Identification of an extracellular matrix protein adhesin, EmaA, which mediates the adhesion of Actinobacillus actinomycetemcomitans to collagen

Keith P. Mintz

Department of Microbiology and Molecular Genetics, Rm 110 Stafford Hall, University of Vermont, Burlington, VT 05405, USA

Correspondence
Keith P. Mintz
Keith.Mintz{at}uvm.edu


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Actinobacillus actinomycetemcomitans is an aetiologic agent in the development of periodontal and some systemic diseases in humans. This pathogen localizes to the underlying connective tissue of the oral cavity in individuals with periodontal disease. The adhesion of A. actinomycetemcomitans to extracellular matrix components of the connective tissue prompted this study to identify gene products mediating the interaction of A. actinomycetemcomitans to these molecules. A transposon mutagenesis system was optimized for use in A. actinomycetemcomitans and used to generate an insertional mutant library. A total of 2300 individual insertion transposon mutants were screened for changes in the adhesion to collagen and fibronectin. Mutants were identified which exhibited the following phenotypes: a decrease in collagen binding; a decrease in fibronectin binding; a decrease in binding to both proteins; and an increase in binding to both collagen and fibronectin. The identification of mutants defective in adhesion to the individual proteins indicates that distinct adhesins are expressed by this organism. Molecular analysis of these mutants implicated 11 independent loci in protein adhesion. One gene, emaA, is likely to encode a direct mediator of collagen adhesion, based on predicted protein features homologous to the collagen-binding protein YadA of Yersinia enterocolitica. EmaA was localized to the outer membrane, as expected for an adhesin. Reduction in fibronectin adhesion appeared to be influenced by abrogation of proteins involved in molybdenum-cofactor biosynthesis. Several other loci identified as reducing or increasing adhesion to both collagen and fibronectin are suggested to be involved in regulatory cascades that promote or repress expression of collagen and fibronectin adhesins. Collectively, the results support the hypothesis that A. actinomycetemcomitans host colonization involves afimbrial adhesins for extracellular matrix proteins, and that the expression of adhesion is modulated by global regulatory mechanisms.


Abbreviations: ECM, extracellular matrix

The GenBank/EMBL/DDBJ accession number for the complete A. actinomycetemcomitans emaA sequence reported in this paper is AY344064.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Periodontal diseases are a group of chronic inflammatory diseases of the gingiva and the supporting structures of the periodontium. These diseases are associated with the colonization of the subgingival crevice with specific anaerobic Gram-negative bacteria (Socransky & Haffajee, 1992). Activation of host immune-surveillance cells by these bacteria, or bacterial products, results in cell activation and the subsequent release of effector molecules that cause tissue damage. In concert with bacterial factors, the net result is the degradation of connective tissue and the destruction of the supporting bone, which leads to tooth loss (Baker, 2000).

Actinobacillus actinomycetemcomitans is a Gram-negative facultative anaerobic bacterium that colonizes the human oral cavity and upper respiratory tract. This bacterium is strongly associated with periodontitis in young individuals and with cases of adult periodontitis (Slots et al., 1980; Slots, 1982; Slots & Listgarten, 1988; Zambon, 1985). This pathogen has been associated with other serious human infections, including infectious endocarditis, soft-tissue abscesses and pneumonia, and it may contribute to cardiovascular disease (Beck et al., 1996; Haraszthy et al., 2000; Kaplan et al., 1989). The periodontium is believed to be the source for these non-oral diseases, but little is known about the tropism used by A. actinomycetemcomitans to colonize the oral cavity and to infiltrate and disseminate in tissues.

Pathogens have evolved diverse strategies to be successful in colonization of host tissues. A common theme amongst these pathogens is the ability to initiate infection by adhesion to specific host macromolecules under stringent or hostile conditions (Finlay, 1990). These molecules include the proteins, secreted by host cells, that form the extracellular matrix (ECM). The ECM is a biologically active tissue composed of a complex mixture of macromolecules, including multiple collagen types, fibronectin, laminin and glycosaminoglycans. The ECM not only serves a structural function but also affects a number of cellular activities, including migration, proliferation and differentiation. ECM proteins that have been described to act as a substrate for bacterial adhesion include collagens, laminin, fibronectin, fibrinogen, vitronectin and heparan sulfate (Patti et al., 1994). A large number of intracellular and extracellular human pathogens adhere to mammalian ECM proteins and have been shown to contribute to the virulence of the micro-organism (Patti et al., 1994; Westerlund & Korhonen, 1993).

A. actinomycetemcomitans invades and migrates through epithelial cells (Fives-Taylor et al., 1999), and is eventually found in contact with collagen fibres of connective tissue (Carranza et al., 1983; Gillett & Johnson, 1982). Previously, we have demonstrated the adhesion of A. actinomycetemcomitans to fibrillar collagens, with the greatest affinity for type V collagen and fibronectin. In the present study, we describe the generation and screening of a transposon mutant library to isolate variants of A. actinomycetemcomitans with altered binding to collagen and fibronectin. Mutants deficient for adhesion to each ECM protein were identified, showing that independent adhesins mediate the binding of this bacterium to collagen and fibronectin. A mediator required for collagen adhesion, EmaA (extracellular matrix protein adhesin), was identified and described. Based on the protein sequence, EmaA is predicted to be structurally related to the collagen-binding protein YadA of Yersinia enterocolitica (Roggenkamp et al., 2003), and is likely to directly mediate adhesion to collagen.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains and plasmids.
The A. actinomycetemcomitans strains developed in this study are derived from the clinical strain SUNY 465. A spontaneous rifampicin- and nalidixic acid-resistant strain, VT 1169, of A. actinomycetemcomitans (Mintz & Fives-Taylor, 2000) was utilized as the recipient strain in all conjugation experiments. All A. actinomycetemcomitans strains were grown statically in trypticase soy broth supplemented with 0·6 % yeast extract (TSBYE) in a humidified 10 % CO2 atmosphere at 37 °C. All Escherichia coli strains were grown in Luria–Bertani (LB) broth at 37 °C with aeration. The Tn10-based transposon vector pLOF/Km was a gift from Dr V. de Lorenzo, Centor de Investigatciones Biológicas, Madrid, Spain and Dr K. Timmis, GBF-National Research Centre for Biotechnology, Braunschweig, Germany. The bacterial strains and plasmids used in this study are listed in Table 1.


