Abstract
Activation of the noncanonical pathway through the interaction of lymphotoxin (LT)-α1β2 and LT-βR is essential for the development of secondary lymphoid organs including lymph nodes (LN) and Peyer’s patches (PP). Although TNFR-associated factor (TRAF) 2 and TRAF5 were identified as signal transducers for the LT-βR, roles for TRAF2 and TRAF5 in the development of secondary lymphoid organs remain obscure. In this study, we show that PP but not mesenteric LN development is severely impaired in traf2−/− and traf2−/−traf5−/− mice. Development of VCAM-1+ and ICAM-1+ mesenchymal cells and expression of CXCL13, a crucial chemokine for the development of PP, are severely impaired in PP anlagen in the intestines of traf2−/− mice. Surprisingly, TNF-α stimulation potently up-regulates cxcl13 mRNA expression in wild-type murine embryonic fibroblasts, which is impaired in traf2−/− and relA−/− murine embryonic fibroblasts. Moreover, RelA is recruited to the promoter of cxcl13 gene upon TNF-α stimulation and PP development is impaired in TNFR type 1 (tnfr1)−/− mice. These results underscore a crucial role for the TNFR1-TRAF2-RelA-dependent canonical pathway in the development of PP through up-regulation of cxcl13 mRNA.
The collective term NF-κB refers to dimeric transcription factors that consist of five members: NF-κB1 (p105/p50), NF-κB2 (p100/p52), RelA, RelB, and c-Rel (1). NF-κB1 and NF-κB2 are synthesized as large precursors, p105 and p100, which are posttranscriptionally processed to DNA binding subunits p50 and p52, respectively. Numerous studies have shown that there are two signaling pathways leading to NF-κB activation termed the canonical and noncanonical pathways (1, 2). Proinflammatory cytokines, such as TNF-α and IL-1, activate the canonical pathway in which the activated IκB kinase complex phosphorylates and induces degradation of inhibitor protein IκB, resulting in nuclear translocation of RelA/p50 dimer. The biological consequence of this pathway is to produce inflammatory cytokines and to up-regulate adhesion molecules and anti-apoptotic genes.
In contrast, some members of the TNFR family, such as CD27, CD40, and lymphotoxin (LT)3-βR but not TNFR type 1 (TNFR1), activate the noncanonical pathway (1, 2). The activation of the noncanonical pathway depends on NF-κB-inducing kinase- dependent activation of the IκB kinase α homodimer. Then, activated IκB kinase α subsequently phosphorylates p100, resulting in partial degradation of p100 to generate p52 that translocates into the nucleus as RelB/p52 dimer. The noncanonical pathway has been implicated in production of organogenic chemokines, such as CXCL13, CCL19, and CCL21, and thereby plays a central role in the development of secondary lymphoid organs (2, 3).
Peyer’s patches (PP) development is regulated by an intimate interaction of CD3−CD4+ IL-7Rα+ hemopoietic cells with mesenchymal cells expressing VCAM-1+ and ICAM-1+ cells (4, 5). CD3−CD4+ IL-7Rα+ cells appear in the fetal liver at embryonic day 13.5 (E13.5), and then are recruited to PP anlagen at E15.5. The CD3−CD4+ IL-7Rα+ cells express LT-α1β2 on the cell surface and transmit a signal to VCAM-1+ and ICAM-1+ cells that express LT-βR, resulting in further up-regulation of VCAM-1 and production of various chemokines including CXCL13, CCL19, and CCL21. Chemokines produced by VCAM-1+ and ICAM-1+ cells further promote the recruitment of CD3−CD4+ IL-7Rα+ cells, finally followed by accumulation of T and B cells. CXCL13 and its receptor, CXCR5, play a crucial role in the development of PP because it is severely impaired in cxcl13−/− or cxcr5−/− mice (6, 7).
