Association of Bordetella dermonecrotic toxin with the extracellular matrix
© Fukui-Miyazaki et al; licensee BioMed Central Ltd. 2010
Received: 20 May 2010
Accepted: 25 September 2010
Published: 25 September 2010
Bordetella dermonecrotic toxin (DNT) causes the turbinate atrophy in swine atrophic rhinitis, caused by a Bordetella bronchiseptica infection of pigs, by inhibiting osteoblastic differentiation. The toxin is not actively secreted from the bacteria, and is presumed to be present in only small amounts in infected areas. How such small amounts can affect target tissues is unknown.
Fluorescence microscopy revealed that DNT associated with a fibrillar structure developed on cultured cells. A cellular component cross-linked with DNT conjugated with a cross-linker was identified as fibronectin by mass spectrometry. Colocalization of the fibronectin network on the cells with DNT was also observed by fluorescence microscope. Several lines of evidence suggested that DNT interacts with fibronectin not directly, but through another cellular component that remains to be identified. The colocalization was observed in not only DNT-sensitive cells but also insensitive cells, indicating that the fibronectin network neither serves as a receptor for the toxin nor is involved in the intoxicating procedures. The fibronectin network-associated toxin was easily liberated when the concentration of toxin in the local environment decreased, and was still active.
Components in the extracellular matrix are known to regulate activities of various growth factors by binding and liberating them in response to alterations in the extracellular environment. Similarly, the fibronectin-based extracellular matrix may function as a temporary storage system for DNT, enabling small amounts of the toxin to efficiently affect target tissues or cells.
Pathogenic bacteria of the genus Bordetella produce dermonecrotic toxin (DNT), which activates Rho GTPases through its transglutaminase activity resulting in deamidation or polyamination [1–3]. DNT is a single chain polypeptide of 1,464 amino acids, with an N-terminal region of at least 54 amino acids responsible for binding to a receptor on target cells  and a C-terminal region of about 300 amino acids conferring the transglutaminase activity . The receptor for DNT is still unknown. The activated Rho GTPases cause aberrant Rho-dependent phenotypes [6, 7], which likely lead to some of the pathological changes observed during Bordetella infections. For example, the turbinate atrophy in atrophic rhinitis, a Bordetella infection of pigs, is caused by DNT acting on osteoblastic cells [8–13]. However, there has been no evidence that DNT is actively secreted from the bacteria, and less than 0.75% (0.60 ng/109 CFU) of produced DNT was detected in culture supernatant of B. bronchiseptica and B. pertussis (unpublished data). It is unknown how this small amount of DNT exerts toxicity against target cells such as osteoblasts covered by epithelial cells and connective tissue.
While attempting to identify the receptor for DNT, we found that DNT associated temporarily with fibronectin (FN)-based extracellular matrix (ECM), on both DNT-sensitive and insensitive cells, indicating that the FN network does not serve as a functional receptor for DNT. We hypothesized that the FN network functions as a temporary storage system for DNT, enabling the small amount of the toxin to effectively reach target cells across the epithelia and connective tissue.
DNT binds to the FN-based ECM network
Screening for a molecule mediating the colocalization of DNT and the FN network
DNT is liberated from the FN network and affects sensitive cells
Here, we found that DNT temporarily associated with the FN network on cells. FN, a major component of the ECM, is mainly produced by fibroblasts and organized into a fibrillar network through binding to cell surface receptors, integrins [14–16]. A DNT mutant deficient in transglutaminase activity was also associated with the FN network (data not shown), indicating that the enzymatic activity of DNT is not required for the association. Because deletion mutants of DNT, in which any of the regions is missing, and heat-inactivated DNT did not associated with the FN network (data not shown), the overall structure of the toxin may be crucial to the association. DNT did not colocalize with the FN network generated by MRC-5 cells, suggesting that it interacts with FN not directly, but via another cellular component. Nidogen-2 in an N-terminally truncated could be a candidate for the component, because it was present in only the fraction which induced the association of DNT with the FN network on MRC-5 cells, whereas full-length nidogen-2 did not. Although its biological importance is not fully understood, nidogen-2 is known to interact with various molecules in the ECM . The nature of the truncated nidogen-2 is currently unknown. How the truncated nidogen-2 mediates the association between DNT and the FN network is not known either. At least, we observed that nidogen-2 was colocalized with not only FN but also DNT in the fibrillar structure. SBED-DNT crosslinked to two distinct components in addition to FN (Fig. 1C). These two components might be other candidates to intermediate the association between DNT and the FN network. However, they could not be isolated by combinations of anion- and cation-exchange chromatographies, probably because of their instability. In addition, the living cells, some cell membrane proteins, and/or the fibrillar structure of FN may be also required, because we could not reproduce the association of DNT with FN in the presence of the culture supernatant of FN-null cells by in vitro techniques such as ELISA and immunoprecipitation (data not shown). DNT may associate with the FN network by a complicated mechanism involving the truncated nidogen-2 and other cellular components. We are now conducting further work to elucidate this issue.
