VP1686, a Vibrio Type III Secretion Protein, Induces Toll-like Receptor-independent Apoptosis in Macrophage through NF-κB Inhibition

Vibrio parahaemolyticus, causative agent of human gastrointestinal diseases, possesses several virulent machineries including thermostable direct hemolysin and type III secretion systems (TTSS1 and -2). In this report, we establish that TTSS1-dependent secretion and translocation of a V. parahaemolyticus effector protein VP1686 into the cytosol induces DNA fragmentation in macrophages. We performed yeast two-hybrid screening to identify the molecules involved in VP1686-mediated cell death pathways and showed that nuclear factor RelA p65/NF-κB physically interacts with VP1686. To understand the impact of this interaction on the NF-κB DNA binding activities in infected macrophages, we analyzed a series of deletion mutants for the TTSS and its secreted proteins. Induction of DNA binding activity of NF-κB was significantly suppressed, and increased macrophage apoptosis has been associated with V. parahaemolyticus strain, which contains both VP1686 and TTSS1. Macrophages lacking Toll-like receptor adaptor molecules MyD88 (myeloid differentiation primary response protein 88) or TRIF (TIR domain-containing adapter-inducing interferon β) showed similar sensitivity to VP1686. As a consequence of NF-κB suppression, microarray analysis has revealed that VP1686 translocation alerted the expression of many genes that have known functions in cellular responses to apoptosis, cell growth, and transcriptional regulation. Our results suggest an important role for Vibrio effector protein VP1686 that activate a conserved apoptotic pathway in macrophages through suppression of NF-κB activation independent of Toll-like receptor signaling. Vibrio parahaemolyticus, causative agent of human gastrointestinal diseases, possesses several virulent machineries including thermostable direct hemolysin and type III secretion systems (TTSS1 and -2). In this report, we establish that TTSS1-dependent secretion and translocation of a V. parahaemolyticus effector protein VP1686 into the cytosol induces DNA fragmentation in macrophages. We performed yeast two-hybrid screening to identify the molecules involved in VP1686-mediated cell death pathways and showed that nuclear factor RelA p65/NF-κB physically interacts with VP1686. To understand the impact of this interaction on the NF-κB DNA binding activities in infected macrophages, we analyzed a series of deletion mutants for the TTSS and its secreted proteins. Induction of DNA binding activity of NF-κB was significantly suppressed, and increased macrophage apoptosis has been associated with V. parahaemolyticus strain, which contains both VP1686 and TTSS1. Macrophages lacking Toll-like receptor adaptor molecules MyD88 (myeloid differentiation primary response protein 88) or TRIF (TIR domain-containing adapter-inducing interferon β) showed similar sensitivity to VP1686. As a consequence of NF-κB suppression, microarray analysis has revealed that VP1686 translocation alerted the expression of many genes that have known functions in cellular responses to apoptosis, cell growth, and transcriptional regulation. Our results suggest an important role for Vibrio effector protein VP1686 that activate a conserved apoptotic pathway in macrophages through suppression of NF-κB activation independent of Toll-like receptor signaling. When encountered with bacterial pathogens, multicultural organisms raise a series of defense mechanism. To ensure the survival, pathogenic bacteria have also evolved sophisticated invasion strategies that involve neutralization of host defense through inhibition of innate immune system, phagocytosis, or induction of apoptosis in macrophages. A broad range of bacterial pathogens, including Gram-negative bacterium Vibrio parahaemolyticus, uses a type III secretion system (TTSS) 2The abbreviations used are: TTSS, type III secretion systems; MyD88, myeloid differentiation primary response protein 88; TRIF, TIR domain-containing adapter-inducing interferon β; LDH, lactate dehydrogenase; LPS, lipopolysaccharide; TLR, Toll-like receptors; PI, propidium iodide; FITC, fluorescein isothiocyanate; Fic, filamentation induced by cAMP. to deliver bacterial effector proteins into the host cell cytosol where they disrupt the signaling network (1Bhattacharjee R.N. Park K.S. Okada K. Kumagai Y. Uematsu S. Takeuchi O. Akira S. Iida T. Honda T. Biochem. Biophys. Res. Commun. 2005; 335: 328-334Crossref PubMed Scopus (19) Google Scholar, 3Mills S.D. Boland A. van der Sory M.P. Smissen P. Kerbourch C. Finlay B.B. Cornelis G.R. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 12638-12643Crossref PubMed Scopus (260) Google Scholar). The TTSS machinery is a needle-like structural complex formed by about 20 proteins, and its components are highly conserved among the bacterial species (4Cornelis G.R. Van Gijsegem F. Annu. Rev. Microbiol. 2000; 54: 735-774Crossref PubMed Scopus (648) Google Scholar). Recently genome sequencing of the clinical V. parahaemolyticus strain revealed that the strain has two gene clusters, TTSS1 and TTSS2, each encoding distinct type III secretion systems (5Makino K. Oshima K. Kurokawa K. Yokoyama K. Uda T. Tagomori K. Iijima Y. Najima M. Nakano M. Yamashita A. Kubota Y. Kimura S. Yasunaga T. Honda T. Shinagawa H. Hattori M. Iida T. Lancet. 2003; 361: 743-749Abstract Full Text Full Text PDF PubMed Scopus (777) Google Scholar). Of the two TTSS encoding gene clusters of V. parahaemolyticus strain RIMD2210633, TTSS1 is almost similar to those of Yersinia spp. and Pseudomonas aeruginosa in the number of genes, their order, and in the identity of each encoded protein. To date, four TTSS1-secreted effector proteins of V. parahaemolyticus (VP1686, VP1683, VP1680, and VPA450) have been identified by using a two-dimensional gel electrophoresis, but none of them showed any significant homology to known effector proteins of other bacteria within the TTSS1 region (6Ono T. Park K.S. Ueta M. Iida T. Honda T. Infect. Immun. 2006; 74: 1032-1042Crossref PubMed Scopus (117) Google Scholar). In vertebrates, the main function of the innate immune system is to recognize the presence of pathogen-associated molecular patterns on invading microbes and initiate downstream signal from Toll-like receptors (TLRs), which lead to the expression of inflammatory response-related genes (7Takeda K. Kaisho T. Akira S. Annu. Rev. Immunol. 2003; 21: 335-376Crossref PubMed Scopus (4767) Google Scholar). The transcription factor nuclear factor κB (NF-κB) is a critical mediator of TLR signaling that controls the synthesis of cytokines, adhesion molecules, and other anti-apoptotic factors such as inhibitor of apoptosis protein, tumor necrosis factor receptor-associated factor, and Bcl-2 families to ensure cellular survival by prevention of cell death (8Akira S. Takeda K. Nat. Rev. Immunol. 2004; 4: 499-511Crossref PubMed Scopus (6743) Google Scholar, 11Chen C. Edelstein L.C. Gélinas C. Mol. Cell. Biol. 2000; 20: 2687-2695Crossref PubMed Scopus (698) Google Scholar). Up-regulation of antiapoptotic protein synthesis through NF-κB activation is also essential for host survival under versatile stress-induced conditions such as bacteria-faced macrophages. Thus, it is plausible, given the function of NF-κB to cell survival, that improper activities of this protein may also be involved in bacteria-induced macrophage apoptosis. Our previous study on DNA fragmentation patterns in HCT116 cells revealed that the induction of apoptosis by V. parahaemolyticus in these cells requires functional TTSS1 (1Bhattacharjee R.N. Park K.S. Okada K. Kumagai Y. Uematsu S. Takeuchi O. Akira S. Iida T. Honda T. Biochem. Biophys. Res. Commun. 2005; 335: 328-334Crossref PubMed Scopus (19) Google Scholar). But the mechanism by which TTSS1 induces apoptosis in HCT116 cells remains poorly defined. In this study we show that VP1686 is an effector protein of V. parahaemolyticus injected by type III secretion system 1 into the cytoplasm and induces apoptosis in both cultured and thioglycollate-elicited peritoneal macrophages. VP1686 binds with RelA p65/NF-κB in yeast and inhibits the DNA binding activity of NF-κB in macrophages. On the basis of these results, we suggest that inhibition of NF-κB is sufficient to sensitize infected-macrophages to death by preventing induction of NF-κB targeted gene expression related to antiapoptotic function. Bacterial Strains and Culture Conditions—All the V. parahaemolyticus and Vibrio species strains used in this study were obtained from the Laboratory for Culture Collection, Research Institute for Microbial Diseases, Osaka University. V. parahaemolyticus strain RIMD2210633 (KP-positive, serotype O3:K6) was used for the construction of deletion mutants and for functional studies. Escherichia coli DH5α and SM10λpir (12Miller V.L. Mekalanos J.J. J. Bacteriol. 