TRAF2, TRAF5, and TRAF6 have been shown to play crucial roles in the activation of the canonical pathway (1). Although we have shown that TRAF2 and TRAF5 act as adaptor molecules to transmit the signals from LT-βR to NF-κB-inducing kinase (8), neither traf2−/− nor traf5−/− mice showed any defect in lymph node (LN) development (9, 10). However, we cannot formally exclude the possibility that TRAF2 and TRAF5 functionally complement each other. Moreover, a recent study has shown that the noncanonical pathway is constitutively activated in traf2−/− B cells (11). Thus, the roles of TRAF2 and TRAF5 in the development of secondary lymphoid organs remain poorly understood. To explore the role for TRAF2 and TRAF5 in the development of secondary lymphoid organs, we have investigated the development of PP and mesenteric LNs (MLNs) in traf2−/−, traf5−/−, and traf2−/−traf5−/− mice.
Materials and Methods
Reagents and cell culture
Recombinant murine TNF-α were purchased from BD Biosciences. FITC-conjugated and purified anti-VCAM-1 mAbs (M/K2), PE-conjugated anti-ICAM-1 mAb (3E2), PE-conjugated anti-CD4 mAb (L3T4), and allophycocyanin-conjugated anti-CD3ε mAb (2C11) were purchased from BD Biosciences. FITC-conjugated and purified anti-IL-7Rα mAb (A7R34) were purchased from eBioscience. Anti-CXCL13 (Genzyme), HRP-conjugated anti-rat Ig (GE Healthcare), and HRP-conjugated anti-goat Ig (Zymed Laboratories) Abs were purchased from the indicated sources. Anti-LT-βR mAb (AFH6) was provided by J. Browning, Department of Immunology and Inflammation, and P. Rennert (Department of Immunology, Biogen-Idec, Cambridge, MA). Anti-RelA, anti-RelB, and anti-p100 Abs were purchased from Santa Cruz Biotechnology. Wild-type (WT), traf2−/−, traf5−/−, traf2−/−, traf5−/−, and relA−/− murine embryonic fibroblasts (MEFs) were cultured in high glucose DMEM containing 10% FCS. Peritoneal macrophages were cultured in 10% FCS-RPMI 1640.
Mice
The traf2+/−, traf5+/−traf2+/−, traf5+/−, and tnfr1+/− mice are kept in specific pathogen-free conditions. The tnfr1+/− mice were provided by T. W. Mak (Department of Medical Biophysics and Immunology, University of Toronto, Toronto, Ontario, Canada). All animal procedures described in this study were performed in accordance with the guidelines for animal experiments of Juntendo University School of Medicine.
Isolation of CD3−CD4+ IL-7Rα+ cells and VCAM-1+ and ICAM-1+ cells from embryonic intestines
For preparation of CD3−CD4+ IL-7Rα+ cells or VCAM-1+ and ICAM-1+ cells, the small intestines from embryos were detached from the mesenteries, minced into small pieces, and shaken vigorously in the presence of 2.4 U/ml dispase (Invitrogen Life Technologies) at 37°C for 30 min. Then, DNase (Sigma-Aldrich) was added to the cell suspensions to the final concentration of 50 μg/ml, followed by further incubation at 37°C for 10 min. Cell suspensions were passed through nylon meshes to eliminate clumped cells, stained with the indicated mAbs, and analyzed on a FACSCalibur (BD Biosciences).
Immunohistochemistry
Whole-mount immunohistostaining was performed as previously described (12). In brief, specimens were fixed with 2% paraformaldehyde (pH 7.4) for 1 h at 4°C and washed with PBS. After specimens were serially dehydrated with methanol, the intrinsic peroxidase activities were blocked with 0.3% of H2O2, and then rehydrated. After incubation with 1% skim milk plus 0.3% Triton X-100 in PBS to block nonspecific bindings, specimens were incubated with primary Abs overnight at 4°C and washed with 0.3% Triton X-100 in PBS. After incubation with HRP-conjugated secondary Abs, color reactions were developed with diaminobenzidine (Dojindo).
Isolation of peritoneal macrophages
Peritoneal macrophages were isolated 4 days after i.p. thioglycolate injection (Sigma-Aldrich) as previously described (13).