View this table:
[in this window]
[in a new window]
 
Table 1. Strains and plasmids used in this study

 
Transposon mutagenesis in A. actinomycetemcomitans.
The pLOF/Km transposon mutagenesis system (Tascon et al., 1993) was optimized for A. actinomycetemcomitans by replacing the aminoglycoside phosphotransferase gene (kan) within the transposon with spectinomycin adenyltransferase, aad9 (accession no. M69221), from Enterococcus faecalis (LeBlanc et al., 1991). The pLOF/Km vector was incubated with NotI to excise kan, gel purified, and used for ligation with aad9 purified as follows. The aad9 gene was isolated from pDL269 (LeBlanc et al., 1991) by incubation with NdeI and HindIII, followed by treatment with Klenow, and cloned into the EcoRV site of Bluescript SK (Stratagene). aad9 was amplified from pBS : : aad9 with primers based on the Bluescript vector sequence, which included an internal NotI restriction site (5'-CACCGCGGTGCGGCCGCTCTAG-3') and a sequence which contained a NotI restriction site engineered on the 5' end of the sequence (5'-GATAAGAATGCGGCCGCGGTCGACGGTATCGATAAGC-3'). The PCR product was restricted with NotI, gel purified and ligated with the pLOF digest. DH5{alpha} {lambda} pir transformants were selected on LB agar plates containing 50 µg spectinomycin ml–1. The pLOF/Sp (VT1542) construct was verified by restriction mapping and DNA sequencing. The plasmid containing the transposon was transformed into electrocompetent S17-1 {lambda} pir cells for conjugation with A. actinomycetemcomitans strain VT1169. Conjugation was performed as previously described (Mintz & Fives-Taylor, 2000) for 6 h on TSBYE plates containing 0·1 mM IPTG to induce transposition, and plated on selective media containing nalidixic acid and rifampicin to counter-select the E. coli donor strain, and spectinomycin to counter-select the recipient strain. Individual colonies were inoculated into wells of 96-well sterile tissue-culture plastic microtitre plates containing TSBYE/Spc (50 µg spectinomycin ml–1) and grown as described above. DMSO was added to a final concentration of 10 % to the wells, which were stored at –80 °C.

The frequency of transposition was calculated by the number of spectinomycin-resistant cells divided by the total number of cells used in the conjugation mixture.

Detection of A. actinomycetemcomitans mutants altered in binding to ECM proteins.
The transposon library was replica plated to microtitre plates containing 250 µl TSBYE and incubated for 3 days. The bacteria were dispersed by pipetting and the bacterial growth determined by measurement of OD495. Bacterial suspension (100 µl) was transferred to microtitre plates previously coated with either human type V collagen (Sigma) or fibronectin (ICN Pharmaceuticals) and blocked with BSA (1 %) in Tris-buffered saline (TBS: 20 mM Tris, 150 mM NaCl, pH 7·4). Bacterial adhesion was detected by ELISA, as described previously (Mintz & Fives-Taylor, 1999). Following incubation of the bacteria for 2 h at 37 °C, the wells were washed with TBS, and purified polyclonal antisera, raised against the whole bacterium, in TBS containing 1 % BSA, were added to detect bound bacteria. The antibodies were incubated for 1 h at room temperature and the wells washed with TBS containing 0·1 % Tween 20 (TBST). The immune complexes were located by the addition of horseradish-peroxidase-conjugated goat anti-rabbit immunoglobulins (Jackson Laboratories) in TBS containing 1 % BSA. Following incubation for 1 h, the wells were washed with TBST, and the immune complexes were detected by the addition of hydrogen peroxide and o-phenylenediamine in 80 mM citrate phosphate buffer, pH 5·0. The reaction was allowed to proceed and stopped by the addition of 4 M sulfuric acid. The absorbance values were recorded using an EL211s microplate reader (Bio-Tek). The ratio of the individual absorption reading divided by the mean absorption reading of all the wells was used to assess bacterial binding in the initial screens.

The strains from the initial screen that displayed differences in binding to human collagen and/or fibronectin were isolated, purified and screened again as follows. Mutant and parent strains were streaked for isolation and a single colony was inoculated into broth cultures and incubated overnight. The cultures were diluted 1 : 10 and allowed to grow for an additional 3 h under normal culture conditions. The bacterial cell density was calculated from an established standard curve of the parent strain, correlating OD495 with colony forming units. Equal numbers of bacteria from each strain were added to the wells in a final volume of 100 µl TSBYE. Assays were performed in triplicate, as described above, and repeated a minimum of three times.

Southern analysis.
Strains were grown in TSBYE containing 50 µg spectinomycin ml–1, and DNA was isolated using the Puregene DNA extraction Kit (Gentra Systems). Chromosomal DNA was restricted with EcoRI and the fragments separated on a 0·7 % agarose gel in TAE buffer. The DNA fragments were transferred to Hybond nylon membranes (Amersham Life Sciences) and the membranes were treated following the method of Sambrook et al. (1989). The membranes were hybridized with DNA probes conjugated with horseradish peroxidase, using the conditions suggested by the manufacturer (Amersham Life Sciences). Hybridizing fragments were visualized using the ECL detection system (Amersham Life Sciences) and exposure to photographic film (XAR-5, Eastman Kodak).

Determination of transposon integration sites within the chromosome of A. actinomycetemcomitans.
The integration site of the transposon within the A. actinomycetemcomitans chromosome was determined by DNA sequencing of inverse PCR products. Total chromosomal DNA was extracted as described above and restricted with AluI or BanI. The enzymes were inactivated by heating, and the DNA fragments ligated by the addition of T4 DNA ligase. A portion of this material was used as the template for PCR. Oligonucleotide primers were synthesized (Sigma-Genosys) based on the published aad9 nucleotide sequence (complementary strand, bases 48–67, 5'-CTCTTGCCAGTCACGTTACG-3') and a sequence adjacent to the unique AluI restriction site (bases 563–578, 5'-GGAATCATCCTCCCAAACAAG-3') or the BanI restriction sequence within the spc gene (bases 420–441, 5'-AGTCGTCGTATCTGAACCATTG-3'). The PCR products were analysed by agarose gel electrophoresis and the products were either gel purified (Qiagen) and subcloned into the T/A cloning vector pGEM T-Easy (Promega), or purified by QIAquik (Qiagen) and sequenced. DNA sequencing was performed at the University of Vermont Cancer Center DNA Analysis Facility. The generated sequences were used to search the A. actinomycetemcomitans strain HK1651 DNA database at the University of Oklahoma for homologous sequences.

