1130
Jingming Wang et al.
Eur. J. Immunol. 2014. 44: 1130–1136
DOI: 10.1002/eji.201344030
SHORT COMMUNICATION
Bruton’s tyrosine kinase regulates TLR9 but not TLR7 signaling in human plasmacytoid dendritic cells Jingming Wang, Kai-Yeung Lau, Jimmy Jung, Palanikumar Ravindran and Franck J. Barrat Hoffmann-La Roche, Nutley, NJ, USA Plasmacytoid dendritic cells (PDCs) represent a key cell type for both innate and adaptive immunity. PDCs express both TLR7 and TLR9 and the recognition of nucleic acids by these two receptors triggers the production of a large amount of type-I IFN and the induction of PDC maturation into APCs. This unique feature of PDCs is at the basis of clinical development of both TLR7 and TLR9 agonists for infectious diseases, allergy, cancer, and asthma. However, TLR7 and TLR9 recognition of self-nucleic acids is linked to many autoimmune diseases including lupus, and a better understanding of the signaling pathways of these two receptors in PDCs is thus important. We have identified Bruton’s tyrosine kinase (Btk) as an important player for TLR9 but not TLR7 signaling in human PDCs. Blocking Btk using a specific inhibitor leads to the reduction of all TLR9-induced responses in PDCs, including cytokine production and expression of costimulatory molecules, while this has no impact on the TLR7 response. This identifies Btk as a key molecule in TLR9 signaling in PDCs and is the first demonstration that the TLR7 and TLR9 pathways can be dissociated in human PDCs.
Keywords: Autoimmunity r Bruton’s tyrosine kinase r Innate immunity r Plasmacytoid dendritic cells r Toll-like receptors
Additional supporting information may be found in the online version of this article at the publisher’s web-site
Introduction Plasmacytoid dendritic cells (PDCs) have the unique ability to contribute to both innate and adaptive immunity [1]. These cells have a plasma cell-like morphology and are key IFN-producing cells in human blood in response to viruses. The production of type-I IFN by PDCs is dominantly associated with signals induced by the nucleic acid-specific receptors TLR7 and TLR9 [1]. However, upon activation, PDCs can dramatically change their morphology to become potent APCs [1]. This unique feature of PDCs is at the basis of clinical development of TLR7 and TLR9 agonists for infectious diseases, cancer, and asthma [2]. Type-I IFNs have been
Correspondence: Dr. Franck J. Barrat e-mail: [email protected]
C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
implicated as central players in the pathogenesis of lupus [3–6] and TLR7 and TLR9 may be stimulated by self-nucleic acids associated with immune complexes per se [7, 8] or released in the form of neutrophil extracellular traps [9, 10]. Once stimulated, TLR7 and TLR9 activate a common MyD88-dependent pathway, causing the activation of PDCs, production of type-I IFNs and other inflammatory cytokines by activating MAP kinases, NF-κB, and interferon regulatory factor 7 (IRF-7) [1, 11]. Bruton’s tyrosine kinase (Btk), a member of the Tec family nonreceptor tyrosine kinases, is a critical regulatory enzyme in the signaling pathways of multiple activating immunoreceptors containing an immunoreceptor tyrosine-based activation motif domain, including both B-cell receptor on B cells and activating Fc receptors such as FcγR on effector immune cells such as monocytes/macrophages [12, 13]. Btk also seems to be involved in TLR signaling [14, 15] and blocking Btk prevents disease in
www.eji-journal.eu
Eur. J. Immunol. 2014. 44: 1130–1136
Innate immunity
numerous models of autoimmune diseases [12, 13, 16, 17]. The role of Btk in TLR7 and TLR9 signaling in human PDCs has not been defined and could reveal another potential approach for the treatment of systemic lupus erythematosus.
erythematosus patients (18, 20), namely IFIT1, IFIT3, IFI44L, ISG15, MX1, MX2, IFITM1, and IRFs (Fig. 1F). Together, these data demonstrate that Btk affects TLR9- but not TLR7-mediated cytokine production in human PDCs.
