relB-/- BM (relB-/-->wt). CD23- and CD21/CD35-expressing lymphocytes were also positive for B220 (not shown). Splenic CD21/CD35highCD23negative-low MZ B cells (R3) are markedly reduced in both RelB-deficient mice and relB-/-->wt BM chimeras. In contrast, CD21/CD35intermediateCD23high follicular B cells (R1) are only slightly reduced in relB-/- mice and even increased in relB-/-->wt chimeras. R2, newly formed B cells and T lymphocytes. Percentages of cells in gates R1 and R3 are shown as means ± SD.

 
Because the organization of the MZ is severely impaired in RelB-deficient mice, signals that are required for normal differentiation of precursor B cells into MZ-type B cells may be reduced or absent in relB-/- spleen. Therefore, we analyzed whether RelB is required in hemopoietic cells, as opposed to being required in supporting nonlymphoid cells. C57BL/6 wt mice were lethally irradiated and reconstituted with BM from relB-/- animals. As shown in Fig. 4Go, mature follicular CD21/CD35intermediateCD23high B cells were reconstituted in the absence of RelB. However, examination of the CD21/CD35highCD23negative-low population revealed an almost complete lack of MZ B cells in the absence of RelB, demonstrating that RelB is required in hemopoietic cells for the generation of MZ B cells.

Expression of TNF ligand/receptor family members in spleen from RelB-deficient mice

Recent studies in mice with targeted mutations revealed essential roles for TNF, LT{alpha}, LT{beta}, and their receptors in lymphoid organ development and function (33, 34, 35). Because TNF and LT{alpha} have been reported to be transcriptionally regulated by NF-{kappa}B (4), we compared mRNA levels of TNF ligand/receptor family members in spleen from wt, nfkb1-/-, nfkb2-/-, and relB-/- mice by semiquantitative RT-PCR analysis. As shown in Fig. 5GoA, expression of the ligands TNF, LT{alpha}, LT{beta}, and LIGHT as well as their receptors TNFR-I and LT{beta}R was readily detected in spleen from all mutant mice. Although expression of LT{alpha} and LT{beta} was reduced 40–50% in relB-/- spleen compared with wt controls, mRNA levels of LIGHT and LT{beta}R were increased {approx}2-fold. Interestingly, spleens from nfkb1-/- and nfkb2-/- mice also showed reduced LT{alpha} mRNA levels, whereas expression of the other family members was not significantly altered.



View larger version (44K):
[in this window]
[in a new window]
  		FIGURE 5. Expression of TNF family members in spleen from RelB-deficient mice. A, Total spleen RNA from wt, nfkb1-/-, nfkb2-/-, and relB-/- mice was reverse transcribed and examined by PCR analysis using primers specific for TNF, LT{alpha}, LT{beta}, LIGHT, TNFR-I, and LT{beta}R. Mean values of mRNA levels from two animals per genotype were normalized for {beta}-actin expression, and samples from wt mice were set to 100%. Only the quantification of the experiment is shown. B, Expression of LT{beta}R ligands on lymphocytes from relB-/- mice. Splenocytes were isolated from wt, nfkb1-/-, and relB-/- mice, cultured overnight, and either induced with PMA and ionomycin or treated with DMSO as a solvent control. Surface expression of ligands binding to the LT{beta}R was detected with a rLT{beta}R-human IgG1 fusion protein. Analysis was restricted to CD4+ T cells. Thin line, Uninduced; thick line, PMA + ionomycin.

 
The LT{beta}R specifically binds the LT{alpha}1{beta}2 heterotrimer and LIGHT, which are expressed on activated lymphocytes (36, 37, 38). Activation of T cells from wt, nfkb1-/-, and relB-/- mice with PMA and ionomycin resulted in increased binding of a LT{beta}R-human IgG1 fusion protein (Fig. 5GoB). We also observed induction of LT{beta}R binding to B cells, although to a lesser extent compared with T cells, and no significant difference was detected between control and mutant mice. However, because GC B cells have not been assessed directly, reduced LT{beta}R binding to this B cell subpopulation in relB-/- mice cannot completely be ruled out. Lymphocytes from nfkb2-/- mice had similar LT{beta}R binding as wt controls (data not shown).

