TRIF–GEFH1–RhoB pathway is involved in MHCII expression on dendritic cells that is critical for CD4 T-cell activation
2006; Springer Nature; Volume: 25; Issue: 17 Linguagem: Inglês
10.1038/sj.emboj.7601286
ISSN1460-2075
AutoresHokuto Kamon, Takaya Kawabe, Hidemitsu Kitamura, Jihye Lee, Daisuke Kamimura, Tsuneyasu Kaisho, Shizuo Akira, Akihiro Iwamatsu, Hisashi Koga, Masaaki Murakami, Toshio Hirano,
Tópico(s)Immunotherapy and Immune Responses
ResumoArticle17 August 2006free access TRIF–GEFH1–RhoB pathway is involved in MHCII expression on dendritic cells that is critical for CD4 T-cell activation Hokuto Kamon Hokuto Kamon Laboratory of Developmental Immunology, Graduate School of Frontier Biosciences and Graduate School of Medicine, Osaka University, Osaka, Japan Search for more papers by this author Takaya Kawabe Takaya Kawabe Laboratory of Developmental Immunology, Graduate School of Frontier Biosciences and Graduate School of Medicine, Osaka University, Osaka, Japan Search for more papers by this author Hidemitsu Kitamura Hidemitsu Kitamura Laboratory for Cytokine Signaling, RIKEN Research Center for Allergy and Immunology, Yokohama, Japan Search for more papers by this author Jihye Lee Jihye Lee Laboratory of Developmental Immunology, Graduate School of Frontier Biosciences and Graduate School of Medicine, Osaka University, Osaka, Japan Search for more papers by this author Daisuke Kamimura Daisuke Kamimura Laboratory for Cytokine Signaling, RIKEN Research Center for Allergy and Immunology, Yokohama, Japan Search for more papers by this author Tsuneyasu Kaisho Tsuneyasu Kaisho Laboratory for Host Defense, RIKEN Research Center for Allergy and Immunology, Yokohama, Japan Search for more papers by this author Shizuo Akira Shizuo Akira Research Institute for Microbial Diseases, Osaka University, Osaka, Japan Search for more papers by this author Akihiro Iwamatsu Akihiro Iwamatsu Protein research network, Yokohama, Japan Search for more papers by this author Hisashi Koga Hisashi Koga Kazusa DNA Research Institute, Kisarazu, Japan Search for more papers by this author Masaaki Murakami Masaaki Murakami Laboratory of Developmental Immunology, Graduate School of Frontier Biosciences and Graduate School of Medicine, Osaka University, Osaka, Japan Search for more papers by this author Toshio Hirano Corresponding Author Toshio Hirano Laboratory of Developmental Immunology, Graduate School of Frontier Biosciences and Graduate School of Medicine, Osaka University, Osaka, Japan Laboratory for Cytokine Signaling, RIKEN Research Center for Allergy and Immunology, Yokohama, Japan Search for more papers by this author Hokuto Kamon Hokuto Kamon Laboratory of Developmental Immunology, Graduate School of Frontier Biosciences and Graduate School of Medicine, Osaka University, Osaka, Japan Search for more papers by this author Takaya Kawabe Takaya Kawabe Laboratory of Developmental Immunology, Graduate School of Frontier Biosciences and Graduate School of Medicine, Osaka University, Osaka, Japan Search for more papers by this author Hidemitsu Kitamura Hidemitsu Kitamura Laboratory for Cytokine Signaling, RIKEN Research Center for Allergy and Immunology, Yokohama, Japan Search for more papers by this author Jihye Lee Jihye Lee Laboratory of Developmental Immunology, Graduate School of Frontier Biosciences and Graduate School of Medicine, Osaka University, Osaka, Japan Search for more papers by this author Daisuke Kamimura Daisuke Kamimura Laboratory for Cytokine Signaling, RIKEN Research Center for Allergy and Immunology, Yokohama, Japan Search for more papers by this author Tsuneyasu Kaisho Tsuneyasu Kaisho Laboratory for Host Defense, RIKEN Research Center for Allergy and Immunology, Yokohama, Japan Search for more papers by this author Shizuo Akira Shizuo Akira Research Institute for Microbial Diseases, Osaka University, Osaka, Japan Search for more papers by this author Akihiro Iwamatsu Akihiro Iwamatsu Protein research network, Yokohama, Japan Search for more papers by this author Hisashi Koga Hisashi Koga Kazusa DNA Research Institute, Kisarazu, Japan Search for more papers by this author Masaaki Murakami Masaaki Murakami Laboratory of Developmental Immunology, Graduate School of Frontier Biosciences and Graduate School of Medicine, Osaka