Results and discussion
Btk is required for the TLR9- but not TLR7-mediated induction of maturation makers on PDCs
Btk regulates TLR9-mediated induction of cytokines in PDCs but has no role in TLR7 signaling We first investigated the role of Btk in the activation of human PDCs through the TLR7/9 signaling pathways by determining the effects of RN486, a specific Btk inhibitor [17], on cytokine and chemokine production by PDCs. Stimulation of purified human PDCs with either gardiquimod (TLR7 agonist) or the CpG-A ODN2216 (TLR9 agonist) resulted in the production of multiple proinflammatory cytokines and chemokines including IFN-α, TNF, IL-6 (Fig. 1A and B) and IP-10, IL-8, MIP-1α, or MIP-1β (data not shown). Blocking Btk resulted in a significant reduction of all the cytokines and chemokines induced by the TLR9 agonist but not by TLR7 (Fig. 1A and B, and data not shown). The activity and selectivity of the Btk inhibitor RN486 has been characterized in solid-based binding assay [17] but there was no data on the activity of this compound in PDC. We have investigated the effect of RN486 at different concentrations in PDCs and showed the IC50 of RN486 to be 0.4–0.6 μM (Supporting Information Fig. 1). As the activity of RN486 on other kinases is at least 139-fold less potent [17], this demonstrates that the effect that we have observed in our experiments using 1 μM is specific to Btk. To further investigate the role of Btk on TLR7- and TLR9-mediated cytokine production, we looked at the effect of Btk inhibition at the single-cell level by measuring IFN-α- and TNF-producing cells by ELISPOT. The data confirmed our initial observation, as we observed a significant reduction of TLR9-induced response and little to no effect on TLR7-induced response (Fig. 1C). We then measured the impact of Btk inhibition at the gene expression level. Similar to what we observed at the protein level, the induction of cytokine release by the TLR9 agonist was dramatically reduced (Fig. 1D) with little to no effect when a TLR7 agonist was used (Fig. 1E). This effect extended to multiple proinflammatory chemokines (data not shown). Furthermore, we explored the impact of Btk inhibition on the overall gene profile of activated PDCs by performing a microarray with mRNA isolated from human PDCs 3 h after stimulation with either gardiquimod or ODN2216 in the absence or presence of Btk inhibition (Fig. 1F). Of the 5214 genes that passed quality control, 888 and 237 genes were upregulated by at least 1.5-fold, and 1386 and 237 downregulated by at least 1.5-fold with gardiquimod and ODN2216 stimulation, respectively. Out of all upregulated genes, 139 TLR9-induced genes were inhibited by the Btk inhibitor RN486. In contrast, only ten TLR7-induced genes were inhibited by the compound. Pathway analysis demonstrated that many of the TLR9-induced genes inhibited by RN486 are related to immune responses, including a group of IFN-α inducible genes previously shown to be elevated in systemic lupus C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Activation of PDCs by synthetic TLR7 and TLR9 agonists is associated with increased expression of a series of activation markers and costimulatory molecules such as CD40, CD80, and CD86 as previously described [18]. The upregulation of these costimulatory molecules is part of a process leading to dramatic morphological changes of PDCs and the acquisition of strong antigen-presentation function [19]. We observed that PDC activation is also accompanied by the expression of CD69, an early activation marker known to be associated with the activation of most immune cells including T cells, B cells, NK cells, and monocytes, [20] but which expression was not defined in PDCs (Fig. 2). We observed that blocking Btk had a dramatic effect on the expression level of the costimulatory molecules and activation markers CD40, CD86, and CD69 induced by TLR9 but not by TLR7 (Fig. 2A–D). This selective role of Btk on the TLR9 signaling pathway was also observed and further confirmed when PDCs were stimulated with the CpG-B ODN2006 (Fig. 2E). Similar to the CpG-A ODN2216, the induction of costimulatory molecules (CD40, CD69, CD86) by a CpG-B was abrogated by the Btk inhibitor. We also observed a significant reduction in the cytokine and chemokine response when CpG-B-activated PDCs were treated with RN486 (data not shown). This means that Btk is a critical molecule in the TLR9 signaling pathway for both classes of CpGs. Both TLR7 and TLR9 agonists induced these cytokines and chemokines with comparable amplitude so the differences observed with RN486 cannot be attributed to difference in the qualitative response of PDCs to the used ligands.