Allelic replacement mutagenesis of emaA of A. actinomycetemcomitans.
Directed gene mutagenesis was achieved by conjugative transfer of a non-replicating broad-host-range plasmid modified for efficient use in A. actinomycetemcomitans (Mintz et al., 2002). An internal fragment of emaA was generated by PCR using the primers 5'-CCCTTTCTACCACTACAGATATACC-3' (corresponding to bases 80–104) and 5'-CACTGCATCAGTATCATTACGACCAC-3' (corresponding to bases 1260–1285 of the complementary stand) and SUNY 465 DNA as the template. A unique HindIII restriction site present at base 609 was used to introduce aad9 by blunt-end ligation, resulting in the disruption of the emaA sequence. The PCR fragment was ligated with the T/A cloning vector pGEM T-Easy (Promega) and transformed. The purified plasmid was incubated with HindIII and treated with Klenow (Sambrook et al., 1989). Following treatment of the plasmid with shrimp alkaline phosphatase (Roche Diagnostics), the DNA was incubated with aad9, transformed into E. coli JM109 and grown on LB agar plates containing 50 µg spectinomycin ml–1. The DNA was released by restriction with EcoRI and ligated with the mobilizable plasmid pVT1460 for conjugation (Mintz et al., 2002). Spectinomycin-resistant colonies were replica-plated on TSBYE plates containing 100 µg kanamycin ml–1. Bacteria that were Spcr, Kans were analysed by colony PCR to confirm the presence of mutant emaA in the chromosome. The complete emaA sequence was submitted to GenBank (accession no. AY344064).

RT-PCR.
RNA was isolated (Gentra Systems) following the manufacturer's instructions, and treated with amplification grade DNase I (Invitrogen) to degrade any contaminating DNA. RNA was quantified spectrophotometrically, and equal amounts of RNA from the mutant and the parent strain were used as template for RT-PCR. Sequences within the putative long-chain fatty acid CoA ligase upstream of emaA, (5'-TACGGCATGACGGAAACCAC-3' and 5'-CGCATAGCAAGGCACGATAAG-3', corresponding to the forward and reverse sequences, respectively) and the putative downstream gene, an anaerobic ribonucleoside-triphosphate reductase, (5'-GACTATTCGCCCTTCTTCCC-3' and 5'-TTGTGCCCAAAGGCAAACC-3', corresponding to the forward and reverse sequences, respectively) were used as the primers, and conditions were optimized for use with SuperScript one-step RT-PCR using Platinum Taq (Invitrogen).

Production of recombinant EmaA and generation of polyclonal antibodies.
A 21 kDa expressed fragment of EmaA, corresponding to amino acids 420–631, was used to generate polyclonal antibodies in New Zealand white rabbits (Cocalico Biologicals Inc.). The protein was expressed in E. coli BL2(DE3) cells as a GST fusion protein from pGEX-6P-1 (Amersham Biosciences). The 636 bp fragment was amplified using Vent polymerase and chromosomal DNA as the template containing engineered restriction sites for directionally cloning into the EcoRI and BamHI sites of the plasmid. Maximum protein expression was achieved by induction of exponential-phase E. coli cells with 0·1 mM IPTG after 2 h culture at 37 °C. Following induction, the cells were collected by centrifugation, the cell pellet resuspended in PBS, and the cells lysed using a French pressure mini-cell at 18 000 p.s.i. (124 200 kPa) in the presence of a protease inhibitor cocktail for bacterial lysates (Sigma). The GST-hybrid protein was purified by affinity chromatography with Sepharose 4B-glutathione resin, using the column method detailed in the manufacturer's instructions (Amersham Biosciences) at 4 °C. The immobilized GST fusion protein was cleaved with PreScission protease (Amersham Biosciences) overnight, and the GST-free recombinant protein was collected and resolved by SDS-PAGE (Laemmli, 1970). The gel was briefly stained with Coomassie Brilliant Blue, destained, and the band corresponding to the EmaA fragment was excised, soaked in water and used as the immunogen. Amino-terminal sequence analysis was performed, and the sequence was found to be identical to the predicted protein sequence.

The expressed protein immobilized on nitrocellulose was used to affinity-purify antibodies specific for the protein fragment, using a modification of the method of Talian et al. (1983). Briefly, the protein was resolved by SDS-PAGE, transferred to nitrocellulose (Towbin et al., 1979), the region of the membrane corresponding to the protein excised and treated as described previously, and blocked with 5 % non-fat dried milk dissolved in TBS. The antiserum was applied to the nitrocellulose and incubated for 2 h at room temperature. The antibodies were dissociated from the protein by incubation with 0·2 M glycine, pH 2·7, and neutralized by the addition of 5 M NaOH.

Isolation of outer-membrane proteins.
Cells from a 200 ml overnight culture of A. actinomycetemcomitans were collected by centrifugation (5860 g, 15 min) and washed with PBS (10 mM phosphate, 150 mM NaCl, pH 7·4). The resulting cell pellet was resuspended in 3 ml 10 mM HEPES, pH 7·4, containing 1 mM phenylmethanesulfonyl fluoride. Bacterial membranes were isolated by disruption of the bacteria using a French pressure mini-cell at 18 000 p.s.i. (124 200 kPa), following the protocol of Nikaido et al. (1997). Intact bacteria and debris were collected by centrifugation at 7650 g for 10 min. The supernatant was carefully removed and centrifuged at 100 000 g for 30 min. The pellet was resuspended in 10 mM HEPES, pH 7·4, and recentrifuged. The membranes were resuspended in 10 mM HEPES, pH 7·4, and sodium N-lauroylsarcosine was added to a final concentration of 1 % and incubated at room temperature for 30 min. Based on bacterial membrane protein solubility in sarcosinate (Filip et al., 1973; Nikaido, 1997), outer-membrane proteins are insoluble in the detergent. The mixture was centrifuged at 15 600 g for 30 min and the pellet was resuspended in 10 mM HEPES, pH 7·4, and centrifuged as above. The final pellet was resuspended in 10 mM HEPES, pH 7·4. Protein concentration was determined by absorbance at 280 nm. The insoluble proteins were solubilized in 1 % SDS before determination of the protein concentration.

Polyacrylamide gel electrophoresis and immunoblot analysis of proteins.
Equal concentrations of proteins were boiled for 10 min in Laemmli sample buffer, before application to 5–15 % polyacrylamide-SDS gels. The separated proteins were transferred to nitrocellulose and probed with the affinity-purified antibodies. The immune complexes were detected using horseradish peroxidase-conjugated goat anti-rabbit immunoglobulins (Jackson Laboratories) and visualized using the ECL detection system (Amersham Life Sciences) and exposure to photographic film (XAR-5, Eastman Kodak).


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Development and characterization of a transposon mutagenesis system for A. actinomycetemcomitans
During the initial phase of this work, a functional transposon mutagenesis system in this strain of A. actinomycetemcomitans (VT1169) was not readily available. Subsequently, a mutagenesis system was published for A. actinomycetemcomitans (Thomson et al., 1999). Therefore, we employed a transposon system that had been successfully used in the swine pathogen A. pleuropneumoniae (Tascon et al., 1993) to generate transposon mutants in A. actinomycetemcomitans. This system involved the use of a suicide conjugative plasmid, pLOF/Km, carrying a mini-Tn10 with an IPTG-inducible transposase located outside the mobile element (Herrero et al., 1990).