University, Osaka, Japan Search for more papers by this author Toshio Hirano Corresponding Author Toshio Hirano Laboratory of Developmental Immunology, Graduate School of Frontier Biosciences and Graduate School of Medicine, Osaka University, Osaka, Japan Laboratory for Cytokine Signaling, RIKEN Research Center for Allergy and Immunology, Yokohama, Japan Search for more papers by this author Author Information Hokuto Kamon1,‡, Takaya Kawabe1,‡, Hidemitsu Kitamura2, Jihye Lee1, Daisuke Kamimura2, Tsuneyasu Kaisho3, Shizuo Akira4, Akihiro Iwamatsu5, Hisashi Koga6, Masaaki Murakami1 and Toshio Hirano 1,2 1Laboratory of Developmental Immunology, Graduate School of Frontier Biosciences and Graduate School of Medicine, Osaka University, Osaka, Japan 2Laboratory for Cytokine Signaling, RIKEN Research Center for Allergy and Immunology, Yokohama, Japan 3Laboratory for Host Defense, RIKEN Research Center for Allergy and Immunology, Yokohama, Japan 4Research Institute for Microbial Diseases, Osaka University, Osaka, Japan 5Protein research network, Yokohama, Japan 6Kazusa DNA Research Institute, Kisarazu, Japan ‡These authors contributed equally to the work *Corresponding author. Department of Molecular Oncology (C-7), Graduate School of Medicine, Osaka University, Suita, 2-2 Yamada-oka, Suita, Osaka 565-0871, Japan. Tel.: +81 66 879 3880/3881; Fax: +81 66 879 3889; E-mail: [email protected] The EMBO Journal (2006)25:4108-4119https://doi.org/10.1038/sj.emboj.7601286 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Dendritic cells (DC) play a central role in immune responses by presenting antigenic peptides to CD4+ T cells through MHCII molecules. Here, we demonstrate a TRIF–GEFH1–RhoB pathway is involved in MHCII surface expression on DC. We show the TRIF (TIR domain-containing adapter inducing IFNβ)- but not the myeloid differentiation factor 88 (MyD88)-dependent pathway of lipopolysaccharide (LPS)-signaling in DC is crucial for the MHCII surface expression, followed by CD4+ T-cell activation. LPS increased the activity of RhoB, but not of RhoA, Cdc42, or Rac1/2 in a manor dependent on LPS-TRIF- but not LPS-Myd88-signaling. RhoB colocalized with MHCII+ lysosomes in DC. A dominant-negative (DN) form of RhoB (DN-RhoB) or RhoB's RNAi in DC inhibited the LPS-induced MHCII surface expression. Moreover, we found GEFH1 associated with RhoB, and DN-GEFH1 or GEFH1's RNAi suppressed the LPS-mediated RhoB activation and MHCII surface expression. DN-RhoB attenuated the DC's CD4+ T-cell stimulatory activity. Thus, our results provide a molecular mechanism relating how the MHCII surface expression is regulated during the maturation stage of DC. The activation of GEFH1-RhoB through the TRIF-dependent pathway of LPS in DC might be a critical target for controlling the activation of CD4+ T cells. Introduction The initial CD4+ T-cell activation is induced by the presentation of antigen peptides on the MHCII molecules of dentritic cell (DC) (Steinman and Inaba, 1985). In immature DC, MHCII molecules accumulate in late endosomal and lysosomal compartments near the nucleus. Vesicles carrying MHCII molecules are considered to be a primary site for MHCII-peptide complex formation (Peters et al, 1991). The activation stages of CD4+ T cells are regulated by a T-cell receptor (TCR)-induced signaling pathway that is initiated by an interaction with MHCII-peptide complexes on DC; the expression level of the surface MHCII-peptide complexes on DC is therefore important for determining the properties of CD4+ T cells (Robinson and Delvig, 2002). After the DC receive maturation signals through a Toll-like receptor (TLR), huge amounts of MHCII-peptide complexes first appear in lysosome-like compartments and subsequently on the DC plasma membrane, where CD4+ T cells are able to recognize them (Steinman et al, 1997). Several studies have focused on how the MHCII-peptide complexes in MHCII+ lysosome vesicles reach the cell surface. It has been suggested that cytoskeletal components play a critical role in the vesicle trafficking in many types of cells, including DC (Rodriguez-Boulan et al, 2005). It is not clear, however, how the TLR signals molecularly control the surface expression of MHCII molecules in DC. Members of the TLR family recognize conserved microbial