- Email: [email protected]
Identification of an outer membrane protein of Fusobacterium necrophorum subsp. necrophorum that binds with high affinity to bovine endothelial cells
Veterinary Microbiology 176 (2015) 196–201
Contents lists available at ScienceDirect
Veterinary Microbiology journal homepage: www.elsevier.com/locate/vetmic
Short Communication
Identification of an outer membrane protein of Fusobacterium necrophorum subsp. necrophorum that binds with high affinity to bovine endothelial cells Amit Kumar, Sailesh Menon, T.G. Nagaraja, Sanjeev Narayanan * Department of Diagnostic Medicine/Pathobiology, College of Veterinary Medicine, Kansas State University, Manhattan, KS 66506, USA
A R T I C L E I N F O
A B S T R A C T
Article history: Received 24 July 2014 Received in revised form 5 November 2014 Accepted 14 December 2014
Fusobacterium necrophorum, a Gram-negative anaerobe, is the primary etiologic agent of liver abscesses in cattle. There are two subspecies; subsp. necrophorum and subsp. funduliforme, which differ in morphological, biochemical, molecular characteristics, and virulence. The subsp. necrophorum, which is more virulent, occurs more frequently in liver abscesses than the subsp. funduliforme. Bacterial adhesion to the host cell surface is critical to the pathogenesis of several bacterial infections, and in F. necrophorum, outer membrane proteins (OMP) have been shown to mediate adhesion to bovine endothelial cells. The objective of this study was to identify potential adhesins that are involved in adhesion of F. necrophorum subsp. necrophorum to the host cells. An OMP of 42.4 kDa, which binds with high affinity to the bovine endothelial cells and is recognized by the sera from cattle with liver abscesses, was identified. N-terminal sequencing of the protein showed 96% homology to the FomA protein of F. nucleatum. The PCR analysis showed that this fomA gene was present in several strains of subsp. necrophorum, subsp. funduliforme of bovine and subsp. funduliforme of human origin. The purified native and recombinantly expressed protein when preincubated with the endothelial cells, prevented the attachment of subsp. necrophorum significantly. In addition, the polyclonal antibody produced against the protein prevented the binding of subsp. necrophorum to bovine endothelial cells. ß 2014 Elsevier B.V. All rights reserved.
Keywords: F. necrophorum Outer membrane proteins FomA Bovine endothelial cells
1. Introduction Fusobacterium necrophorum is a Gram-negative, anaerobic, and rod-shaped to pleomorphic bacterium. It is associated with a variety of necrotic infections in animals, and is a major pathogen of cattle in which it causes hepatic abscesses, a disease of significant economic importance to the feedlot industry in the US (Nagaraja et al., 2005). Hepatic abscesses result from ruminal acidosis, which
* Corresponding author at: Department of Diagnostic Medicine and Pathobiology, K-246 Mosier Hall, 1800 Denison Avenue, USA. Tel.: +1 785 532 4430; fax: +1 785 532 4039. E-mail address: [email protected] (S. Narayanan). http://dx.doi.org/10.1016/j.vetmic.2014.12.015 0378-1135/ß 2014 Elsevier B.V. All rights reserved.
subsequently leads to rumenitis. F. necrophorum, a normal inhabitant of the rumen, by an unknown mechanism binds to the ruminal epithelium (Takayama et al., 2000), causes micro abscesses in the ruminal wall, and eventually reaches liver via portal circulation to cause abscesses (Scanlan and Hathcock, 1983; Nagaraja and Chengappa, 1998). F. necrophorum is classified into subsp. necrophorum and subsp. funduliforme, which differ in morphological, biochemical, and molecular characteristics, and in virulence (Shinjo et al., 1991; Tadepalli et al., 2008). The subsp. necrophorum, which is more virulent, occurs more frequently in liver abscesses than the subsp. funduliforme. Bacterial adhesion is a critical step in the establishment of infection and disease pathogenesis of many Gram-negative bacterial species (Haase et al., 1999; Rocha-De-Souza
A. Kumar et al. / Veterinary Microbiology 176 (2015) 196–201
et al., 2001; Huntley et al., 2007). Generally, Gram-negative bacterial cell attachment to the host eukaryotic cells is mediated by capsule, fimbrial proteins or outer membrane proteins (OMP) (Bavington and Page, 2005). Outer membrane proteins mediating attachment to epithelial cells have been characterized in another species of Fusobacterium; F. nucleatum, a human oral pathogen (Han et al., 2005). We have shown that OMP mediate attachment of F. necrophorum to bovine endothelial cells (Kumar et al., 2013a). Interestingly, the OMP profile of F. necrophorum subsp. necrophorum is different from that of the subsp. funduliforme (Kumar et al., 2013b). The most prominent bands were 40 and 37.5 kDa in subsp. necrophorum and F. necrophorum subsp. funduliforme, respectively (Kumar et al., 2013b). In this study, we show that a 42.4 kDa OMP of F. necrophorum subsp. necrophorum binds with very high affinity to bovine endothelial cells. 2. Materials and methods 2.1. Bacterial strains and culturing F. necrophorum subsp. necrophorum, strains 8L1, A25, A21, A50, subsp. funduliforme strains B35, B47, B17, B29, isolated from bovine liver abscesses and subsp. necrophorum strain RA15, RA16 and RA17 from ruminal content (Tan et al., 1992; Narayanan et al., 1997) and four human clinical strains of F. necrophorum (RMA10682, RMA14786, RMA16505 and RMA16539; provided by Dr. Diane Citron, R. M. Alden Research Laboratory, Santa Monica, CA) were used in this study. The organisms were cultured in prereduced, anaerobically sterilized brain heart infusion (PRAS-BHI) broth (Tan et al., 1992). Bacterial cells for attachment assay were obtained by inoculating 0.1 ml of an overnight culture derived from a single colony into a 10 ml PRAS-BHI and grown to an absorbance of 0.6 at 600 nm, which corresponded to 1 107 cells/ml. 2.2. Isolation of high-affinity binding OMP Bovine adrenal gland capillary endothelial cells (EJG cells, CRL-8659, American type culture collection, Manassas, VA) were grown in EMEM medium with 10% fetal bovine serum (FBS) supplemented with 10 ml of 200 mM L-glutamine (Thermo Fisher Scientific Inc., Waltham, MA), 5 ml of amphotericin B (Thermo Fisher), 5 ml of penicillin (10,000 U/ml; Thermo Fisher) and streptomycin (10 mg/ ml; Thermo Fisher) and 2 ml of 1% solution of ciprofloxacin HCl (10 mg/ml) (Cellgro Mediatech Inc., Manassas, VA) per 1 l of EMEM medium. The medium was changed every 3 or 4 days until the cells formed a monolayer. The cells were then trypsinized and seeded in a six-well plate (Corning Inc. Lowell, MA) with a cell suspension containing 1 105 cells/ml and incubated at 37 8C in EMEM medium containing 10% FBS for 48 h to remove any effect of trypsinization on the eukaryotic cell surface proteins. The EJG cells seeded in a six-well plate were then fixed with modified Karnovsky’s fixative (0.1 M cacodylate buffer, 2.5% gluteraldehyde, 2% paraformaldehyde), and incubated with OMP extracted from subsp. necrophorum, strain 8L1 (500 ml of extract containing 400 mg of protein/ well). We used endothelial cells treated with Karnovsky’s
197