1988; 170: 2575-2583Crossref PubMed Scopus (1715) Google Scholar) strains were used for the general manipulation of plasmids and the mobilization of plasmids into V. parahaemolyticus, respectively. The bacteria were cultured at 37 °C with shaking in Luria-Bertani (LB) medium (for E. coli) and LB supplemented with 3% NaCl (for V. parahaemolyticus). Thiosulfate citrate bile salts sucrose agar (Nissui, Tokyo, Japan) was used for the screening of mutant strains. Antibiotics were used at the following concentrations: ampicillin (100 μg/ml), kanamycin (50 μg/ml), and chloramphenicol (5 μg/ml). Cell Lines—RAW264.7 cells were grown as monolayer in Dulbecco's modified Eagle's medium (Nissui, Tokyo, Japan) supplemented with 10% fetal bovine serum, 100 μg/ml streptomycin, and 10 units/ml penicillin G. Mouse peritoneal macrophages were collected by peritoneal lavage with Hanks' balanced salt solution at 3 days after intraperitoneal injection of 2 ml of 4% sterile thioglycollate into 8-12-week-old mice. Peritoneal macrophages were cultured in RPMI 1640 medium with 10% fetal bovine serum, 100 μg/ml streptomycin, and 10 units/ml penicillin G. VP1686 Complementation—The VP1686 gene (operon) containing a putative promoter region from the KXV-237 (RIMD2210633) strain was amplified by PCR using the following oligonucleotide primers 5′-GAGTAGGGGATCCCCGCCAAA-3′ and 5′-AATACCAACGTCGACAATCAC-3′. The amplified fragment was cloned into a pT7blue T vector and digested with BamHI and SalI. The digested fragment was then cloned into the pSA19Cm-MCS (13Park K.S. Arita M. Iida T. Honda T. Infect. Immun. 2005; 73: 5754-5761Crossref PubMed Scopus (22) Google Scholar). The plasmid was introduced into KX-V237 VP1686 deletion mutant by electroporation (1.5 kV, 1000 ohms, 25 microfarads). Infections of MyD88-/-, TRIF-/- Macrophages—MyD88 and TRIF knock-out mouse and Balb/c mice were injected with 4% thioglycollate 3 days before harvesting peritoneal macrophages. Approximately 2 × 105 peritoneal macrophages were seeded into a 10-mm dish and infected as described above. DNA Fragmentation—Both fragmented DNA and high molecular weight intact genomic DNA were extracted from 2 × 105 cells using the suicide-Track™ DNA ladder isolation kit (Oncogene, Madison, WI). 1.5% agarose gel electrophoresis was followed by ethidium bromide staining. Annexin V-FITC Apoptosis Detection Assay—The cells were harvested and infected with V. parahaemolyticus as described above and washed three times with phosphate-buffered saline. Staining was carried out using the annexin V-FITC apoptosis detection kit (BioVision, Mountain View, CA). Briefly, 2 × 105 cells were resuspended in 1× binding buffer and incubated with annexin V-FITC and propidium iodide (PI) for 5 min in darkness at room temperature. Annexin V binding was analyzed by FACScan cytometer (BD Biosciences) equipped with a FITC signal detector FL1 (excitation = 488 nm, green) and PI staining by the phycoerythrin emission signal detector FL2 (excitation = 585 nm, red). The percentage of apoptotic cells was calculated from the total (104 cells) using FlowJo 4.5.2 software. Cytotoxicity assays—RAW 264.7 macrophages (2×104 cells) were seeded onto a 96-well plate and incubated overnight at 37 °C. Before infection with the bacteria, cells were washed with phosphate-buffered saline, pH 7.2, and further incubated with Dulbecco's modified Eagle's medium without phenol red and antibiotics. The release of lactate dehydrogenase (LDH) into the medium was assayed using the CytoTox96 nonradioactive cytotoxicity kit (Promega) according to the manufacturer's instructions. At 4 h post-infection with bacteria, the supernatants were collected, and the release of LDH was quantified. LDH release (% cytotoxicity) was calculated using the equation ((A490 experimental release - A490 spontaneous release)/(A490 maximum release - A490 spontaneous release)) × 100. The spontaneous release is the amount of LDH released from the cytoplasm of uninfected cells, whereas the maximum release is the amount released by total lysates of uninfected cells by Triton X-100. Yeast Two-hybrid Screening—Yeast two-hybrid screening was performed as described previously (14Kawai T. Sato S. Ishii K.J. Coban C. Hemmi H. Yamamoto M. Terai K. Matsuda M. Inoue J. Uematsu S. Takeuchi O. Akira S. Nat. Immunol. 2004; 5: 1061-1068Crossref PubMed Scopus (834) Google Scholar). For the construction of bait plasmid, full-length VP1686 gene was cloned in-frame into GAL4 DNA binding domain of pGBKT7. Immunoblotting—Cells were infected with or without V. parahaemolyticus were washed with cold phosphate-buffered saline and pelleted. The cells were lysed in buffer X + bovine serum albumin (BSA; 100 mm Tris-HCl, pH 8.5, 250 mm NaCl, 1% (v/v) Nonidet P-40, 1 mm EDTA, 1 μg/ml aprotinin, 2 mg/ml BSA). Whole cell protein lysates were solubilized in loading buffer, subjected to SDS-PAGE, and transferred to nitrocellulose followed by incubation with the antibodies as mentioned in the figure legends. Electrophoretic Mobility Shift Assay—RAW macrophages (1×106) were stimulated with 50 ng/ml lipopolysaccharide (LPS) or infected with V. parahaemolyticus for 2 h. Nuclear proteins were extracted and then incubated with an end-labeled, double-stranded oligonucleotide containing a NF-κB binding site of the tumor necrosis factor α promoter in 25 μlof binding buffer (10 mm