structures and activate both innate and adaptive immune responses through the activation of DC (Medzhitov and Janeway, 1997; Schnare et al, 2001). After TLR4 binds lipopolysaccharide (LPS), at least two signaling pathways are activated. One is dependent on myeloid differentiation factor 88 (MyD88) and the other on TRIF (TIR domain-containing adapter inducing IFNβ) (Beutler, 2004). Myd88, a TLR-binding protein, contributes to the LPS-induced activation of NF-κB and mitogen-activated protein kinases. Previous studies indicated the LPS-mediated activation of Myd88-dependent signaling induces inflammatory cytokines such as IL-6, TNFα, and IL-1. Myd88-deficient DC normally show an enhanced surface expression of MHCII and costimulatory molecules after LPS treatment (Horng et al, 2002). On the other hand, TRIF-dependent signaling, which is identified as Myd88-independent signaling from the TLR4 receptor, leads to IFNβ expression via activation of IRF3 (Hoebe et al, 2003a). The TRIF–IFNβ pathway induces the upregulation of costimulatory molecules such as CD86, CD80, and CD40; thus, IFNαβR-deficient macrophages fail to enhance costimulatory molecule expression after LPS stimulation (Hoebe et al, 2003b). Recently, the modification of actin organization by LPS-TLR4-Myd88-independent signaling was shown to play an important role in antigen uptake by DC (West et al, 2004). It is also demonstrated that TRIF-mediated signaling is regulated by TRAF6, TRAF3, RIP, TBK1, IKK-ε, and IRF3 (Barton and Medzhitov, 2004; Hacker et al, 2005; Oganesyan et al, 2005). However, involvement of TRIF pathway in MHCII vesicle movement has not been described. The Rho family of GTPases, Rho, Rac, and Cdc42, play key roles in intracellular vesicle trafficking, including endocytosis and exocytosis, through dynamic regulation of the actin and tubulin cytoskeleton (Ridley, 2001). RhoA, RhoB, and RhoC are all similar in primary structure, but they appear to have some functional differences (Ridley, 2001; Burridge and Wennerberg, 2004). Activated RhoB molecules inhibit the transport of vesicles carrying EGF receptors to lysosomes (Gampel et al, 1999). Some Rho family members have been shown to function in DC (Garrett et al, 2000; Burns et al, 2001). Garrett et al (2000) showed Cdc42 is critical for endocytosis in immature DC, and Benvenuti et al (2004) showed Rac1/2 control mature DC migration towards and contact with T cells. The activation of Rho family proteins is regulated by their binding with GDP or GTP, which is mediated by various cell-surface receptors (Kjoller and Hall, 1999). Cycling between the GDP- and GTP-bound state of Rho family members is controlled primarily by two classes of regulatory molecules: GTPase-activating proteins (GAPs), which enhance the relatively slow intrinsic GTPase activity of Rho proteins, and guanine nucleotide-exchange factors (GEFs), which catalyze the exchange of GDP for GTP. GAPs suppress Rho activity, whereas GEFs promote it. GEFH1, which associates with tubulin in a manner dependent on a Zn-finger motif, is a unique regulator of the Rho family (Ren et al, 1998). GEFH1 is known to activate Rho A, B, C, and Rac1 (Ren et al, 1998). Although roles of this molecule in DC have not been investigated, GEFH1 is known to play a role in morphological changes and cell–cell interaction in other cell types (Benais-Pont et al, 2003; Aijaz et al, 2005; Birukova et al, 2006). Here, we showed TRIF-dependent but Myd88-independent activation of a GEFH1–RhoB pathway by LPS is involved in the surface expression of MHCII molecules, which induces the optimal CD4+ T cell activation in DC. Results LPS–TRIF pathway is critical for the upregulation of MHCII molecules on DC, followed by CD4+ T-cell activation LPS-mediated TLR4 signaling stimulates two pathways: (i) MyD88-mediated and (ii) TRIF-mediated (Hoebe et al, 2003a; Barton and Medzhitov, 2004). We investigated which pathway was important for the LPS-mediated upregulation of MHCII molecules on DC. We prepared DC from TRIFKO, Myd88KO, and control mice. As TRIF-deficiency might reduce total amount of MHCII molecules in DC, we investigated total MHCII level in DC by Western blotting as described previously (Kitamura et al, 2005) and showed MHCII levels were comparable between DC derived from