Btk acts upstream of both IRF-7 and NF-κB in the TLR9 signaling pathway To gain insight into the mechanism of action of Btk inhibition in PDCs, we examined the effects of blocking Btk on the phosphorylation of the two key transcription factors IRF-7 and NF-κB, which respectively contribute to the activation of type-I IFN and other proinflammatory cytokines in PDCs [21]. When activated via TLR9, we observed an increased in IRF-7 and NF-κB phosphorylation in primary PDCs (Fig. 3). There was a clear correlation between the level of phosphorylation of IRF-7 or NF-κB and the presence of costimulatory molecules (Fig. 3A and B). The levels of these activation markers were low and somewhat difficult to detect in a consistent matter due to the timing of the experiment (5 h post activation), which is required in order to detect IRF-7 and NF-κB. However, the data clearly indicated that the measure of the phosphorylation level of IRF-7 and NF-κB is reflective of the level of activation of PDCs. The inhibition of Btk attenuated both www.eji-journal.eu
1131
1132
Jingming Wang et al.
Eur. J. Immunol. 2014. 44: 1130–1136
Figure 1. Btk is a key signaling component of TLR9- but not TLR7-induced cytokine response by PDCs. Purified PDCs were cultured for 18 h with (A) CpG-A ODN2216 (1 μM) or (B) gardiquimod (5 μM) either alone or in the presence of the Btk inhibitor RN486 (1 μM). Levels of cytokines in culture supernatants were then measured by Luminex. Each box represents an individual donor and mean values from four to eight different donors are represented by horizontal lines. *p ≤ 0.05, **p ≤ 0.01, and ***p ≤ 0.001 (Student’s t-test). (C) Purified PDCs were seeded in a 96-well plate precoated with capture Ab, stimulated with ODN2216 (3 μM) or gardiquimod (5 μM) for 20 h either alone or in the presence of the Btk inhibitor RN486 (1 μM) and then removed. The detection antibodies conjugated with enzyme were added, followed by color development. The colored spots were counted as cytokine-positive cells by ELISPOT reader. Samples for each condition were tested in duplicate and data are shown mean + SD of four donors from two separate experiments, two donors per experiment. *p ≤ 0.05, **p ≤ 0.01 (Student’s t-test). (D, E) Purified PDCs were stimulated with (D) TLR9 (CpG-A ODN2216) or (E) TLR7 (Gardiquimod) agonist either alone or in the presence of the Btk inhibitor RN486 (1 μM). Total RNA was isolated and cytokine expression (IFN-α, TNF, and IL-6) was measured by real-time RT-PCR. Each line represents a different donor (N = 4) and data show fold increase over unstimulated cells. (F) Heat map of microarray analysis shows the normalized expression levels of a selected panel of genes that were upregulated by both gardiquimod and ODN2216, and inhibited by RN486 by at least 50% in ODN2216-stimulated cells. Genes are grouped into four different categories as indicated based on pathway analysis. C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eji-journal.eu
Eur. J. Immunol. 2014. 44: 1130–1136
Innate immunity
Figure 2. Blockade of Btk inhibits the activation of PDCs by TLR9 but not TLR7. (A–D) Purified PDCs were cultured for 18 h with (A, B) ODN2216 (1 μM) or (C, D) gardiquimod (5 μM), either alone or in the presence of RN486 at the indicated concentrations. Cells were characterized for CD40, CD86, and CD69 expression by flow cytometry analysis. Example of the effect of blocking Btk (A, C) on TLR9 or TLR7 and mean fluorescence intensity (MFI) (B, D) is shown. Shaded gray histograms represent basal expression of all cell surface markers in unstimulated cells. Solid black line, solid gray line, dotted black line, and dotted gray line represent in sequential order vehicle, 0.3, 1, and 3 μM of RN486, respectively. Data in (B) and (D) are shown as the mean + SD of six donors pooled from three independent experiments *p ≤ 0.05, **p ≤ 0.01 (Student’s t-test). (E) Effect of RN486 (3 μM) on the activation of PDCs by TLR9 agonists CpG-A (ODN2216) or CpG-B (ODN2006). Data show the mean of a total of six independent donors pooled from two independent experiments.