Western blotting
Cells were lysed in RIPA buffer (50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 1% Nonidet P-40, 0.5% deoxycholate, 0.1% SDS, 25 mM β-glycerophosphate, 1 mM sodium orthovanadate, 1 mM sodium fluoride, 1 mM PMSF, 1 μg/ml aprotinin, and 1 μg/ml leupeptin). Western blotting was performed as previously described (14).
Real-time PCR
Total RNAs were extracted by using RNA STAT60 (Tel-Test). First-strand cDNAs were synthesized by using SuperScript II (Invitrogen Life Technologies). Real-time PCR was performed in duplicates in the 7500 Real-Time PCR Detection System using TaqMan Universal PCR Master Mix and Assays-on-Demand gene expression products of the mouse target genes including the cxcl13 (Mm 00444533_m1), ccl19 (Mm 00839967_g1), il6 (Mm 446190_m1), p100 (Mm 00438867_m1), or relB (Mm 00485672_m1) along with an endogenous control (gapdh, Mm99 999 915_g1; Applied Biosystems). The expression levels of these genes were expressed relative to those of gapdh using a 7500 SDS software (Applied Biosystems).
Chromatin immunoprecipitation (ChIP) assay
ChIP assay was performed using a ChIP Assay kit (Upstate Biotechnology) according to the manufacturer’s instruction. The primers used for amplification of each gene promoter are cxcl13 5′-TTGCAGGTGCCAGGGACATAA-3′ and 5′-CTGCCTGGAGGTGGAGTTCAA-3′; and il6 5′-TTCTTAGGGCTAGCCTCAAGG-3′ and 5′-ATGAGCTACAGACATC CCCAG-3′.
Results
PP, but not MLN, development is severely impaired in traf2−/− and traf2−/−traf5−/− mice
To investigate the possibility that TRAF2 and TRAF5 functionally complement each other in the LT-βR-mediated signaling pathway, we first examined the development of MLNs in WT, traf2−/−, traf5−/−, and traf2−/−traf5−/− mice. As we reported previously (9, 10), the development of MLNs and other LNs was not impaired in traf2−/− mice or traf5−/− mice (Fig. 1⇓A and data not shown). Moreover, in contrast to our expectation, the development of MLNs appeared normal even in traf2−/−traf5−/− mice (Fig. 1⇓A), indicating that neither TRAF2 nor TRAF5 play any role in the development of MLNs. Because traf2−/− and traf2−/− traf5−/− mice died at 2∼3 wk after birth, we next examined PP anlagen in the intestines of newborn mice by whole-mount immunostaining with anti-VCAM-1 Ab. As we reported previously (12), PP anlagen were detected as clusters of VCAM-1+ cells in the intestines of newborn WT mice (Fig. 1⇓B). Intriguingly, the number of VCAM-1+ cells was severely reduced in traf2−/− and traf2−/− traf5−/− mice (Fig. 1⇓B). The decrease of VCAM-1+ cells in traf5−/− mice was less severe than in traf2−/− mice, indicating that TRAF2 plays a dominant role in the development of PP. Microscopic analysis of PP using serial sections revealed that large follicles protruding to the abdominal cavity were easily detected in the intestines of WT mice, whereas only small and rudimentary follicles were detected in the intestines of traf2−/−traf5−/− mice (Fig. 1⇓C). Although the defect in PP development in traf2−/− traf5−/− mice was slightly more severe than the defect found in traf2−/− mice (Fig. 1⇓B), we compared the phenotype of WT vs traf2−/− mice in the subsequent analysis.
A severe defect of PP, but not MLN, in traf2−/− and traf2−/−traf5−/− mice. A, Normal development of MLNs. Macroscopic examination of MLNs in WT, traf2−/−, traf5−/−, and traf2−/−traf5−/− mice. The arrows indicate MLNs. B, Expression of VCAM-1 in PP anlagen. The intestines from newborn WT, traf2−/−, traf5−/−, and traf2−/−traf5−/− mice were analyzed by whole-mount immunostaining with anti-VCAM-1 Ab. Arrows indicate the VCAM-1+ spots. C, Histologic analysis of PP. The intestines of 10-day-old WT and traf2−/−traf5−/− mice were serially sectioned and stained with H&E. The original magnifications are ×40 (top panels) and ×100 (bottom panels). Arrows indicate PP.