TRIFKO, Myd88KO, and control mice (Supplementary Figure S1). Therefore, we stimulated them with LPS to investigate MHCII expression. LPS induced the upregulation of MHCII molecules on DC from control and Myd88KO mice but not on DC from TRIFKO mice (Figure 1A and B). Moreover, we showed LPS-mediated upregulation of MHCII suppressed in TRIFKO DCs compared to other ones from 6 h after stimulation (Figure 1C). Figure 1.A TRIF-mediated pathway is critical for the LPS-mediated upregulation of MHCII molecules on DC and division of CD4+ T cells in vivo. (A) BMDC were prepared from TRIFKO, Myd88KO, and control mice, and stimulated with LPS. The surface expression of MHCII on CD11c+7AAD− cells was measured by a flow cytometer. The values on the profiles are the percentage of cells represented by the MHCIIhigh population. (B) The average of the percent MHCIIhigh CD11c+7AAD− cells with or without LPS treatment in five independent experiments is shown with error bars indicating 1 s.d. *P<0.05) was calculated by Student's t-test. (C) The percent MHCIIhigh CD11c+7AAD− cells with or without LPS treatment is shown. These experiments were performed four times independently, and representative data are shown. (D) BMDC were prepared from TRIFKO and control mice, stimulated with LPS and loaded with peptide-P25. The cells were injected at footpads of C57BL/6 mice that were transferred with CFSE-P25 TCR-Tg CD4+ T cells (CD45.1+). T-cell division in the inguinal lymphonodes was analyzed by CFSE profiles of CD45.1+CD4+ cells. The values in the FACS profiles are the percentage of cells represented by the CD45.1+CFSE+ population. The average of the number of CFSE+CD45.1+CD4+T cells in the each condition of three independent experiments is shown with error bars indicating 1 s.d. The P-value (<0.05) was calculated by Student's t-test and is indicated by *. (E) CD4+ T cells of p25 TCR-Tg mice were co-cultured with BMDC that were prepared from TRIFKO and control mice, and incubated with or without LPS in the presence of the antigenic peptide (1–10 000 ng/ml) for 2 days. IL-2 in the culture supernatants was measured by ELISA. These experiments were performed three times, independently, and representative data are shown. The average IL-2 production is shown with error bars indicating 1 s.d. The P-value (<0.05 or <0.01) was calculated by Student's t-test and is indicated by * or **. Download figure Download PowerPoint We next investigated whether TRIFKO-DC have normal level of CD4+ T-cell stimulatory activity. LPS-stimulated wild-type or TRIFKO DC were pulsed with peptide-P25 and injected in footpads of C57BL6 mice, which were transferred with CFSE (5- and 6-carboxyfluorescein diacetate succinimidyl ester)-labeled p25-CD4+ T cells (CD45.1+). While the wild type DC induced strong division of p25-CD4+ T cells in vivo, the TRIFKO ones hardly enhanced T-cell division (Figure 1D). We also showed both TRIFKO DCs and control DCs reached almost similarly to inguinal lymphnodes in this experiment (Supplementary Figure S2). However, the defect of in vivo T-cell activation could be due to too many other factors besides MHCII expression. To rule out these possibilities better, we performed an in vitro assay of peptide presentation and showed TRIFKO DCs had reduced T-cell stimulatory activity compared to the control ones at any concentration of the antigen peptide (Figure 1E). From these results, it is reasonable that defect of LPS-mediated MHCII upregulation is at least in part involved in reduced T-cell stimulatory activity of TRIFKO DCs, although we also observed TRIFKO DCs defect LPS-mediated upregulation of costimulatory molecules (Supplementary Figure S3). All these results suggested that a TRIF-dependent pathway of TLR4 signaling in DC is critical for the surface expression of MHCII, which at least in part plays a role for activation of CD4+ T cells. Activity of RhoB but not of RhoA, Cdc42, or Rac1/2 is enhanced in DC by LPS The Rho family of small GTPases plays a pivotal role in the dynamic regulation of vesicle transport through the remodeling of cytoskeletal components (Camera et al, 2003). We confirmed that cytoskeletal components were rearranged after LPS treatment in DC (West et al, 2004) (data not shown). We hypothesized that the TRIF-mediated pathway activates