SEARCH

SEARCH BY CITATION

Keywords:

  • CpG-DNA;
  • HLA-DRA;
  • Lipopolysaccharide;
  • NF-κB;
  • Toll-like receptor

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements

Microbial components, such as DNA containing immunostimulatory CpG motifs (CpG-DNA) and lipopolysaccharides (LPS), elicit the cell surface expression of MHC class II (MHC-II) through Toll-like receptor (TLR)/IL-1R. Here, we show that CpG-DNA and LPS induce expression of the HLA-DRA in the human B cell line, RPMI 8226. Ectopic expression of the dominant negative mutant of CIITA and RNA interference targeting the CIITA gene indicate that CIITA activation is not enough for the maximal MHC-II expression induced by CpG-DNA and LPS. Additionally, nuclear factor (NF)-κB activation is required for the CpG-DNA-activated and LPS-activated HLA-DRA expression, whereas IFN-γ-induced MHC-II expression depends on CIITA rather than on NF-κB. Comprehensive mutant analyses, electrophoretic mobility shift assays and chromatin immunoprecipitation assays, reveal that the functional interaction of NF-κB with the promoter element is necessary for the TLR-mediated HLA-DRA induction by CpG-DNA and LPS. This novel mechanism provides the regulation of MHC-II gene expression with complexity and functional diversity.

Abbreviations:
ChIP:

chromatin immunoprecipitation

CpG-DNA:

DNA containing immunostimulatory CpG motifs

EMSA:

electrophoretic mobility shift assay

IκBαSR:

IκBα super repressor

IKK:

IκB kinase

MHC-II:

MHC class II

PAMP:

pathogen-associated molecule pattern

RFX:

regulatory factor binding to the X box

siRNA:

small interfering RNA

TRAF:

TNFR-associated factor

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements

The MHC class II (MHC-II) molecules are cell surface glycoproteins that are pivotal for inducing and regulating immune responses through their ability to present peptides derived from extracellular pathogens to CD4+ T lymphocytes. The expression of MHC-II genes is enhanced by inflammatory and immune stimuli, particularly cytokines such as IFN-γ, which stimulate the transcription of MHC-II genes 1, 2. MHC-II genes are expressed in professional APC such as DC, macrophages and B cells, and regulated tightly to ensure a specific response to foreign pathogens while minimizing the damage to host tissues. Such regulation is predominantly implemented at the transcription level by the master regulator, the CIITA, and includes mechanisms that involve the epigenetic modification and remodeling of chromatin 3, 4.

The MHC-II promoters share a common set of cis-acting elements and the set includes the W/S, X1, X2, and Y boxes, which are termed the S-X-Y modules 2, 5. The regulatory factor binding to the X box (RFX) complex, which comprises the RFX-5, RFXANK and RFX-AP components, binds to the X1 box of the MHC-II promoter, and the downstream X2 box is bound by the cyclic AMP response element-binding protein CREB 69. The Y box binds the heterotrimeric transcription factor, NF-Y, which consists of A, B and C subunits 10, 11. These transcription factors form a multiprotein complex, known as the MHC-II enhanceosome, which cooperatively binds the S-X-Y modules, and the multiprotein complex provides the appropriate interaction surface for recruiting the CIITA 12, 13.

Although constitutively expressed in the restricted APC, the MHC-II gene can be induced in most cell types following stimulation with IFN-γ 1, 3. The binding of IFN-γ to its cell-surface receptor activates the tyrosine and serine phosphorylation of components of the JAK-STAT pathway 14. The activated STAT1 then binds to the IFN-γ-activated sequence element of the IFN-γ-responsive promoter, leading to the expression of genes such as CIITA 15, 16. Upon IFN-γ induction, the CIITA is produced, and induces or enhances the association of transcription factors with the promoter to form enhanceosome 12, 13. The CIITA interacts with numerous cofactors, including chromatin-modifying enzymes and components of the basal transcription machinery 1719.

In addition to IFN-γ, pathogen-associated molecule patterns (PAMP), including LPS and bacterial DNA, are capable of inducing MHC-II gene expression 20, 21. LPS is the major component of the outer surface of Gram-negative bacteria and a typical example of potent immune activator. Bacterial DNA and synthetic oligodeoxynucleotides, which contain unmethylated CpG dinucleotides in the context of particular base sequences (CpG-DNA), are also recognized by innate immune cells 22. Immune activation by CpG-DNA depends on TLR9, which shares a high degree of homology with TLR4 23, 24. Furthermore, CpG-DNA and LPS can induce proliferation of B cells and enhance the surface expression of MHC-II molecules 20, 25.

Previously, we reported that phosphorothioate backbone-modified CpG-DNA is a powerful mediator of HLA-DRA gene expression in the human B cell line RPMI 8226 21. However, the mechanism by which the TLR/IL-1R pathway controls the MHC-II gene expression has not been clearly examined. Here, we show that the CpG-DNA-induced and LPS-induced HLA-DRA expression in RPMI 8226 cells differ from the IFN-γ-activated pathway. Experimental evidence shows that NF-κB is another key transcription factor in the expression of the HLA-DRA gene, and that NF-κB is directly involved in TLR/IL-1R-mediated MHC-II gene expression in human B cells.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements

Expression of HLA-DRA in CpG-DNA-stimulated and LPS-stimulated RPMI 8226 cells

First, we determined induction of the endogenous HLA-DRA gene expression in CpG-DNA-treated and LPS-treated human RPMI 8226 B cell line. To examine CpG-DNA-induced and LPS-induced HLA-DRA mRNA expression, we used an RT-PCR and quantitative real-time PCR assays. As shown in Fig. 1A and B, the HLA-DRA gene is endogenously expressed in RPMI 8226 B cells, and CpG-DNA 21 and LPS each greatly enhance the gene expression in a time-dependent manner. We also examined the HLA-DRA mRNA expression in IFN-γ-treated cells. The expression level in IFN-γ-treated RPMI 8226 cells is lower than the expression levels in CpG-DNA-activated cells and LPS-activated cells. In addition, the TLR4 and TLR9 genes are constitutively expressed in the RPMI 8226 cells regardless of the cell stimulations (Fig. 1A).