fixative because fixation protects the cell ultrastructure by crosslinking the proteins and hence, preventing them from getting loose during washing. The extraction of OMP from F. necrophorum subsp. necrophorum, strain 8L1 was carried out using the method published previously (Kumar et al., 2013b). After overnight incubation of endothelial cells with OMP, the unbound fraction was removed from the wells. The cells were washed twice with PBS, and then twice each with buffers containing increasing strengths of detergents; first with PBS containing 0.1% nonyl phenoxypolyethoxylethanol-40 (NP-40; Thermo Fisher) followed by modified radio-immunoprecipitation assay (RIPA) buffer (25 mM Tris–HCl pH 7.6, 150 mM NaCl, 1% NP-40, and 1% sodium deoxycholate). Finally, the bound OMP was collected by washing the wells with 200 ml SDS sample buffer (62.5 mM Tris–HCl, pH 6.8, 25% glycerol and 10% SDS). Each buffer wash was concentrated to 200 ml volume as of SDS sample buffer and separated on a SDS-PAGE gel. The EJG cells incubated without OMP, but subjected to buffers served as the negative control. 2.3. N-terminal sequencing of the high-affinity bound OMP The isolated high-affinity binding OMP in SDS-PAGE buffer was run on a 4% stacking and 10% separation gel after overnight polymerization to avoid an N-terminal blockage. The separated proteins were transferred on to a polyvinylidene fluoride (PVDF) membrane overnight by electroblotting and stained with Coomassie blue. The membrane was then destained for 15 min in 40% methanol/10% acetic acid followed by rinsing in 90% methanol/5% acetic acid followed by multiple rinses in distilled water for a total of 4 h. Finally, the membrane was dried using Whatman No. 1 filter paper and stored. The 42.4 kDa protein present on dried PVDF membrane was excised and submitted on dry ice for sequencing (The Protein Facility, Iowa State University, Ames, IA). The N-terminal protein sequencing was carried out with 494 Procise Protein/ peptide Sequencer/140C Analyzer (Perkin Elmer Inc., San Jose, CA) using Edman degradation method. The identified amino acid sequence was searched by blastp against GenBank protein database in PubMed and investigated for similar proteins present in other bacterial species. The protein was designated FomA because of the high degree of similarity to FomA OMP of F. nucleatum. 2.4. Adhesion inhibition assay with FomA The OMP of subsp. necrophorum strain 8L1 (400 mg) were incubated with EJG cells seeded in a 6-well plate treated with Karnovsky’s fixative, as mentioned previously. After overnight incubation, cells were washed with PBS twice to remove unbound proteins, followed by two washes with PBS + 0.1% NP-40 to remove OMP other than the FomA bound to the fixed cells. Subsp. necrophorum strain 8L1 was grown to an absorbance of 0.6 at 600 nm in PRAS-BHI medium and 1 ml of the culture (1 107 cells/ ml) was inoculated into each well. The bacterial cells were incubated with fixed cells for 1 h followed by vigorous washing with PBS for 3 to 4 times to remove unbound bacterial cells. The bacterial cells bound to the EJG cells
198
A. Kumar et al. / Veterinary Microbiology 176 (2015) 196–201
were collected by scraping the wells after adding 500 ml of trypsin and quickly inactivating the trypsin with 1.5 ml of anaerobic EMEM medium containing 10% heat inactivated FBS. An aliquot of 100 ml of the cell suspension containing the bacterial cells was serially diluted in PRAS-BHI and spread-plated on blood agar, and incubated in an anaerobic glove box (Forma Scientific, Marietta, OH) to enumerate the bacterial concentration. 2.5. Identification, cloning, sequencing and expression of FomA gene The primers to amplify the gene for the FomA was designed based on the published sequence of fomA of F. nucleatum. Forward primer was designed starting from the amino acid immediately following 20-amino acid signal sequence of fomA (50 -GAAGTTATGCCAGCACCTATGCCAGAA-30 ) and the reverse primer including the stop codon of the protein (50 -TTAGAAGCTAACTTTCATACC AG-30 ). The PCR conditions were as follows: 1 cycle 94 8C for 3 min followed by 35 cycles of denaturation at 94 8C for 1 min, annealing at 52 8C for 30 s, and extension at 72 8C for 2 min and a final extension at 72 8C for 4 min using Ex taq polymerase (Takara Inc., Mountain View, CA). The amplified gene of fomA from subsp. necrophorum strain 8L1 was cloned in-frame into pET45b+ vector (EMD Bioscience, Billerica, MA) by adding restriction sites at the 50 end of the primers (50 -CGGGATCCAGAAGTTATGCCTGCACC-30 with BamHI site and reverse primer 50 -AACTGCAGTTAGAAGCTAACTTTCATA-30 with PstI site). The pET45b+ vector has an N-terminal histidine tag for protein purification using a Ni+ affinity column. The cloned vector was transformed into Escherichia coli BL21 (DE3) cells and the clones were verified by PCR of several random colonies. The DNA sequence of the insert from the clone expressing the protein was determined using DNA sequencing (ACGT Inc., Wheeling, IL). The amino acid sequence of the cloned DNA was deduced from the DNA sequence and analyzed further for its similarity with homologous proteins present in the database using blastp at PubMed database. 2.6. fomA gene and FomA protein of F. necrophorum subspecies The PCR analysis was carried out with DNA extracted (Narayanan et al., 2001) from four strains of subsp. necrophorum (A25, A21, A50 and 8L1), four strains of subsp. funduliforme (B17, B29, B47, and B35) and four human clinical strains of F. necrophorum (RMA10682, RMA14786, RMA16505 and RMA16539) to determine if fomA gene is present. Further, the presence of FomA protein in bovine subsp. funduliforme and human strains of F. necrophorum was determined by SDS-PAGE analysis of OMP. 2.7. Adhesion inhibition assay with polyclonal antibody raised against the native and recombinant FomA of F. necrophorum The FomA native protein obtained from the SDS-PAGE gel and the recombinant protein expressed and purified
were sent to Cocalico Biologicals Inc. (Reamstown, PA) for polyclonal antibody production in rabbit. The cultured EJG cells were seeded in a 6-well plate (Corning Inc.) by adding 3 ml of cells at a concentration of 1 105 cells/ml and incubated in EMEM medium for 48 h. After 48 h, the cells were washed three times with sterile PBS. Bacteria (subsp. necrophorum or subsp. funduliforme) were grown to an OD600 of 0.6, washed, incubated with 1:100 dilution of the polyclonal antiserum for 1 h, and washed three times in PBS to remove unbound antibodies. These pretreated bacteria were then added to EJG cell monolayers at a multiplicity of infection of 100:1 and allowed to attach for 1 h at 37 8C. Cells were washed 4 times with sterile PBS and the EJG cells along with attached bacteria were collected using 500 ml of Trypsin-Versene (Lonza Walkersville Inc., Walkersville, MD), and neutralized quickly with 1.5 ml of anaerobic EMEM containing 10% fetal bovine serum The attached bacteria were enumerated following serial dilution and spread-plating on blood agar and incubating for 2 days at 39 8C in an anaerobic glove box. 2.8. Statistical analysis All adhesion inhibition assays of F. necrophorum to EJG cells set up in triplicates were performed a minimum of three independent replications. The data were analyzed by PROC MIXED procedure of SAS (ver. 9.2; Cary, NC) with colony counts as the outcome variable and EJG cell treatments (preincubation with OMP or anti-FomA antiserum) as the predictor variable. A P value of <0.05 was considered significant. 