Control of Toll-like Receptor-mediated T Cell-independent Type 1 Antibody Responses by the Inducible Nuclear Protein IκB-ζ

Background: IκB-ζ, a member of the IκB family of nuclear proteins that regulates transcription, can be induced by Toll-like receptor (TLR) signaling. Results: IκB-ζ deficiency in B cells reduced T cell-independent antibody response-1. Conclusion: IκB-ζ is a key regulator of TLR-mediated class switch recombination (CSR) in B cells. Significance: T cell-dependent and -independent antibody responses are regulated by different mechanisms.


Antibody responses have been classified as being either T celldependent or T cell-independent (TI). TI antibody responses are further classified as being either type 1 (TI-1) or type 2 (TI-2), depending on their requirement for B cell-mediated antigen receptor signaling. Although the mechanistic basis of antibody responses has been studied extensively, it remains unclear whether different antibody responses share similarities in their transcriptional regulation. Here, we show that mice deficient in IB-, specifically in their B cells, have impaired TI-1 antibody responses but normal T cell-dependent and TI-2 antibody responses. The absence of IB-in B cells also impaired proliferation triggered by Toll-like receptor (TLR) activation, plasma cell differentiation, and class switch recombination (CSR). Mechanistically, IB--deficient B cells could not induce TLRmediated induction of activation-induced cytidine deaminase (AID), a class-switch DNA recombinase. Retroviral transduction of AID in IB--deficient B cells restored CSR activity. Furthermore, acetylation of histone H3 in the vicinity of the transcription start site of the gene that encodes AID was reduced in IB--deficient B cells relative to IB--expressing B cells. These results indicate that IB-regulates TLR-mediated CSR by inducing AID. Moreover, IB-defines differences in the transcriptional regulation of different antibody responses.
Antibody responses are central to protecting hosts from pathogen infection. After B cells recognize antigens, they initiate three steps, proliferation, class switch recombination (CSR), 2 and plasma cell differentiation, that are required for antibody production. In terms of antibody responses, antigens are typically classified as being either T cell-dependent (TD) or T cell-independent (TI) antigens (1). TD antigens are soluble proteins or peptides that are recognized by specific B cell receptors and induce clonal activation of B cells; TD antibody responses require the interaction of the CD40 ligand on a T cell with a CD40 receptor on a B cell (2,3). In contrast, TI antigens can initiate antibody responses independently of T cells. TI antibody responses are classically defined as TI type 1 (TI-1) antigens and TI type 2 (TI-2) antigens, depending on their requirement for Btk, which is a key kinase needed for B cell antigen receptor (BCR) signaling (4,5). The TI-1 antigen TNP-LPS, but not the TI-2 antigen TNP-Ficoll, can elicit anti-TNP plaque-forming cell responses in Btk-deficient mice (4). Thus, BCR signaling is necessary for responses triggered by TI-2 antigen but dispensable for responses triggered by the TI-1 antigen. TI-2 antigens, which contain a repetitive epitope such as capsular polysaccharide, induce strong BCR signaling by engaging multiple BCRs, which induces clonal B cell activation and antigen-specific immunoglobulin (Ig) production (6). TI-1 antigens, such as LPS, are considered to act as mitogens that stimulate B cells to produce polyclonal antibodies following Toll-like receptor (TLR) stimulation (7,8). However, the polysaccharide moiety of the LPS binds to the BCRs of multiple B cells (9). As a consequence, LPS can induce the production of not only polyclonal Igs but also antigen-specific Igs by co-en-gaging TLR4 and BCR. In addition, co-stimulation of other TLR ligands and BCR induces strong activation-induced cytidine deaminase (AID) expression and a high rate of CSR. Thus, TLRmediated antibody responses are divided into BCR-independent polyclonal responses and BCR-dependent clonal responses.
Although the mechanism of antibody responses varies widely between the types of antigens described above, it remains unclear whether common transcriptional factors regulate both TD and TI antibody responses. CSR in B cells switches one isotype of an antibody to another. AID is thought to be a master regulator of CSR, which is regulated by transcriptional factors that include Bach2, IRF4, and Hoxc4 (10 -16).