Expression of CXCL13 in the PP anlagen is severely reduced in traf2−/− mice
We have previously shown that VCAM-1+ and ICAM-1+ mesenchymal cells produce CXCL13, and the CD3−CD4+ IL-7Rα+ hemopoietic cells express its receptor CXCR5 (12). Because cxcl13−/− or cxcr5−/− mice showed a severe defect in PP development (6, 7), interaction between CXCR5 and CXCL13 is considered to be essential for the recruitment of IL-7Rα+ cells to PP anlagen (4, 5). We first examined the accumulation of CD3−CD4+ IL-7Rα+ cells in PP anlagen of traf2−/− mice by whole-mount immunostaining with anti-IL-7Rα Ab. Consistent with a decrease of VCAM-1+ cells, the number of CD3−CD4+ IL-7Rα+ cells was similarly reduced in traf2−/− mice (Fig. 2⇓A, top panels). Thus, we next examined the expression of CXCL13 in PP anlagen of traf2−/− mice. As expected, expression of CXCL13 was severely reduced in PP anlagen of traf2−/− mice (Fig. 2⇓A, bottom panels). To discriminate between the hypotheses that the observed reduction of IL-7Rα+ and VCAM-1+ cells in PP anlagen of traf2−/− mice resulted from a defect of their development, or was the consequence of reduced recruitment to PP anlagen, we investigated the proportion of CD3−CD4+ IL-7Rα+ and VCAM-1+ and ICAM-1+ cells in the whole intestines of embryos at E17.5 and E19.5. The proportion of VCAM-1+ and ICAM-1+ cells was significantly reduced in traf2−/− mice (Fig. 2⇓B, top plots), suggesting that TRAF2 is crucial for the development of VCAM-1+ and ICAM-1+ cells. In contrast, the proportion of CD3−CD4+ IL-7Rα+ cells present in the intestines was not decreased, but rather slightly increased in traf2−/− mice as compared with WT mice (Fig. 2⇓B, bottom plots), suggesting that the recruitment of CD3−CD4+ IL-7Rα+ cells to PP anlagen, but not the development of CD3−CD4+ IL-7Rα+ cells in the fetal liver, is impaired in traf2−/− mice. Similar findings were observed in tnfr1−/−relA−/− mice, in which the proportion of CD3−CD4+ IL-7Rα+ cells was not reduced in the whole intestines, but PP were absent (15).
Development of CXCL13-producing VCAM-1+ and ICAM-1+ cells, but not CD3−CD4+ IL-7Rα+ cells, is impaired in traf2−/− mice. A, IL-7Rα+ and CXCL13+ spots in PP anlagen in WT and traf2−/− mice. The intestines from newborn WT and traf2−/− mice were analyzed by whole-mount immunostaining with anti-IL-7Rα Abs (top panels) and anti-CXCL13 Abs (bottom panels). Arrows indicate the IL-7Rα+ and CXCL13+ spots. B, The proportion of VCAM-1+ to ICAM-1+ cells and CD4+ to IL-7Rα+ cells in the intestines of WT and traf2−/− mice. Single-cell suspensions were prepared from the intestines of WT and traf2−/− embryos at E17.5 and E19.5, and the percentage of VCAM-1+ and ICAM-1+ cells (top plots) and CD4+ and IL-7Rα+ cells (bottom plots) were measured by flow cytometry. We confirmed that CD4+ IL-7Rα+ cells were CD3− (data not shown). Similar results were obtained with five independent experiments.