thumbnail image

Figure 1. CpG-DNA-induced and LPS-induced HLA-DRA gene expression in RPMI 8226 B cells. (A) The cells were treated with IFN-γ, CpG-DNA, or LPS for the indicated periods, and the total RNA was then extracted. After RT, the expressions of HLA-DRA, human TLR4, and TLR9 genes were analyzed by PCR using the primer sets shown in the Materials and methods. The expression level of human actin mRNA was used as an internal control. M, DNA standard marker. (B) Data were also presented as a relative mRNA expression to the unstimulated control by quantitative real-time PCR. (C and D) Cells were transiently transfected with pHLA-DRA-Luc for 24 h, and stimulated with an increasing amount of CpG-DNA for 24 h (C), and with IFN-γ, CpG-DNA, or LPS for different periods (D). Cultures were harvested and assayed for luciferase activity. The luciferase activity, which was measured as a relative light unit, was normalized to Renilla activity.

Download figure to PowerPoint

Next, we also confirmed the activation of the HLA-DRA promoter, which uses the promoter-reporter construct containing 366 bp 5′ of the transcription start site linked to a luciferase gene. This promoter-reporter construct was transiently transfected into RPMI 8226 B cells, which were treated with CpG-DNA, LPS, or IFN-γ. As shown in Fig. 1C and D, we found that CpG-DNA could activate the HLA-DRA promoter in a dose- and time-dependent manner. Moreover, 24 h after treating the cells with CpG-DNA or LPS, we observed maximal promoter activation, and the expression persisted for 48 h after the stimulation (data not shown). As with the mRNA expression shown in Fig. 1A and B, the CpG-DNA and LPS were more potent than IFN-γ in influencing the HLA-DRA promoter activity (Fig. 1D).

CIITA-dependent and CIITA-independent induction of HLA-DRA expression in TLR-triggered RPMI 8226 cells

To investigate how CpG-DNA and LPS induce HLA-DRA expression, we examined whether TLR-mediated stimulation affects the expression of the CIITA, which is known to be a main transcriptional coactivator essential for the constitutive and IFN-γ-induced expression of MHC-II 1. After stimulation of the RPMI 8226 cells with IFN-γ, the mRNA expression of CIITA increased in a time-dependent manner (Fig. 2A), and in a pattern similar to that of the HLA-DRA expression (Fig. 1A). However, the CIITA mRNA was barely induced when the cells were treated with CpG-DNA (Fig. 2A and E) or LPS (Fig. 2E). Furthermore, we were not able to detect the induction of CIITA protein in the cells stimulated with CpG-DNA in contrast to the cells treated with IFN-γ (Fig. 2B). We also defined whether CpG-DNA and LPS influence the CIITA promoter activity using the luciferase-reporter construct. In contrast to the results of the IFN-γ stimulation, the promoter activity of the CIITA gene was not affected in CpG-DNA-treated and LPS-treated RPMI 8226 cells (Fig. 2C).

thumbnail image

Figure 2. CIITA-independent HLA-DRA expression induced by CpG-DNA and LPS. (A and B) RPMI 8226 cells were treated with IFN-γ or CpG-DNA for the indicated periods, and the expression of human CIITA was analyzed by RT-PCR and Western blotting. (C) The cells were transiently transfected with phCIITA-Luc for 24 h, and then stimulated with IFN-γ, CpG-DNA, or LPS for the indicated periods. Cultures were harvested and assayed for luciferase activity. (D) The expression plasmid encoding human CIITA (hCIITA) or dominant negative version of the CIITA (ΔCIITA) was transiently cotransfected with pHLA-DRA-Luc into the cells, and then stimulated with IFN-γ, CpG-DNA, or LPS for 24 h. The luciferase activity was measured. The results are represented as a fold activation compared with the control vector alone. (E and F) Cells were transfected with an siRNA duplex for CIITA gene for 24 h, and then stimulated with IFN-γ, CpG-DNA, or LPS for 12 h. The cells were harvested and the expressions of HLA-DRA and CIITA genes were analyzed by RT-PCR and by Western blotting. (G) The siRNA duplex for CIITA gene was transiently cotransfected with pHLA-DRA-Luc into the cells, and then stimulated with IFN-γ, CpG-DNA, or LPS for 24 h. Cultures were harvested and assayed for luciferase activity. Negative control siRNA duplex (Ambion) was used as an internal control.

Download figure to PowerPoint

To examine whether the CIITA is involved in the TLR-activated MHC-II expression, we used an expression vector encoded with a CIITA or with the CIITA dominant negative version, ΔCIITA, to cotransfect the HLA-DRA promoter-reporter construct into the cells 26. As shown in Fig. 2D, the ectopic expression of the CIITA significantly enhanced the HLA-DRA promoter activity. Following the cotransfection of the ΔCIITA with the reporter construct, the luciferase activity was reduced to a background level in the cells stimulated with IFN-γ. On the other hand, the ectopic expression of the ΔCIITA partially decreased the CpG-DNA-induced and LPS-induced promoter activation by about 50% (Fig. 2D). Nonetheless, the CpG-DNA-induced and LPS-induced promoter activation in the presence of the ΔCIITA were still more than four times greater than the unstimulated control. These results suggest that there is a CIITA-independent pathway in addition to a CIITA-dependent pathway.