HEPES-KOH, pH 7.8, 50 mm KCl, 1 mm EDTA EDTA, pH 8.0, 5 mm MgCl2, and 10% glycerol) for 20 min at room temperature and loaded on a native 5% polyacrylamide gel. The DNA-protein complex was visualized by autoradiography. The specificities of the shifted band were confirmed by adding antibodies specific for p65 NF-κB (Santa Cruz Biotechnology, Santa Cruz, CA). Microarray Gene Chip Analysis—RAW cells were infected with TTSS deletion mutants of V. parahaemolyticus (m.o.i. 2) for 2 h. Total RNA was extracted (RNeasy kit, Qiagen), and double-stranded cDNA was synthesized from 5 μg of total RNA using the Superscript system (Invitrogen) primed with a T7-(dT)24 primer. To prepare biotin-labeled cRNA from this cDNA, an in vitro transcription reaction was performed in the presence of T7 RNA polymerase and biotinylated ribonucleotides (Enzo Diagnostics). The cRNA product was purified (RNeasy kit), fragmented, and hybridized to a mouse genome 430 2.0 gene array chip as per the manufacturer's instruction (Affymetrix). The chips were washed and scanned with a Gene-Array scanner (Affymetrix). The color intensity of gene expression was generated with the R (Version 2.2.1) and Bioconductor software. Identification of VP1686—Recent genome sequencing of the clinical V. parahaemolyticus strain RIMD2210633 identified two sets of genes for the type III secretion system (TTSS), TTSS1 and TTSS2 (5Makino K. Oshima K. Kurokawa K. Yokoyama K. Uda T. Tagomori K. Iijima Y. Najima M. Nakano M. Yamashita A. Kubota Y. Kimura S. Yasunaga T. Honda T. Shinagawa H. Hattori M. Iida T. Lancet. 2003; 361: 743-749Abstract Full Text Full Text PDF PubMed Scopus (777) Google Scholar). Previously we have shown that the cytotoxic activity of mutant strains having a deletion in the TTSS1 gene was significantly decreased as compared with that of the parent and TTSS2-mutant strains (1Bhattacharjee R.N. Park K.S. Okada K. Kumagai Y. Uematsu S. Takeuchi O. Akira S. Iida T. Honda T. Biochem. Biophys. Res. Commun. 2005; 335: 328-334Crossref PubMed Scopus (19) Google Scholar). To understand the role of TTSS in bacterial pathogenesis and in immunity, we infected macrophage cell line RAW264.7 with the deletion mutants for TTSS1 or TTSS2 for 4 h at the multiplicity of infection (m.o.i. 2). By Western blot analysis, we have confirmed the secretion and subsequent translocation of a bacterial effector protein called VP1686 in both culture medium and in RAW cell extract upon infection by the parent and TTSS2-deleted mutant strains of V. parahaemolyticus. However, no VP1686 was detected when macrophages were infected with the bacterial strain that is lacking TTSS1 (Fig. 1A). The data suggest that VP1686 was secreted and injected inside macrophages in a TTSS1-dependent manner. Previously several TTSS1-dependent secretion proteins (VP1680, VPA450) including VP1686 were also identified in bacterial culture supernatant by two-dimensional gel electrophoresis (6Ono T. Park K.S. Ueta M. Iida T. Honda T. Infect. Immun. 2006; 74: 1032-1042Crossref PubMed Scopus (117) Google Scholar). Using the proposed amino acid sequence of VP1686 (5Makino K. Oshima K. Kurokawa K. Yokoyama K. Uda T. Tagomori K. Iijima Y. Najima M. Nakano M. Yamashita A. Kubota Y. Kimura S. Yasunaga T. Honda T. Shinagawa H. Hattori M. Iida T. Lancet. 2003; 361: 743-749Abstract Full Text Full Text PDF PubMed Scopus (777) Google Scholar), our conserved domain search using protein-protein BLAST detected that VP1686 belongs to the family of proteins called Fic (filamentation induced by cAMP) (Fig. 1, B and C). This family contains a central conserved motif HPFXXGNG in most of its diverse members and is suggested to be involved in regulatory mechanism of cell division, although exact molecular function of these proteins is still uncertain (Fig. 1D). Induction of DNA Fragmentation in V. parahaemolyticus-infected macrophages is TLR-independent and Requires Both TTSS1 and VP1686—In our previous studies, we have shown that one of the characteristic effects of TTSS1 is cytotoxicity to eukaryotic cells (1Bhattacharjee R.N. Park K.S. Okada K. Kumagai Y. Uematsu S. Takeuchi O. Akira S. Iida T. Honda T. Biochem. Biophys. Res. Commun. 2005; 335: 328-334Crossref PubMed Scopus (19) Google Scholar, 15Park K.S. Ono T. Rokuda M. Jang M.H. Iida T. Honda T. Microbiol. Immunol. 