3. Results 3.1. Identification of an OMP with high binding affinity The fraction of extracted OMP removed after overnight incubation with Karnovsky’s fixed EJG cells showed several protein bands that did not bind to the fixed cells (lane 2 in Fig. 1). The PBS wash of Karnovsky’s fixed cells incubated with extracted OMP of F. necrophorum subsp. necrophorum revealed the proteins loosely bound to the fixed cells (lane 3 in Fig. 1). However, no protein band was detected in washing with PBS + 0.1% NP-40 (lane 5 from treated cells and 6 from control cells in Fig. 1), modified RIPA (lane 7 from treated cells and 8 from control cells in Fig. 1). The final wash with SDS-PAGE buffer containing 10% SDS revealed a band of 42.4 kDa protein (lane 10 from treated cells and lane 11 from control cells in Fig. 1). The N-terminal sequence of the identified protein was K-E-V-M-P-A-P-M-P-E-D-E. A blastp search revealed that the identified adhesin protein sequence was similar to FomA protein of F. nucleatum and Fusobacterium periodontium, and an OMP protein in Fusobacterium varium. The amino acid sequence comparison of the identified adhesin protein and homologous proteins of other fusobacterial species revealed that the adhesin protein of F. necrophorum has a 20-aa signal peptide, similar to FomA (Nakagaki et al., 2010). This signal peptide is cleaved upon its expression into the periplasm and hence, the N-terminal sequencing showed the starting amino acid as lysine instead of methionine. Further analyses using inverse PCR identified
A. Kumar et al. / Veterinary Microbiology 176 (2015) 196–201
199
the DNA sequence of signal peptide as ‘ATGAAAAAATTAGCACTTGTATTAGGTTCTTTATTAGTCATCGGTTCT GCCGCTTCTGCT’ encoding MKKLALVLGSLLVIGSAASA as the predicted signal peptide. Western blot analysis with sera from steers that were experimentally induced with liver abscesses via intraperitoneal inoculation with subsp. necrophorum strain 8L1 in a separate study showed that the protein is immunodominant (Fig. 1B). 3.2. Adhesion inhibition assay When F. necrophorum subsp. necrophorum strain 8L1 was incubated with fixed EJG cells that were saturated with the FomA protein of F. necrophorum (after removing all loosely bound OMP with stringent wash leaving only the FomA protein), there was four fold (P < 0.0001) decrease in the binding of the bacterial cells compared to the control EJG cells that were not saturated with the FomA protein (Fig. 2). Other strains of subsp. necrophorum showed similar results (data not shown). 3.3. FomA in other subspecies of F. necrophorum The PCR assay of four bovine strains of subsp. necrophorum, four bovine strains of subs. funduliforme showed that the FomA gene was present in both subspecies (Fig. 3A). However, the size of the amplified PCR product was slightly lower, in subsp. funduliforme than that of subsp. necrophorum (Fig. 3A). F. necrophorum of human origin also showed PCR product of similar size (Fig. 3A).
Fig. 2. Inhibition assay with FomA. The EJG cells were saturated with FomA and attachment study was carried out with subsp. necrophorum strain 8L1. The bacteria were incubated with fixed EJG cells and the unbound bacteria were washed off using PBS. The cells were scraped along with bound bacteria and serially diluted in PRAS-BHI for enumeration and plated on blood agar to enumerate surface bound bacteria. Error bars represent standard error.
However, the protein band of 40–45 kDa range was only present in one (RMA16505) of the four human strains (Fig. 3B). All the bovine strains of subsp. necrophorum (RA15, RA16, RA17, 8L1, A50 and A25) showed a protein band near 40–45 kDa which was absent in subsp. funduliforme (B17, B30, B35, B36) (Fig. 3B). 3.4. Cloning, sequencing and expression of FomA from subsp. necrophorum Cloning of FomA gene, using primers that were designed based on the N-terminal sequence of FomA protein of F. nucleatum, in pET45b+ vector resulted in expression of a protein containing 363 amino acids with deduced molecular weight of 42.4 kDa. The protein had a 96% homology in C and N-terminal amino acid sequences to FomA of F. nucleatum. 3.5. Adhesion inhibition assay with polyclonal antibody raised against recombinant FomA
Fig. 1. (A) Outer membrane proteins (OMP) from subsp. necrophorum were incubated overnight with fixed EJG cells and unbound protein was removed (lane 2) and washed with PBS (lanes 3 and 4) and detergents with increasing stringency (PBS + 0.1% NP-40: lanes 5 and 6, modified RIPA: lanes 7 and 8, and SDS-PAGE buffer with 10% SDS: lanes 10 and 11). No protein was detected in washing with PBS + 0.1% NP-40 and modified RIPA buffer. The final wash with SDS-PAGE buffer containing 10% SDS revealed a band of 42.4 kDa protein in cells treated with OMP (lane 10). Lane 4, 6, 8 and 11 are control cells with no added OMP. (B) Western blot of the purified protein with serum from a steer with experimentally induced liver abscesses with subsp. necrophorum.
When subsp. necrophorum strain 8L1was preincubated with polyclonal antibodies raised against the recombinant FomA, there was up to 90% reduction in the binding of bacterial cells to EJG cells (Fig. 4b). The inhibition with polyclonal antibody raised against purified native FomA protein from subsp. necrophorum (>97% as compared to serum before immunization and >98% compared to no antibody control) was slightly higher than inhibition shown with antibody raised against the recombinant FomA protein (>92% as compared with pre-immunized serum and >94% with no antibody control). 4. Discussion We have shown previously that OMP of F. necrophorum mediate bacterial adhesion to bovine endothelial cells (Kumar et al., 2013a). Trypsin treatment of F. necrophorum, preincubation of endothelial cells with OMP extracted from F. necrophorum, or polyclonal antibody raised against
200
A. Kumar et al. / Veterinary Microbiology 176 (2015) 196–201
Fig. 3. (A) The presence of FomA gene in human F. necrophorum subsp. funduliforme strains (L: 1KB DNA ladder, 1-4: 1: RMA10682, 2: RMA16505, 3: RMA14786, 4: RMA16539), bovine isolates of subsp. funduliforme (Lanes 5–8: B17, 7: B29, 8: B47, 9: B35) and subsp. necrophorum (Lanes 9–12: A25, 10: A21, 11: A50, 12: 8L1) 13: negative control. (B) On SDS-PAGE analysis of OMPs, the 42.4 kD (FomA) protein was present in all strains of subsp. necrophorum. However, among subsp. funduliforme of human and bovine origins, only human strain RMA16505 showed protein of 42.4 kDa. The arrow indicates the 42.4 kDa protein band.
the OMP decreased the binding of bacterial cells (Kumar et al., 2013a). We designed a protocol to identify an OMP that binds with high affinity to the endothelial cells. The high affinity binding OMP was isolated after serially washing the cells with detergents of increasing stringency and finally with SDS-PAGE buffer which contained 10% SDS. No protein was detected in any of the fractions except in the highly stringent SDS-PAGE buffer that contained 10% SDS which suggested that the isolated protein was tightly bound to the EJG cells. This N-terminal protein sequencing of the adhesin and further analysis using BLASTP tool showed that this protein is present in other species of Fusobacterium, such as FomA of F. nucleatum, F. periodonticum and F. varium. The FomA of F. nucleatum is a porin protein and has been shown to act as an adhesin in
Fig. 4. Inhibition assay with polyclonal antibody raised against the adhesin protein. Subsp. necrophorum when preincubated with the antibody raised against native 42.4 kDa protein or against recombinant 42.4 kDa protein, showed a significant decrease in bacterial binding to bovine endothelial cells compared to subsp. necrophorum not treated with any antibody (control) or when preincubated with rabbit preimmunization sera. Error bars represent standard error.