The association of DNT with the FN network was seen in not only DNT-sensitive cells but also insensitive cells, which indicates that the FN network neither serves as a receptor for the toxin nor is involved in the intoxicating procedures of the toxin on sensitive cells. Instead, the FN network may serve as a transient storage site for DNT, allowing the toxin to reach target cells or tissues more efficiently. Components of the ECM including FN are known to bind and regulate various growth factors such as insulin-like growth factor (IGF), fibroblast growth factor (FGF), transforming growth factor-beta (TGF-β), hepatocyte growth factor (HGF), and vascular endothelial growth factor (VEGF) [18, 19]. These growth factors are released from the ECM in response to alterations in the extracellular environment and exert biological effects to regulate cell survival, proliferation, and differentiation. For example, VEGF is associated with the ECM via FN or heparan sulfate at acidic pH. When the pH of the extracellular milieu increases, VEGF is released from the ECM network and activates its functional receptor to induce angiogenesis [20, 21]. This pH-dependent association of VEGF is considered a key mechanism determining the direction of newly developed blood vessels in wound healing and tumor metastasis. The association of DNT with the FN network was also dependent on the pH of the extracellular environment. Bordetella infections are reported to be accompanied by necrosis or the desquamation of superficial epithelial layers with inflammatory responses [22, 23]. These events may facilitate the exposure of newly generated ECM containing FN. The inflammatory locus is reportedly characterized by local acidosis due to lactic acid production . FN is actively produced by fibroblasts and osteoblasts, mesenchymal cells, which could be targets for DNT. Therefore, it is conceivable that DNT binds to the ECM containing FN at low pH in inflammatory areas during an infection, and by repeatedly associating with and diffusing from the FN network, moves deep into tissue where the density of FN should be higher, eventually reaching target cells. This may explain how DNT, which is not secreted by bacteria and is present at low concentrations in extrabacterial milieus, can affect target tissues in Bordetella infections such as atrophic rhinitis.
DNT associates temporarily with FN-based ECM network. The association seems to be mediated by the truncated-form of nidogen-2 and/or some cellular components, which have an affinity to the FN network. It is likely that the FN network does not function as a specific receptor but serves as a temporary storage system for DNT, enabling the small amount of the toxin to effectively reach target cells across the epithelia and connective tissue.
Mouse preosteoblastic cells MC3T3-E1 were cultured in alpha modified Eagle's medium (α-MEM) supplemented with 10% fetal calf serum (FCS). Mouse embryonic fibroblasts Balb3T3 and human fibroblasts MRC-5 were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% FCS. FN-null cells, which were kindly provided by Dr. Sottile , were maintained in an 1:1 mixture of Cellgro (Mediatech) and Aim V (GIBCO/Invitrogen).
Reagents and antibodies
Human plasma FN was purchased from Sigma. Collagen to coat culture plates and slide glasses was from Nitta gelatin Co. (Cellmatrix, Osaka, Japan). The monoclonal (FN-15, F7387) and polyclonal (F3648) antibodies against FN and polyclonal antibody against laminin (L9393) were obtained from Sigma. The anti-mouse nidogen-2 (M-300, sc-33143) and anti-collagen type I (234168) antibodies were from Santa Cruz and Calbiochem, respectively. The anti-DNT monoclonal antibody 2B3 and anti-DNT polyclonal antibody were prepared as reported [4, 26]. Alexa 488-conjugated goat anti-rabbit IgG, Alexa 546-conjugated goat anti-mouse IgG, and Alexa 488-conjugated streptavidin were from Molecular Probes/Invitrogen. Horseradish peroxidase (HRP)-conjugated streptavidin was from Chemicon. DNT that is N-terminally fused with hexahistidine was obtained as reported .
Sulfo-SBED, a trifunctional cross-linking reagent, was purchased from Thermo scientific. 5-carboxyfluorescein, succinimidyl ester (5-FAM, SE) was obtained from Molecular Probes/Invitrogen. For conjugation, DNT was dialyzed against 0.1 M NaHCO3, pH 8.3, mixed with Sulfo-SBED or 5-FAM, SE at a molar ratio of 1:32, and incubated at room temperature for 30 min. After incubation, the unconjugated reagent was removed by gel filtration with a PD-10 column (GE Healthcare).