2004; 48: 313-318Crossref PubMed Scopus (102) Google Scholar). We performed DNA fragmentation and annexin V staining assays of macrophages infected with mutant V. parahaemolyticus strain carrying deleted genes that encode TTSS system 1 and 2 as well as individual secretion proteins to determine whether any of them play a role in initiating programmed cell death in macrophages. Primarily, mutant strains lacking TTSS1, TTSS2, and each of the three proteins encoded by VP1686, VP1680, VPA450 were exposed to RAW macrophages (m.o.i. 2) (Table 1). Total DNA (genomic and fragmented) was prepared from infected cells after 4 h. Electrophoretic patterns indicate the presence of oligonucleosomal length DNA fragmentation (≤190 bp) in cells infected with TTSS2-deleted strain but not in uninfected cells or cells infected with TTSS1-deletion mutants (Fig. 2, A and B). Neither heat-killed bacteria nor bacteria-free cultural supernatant of 4-h post-inoculation were able to induce DNA fragmentation in RAW cells (data not shown). These results indicate that intact live bacteria and its TTSS1 components (not TTSS2) are involved in the apoptosis of macrophages and, thus, the cytotoxic activity of the parental strain.TABLE 1Bacterial strains used in this study tdh, thermostable direct hemolysis.Strain/plasmidGenotype or relevant phenotypeSourceV. parahaemolyticus strainsRIMD2210633Clinical isolate;tdhA+,tdhS+25POR-1Both tdh(AS) deleted from RIMD221063315ΔTTSS1POR-1ΔvcrD-12ΔTTSS2POR-1ΔvcrD-22VPΔ1686POR-1ΔVP16866VPΔ1680POR-1ΔVP16806VPAΔ450POR-1ΔVPA4506VP1686 complimentedPOR-1ΔVP1686 + VP1686This study Open table in a new tab To test whether the phenotype described above resulted from the specific function exerted by TTSS1, RAW cells were infected with TTSS1-containing but TTSS2-deleted mutants of V. parahaemolyticus for 1.5 h. No DNA fragmentation was seen in these cells in this stage. We then washed the cells extensively to remove the extracellular bacteria or kill bacteria by incubating cells in fresh medium with 100 μg/ml gentamycin for 2, 3, and 6 h. In the absence of any surrounding live bacteria, apoptotic DNA fragmentation started to appear in cells 3 h after gentamycin treatment, indicating that the removal/killing of extracellular bacteria was unable to halt the signaling events initiated in these cells in the first 1.5 h (Fig. 2C). We hypothesize that once V. parahaemolyticus injects VP1686 inside the host cell cytoplasm, subsequent action of VP1686 were irreversible. With the help of Western blot analysis in an earlier experiment, we have indeed confirmed the presence of TTSS1-dependent VP1686 translocation in apoptotic RAW whole cell extract and in cell culture supernatant (Fig. 1A). Several previous studies demonstrated that activated TLRs could act as a potent inducer of apoptosis in macrophages that encounter a bacterial pathogen (16Hsu L.C. Park J.M. Zhang K. Luo J.L. Maeda S. Kaufman R.J. Eckmann L. Guiney D.G. Karin M. Nature. 2004; 428: 341-345Crossref PubMed Scopus (320) Google Scholar, 17Haase R Kirschning C.J. Sing A. Schrottner P. Fukase K. Kusumoto S. Wagner H. Heesemann J. Ruckdeschel K. J. Immunol. 2003; 171: 4294-4303Crossref PubMed Scopus (116) Google Scholar). To determine whether TLR signaling could play a role in the mechanism of apoptosis induction by V. parahaemolyticus we investigated DNA fragmentation pattern in primary mouse peritoneal macrophages that are deficient for functional TLR adapters MyD88 (myeloid differentiation primary response protein 88) and TRIF (TIR domain-containing adapter inducing interferon β). MyD88 is a central adapter shared by almost all TLRs (18Takeuchi O. Akira S. Curr. Top. Microbiol. Immunol. 2002; 270: 155-156Crossref PubMed Scopus (131) Google Scholar), and TRIF is used in TLR4 signaling activated by bacterial LPS independent of MyD88 (19Yamamoto M. Sato S. Hemmi H. Hoshino K. Kaisho T. Sanjo H. Takeuchi O. Sugiyama M. Okabe M. Takeda K. Akira S. Science. 2003; 301: 640-643Crossref PubMed Scopus (2527) Google Scholar). Surprisingly, both MyD88-/- and TRIF-/- macrophages underwent apoptosis upon infection with TTSS1-containing mutant (Fig. 2D) Thus, signaling events originated from TLR due to bacterial recognition is not essential for VP1686-induced macrophage apoptosis. Annexin V Binding Analysis Shows VP1686-mediated Induction of Apoptosis—To support and extend the DNA fragmentation experiments, we performed double-staining using annexin V-FITC and PI followed by flow cytometry. Fig. 3 shows that in RAW cells, when infected with mutant strains that lack either TTSS1 or its effector VP1686, only 7.6 and 3.9% cells were stained positive (right-bottom quadrant) for phosphatidylserine, a marker of early stages of apoptosis. Almost an equal number of the apoptotic cells were stained for asynchronously grown uninfected cells (5.7%), which could be the cells committing natural cell death without any external induction. Interestingly, a considerable increase in apoptotic cells was found when macrophages were infected with the mutant strains lacking TTSS2, VP1680, or VPA450, but TTSS1 and VP1686 were retained (39.6, 37.8, and 26.2%, respectively). These results strongly suggest that the remaining