The noncanonical pathway is constitutively activated in traf2−/− and traf2−/−traf5−/− mice
A previous study has shown that the noncanonical pathway is constitutively activated in traf2−/− B cells (11). However, because PP development was impaired in traf2−/− mice (Fig. 1⇑), we hypothesized that activation of the noncanonical pathway might be impaired in other cell types in traf2−/− mice. In contrast to our expectation, the processing of p100 to p52, a hallmark of activation of the noncanonical pathway, and the nuclear accumulation of RelB were enhanced in the nucleus of splenocytes isolated from 2-wk-old traf2−/− and traf2−/−traf5−/− mice (Fig. 3⇓A). Similarly, the processing of p100 to p52 and the nuclear translocation of p52 and RelB were observed in MEFs from traf2−/− and traf2−/− traf5−/− mice before anti-LT-βR Ab stimulation (Fig. 3⇓, B and C). Collectively, these results show that TRAF2 negatively regulates activation of the noncanonical pathway in various cell types as well as B cells, and TRAF5 does not play any role in the activation of the noncanonical pathway. Moreover, these results indicate that the defect in PP development in traf2−/− and traf2−/− traf5−/− mice might not be caused by a defect in the noncanonical pathway, and a constitutive activation of the noncanonical pathway is not sufficient for the development of PP.
Constitutive activation of the noncanonical pathway in traf2−/− and traf2−/−traf5−/− mice. A, Nuclear translocation of p52 and RelB is enhanced in splenocytes of traf2−/− and traf2−/−traf5−/− mice. Nuclear extracts were prepared from the splenocytes of mice showing the indicated genotype, and the levels of p52 and RelB were analyzed by immunoblotting (IB) with anti-p100 and anti-RelB Abs. The number represents each individual mouse showing the indicated genotype. B, Anti-LT-βR Ab-induced processing of p100. WT, traf2−/−, traf5−/−, and traf2−/−traf5−/− MEFs were stimulated with anti-LT-βR Ab (5 μg/ml) for the indicated time points. The whole cell lysates were analyzed by immunoblotting with anti-p100 Ab. C, Anti-LT-βR Ab-induced nuclear translocation of p52 and RelB. MEFs were stimulated as in B and the nuclear extracts were analyzed as in A.
PP development is impaired in tnfr1−/− mice
Because the TNF-α-induced canonical pathway is impaired in traf2−/− mice (9), we speculated that a defect in the TNF-α-dependent canonical pathway might be responsible for the defective PP development in traf2−/− mice. So far, the contribution of TNF-α-TNFR1 signals to the development of PP has been controversial (16, 17, 18). To directly address this issue, we performed whole-mount immunohistochemical analysis using the intestines of newborn mice. Similar to traf2−/− mice, the number of VCAM-1+ cells and the expression of CXCL13 in PP anlagen were severely reduced in tnfr1−/− mice (Fig. 4⇓A). Similarly, the proportion of VCAM-1+ and ICAM-1+ cells was significantly reduced in the whole intestines of tnfr1−/− embryos at E19.5 (Fig. 4⇓B). In contrast to a previous study (15), the proportion of CD3−CD4+ IL-7Rα+ cells in tnfr1−/− mice was reduced compared with WT mice. The reason for this discrepancy is currently unknown. Collectively, these results show that the TNFR1-mediated signaling pathway plays a crucial role in the development of both VCAM-1+ and ICAM-1+ cells and PPs.
A severe defect of PP in tnfr1−/− mice. A, VCAM-1+ and CXCL13+ spots in PP anlagen in WT and tnfr1−/− mice. The intestines from newborn WT and tnfr1−/− mice were analyzed by whole-mount immunostaining with anti-VCAM-1 Abs (top panels) and anti-CXCL13 Abs (bottom panels). Arrows indicate the VCAM-1+ and CXCL13+ spots. B, The proportion of VCAM-1+ to ICAM-1+ cells and CD4+ to IL-7Rα+ cells in the intestines of WT and tnfr1−/− mice. Single-cell suspensions were prepared from the intestines of WT and tnfr1−/− embryos at E19.5, and the percentage of VCAM-1+ and ICAM-1+ cells (top plots) and CD4+ and IL-7Rα+ cells (bottom plots) were measured as in Fig. 2⇑B. Similar results were obtained with two independent experiments.