GC and splenic MZ formation requires RelB expression in radioresistant cells

To examine which cell types have to express RelB to initiate and maintain GC reactions, reciprocal BM transplantation experiments were performed and chimeric mice were immunized with SRBCs. BM from donors was transferred into lethally irradiated relB-/- recipients (wt->relB-/-) and vice versa (relB-/-->wt). As a control, BM from wt donors was used to reconstitute irradiated wt recipients (wt->wt). Typical GCs with PNA+ clusters surrounded by IgD+ B cells developed in spleens from both wt->wt and relB-/-->wt mice (Fig. 6Go, A and B), confirming previous results (39). However, similar to relB-/- mice, spleen sections from wt->relB-/- animals had only few, very small PNA+ aggregates (Fig. 6GoC). In addition, FDC networks were absent from wt->relB-/- spleen (Fig. 6GoF), whereas FDC-M1 staining in relB-/-->wt mice was similar to wt->wt controls (Fig. 6Go, D and E).



View larger version (144K):
[in this window]
[in a new window]
  		FIGURE 6. GC and splenic MZ formation requires RelB expression in stromal cells. Irradiated recipients were reconstituted with BM from donors, as indicated. After 6–8 wk, chimeras were immunized i.p. with SRBCs. Ten days later, animals were sacrificed and analyzed for their degree of chimerism, and spleen cryosections were stained with both anti-IgD (red) and PNA (dark blue, A–C), or with FDC-M1 (brown, D–F), or with MOMA-1 (red, G–I). Sections in D–I were counterstained with hematoxylin. WP, White pulp; RP, red pulp.

 
BM chimeras were also analyzed to determine which RelB-positive cell types are required for MZ formation. Similar to the result of relB-/- mice shown in Fig. 2GoF, MOMA-1+ MMM did not form a MZ structure, but were scattered throughout the splenic red pulp of wt->relB-/- animals (Fig. 6GoI). In contrast, MOMA-1 staining in relB-/-->wt mice was comparable with wt->wt controls (Fig. 6Go, G and H). Analysis of 4-wk-old relB-/- mice that at the day of birth were irradiated and injected intrahepatically with wt fetal liver cells confirmed the splenic defects, despite the presence of RelB-positive transferred cells in the white pulp of the recipients (data not shown). Together, these data indicate an essential role of RelB in nonhemopoietic cells for the development of proper B cell follicles, GCs, and FDC networks. Moreover, RelB expression in both radioresistant and hemopoietic cells is required for normal MZ formation and its population by macrophages and B cells.

Reduced expression of homing chemokines in relB-/- spleen

One possible explanation for the observed defects in the organization and microarchitecture of the spleen in relB-/- mice is that radioresistant stromal cells require RelB to generate signals that are needed for proper lymphocyte migration. Recently, several reports have shown an essential role for chemokines and their receptors in cell migration and organization of secondary lymphoid organs (40, 41, 42). We focused our analysis on three chemokines that are constitutively expressed in lymphoid organs. Both ELC and SLC bind to the chemokine receptor CCR7, whereas BLC interacts with CXCR5, also termed BLR-1 (40, 41, 42).

Semiquantitative RT-PCR analysis from total spleen RNA revealed that expression of all chemokines and chemokine receptors was comparable between nfkb1-/- mice and wt controls. In contrast, a marked reduction in BLC mRNA levels was observed in spleens from nfkb2-/- (3.5-fold) and even more pronounced in relB-/- mice (15-fold). ELC and SLC expression was also reduced in spleen from nfkb2-/- and relB-/- mice, but to a lesser extent. Expression of the chemokine receptors CXCR5 and CCR7 was only slightly affected in relB-/- mice and normal in nfkb2-/- animals (Fig. 7Go, A and B). Interestingly, BLC and SLC mRNA levels were reduced to a similar level in both relB-/- mice and wt->relB-/- chimeras, whereas the reduced expression of ELC in relB-/- spleen was partially rescued by adoptive transfer of wt BM (data not shown).



View larger version (71K):
[in this window]
[in a new window]
  		FIGURE 7. Reduced chemokine expression in relB-/- spleen. A, Total spleen RNA from wt, nfkb1-/-, nfkb2-/-, and relB-/- mice was reverse transcribed and examined by PCR using primers specific for the chemokines BLC, ELC, and SLC and their receptors CXCR5 and CCR7. Expression of {beta}-actin is shown as an amplification control. B, Quantification of the experiment shown in A. Mean values of mRNA levels from two animals per genotype were normalized for {beta}-actin expression, and samples from wt mice were set to 100%. C, Northern analysis of BLC and SLC mRNA levels in spleen from wt, nfkb1-/-, nfkb2-/-, relB-/-, nur77TgrelB-/+, and nur77TgrelB-/- mice. Numbers represent mean values of mRNA levels from two animals per genotype (wt mice were set to 100%).