Initial conjugation studies with this plasmid and the recipient A. actinomycetemcomitans strain VT1169 resulted in the generation of thousands of transconjugants. However, the frequency of transposition events decreased with repeated studies. The reason for this phenomenon remains elusive. To generate consistent numbers of transconjugants in the A. actinomycetemcomitans strain used in this study, it was necessary to exchange the antibiotic cassette encoding kanamycin resistance for a cassette encoding spectinomycin resistance. The kanamycin gene in the mini-Tn10 was replaced with the spectinomycin gene (aad9) from Enterococcus faecalis, which has been used extensively in this strain of A. actinomycetemcomitans. This modified plasmid, pLOF/Sp (pVT1542), gave consistent high levels of transposon mutants not obtained with pLOF/Km. Characterization of this transposon system in A. actinomycetemcomitans is described below.

Transposon insertion was efficient and appeared random. Of the transconjugants, 98 % were sensitive to ampicillin (plasmid marker), which indicated that the plasmid was not present in the majority of the transconjugants. Southern analysis of 15 randomly selected transconjugants demonstrated that the transposon integrated randomly into the chromosome of these transconjugants (Fig. 1). In most instances, a single copy of the transposon was present in the chromosome. However, infrequent integration of the plasmid was observed (Fig. 1, lane 12, the faster-migrating hybridizing band corresponded to DNA containing vector sequence following Southern analysis with the vector as the probe). The stability of the transposon within the chromosome of the transconjugants was determined by growing 10 transconjugants in non-selective media for 8 days, followed by replica plating on TSBYE agar plates with or without spectinomycin and Southern blot analysis of the same strains at the start and end of the incubation period. The hybridization pattern of the strains grown in the absence of antibiotic was identical to the same strains grown in the presence of spectinomycin. Collectively, the spc-modified transposon integrates as a stable, single copy genetic element, randomly in the genome. The apparent transposition frequency was calculated to be 2·1x10–4. This is similar to the frequency of transposition determined for another transposon mutagenesis system developed for A. actinomycetemcomitans (Thomson et al., 1999).



View larger version (134K):
[in this window]
[in a new window]
 
Fig. 1. Southern blot analysis of random A. actinomycetemcomitans transposon mutants. DNA was isolated from randomly selected transposon insertion mutants and restricted with EcoRI. The DNA fragments were separated by electrophoresis on a 0·7 % agarose gel and transferred to a charged nylon membrane by diffusion. The membrane was probed with the DNA of the antibiotic marker of the transposon labelled with horseradish peroxidase. The first lane represents parent strain VT1169 chromosomal DNA. Lanes 1–15 represent DNA isolated from the transposon mutants.

 
Identification of A. actinomycetemcomitans strains altered in binding to collagen and fibronectin
The transposon mutants derived from the library were characterized by ELISA for adhesion to the ECM proteins collagen and fibronectin (Mintz & Fives-Taylor, 1999). Sixteen mutants were chosen for further analysis. The chromosomal DNA was extracted, and the integration site of the transposon was determined by inverse PCR followed by sequencing of the PCR product. Search for homologous DNA sequences in the A. actinomycetemcomitans genome database resulted in high sequence homology with the prototypic strain for 15 of the 16 transposon mutants. The DNA sequence derived from the remaining mutant (1D2) was absent from the genomic database sequence. The ORFs of the affected genes were determined, and 11 individual ORFs were identified. The translated protein sequence of the mutants was compared for homologous sequences in the available protein sequence databases (Table 2). The level of adhesion to collagen (col) and fibronectin (fn) of the mutants was designated as normal (+), decreased (–) or increased (++), as compared with the wild-type control. The mutants were classified into four categories: 1) col fn+, 2) col+ fn, 3) col fn and 4) col++ fn++ (Fig. 2).


View this table:
[in this window]
[in a new window]
 
Table 2. Protein homology of ECM-protein-adhesion transposon mutants

Percentage binding is relative to the parental strain.

 


View larger version (33K):
[in this window]
[in a new window]
 
Fig. 2. Adhesion of A. actinomycetemcomitans transposon mutant strains to collagen and fibronectin. Insertional mutants were tested for adhesion to the ECM proteins type V collagen and fibronectin, as described in Methods. Data are represented as a percentage of adhesion of the parent strain, normalized to 100 %. The level of adhesion to collagen (col) and fibronectin (fn) of the mutants is designated as: normal (+), decreased (–) or increased (++). Four categories of A. actinomycetemcomitans adhesion mutants were delineated. The bars represent an individual strain, representative of each of the four phenotypes. Error bars represent the standard deviation of the mean.

 
The existence of mutants that displayed an exclusive fibronectin-adhesion defect (col+ fn) was attributed to the insertion of the transposon in two ORFs, moaA and moeA, associated with the biosynthesis of a molybdenum cofactor (MoCo). MoCo is a crucial component of various enzymes catalysing important redox reactions (Kisker et al., 1997). The A. actinomycetemcomitans MoaA (molybdenum cofactor biosynthesis protein A) deduced sequence shares 78–79 % sequence identity with the Haemophilus influenzae and Pasteurella multocida proteins. MoaA is involved in the first step of MoCo biosynthesis: formation of the sulfur-free pterin precursor Z from a guanosine derivative (Rajagopalan, 1996). The MoeA protein of A. actinomycetemcomitans also shares a high degree of identity to the proteins of H. influenzae and P. multocida. Although this protein is reported to be involved in MoCo biosynthesis, the exact function of this protein is unknown (Rajagopalan, 1996).

ECM-binding mutants reduced for both collagen and fibronectin (col fn) were associated with insertion of the transposon in genes transcribing global regulatory proteins most closely related to P. multocida and Haemophilus somnus cAMP receptor protein (Crp), also known as catabolite activator protein (CAP), and Mlc. These proteins have been thoroughly studied in the regulation of carbohydrate metabolism (Bruckner & Titgemeyer, 2002). However, they may be pleiotropic regulators for other biosynthetic pathways.

Several mutants were identified where insertion resulted in a gain of function: hyper-binding to both substrates (col++ fn++). These proteins include: 1) the heat-modifiable protein (OMP34) of A. actinomycetemcomitans (Wilson, 1991), 2) a protein homologous to the putative membrane protein (YbjE) of Shigella flexneri, 3) a protein homologous to a putative N6-adenine-specific DNA methylase of Salmonella typhimurium, and 4) adenylate cyclase (CyaA) of multiple bacterial taxa, which generates cAMP, a common global regulatory molecule in biological systems (Botsford & Harman, 1992).

The mutants that displayed a decrease in collagen binding, without an appreciable effect on fibronectin binding (col fn+), were found to have the transposon inserted in the same ORF. Two distinct integration sites were identified, with one 184 bp downstream to the other, within the amino terminus of the predicted protein. The gene associated with the phenotype of these mutants was designated emaA (extracellular matrix protein adhesin A).