TNF-α induces expression of cxcl13 mRNA in WT MEFs, which is impaired in traf2−/− cells
Given that TNFR1 plays a crucial role in the development of PP (Fig. 4⇑A), we surmised that TNF-α stimulation also induces expression of cxcl13 mRNA. Consistent with a previous study (19), anti-LT-βR Ab induced up-regulation of cxcl13 mRNA in WT MEFs (Fig. 5⇓A). Interestingly, TNF-α stimulation potently up-regulated cxcl13 mRNA and this induction was significantly impaired in traf2−/− MEFs (Fig. 5⇓A). In contrast to cxcl13 mRNA, expression of ccl19 mRNA, whose expression is also regulated by the noncanonical pathway, was up-regulated in traf2−/− MEFs before stimulation, and further enhanced upon TNF-α or anti-LT-βR Ab stimulation (Fig. 5⇓A). Moreover, expression of p100 and relB mRNA was slightly up-regulated before stimulation, and the induction levels of these mRNAs in traf2−/− MEFs were comparable to or slightly higher than WT MEFs upon stimulation (Fig. 5⇓A). Taken that the noncanonical pathway was constitutively activated in traf2−/− MEFs, these results suggest that the activation of the noncanonical pathway might be also involved in up-regulation of p100 and relB mRNA. Although TNF-α-induced nuclear translocation of NF-κB was slightly delayed (9), TNF-α-induced il6 mRNA expression was severely impaired in traf2−/− MEFs (Fig. 5⇓A). These results suggest that TNF-α-induced activation of the canonical pathway is impaired in traf2−/− mice. Furthermore, TNF-α-dependent up-regulation of cxcl13 and ccl19 mRNA was also observed in peritoneal macrophages (Fig. 5⇓B).
TNF-α and anti-LT-βR Abs induce up-regulation of cxcl13 mRNA, and TNF-α-induced up-regulation of cxcl13 mRNA is severely impaired in traf2−/− and relA−/− MEFs. A, Total RNA was extracted from either WT or traf2−/− MEFs before (−) and after stimulation with TNF-α Ab (T, 10 ng/ml) for 6 h or anti-LT-βR Ab (L, 5 μg/ml) for 24 h. Expression of the indicated genes was analyzed by real-time PCR. The experiments were performed in duplicate using three different independently prepared MEFs with similar results (data not shown). B, Total RNA was extracted from peritoneal macrophages of WT mice before (−) and after stimulation with TNF-α (T, 10 ng/ml) for 6 h, and gene expression was analyzed as in A. C, WT or relA−/− MEFs were stimulated, and gene expression was analyzed as in A. D, RelA is recruited to the promoter of cxcl13 and il6 genes upon TNF-α stimulation. WT and traf2−/− MEFs were stimulated with TNF-α (10 ng/ml) for the indicated time points, and the recruitment of RelA to the promoter of cxcl13 or il6 gene was examined by ChIP analysis.
RelA is required for up-regulation of cxcl13 mRNA and is recruited to the promoter of cxcl13 gene
TNF-α stimulation induced cxcl13 mRNA expression and this induction was impaired in traf2−/− MEFs (Fig. 5⇑A), suggesting that activation of the canonical pathway might play a crucial role in the induction of cxcl13 mRNA. To directly test this possibility, we stimulated relA−/− MEFs with TNF-α and examined cxcl13 and il6 mRNA expression. As we previously reported (20), TNF-α-induced il6 mRNA expression was severely impaired in relA−/− MEFs (Fig. 5⇑C). Intriguingly, induction of cxcl13 mRNA was absent in relA−/− MEFs upon TNF-α as well as anti-LT-βR Ab stimulation (Fig. 5⇑C). Consistent with a previous study (19), TNF-α Ab-induced and anti-LT-βR Ab-induced up-regulation of p100 and relB mRNA was moderately reduced in relA−/− MEFs (Fig. 5⇑C). However, taken that TNF-α did not induce the processing of p100 to p52 (data not shown), relA might also control cxcl13 mRNA expression by a mechanism other than up-regulation of p100 and relB mRNA. To examine whether RelA directly controls the expression of cxcl13 mRNA, we performed ChIP analysis using the cxcl13 gene promoter. Interestingly, RelA was recruited to the promoter region containing an NF-κB site of cxcl13 gene at 10 min after TNF-α stimulation in WT MEFs (Fig. 5⇑D). Notably, this recruitment of RelA was significantly reduced in traf2−/− MEFs and such recruitment was not detectable at 5 h after TNF-α stimulation. Together, these results show that RelA controls the expression of cxcl13 mRNA through direct binding to its promoter. Similar results were obtained with the promoter of il6 gene (Fig. 5⇑D). However, these results appear to be inconsistent with a previous study (21), which showed that RelA was not recruited to the promoter of cxcl13 upon TNF-α or anti-LT-βR stimulation. These discrepancies might be caused by the different time points (10 min vs 1 and 6 h after stimulation) used for the ChIP assay. Consistent with this notion, TNF-α-induced recruitment of RelA to the promoter of cxcl13 gene was observed at 10 min, but not 5 h after stimulation in WT MEFs (Fig. 5⇑D).