TLR9-induced IRF-7 and NF-κB phosphorylation in a dosedependent manner (Fig. 3C). We observed no effect when a TLR7 agonist was used (data not shown). This indicates that (i) Btk acts upstream of both the IRF-7 and NF-κB pathways in the TLR9 signaling pathway in PDCs and (ii) that both pathways are affected by Btk inhibition.
C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Concluding remarks Both TLR7 and TLR9 have been shown to recognize self-RNA and self-DNA, respectively, and to induce IFN production by PDCs [22]. The signaling of TLR7 and TLR9 seems comparable and involves both NF-κB and IRF-7 transcription factors. The induction
www.eji-journal.eu
1133
1134
Jingming Wang et al.
Eur. J. Immunol. 2014. 44: 1130–1136
Figure 3. Effects of RN486 on TLR9 agonist-stimulated phosphorylation of IRF-7 and NF-κB. (A, B) Purified PDCs were cultured for 5 h with ODN2216 either alone or in the presence of RN486. Cells were then characterized for the surface markers CD123, CD69, and CD40 as well as for phospho-IRF-7 or phospho-NF-κB and analyzed by flow cytometry. Differential expression of surface activation markers CD40 and CD69 in (A) phospho-IRF-7positive or -negative PDCs and in (B) phospho-NF-κB-positive or -negative PDCs is shown. Data are representative of two separate experiments. (C) Purified PDCs were cultured for 5 h with ODN2216 either alone or in the presence of RN486 at the indicated concentrations. Cells were then characterized for the surface markers CD123, CD69, and CD40 as well as for phospho-IRF-7 or phospho-NF-κB and analyzed by flow cytometry. Data are representative of two separate experiments with two donors per experiments.
of PDC maturation and upregulation of costimulatory molecules appear to be dependent on NF-κB activation while IRF-7 is the key transcription factor leading to type-I IFN by PDCs. One key differentiation element in the triggering of these two pathways relies on the nature of the endosome where the recognition of the ligand occurs [18]. The implication of TLR7 and TLR9 in self-nucleic acid recognition in disease such as lupus has been characterized [22]. In mouse lupus mouse models, it seems that TLR9 and TLR7 may have different role with TLR9 having a protective role in some models [23, 24]. It is unclear whether these mouse models are
C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
reflective of the human situation but it stresses the importance of a better understanding of the difference in the signaling pathways of these receptors. It is unclear how TLR7 and TLR9 signaling differ in PDCs beside their difference in ligand recognition. Here, we show that Btk is critical for TLR9 signaling, controlling both IRF-7 and NF-κB phosphorylation and activation while we observed no effect on TLR7-induced PDC response suggesting that these two receptors have a unique signaling pathway. We have used PDCs as these cells are the most relevant to study TLR7 and TLR9 signaling and can produce both type-I IFN through the IRF-7 pathway
www.eji-journal.eu
Eur. J. Immunol. 2014. 44: 1130–1136
Innate immunity
but also a set of proinflammatory cytokines which are regulated by NF-κB. However, PDCs represent less than 1% of blood cells which prevented us from performing direct binding studies with Btk and other players of the TLR signaling pathway or the TLRs themselves. These data however clearly show that the TLR9- but not TLR7-induced response can be blocked using a specific Btk inhibitor and reveal a differential role of Btk in regulating the two closely linked signaling pathways.
A total of 8185 probe sets representing 5214 genes passed the quality check and curation process. The criteria for calling a gene modulated by RN486 are (i) the fold change between stimulated and unstimulated samples is >1.5 and corresponding false discovery rate is
Jingming Wang et al.