The observation that deficiencies in any of these four transcriptional factors impair both TD-and TI-induced AID and CSR suggests that the same mechanisms of transcriptional regulation operate both in TD and TI antibody responses. However, the signaling pathway activated by CD40, which is a key receptor for the TD antibody response, clearly differs from that triggered by TLR activation. Thus, it is possible that transcriptional regulation of AID is regulated by factors that differ between the TD and TI antibody responses.
This study focused on the function of nuclear IB family member IB-in B cell-mediated antibody responses. IB-is a transcriptional regulator that interacts with NF-B in macrophages (17,18). Previous studies showed that IBis key regulator of innate and adaptive immune responses, such as Th17 development, NK cell-derived IFN-␥ production, and IL-6 production in macrophages (19 -22). In epithelial cells, a deficiency in IB-causes apoptosis, which induces Sjögren's syndromelike inflammation (23). We have recently shown that IB-controls TLR-induced IL-10 production in B cells (24). However, a role for IB-in B cell antibody responses has never been reported. Here, we report that a deficiency of IB-specifically in B cells impaired TI-1, but not TD and TI-2, antibody responses both in vitro and in vivo. Furthermore, we showed that a deficiency in IB--impaired TLR induced proliferation, CSR, and differentiation of plasma cells. Notably, IB--deficient B cells did exhibit AID expression by anti-CD40 stimulation but not LPS stimulation. Furthermore, IB-is essential for the co-stimulation of either TLR2 or TLR9 with BCR to ensure CSR. These findings indicate that the IB--regulated transcriptional network controls TLR-mediated antibody responses. These results reveal that IB-defines a key distinction between TD and TI antibody responses.

EXPERIMENTAL PROCEDURES
Mice-The loxP-flanked Nfkbiz allele has been described previously (23). We generated Nfkbiz fl/⌬ Mb1 cre/ϩ mice by crossing of Nfkbiz fl/⌬ mice with Mb-1 cre mice (25). All mice were kept under specific pathogen-free conditions in the animal facilities of Tohoku University. All animal protocols were approved by the Institutional Animal Care and Use Committee.
Cells-B cells were purified from the spleen by using a B cell isolation kit for negative depletion of cells that express CD43, CD4, or Ter-119 (Miltenyi Biotech, Bergisch Gladbach, Germany). Use of the kit according to the manufacturer's protocol resulted in a purity of Ͼ95% of B220 ϩ B cells. The murine B lymphoma cell line CH12F3-2A (Riken Cell Bank, Tsukuba, Japan) was cultured in RPMI 1640 medium supplemented with 10% heat-inactivated FCS, 100 units/ml penicillin, 100 g/ml streptomycin, 2 mM L-glutamine, 0.1 mM nonessential amino acids, 1 mM sodium pyruvate, and 50 M 2-mercaptoethanol at 37°C in 5% CO 2 .
Immunization and ELISA-Basal serum Ig titers were quantified by ELISA by using HRP-conjugated Ig from Southern Biotech (Victoria, Australia). To evaluate T cell-dependent or T cell-independent antibody responses, mice were administered intraperitoneal doses of the following: 100 g of TNP-KLH in alum; 50 g of TNP-Ficoll; or 50 g of TNP-LPS. Titers of antibodies to TNP were measured by ELISA with plate-bound TNP-conjugated BSA (Biosearch Technologies) and isotypespecific horseradish peroxidase-conjugated secondary antibodies (Southern Biotech).
Flow Cytometry-Cell surface antigens were stained in the dark at 4°C with antibodies diluted in PBS that contained 0.5% bovine serum albumin. Cells were analyzed using a Galios instrument (Beckman Coulter). Dead cells (DAPI ϩ ) were excluded from the analysis. B cells, T cells, dendritic cells, and macrophages with the B220 ϩ , CD3 ϩ , CD11c ϩ , or CD11b ϩ genotype were purified (Ͼ95%) from Nfkbiz fl/⌬ or Nfkbiz fl/⌬ Mb1 cre/ϩ mice using a Aria II cell sorter (BD Biosciences).
Analysis of in Vitro CSR-Splenic B cells were stimulated with LPS (20 g/ml) or anti-CD40 antibodies and the additional reagents indicated below. No additional reagents were added for CSR to IgG3 and IgG2b, but 5 ng/ml mouse IL-4 (5 ng/ml) was added for CSR to IgG1, 50 ng/ml mouse IFN-␥ (PeproTech) was added for CSR to IgG2a, and 1 ng/ml TGF-␤1 (PeproTech) was added for CSR to IgA. Supernatants from cell cultures were collected on day 7 to analyze the secretion of Igs. To analyze surface Igs, cells were collected on day 3 and stained with phycorerythrin-labeled rat mAb to mouse IgG1.