Immunofluorescent staining of DNT-treated cells
MC3T3-E1, Balb3T3, and MRC-5 cells were seeded at 50,000 cells/cm2 in wells of a 24-well plate with cover glasses and grown overnight. FN-null cells were cultured overnight on collagen-coated cover glasses in Cellgro-Aim V with or without 10 μg/ml of human FN. The next day, the medium was replaced with a fresh batch containing 2 μg/ml of DNT, 5-FAM-conjugated DNT (5-FAM-DNT) or SBED-conjugated DNT (SBED-DNT), and the cells were incubated for 15 min at 37°C. The cells were then fixed with 3% paraformaldehyde in Dulbecco's modified phosphate-buffered saline (D-PBS (-)) for 10 min and treated with primary antibodies for 1 h, and subsequently secondary antibodies for 30 min in the presence of 10% FCS. The cells were washed three times with D-PBS (-) after each procedure. The cells were mounted in Fluoromount (Diagnostic BioSystems) and imaged with an OLYMPUS BX50 microscope controlled by SlideBook 4.0 (Intelligent Imaging Innovation, Inc.).
Anti-DNT polyclonal or monoclonal antibodies were used at 10 μg/ml for DNT staining. FN, collagen typeI, laminin, and nidogen-2 were stained with the respective antibodies at concentrations indicated in the instruction manuals.
Cross-linking of MC3T3-E1 cells with SBED-conjugated DNT
Confluent MC3T3-E1 cells in a 10-cm dish were treated with 2.5 μg/ml of SBED-DNT at 37°C for 15 min and then exposed to UV light at 365 nm for 5 min. The cells were washed with D-PBS (-) twice and solubilized with D-PBS (-) containing 1% NP-40 and 1% protease inhibitor cocktail (Nacalai, Kyoto, Japan) at 4°C for 60 min. The solubilized cells were centrifuged at 10,000 × g for 20 min at 4°C and the supernatant and pellet were collected as the soluble fraction and insoluble fraction, respectively. The insoluble fraction was sonicated in D-PBS (-) containing 5 μg/ml of DNase I and 8 M urea. After centrifugation, the supernatant was injected into a Mini Q column (0.32 × 3 cm, GE Healthcare), and eluted with a gradient of 0-1 M NaCl in 20 mM Tris-HCl (pH 8.5), containing 8 M urea, using the SMART system (GE Healthcare).
Screening for components intermediating the association between DNT and the FN network
FN-null cells or MC3T3-E1 cells were cultured in FCS-free DMEM or α-MEM for 72 h. The supernatant of the culture was dialyzed against 20 mM Tris-HCl, pH 8.5 containing 0.5 M NaCl, and subjected to anion-exchange chromatography with a HiTrap Q column (0.7 × 2.5 cm, GE Healthcare). The materials absorbed to the column were eluted in 1-ml fractions with a linear gradient of 0.5-1 M NaCl, and each fraction was tested for the ability to recruit DNT to the fibrillar structure on MRC-5 cells using immunofluorescence microscopy. The positive fractions were collected, appropriately diluted, and mixed with 5% CHAPS and 10 M urea to make a solution of 20 mM Tris-HCl, pH 8.5, containing 50 mM NaCl, 0.5% CHAPS and 6 M urea. The sample was subjected to Mono Q anion-exchange chromatography, and eluted with a linear gradient of 0.05-1 M NaCl. The eluted fractions were examined again for the ability to recruit DNT to the fibrillar structure on MRC-5 cells. Proteins contained in the positive fraction were identified by mass spectrometry as mentioned below.
DNT diffusion assay
FN-null cells were seeded in wells of a 24-well plate at 25,000 cells/cm2 and grown overnight. The next day, the cells were washed well with Cellgro-Aim V and incubated overnight in the same medium with or without 10 μg/ml of human FN. The culture medium was replaced with a fresh batch containing 2.5 μg/ml of DNT and the cells were incubated for 15 min at 37°C. After three washes with FCS-free DMEM, the cells were further incubated in the fresh medium. The culture supernatant was taken at the indicated time point, and an aliquot was applied to MC3T3-E1 cells without dilution. After incubation at 37°C overnight, the cells were examined for actin stress fibers as described previously . Another aliquot of the culture supernatant was examined for DNT by sandwich-ELISA, performed with a 96-well plate coated with anti-DNT polyclonal antibody. After blocking with 0.2% BSA at 4°C overnight, each sample was added to the plate in triplicate and incubated for 2 h at 37°C. The plate was treated with biotin-labeled anti-DNT antibody, followed by HRP-conjugated streptavidin for 1 h at 37°C. BM Blue POD substrate (Roche) was used as an HRP substrate and the reaction was stopped by 1 M H2SO4. The wells were washed four times with D-PBS (-) containing 0.05% Tween-20 between each step. The concentration of DNT was estimated from a standard curve made with a DNT preparation.