Discussion
In this study, we have shown that TRAF2-dependent activation of the canonical pathway plays a crucial role in the development of PP through up-regulation of cxcl13 mRNA. TNF-α stimulation induces up-regulation of cxcl13 mRNA, and this induction is severely impaired in traf2−/− and relA−/− MEFs. Together, these results have underscored a role for the canonical pathway in the development of PP and the induction of cxcl13 mRNA.
Although previous studies have suggested that TRAF2 and TRAF5 might be signal transducers for the LT-βR, traf2−/− or traf5−/− mice did not show an apparent defect in the development of peripheral LNs or MLNs (8, 9, 10). These results suggested that TRAF2 and TRAF5 functionally complement each other in the LT-βR-mediated signaling pathway. However, in contrast to our expectation, development of MLNs was not impaired even in traf2−/−traf5−/− mice (Fig. 1⇑). Interestingly, a defect in PP development was observed in traf2−/− and traf2−/−traf5−/− mice (Fig. 1⇑). Taken that PP and MLN development are impaired in ltβr−/− mice and the noncanonical pathway is constitutively activated in traf2−/− and traf2−/−traf5−/− cells (11, 22) (Fig. 3⇑), it appears that the constitutively activated noncanonical pathway is sufficient for MLN development in the absence of TRAF2. Moreover, PP but not MLN development was also impaired in tnfr1−/− mice (Fig. 4⇑ and data not shown). Some biological functions of the TNFR1-dependent canonical pathway are impaired in traf2−/− mice (9, 23) (see below), suggesting that the defective PP development in traf2−/− mice might not be due to a defect in the LT-βR-dependent signaling pathway, but rather the TNFR1-dependent canonical pathway.
We have previously reported that TNF-α-induced nuclear translocation of NF-κB is slightly delayed in traf2−/− MEFs, but severely impaired in traf2−/−traf5−/− MEFs, suggesting that TRAF2 and TRAF5 might have redundant function in TNF-α-induced nuclear translocation of NF-κB (9, 23). However, susceptibility to TNF-α plus cycloheximide-induced cell death is enhanced and TNF-α-induced IL-6 production is impaired in traf2−/− MEFs compared with WT MEFs (9, 23) (Fig. 5⇑A). Given that protection from TNF-α-induced cell death and IL-6 production largely depend on activation of the canonical pathway (20, 24), we have no explanation why such a slight delay of NF-κB translocation observed in traf2−/− MEFs might impair the biological functions of the canonical pathway. One possible explanation would be that TRAF2-dependent signaling pathway regulates phosphorylation of RelA, which has been shown to be essential for RelA-dependent transcriptional activity (1). Further study will be required to address this issue. Collectively, these results suggest that TRAF2 plays a redundant role in TNF-α-induced nuclear translocation of NF-κB, but plays critical roles in some RelA-dependent biological functions. In addition, taken that the defect of PP development in traf2−/−traf5−/− mice was more severe than traf2−/− mice and the mild defect of PP development was observed in traf5−/− mice (Fig. 1⇑C), TRAF5 plays some role in the development of PP through activation of the TNFR1-dependent canonical pathway.