To further evaluate the CIITA-independent pathway of TLR-mediated HLA-DRA gene expression, we used a small interfering RNA (siRNA) duplex targeting the human CIITA gene. As CIITA protein was hardly detected by Western blotting, we presume that the basal level of CIITA is very low. When the RPMI 8226 cells were transfected with the CIITA siRNA duplex, IFN-γ failed to induce HLA-DRA mRNA expression as well as CIITA induction (Fig. 2E and F). However, the CIITA siRNA transfection inhibited but did not abolish CpG-DNA- and LPS-induced expression of HLA-DRA gene (Fig. 2E). These results were also confirmed by the luciferase assay, in which the CIITA siRNA duplex was cotransfected with HLA-DRA promoter-reporter construct (Fig. 2G). As with the luciferase activity shown in Fig. 2D, the cotransfection of the CIITA siRNA duplex abolished IFN-γ−induced HLA promoter activation, but partially decreased the CpG-DNA-induced and LPS-induced promoter activation. Therefore, it is likely that CIITA contributes only partially to MHC-II expression induced by CpG-DNA and LPS, that a critical regulatory factor other than CIITA enables the TLR-induced MHC-II expression to achieve its maximum activity. Interesting point is that CIITA contributes to HLA induction even though LPS or CpG-DNA stimulation does not induce CIITA. We believe that basal level of CIITA can contribute to HLA induction. Whether the effect is direct or indirect is to be determined. These results also suggest a possibility that combined stimulation with IFN-γ and either LPS or CpG-DNA enhance the MHC-II expression compared with stimulation with each alone. As shown in Fig. 3, the combined stimulation with IFN-γ and CpG-DNA or LPS did show at least an additive effect on activation of the HLA-DRA promoter.

thumbnail image

Figure 3. Enhanced activation of HLA-DRA promoter by combined stimulation with IFN-γ plus either CpG-DNA or LPS. Cells were transiently transfected with pHLA-DRA-Luc for 24 h and stimulated with IFN-γ, CpG-DNA, or LPS alone or combination of IFN-γ with CpG-DNA or LPS for 24 h. Cultures were harvested and assayed for luciferase activity.

Download figure to PowerPoint

Requirement of NF-κB for the HLA-DRA promoter activation induced by CpG-DNA and LPS

The promoters of MHC-II contain three conserved sets of cis-acting elements. These sets, termed the S-X-Y module, bind a multiprotein complex known as the MHC-II enhanceosome, and they provide the appropriate interactive surface for recruiting CIITA (Fig. 4A) 18. To identify other DNA elements necessary for CpG-DNA-induced and LPS-induced MHC-II promoter activation, we performed a computer-assisted analysis of the HLA-DRA gene promoter. In addition to the S-X-Y module, we observed an element similar to a recognition site for the NF-κB protein in the promoter region between –301 and –292 of the HLA-DRA gene (Fig. 4A and B). We also noticed that the NF-κB motifs are conserved in MHC-II and HLA-DM genes (Fig. 4B).

thumbnail image

Figure 4. Requirement of NF-κB for the HLA-DRA expression induced by CpG-DNA and LPS. (A) Sequence and putative transcription factor binding elements of the HLA-DRA gene promoter. (B) Sequence alignment of the conserved NF-κB-binding elements present in MHC-II promoters and HLA-DM genes. (C) The structures of luciferase reporter plasmids are shown. The mutated NF-κB binding site is indicated by a cross. The RPMI 8226 cells were transfected with each reporter construct, and then cultured in the presence of IFN-γ, CpG-DNA, or LPS for 24 h. The luciferase activity was measured; the fold activation represents the ratio of the luciferase activity in stimulated and unstimulated cells. (D) The RPMI 8226 cells were cotransfected with pHLA-DRA-Luc and expression plasmid encoding either NF-κB p65 or IκBαSR, and then stimulated with IFN-γ, CpG-DNA, or LPS for 24 h. The luciferase activity was measured. The results are represented as fold activation compared with control vector alone. (E) The cells were pretreated with 50 μM of BMS-345541 for 1 h, and then cultured in the presence of IFN-γ, CpG-DNA, or LPS for 12 h. The expression of HLA-DRA mRNA and CIITA mRNA was analyzed by RT-PCR.

Download figure to PowerPoint

To evaluate the possibility that the NF-κB element contributes to the activity of the HLA-DRA promoter, we introduced a mutation of the NF-κB-binding element into the promoter-reporter construct. As shown in Fig. 4C, CpG-DNA- and LPS-induced HLA-DRA promoter activity decreased by about 40% when we introduced a mutation of the NF-κB element into the construct. However, the mutation of the NF-κB element did not affect the HLA-DRA promoter activity in the cells stimulated with IFN-γ. These results suggest that the NF-κB-binding element in the TLR-stimulated RPMI 8226 cells regulate the HLA-DRA promoter activation.

To further confirm the contribution of NF-κB in the CpG-DNA- and LPS-induced HLA-DRA expression, we used ectopic expression of NF-κB p65 and IκBα super repressor (IκBαSR), a nondegrading mutant of IκB. When the HLA-DRA promoter-reporter construct was cotransfected with the expression plasmid encoded with NF-κB p65, the luciferase activity increased to a level comparable to the levels induced by CpG-DNA and LPS (Fig. 4D). IκBαSR significantly inhibits the CpG-DNA- and LPS-induced HLA-DRA promoter activation, though it does not affect the IFN-γ-induced luciferase activity (Fig. 4D). To substantiate the requirement for NF-κB activation in the transcriptional up-regulation of HLA-DRA following TLR triggering by CpG-DNA and LPS, we tested the effect of BMS-345541, an IκB kinase-2 (IKK-2) inhibitor 27 on the expression of HLA-DRA mRNA. Fig. 4E shows that the CpG-DNA-induced and LPS-induced expression of HLA-DRA was markedly reduced when the cells were pretreated with BMS-345541, but IFN-γ induced the gene expression independently of BMS-345541. These results support that, in contrast to the response of IFN-γ, NF-κB is essential for the HLA-DRA expression induced by CpG-DNA and LPS.