coaggregation of F. nucleatum with other bacteria to form biofilm in oral infections in humans (Liu et al., 2010). The EJG cells saturated with adhesin showed reduced binding of subsp. necrophorum indicating that the protein blocked the binding sites of endothelial cells. The western blot analysis of the protein using sera from steers with liver abscesses caused by subsp. necrophorum showed that this protein is immunodominant. Based on the PCR analysis, strains of subsp. necrophorum and subsp. funduliforme of cattle origin, and human strains of subsp. funduliforme carried the fomA gene. However, the SDS-PAGE analysis of the OMP of strains of subsp. funduliforme of bovine and human origin did not show any dominant band in the 40–45 kDa range, except strain 16505 of human origin. The absence of the protein could either be due to low expression or mutation in the gene causing truncation of this protein in subsp. funduliforme. Further work is needed to analyze if this protein is actually expressed in subsp. funduliforme. Subsp. necrophorum is more virulent and is more frequently isolated in the liver abscesses of cattle (Nagaraja and Chengappa, 1998). It is likely that the FomA protein of subsp. necrophorum contributes to the binding of this bacterium to the host cells surfaces and facilitate its transport from the rumen to the liver to establish hepatic infection. The polyclonal antibody raised against the purified native and recombinant FomA in rabbit when incubated with subsp. necrophorum, showed significant decrease in the bacterial binding indicating that the antibody blocked the adhesin on the surface of subsp. necrophorum. The recognition of FomA protein by sera of steers infected with subsp. necrophorum and the ability of the antisera raised
A. Kumar et al. / Veterinary Microbiology 176 (2015) 196–201
against FomA to reduce binding of subsp. necrophorum to EJG cells show that it can be used as a potential candidate for vaccine against liver abscesses in cattle. A vaccine prepared using FomA of F. nucleatum has been shown to decrease the oral infections in a mouse model (Liu et al., 2010). Studies to create a mutant and subsequent complementation analysis were not carried out because there are no established systems available to genetically manipulate F. necrophorum genome. In conclusion, we have identified an outer membrane protein of F. necrophorum which binds with high affinity to the bovine endothelial cells. This protein may be a potential candidate for the development of a vaccine to prevent fusobacterial infections in cattle. Conflict of interest statement The authors declare that there are no conflicts of interest. Acknowledgements The authors acknowledge the assistance of Mr. Joe Anderson from the Virology Department at the Kansas State University-Veterinary Diagnostic Laboratory in culturing endothelial cells. We greatly appreciate financial assistance from the College of Veterinary Medicine and the Department of Diagnostic Medicine and Pathobiology at Kansas State University. References Bavington, C., Page, C., 2005. Stopping bacterial adhesion: a novel approach to treating infections. Respiration 72, 335–344. Haase, E.M., Zmuda, J.L., Scannapieco, F.A., 1999. Identification and molecular analysis of rough-colony-specific outer membrane proteins of Actinobacillus actinomycetemcomitans. Infect. Immun. 67, 2901–2908. Han, Y.W., Ikeami, A., Rajanna, C., Kawsar, H.I., Zhou, Y., Li, M., Sojar, H.T., Genco, R.J., Kuramitsu, H.K., Deng, C.X., 2005. Identification and characterization of a novel adhesin unique to oral fusobacteria. J. Bacteriol 187, 5330–5340.
201
Huntley, J.F., Conley, P.G., Hagman, K.E., Norgard, M.V., 2007. Characterization of Francisella tularensis outer membrane proteins. J. Bacteriol. 189, 561–574. Kumar, A., Gart, E., Nagaraja, T.G., Narayanan, S., 2013a. Adhesion of Fusobacterium necrophorum to bovine endothelial cells is mediated by outer membrane proteins. Vet. Microbiol. 162, 813–818. Kumar, A., Peterson, G., Nagaraja, T.G., Narayanan, S., 2013b. Outer membrane proteins of Fusobacterium necrophorum subsp. necrophorum and subsp. funduliforme. J. Basic Microbiol., http://dx.doi.org/10.1002/jobm. 201200748. Liu, P.F., Shi, W., Zhu, W., Smith, J.W., Hsieh, S.L., Gallo, R.L., Huang, C.M., 2010. Vaccination targeting surface FomA of Fusobacterium nucleatum against bacterial co-aggregation: implication for treatment of periodontal infection and halitosis. Vaccine 28, 3496–3505. Nagaraja, T.G., Narayanan, S.K., Stewart, G.C., Chengappa, M.M., 2005. Fusobacterium necrophorum infections in animals: pathogenesis and pathogenic mechanisms. Anaerobe 11, 239–246. Nagaraja, T.G., Chengappa, M.M., 1998. Liver abscesses in feedlot cattle: a review. J. Anim. Sci. 76, 287–298. Nakagaki, H., Sekine, S., Terao, Y., Toe, M., Tanaka, M., Ito, H.O., Kawabata, S., Shizukuishi, S., Fujihashi, K., Kataoka, K., 2010. Fusobacterium nucleatum envelope protein FomA is immunogenic and binds to the salivary statherin-derived peptide. Infect. Immun. 78, 1185–1192. Narayanan, S.K., Nagaraja, T.G., Okwumabua, O., Staats, J., Chengappa, M.M., Oberst, R.D., 1997. Ribotyping to compare Fusobacterium necrophorum isolates from bovin liver abscesses, ruminal walls and ruminal contents. Appl. Environ. Microbiol. 63, 4671–4678. Narayanan, S.K., Nagaraja, T.G., Chengappa, M.M., Stewart, G.C., 2001. Cloning, sequencing, and expression of the leukotoxin gene from Fusobacterium necrophorum. Infect. Immun. 69, 5447–5455. Rocha-De-Souza, C.M., Colombo, A.V., Hirata, R., Mattos-Guaraldi, A.L., Monteiro-Leal, L.H., Previato, J.O., Freitas, A.C., Andrade, A.F., 2001. Identification of a 43-kDa outer-membrane protein as an adhesin in Aeromonas caviae. J. Med. Microbiol. 50, 313–319. Scanlan, C.M., Hathcock, T.L., 1983. Bovine rumenitis-liver abscess complex: a bacteriological review. Cornell Vet. 73, 288–297. Shinjo, T., Fujisawa, T., Mitsuoka, T., 1991. Proposal of two subspecies of Fusobacterium necrophorum (Flugge) Moore and Holdeman: Fusobacterium necrophorum subsp. necrophorum subsp. nov., nom. rev (ex Flugge 1886), and Fusobacterium necrophorum subsp. funduliforme subsp. nov., nom. rev. (ex Halle 1898). Int. J. Syst. Bacteriol. 41, 395–397. Tadepalli, S., Stewart, G.C., Nagaraja, T.G., Narayanan, S.K., 2008. Leukotoxin operon and differential expressions of the leukotoxin gene in bovine Fusobacterium necrophorum subspecies. Anaerobe 14, 13–18. Takayama, Y., Kanoe, M., Maeda, K., Okada, Y., Kai, K., 2000. Adherence of Fusobacterium necrophorum subsp. necrophorum to ruminal cells derived from bovine rumenitis. Lett. Appl. Microbiol. 30, 308–311. Tan, Z.L., Nagaraja, T.G., Chengappa, M.M., 1992. Factors affecting leukotoxin activity of Fusobacterium necrophorum. Vet. Microbiol. 32, 15–28.