 
To rule out that the reduced BLC and SLC expression in relB-/- spleen results from a dilution effect due to a relative decrease of lymphoid areas in enlarged spleens, we also analyzed nur77-transgenic (Tg) relB-/- mice (nur77TgrelB-/-). The constitutive overexpression of the orphan nuclear receptor Nur77 in thymocytes results in massive apoptosis and a dramatic reduction of peripheral T cells (43). We have previously shown that nur77TgrelB-/- mice do not develop multiorgan inflammation or splenomegaly (44, 45). Northern analysis confirmed the reduced mRNA levels of BLC (20-fold) and SLC (10-fold) in relB-/- spleen (Fig. 7GoC). Moreover, BLC and SLC expression was also clearly reduced in nur77TgrelB-/- spleen (15- and 3.5-fold, respectively) compared with Tg controls. These results indicate that decreased expression of the homing chemokines BLC and SLC by stromal cells contributes to the defective splenic microarchitecture in RelB-deficient mice.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
RelB-deficient mice develop a complex phenotype, including an autoimmune-like inflammatory syndrome, myeloid hyperplasia, multifocal defects in immune responses, and impaired development of lymphoid organs (9, 10, 11). In this study, we focused our analysis on the phenotypical changes in the spleen microarchitecture in mice lacking RelB or the p50 subunit of NF-{kappa}B. Immunohistochemical analysis of wt spleens showed strong RelB expression in the PALS that most likely stems from interdigitating lymphoid DCs (22, 46). T cell area stromal cells also account for the positive RelB staining because RelB expression was observed in the PALS of relB-/-->wt BM chimeras (data not shown). We also found RelB expression in GCs, but not in primary B cell follicles. Clusters of strongly RelB-positive cells bordering the MZ and the white pulp are probably myeloid DCs due to their location interrupting the ring of the MZ (46, 47, 48). It is possible that MAdCAM-1+ marginal sinus cells and MZM express lower levels of RelB. Although costaining experiments revealed RelB expression in a small subset of MZM (data not shown), we still do not know the identity of all RelB-positive cells in the splenic MZ.

Our data demonstrate that RelB is not only required for the normal formation of B cell follicles, PNA-positive GCs, and FDC networks, but it also plays an essential role in the development of the splenic MZ. In contrast, p50-deficient mice have only minimal defects in the development of primary and secondary B cell follicles and in MZ formation. The histopathological changes observed in the spleen of relB-/- mice are also clearly distinct from animals with a targeted mutation of the family member c-Rel (49, 50, 51) (Table IGo and data not shown). However, mice lacking the p52/NF-{kappa}B2 subunit of NF-{kappa}B have defects in the spleen microarchitecture very similar to the ones observed in RelB-deficient mice (16, 25, 52) (Table IGo). In this respect, it is important to note that RelB does not homodimerize and requires p50 or p52 to form transcriptionally active complexes (5). Moreover, adoptive transfers of wt BM into either p52 (52)- or RelB-deficient mice (this study) fail to restore B cell follicles or FDC networks, indicating that expression of p52-RelB complexes in radioresistant stromal cells is required for a proper splenic microarchitecture.


View this table:
[in this window]
[in a new window]
  		Table I. Comparison of spleen phenotypes of mice lacking Rel/NF-{kappa}B or TNF ligand/receptor family membersa

 
Lack of GCs and FDC networks in RelB-deficient mice

Several other mutant mouse lines have defective secondary lymphoid organ microarchitecture. In particular, mice with targeted disruptions of TNF, LT{alpha}, LT{beta}, TNFR-I, or LT{beta}R lack organized GCs and FDC networks (33, 34). Although all of these mutants present unique phenotypes, they also have pathological changes similar to RelB-deficient mice. In addition, the spontaneous alymphoplasia (aly) mutant has severe defects in lymphoid organ development (53), with a complete lack of GCs and MZ structures (54, 55, 56). Importantly, the aly allele carries a point mutation, causing an amino acid substitution in the carboxyl-terminal interaction domain of the NF-{kappa}B-inducing kinase (NIK) that interferes with LT{beta}R-mediated NF-{kappa}B activation (57). The similarities between knockout lines with a defective hemopoietic compartment (i.e., tnf-/-, lta-/-, and ltb-/-) and mutant mice with predominantly stromal defects (i.e., ltbr-/-, aly/aly, nfkb2-/-, and relB-/-) suggest an interdependent network of TNF/LT ligands, their receptors, and NIK, with an important role of p52-RelB heterodimers in the development of a proper microarchitecture in the spleen (summarized in Table IGo).