Binding of NF-κB to the HLA-DRA promoter by the stimulation of CpG-DNA and LPS

To determine whether NF-κB directly binds to the HLA-DRA promoter after signal-induced nuclear accumulation in CpG-DNA-stimulated and LPS-stimulated cells, we performed electrophoretic mobility shift assay (EMSA) experiments with a consensus NF-κB binding sequence as a probe. We observed a protein-DNA complex in the nuclear extracts from CpG-DNA-treated and LPS-treated cells, whereas the cell stimulated with IFN-γ had no effect (Fig. 5A). To analyze the specificity of the protein-DNA complex, we conducted competition experiments with the HLA-DRA promoter sequence –323 to –282, which contained either NF-κB-binding element or its point mutation sequence. The results show that the protein-DNA complex was not formed in the presence of an excess of an unlabeled HLA-DRA promoter sequence, and that the addition of the point mutation sequence had no competition effect (Fig. 5B). For further analysis, we conducted a supershift assay with the aid of an antibody specific to NF-κB p65, and we found that a new supershift band appeared with a concurrently decreased the protein-DNA complex in the presence of the antibody (Fig. 5B).

thumbnail image

Figure 5. Direct interaction of NF-κB with the HLA-DRA promoter. (A and B) EMSA was performed using a consensus NF-κB oligonucleotide as probe with nuclear extracts from the cells treated with INF-γ, CpG-DNA, or LPS. The DNA-protein complexes were analyzed by competition assay with a cold oligonucleotide for HLA-DRA NF-κB or its mutant for the NF-κB binding site as a competitor. (C) Nuclear extracts from the cells treated with INF-γ, CpG-DNA, or LPS were also analyzed by the EMSA using an HLA-DRA NF-κB oligonucleotide as a probe. For the supershift assay, an antibody to NF-κB p65, p50, c-Rel, or p52 was included in the reaction mixture. The DNA-protein complexes are denoted by arrows and the supershift complexes are denoted by arrowheads. (D) Cross-linked chromatin isolated from IFN-γ, CpG-DNA, or LPS-treated RPMI 8226 cells was immunoprecipitated with antibody to NF-κB p65. Immunoprecipitates were analyzed by PCR for the presence of promoter sequences of the MHC-II genes HLA-DRA, HLA-DRB5, and HLA-DPA using the primer pairs described in the Materials and methods (Bound). The DNA purified form the sonicated chromatin was directly analyzed by PCR using the same primer sets, which was used as an input control (Input).

Download figure to PowerPoint

We then used a probe containing HLA-DRA promoter sequence –323 to –282 to analyze whether NF-κB actually binds to the binding element of the HLA-DRA promoter in the TLR-triggered RPMI 8226 cells. The results of this analysis are consistent with the EMSA results, for which we used the consensus NF-κB binding sequence as a probe (Fig. 5A and C). We found, in particular, that the protein-DNA complex is specifically disrupted by the addition of antibody against NF-κB p65, p50 and c-Rel (Fig. 5C, complex 1), and that the antibody against NF-κB p52 induces the supershift complex with a concurrently decreased the protein-DNA complex (Fig. 5C, complex 2). These results suggest that NF-κB directly binds to the NF-κB binding site of the HLA-DRA promoter when CpG-DNA and LPS, but not IFN-γ, activate the RPMI 8226 cells.

To confirm the interactions of NF-κB with MHC-II promoters in vivo, we performed chromatin immunoprecipitation (ChIP) assay. Formaldehyde cross-linked chromatin fragments were isolated from RPMI 8226 cells, immunoprecipitated with the antibody to NF-κB p65, and analyzed by PCR for the presence of specific DNA sequences corresponding to putative NF-κB-binding sites in MHC-II promoters (Fig. 4B). The NF-κB binding sites in promoters of three MHC-II genes (HLA-DRA, HLA-DRB5, and HLA-DPA) and W-X-Y module in HLA-DRA promoter were analyzed. As expected, all of the NF-κB binding regions in the three MHC-II promoters were immunoprecipitated with anti-NF-κB p65 Ab when the cells were treated with CpG-DNA or LPS (Fig. 5D). However, we did not detect the PCR products of NF-κB binding sequences in the MHC-II promoters when the cells were treated with IFN-γ (Fig. 5D). We also did not detect the PCR products of W-X-Y module in HLA-DRA promoter from all of the chromatin fragments immunoprecipitated with the antibody to NF-κB p65.

NF-κB-dependent expression of the HLA-DRA gene in the primary human B cells activated by CpG-DNA

In order to investigate whether the requirement of NF-κB activation in the HLA-DRA expression is relevant in vivo, we used primary B cells purified from human blood. Human B cells were isolated by depletion of non-B cells, and the cells isolated by this method were routinely 95–98% CD19+ (data not shown). The ability of CpG-DNA to up-regulate the surface expression of HLA-DR on the primary human B cells is shown in Fig. 6A. As with the results of HLA-DRA mRNA expression in RPMI 8226 cells (Fig. 1B), HLA-DR expression in the primary B cells was induced more potently by CpG-DNA than by IFN-γ. We also examined the HLA-DRA mRNA expression in the primary human B cells. The expression level was much higher in the CpG-DNA-treated cells than in the IFN-γ-treated cells (Fig. 6B and C). Furthermore, surface expression as well as the mRNA expression of HLA-DRA after stimulation with CpG-DNA decreased markedly by the pretreatment with an IKK-2 inhibitor BMS-345541 (Fig. 6). Based on these results, it is evident that NF-κB activation is necessary to induce the HLA-DRA expression triggered by CpG-DNA and LPS in B cells.