Eur. J. Immunol. 2014. 44: 1130–1136
DOI: 10.1002/eji.201344030
SHORT COMMUNICATION
Bruton’s tyrosine kinase regulates TLR9 but not TLR7 signaling in human plasmacytoid dendritic cells Jingming Wang, Kai-Yeung Lau, Jimmy Jung, Palanikumar Ravindran and Franck J. Barrat Hoffmann-La Roche, Nutley, NJ, USA Plasmacytoid dendritic cells (PDCs) represent a key cell type for both innate and adaptive immunity. PDCs express both TLR7 and TLR9 and the recognition of nucleic acids by these two receptors triggers the production of a large amount of type-I IFN and the induction of PDC maturation into APCs. This unique feature of PDCs is at the basis of clinical development of both TLR7 and TLR9 agonists for infectious diseases, allergy, cancer, and asthma. However, TLR7 and TLR9 recognition of self-nucleic acids is linked to many autoimmune diseases including lupus, and a better understanding of the signaling pathways of these two receptors in PDCs is thus important. We have identified Bruton’s tyrosine kinase (Btk) as an important player for TLR9 but not TLR7 signaling in human PDCs. Blocking Btk using a specific inhibitor leads to the reduction of all TLR9-induced responses in PDCs, including cytokine production and expression of costimulatory molecules, while this has no impact on the TLR7 response. This identifies Btk as a key molecule in TLR9 signaling in PDCs and is the first demonstration that the TLR7 and TLR9 pathways can be dissociated in human PDCs.
Keywords: Autoimmunity r Bruton’s tyrosine kinase r Innate immunity r Plasmacytoid dendritic cells r Toll-like receptors
Additional supporting information may be found in the online version of this article at the publisher’s web-site
Introduction Plasmacytoid dendritic cells (PDCs) have the unique ability to contribute to both innate and adaptive immunity [1]. These cells have a plasma cell-like morphology and are key IFN-producing cells in human blood in response to viruses. The production of type-I IFN by PDCs is dominantly associated with signals induced by the nucleic acid-specific receptors TLR7 and TLR9 [1]. However, upon activation, PDCs can dramatically change their morphology to become potent APCs [1]. This unique feature of PDCs is at the basis of clinical development of TLR7 and TLR9 agonists for infectious diseases, cancer, and asthma [2]. Type-I IFNs have been
Correspondence: Dr. Franck J. Barrat e-mail: [email protected]
C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
implicated as central players in the pathogenesis of lupus [3–6] and TLR7 and TLR9 may be stimulated by self-nucleic acids associated with immune complexes per se [7, 8] or released in the form of neutrophil extracellular traps [9, 10]. Once stimulated, TLR7 and TLR9 activate a common MyD88-dependent pathway, causing the activation of PDCs, production of type-I IFNs and other inflammatory cytokines by activating MAP kinases, NF-κB, and interferon regulatory factor 7 (IRF-7) [1, 11]. Bruton’s tyrosine kinase (Btk), a member of the Tec family nonreceptor tyrosine kinases, is a critical regulatory enzyme in the signaling pathways of multiple activating immunoreceptors containing an immunoreceptor tyrosine-based activation motif domain, including both B-cell receptor on B cells and activating Fc receptors such as FcγR on effector immune cells such as monocytes/macrophages [12, 13]. Btk also seems to be involved in TLR signaling [14, 15] and blocking Btk prevents disease in
www.eji-journal.eu
Eur. J. Immunol. 2014. 44: 1130–1136
Innate immunity
numerous models of autoimmune diseases [12, 13, 16, 17]. The role of Btk in TLR7 and TLR9 signaling in human PDCs has not been defined and could reveal another potential approach for the treatment of systemic lupus erythematosus.
erythematosus patients (18, 20), namely IFIT1, IFIT3, IFI44L, ISG15, MX1, MX2, IFITM1, and IRFs (Fig. 1F). Together, these data demonstrate that Btk affects TLR9- but not TLR7-mediated cytokine production in human PDCs.