Because it has recently been shown that GCs and FDC networks depend on the production of TNF, LT{alpha}, and LT{beta} by B cells (58, 59), it was important to rule out intrinsic defects in relB-/- B cells. The reciprocal transplantation experiments demonstrate that the lack of RelB in BM does not affect the ability of B cells to form follicles into which Ag-activated PNA+ B cells can migrate to form GCs. Also, the capacity of relB-/- hemopoietic cells to induce FDC networks in wt recipients is comparable with wt->wt controls. Thus, the source of the defects leading to the lack of GCs and FDC networks cannot be readily transferable hemopoietic cells, but instead appears to stem from radioresistant nonhemopoietic cells. This is also supported by the normal or only moderately reduced TNF, LT{alpha}, and LT{beta} mRNA levels in spleen from p52- and RelB-deficient mice, although we cannot exclude a compensatory role of LIGHT in relB-/- spleen. The reason for the increased LIGHT and LT{beta}R mRNA levels in relB-/- spleen is unclear and requires further investigation. Because FDCs are known to resist high doses of irradiation (60), it is likely that FDCs or FDC precursors are the radioresistant cells that depend on p52-RelB heterodimers for survival or differentiation. Consistent with intrinsic defects in FDC differentiation, we found RelB expression in FDC-like cells within GCs. In addition, both p52 and RelB have previously been reported to be expressed in FDCs (52, 61). Alternatively, the defects may reside in stromal cells distinct from FDCs that are necessary for the formation of normal FDC networks.

Despite the complete lack of GCs or FDC networks, RelB-deficient mice show Ig isotype switching in response to TD Ags, although at reduced levels and with delayed kinetics (11). This finding correlates with normal in vitro maturation of relB-/- B cells to Ig secretion and Ig class switching in response to distinct activators in combination with various cytokines (62). Some affinity maturation and isotype switching in the absence of proper GCs and FDC networks have also been reported in other mutant lines, such as mice deficient for p52, LT{alpha}{beta}, TNFR-I, or LT{beta}R, in particular after repeated immunization with high doses of Ag (52, 63, 64, 65, 66). Thus, one role of GCs and FDC networks may be to provide an optimal environment for effective Ab responses to limiting, and presumably more physiological, levels of Ags.

RelB-deficient mice lack splenic MZ

The spleen plays a major role in the protection against bacterial infections, and the MZ in particular participates in immune responses against TI polysaccharide Ags (13, 67). Using a panel of mAbs against MZM, MMM, and marginal sinus-lining stromal cells, we found that MZ organization in relB-/- spleen is severely impaired, correlating with high susceptibility to bacterial infections and markedly reduced isotype switching in response to TI Ags (11). In addition, RelB-deficient mice lack MZ B cells, a specialized cell type that is required for normal TI type II humoral responses to polysaccharide Ags (31). A marked reduction of MZ B cells was also observed in relB-/-->wt chimeras, indicating an intrinsic defect in MZ B cell development. Similarly, it has been shown that p50/NF-{kappa}B1 is required for MZ B cell generation, and that RelA and c-Rel play significant, but less critical roles in this process (68). Together these findings demonstrate that NF-{kappa}B is required for MZ B cell development and suggest a particular role of p50-RelB complexes.

RelB-deficient mice also completely lack MECA-367+ marginal sinus-lining cells and have markedly reduced numbers of ER-TR9+ MZM. Transfer of wt BM into RelB-deficient recipients does not restore the disrupted marginal sinus structures, suggesting that stromal cells are responsible for the MZ defects in RelB-deficient mice. The MECA-367 mAb recognizes an epitope of the addressin MAdCAM-1 on sinus-lining cells in the MZ. MAdCAM-1 has been shown to be important for extravasation of blood lymphocytes into Peyer’s patches (69), however, not for homing of lymphocytes into the white pulp of adult spleen (28). It remains to be investigated whether MAdCAM-1 expression on sinus-lining cells is required for proper MZ formation during early stages of spleen development.

MOMA-1+ MMM were found scattered throughout the red pulp in relB-/- spleen, indicating that RelB is not essential for the development of MMM, but rather for their proper localization within the splenic MZ. This is supported by adoptive transfer experiments in which relB-/-->wt chimeras showed MOMA-1 staining in the MZ similar to wt->wt controls. In contrast, p52-deficient mice lack MMM and have normal or only slightly reduced numbers of ER-TR9+ MZM (25). BM transfers restore MMM to the MZ in p52-deficient recipients, indicating intrinsic defects in this hemopoietic cell (52). Both p50 and p52 can also interact with the I{kappa}B family member Bcl-3 to form transcriptionally active complexes (70, 71). Interestingly, Bcl-3-deficient mice have dramatically reduced numbers of MMM, and ER-TR9+ MZM are virtually absent in spleen (72). These results suggest different requirements for the development of these two distinct MZM populations with a particular role of p52 and Bcl-3 for MMM, whereas MZM depend on Bcl-3 and RelB.