thumbnail image

Figure 6. NF-κB-dependent expression of the HLA-DRA gene in the primary human B cells. (A) Purified human B cells were cultured with medium only (open plots) or stimulated for 48 h with of IFN-γ (200 U/mL) or CpG-DNA (3 μM) (filled plots) in the absence or presence of 50 μM of BMS-345541. The cells were stained with an FITC-conjugated antibody to HLA-DR and analyzed with a FACScan flow cytometer. FITC-conjugated mouse IgG2a κ was used as an isotype control (dashed plots). (B and C) The cells were pretreated with DMSO control or 50 μM of BMS-345541 for 1 h, and then cultured in the presence of IFN-γ or CpG-DNA for 12 h. The expression of HLA-DRA mRNA and CIITA mRNA was analyzed by RT-PCR (B) and HLA-DRA mRNA was analyzed by quantitative real-time PCR (C).

Download figure to PowerPoint

MyD88-dependent and TRAF6-dependent activation of the HLA-DRA promoter by CpG-DNA and LPS

MyD88 and TNFR-associated factor 6 (TRAF6) are necessary in TLR-mediated intracellular signaling, and the signaling causes activation of the common transcription factors NF-κB and AP-1 21, 28, 29. We therefore used an expression plasmid that encodes a dominant negative version of MyD88 (ΔMyD88) to investigate the involvement of MyD88 in CpG-DNA-induced or LPS-induced HLA-DRA gene expression. Cotransfection of ΔMyD88 with the promoter-reporter construct markedly inhibits the HLA-DRA promoter activation in RPMI 8226 cells treated with CpG-DNA and LPS (Fig. 7A and B).

thumbnail image

Figure 7. Inhibition of the TLR-mediated HLA-DRA promoter activation by ΔMyD88 and ΔTRAF6. The cells were transiently cotransfected with pHLA-DRA-Luc and the expression plasmid encoding dominant negative version of MyD88 (ΔMyD88), TRAF2 (ΔTRAF2), or TRAF6 (ΔTRAF6). After the transfection, the cultures were treated with CpG-DNA (A), LPS (B), or IFN-γ (C) for 24 h, and then the luciferase activity was measured. The results are represented as a fold activation compared with the control vector alone.

Download figure to PowerPoint

We also attempted to determine the requirements of TRAF6 in the HLA-DRA promoter activation induced by CpG-DNA and LPS. As shown in Fig. 7A and B, ΔTRAF6, a dominant negative version of TRAF6, significantly reduces the promoter activation that is stimulated by CpG-DNA and LPS. However, ΔTRAF2, a dominant negative version of TRAF2, does not affect TLR-triggered HLA-DRA promoter activation in the RPMI 8226 cells. These data indicate that the signal transduction molecules MyD88 and TRAF6 in the TLR/IL-1R pathway are commonly required for the HLA-DRA promoter activation in a human B cell line treated with CpG-DNA and LPS. In contrast, cotransfection of RPMI 8226 cells with ΔMyD88, ΔTRAF2, or ΔTRAF6 did not affect the IFN-γ-induced HLA-DRA promoter activation (Fig. 7C).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements

Microbial components transmit a signal through their cognate TLR, cell surface receptors which have a high degree of homology with Drosophila Toll 30. The exposure of APC to PAMP such as CpG-DNA or LPS results in activation of the MyD88/IL-1R-associated kinase (IRAK) pathway, which is thought to be a downstream effector of TLR9 and TLR4. Activation of the MyD88/IRAK pathway stimulates the transcription factor NF-κB, and this transcription factor in turn enhances the transcriptional up-regulation of genes downstream from the κB motifs, which include cytokines, cell adhesion molecules, and cell surface molecules 28, 29, 31. The TLR stimulation by CpG-DNA or LPS also increases the MHC-II expression more potently than the IFN-γ stimulation (Fig. 1) 20, 25. It is therefore reasonable to postulate that the expression of MHC-II molecules may also be controlled by PAMP via activation of the transcriptional factor NF-κB. However, the validity of this postulation has not been examined.

Here, we suggest that the HLA-DRA expressions by CpG-DNA and LPS differ from the gene expression induced by IFN-γ. The constitutive and inducible MHC-II expression is quantitatively regulated by the essential transcriptional coactivator CIITA, and therefore depends on the control of CIITA expression 3, 32. As reported previously 16, 33, activation of the CIITA is necessary for the HLA-DRA expression by IFN-γ stimulation through the STAT1-mediated pathway in the human B cells. However, there is no evidence that CpG-DNA or LPS up-regulates the expression of the CIITA. Importantly, the CpG-DNA- and LPS-induced expression of HLA-DRA shows interesting features in this report. The CIITA activation is required for maximal activation; however, about 50% of the maximal activity is seen independently of CIITA (Fig. 2). Even in the ectopic expression of the ΔCIITA, the ability of CpG-DNA and LPS to induce the promoter activation is comparable to the activity induced by IFN-γ. Moreover, the RNA interference targeting the CIITA gene did not abolish the HLA-DRA expression induced by CpG-DNA and LPS in contrast to the mRNA expression induced by IFN-γ (Fig. 2E and G).

Recently, it has been reported that CpG-DNA enhances the MHC-II protein expression because of the increased stability of peptide-MHC-II complexes in the maturation of DC 34. In a contrasting view of the regulation in DC, Kurchtey et al.35 suggested that a CpG-DNA-induced increase in MHC-II expression by B lymphocytes is not due to protein stabilization, but to the increased stability of MHC-II mRNA. Based on our study here, we propose another transcriptional regulation through which CpG-DNA and LPS can enhance the HLA-DRA expression. NF-κB, in addition to the CIITA, is an essential factor for the CpG-DNA-induced and LPS-induced expression of HLA-DRA (Fig. 4).