Results and discussion
Btk is required for the TLR9- but not TLR7-mediated induction of maturation makers on PDCs
Btk regulates TLR9-mediated induction of cytokines in PDCs but has no role in TLR7 signaling We first investigated the role of Btk in the activation of human PDCs through the TLR7/9 signaling pathways by determining the effects of RN486, a specific Btk inhibitor [17], on cytokine and chemokine production by PDCs. Stimulation of purified human PDCs with either gardiquimod (TLR7 agonist) or the CpG-A ODN2216 (TLR9 agonist) resulted in the production of multiple proinflammatory cytokines and chemokines including IFN-α, TNF, IL-6 (Fig. 1A and B) and IP-10, IL-8, MIP-1α, or MIP-1β (data not shown). Blocking Btk resulted in a significant reduction of all the cytokines and chemokines induced by the TLR9 agonist but not by TLR7 (Fig. 1A and B, and data not shown). The activity and selectivity of the Btk inhibitor RN486 has been characterized in solid-based binding assay [17] but there was no data on the activity of this compound in PDC. We have investigated the effect of RN486 at different concentrations in PDCs and showed the IC50 of RN486 to be 0.4–0.6 μM (Supporting Information Fig. 1). As the activity of RN486 on other kinases is at least 139-fold less potent [17], this demonstrates that the effect that we have observed in our experiments using 1 μM is specific to Btk. To further investigate the role of Btk on TLR7- and TLR9-mediated cytokine production, we looked at the effect of Btk inhibition at the single-cell level by measuring IFN-α- and TNF-producing cells by ELISPOT. The data confirmed our initial observation, as we observed a significant reduction of TLR9-induced response and little to no effect on TLR7-induced response (Fig. 1C). We then measured the impact of Btk inhibition at the gene expression level. Similar to what we observed at the protein level, the induction of cytokine release by the TLR9 agonist was dramatically reduced (Fig. 1D) with little to no effect when a TLR7 agonist was used (Fig. 1E). This effect extended to multiple proinflammatory chemokines (data not shown). Furthermore, we explored the impact of Btk inhibition on the overall gene profile of activated PDCs by performing a microarray with mRNA isolated from human PDCs 3 h after stimulation with either gardiquimod or ODN2216 in the absence or presence of Btk inhibition (Fig. 1F). Of the 5214 genes that passed quality control, 888 and 237 genes were upregulated by at least 1.5-fold, and 1386 and 237 downregulated by at least 1.5-fold with gardiquimod and ODN2216 stimulation, respectively. Out of all upregulated genes, 139 TLR9-induced genes were inhibited by the Btk inhibitor RN486. In contrast, only ten TLR7-induced genes were inhibited by the compound. Pathway analysis demonstrated that many of the TLR9-induced genes inhibited by RN486 are related to immune responses, including a group of IFN-α inducible genes previously shown to be elevated in systemic lupus C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Activation of PDCs by synthetic TLR7 and TLR9 agonists is associated with increased expression of a series of activation markers and costimulatory molecules such as CD40, CD80, and CD86 as previously described [18]. The upregulation of these costimulatory molecules is part of a process leading to dramatic morphological changes of PDCs and the acquisition of strong antigen-presentation function [19]. We observed that PDC activation is also accompanied by the expression of CD69, an early activation marker known to be associated with the activation of most immune cells including T cells, B cells, NK cells, and monocytes, [20] but which expression was not defined in PDCs (Fig. 2). We observed that blocking Btk had a dramatic effect on the expression level of the costimulatory molecules and activation markers CD40, CD86, and CD69 induced by TLR9 but not by TLR7 (Fig. 2A–D). This selective role of Btk on the TLR9 signaling pathway was also observed and further confirmed when PDCs were stimulated with the CpG-B ODN2006 (Fig. 2E). Similar to the CpG-A ODN2216, the induction of costimulatory molecules (CD40, CD69, CD86) by a CpG-B was abrogated by the Btk inhibitor. We also observed a significant reduction in the cytokine and chemokine response when CpG-B-activated PDCs were treated with RN486 (data not shown). This means that Btk is a critical molecule in the TLR9 signaling pathway for both classes of CpGs. Both TLR7 and TLR9 agonists induced these cytokines and chemokines with comparable amplitude so the differences observed with RN486 cannot be attributed to difference in the qualitative response of PDCs to the used ligands.