Rho family member(s), which are necessary for the alteration of cytoskeletal components to traffic intracellular MHCII+ lysosome vesicles toward the cell surface of DC after LPS stimulation. RhoB protein level was increased until 3 h after LPS stimulation (Figure 2A). Importantly, LPS-mediated RhoB expression and RhoB activity per cells dramatically increased 3 h after LPS stimulation (Figure 2A). As we wish to analyze the activity of each Rho family member per unit of RhoB protein, the samples were adjusted so that each lane contained the same amount of total RhoB, based on the results of Western blotting before the real part. We found that the RhoB-activity per unit was enhanced starting at 1.5 h and peaked 3 h at least until 6 h after LPS treatment in DC (Figure 2B and C). In contrast, we did not observe any increase in RhoA, Cdc42, or Rac1/2 activity per unit in DC (Figure 2B and C). The Cdc42-activity per unit gradually decreased after LPS treatment, as described by Mellman's group (Garrett et al, 2000; Figure 2B and C). We also analyzed the activity of RhoGTPases short time points, 5 and 15 min, after LPS stimulation and showed they did not significantly alter their activity within the short time periods (Supplementary Figure S4). Consistent with this result, Watts and co-workers demonstrated that LPS-mediated signal slightly increase Cdc42 but did not increase Rac1/2 activity short time periods (West et al, 2004). Moreover, as described above, we observed that RhoB mRNA and its product increased after LPS treatment (Figure 2A and data not shown); the results indicated the total RhoB activity in one dendritic cell dramatically increased after LPS treatment (see also Figure 2A, right panel). Together with the results shown in Figure 1, these data suggested that RhoB activation by LPS might play an important role in the upregulation of MHCII molecules in DC. Figure 2.RhoB activity is enhanced by the LPS–TRIF pathway in DC. (A) (Left panel) BMDC were prepared from wild-type mice and stimulated with LPS. The protein levels of RhoB was investigated. We put the same number of CD11c+ cells that were magnetically sorted in each lane. Western blotting for RhoB is performed and relative expression of RhoB is shown in the figure. We calculated 'time 0' as 1. Representative data from three independent experiments are shown. The relative amount of RhoB is shown with error bars indicating 1 s.d. The P-value (<0.05) was calculated by Student's t-test and is indicated by *. (Right panel) BMDC were prepared from wild-type mice and stimulated with LPS for 3 h. The protein levels of RhoB (activated or total) was investigated. The same number of CD11c+ cells that were magnetically sorted is applied in each lane. These experiments were performed at least three times independently, and representative data are shown. (B, C) BMDC were prepared from wild-type mice and stimulated with LPS for the indicated time period. CD11c cells were sorted and the activities of Rho family were investigated. The amount of total RhoB was adjusted to be the same in each lane in this experiment. The activities were compared with each other by defining the LPS-untreated control as 1. These experiments were performed three times independently, and representative data are shown. The relative activation level is shown with error bars indicating 1 s.d. The P-value (<0.05) was calculated by Student's t-test and is indicated by *. (D) BMDC were prepared from TRIFKO, Myd88KO, and control mice and stimulated with LPS for 6 h. CD11c cells were sorted and the activities of RhoB were investigated. The activities were compared with each other by defining the LPS-untreated control as 1. These experiments were performed three times independently, and representative data are shown. The relative activation level is shown with error bars indicating 1 s.d. The P-value (<0.05) was calculated by Student's t-test and is indicated by *. (E) BMDC were prepared from wild type and TRIFKO mice and stimulated with LPS. Real-time PCR of RhoB was performed. We showed relative expression of RhoB in the figure. We calculated 'time 0' as 1. Representative data from three independent experiments are shown. The relative amount of RhoB is shown with error bars indicating 1 s.d. The P-value (<0.01) was calculated by Student's t-test and is indicated by **. Download figure Download PowerPoint RhoB activation by LPS is dependent on the TRIF-mediated pathway in DC We next investigated whether the TRIF-dependent but Myd88-independent pathway enhanced the RhoB activity in DC after LPS stimulation. We prepared DC from TRIFKO, Myd88KO, and wild-type mice and stimulated them with LPS. We again analyzed the activity of RhoB per unit. RhoB activation per unit increased in the DC derived from MyD88KO but not TRIFKO mice after LPS stimulation (Figure 2D). We also confirmed that RhoA, Cdc42, or Rac1/2 activity did not increase in DC from TRIFKO mice after LPS treatment (data not shown). Moreover, we observed that LPS-mediated increase of RhoB mRNA and its product was dependent on TRIF-mediated signaling (Figure 2E and data not shown). All these results clearly showed the TRIF-mediated pathway of LPS signaling is important for RhoB activation in DC. RhoB colocalized with MHCII in DC The data described above suggested that RhoB might play a key role in the LPS-TRIF-mediated surface expression of MHCII molecules in DC. This hypothesis was supported by the observation that RhoB distribution was highly correlated with the localization of MHCII+ lysosomes either with or without LPS treatment (Figure 3A, see arrow heads). Moreover, a detailed colocalization analysis of MHCII and RhoB was performed with the profiles using TCS-SP2 and we confirmed a strong colocalization between MHCII and RhoB in LPS-treated DC (Figure 3B). We also counted BMDC after staining with the anti-RhoB antibody and showed a significant colocalization between RhoB and MHCII+ lysosome tubules after LPS treatment in the outside of the perinuclear area in DC (Figure 3B). In contrast, RhoA was mainly found in the cytoplasm, but partially colocalized with MHCII+ lysosomes regardless of LPS treatment (data not shown). All these data support the idea that RhoB is involved in a regulatory mechanism of the surface expression of MHCII on DC after LPS treatment. Figure 3.RhoB colocalized with MHCII+ lysosomes in DC. (A) BMDC were prepared from wild-type mice, incubated with or without LPS for 3 h, and fixed. The I-Ab and RhoB molecules were analyzed by confocal microscopy. Arrowheads indicate MHCII+ lysosomes colocalized with RhoB. Representative data from more than five independent experiments are shown. (B) A detailed colocalization analysis of MHCII and RhoB was performed with the profiles using TCS-SP2 (Leica) (upper graph). (C) The amount of DC that had RhoB molecules colocalized with MHCII+ lysosomes in the outside of the perinuclear area after LPS treatment was calculated by the confocal microscopy results (bottom graph). We analyzed total 50–100 MHCII+ cells in three independent experiments of day 6-BMDC cultures and shown with error bars indicating 1 s.d. Download figure Download PowerPoint A dominant negative form of RhoB (DN-RhoB) and RNAi of RhoB suppressed the LPS-mediated surface expression of MHCII molecules on DC To confirm the importance of RhoB activity for the surface expression of MHCII, we used a retroviral gene delivery system for DN-RhoB. The infection efficiency, as evaluated by Thy1.1 expression on the cell surface, was routinely 3–6% in CD11c+ cells (data not shown) and we used only Thy1.1+ cells, which were sorted by Moflo, in the analysis described below. DN-Rho molecules might be cytotoxic as reported previously (Gomez et al, 1997), but DN-RhoB expression in DC did not increase apoptotic cells (7AAD+ cells) (data not shown). Then, we analyzed the activity of RhoB per unit. DN-RhoB, but not a control vector (Mock), significantly inhibited the RhoB activity in DC after LPS treatment (more than 50% reduction of RhoB activity) (Figure 4A). As shown in Figure 4B, DN-RhoB but not DN-RhoA, or Mock infection significantly suppressed the LPS-mediated surface expression of MHCII molecules, indicating that RhoB, but not RhoA, is critical for the LPS-mediated surface expression of MHCII. This was consistent with the fact that LPS stimulation activated RhoB but not RhoA as shown in Figure 2B. Moreover, we observed that siRNA of RhoB significantly decreased both RhoB mRNA level and LPS-induced surface expression of MHCII on DC (Figure 4C and Supplementary