Real Time RT-PCR-Total RNA was prepared using RNAiso Plus. Levels of mRNA were quantified by real time RT-PCR using the High Capacity cDNA reverse transcription kit (Applied Biosystems, Foster City, CA) and SYBR Premix EX TaqII (Takara Bio Inc., Otsu, Japan) with a LightCycler 3302 instrument (Roche Diagnostics). The primer sequences used are shown in Table 2.
RNA Sequence-Total RNA were purified from LPS-stimulated (20 g/ml) splenic B cells on day 3 by RNeasy (Qiagen, Venlo, Netherlands), according to this study. Poly(A) mRNAs were purified from total RNA using the poly(A) mRNA magnetic isolation module (New England Biolabs, Ipswich, MA). Libraries were prepared using the Next Ultra RNA library prep kit for Illumina (New England Biolabs). After the preparation of the RNA library, we performed sequencing using an Illumina IIx genome analyzer. Reads (38 bp) were mapped to the mouse genome (mm9 from University of California at Santa Cruz genome browser database) using the TopHat Version 2.0.0 algorithm with default settings. Only reads with a Phred quality score greater than or equal to 25 were analyzed. The BED Tools package (27) was used to filter rRNA (ribosomal RNA) and tRNA (transfer RNA), with rRNA and tRNA annotations downloaded from the University of California at Santa Cruz table browser. The data have been entered into the NCBI Gene Expression Omnibus (accession number GSE57837). The data were modified and shown in Table 3. (In order to exclude those genes with very low expression, only genes with a RNA-seq score of Ͼ0.05 in at least one sample were chosen. Of this gene set, those genes with a Ͻ0.5-fold change in expression in the sample from B-cell-specific Nfkbiz-deficient (cKO) mice compared to the sample from control are shown.) Immunoblotting-Cells were lysed, subjected to 10% SDS-PAGE, and analyzed by immunoblotting with anti-IBor anti-␤-actin antibodies, and secondary antibodies were conjugated with horseradish peroxidase. Bound antibodies were visualized by chemiluminescence after incubation with Immobilon Western Chemiluminescent HRP substrate.
Retroviral Transduction-The cDNAs that encode BATF or AID were cloned into pMY-IREIS-EGFP (28). Recombinant retroviruses were prepared by transfecting the Plat-E packaging cells with plasmid, using calcium phosphate transfection. B cells were stimulated with anti-IgD for 24 h and were infected with the viral supernatants in the presence of Polybrene (5 g/ml) by spin infection for 90 min at 800 ϫ g at 32°C. The cells

Gene Orientation Sequence
were incubated at 37°C in 5% CO 2 for 2 h and stimulated by exposure to both LPS and IL-4 to induce CSR. Transfection-CH12F3-2A cells were transfected by electroporation with each reporter plus phRL-TK (Promega Corp., Madison, WI). One day after electroporation, the cells were stimulated either with LPS plus IL-4 or with anti-CD40 plus IL-4.
Luciferase Assay-Cells were stimulated as indicated and lysed for luciferase assay. Luciferase activity was measured by the Dual-Luciferase TM reporter assay system according to the manufacturer's instructions (Promega Corp.).
ChIP Assay-Splenic B cells were activated with LPS plus IL-4 for 3 days. Cells were fixed for 10 min at 25°C in 1% (w/v) formaldehyde. Cross-linking was terminated by the addition of 150 mM glycine. After being washed with ice-cold PBS containing 0.5% BSA, cells were lysed by sonication in SDS lysis buffer (1% (w/v) SDS, 10 mM EDTA, and 50 mM Tris, pH 8.0). Debris was removed by centrifugation. Lysates were cleared by mixing with Protein G-Sepharose (GE Healthcare) plus salmon sperm DNA (Invitrogen). A ChIP assay was performed using antibodies against acetyl-histone H3 (Lys-27) and normal rabbit IgG. Quantitative PCR was performed with a LightCycler using the primers described in Table 2.
Statistical Analysis-Paired data were evaluated with Student's t test. A value of p Ͻ 0.05 was considered statistically significant.