Btk acts upstream of both IRF-7 and NF-κB in the TLR9 signaling pathway To gain insight into the mechanism of action of Btk inhibition in PDCs, we examined the effects of blocking Btk on the phosphorylation of the two key transcription factors IRF-7 and NF-κB, which respectively contribute to the activation of type-I IFN and other proinflammatory cytokines in PDCs [21]. When activated via TLR9, we observed an increased in IRF-7 and NF-κB phosphorylation in primary PDCs (Fig. 3). There was a clear correlation between the level of phosphorylation of IRF-7 or NF-κB and the presence of costimulatory molecules (Fig. 3A and B). The levels of these activation markers were low and somewhat difficult to detect in a consistent matter due to the timing of the experiment (5 h post activation), which is required in order to detect IRF-7 and NF-κB. However, the data clearly indicated that the measure of the phosphorylation level of IRF-7 and NF-κB is reflective of the level of activation of PDCs. The inhibition of Btk attenuated both www.eji-journal.eu
1131
1132
Jingming Wang et al.
Eur. J. Immunol. 2014. 44: 1130–1136
Figure 1. Btk is a key signaling component of TLR9- but not TLR7-induced cytokine response by PDCs. Purified PDCs were cultured for 18 h with (A) CpG-A ODN2216 (1 μM) or (B) gardiquimod (5 μM) either alone or in the presence of the Btk inhibitor RN486 (1 μM). Levels of cytokines in culture supernatants were then measured by Luminex. Each box represents an individual donor and mean values from four to eight different donors are represented by horizontal lines. *p ≤ 0.05, **p ≤ 0.01, and ***p ≤ 0.001 (Student’s t-test). (C) Purified PDCs were seeded in a 96-well plate precoated with capture Ab, stimulated with ODN2216 (3 μM) or gardiquimod (5 μM) for 20 h either alone or in the presence of the Btk inhibitor RN486 (1 μM) and then removed. The detection antibodies conjugated with enzyme were added, followed by color development. The colored spots were counted as cytokine-positive cells by ELISPOT reader. Samples for each condition were tested in duplicate and data are shown mean + SD of four donors from two separate experiments, two donors per experiment. *p ≤ 0.05, **p ≤ 0.01 (Student’s t-test). (D, E) Purified PDCs were stimulated with (D) TLR9 (CpG-A ODN2216) or (E) TLR7 (Gardiquimod) agonist either alone or in the presence of the Btk inhibitor RN486 (1 μM). Total RNA was isolated and cytokine expression (IFN-α, TNF, and IL-6) was measured by real-time RT-PCR. Each line represents a different donor (N = 4) and data show fold increase over unstimulated cells. (F) Heat map of microarray analysis shows the normalized expression levels of a selected panel of genes that were upregulated by both gardiquimod and ODN2216, and inhibited by RN486 by at least 50% in ODN2216-stimulated cells. Genes are grouped into four different categories as indicated based on pathway analysis. C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eji-journal.eu
Eur. J. Immunol. 2014. 44: 1130–1136
Innate immunity
Figure 2. Blockade of Btk inhibits the activation of PDCs by TLR9 but not TLR7. (A–D) Purified PDCs were cultured for 18 h with (A, B) ODN2216 (1 μM) or (C, D) gardiquimod (5 μM), either alone or in the presence of RN486 at the indicated concentrations. Cells were characterized for CD40, CD86, and CD69 expression by flow cytometry analysis. Example of the effect of blocking Btk (A, C) on TLR9 or TLR7 and mean fluorescence intensity (MFI) (B, D) is shown. Shaded gray histograms represent basal expression of all cell surface markers in unstimulated cells. Solid black line, solid gray line, dotted black line, and dotted gray line represent in sequential order vehicle, 0.3, 1, and 3 μM of RN486, respectively. Data in (B) and (D) are shown as the mean + SD of six donors pooled from three independent experiments *p ≤ 0.05, **p ≤ 0.01 (Student’s t-test). (E) Effect of RN486 (3 μM) on the activation of PDCs by TLR9 agonists CpG-A (ODN2216) or CpG-B (ODN2006). Data show the mean of a total of six independent donors pooled from two independent experiments.