Contents lists available at ScienceDirect
Veterinary Microbiology journal homepage: www.elsevier.com/locate/vetmic
Short Communication
Identification of an outer membrane protein of Fusobacterium necrophorum subsp. necrophorum that binds with high affinity to bovine endothelial cells Amit Kumar, Sailesh Menon, T.G. Nagaraja, Sanjeev Narayanan * Department of Diagnostic Medicine/Pathobiology, College of Veterinary Medicine, Kansas State University, Manhattan, KS 66506, USA
A R T I C L E I N F O
A B S T R A C T
Article history: Received 24 July 2014 Received in revised form 5 November 2014 Accepted 14 December 2014
Fusobacterium necrophorum, a Gram-negative anaerobe, is the primary etiologic agent of liver abscesses in cattle. There are two subspecies; subsp. necrophorum and subsp. funduliforme, which differ in morphological, biochemical, molecular characteristics, and virulence. The subsp. necrophorum, which is more virulent, occurs more frequently in liver abscesses than the subsp. funduliforme. Bacterial adhesion to the host cell surface is critical to the pathogenesis of several bacterial infections, and in F. necrophorum, outer membrane proteins (OMP) have been shown to mediate adhesion to bovine endothelial cells. The objective of this study was to identify potential adhesins that are involved in adhesion of F. necrophorum subsp. necrophorum to the host cells. An OMP of 42.4 kDa, which binds with high affinity to the bovine endothelial cells and is recognized by the sera from cattle with liver abscesses, was identified. N-terminal sequencing of the protein showed 96% homology to the FomA protein of F. nucleatum. The PCR analysis showed that this fomA gene was present in several strains of subsp. necrophorum, subsp. funduliforme of bovine and subsp. funduliforme of human origin. The purified native and recombinantly expressed protein when preincubated with the endothelial cells, prevented the attachment of subsp. necrophorum significantly. In addition, the polyclonal antibody produced against the protein prevented the binding of subsp. necrophorum to bovine endothelial cells. ß 2014 Elsevier B.V. All rights reserved.
Keywords: F. necrophorum Outer membrane proteins FomA Bovine endothelial cells
1. Introduction Fusobacterium necrophorum is a Gram-negative, anaerobic, and rod-shaped to pleomorphic bacterium. It is associated with a variety of necrotic infections in animals, and is a major pathogen of cattle in which it causes hepatic abscesses, a disease of significant economic importance to the feedlot industry in the US (Nagaraja et al., 2005). Hepatic abscesses result from ruminal acidosis, which
* Corresponding author at: Department of Diagnostic Medicine and Pathobiology, K-246 Mosier Hall, 1800 Denison Avenue, USA. Tel.: +1 785 532 4430; fax: +1 785 532 4039. E-mail address: [email protected] (S. Narayanan). http://dx.doi.org/10.1016/j.vetmic.2014.12.015 0378-1135/ß 2014 Elsevier B.V. All rights reserved.
subsequently leads to rumenitis. F. necrophorum, a normal inhabitant of the rumen, by an unknown mechanism binds to the ruminal epithelium (Takayama et al., 2000), causes micro abscesses in the ruminal wall, and eventually reaches liver via portal circulation to cause abscesses (Scanlan and Hathcock, 1983; Nagaraja and Chengappa, 1998). F. necrophorum is classified into subsp. necrophorum and subsp. funduliforme, which differ in morphological, biochemical, and molecular characteristics, and in virulence (Shinjo et al., 1991; Tadepalli et al., 2008). The subsp. necrophorum, which is more virulent, occurs more frequently in liver abscesses than the subsp. funduliforme. Bacterial adhesion is a critical step in the establishment of infection and disease pathogenesis of many Gram-negative bacterial species (Haase et al., 1999; Rocha-De-Souza
A. Kumar et al. / Veterinary Microbiology 176 (2015) 196–201
et al., 2001; Huntley et al., 2007). Generally, Gram-negative bacterial cell attachment to the host eukaryotic cells is mediated by capsule, fimbrial proteins or outer membrane proteins (OMP) (Bavington and Page, 2005). Outer membrane proteins mediating attachment to epithelial cells have been characterized in another species of Fusobacterium; F. nucleatum, a human oral pathogen (Han et al., 2005). We have shown that OMP mediate attachment of F. necrophorum to bovine endothelial cells (Kumar et al., 2013a). Interestingly, the OMP profile of F. necrophorum subsp. necrophorum is different from that of the subsp. funduliforme (Kumar et al., 2013b). The most prominent bands were 40 and 37.5 kDa in subsp. necrophorum and F. necrophorum subsp. funduliforme, respectively (Kumar et al., 2013b). In this study, we show that a 42.4 kDa OMP of F. necrophorum subsp. necrophorum binds with very high affinity to bovine endothelial cells. 2. Materials and methods 2.1. Bacterial strains and culturing F. necrophorum subsp. necrophorum, strains 8L1, A25, A21, A50, subsp. funduliforme strains B35, B47, B17, B29, isolated from bovine liver abscesses and subsp. necrophorum strain RA15, RA16 and RA17 from ruminal content (Tan et al., 1992; Narayanan et al., 1997) and four human clinical strains of F. necrophorum (RMA10682, RMA14786, RMA16505 and RMA16539; provided by Dr. Diane Citron, R. M. Alden Research Laboratory, Santa Monica, CA) were used in this study. The organisms were cultured in prereduced, anaerobically sterilized brain heart infusion (PRAS-BHI) broth (Tan et al., 1992). Bacterial cells for attachment assay were obtained by inoculating 0.1 ml of an overnight culture derived from a single colony into a 10 ml PRAS-BHI and grown to an absorbance of 0.6 at 600 nm, which corresponded to 1 107 cells/ml. 2.2. Isolation of high-affinity binding OMP Bovine adrenal gland capillary endothelial cells (EJG cells, CRL-8659, American type culture collection, Manassas, VA) were grown in EMEM medium with 10% fetal bovine serum (FBS) supplemented with 10 ml of 200 mM L-glutamine (Thermo Fisher Scientific Inc., Waltham, MA), 5 ml of amphotericin B (Thermo Fisher), 5 ml of penicillin (10,000 U/ml; Thermo Fisher) and streptomycin (10 mg/ ml; Thermo Fisher) and 2 ml of 1% solution of ciprofloxacin HCl (10 mg/ml) (Cellgro Mediatech Inc., Manassas, VA) per 1 l of EMEM medium. The medium was changed every 3 or 4 days until the cells formed a monolayer. The cells were then trypsinized and seeded in a six-well plate (Corning Inc. Lowell, MA) with a cell suspension containing 1 105 cells/ml and incubated at 37 8C in EMEM medium containing 10% FBS for 48 h to remove any effect of trypsinization on the eukaryotic cell surface proteins. The EJG cells seeded in a six-well plate were then fixed with modified Karnovsky’s fixative (0.1 M cacodylate buffer, 2.5% gluteraldehyde, 2% paraformaldehyde), and incubated with OMP extracted from subsp. necrophorum, strain 8L1 (500 ml of extract containing 400 mg of protein/ well). We used endothelial cells treated with Karnovsky’s
197