secretory components such as VP1686 and its injection machinery TTSS1 in the parental strain might be the principal apoptosis inducer in macrophages. Complementation of VP1686 in VP1686-deleted strain Reverses Cytotoxicity—In an attempt to reveal whether Vibrio-induced apoptosis is specific to VP1686 or a more general phenomenon in apoptosis provocation, we complemented the gene for VP1686 in VP1686-deleted mutant strain because this strain showed diminished cytotoxic potential. By assaying LDH release from the RAW macrophages, we see the changes in reversion of its cell death ability due to VP1686 complementation. LDH, a stable cytosolic enzyme, was shown to be released upon cell lysis during the later stages of apoptosis as well as in early stages of necrosis (20Bhattacharjee R.N. Park K.S. Uematsu S. Okada K. Hoshino K. Takeda K. Takeuchi O. Akira S. Iida T. Honda T. FEBS Lett. 2005; 579: 6604-6610Crossref PubMed Scopus (25) Google Scholar). To confirm the successful genetic complementation, we detected the expression of VP1686 protein by Western blot analysis in previously VP1686 non-producing strains (Fig. 4A). Consistent with VP1686 carrying wild type strains, as shown in Fig. 4B, the wild type and the VP1686-complimented strains showed increased and almost indistinguishable levels of macrophage cell death (75 and 85%, respectively) after 4 h of infection (m.o.i. 2). The LDH experiment also supports our existing annexin V staining result that both the TTSS1-deleted mutant and VP1686-deleted mutants showed considerable decreases in cellular cytotoxicity (20-25%). Interestingly, macrophages when incubated with LPS for 8 h (50 ng/ml) did not show any increase in cell death. A restored level of LDH release attributed to VP1686-complemented strains prompted us to conclude that this protein plays a specific role in macrophage cell death. VP1686 Interacts with NF-κB p65 in Yeast and Suppresses DNA Binding Activity of NF-κB in Macrophages—Unlike several other bacterial infection-induced apoptosis in macrophages, the mechanism of VP1686-induced apoptosis was not dependent on cell cycle arrest, caspase-1 activity (data not shown), and TLR-mediated signaling. In an effort to understand the mechanism of VP1686-induced apoptosis in macrophages, we employed the yeast two-hybrid method to identify VP1686-interacting protein(s) by using full-length VP1686 as bait. From a screen of ∼4×105 yeast transformants, 14 cDNA clones scored positive for reporter gene activities. Sequence analysis revealed that two of these clones encoded a similar portion of NF-κB. To further explore this interaction, full-length VP1686, full length of another TTSS1-dependent secretor protein VP1680, the N terminus of VP1686, and the C terminus of VP1686 were studied in the yeast two-hybrid systems. Fig. 5A showed that NF-κB interacted with the full-length VP1686 as well as with the N terminus of VP1686 but not with the VP1680 nor with the C terminus of VP1686. The result indicates that the well conserved Fic domain in VP1686 is not necessary for this interaction. NF-κB is an important transcriptional regulator of inducible expression of numerous genes involved in apoptosis, inflammation, and innate immune response. To investigate whether V. parahaemolyticus induces NF-κB DNA binding activity consistent with Entero-pathogenic Eschericia coli, Yersinia, and Salmonella, RAW cells were challenged with V. parahaemolyticus for 2 h. Nuclear extracts were prepared and subjected to electrophoretic mobility shift assays using consensus oligonucleotide NF-κB probe. Uninfected cells served as a negative control, whereas macrophage incubated with LPS for 2 h was used as a positive control. Fig. 5B shows that the VP1686 inhibits DNA binding activity of the NF-κB because as a result of TTSS1 deletion from the strain that no longer can secret VP1686, NF-κB was able to bind with DNA in macrophages. As expected, NF-κB activation was severely impaired in macrophages infected with TTSS2-deleted strain, which contains functional TTSS1. Similar to TTSS1 mutant, VP1686-deleted strain also failed to block NF-κB activation. In addition, we have found that upon re-introduction of the VP1686 gene in the VP1686-deleted strain by genetic complementation, the capability to suppress activation of NF-κB was restored. The specificity of the band was confirmed by a shift by treatment with anti-p65 antibodies (Fig. 5C). Together, the analysis of the different mutant Vibrio strains indicates that VP1686 impairs NF-κB activation and subsequently mediates apoptosis in macrophages. Differential Gene Expression Pattern in RAW Macrophage Infected with TTSS1 and TTSS1 Deletion Mutants—The transcription factor NF-κB is involved in dozens of signaling pathways regulating