Mice Deficient in IB-Specifically in Their B Cells Have
Impaired TI-1 Antibody Responses-The transcriptional regulator IB-can be up-regulated by BCR-or LPS-mediated stimulation of B cells through transcriptional and/or posttranscriptional regulation (24). IB--deficient mice exhibit Sjögren's syndrome-like autoimmune disease and abnormal B cell activation (23). However, given that those phenotypes are triggered by epithelial cell death in lacrimal gland, the role of IB-in B cells remains poorly defined. To better understand the role of IB-in B cells, we took advantage of Cre-lox technology to generate a B cell-specific deletion of the Nfkbiz gene by crossing mice with the Nfkbiz flox allele to mice that express the Cre recombinase under the control of the murine Cd79a promoter (Cd79a-Cre, also known as Mb1-Cre). This confirmed that Nfkbiz expression in cKO mice was reduced in B cells but not in other immune cells (Fig. 1A) (28). These mice appeared healthy and grew without any phenotypic abnormalities (23). Examination of the serum Ig concentration in cKO (Mb-1 Cre;Nfkbiz fl/⌬ ) mice revealed that levels of IgM, IgG1, IgG2b, IgG3, and IgA were comparable in cKO and control (Nfkbiz fl/⌬ ) mice (Fig. 1B).
Next, we analyzed the role of IB-in antigen-specific B cell responses by administration of either a TD antigen (TNP-KLH in alum), a TI-2 antigen (TNP-Ficoll), or a TI-1 antigen (TNP-LPS) in vivo. In the cases of TNP-KLH and TNP-Ficoll, levels of TNP-specific antibody production were comparable in control and cKO mice (Fig. 1, C and D). Surprisingly, TNP-specific IgM production induced by TNP-LPS was modestly reduced, and IgG3 production was completely impaired in cKO mice (Fig.  (29), we next examined whether IBdeficiency affects the development of subsets of peripheral B cells. The numbers of B220 ϩ B cells and B220 ϩ AA4.1 ϩ immature B cells in the spleens of cKO mice were identical to those in control mice ( Fig. 2A). Likewise, the numbers of sIgM-sIgD ϩ mature B cells, CD21 high CD23 low marginal zone B cells, and CD21 low CD23 high follicular B cells were also the same in the two groups of mice (Fig. 2, B and C). However, cKO mice had slightly fewer sIgM ϩ sIgD ϩ B cells than control mice. These results suggest that IB-is dispensable for the development of marginal zone and follicular B cells. In addition, subsets of B cells in the peritoneal cavity, such as B1a (B220 low CD5 hi ), B1b (B220 low CD5 low ), and B2 (B220 hi CD5 low ), were equally abundant in cKO and control mice (Fig. 2D). These results suggested that B cell maturation does not play a critical role in impairing TI-1 antibody responses in cKO mice.

Stimulation of TLR, but Not CD40, Induces IB-via Post-
transcriptional Regulation-We next investigated why IBdeficiency only affects TI-1 antibody responses. Our previous study demonstrated that the induction of IB-protein following BCR stimulation was weaker than that after TLR stimulation even though the increase in the level of the mRNA that encodes IB-after BCR stimulation was sufficient to support similar accumulation of IB-protein (24). The observed differences between transcript abundances and protein levels might thus be attributed to differences in translational regulation after BCR stimulation or TLR stimulation. To examine whether IB-was induced upon anti-CD40 stimulation, purified splenic B cells were stimulated either with LPS plus IL-4 or with anti-CD40 plus IL-4. As expected, IB-(85 kDa) was induced only after stimulation with LPS plus IL-4 (Fig. 3A). In addition, we found that a 90-kDa modified protein was induced after combined exposure to LPS and IL-4. Although the induction of this modified protein by LPS stimulation was reported previously (18), the nature of the modification remains poorly defined. Given that post-transcriptional regulation of IB-is activated by TLR/IL-1R but not by stimulation with TNF-␣ (18, 26), we compared post-transcriptional regulation of IB-after treatment either with LPS plus IL-4 or with anti-CD40 plus IL-4. Given our previous demonstration that transcriptional activity of the SV40 promoter was dispensable for LPS stimulation (26), we prepared SV40 promoter-driven reporters that expressed an mRNA that included a fusion of the coding sequence of luciferase to the 3Ј-UTR of the transcript that encodes IB-. The promoter activity was thus a reliable indicator of the post-transcriptional regulation of IBexpression. We found that luciferase activity of the IB-3Ј-UTR fusion reporter was activated only upon exposure to LPS plus IL-4 and not after stimulation by anti-CD40 plus IL-4 (Fig. 3B). Thus, these results indicated that the 3Ј-UTR-mediated post-transcriptional regulation of IBdefines LPS-specific, but not anti-CD40-mediated, induction of IBin B cells.