Chemokine expression in RelB-deficient mice

Chemokines provide important signals for the proper localization of lymphocytes in specialized compartments within lymphoid organs (33, 40). In particular, BLC expressed by follicular stromal cells selectively attracts B cells via CXCR5, whereas expression of SLC by T zone stromal cells attracts T cells and DCs via CCR7 (73). BLC-deficient mice fail to organize B cells into polarized follicular clusters and lack follicle FDCs (74). In addition, mice lacking SLC expression have defects in T cell homing and DC localization in secondary lymphoid organs (75). Thus, BLC and SLC play an essential role in the formation of organized lymphoid tissue.

BLC and SLC expression is markedly reduced in both relB-/- and nfkb2-/- spleen, whereas expression of the corresponding receptors CXCR5 and CCR7 is only slightly affected in relB-/- and normal in nfkb2-/- mice. Reduced SLC levels have also been observed in thymus of RelB-deficient animals (76). Interestingly, the reduction in BLC and SLC mRNA levels in spleen is comparable between relB-/- mice and wt->relB-/- chimeras, indicating that RelB complexes in stromal cells have a role upstream of BLC and SLC. This is consistent with the observation that RelB and BLC are coexpressed in some GC cells, although not all BLC-positive cells also show RelB staining (data not shown). The chemokine ELC is expressed by stromal cells and DCs in lymphoid tissues and strongly attracts naive T cells and activated B cells (77). The reduced expression of ELC correlates with reduced numbers and abnormal distribution of DCs in spleen of RelB-deficient mice (9, 10, 47, 78). Although the reduced ELC levels may contribute to the disrupted microarchitecture in relB-/- spleen, adoptive transfer of wt BM into RelB-deficient recipients partially restored ELC levels, but not GC and FDC network formation.

Recently, it has been shown that LT{alpha}{beta} and TNF on hemopoietic cells are required for stromal cell expression of BLC, ELC, and SLC homing chemokines in B and T cell areas of the spleen (73). Lower ELC and SLC mRNA levels and drastically reduced BLC mRNA expression have also been reported in spleen from aly/aly mice (79). Thus, our data indicate that activation of p52-RelB heterodimers in stromal cells downstream of TNF and LT{alpha}{beta}, probably mediated by TNFR-I/LT{beta}R and NIK, is required directly or indirectly for the expression of factors, such as BLC and SLC, that play an essential role in B cell follicle and FDC network development and lymphocyte compartmentalization in the spleen.


   Acknowledgments
 
We gratefully acknowledge Claudia Stoll, Monika Pech, Norma Howells, and Reina Mebius for excellent technical assistance and advice. We are indebted to Marie Kosco-Vilbois for providing FDC-M1 mAb; Jason G. Cyster for BLC, ELC, and SLC cDNAs and advice; and Jeffrey L. Browning and Paul D. Rennert for anti-LT{alpha} and anti-LT{beta} mAbs as well as LT{beta}R-human IgG fusion proteins. We also thank Harm HogenEsch for valuable comments on this manuscript, Peter Herrlich for continuing support, and all the staff in the animal facility at the Institute of Toxicology and Genetics.


   Footnotes
 
1 This work was supported by the Deutsche Forschungsgemeinschaft (Grants We2224/1-1 and We2224/2-1). Back

2 Address correspondence and reprint requests to Dr. Falk Weih, Forschungszentrum Karlsruhe, Institute of Toxicology and Genetics, P.O. Box 3640, 76021 Karlsruhe, Germany. E-mail address: falk.weih{at}itg.fzk.de Back

3 Abbreviations used in this paper: BM, bone marrow; BLC, B lymphocyte chemoattractant; DAB, diaminobenzidine; DC, dendritic cell; ELC, EBV-induced molecule 1 ligand chemokine; FDC, follicular DC; GC, germinal center; IC, immune complex; LT, lymphotoxin; MAdCAM-1, mucosal addressin cellular adhesion molecule-1; MMM, metallophilic marginal macrophage; MZ, marginal zone; MZM, MZ macrophage; NIK, NF-{kappa}B-inducing kinase; PALS, periarteriolar lymphatic sheath; PNA, peanut agglutinin; SLC, secondary lymphoid organ chemokine; TD, T cell-dependent; Tg, transgenic; TI, T cell-independent; wt, wild type. Back