By using promoter-reporter assay, we provide evidence that NF-κB directly induces the HLA-DRA promoter activation in response to CpG-DNA and LPS. When we analyzed the sequence of the HLA-DRA promoter region, an NF-κB-binding element appeared (Fig. 4A and B). Furthermore, mutation of the NF-κB-binding site in the promoter region reduced the CpG-DNA- and LPS-induced promoter activity (Fig. 4C). We also found that the expression of HLA-DRA in the TLR-stimulated B cells was tightly regulated via the NF-κB activation by using IκBαSR, a nondegrading mutant of IκBα and by using BMS-345541, an IKK-2 inhibitor (Fig. 4 and 6). Furthermore, direct binding of NF-κB to the NF-κB binding site of HLA-DRA promoter was detected by the EMSA and ChIP assays (Fig. 5). These results indicate that the NF-κB binding element and its interaction with NF-κB are essential for the CpG-DNA-induced and LPS-induced transcriptional activity of the HLA-DRA promoter (Fig. 8).

thumbnail image

Figure 8. A model for HLA-DRA expression induced by TLR triggering.

Download figure to PowerPoint

A necessary and major role is played by INF-γ in priming APC to defend certain microbes in both innate and adaptive immune responses. It is well known that IFN-γ interacts with a heterodimeric type II cytokine receptor, resulting in activation of JAK 1 and 2 as well as Y701 phosphorylation of STAT1 15. Tyrosine-phosphorylated STAT1 in turn dimerizes and translocates into the nucleus, where it initiates gene transcription. When we treated RPMI 8226 cells with either CpG-DNA or LPS, we observed no rapid tyrosine phosphorylation of STAT1. However, as reported elsewhere, prolonged stimulation causes the phosphorylation of STAT1 after the TLR triggering (data not shown) 36. When combined with the data from the inhibitory effect on the CIITA expression (Fig. 2), these results imply that the TLR-induced HLA-DRA expression requires basal activation of the CIITA for maximal activation even though CIITA expression is not induced by TLR stimulation. As a possible mechanism, the TLR/IL-1R signaling may generate intracellular cross-talk between the signaling pathways that regulates the MHC-II expression (Fig. 8). The concerted action of NF-κB and CIITA may result in the maximal response to TLR stimulation.

The induction of MHC-II expression is extremely complex and requires the coordinate engagement of several distinct trans-acting factors to form an enhanceosome 12, 13. An alternative mechanism for MHC-II gene expression by TLR triggering reflects the fact that activation of innate immunity is a prerequisite for the optimal induction of adaptive immune responses. In this respect, the microenvironment can modulate the cellular responses toward protective cytokines and surface molecules. According to Pai et al. 37, Mycobacterium tuberculosis and its 19-kDa lipoprotein inhibit IFN-γ induction of mRNA for CIITA, IFN-regulatory factor-1, and MHC-II by a TLR2-dependent mechanism. This may suggest a potential mechanism for immune evasion by M. tuberculosis. Here, we show that selective activation of NF-κB by CpG-DNA and LPS could result in accelerated up-regulation of MHC-II expression independently of IFN-γ signaling pathway activation. Several lines of our experimental evidence suggest there is a strong possibility that a finely tuned combination of CpG-DNA and IFN-γ effectively triggers pathogen-specific adaptive immune responses via regulation of the MHC-II expression. At least an additive effect after combined stimulation with IFN-γ and either LPS or CpG-DNA strongly supports this idea (Fig. 3). Future studies will provide further insight into this aspect of MHC-II promoter activity.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements

Cell culture and reagents

We obtained the human B cell line RPMI 8226 from the American Type Culture Collection (Manassas, VA), and maintained it in an RPMI 1640 medium with 10% fetal bovine serum. We used trypan blue dye exclusion to assay the viability, which was typically greater than 95%. Phosphorothioate backbone-modified oligodeoxynucleotide was purchased from GenoTech (Daejeon, Korea). The CpG-DNA 2006(S) consisted of 24 bases that contained three CpG motifs (underlined): TCGTCGTTTTGTCGTTTTGTCGTT. The Escherichia coli LPS (Sigma, St. Louis, MO) was suspended in sterile water and added to the cells. We obtained the human IFN-γ from Roche (Indianapolis, IN), and purchased BMS-345541, an IKK-2 inhibitor from Calbiochem (Darmstadt, Germany). We purchased the anti-NF-κB p65 Ab and anti-NF-κB p52 Ab from Delta Biolabs (Campbell, CA), and the anti-NF-κB p50 Ab and anti-c-Rel Ab from Santa Cruz Biotechnology (Santa Cruz, CA).