The emaA sequence from VT1169 was determined to be 99 % identical to the sequence found in the genomic sequence for the prototypic A. actinomycetemcomitans strain HK1651. The gene encodes a putative 201 kDa protein. The protein sequence (Fig. 3) shares sequence homology with the well-characterized collagen-binding protein, YadA, of Yersinia enterocolitica. The amino terminus of the EmaA protein contains sequences that are identical to those required for the binding of YadA to collagen, or contains conserved amino-acid substitutions within these sequences (Tahir et al., 2000). The carboxyl terminus of EmaA contains a conserved domain homologous with the cell-membrane anchor domain of YadA (Tamm et al., 1993). Other conserved features or structural elements, similar to those of the Oca family of adhesins (which includes YadA), are contained in this sequence (Roggenkamp et al., 2003).



View larger version (65K):
[in this window]
[in a new window]
 
Fig. 3. Deduced A. actinomycetemcomitans EmaA protein sequence. The amino-acid sequence was derived from the A. actinomycetemcomitans VT1169 genetic sequence of emaA. Italic type, putative signal peptide sequence; bold type, putative collagen-binding sequences; underlined type, degenerate ~14-amino-acid repeat sequences; underlined and italic type, A. actinomycetemcomitans amino-acid sequences conserved in the ‘neck’ region of YadA; bold and italic type, coil-coiled motifs, as determined by COILS version 2.1; bold and underlined type, sequence homologous to membrane-binding domain of YadA.

 
The emaA ORF identified in SUNY 465 encodes a single protein. In contrast, the sequence from A. actinomycetemcomitans strain UT32 (Li et al., 2004) describes two translation products, ApiB and C, from the emaA ORF. These two sequences are designated surface proteins which may be involved with the invasion of this pathogen into epithelial cells. One protein (ApiB) is identical to amino acids 1–1247 of EmaA, but contains an additional 22 amino acids absent from the EmaA sequence. The second UT32 sequence (ApiC) is identical to amino acids 1245–1932, but EmaA contains an additional 33 amino acids at the carboxyl terminus.

Characterization of a collagen-binding protein of A. actinomycetemcomitans
The loss-of-function phenotype related to emaA : : Tn10dSpc was confirmed by generating an emaA-directed insertion mutant. An allelic replacement mutant was generated and tested for binding to collagen and fibronectin. The emaA : : spc and emaA : : Tn10dSpc mutants both displayed a similar decrease in binding to collagen, as compared with the parent strain (51±6 and 46±4 %, respectively), while maintaining wild-type levels of binding to fibronectin (91±10 and 105±15 %, respectively). The binding of the emaA : : spc mutant to other fibrillar collagen types (I, II and III) was also reduced when compared with the parent strain (data not shown). These data indicate that inactivation of emaA is not attributable to pleiotropic effects of the integration of the transposon, and that disruption of this gene results in a general defect for collagen binding.

Putative polar effects of the inactivation of emaA on the transcription of upstream and downstream genes were investigated by RT-PCR. Primer sets selected to generate ~400 bp internal fragments were used in these reactions with RNA extracted from emaA : : spc and the parent strain. Products were generated for the putative long-chain fatty acid CoA ligase (AAC21681), upstream of the start codon of emaA, and a putative anaerobic ribonucleoside triphosphate reductase, NrdD (AAK03024), downstream of emaA, from RNA extracted from both the emaA mutant and the parent strain (Fig. 4). In addition, analysis of the emaA sequence indicates the presence of a stem–loop structure ({Delta}G=–13 kcal mol–1) immediately following the transcriptional stop codon, consistent with a Rho-independent transcriptional termination signal. These data suggest that the disruption of emaA does not interfere with the gene expression of the surrounding genes, and that the reduction in the binding to collagen is exclusively associated with the inactivation of emaA.



View larger version (90K):
[in this window]
[in a new window]
 
Fig. 4. Transcriptional analysis of surrounding genes following inactivation of A. actinomycetemcomitans emaA. RNA was extracted from emaA : : spc and the parent strain and used in RT-PCR reactions using primer sets to generate products corresponding to the putative genes upstream and downstream of emaA. Products were separated on 0·7 % agarose gels containing ethidium bromide. (A) RT-PCR products generated with RNA from the parent strain (+) or the mutant strain (–) using primers corresponding to the gene upstream of emaA; (B) products generated using the primer set corresponding to the ORF downstream of emaA.

 
The localization of EmaA and verification of the predicted molecular mass was determined by immunoblotting. Polyclonal antibodies were generated against a 21 kDa EmaA internal protein fragment expressed in E. coli as a GST fusion protein and cleaved with PreScission protease (Amersham Biosciences) to remove the GST. The product was further purified by SDS-PAGE, and the band corresponding to the EmaA fragment was excised and used as the immunogen in rabbits. Sera were affinity purified and used to detect EmaA in immunoblots.

Membrane fractions of both the wild-type and mutant emaA strain were generated and the outer-membrane proteins recovered by differential solubilization in sodium lauroylsarcosine. Equivalent amounts of outer-membrane proteins were separated by SDS-PAGE and either stained for protein (Fig. 5, panel A) or transferred to nitrocellulose and probed with affinity-purified antibodies (Fig. 5, panel B). Protein staining (Coomassie Brilliant Blue) revealed very little difference between the profiles of the membrane preparations, which contained few proteins with a molecular mass greater than 90 kDa. However, the immunoblots clearly showed a difference between the two strains. An immunoreactive species greater than 200 kDa was found in the outer-membrane fraction of the parent strain, but was completely absent in the emaA mutant. Some cross-reactivity of the anti-EmaA antibodies with other membrane proteins was present. The band between the 29 kDa and 36 kDa markers was the heat-modifiable protein Omp34, which has been demonstrated to bind the Fc portion of immunoglobulins (Mintz & Fives-Taylor, 1994). Collectively, these data suggest that emaA encodes a ~200 kDa outer-membrane protein, which mediates the adhesion of A. actinomycetemcomitans to collagen.