Received for publication December 14, 2000. Accepted for publication June 5, 2001.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Baeuerle, P. A., T. Henkel. 1994. Function and activation of NF-{kappa}B in the immune system. Annu. Rev. Immunol. 12:141.[Medline]
  2. Ghosh, S., M. J. May, E. B. Kopp. 1998. NF-{kappa}B and Rel proteins: evolutionarily conserved mediators of immune responses. Annu. Rev. Immunol. 16:225.[Medline]
  3. Karin, M.. 1999. How NF-{kappa}B is activated: the role of the I{kappa}B kinase (IKK) complex. Oncogene 18:6867.[Medline]
  4. Pahl, H. L.. 1999. Activators and target genes of Rel/NF-{kappa}B transcription factors. Oncogene 18:6853.[Medline]
  5. Ryseck, R.-P., F. Weih, D. Carrasco, R. Bravo. 1996. RelB, a member of the Rel/NF-{kappa}B family of transcription factors. Braz. J. Med. Biol. Res. 29:895.[Medline]
  6. Attar, R. M., J. Caamaño, D. Carrasco, V. Iotsova, H. Ishikawa, R.-P. Ryseck, F. Weih, R. Bravo. 1997. Genetic approaches to study Rel/NF-{kappa}B/I{kappa}B function in mice. Semin. Cancer Biol. 8:93.[Medline]
  7. Gerondakis, S., M. Grossmann, Y. Nakamura, T. Pohl, R. Grumont. 1999. Genetic approaches in mice to understand Rel/NF-{kappa}B and I{kappa}B function: transgenics and knockouts. Oncogene 18:6888.[Medline]
  8. Lo, D., H. Quill, L. Burkly, B. Scott, R. D. Palmiter, R. L. Brinster. 1992. A recessive defect in lymphocyte or granulocyte function caused by an integrated transgene. Am. J. Pathol. 141:1237.[Abstract]
  9. Burkly, L., C. Hession, L. Ogata, C. Reilly, L. A. Marconi, D. Olson, R. Tizard, R. Cate, D. Lo. 1995. Expression of relB is required for the development of thymic medulla and dendritic cells. Nature 373:531.[Medline]
  10. Weih, F., D. Carrasco, S. K. Durham, D. S. Barton, C. A. Rizzo, R.-P. Ryseck, S. A. Lira, R. Bravo. 1995. Multiorgan inflammation and hematopoietic abnormalities in mice with a targeted disruption of RelB, a member of the NF-{kappa}B/Rel family. Cell 80:331.[Medline]
  11. Weih, F., G. Warr, H. Yang, R. Bravo. 1997. Multifocal defects in immune responses in RelB-deficient mice. J. Immunol. 158:5211.[Abstract]
  12. Kosco-Vilbois, M. H., J. Y. Bonnefoy, Y. Chvatchko. 1997. The physiology of murine germinal center reactions. Immunol. Rev. 156:127.[Medline]
  13. Kraal, G.. 1992. Cells in the marginal zone of the spleen. Int. Rev. Cytol. 132:31.[Medline]
  14. Cyster, J. G.. 2000. B cells on the front line. Nat. Immunol. 1:9.[Medline]
  15. Sha, W. C., H.-C. Liou, E. I. Tuomanen, D. Baltimore. 1995. Targeted disruption of the p50 subunit of NF-{kappa}B leads to multifocal defects in immune responses. Cell 80:321.[Medline]
  16. Caamaño, J. H., C. A. Rizzo, S. K. Durham, D. S. Barton, C. Raventos-Suarez, C. M. Snapper, R. Bravo. 1998. Nuclear factor (NF)-{kappa}B2 (p100/p52) is required for normal splenic microarchitecture and B cell-mediated immune responses. J. Exp. Med. 187:185.[Abstract/Free Full Text]
  17. Chen, L. L., A. M. Frank, J. C. Adams, R. M. Steinman. 1978. Distribution of horseradish peroxidase (HRP)-anti-HRP immune complexes in mouse spleen with special reference to follicular dendritic cells. J. Cell Biol. 79:184.[Abstract/Free Full Text]
  18. Ryseck, R.-P., H. Macdonald-Bravo, M.-G. Mattei, S. Ruppert, R. Bravo. 1989. Structure, mapping and expression of a growth factor inducible gene encoding a putative nuclear hormonal binding receptor. EMBO J. 8:3327.[Medline]