RT-PCR analysis

After treating the cells with IFN-γ (200 U/mL), CpG-DNA (3 μM) or LPS (100 ng/mL) for the indicated periods, we extracted the total RNA with a MicroRNA Isolation Kit (Stratagene, La Jolla, CA) in accordance with the manufacturer's instructions. We then reverse-transcribed 7 μg of total RNA in the first-strand synthesis buffer containing 6 μg/mL of oligo(dT) primer, 50 U of StrataScriptTM reverse transcriptase, 4 mM of dNTP, and 40 U of RNase inhibitor. The primer pairs for HLA-DRA, CIITA, TLR4, TLR9, and actin were as follows: HLA-DRA, 5′-CGAGTTCTATCTGAATCCTG-3′ and 5′-GTTCTGCTGCATTGCTTTTGC-3′; CIITA, 5′-AAGGTGGCTACCTGGAGCTT-3′ and 5′-GTCCTTGCTCAGGCCCTC-3′; TLR4, 5′-ACCTCCCCTTCTCAACCAAGAAC-3′ and 5′-GATTGTGAGCCACATTAAGTTCTTTC-3′; TLR9, 5′-TGCCTGGCAAAACCCTCTTTGAGA-3′ and 5′-TCGTAGCGGGAGCGGCGGCC-3′; actin, 5′-CCGACACAGACACCATCA AC-3′ and 5′-CCTTAATGTCACGCACGATTT-3′. For the quantitative analysis of HLA-DRA expression, real-time PCR analysis was performed using CYBR Green Master Mix and run in the ABI 7000 machine (Applied Biosystems, Foster city, CA). The method uses the following primer pair for HLA-DRA: 5′-TGGACAAAGCCAACCTGGAAA-3′ and 5′-AGGACGTTGGGCTCTCTCAG-3′. The mRNA levels were normalized using GAPDH as an internal control, and relative expression was determined by dividing all normalized values within a data set by the normalized arbitrary units of the unstimulated control.

Construction of luciferase reporter plasmids

To amplify the HLA-DRA promoter that contained –323 to +43, we used PCR with an aid of human genomic DNA (Clontech, Palo Alto, CA) as a template with the following primer set: DRA(–323), 5′-GGTACCATCTGTATTTAGCTCTCACTTTAG-3′ and DRA(+43), 5′-AAGCTTGGGAGTGAGGCAGAACAGAC-3′. We then ligated this fragment into KpnI and HindIII sites of the luciferase reporter plasmid pGL3-Basic vector (Promega, Madison, WI), which yielded the reporter construct pHLA-DRA-Luc. To introduce a site-specific mutation in NF-κB-binding site, we abrogated the transcription factor recognition site and replaced with non-functional sequence by two-step PCR mutagenesis method 38. The method uses 5′-primer DRA(–323) and 3′-primer DRA(+43), along with primers that encode the following sequences in sense or antisense orientation: mNF-κB, –312GCTCTCACTTTAccTGTTTggATTGATTCTATTC-279. The mutated sites are indicated with lowercase letters. We cloned the promoter sequence, including the mutation, into the pGL3-Basic vector, thereby yielding the reporter construct pHLA-DRA(mκB)-Luc. With the aid of PCR, we amplified the human CIITA promoter fragment –354 to +44 by using human genomic DNA (Clontech) as a template with the following primer set: 5′-CGGTACCACTGAGTTGGAGAGAAA-3′ and 5′-GAAGCTTCTCCCTCCCGCCAGCTC-3′. This fragment, which we cloned into KpnI and HindIII sites of the pGL3-Basic vector, yielded the luciferase reporter construct phCIITA-Luc.

Construction of expression plasmids

With the aid of PCR, we amplified the human CIITA cDNA by using the human fetal liver cDNA (Clontech) as a template with the following primer sets: 5′-primer, 5′-CGGAATTCCTACACAATGCGTTGCCTG-3′ and 3′- primer, 5′-CGGCTCGAGCACAGCTGGGATCATCTCAG-3′. The PCR product, which we cloned into EcoRI and XhoI sites of the expression plasmid pcDNA 3.0 (Invitrogen, Carlsbad, CA), yielded the expression construct phCIITA. The pΔCIITA is the mutant construct that deleted a GTP-binding site (451–460 of the amino acid sequence) of human CIITA protein 26. To delete the GTP-binding site, the two primer sets used for PCR: 5′-primer, 5′-GAATCGATTGCTGTGCTGGGCAA-3′ and 3′-primer, 5′-CCATCGATGCAGCCAACAGCACCTCAG-3′. The resulting fragment, which was digested with ClaI, ligated, and cloned into a pcDNA 3.0 vector, yielded the expression construct pΔCIITA. The following expression plasmids were used for the transfection 21: NF-κB p65, IκBαSR, ΔMyD88, ΔTRAF2, and ΔTRAF6.

Transfection and luciferase assay

A day before transfection, we placed the RPMI 8226 cells into 12-well plates at the concentration of 2 × 105 cells/well. The cells were used for transfection with FuGene 6 Transfection Reagent (Roche) in accordance with the manufacturer's instructions. We confirmed the equivalent transfection efficiency by cotransfecting the promoterless Renilla luciferase vector pRL-null (Promega) 39. After placing the transfected cells in a complete medium for 24 h, we added IFN-γ (200 U/mL), CpG-DNA (3 μM) or LPS (100 ng/mL) for the indicated periods. We then harvested the cells, and measured the luciferase activity by using the Dual-Luciferase Reporter Assay System (Promega) with a TD-20/20 luminometer (Tuner Designs, Sunnyvale, CA) in accordance with the manufacturer's specifications.

Western blotting

We performed SDS-PAGE and Western blot analysis as described elsewhere 38. After treating the cells with IFN-γ (200 U/mL), CpG-DNA (3 μM) or LPS (100 ng/mL) for the indicated periods, we harvested the cells and lysed them in 10 mM HEPES, pH 7.4, 150 mM NaCl, 5 mM EDTA, 1% Triton X-100, and Complete Protease Inhibitor Cocktail (Roche). We purchase the anti-CIITA Ab from Rockland (Gilbertsville, PA) and anti-actin Ab form Sigma.