Numerous studies have shown a critical role for the noncanonical pathway in the development of secondary lymphoid organs (2, 4). Conversely, a previous study reported that the development of secondary lymphoid organs including MLNs and PP is impaired in tnfr1−/−relA−/− mice (15), suggesting an essential role for the RelA-dependent signaling pathway in the development of secondary lymphoid organs. However, the detailed molecular mechanism remains obscure. Our present study has clearly demonstrated that RelA directly controls the expression of cxcl13 mRNA (Fig. 5⇑). It is possible that the canonical pathway might regulate the expression of cxcl13 mRNA by up-regulating p100 and relB mRNA because expression of p100 and relB mRNA has been shown to be regulated in a RelA-dependent fashion (19). However, this scenario does not fully explain our observation. Indeed, TNF-α- and anti-LT-βR-induced expression of p100 and relB mRNA was moderately reduced in relA−/− MEFs compared with WT MEFs. However, induction of relB but not p100 mRNA was only slightly reduced in traf2−/− MEFs (Fig. 5⇑). Notably, TNF-α stimulation could not induce the processing of p100 to p52, a critical step for activation of the noncanonical pathway (data not shown); nevertheless, TNF-α did up-regulate cxcl13 mRNA in WT MEFs and macrophages (Fig. 5⇑, A and B). Taken that RelA was recruited to the promoter of cxcl13 gene at 10 min after TNF-α stimulation together (Fig. 5⇑D), we have concluded that RelA directly controls the expression of cxcl13 mRNA.
We have previously shown that the intimate interaction of IL-7Rα+ inducer cells and VCAM-1+ and ICAM-1+ mesenchymal cells plays a crucial role in the development of MLNs as well as PP (25). In contrast to tnfr1−/−relA−/− mice (15), the development of PP, but not MLNs, was impaired in traf2−/−, traf2−/−traf5−/−, and tnfr1−/− mice. These results indicate that a receptor other than TNFR1 might independently transmit the signals to develop MLNs even in the absence of TRAF2. The most probable candidate is RANK because the development of MLNs but not PP is severely impaired in rankl−/− and rank−/− mice (26, 27, 28, 29). Taken that TRAF6, but not TRAF2, plays a major role in the RANK-dependent signaling pathway (29, 30), the RANK-TRAF6-dependent, but not TNFR1-TRAF2-dependent, canonical signaling pathway may play a crucial role in the development of MLNs. Collectively, the development of PP and MLNs is differentially regulated by the TRAF family proteins, TRAF2 and TRAF6. In summary, our present study underscores that the TNFR1-TRAF2-RelA-dependent signaling pathway plays a critical role in the development of PP through up-regulation of CXCL13.
Acknowledgments
We thank T. W. Mak, J. L. Browning, K. Honda, M. Matsumoto, S. Yamaoka, T. Saitoh, J. Inoue, J. Gohda, H. Nishina, K. Maeda, and H. Akiba for providing reagents and helpful discussion. We also thank P. Rennert for critically reading the manuscript.
Disclosures
The authors have no financial conflict of interest.
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 grants-in-aid for 21st Century Centers of Excellence Research, Scientific Research (B) from the Japan Society for the Promotion of Science, and grants from the Takeda Science Foundation and the Tokyo Biochemical Research Foundation by a grant from the Human Frontier Science Program, and by a High Technology Research Center Grant from the Ministry of Education, Culture, Sport, Science and Technology, Japan.
↵2 Address correspondence and reprint requests to Dr. Hiroyasu Nakano, Department of Immunology, Juntendo University School of Medicine, 2-1-1 Hongo, Bunkyo-ku, Tokyo 113-8421, Japan. E-mail address: hnakano{at}med.juntendo.ac.jp
↵3 Abbreviations used in this paper: LT, lymphotoxin; TNFR1, TNFR type 1; TRAF, TNFR-associated factor; PP, Peyer’s patch; MEF, murine embryonic fibroblast; LN, lymph node; MLN, mesenteric LN; ChIP, chromatin immunoprecipitation; WT, wild type.
- Received August 21, 2006.
- Accepted November 28, 2006.
- Copyright © 2007 by The American Association of Immunologists
References
In this issue