TLR9-induced IRF-7 and NF-κB phosphorylation in a dosedependent manner (Fig. 3C). We observed no effect when a TLR7 agonist was used (data not shown). This indicates that (i) Btk acts upstream of both the IRF-7 and NF-κB pathways in the TLR9 signaling pathway in PDCs and (ii) that both pathways are affected by Btk inhibition.
C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Concluding remarks Both TLR7 and TLR9 have been shown to recognize self-RNA and self-DNA, respectively, and to induce IFN production by PDCs [22]. The signaling of TLR7 and TLR9 seems comparable and involves both NF-κB and IRF-7 transcription factors. The induction
www.eji-journal.eu
1133
1134
Jingming Wang et al.
Eur. J. Immunol. 2014. 44: 1130–1136
Figure 3. Effects of RN486 on TLR9 agonist-stimulated phosphorylation of IRF-7 and NF-κB. (A, B) Purified PDCs were cultured for 5 h with ODN2216 either alone or in the presence of RN486. Cells were then characterized for the surface markers CD123, CD69, and CD40 as well as for phospho-IRF-7 or phospho-NF-κB and analyzed by flow cytometry. Differential expression of surface activation markers CD40 and CD69 in (A) phospho-IRF-7positive or -negative PDCs and in (B) phospho-NF-κB-positive or -negative PDCs is shown. Data are representative of two separate experiments. (C) Purified PDCs were cultured for 5 h with ODN2216 either alone or in the presence of RN486 at the indicated concentrations. Cells were then characterized for the surface markers CD123, CD69, and CD40 as well as for phospho-IRF-7 or phospho-NF-κB and analyzed by flow cytometry. Data are representative of two separate experiments with two donors per experiments.
of PDC maturation and upregulation of costimulatory molecules appear to be dependent on NF-κB activation while IRF-7 is the key transcription factor leading to type-I IFN by PDCs. One key differentiation element in the triggering of these two pathways relies on the nature of the endosome where the recognition of the ligand occurs [18]. The implication of TLR7 and TLR9 in self-nucleic acid recognition in disease such as lupus has been characterized [22]. In mouse lupus mouse models, it seems that TLR9 and TLR7 may have different role with TLR9 having a protective role in some models [23, 24]. It is unclear whether these mouse models are
C 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
reflective of the human situation but it stresses the importance of a better understanding of the difference in the signaling pathways of these receptors. It is unclear how TLR7 and TLR9 signaling differ in PDCs beside their difference in ligand recognition. Here, we show that Btk is critical for TLR9 signaling, controlling both IRF-7 and NF-κB phosphorylation and activation while we observed no effect on TLR7-induced PDC response suggesting that these two receptors have a unique signaling pathway. We have used PDCs as these cells are the most relevant to study TLR7 and TLR9 signaling and can produce both type-I IFN through the IRF-7 pathway
www.eji-journal.eu
Eur. J. Immunol. 2014. 44: 1130–1136
Innate immunity
but also a set of proinflammatory cytokines which are regulated by NF-κB. However, PDCs represent less than 1% of blood cells which prevented us from performing direct binding studies with Btk and other players of the TLR signaling pathway or the TLRs themselves. These data however clearly show that the TLR9- but not TLR7-induced response can be blocked using a specific Btk inhibitor and reveal a differential role of Btk in regulating the two closely linked signaling pathways.
A total of 8185 probe sets representing 5214 genes passed the quality check and curation process. The criteria for calling a gene modulated by RN486 are (i) the fold change between stimulated and unstimulated samples is >1.5 and corresponding false discovery rate is