Deficiency of IB-Impairs TLR-mediated in Vitro Antibody Secretion and B Cell Proliferation-To establish the mechanistic basis of the defective TI-1 antibody responses in cKO mice, we examined whether purified IB--deficient B cells were also impaired in in vitro antibody production triggered by LPS stimulation in either the presence or absence of cytokine. After stimulation by exposure to various conditions, we measured levels of Igs secreted into the culture medium. This indicated that IB--deficient B cells secreted less IgM, IgG1, IgG2b, IgG3, and IgA than control B cells (Fig. 4A). This provided in vitro confirmation of the defect of TLR-mediated antibody  responses observed in cKO mice. Next, we examined whether the reduced antibody production could be attributed to changes in the proliferation of B cells.
We analyzed the effects of IB-deficiency on B cell proliferation by monitoring rates of incorporation of the vital dye CFSE. Rates of cell division in control B cells and cKO B cells were indistinguishable following stimulation with either BCR or CD40. However, after stimulation with LPS or CpG-DNA, the rate of division of cKO B cells was considerably less than that of control B cells (Fig. 4B). These results demonstrated that IBis required for B cell proliferation triggered by TLR stimulation.

IB-Is Essential for TLR-mediated Differentiation of B Cells into Plasma
Cells-To analyze whether IB-regulates the differentiation of plasma cells, purified splenic B cells from control or cKO mice were cultured for 3 days in the presence of either LPS alone, LPS plus IL-4, or anti-CD40 plus IL-4. The results indicated that cKO B cells expressed lower levels of the plasma cell marker CD138 than control B cells (Fig. 5, A and B). To clarify the molecular mechanism involved, we examined the RNA sequence and found the differences in the levels of Prdm1 (Coding for Blimp-1), a transcriptional factor required for the differentiation of B cells into plasma cells (Table 3) (30). This indicated that IB--deficient B cells failed to express Blimp-1 after stimulation with LPS (Fig. 5C). In addition, reduced levels of acetylation of histone H3 in the Blimp-1 promoter region in IB--deficient B cells suggested that they contained more active chromatin than unmodified B cells (Fig. 5D); however, this difference was not significant (p ϭ 0.1865). Thus, IBprobably controls the differentiation of B cells into plasma cells through inducing Blimp-1 expression.
IB-Is Essential for TLR-mediated CSR-To assess the effects of IBdeficiency on CSR, splenic B cells were stimulated either with LPS plus IL-4 or with anti-CD40 plus IL-4 (to induce switching to IgG1). After 3 days of stimulation by LPS plus IL-4, levels of surface IgG1 were lower in cKO B cells than in control B cells (Fig. 6A). However, following stimulation with anti-CD40 plus IL-4, levels of surface IgG1 were identical in cKO B cells and control B cells. The impairment of CSR observed in IB--deficient B cells could not be attributed to a change in the rate of their proliferation, because there were fewer IgG1-positive B cells in each cell division in populations of cKO B cells than in populations of control B cells (Fig. 6B).
Similarly, LPS induced fewer IgG3-positive cells when administered to cKO B cells than when administered to control B cells (Fig. 6, C and D).
To establish what impairs CSR in cKO B cells, we examined whether a reduced rate of CSR in cKO B cells resulted from reduced accumulation of germ line transcripts that encode the intervening heavy chain region and the constant heavy chain region (I H -C H ), which is necessary for CSR (31). Real time quantitative RT-PCR showed that, after stimulation for 3 days with LPS plus IL-4, the abundance of germ line transcripts that encode I␥1-C␥1 was similar in cKO B cells and control B cells. In contrast, post-recombination I -C ␥ 1 transcripts, which are generated by CSR, were significantly less abundant in cKO B cells than in control B cells (Fig. 6E). Co-engagement of BCR and TLR induces CSR through a noncanonical NF-B pathway (9). We examined whether IB-deficiency affects CSR triggered by simultaneous exposure to BCRs and TLRs. Stimulation of control B cells either with anti-IgD-dextran plus Pam3CSK4 (TLR2 ligand) or with CpG-DNA (TLR9 ligand) plus IL-4 caused a strong induction of CSR to IgG1. However, cKO B cells failed to induce CSR (Fig. 6F). Taken together, these results indicate that IBis essential for the induction of CSR through the co-engagement of BCR and TLR.