many aspect of cellular activities such as cell growth and differentiation, apoptosis, stress, immune response, inflammation, and adhesion. To determine the effect of VP1686-induced NF-κB suppression on NF-κB-mediated gene expression, we infected macrophages with the mutant strains containing or lacking TTSS1 (at m.o.i. 2 for 2 h) and employed microarray technology. Using this technology, the expression pattern of more than 34,000 genes was analyzed and compared with un-infected control. Data analysis was focused on genes that were altered with more than a 2-fold change. There were about 235 genes that were differentially expressed due to the inactivity of NF-κB which are presented as a hierarchical clustering in Fig. 6 for the genes which appeared in 7 different clusters. These genes belong to different biological processes, such as apoptosis, cell death, response to stress, and transcriptional regulation. Expression profile of selected genes and specific pathways in which the genes are involved were classified by gene ontology (Table 2 and 3).TABLE 2Signaling pathways modulated in macrophages infected with TTSS1 mutantsGene nameGenBank™ accession no.ΔTT1/ConΔTT2/ConΔTT2/ΔTT1Signaling pathwaysRgs1NM_0158112.5433.7513.3G-protein-coupled receptor protein signalingOpn3NM_0100987.3810.461.4G-protein-coupled receptor protein signalingKlf6NM_0118031.474.192.9Cytokine- and chemokine-mediated signalingCsnk2a1BB2837592.454.131.7Wnt receptor signaling pathwayRgs20.583.225.5G-protein/cell cycle signal transductionCd47NM_0105813.052.510.8Integrin-mediated signaling pathwayCcr1AV2316482.5019.707.9Myeloid cell differentiation signaling pathwayEntpd1BI15144014.506.750.5Platelet activation signaling pathwayPtger3D174060.230.271.2Cytosolic calcium ion regulation pathwayJag2AV2646810.420.210.5Cell communication/Notch signaling pathwayGria3BM2205765.546.911.2Potassium ion transport pathwayAkt1s1BF0198393.724.401.2Notch signaling pathwayLtbp3BB3248230.190.100.5Transforming growth factor-α receptor signaling pathway Open table in a new tab TABLE 3Functional categorization of differentially expressed genes in macrophages infected with TTSS1 mutantsGene nameGenBank™ accession no.ΔTT1/conΔTT2/conΔTT2/ΔTT1ApoptosisBnip3lAK0186681.33.22.4PycardBG0842300.560.260.46Bcl2aL164626.135.350.87Pdcd2BI5261953.154.061.29Casp3D8635211.8519.461.64BfarAK0138742.193.931.80Dapk1BC0266710.650.270.41Dnaja3AK0078520.370.270.74Als2cr2BB2779120.300.592.01Igf1BG0926771.6011.857.41Irak3AV2284930.510.320.63Mcl1AV3184944.8725.925.32Tnfrsf1bM604696.236.241.00Bhlhb4NM_0806410.600.190.32Cell growth and differentiationEmp1U256331.033.933.84Ifrd1NM_0135622.864.261.49Egr1NM_0079132.324.161.79Ccr1AV2316482.5019.707.88Smap1BC0069461.534.733.09Rtel1BG0710280.300.602.01Regulation of transcriptionCdkn2cBC0270260.380.190.50RxrbBC0194320.281.093.84Rnf12NM_0139161.503.592.39Klf4BG0694130.726.529.00NfkbizAB02655123.7523.550.99Bcl3NM_0336013.285.141.57Irf2NM_0083910.480.210.44Ddx5NM_0078401.633.532.17Runx3NM_01973255.4376.331.38Hmgb1AI6487591.104.193.81ArntAV2337930.700.250.36Ets2AV2967030.300.752.53 Open table in a new tab We have identified the secreted protein VP1686 as the effector molecule responsible for the induction of apoptosis in macrophages infected by V. parahaemolyticus and that the effect depends on TTSS1. VP1686 belongs to the Fic protein family. This family contains a central conserved motif in most diverse members including Caenorhabditis elegans and Drosophila melanogaster and suggested to be involved in the regulatory mechanism of cell division. To date, the exact molecular function of these proteins is unknown. However, we have first identified VP1686 as an inducer of apoptosis in macrophages. A previous report from our group has shown that protein VP1680 of V. parahaemolyticus, also secreted via TTSS1, is associated with HeLa cell toxicity (6Ono T. Park K.S. Ueta M. Iida T. Honda T. Infect. Immun. 2006; 74: 1032-1042Crossref PubMed Scopus (117) Google Scholar). In this study we confirmed that the deletion mutant of VP1686 but not VP1680 fails to elicit apoptosis in macrophages. Therefore, VP1686 action may be specific to macrophages. Macrophages are essential components of the innate immune system, vital for recognition and elimination of microbial pathogens. Macrophages use TLRs to identify common pattern of pathogens, such as LPS, and in turn activate intracellular signaling pathways related to inflammation, immunity, and pro-apoptosis as well as anti-apoptosis (21Medzhitov R. Nat. Rev. Immunol. 