fixative because fixation protects the cell ultrastructure by crosslinking the proteins and hence, preventing them from getting loose during washing. The extraction of OMP from F. necrophorum subsp. necrophorum, strain 8L1 was carried out using the method published previously (Kumar et al., 2013b). After overnight incubation of endothelial cells with OMP, the unbound fraction was removed from the wells. The cells were washed twice with PBS, and then twice each with buffers containing increasing strengths of detergents; first with PBS containing 0.1% nonyl phenoxypolyethoxylethanol-40 (NP-40; Thermo Fisher) followed by modified radio-immunoprecipitation assay (RIPA) buffer (25 mM Tris–HCl pH 7.6, 150 mM NaCl, 1% NP-40, and 1% sodium deoxycholate). Finally, the bound OMP was collected by washing the wells with 200 ml SDS sample buffer (62.5 mM Tris–HCl, pH 6.8, 25% glycerol and 10% SDS). Each buffer wash was concentrated to 200 ml volume as of SDS sample buffer and separated on a SDS-PAGE gel. The EJG cells incubated without OMP, but subjected to buffers served as the negative control. 2.3. N-terminal sequencing of the high-affinity bound OMP The isolated high-affinity binding OMP in SDS-PAGE buffer was run on a 4% stacking and 10% separation gel after overnight polymerization to avoid an N-terminal blockage. The separated proteins were transferred on to a polyvinylidene fluoride (PVDF) membrane overnight by electroblotting and stained with Coomassie blue. The membrane was then destained for 15 min in 40% methanol/10% acetic acid followed by rinsing in 90% methanol/5% acetic acid followed by multiple rinses in distilled water for a total of 4 h. Finally, the membrane was dried using Whatman No. 1 filter paper and stored. The 42.4 kDa protein present on dried PVDF membrane was excised and submitted on dry ice for sequencing (The Protein Facility, Iowa State University, Ames, IA). The N-terminal protein sequencing was carried out with 494 Procise Protein/ peptide Sequencer/140C Analyzer (Perkin Elmer Inc., San Jose, CA) using Edman degradation method. The identified amino acid sequence was searched by blastp against GenBank protein database in PubMed and investigated for similar proteins present in other bacterial species. The protein was designated FomA because of the high degree of similarity to FomA OMP of F. nucleatum. 2.4. Adhesion inhibition assay with FomA The OMP of subsp. necrophorum strain 8L1 (400 mg) were incubated with EJG cells seeded in a 6-well plate treated with Karnovsky’s fixative, as mentioned previously. After overnight incubation, cells were washed with PBS twice to remove unbound proteins, followed by two washes with PBS + 0.1% NP-40 to remove OMP other than the FomA bound to the fixed cells. Subsp. necrophorum strain 8L1 was grown to an absorbance of 0.6 at 600 nm in PRAS-BHI medium and 1 ml of the culture (1 107 cells/ ml) was inoculated into each well. The bacterial cells were incubated with fixed cells for 1 h followed by vigorous washing with PBS for 3 to 4 times to remove unbound bacterial cells. The bacterial cells bound to the EJG cells
198
A. Kumar et al. / Veterinary Microbiology 176 (2015) 196–201
were collected by scraping the wells after adding 500 ml of trypsin and quickly inactivating the trypsin with 1.5 ml of anaerobic EMEM medium containing 10% heat inactivated FBS. An aliquot of 100 ml of the cell suspension containing the bacterial cells was serially diluted in PRAS-BHI and spread-plated on blood agar, and incubated in an anaerobic glove box (Forma Scientific, Marietta, OH) to enumerate the bacterial concentration. 2.5. Identification, cloning, sequencing and expression of FomA gene The primers to amplify the gene for the FomA was designed based on the published sequence of fomA of F. nucleatum. Forward primer was designed starting from the amino acid immediately following 20-amino acid signal sequence of fomA (50 -GAAGTTATGCCAGCACCTATGCCAGAA-30 ) and the reverse primer including the stop codon of the protein (50 -TTAGAAGCTAACTTTCATACC AG-30 ). The PCR conditions were as follows: 1 cycle 94 8C for 3 min followed by 35 cycles of denaturation at 94 8C for 1 min, annealing at 52 8C for 30 s, and extension at 72 8C for 2 min and a final extension at 72 8C for 4 min using Ex taq polymerase (Takara Inc., Mountain View, CA). The amplified gene of fomA from subsp. necrophorum strain 8L1 was cloned in-frame into pET45b+ vector (EMD Bioscience, Billerica, MA) by adding restriction sites at the 50 end of the primers (50 -CGGGATCCAGAAGTTATGCCTGCACC-30 with BamHI site and reverse primer 50 -AACTGCAGTTAGAAGCTAACTTTCATA-30 with PstI site). The pET45b+ vector has an N-terminal histidine tag for protein purification using a Ni+ affinity column. The cloned vector was transformed into Escherichia coli BL21 (DE3) cells and the clones were verified by PCR of several random colonies. The DNA sequence of the insert from the clone expressing the protein was determined using DNA sequencing (ACGT Inc., Wheeling, IL). The amino acid sequence of the cloned DNA was deduced from the DNA sequence and analyzed further for its similarity with homologous proteins present in the database using blastp at PubMed database. 2.6. fomA gene and FomA protein of F. necrophorum subspecies The PCR analysis was carried out with DNA extracted (Narayanan et al., 2001) from four strains of subsp. necrophorum (A25, A21, A50 and 8L1), four strains of subsp. funduliforme (B17, B29, B47, and B35) and four human clinical strains of F. necrophorum (RMA10682, RMA14786, RMA16505 and RMA16539) to determine if fomA gene is present. Further, the presence of FomA protein in bovine subsp. funduliforme and human strains of F. necrophorum was determined by SDS-PAGE analysis of OMP. 2.7. Adhesion inhibition assay with polyclonal antibody raised against the native and recombinant FomA of F. necrophorum The FomA native protein obtained from the SDS-PAGE gel and the recombinant protein expressed and purified
were sent to Cocalico Biologicals Inc. (Reamstown, PA) for polyclonal antibody production in rabbit. The cultured EJG cells were seeded in a 6-well plate (Corning Inc.) by adding 3 ml of cells at a concentration of 1 105 cells/ml and incubated in EMEM medium for 48 h. After 48 h, the cells were washed three times with sterile PBS. Bacteria (subsp. necrophorum or subsp. funduliforme) were grown to an OD600 of 0.6, washed, incubated with 1:100 dilution of the polyclonal antiserum for 1 h, and washed three times in PBS to remove unbound antibodies. These pretreated bacteria were then added to EJG cell monolayers at a multiplicity of infection of 100:1 and allowed to attach for 1 h at 37 8C. Cells were washed 4 times with sterile PBS and the EJG cells along with attached bacteria were collected using 500 ml of Trypsin-Versene (Lonza Walkersville Inc., Walkersville, MD), and neutralized quickly with 1.5 ml of anaerobic EMEM containing 10% fetal bovine serum The attached bacteria were enumerated following serial dilution and spread-plating on blood agar and incubating for 2 days at 39 8C in an anaerobic glove box. 2.8. Statistical analysis All adhesion inhibition assays of F. necrophorum to EJG cells set up in triplicates were performed a minimum of three independent replications. The data were analyzed by PROC MIXED procedure of SAS (ver. 9.2; Cary, NC) with colony counts as the outcome variable and EJG cell treatments (preincubation with OMP or anti-FomA antiserum) as the predictor variable. A P value of <0.05 was considered significant. 