View larger version (56K):
[in this window]
[in a new window]
 
Fig. 5. Localization of EmaA to the outer membrane fraction of A. actinomycetemcomitans. Total membrane fractions were obtained from the parent strain (+) and the isogenic mutant (–) by cell lysis in a French press, followed by differential centrifugation. Membranes were solubilized in sodium N-lauroylsarcosine to generate outer-membrane proteins. Equal amounts of proteins were separated in 5–15 % gradient polyacrylamide gels by SDS-PAGE. The gels were either stained with Coomassie Brilliant Blue for total protein (A) or transferred to nitrocellulose and probed with affinity-purified anti-EmaA antibodies (B). The arrow denotes the presence of EmaA protein.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In Gram-negative bacteria, specific ECM proteins may be recognized by single or multipurpose adhesins, which in turn may interact with several ECM components. The plasmid-encoded outer-membrane protein YadA mediates the specific binding of enteropathogenic yersiniae to various types of collagen, as well as binding to laminin and the cellular form of fibronectin (Schulze-Koops et al., 1992, 1993; Tamm et al., 1993; Tertti et al., 1992). This is in contrast to the Dr fimbriae of E. coli and the type 3 fimbriae of Klebsiella pneumoniae, which bind specifically to one type of collagen (Tarkkanen et al., 1990; Westerlund et al., 1989). The CadF protein of Campylobacter jejuni is specific for the binding of this enteric pathogen to fibronectin (Konkel et al., 1997). In addition, bacteria may contain multiple surface proteins that interact with several ECM proteins. The oral spirochaete Treponema denticola expresses several proteins that appear to bind to both fibronectin and laminin (Fenno et al., 1996; Umemoto et al., 1993). In this study, we have isolated mutants that display a wild-type level of adhesion to one, but reduced adhesion to another ECM protein (col fn+, col+ fn). In our previous study (Mintz & Fives-Taylor, 1999), we demonstrated differences in the binding isotherms to collagen and fibronectin of trypsin-treated bacteria. In addition, differences found in the binding phenotype to these proteins among strains of A. actinomycetemcomitans suggest the presence of separate ECM protein adhesins. The biochemical and genetic data strongly suggest that A. actinomycetemcomitans expresses separate adhesins to mediate the interaction with individual ECM proteins. However, the presence of residual binding activity in these mutants suggests the existence of additional adhesins for collagen and fibronectin. Therefore, multiple adhesins may exist for each specific ECM protein, but one cannot exclude the possibility that other putative adhesins have broader substrate specificity than the adhesins identified in this study.

Excluding emaA, the insertion of the transposon in the other mutants can have polar effects on the transcription of surrounding genes. Transcriptional analysis must be performed to define the effects of these insertional mutations. With this caveat stated, the ECM protein binding mutants identified by this genetic screen are indicative of structural proteins and proteins which may act directly or indirectly to regulate adhesin gene expression and/or synthesis. The transposon insertions associated with a defect exclusively in fibronectin adhesion were found in distinct, non-contiguous operons associated with molybdenum cofactor (MoCo) biosynthesis. MoCo is an important cofactor of a number of biosynthetic pathways in which the cofactor itself may be required for adhesin synthesis. Alternatively, the metal moiety of the cofactor (Mo2+) may act as a signalling molecule, similar to other metal ions that modulate virulence-associated factors (Groisman, 2001; Hantke, 2001; Kehres & Maguire, 2003; Straley et al., 1993). In addition, MoeA, which is structurally related to the carboxyl-terminal domain of gephyrin (Sola et al., 2001), may serve a structural role in fibronectin adhesion. Gephyrin regulates receptor clustering by forming a subsynaptic protein scaffolding, which anchors the receptors to the cytoskeleton (Kneussel & Betz, 2000). This implies that MoeA may act to stabilize or anchor the fibronectin adhesin or associated molecules.

Loci that reduce or enhance both fibronectin and collagen adhesion (mlc, crp, cyaA) are associated with global regulatory molecules that impact the expression of the actual adhesin. Both Mlc and CRP are usually associated with carbohydrate metabolism, but can activate or repress a large number of promoters outside carbohydrate regulation (Bruckner & Titgemeyer, 2002). The adenylate cyclase and the receptor protein signalling pathway are essential for the global regulation of transcription of the type III secretion systems of both Pseudomonas aeruginosa and Y. enterocolitica (Petersen & Young, 2002; Wolfgang et al., 2003). The inactivation of this signalling pathway in Y. enterocolitica results in a reduction in the virulence of this pathogen (Petersen & Young, 2002).

Several other loci that impacted ECM protein adhesion cannot be readily classified by probable function, but speculation is still warranted. Increased collagen and fibronectin adhesion was associated with loci encoding outer-membrane proteins. It is speculated that absence of the proteins influences the membrane adhesins or allows more adhesins to be integrated into the outer membrane. These mutants may prove to be of value in identifying additional ECM protein mediators if a greater number of adhesins are expressed.

The collagen adhesin identified in this study is likely a member of a novel class of non-fimbrial oligomeric coiled-coil adhesins (Oca) (Roggenkamp et al., 2003). The members of this family, in addition to YadA, include UspA1 and UspA2 of Moraxella catarrhalis, and the autotransporter Hia of H. influenzae (Hoiczyk et al., 2000). These proteins share conserved features or structural elements. These include: (i) an amino-terminal Sec-dependent secretion signal, (ii) a head domain consisting of degenerate 14-residue repeats, (iii) a highly conserved neck region(s), (iv) a stalk domain of variable length with a high probability of coiled-coil formation, and (v) a carboxyl-terminal membrane-anchor domain with conserved structural features. In addition, members of this protein class do not contain cysteine residues in the mature polypeptide. The known functions of these adhesins are binding to eukaryotic cells and ECM proteins. The binding domains mediating adherence are usually located in non-conserved regions of the amino terminus of the protein (Roggenkamp et al., 2003).

The EmaA protein consists of 1965 amino acids, and there are no cysteine residues in the mature protein. A single cysteine is present in the putative signal sequence. The amino terminus of this protein contains the approximately 14-residue degenerate repeats (alternating pattern of branched aliphatic and small residues, followed by a position consisting mainly of Ala, Gly, Ser or Thr), as found in the head region of YadA (Hoiczyk et al., 2000). Surrounding and included within this region are multiple sequences identical to the collagen-binding consensus sequence NSVAIG-S, or displaying conserved amino-acid substitutions (Tahir et al., 2000). Juxtaposed to the putative collagen-binding domain is the sequence TDAVNVAQL, which is identical to the neck sequence of YadA (Hoiczyk et al., 2000). A putative stalk region is present, which contains one region with a high probability of a coiled-coil domain, as determined by COILS version 2.1 (Window=21 score of 0·62). The carboxyl terminus is separated from the stalk region by another neck sequence followed by a coiled-coil segment (Window=21 score of 0·75) and a membrane-anchor domain formed by four transmembrane {beta}-strands, as suggested by the A. actinomycetemcomitans sequence alignment with the YadA sequence listed in Hoiczyk et al. (2000).