2001; 1: 135-145Crossref PubMed Scopus (3279) Google Scholar, 22Akira S. Uematsu S. Takeuchi O. Cell. 2006; 124: 783-801Abstract Full Text Full Text PDF PubMed Scopus (8883) Google Scholar). Numerous studies indicate that TLR- and MyD88-mediated signaling play an essential role in the initiation of apoptosis in bacteria-faced macrophages. Several reports showed that macrophage apoptosis by either Gram-positive (Bacillus anthracis) or Gram-negative (Yersinia, Salmonella) pathogens required activation of TLR4 by LPS (16Hsu L.C. Park J.M. Zhang K. Luo J.L. Maeda S. Kaufman R.J. Eckmann L. Guiney D.G. Karin M. Nature. 2004; 428: 341-345Crossref PubMed Scopus (320) Google Scholar). In the case of Shigella-, Salmonella-, and Yersinia-induced cytotoxicity of macrophages, type III secretion system was also required (1Bhattacharjee R.N. Park K.S. Okada K. Kumagai Y. Uematsu S. Takeuchi O. Akira S. Iida T. Honda T. Biochem. Biophys. Res. Commun. 2005; 335: 328-334Crossref PubMed Scopus (19) Google Scholar, 21Medzhitov R. Nat. Rev. Immunol. 2001; 1: 135-145Crossref PubMed Scopus (3279) Google Scholar, 23Ruckdeschel K. Mannel O. Richter K. Jacobi C.A. Trulzsch K. Rouot B. Heesemann J. J. Immunol. 2001; 166: 1823-1831Crossref PubMed Scopus (129) Google Scholar). We have shown here that macrophages from mice lacking MyD88 and TRIF, two signaling adapter proteins that act downstream of TLR4 (MyD88 is shared by almost all TLRs), are equally sensitive to apoptosis induction by V. parahaemolyticus infection and conclude that VP1686 action does not require TLRs or LPS. In earlier reports using different infection models there was increasing evidence that NF-κB activation is important for self-defense and survival of macrophages when encountered with bacteria (16Hsu L.C. Park J.M. Zhang K. Luo J.L. Maeda S. Kaufman R.J. Eckmann L. Guiney D.G. Karin M. Nature. 2004; 428: 341-345Crossref PubMed Scopus (320) Google Scholar, 21Medzhitov R. Nat. Rev. Immunol. 2001; 1: 135-145Crossref PubMed Scopus (3279) Google Scholar). The NF-κB system also plays a central role in innate immunity that systematically detects and eliminates microbial pathogens by TLR-mediated gene expression. Therefore, it is conceivable that to establish pathogenic action in such a hostile environment bacteria require the delivery of a unique virulent mechanism(s). Triggering the activation of proapoptotic signals, hindering their cytotoxic effects by the antiapoptotic activity of the host, such as mediated by the NF-κB.A in a series of studies, suggests that Yersinia type III secretion machinery injects YopP/YopJ protein into the macrophage, where it binds and inhibits the NF-κB-activating inhibitory κβ kinase, leading to down-regulation of NF-κB activation (17Haase R Kirschning C.J. Sing A. Schrottner P. Fukase K. Kusumoto S. Wagner H. Heesemann J. Ruckdeschel K. J. Immunol. 2003; 171: 4294-4303Crossref PubMed Scopus (116) Google Scholar, 22Akira S. Uematsu S. Takeuchi O. Cell. 2006; 124: 783-801Abstract Full Text Full Text PDF PubMed Scopus (8883) Google Scholar). DNA binding of NF-κB was also actively suppressed in apoptotic macrophages by viable E. coli-secreted and -translocated effector protein B (Esp-B) (24Hauf N. Chakraborty T. J. Immunol. 2003; 170: 2074-2082Crossref PubMed Scopus (72) Google Scholar). Our data are consistent with the notion that translocated Vibrio effector protein VP1686 shares common property of Yersinia, Escherichia, or Shigella to mediate suppression of both basal and signal induced NF-κB activity through interference with NF-κB signaling. Unlike Yersinia or Escherichia, prior activation of macrophages by LPS was not required for VP1686-induced apoptosis. Gene expression studies have revealed differential expression of more than 235 genes between TTSS1-containing or TTSS1-lacking V. parahaemolyticus infection of macrophages. Genes that are up-regulated or down-regulated after infection include genes that participate in diverse biological processes, such as cell death, cell adhesion, signal transduction, response to stress, cell growth and/or maintenance, and transcriptional regulation (Tables 2 and 3). More importantly, about 15 different signaling pathways and 29 genes involved in apoptotic and cell growth pathways were altered in their expression profiles. In summary, this study provides new insights into the mechanism by which V. parahaemolyticus triggers apoptosis in macrophages. By injection of VP1686, Vibrio affects the signaling networks of a highly conserved host defense system mediated by NF-κB. Although TLR represents a key inducer of NF-κB pathway, cytosolic action of VP1686 in macrophages may disrupt NF-κB activities directly without the signaling dependence from TLRs. Further investigation is required to understand the role of specific NF-κB target molecule(s) in the apoptotic pathway induced by V. parahaemolyticus. We thank Akira laboratory members for technical support and Dr. Manoor P. Hande for critical reading of the manuscript.