RNA interference experiments

Pre-designed siRNA duplex targeted for human CIITA was purchased from Ambion (Austin, TX). A 21-nucleotide siRNA duplex with dTdT overhangs of the target CIITA mRNA coding sequence (5′-CCACTTTTCATAACCTAGCTT-3′) was synthesized, purified, and annealed by Ambion. RPMI 8226 cells were transfected with the synthetic siRNA duplex at a final concentration of 50 nM using FuGene 6 Transfection Reagent (Roche). After placing the transfected cells in a complete medium for 24 h, we added IFN-γ (200 U/mL), CpG-DNA (3 μM) or LPS (100 ng/mL) for 12 h. We then harvested the cells, and performed the RT-PCR analysis or luciferase assay.

EMSA

Nuclear extracts of the cells treated with IFN-γ, CpG-DNA, or LPS for 30 min were prepared as described elsewhere 38. We purchased the NF-κB consensus oligonucleotide from Santa Cruz Biotechnology as a probe, and synthesized HLA-DRA promoter fragments that encompassed –323 to –282 (the HLA-DRA NF-κB) by the customer service of GenoTech. For the EMSA, we incubated for 30 min 20 000 cpm of the 32P-labeled probe with 20 μg of nuclear extract proteins in 20 μL of a binding buffer that contained 10 mM HEPES, pH 7.9, 65 mM NaCl, 1 mM dithiothreitol, 0.2 mM EDTA, 0.02% NP-40, 50 μg/mL poly(dI·dC):poly(dI·dC) and 8% glycerol. We then resolved this mixture on 4% PAGE. Next, we conducted an oligonucleotide competition by pre-incubating nuclear extracts with the cold probe (50-fold excess HLA-DRA NF-κB or its mutant for the NF-κB-binding site) for 30 min prior to the addition of the labeled probe, and carried out the NF-κB antibody supershift assays by adding 1 μg of the antibody to the reaction mixture.

ChIP assay

Chromatin from RPMI 8226 cells treated with IFN-γ, CpG-DNA, and LPS for 30 min was prepared as described elsewhere 40 with several modifications. The cells were cross-linked with 1% formaldehyde for 10 min, and then lysed in TE (10 mM Tris-HCl, pH 8.0 and 1 mM EDTA) containing 0.5% NP-40 and protease inhibitors (Roche). Nuclei were pelleted and lysed in TE containing 1% SDS and protease inhibitors. Lysed nuclei were sonicated using a microtip until the average DNA fragment was approximately 600 bp, and cleared by centrifugation. The resulting chromatin supernatants were diluted ten times with dilution buffer (TE containing 0.01% SDS, 1% Triton X-100, 150 mM NaCl, 50 μg/mL salmon sperm DNA and protease inhibitors), and ChIP was performed with 5 μg of anti-NF-κB p65 Ab. After the incubation for 3 h at 4°C, protein A-Sepharose beads (Amersham Biosciences, Piscataway, NJ) were added and incubated for another 1 h. Immune complexes were washed twice with dilution buffer, twice with dilution buffer containing 500 mM NaCl, and twice with TE containing 0.5% NP-40, 0.5% deoxycholic acid and 250 mM LiCl, followed by a final wash with TE. Immune complexes were eluted with 50 mM NaHCO3 and 1% SDS, and then non-covalent links were reversed by addition of NaCl to a final concentration of 300 mM and heating to 65°C for 5 h. DNA was purified by proteinase K digestion, phenol/chloroform extraction, and ethanol precipitation. Immunoprecipitated DNA was analyzed by PCR using promoter-specific primer pairs for the putative NF-κB-binding sites or S-X-Y module of MHC-II promoters. The following primers were used: HLA-DRA(S-X-Y), 5′-CTTCTTTATCCAATGAACGGAGT-3′ and 5′-TGGGGTGTAATAGAGTCTGACCA-3′; HLA-DRA(NF-κB), 5′-TCCCTTACGCAAACTCTCCA-3′ and 5′-GTTGGTCAATGACGGACAGA-3′; HLA-DRB5(NF-κB), 5′-CCTGACTGGTTTTCGCATTTT-3′ and 5′-AGTGAGCGAGACTCCGTCTG-3′; HLA-DPA(NF-κB), 5′-CTGGGTGGGACTTTTCTCCT-3′ and 5′-CTGGGTGGGACTTTTCTCCT-3′. The DNA purified from the sonicated nuclear lysates, was directly analyzed by PCR using the same primer sets, and was used as an input control.

Isolation of primary human B cells and FACS analysis

Peripheral blood was obtained from Korean National Red Cross Central Blood Center of Korea with institutional review board approval. We isolated PBMC using density gradient centrifugation with Ficoll-Paque (Amersham Biosciences). We then purified the B cells using the Human B Cell Isolation Kit (Miltenyi Biotec, Germany) in accordance with the manufacturer's specifications. Surface expression of HLA-DR antigens was analyzed by FACS with a FACScan flow cytometer (BD Biosciences, San Diego, CA). Briefly, cells were washed with PBS containing 0.1% BSA and incubated for 20 min at 4°C with 10 μg/mL of human IgG (Sigma) to block Fc receptors. After blocking, cells were incubated with FITC-conjugated anti-HLA-DR (BD Biosciences) for 1 h at 4°C. FITC-conjugated mouse IgG2a κ was used as an isotype control. FACS data were analyzed using WinMDI 2.8 FACS software.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements

This work was supported by a MOST grant of the Next Generation Growth Engine Program of Korea (M10416020002–05N1602). Doo-Sik Kim was also supported by Research Funds of the National Research Laboratory Program (M1–0400–00–0043) from MOST in Korea. We thank GenoCheck, Inc (Ansan, Korea) for the technical assistance in quantitative real-time PCR analyses.

  • 1

    WILEY-VCH

  • 2

    WILEY-VCH

  • 3

    WILEY-VCH

  • 4

    WILEY-VCH

  • 5

    WILEY-VCH

  • 6

    WILEY-VCH

  • 7

    WILEY-VCH

  • 8

    WILEY-VCH