3. Results 3.1. Identification of an OMP with high binding affinity The fraction of extracted OMP removed after overnight incubation with Karnovsky’s fixed EJG cells showed several protein bands that did not bind to the fixed cells (lane 2 in Fig. 1). The PBS wash of Karnovsky’s fixed cells incubated with extracted OMP of F. necrophorum subsp. necrophorum revealed the proteins loosely bound to the fixed cells (lane 3 in Fig. 1). However, no protein band was detected in washing with PBS + 0.1% NP-40 (lane 5 from treated cells and 6 from control cells in Fig. 1), modified RIPA (lane 7 from treated cells and 8 from control cells in Fig. 1). The final wash with SDS-PAGE buffer containing 10% SDS revealed a band of 42.4 kDa protein (lane 10 from treated cells and lane 11 from control cells in Fig. 1). The N-terminal sequence of the identified protein was K-E-V-M-P-A-P-M-P-E-D-E. A blastp search revealed that the identified adhesin protein sequence was similar to FomA protein of F. nucleatum and Fusobacterium periodontium, and an OMP protein in Fusobacterium varium. The amino acid sequence comparison of the identified adhesin protein and homologous proteins of other fusobacterial species revealed that the adhesin protein of F. necrophorum has a 20-aa signal peptide, similar to FomA (Nakagaki et al., 2010). This signal peptide is cleaved upon its expression into the periplasm and hence, the N-terminal sequencing showed the starting amino acid as lysine instead of methionine. Further analyses using inverse PCR identified
A. Kumar et al. / Veterinary Microbiology 176 (2015) 196–201
199
the DNA sequence of signal peptide as ‘ATGAAAAAATTAGCACTTGTATTAGGTTCTTTATTAGTCATCGGTTCT GCCGCTTCTGCT’ encoding MKKLALVLGSLLVIGSAASA as the predicted signal peptide. Western blot analysis with sera from steers that were experimentally induced with liver abscesses via intraperitoneal inoculation with subsp. necrophorum strain 8L1 in a separate study showed that the protein is immunodominant (Fig. 1B). 3.2. Adhesion inhibition assay When F. necrophorum subsp. necrophorum strain 8L1 was incubated with fixed EJG cells that were saturated with the FomA protein of F. necrophorum (after removing all loosely bound OMP with stringent wash leaving only the FomA protein), there was four fold (P < 0.0001) decrease in the binding of the bacterial cells compared to the control EJG cells that were not saturated with the FomA protein (Fig. 2). Other strains of subsp. necrophorum showed similar results (data not shown). 3.3. FomA in other subspecies of F. necrophorum The PCR assay of four bovine strains of subsp. necrophorum, four bovine strains of subs. funduliforme showed that the FomA gene was present in both subspecies (Fig. 3A). However, the size of the amplified PCR product was slightly lower, in subsp. funduliforme than that of subsp. necrophorum (Fig. 3A). F. necrophorum of human origin also showed PCR product of similar size (Fig. 3A).
Fig. 2. Inhibition assay with FomA. The EJG cells were saturated with FomA and attachment study was carried out with subsp. necrophorum strain 8L1. The bacteria were incubated with fixed EJG cells and the unbound bacteria were washed off using PBS. The cells were scraped along with bound bacteria and serially diluted in PRAS-BHI for enumeration and plated on blood agar to enumerate surface bound bacteria. Error bars represent standard error.
However, the protein band of 40–45 kDa range was only present in one (RMA16505) of the four human strains (Fig. 3B). All the bovine strains of subsp. necrophorum (RA15, RA16, RA17, 8L1, A50 and A25) showed a protein band near 40–45 kDa which was absent in subsp. funduliforme (B17, B30, B35, B36) (Fig. 3B). 3.4. Cloning, sequencing and expression of FomA from subsp. necrophorum Cloning of FomA gene, using primers that were designed based on the N-terminal sequence of FomA protein of F. nucleatum, in pET45b+ vector resulted in expression of a protein containing 363 amino acids with deduced molecular weight of 42.4 kDa. The protein had a 96% homology in C and N-terminal amino acid sequences to FomA of F. nucleatum. 3.5. Adhesion inhibition assay with polyclonal antibody raised against recombinant FomA
Fig. 1. (A) Outer membrane proteins (OMP) from subsp. necrophorum were incubated overnight with fixed EJG cells and unbound protein was removed (lane 2) and washed with PBS (lanes 3 and 4) and detergents with increasing stringency (PBS + 0.1% NP-40: lanes 5 and 6, modified RIPA: lanes 7 and 8, and SDS-PAGE buffer with 10% SDS: lanes 10 and 11). No protein was detected in washing with PBS + 0.1% NP-40 and modified RIPA buffer. The final wash with SDS-PAGE buffer containing 10% SDS revealed a band of 42.4 kDa protein in cells treated with OMP (lane 10). Lane 4, 6, 8 and 11 are control cells with no added OMP. (B) Western blot of the purified protein with serum from a steer with experimentally induced liver abscesses with subsp. necrophorum.
When subsp. necrophorum strain 8L1was preincubated with polyclonal antibodies raised against the recombinant FomA, there was up to 90% reduction in the binding of bacterial cells to EJG cells (Fig. 4b). The inhibition with polyclonal antibody raised against purified native FomA protein from subsp. necrophorum (>97% as compared to serum before immunization and >98% compared to no antibody control) was slightly higher than inhibition shown with antibody raised against the recombinant FomA protein (>92% as compared with pre-immunized serum and >94% with no antibody control). 4. Discussion We have shown previously that OMP of F. necrophorum mediate bacterial adhesion to bovine endothelial cells (Kumar et al., 2013a). Trypsin treatment of F. necrophorum, preincubation of endothelial cells with OMP extracted from F. necrophorum, or polyclonal antibody raised against
200
A. Kumar et al. / Veterinary Microbiology 176 (2015) 196–201
Fig. 3. (A) The presence of FomA gene in human F. necrophorum subsp. funduliforme strains (L: 1KB DNA ladder, 1-4: 1: RMA10682, 2: RMA16505, 3: RMA14786, 4: RMA16539), bovine isolates of subsp. funduliforme (Lanes 5–8: B17, 7: B29, 8: B47, 9: B35) and subsp. necrophorum (Lanes 9–12: A25, 10: A21, 11: A50, 12: 8L1) 13: negative control. (B) On SDS-PAGE analysis of OMPs, the 42.4 kD (FomA) protein was present in all strains of subsp. necrophorum. However, among subsp. funduliforme of human and bovine origins, only human strain RMA16505 showed protein of 42.4 kDa. The arrow indicates the 42.4 kDa protein band.
the OMP decreased the binding of bacterial cells (Kumar et al., 2013a). We designed a protocol to identify an OMP that binds with high affinity to the endothelial cells. The high affinity binding OMP was isolated after serially washing the cells with detergents of increasing stringency and finally with SDS-PAGE buffer which contained 10% SDS. No protein was detected in any of the fractions except in the highly stringent SDS-PAGE buffer that contained 10% SDS which suggested that the isolated protein was tightly bound to the EJG cells. This N-terminal protein sequencing of the adhesin and further analysis using BLASTP tool showed that this protein is present in other species of Fusobacterium, such as FomA of F. nucleatum, F. periodonticum and F. varium. The FomA of F. nucleatum is a porin protein and has been shown to act as an adhesin in
Fig. 4. Inhibition assay with polyclonal antibody raised against the adhesin protein. Subsp. necrophorum when preincubated with the antibody raised against native 42.4 kDa protein or against recombinant 42.4 kDa protein, showed a significant decrease in bacterial binding to bovine endothelial cells compared to subsp. necrophorum not treated with any antibody (control) or when preincubated with rabbit preimmunization sera. Error bars represent standard error.