  19. Coligan, J. E., A. M. Kruisbeek, D. H. Margulies, E. M. Shevach, and W. Strober. 1992. Isolation and fractionation of mononuclear cell populations. In Current Protocols in Immunology. J. Wiley and Sons, eds. Greene Publishing Associates & Wiley-Interscience, New York.
  20. Force, W. R., B. N. Walter, C. Hession, R. Tizard, C. A. Kozak, J. L. Browning, C. F. Ware. 1995. Mouse lymphotoxin-{beta} receptor: molecular genetics, ligand binding, and expression. J. Immunol. 155:5280.[Abstract]
  21. Weih, F., S. K. Durham, D. S. Barton, W. C. Sha, D. Baltimore, R. Bravo. 1997. p50-NF-{kappa}B complexes partially compensate for the absence of RelB: severely increased pathology in p50-/-relB-/- double-knockout mice. J. Exp. Med. 185:1359.[Abstract/Free Full Text]
  22. Carrasco, D., R.-P. Ryseck, R. Bravo. 1993. Expression of relB transcripts during lymphoid organ development: specific expression in dendritic antigen-presenting cells. Development 118:1221.[Abstract]
  23. Weih, F., D. Carrasco, R. Bravo. 1994. Constitutive and inducible Rel/NF-{kappa}B activities in mouse thymus and spleen. Oncogene 9:3289.[Medline]
  24. Schwarz, E. M., P. Krimpenfort, A. Berns, I. M. Verma. 1997. Immunological defects in mice with a targeted disruption in Bcl-3. Genes Dev. 11:187.[Abstract/Free Full Text]
  25. Franzoso, G., L. Carlson, L. Poljak, E. W. Shores, S. Epstein, A. Leonardi, A. Grinberg, T. Tran, T. Scharton-Kersten, M. Anver, et al 1998. Mice deficient in nuclear factor (NF)-{kappa}B/p52 present with defects in humoral responses, germinal center reactions, and splenic microarchitecture. J. Exp. Med. 187:147.[Abstract/Free Full Text]
  26. van Vliet, E., M. Melis, W. van Ewijk. 1985. Marginal zone macrophages in the mouse spleen identified by a monoclonal antibody: anatomical correlation with a B cell subpopulation. J. Histochem. Cytochem. 33:40.[Abstract]
  27. Kraal, G., M. Janse. 1986. Marginal metallophilic cells of the mouse spleen identified by a monoclonal antibody. Immunology 58:665.[Medline]
  28. Kraal, G., K. Schornagel, P. R. Streeter, B. Holzmann, E. C. Butcher. 1995. Expression of the mucosal vascular addressin, MAdCAM-1, on sinus-lining cells in the spleen. Am. J. Pathol. 147:763.[Abstract]
  29. Szabo, M. C., E. C. Butcher, L. M. McEvoy. 1997. Specialization of mucosal follicular dendritic cells revealed by mucosal addressin-cell adhesion molecule-1 display. J. Immunol. 158:5584.[Abstract]
  30. van Vliet, E., M. Melis, J. M. Foidart, W. Van Ewijk. 1986. Reticular fibroblasts in peripheral lymphoid organs identified by a monoclonal antibody. J. Histochem. Cytochem. 34:883.[Abstract]
  31. Guinamard, R., M. Okigaki, J. Schlessinger, J. V. Ravetch. 2000. Absence of marginal zone B cells in Pyk-2-deficient mice defines their role in the humoral response. Nat. Immunol. 1:31.[Medline]
  32. Oliver, A. M., F. Martin, G. L. Gartland, R. H. Carter, J. F. Kearney. 1997. Marginal zone B cells exhibit unique activation, proliferative and immunoglobulin secretory responses. Eur. J. Immunol. 27:2366.[Medline]
  33. Fu, Y. X., D. D. Chaplin. 1999. Development and maturation of secondary lymphoid tissues. Annu. Rev. Immunol. 17:399.[Medline]
  34. Matsumoto, M.. 1999. Role of TNF ligand and receptor family in the lymphoid organogenesis defined by gene targeting. J. Med. Invest. 46:141.[Medline]
  35. Pasparakis, M., S. Kousteni, J. Peschon, G. Kollias. 2000. Tumor necrosis factor and the p55TNF receptor are required for optimal development of the marginal sinus and for migration of follicular dendritic cell precursors into splenic follicles. Cell. Immunol. 201:33.[Medline]