Figure S5) like DN-RhoB transduction (Figure 4B). Thus, we concluded that effects of the RhoB's RNAi and the DN-RhoB are the same at least on MHCII movement in DCs. Figure 4.DN-RhoB expression and RNAi of RhoB treatment inhibited the LPS-mediated surface expression of MHCII in DC. BMDC were prepared from wild-type mice, infected with Thy1.1+ retrovirus carrying or lacking DN-RhoB or carrying DN-RhoA, and then incubated with or without LPS stimulation for 12 h. (A) The RhoB activity was evaluated. We calculated the increase amount of activated RhoB by intensity of the bands. We calculated 'the intensity increased after LPS stimulation in DC Mock-infected' as 100. The average relative activation level from three independent experiments is shown with error bars indicating 1 s.d. The P-value (<0.05) was calculated by Student's t-test and is indicated by *. (B) The resulting DC were processed by flow cytometer. The values on the profiles are the percentage of cells represented by the MHCIIhigh population in the gate for Thy1.1+CD11c+ cells. These experiments were performed five times independently, and representative data are shown. The average relative increase of MHCIIhigh population after LPS treatment in five independent experiments is shown in the figure with error bars indicating 1 s.d. The increases were compared with each other by defining the Mock-infected control as 1. The P-value (<0.05) was calculated by Student's t-test and is indicated by *. (C) BMDC were prepared from wild-type mice, treated them with RNAi for RhoB from 36 h before LPS treatement, and then incubated with or without LPS stimulation for 12 h. Surface expression of MHCII was analyzed in the presence or absence of LPS. We calculated the increase of MHCIIhigh population in each profile (see the bar). We calculated 'the percent increased after LPS stimulation in DC control RNAi treated ' as 100. The average relative activation level from three independent experiments is shown with error bars indicating 1 s.d. The P-value (<0.05) was calculated by Student's t-test and is indicated by *. (D) BMDCs were prepared from wild-type mice, infected with Thy1.1+ retrovirus carrying or lacking DN-RhoB, and then incubated with or without LPS stimulation for 12 h. The resulting DCs were processed by flow cytometer. These experiments were performed four times independently, and representative data are shown. Download figure Download PowerPoint Although we showed a transduction of constitutive active form (CA) RhoB in DC altered actin organization and MHCII+ vesicle positioning (Supplementary Figure S6), surface expression of MHCII molecules just slightly increased (Supplementary Figure S6), suggesting that another molecular mechanism in addition to activation of RhoB is necessary to induce full expression of MHCII on DC. The inhibitory effect of DN-RhoB was also observed in LPS-induced expression of CD86 but not CD80 and CD40, while transduction of DN-RhoA had no effect on these expressions (Figure 4D and data not shown). This result suggested that not only MHCII but also CD86 surface expression is regulated by RhoB molecules and, in other words, RhoB-mediated regulation is specific at least for MHCII and CD86 but not for CD80 and CD40 after LPS stimulation. GEFH1 associated and colocalized with RhoB in DC We next wished to identify molecules that might regulate RhoB in DC. We therefore investigated molecules that were associated with RhoB in DC. We focused on GEF, which usually have around 100 kDa, because Rho activity is mainly enhanced by GEFs (Rossman et al, 2005). We prepared a GST-fusion protein of the activated form of RhoB (GST-RhoB) and incubated it with the lysates of LPS-stimulated DC. A 100-kDa molecule was associated with GST-RhoB in the lysates from LPS-stimulated DC (see arrowhead, Supplementary Figure S7). MALDI-TOF mass analysis revealed that the 100-kDa molecule was GEFH1 (Ren et al, 1998). These results suggested that GEFH1 is associated with RhoB in DC. Additionally, a GST-fusion protein of the wild type of RhoB was also associated with GEFH1 (data not shown), suggesting GEFH1 associates with both active and inactive forms of Rho as described previously (Ren et al, 1998). We further investigated whether GEFH1 was colocalized with R
Referência(s)