IB-Regulates CSR through AID Induction-We next clarified the molecular mechanisms of class switch recombination and found that expression of Aicda (coding for AID), the enzyme that induces DNA cleavage in the switch region of the Ig heavy chain locus (so-called CSR), was less in cKO B cells (Table 3) (10,14). Detection of Aicda mRNA by real time quantitative RT-PCR indicated that its abundance peaked within 48 -72 h after the stimulation of control B cells induced either by LPS plus IL-4 or by CD40 plus IL-4. However, AID expression in cKO B cells was less than 60% that in control B cells (Fig.  7, A and B). In addition, LPS failed to induce AID in cKO B cells (Fig. 7C). We thus investigated whether the defective CSR in IB--deficient B cells resulted from impaired expression of AID. To test this hypothesis, we used retroviral transfection to overexpress AID in cKO B cells and measured isotype switching in response to stimulation by LPS plus IL-4. Consistent with the results shown in Fig. 6A, rates of CSR were much lower in cKO B cells transduced with the control retrovirus than in control B cells (Fig. 7D). In contrast, overexpression of AID in cKO B cells  NOVEMBER 7, 2014 • VOLUME 289 • NUMBER 45 restored CSR, as shown in control B cells. In addition, retroviral transduction of BATF, which is a key regulator of AID expression (16), did not rescue CSR in cKO B cells. These results suggested that IBcontrols CSR by direct regulation of AID expression.

Control of Antibody Responses by IB-
To assess the role of IB-in the induction of AID, we examined whether overexpression of IB-affects the expression of a reporter gene placed under the control of the AID regulatory region. Four regions within the genomic Aicda locus are highly conserved among many species (10). These regions are called region 1 (positions Ϫ1500 to ϩ101), region 2 (positions ϩ121 to ϩ2221), region 3 (positions ϩ16,278 to ϩ18,378), and region 4 (positions Ϫ9224 to Ϫ7424). When cells were co-transfected with IBand each of the four reporters that contain an AID regulatory region, the region 1-containing AID reporter was most significantly activated in the presence of IB-in HEK293 cells (Fig. 7E). Additionally, the region 4-containing AID reporter was activated in the presence of IB-in HEK293 cells. To further confirm these findings, we used the B cell line CH12F3-2A and found that the region 1-containing AID reporter, but not the region 4-containing AID reporter, was significantly activated in the presence of IB-. Therefore, region 1 is more important than region 4 for AID gene expression in the response to IB- (Fig. 7F). It has been reported that IBcontrols NF-B target gene expression (21). In addition, the NF-B subunit p65 plays an important role in AID expression (32). We found that overexpression of the NF-B p65 subunit did not further elevate the activity of the region 1-containing AID reporter in the presence of IB-, indicating that NF-B may not have been involved in the effect of IB-on AID transcription (Fig. 7G).
We next analyzed the chromatin structure of the genomic Aicda locus in activated B cells. When B cells were activated, histone H3 in the conserved region of the genomic Aicda locus was highly acetylated; acetylation of histone H3 is a mark of active chromatin (33). ChIP analysis indicated that histone H3 in the vicinity of the transcriptional starting site (region 1) and ϩ0.2-kb area (region 2) was highly acetylated in control B cells but not in cKO B cells after stimulation with LPS plus IL-4 ( Fig.  7H). Taken together, these results suggested that IBregulates AID expression by controlling access to region 1 and modulating histone acetylation around the transcriptional starting site (region 1).