  36. Crowe, P. D., T. L. VanArsdale, B. N. Walter, C. F. Ware, C. Hession, B. Ehrenfels, J. L. Browning, W. S. Din, R. G. Goodwin, C. A. Smith. 1994. A lymphotoxin-{beta}-specific receptor. Science 264:707.[Abstract/Free Full Text]
  37. Mauri, D. N., R. Ebner, R. I. Montgomery, K. D. Kochel, T. C. Cheung, G. L. Yu, S. Ruben, M. Murphy, R. J. Eisenberg, G. H. Cohen, et al 1998. LIGHT, a new member of the TNF superfamily, and lymphotoxin {alpha} are ligands for herpesvirus entry mediator. Immunity 8:21.[Medline]
  38. Zhai, Y., R. Guo, T. L. Hsu, G. L. Yu, J. Ni, B. S. Kwon, G. W. Jiang, J. Lu, J. Tan, M. Ugustus, et al 1998. LIGHT, a novel ligand for lymphotoxin {beta} receptor and TR2/HVEM induces apoptosis and suppresses in vivo tumor formation via gene transfer. J. Clin. Invest. 102:1142.[Medline]
  39. Gerloni, M., D. Lo, M. Zanetti. 1998. DNA immunization in relB-deficient mice discloses a role for dendritic cells in IgM->IgG1 switch in vivo. Eur. J. Immunol. 28:516.[Medline]
  40. Cyster, J. G.. 1999. Chemokines and cell migration in secondary lymphoid organs. Science 286:2098.[Abstract/Free Full Text]
  41. Zlotnik, A., O. Yoshie. 2000. Chemokines: a new classification system and their role in immunity. Immunity 12:121.[Medline]
  42. Cyster, J. G.. 2000. Leukocyte migration: scent of the T zone. Curr. Biol. 10:R30.[Medline]
  43. Weih, F., R.-P. Ryseck, L. Chen, R. Bravo. 1996. Apoptosis of nur77/N10-transgenic thymocytes involves the Fas/Fas ligand pathway. Proc. Natl. Acad. Sci. USA 93:5533.[Abstract/Free Full Text]
  44. Weih, F., S. K. Durham, D. S. Barton, W. C. Sha, D. Baltimore, R. Bravo. 1996. Both multiorgan inflammation and myeloid hyperplasia in RelB-deficient mice are T cell dependent. J. Immunol. 157:3974.[Abstract]
  45. Barton, D., H. HogenEsch, F. Weih. 2000. Mice lacking the transcription factor RelB develop T cell-dependent skin lesions similar to human atopic dermatitis. Eur. J. Immunol. 30:2323.[Medline]
  46. Steinman, R. M., M. Pack, K. Inaba. 1997. Dendritic cells in the T-cell areas of lymphoid organs. Immunol. Rev. 156:25.[Medline]
  47. Wu, L., A. D’Amico, K. D. Winkel, M. Suter, D. Lo, K. Shortman. 1998. RelB is essential for the development of myeloid-related CD8{alpha}- dendritic cells but not of lymphoid-related CD8{alpha}+ dendritic cells. Immunity 9:839.[Medline]
  48. Crowley, M. T., C. R. Reilly, D. Lo. 1999. Influence of lymphocytes on the presence and organization of dendritic cell subsets in the spleen. J. Immunol. 163:4894.[Abstract/Free Full Text]
  49. Köntgen, F., R. J. Grumont, A. Strasser, D. Metcalf, R. Li, D. Tarlinton, S. Gerondakis. 1995. Mice lacking the c-rel proto-oncogene exhibit defects in lymphocyte proliferation, humoral immunity, and interleukin-2 expression. Genes Dev. 9:1965.[Abstract/Free Full Text]
  50. Carrasco, D., J. Cheng, A. Lewin, G. Warr, H. Yang, C. Rizzo, F. Rosas, C. Snapper, R. Bravo. 1998. Multiple hemopoietic defects and lymphoid hyperplasia in mice lacking the transcriptional activation domain of the c-Rel protein. J. Exp. Med. 187:973.[Abstract/Free Full Text]
  51. Liou, H.-C., Z. Jin, J. Tumang, S. Andjelic, K. A. Smith, M.-L. Liou. 1999. c-Rel is crucial for lymphocyte proliferation but dispensable for T cell effector function. Int. Immunol. 11:361.[Abstract/Free Full Text]
  52. Poljak, L., L. Carlson, K. Cunningham, M. H. Kosco-Vilbois, U. Siebenlist. 1999. Distinct activities of p52/NF-{kappa}B required for proper secondary lymphoid organ microarchitectur