Protein concentrations were determined using BCA Protein Assay Reagents (Thermo Scientific). For the mass spectrometric analysis, samples were subjected to SDS-PAGE, followed by staining using a Silver stain MS kit (Wako). Visualized proteins were exercised from the gels, and digested with trypsin according to a method described elsewhere . Mass spectrometric data were analyzed with the MASCOT program (Matrix Science Ltd.). The statistical differences among groups of data were analyzed by one-way analysis of variance (ANOVA), followed by a Bonferroni posttest, using GraphPad Prism software version 4 (GraphPad Software, Inc.)
We are grateful to Dr. Sottile (University of Rochester Medical Center, NY, USA) for providing the FN-null cells. This work was supported in part by Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Science, and Technology of Japan.
- Fukui A, Horiguchi Y: Bordetella dermonecrotic toxin exerting toxicity through activation of the small GTPase Rho. J Biochem. 2004, 136 (4): 415-419. 10.1093/jb/mvh155.View ArticlePubMed
- Horiguchi Y, Inoue N, Masuda M, Kashimoto T, Katahira J, Sugimoto N, Matsuda M: Bordetella bronchiseptica dermonecrotizing toxin induces reorganization of actin stress fibers through deamidation of Gln-63 of the GTP-binding protein Rho. Proc Natl Acad Sci USA. 1997, 94 (21): 11623-11626. 10.1073/pnas.94.21.11623.PubMed CentralView ArticlePubMed
- Masuda M, Betancourt L, Matsuzawa T, Kashimoto T, Takao T, Shimonishi Y, Horiguchi Y: Activation of rho through a cross-link with polyamines catalyzed by Bordetella dermonecrotizing toxin. Embo J. 2000, 19 (4): 521-530. 10.1093/emboj/19.4.521.PubMed CentralView ArticlePubMed
- Matsuzawa T, Kashimoto T, Katahira J, Horiguchi Y: Identification of a receptor-binding domain of Bordetella dermonecrotic toxin. Infect Immun. 2002, 70 (7): 3427-3432. 10.1128/IAI.70.7.3427-3432.2002.PubMed CentralView ArticlePubMed
- Kashimoto T, Katahira J, Cornejo WR, Masuda M, Fukuoh A, Matsuzawa T, Ohnishi T, Horiguchi Y: Identification of functional domains of Bordetella dermonecrotizing toxin. Infect Immun. 1999, 67 (8): 3727-3732.PubMed CentralPubMed
- Horiguchi Y, Senda T, Sugimoto N, Katahira J, Matsuda M: Bordetella bronchiseptica dermonecrotizing toxin stimulates assembly of actin stress fibers and focal adhesions by modifying the small GTP-binding protein rho. J Cell Sci. 1995, 108 (Pt 10): 3243-3251.PubMed
- Masuda M, Minami M, Shime H, Matsuzawa T, Horiguchi Y: In vivo modifications of small GTPase Rac and Cdc42 by Bordetella dermonecrotic toxin. Infect Immun. 2002, 70 (2): 998-1001.PubMed CentralPubMed
- Brockmeier SL, Register KB, Magyar T, Lax AJ, Pullinger GD, Kunkle RA: Role of the dermonecrotic toxin of Bordetella bronchiseptica in the pathogenesis of respiratory disease in swine. Infect Immun. 2002, 70 (2): 481-490. 10.1128/IAI.70.2.481-490.2002.PubMed CentralView ArticlePubMed
- Hanada M, Shimoda K, Tomita S, Nakase Y, Nishiyama Y: Production of lesions similar to naturally occurring swine atrophic rhinitis by cell-free sonicated extract of Bordetella bronchiseptica. Jpn J vet Sci. 1979, 41 (1): 1-8.View Article
- Horiguchi Y, Okada T, Sugimoto N, Morikawa Y, Katahira J, Matsuda M: Effects of Bordetella bronchiseptica dermonecrotizing toxin on bone formation in calvaria of neonatal rats. FEMS Immunol Med Microbiol. 1995, 12 (1): 29-32. 10.1111/j.1574-695X.1995.tb00170.x.View ArticlePubMed