The emaA sequence predicts a protein with a molecular mass of 201 kDa, and a protein greater than 200 kDa was observed in immunoblots of membrane proteins. Based on the protein staining, EmaA is not a significant protein component of the bacterial membrane. Although this may be indicative of the relative abundance of the adhesin within the membrane of the bacterium, it does not reflect the protein's biological significance. The position of the protein in the immunoblot makes estimation of the molecular mass unreliable. Therefore, EmaA may exist as heat-stable aggregates in SDS-PAGE, which is a characteristic of Oca proteins (Roggenkamp et al., 2003). Alternatively, posttranslational modification of the protein may alter the protein's electrophorectic mobility, which implies that the ORF encodes a single protein. Further biophysical and structural studies are warranted to resolve these issues.

Surface ultrastructual analysis of the A. actinomycetemcomitans strain used in this study demonstrates the presence of small vesicles and fibrillar membranous extensions with knob-like ends on the surface of the bacterium (Meyer & Fives-Taylor, 1994). The YadA and UspA adhesins form ‘lollipop’-shaped structures on the outer membrane of the bacterium, as detected by electron microscopy (Hoiczyk et al., 2000). Based on the structural homology of EmaA with these proteins, one would predict the presence of similar structures on the surface of A. actinomycetemcomitans. Investigations into the surface structures of the wild-type and emaA mutant strains are being conducted.

Bacterial adhesins that specifically recognize ECM proteins have been characterized for a number of pathogens (Patti et al., 1994). The results of this study are consistent with the existence of distinct mediators for fibronectin and collagen adhesion. However, it is possible that an adhesin that mediates the binding to both ECM proteins is also present. A collagen-specific adhesin, EmaA, has been identified which contains protein sequence information and structural characteristics that are analogous to a family of established ECM-binding proteins. In addition, ubiquitous signalling molecules have also been identified that regulate the adhesion of this pathogen to these substrates. Further studies will be required to determine if these pathways are implicated in the regulation of other virulence determinants expressed by this pathogen.


   ACKNOWLEDGEMENTS
 
We are grateful to Michael Florucci and Solomon Lakey for assistance with the development and screening of the transposon library, and to Denise Fernandez and Paula Austin for their expert technical assistance and compiling the emaA sequence. We would also like to thank Murry Stein and Paula Fives-Taylor for helpful discussions and critical reading of the manuscript. This work was supported by Public Health Service grant RO1-DE13824.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Baker, P. J. (2000). The role of immune responses in bone loss during periodontal disease. Microbes Infect 2, 1181–1192.[CrossRef][Medline]

Beck, J., Garcia, R., Heiss, G., Vokonas, P. S. & Offenbacher, S. (1996). Periodontal disease and cardiovascular disease. J Periodontol 67, 1123–1137.[Medline]

Botsford, J. L. & Harman, J. G. (1992). Cyclic AMP in prokaryotes. Microbiol Rev 56, 100–122.[Medline]

Bruckner, R. & Titgemeyer, F. (2002). Carbon catabolite repression in bacteria: choice of the carbon source and autoregulatory limitation of sugar utilization. FEMS Microbiol Lett 209, 141–148.[CrossRef][Medline]

Carranza, F. A., Jr, Saglie, R., Newman, M. G. & Valentin, P. L. (1983). Scanning and transmission electron microscopic study of tissue-invading microorganisms in localized juvenile periodontitis. J Periodontol 54, 598–617.[Medline]

Fenno, J. C., Muller, K. H. & McBride, B. C. (1996). Sequence analysis, expression, and binding activity of recombinant major outer sheath protein (Msp) of Treponema denticola. J Bacteriol 178, 2489–2497.[Abstract]

Filip, C., Fletcher, G., Wulff, J. L. & Earhart, C. F. (1973). Solubilization of the cytoplasmic membrane of Escherichia coli by the ionic detergent sodium-lauryl sarcosinate. J Bacteriol 115, 717–722.[Medline]

Finlay, B. B. (1990). Cell adhesion and invasion mechanisms in microbial pathogenesis. Curr Opin Cell Biol 2, 815–820.[Medline]

Fives-Taylor, P. M., Meyer, D. H., Mintz, K. P. & Brissette, C. (1999). Virulence factors of Actinobacillus actinomycetemcomitans. Periodontology 2000 20, 136–167.[Medline]

Gillett, R. & Johnson, N. W. (1982). Bacterial invasion of the periodontium in a case of juvenile periodontitis. J Clin Periodontol 9, 93–100.[Medline]

Groisman, E. A. (2001). The pleiotropic two-component regulatory system PhoP-PhoQ. J Bacteriol 183, 1835–1842.[Free Full Text]

Hantke, K. (2001). Iron and metal regulation in bacteria. Curr Opin Microbiol 4, 172–177.[CrossRef][Medline]

Haraszthy, V. I., Zambon, J. J., Trevisan, M., Zeid, M. & Genco, R. J. (2000). Identification of periodontal pathogens in atheromatous plaques. J Periodontol 71, 1554–1560.[Medline]

Herrero, M., de Lorenzo, V. & Timmis, K. N. (1990). Transposon vectors containing non-antibiotic resistance selection markers for cloning and stable chromosomal insertion of foreign genes in Gram-negative bacteria. J Bacteriol 172, 6557–6567.[Medline]

Hoiczyk, E., Roggenkamp, A., Reichenbecher, M., Lupas, A. & Heesemann, J. (2000). Structure and sequence analysis of Yersinia YadA and Moraxella UspAs reveal a novel class of adhesins. EMBO J 19, 5989–5999.[Abstract/Free Full Text]

Kaplan, A. H., Weber, D. J., Oddone, E. Z. & Perfect, J. R. (1989). Infection due to Actinobacillus actinomycetemcomitans: 15 cases and review. Rev Infect Dis 11, 46–63.[Medline]

Kehres, D. G. & Maguire, M. E. (2003). Emerging themes in manganese transport, biochemistry and pathogenesis in bacteria. FEMS Microbiol Rev 27, 263–290.[CrossRef][Medline]

Kisker, C., Schindelin, H. & Rees, D. C. (1997). Molybdenum-cofactor-containing enzymes: structure and mechanism. Annu Rev Biochem 66, 233–267.[CrossRef][Medline]

Kneussel, M. & Betz, H. (2000). Clustering of inhibitory neurotransmitter receptors at developing postsynaptic sites: the membrane activation model. Trends Neurosci 23, 429–435.[CrossRef][Medline]

Konkel, M. E., Garvis, S. G., Tipton, S. L., Anderson, D. E., Jr & Cieplak, W., Jr (1997). Identification and molecular cloning of a gene encoding a fibronectin-binding protein (CadF) from Campylobacter jejuni. Mol Microbiol 24, 953–963.[Medline]

Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685.[Medline]

LeBlanc, D. J., Lee, L. N. & Inamine, J. M. (1991). Cloning and nucleotide base sequence analysis of a spectinomycin adenyltransferase AAD(9) determinant from Enterococcus faecalis. Antimicrob Agents Chemother 35, 1804–1810.[Medline]