DISCUSSION
This study sought to analyze the role of IB-in B cell antibody responses by characterizing mice deficient in IB-, specifically in their B cells. In many cases, deficiency of transcriptional regulators impairs both TD and TI antigen responses in precedents (11-13, 15, 16). Here, we have used in vivo and in vitro assays to show that TLR-mediated TI-1, but not TD or TI-2, antibody responses are impaired in cKO mice. These defects were caused by reduced rates of B cell proliferation, differentiation of B cells into plasma cells, and B cell CSR. This TI-1-specific function of IBis assumed to result from TLR-specific induction of IB-. Induction of IB-requires threshold levels of transcriptional activation, Histone acetylation (AcH) enrichment was analyzed by a chromatin immunoprecipitation assay performed using antibody against acetyl-histone H3 (Lys-27). Data represent the mean Ϯ S.E. of triplicate samples and are representative of two independent experiments. **, p Ͻ 0.01. mRNA stabilization, and translational activation (18,34). Although TLR4 stimulation satisfies the criteria needed to induce IB-, stimulation with anti-CD40 antibody failed to initiate post-transcriptional activation of IB-. In addition, we have shown that BCR stimulation can stabilize IB-mRNA, although the increase in levels of IB-protein is less than that triggered by TLR stimulation. This might be caused by reduced rates of translational activation. Mechanistically, the TLR signal molecule MyD88 positively regulates IBprotein expression (19). However, MyD88-deficient B cells show increased IB-mRNA expression in response to LPS stimulation, to a level even higher than that of control B cells (data not shown). Thus, the TLR-MyD88 pathway may control the post-transcriptional regulation of IB-. Therefore, robust induction of IBby TLR might define the TI-1specific function of IB-.
In the case of CSR, impaired induction of AID contributed substantially to the impairment of CSR in cKO B cells, because levels of germ line transcripts for IgG1 were normal. In fact, retroviral transduction of AID rescued CSR in cKO B cells following stimulation by LPS plus IL-4. Reporter analysis indicated that overexpression of IBin HEK293 cells activated AID reporters that contained either region 1 or region 4. However, only the region 1-containing AID reporter was activated in CH12F3-2A cells. Given that CH12F3-2A is a B cell line, IBmight regulate AID expression by affecting region 1 in B cells. Consistent with this notion, levels of acetylation of histone H3 in region 1 and region 2, but not region 4, were lower in cKO B cells than in control B cells. Given that the expression of the region 2 reporter was not affected by IBoverexpression, the reduced rate of histone H3 acetylation in region 2 in cKO B cells might not be physiologically relevant. We previously demonstrated that TLR-mediated NF-B activation was comparable in control and IB--deficient B cells (24). Given that inhibition of histone deacetylase activity induces AID expression (35), histone acetylation in the genomic AICDA locus might promote AID expression. Furthermore, it has been shown that IBand histone deacetylase 5 are co-localized in the nucleus, suggesting that IBmay function by modulating histone deacetylase 5 activity (36). Taken together, our findings suggest that IB-would regulate chromatin structure to activate the expression of the gene that encodes AID. Given that TD antibody responses are independent of IB-, unidentified factors might control AID induction as a substitute for IB-in TD antibody responses.
IBforms a complex with NF-B and controls NF-B gene expression (21,37). A previous study has shown that IB-positively regulates IL-17A gene expression in combination with ROR␥t, which is dispensable for NF-B activation (20,38). Here, we show that IB-transcriptional activity the region 1-containing AID reporter is dispensable for NF-B transcriptional activity.
It is widely thought that TI antibody responses are not as important as TD antibody responses in protecting against infection. However, TI-1 responses are critical for preventing blood-borne infections from evolving into life-threatening conditions (39). In addition, TLR ligands are required for optimal antibody responses against Streptococcus pneumonia and after pneumococcal vaccination (40 -42). A human patient deficient in IRAK4 (a TLR signaling molecule) presented with a suppressed glycan-specific IgG antibody response after administration of an anti-pneumococcal glycan vaccine (43). Although this vaccine is broadly defined as a TI-2 antigen, it contains a TLR2 ligand and requires IRAK4 for the production of specific antibodies (44). Thus, both BCR and TLR signaling are required for a protective response to this vaccine. We have shown that IB-is required to induce BCR-and TLR2-dependent antibody responses in mice. In addition, IRAK4 is a key regulator of the activation of IB-both at the transcriptional and posttranscriptional levels (45). The induction of IB-in B cells by TLR signaling might play an important role in ensuring the efficacy of an anti-pneumococcal vaccine in humans. Given that IB--mediated antibody responses are independent of T cells, obtaining a better understanding of the IB--mediated antibody responses might contribute to the development of vaccines for patients with T cell deficiencies, such as individuals with acquired immune deficiency syndrome.