- Magyar T, Glavits R, Lendvai N, Rethy L: Turbinate atrophy in mice caused by Bordetella pertussis. Acta Vet Hung. 1987, 35 (4): 433-436.PubMed
- Roop RM, Veit HP, Sinsky RJ, Veit SP, Hewlett EL, Kornegay ET: Virulence factors of Bordetella bronchiseptica associated with the production of infectious atrophic rhinitis and pneumonia in experimentally infected neonatal swine. Infect Immun. 1987, 55 (1): 217-222.PubMed CentralPubMed
- Silveira D, Edington N, Smith IM: Ultrastructural changes in the nasal turbinate bones of pigs in early infection with Bordetella bronchiseptica. Res Vet Sci. 1982, 33 (1): 37-42.PubMed
- Bosman FT, Stamenkovic I: Functional structure and composition of the extracellular matrix. J Pathol. 2003, 200 (4): 423-428. 10.1002/path.1437.View ArticlePubMed
- Mao Y, Schwarzbauer JE: Fibronectin fibrillogenesis, a cell-mediated matrix assembly process. Matrix Biol. 2005, 24 (6): 389-399. 10.1016/j.matbio.2005.06.008.View ArticlePubMed
- Pankov R, Yamada KM: Fibronectin at a glance. J Cell Sci. 2002, 115 (Pt 20): 3861-3863. 10.1242/jcs.00059.View ArticlePubMed
- Ho MS, Bose K, Mokkapati S, Nischt R, Smyth N: Nidogens-Extracellular matrix linker molecules. Microsc Res Tech. 2008, 71 (5): 387-395. 10.1002/jemt.20567.View ArticlePubMed
- Flaumenhaft R, Rifkin DB: The extracellular regulation of growth factor action. Mol Biol Cell. 1992, 3 (10): 1057-1065.PubMed CentralView ArticlePubMed
- Taipale J, Keski-Oja J: Growth factors in the extracellular matrix. Faseb J. 1997, 11 (1): 51-59.PubMed
- Goerges AL, Nugent MA: Regulation of vascular endothelial growth factor binding and activity by extracellular pH. J Biol Chem. 2003, 278 (21): 19518-19525. 10.1074/jbc.M211208200.View ArticlePubMed
- Goerges AL, Nugent MA: pH regulates vascular endothelial growth factor binding to fibronectin: a mechanism for control of extracellular matrix storage and release. J Biol Chem. 2004, 279 (3): 2307-2315. 10.1074/jbc.M308482200.View ArticlePubMed
- Mattoo S, Cherry JD: Molecular pathogenesis, epidemiology, and clinical manifestations of respiratory infections due to Bordetella pertussis and other Bordetella subspecies. Clin Microbiol Rev. 2005, 18 (2): 326-382. 10.1128/CMR.18.2.326-382.2005.PubMed CentralView ArticlePubMed
- Mattoo S, Foreman-Wykert AK, Cotter PA, Miller JF: Mechanisms of Bordetella pathogenesis. Front Biosci. 2001, 6: E168-186. 10.2741/Mattoo.View ArticlePubMed
- Lardner A: The effects of extracellular pH on immune function. J Leukoc Biol. 2001, 69 (4): 522-530.PubMed
- Sottile J, Hocking DC, Swiatek PJ: Fibronectin matrix assembly enhances adhesion-dependent cell growth. J Cell Sci. 1998, 111 (Pt 19): 2933-2943.PubMed
- Horiguchi Y, Nakai T, Kume K: Simplified procedure for purification of Bordetella bronchiseptica dermonecrotic toxin. FEMS Microbiol Lett. 1990, 54 (1-3): 39-43. 10.1111/j.1574-6968.1990.tb03969.x.View ArticlePubMed
- Matsuzawa T, Fukui A, Kashimoto T, Nagao K, Oka K, Miyake M, Horiguchi Y: Bordetella dermonecrotic toxin undergoes proteolytic processing to be translocated from a dynamin-related endosome into the cytoplasm in an acidification-independent manner. J Biol Chem. 2004, 279 (4): 2866-2872. 10.1074/jbc.M310340200.View ArticlePubMed
- Satoh T, Arii J, Suenaga T, Wang J, Kogure A, Uehori J, Arase N, Shiratori I, Tanaka S, Kawaguchi Y: PILRalpha is a herpes simplex virus-1 entry coreceptor that associates with glycoprotein B. Cell. 2008, 132 (6): 935-944. 10.1016/j.cell.2008.01.043.PubMed CentralView ArticlePubMed
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.