coaggregation of F. nucleatum with other bacteria to form biofilm in oral infections in humans (Liu et al., 2010). The EJG cells saturated with adhesin showed reduced binding of subsp. necrophorum indicating that the protein blocked the binding sites of endothelial cells. The western blot analysis of the protein using sera from steers with liver abscesses caused by subsp. necrophorum showed that this protein is immunodominant. Based on the PCR analysis, strains of subsp. necrophorum and subsp. funduliforme of cattle origin, and human strains of subsp. funduliforme carried the fomA gene. However, the SDS-PAGE analysis of the OMP of strains of subsp. funduliforme of bovine and human origin did not show any dominant band in the 40–45 kDa range, except strain 16505 of human origin. The absence of the protein could either be due to low expression or mutation in the gene causing truncation of this protein in subsp. funduliforme. Further work is needed to analyze if this protein is actually expressed in subsp. funduliforme. Subsp. necrophorum is more virulent and is more frequently isolated in the liver abscesses of cattle (Nagaraja and Chengappa, 1998). It is likely that the FomA protein of subsp. necrophorum contributes to the binding of this bacterium to the host cells surfaces and facilitate its transport from the rumen to the liver to establish hepatic infection. The polyclonal antibody raised against the purified native and recombinant FomA in rabbit when incubated with subsp. necrophorum, showed significant decrease in the bacterial binding indicating that the antibody blocked the adhesin on the surface of subsp. necrophorum. The recognition of FomA protein by sera of steers infected with subsp. necrophorum and the ability of the antisera raised
A. Kumar et al. / Veterinary Microbiology 176 (2015) 196–201
against FomA to reduce binding of subsp. necrophorum to EJG cells show that it can be used as a potential candidate for vaccine against liver abscesses in cattle. A vaccine prepared using FomA of F. nucleatum has been shown to decrease the oral infections in a mouse model (Liu et al., 2010). Studies to create a mutant and subsequent complementation analysis were not carried out because there are no established systems available to genetically manipulate F. necrophorum genome. In conclusion, we have identified an outer membrane protein of F. necrophorum which binds with high affinity to the bovine endothelial cells. This protein may be a potential candidate for the development of a vaccine to prevent fusobacterial infections in cattle. Conflict of interest statement The authors declare that there are no conflicts of interest. Acknowledgements The authors acknowledge the assistance of Mr. Joe Anderson from the Virology Department at the Kansas State University-Veterinary Diagnostic Laboratory in culturing endothelial cells. We greatly appreciate financial assistance from the College of Veterinary Medicine and the Department of Diagnostic Medicine and Pathobiology at Kansas State University. References Bavington, C., Page, C., 2005. Stopping bacterial adhesion: a novel approach to treating infections. Respiration 72, 335–344. Haase, E.M., Zmuda, J.L., Scannapieco, F.A., 1999. Identification and molecular analysis of rough-colony-specific outer membrane proteins of Actinobacillus actinomycetemcomitans. Infect. Immun. 67, 2901–2908. Han, Y.W., Ikeami, A., Rajanna, C., Kawsar, H.I., Zhou, Y., Li, M., Sojar, H.T., Genco, R.J., Kuramitsu, H.K., Deng, C.X., 2005. Identification and characterization of a novel adhesin unique to oral fusobacteria. J. Bacteriol 187, 5330–5340.
201
Huntley, J.F., Conley, P.G., Hagman, K.E., Norgard, M.V., 2007. Characterization of Francisella tularensis outer membrane proteins. J. Bacteriol. 189, 561–574. Kumar, A., Gart, E., Nagaraja, T.G., Narayanan, S., 2013a. Adhesion of Fusobacterium necrophorum to bovine endothelial cells is mediated by outer membrane proteins. Vet. Microbiol. 162, 813–818. Kumar, A., Peterson, G., Nagaraja, T.G., Narayanan, S., 2013b. Outer membrane proteins of Fusobacterium necrophorum subsp. necrophorum and subsp. funduliforme. J. Basic Microbiol., http://dx.doi.org/10.1002/jobm. 201200748. Liu, P.F., Shi, W., Zhu, W., Smith, J.W., Hsieh, S.L., Gallo, R.L., Huang, C.M., 2010. Vaccination targeting surface FomA of Fusobacterium nucleatum against bacterial co-aggregation: implication for treatment of periodontal infection and halitosis. Vaccine 28, 3496–3505. Nagaraja, T.G., Narayanan, S.K., Stewart, G.C., Chengappa, M.M., 2005. Fusobacterium necrophorum infections in animals: pathogenesis and pathogenic mechanisms. Anaerobe 11, 239–246. Nagaraja, T.G., Chengappa, M.M., 1998. Liver abscesses in feedlot cattle: a review. J. Anim. Sci. 76, 287–298. Nakagaki, H., Sekine, S., Terao, Y., Toe, M., Tanaka, M., Ito, H.O., Kawabata, S., Shizukuishi, S., Fujihashi, K., Kataoka, K., 2010. Fusobacterium nucleatum envelope protein FomA is immunogenic and binds to the salivary statherin-derived peptide. Infect. Immun. 78, 1185–1192. Narayanan, S.K., Nagaraja, T.G., Okwumabua, O., Staats, J., Chengappa, M.M., Oberst, R.D., 1997. Ribotyping to compare Fusobacterium necrophorum isolates from bovin liver abscesses, ruminal walls and ruminal contents. Appl. Environ. Microbiol. 63, 4671–4678. Narayanan, S.K., Nagaraja, T.G., Chengappa, M.M., Stewart, G.C., 2001. Cloning, sequencing, and expression of the leukotoxin gene from Fusobacterium necrophorum. Infect. Immun. 69, 5447–5455. Rocha-De-Souza, C.M., Colombo, A.V., Hirata, R., Mattos-Guaraldi, A.L., Monteiro-Leal, L.H., Previato, J.O., Freitas, A.C., Andrade, A.F., 2001. Identification of a 43-kDa outer-membrane protein as an adhesin in Aeromonas caviae. J. Med. Microbiol. 50, 313–319. Scanlan, C.M., Hathcock, T.L., 1983. Bovine rumenitis-liver abscess complex: a bacteriological review. Cornell Vet. 73, 288–297. Shinjo, T., Fujisawa, T., Mitsuoka, T., 1991. Proposal of two subspecies of Fusobacterium necrophorum (Flugge) Moore and Holdeman: Fusobacterium necrophorum subsp. necrophorum subsp. nov., nom. rev (ex Flugge 1886), and Fusobacterium necrophorum subsp. funduliforme subsp. nov., nom. rev. (ex Halle 1898). Int. J. Syst. Bacteriol. 41, 395–397. Tadepalli, S., Stewart, G.C., Nagaraja, T.G., Narayanan, S.K., 2008. Leukotoxin operon and differential expressions of the leukotoxin gene in bovine Fusobacterium necrophorum subspecies. Anaerobe 14, 13–18. Takayama, Y., Kanoe, M., Maeda, K., Okada, Y., Kai, K., 2000. Adherence of Fusobacterium necrophorum subsp. necrophorum to ruminal cells derived from bovine rumenitis. Lett. Appl. Microbiol. 30, 308–311. Tan, Z.L., Nagaraja, T.G., Chengappa, M.M., 1992. Factors affecting leukotoxin activity of Fusobacterium necrophorum. Vet. Microbiol. 32, 15–28.