Li, L., Matevski, D., Aspiras, M., Ellen, R. P. & Lepine, G. (2004). Two epithelial cell invasion-related loci of the oral pathogen Actinobacillus actinomycetemcomitans. Oral Microbiol Immunol 19, 16–25.[CrossRef][Medline]

Meyer, D. H. & Fives-Taylor, P. M. (1994). Characteristics of adherence of Actinobacillus actinomycetemcomitans to epithelial cells. Infect Immun 62, 928–935.[Abstract]

Mintz, K. P. & Fives-Taylor, P. M. (1994). Identification of an immunoglobulin Fc receptor of Actinobacillus actinomycetemcomitans. Infect Immun 62, 4500–4505.[Abstract]

Mintz, K. P. & Fives-Taylor, P. M. (1999). Binding of the periodontopathogen, Actinobacillus actinomycetemcomitans, to extracellular matrix proteins. Oral Microbiol Immunol 14, 109–116.[CrossRef][Medline]

Mintz, K. P. & Fives-Taylor, P. M. (2000). impA, a gene coding for an inner membrane protein, influences colonial morphology of Actinobacillus actinomycetemcomitans. Infect Immun 68, 6580–6586.[Abstract/Free Full Text]

Mintz, K. P., Brissette, C. & Fives-Taylor, P. M. (2002). A recombinase A-deficient strain of Actinobacillus actinomycetemcomitans constructed by insertional mutagenesis using a mobilizable plasmid. FEMS Microbiol Lett 206, 87–92.[CrossRef][Medline]

Nikaido, H. (1997). Isolation of outer membranes. In Bacterial Pathogenesis: Selected Methods in Enzymology, pp. 113–122. Edited by V. L. Clark & P. M. Bavoil. San Diego, CA: Academic Press.

Patti, J. M., Allen, B. L., McGavin, M. J. & Hook, M. (1994). MSCRAMM-mediated adherence of microorganisms to host tissues. Annu Rev Microbiol 48, 585–617.[CrossRef][Medline]

Petersen, S. & Young, G. M. (2002). Essential role for cyclic AMP and its receptor protein in Yersinia enterocolitica virulence. Infect Immun 70, 3665–3672.[Abstract/Free Full Text]

Rajagopalan, K. V. (1996). Biosynthesis of the molybdenum cofactor. In Escherichia Coli and Salmonella; Cellular and Molecular Biology, pp. 674–679. Edited by F. C. Neidhardt III and others. Washington, DC: American Society for Microbiology.

Roggenkamp, A., Ackermann, N., Jacobi, C. A., Truelzsch, K., Hoffmann, H. & Heesemann, J. (2003). Molecular analysis of transport and oligomerization of the Yersinia enterocolitica adhesin YadA. J Bacteriol 185, 3735–3744.[Abstract/Free Full Text]

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Schulze-Koops, H., Burkhardt, H., Heesemann, J., von der Mark, K. & Emmrich, F. (1992). Plasmid-encoded outer membrane protein YadA mediates specific binding of enteropathogenic yersiniae to various types of collagen. Infect Immun 60, 2153–2159.[Abstract]

Schulze-Koops, H., Burkhardt, H., Heesemann, J., Kirsch, T., Swoboda, B., Bull, C., Goodman, S. & Emmrich, F. (1993). Outer membrane protein YadA of enteropathogenic yersiniae mediates specific binding to cellular but not plasma fibronectin. Infect Immun 61, 2513–2519.[Abstract]

Slots, J. (1982). Selective medium for isolation of Actinobacillus actinomycetemcomitans. J Clin Microbiol 15, 606–609.[Medline]

Slots, J. & Listgarten, M. A. (1988). Bacteroides gingivalis, Bacteroides intermedius and Actinobacillus actinomycetemcomitans in human periodontal diseases. J Clin Periodontol 15, 85–93.[Medline]

Slots, J., Reynolds, H. S. & Genco, R. J. (1980). Actinobacillus actinomycetemcomitans in human periodontal disease: a cross-sectional microbiological investigation. Infect Immun 29, 1013–1020.[Medline]

Socransky, S. S. & Haffajee, A. D. (1992). The bacterial etiology of destructive periodontal disease: current concepts. J Periodontol 63, 322–331.[Medline]

Sola, M., Kneussel, M., Heck, I. S., Betz, H. & Weissenhorn, W. (2001). X-ray crystal structure of the trimeric N-terminal domain of gephyrin. J Biol Chem 276, 25294–25301.[Abstract/Free Full Text]

Straley, S. C., Plano, G. V., Skrzypek, E., Haddix, P. L. & Fields, K. A. (1993). Regulation by Ca2+ in the Yersinia low-Ca2+ response. Mol Microbiol 8, 1005–1010.[Medline]

Tahir, Y. E., Kuusela, P. & Skurnik, M. (2000). Functional mapping of the Yersinia enterocolitica adhesin YadA. Identification of eight NSVAIG - S motifs in the amino-terminal half of the protein involved in collagen binding. Mol Microbiol 37, 192–206.[CrossRef][Medline]

Talian, J. C., Olmsted, J. B. & Goldman, R. D. (1983). A rapid procedure for preparing fluorescein-labeled specific antibodies from whole antiserum: its use in analyzing cytoskeletal architecture. J Cell Biol 97, 1277–1282.[Abstract]

Tamm, A., Tarkkanen, A. M., Korhonen, T. K., Kuusela, P., Toivanen, P. & Skurnik, M. (1993). Hydrophobic domains affect the collagen-binding specificity and surface polymerization as well as the virulence potential of the YadA protein of Yersinia enterocolitica. Mol Microbiol 10, 995–1011.[Medline]

Tarkkanen, A. M., Allen, B. L., Westerlund, B., Holthofer, H., Kuusela, P., Risteli, L., Clegg, S. & Korhonen, T. K. (1990). Type V collagen as the target for type-3 fimbriae, enterobacterial adherence organelles. Mol Microbiol 4, 1353–1361.[Medline]

Tascon, R. I., Rodriguez-Ferri, E. F., Gutierrez-Martin, C. B., Rodriguez-Barbosa, I., Berche, P. & Vazquez-Boland, J. A. (1993). Transposon mutagenesis in Actinobacillus pleuropneumoniae with a Tn10 derivative. J Bacteriol 175, 5717–5722.[Abstract]

Tertti, R., Skurnik, M., Vartio, T. & Kuusela, P. (1992). Adhesion protein YadA of Yersinia species mediates binding of bacteria to fibronectin. Infect Immun 60, 3021–3024.[Abstract]

Thomson, V. J., Bhattacharjee, M. K., Fine, D. H., Derbyshire, K. M. & Figurski, D. H. (1999). Direct selection of IS903 transposon insertions by use of a broad-host-range vector: isolation of catalase-deficient mutants of Actinobacillus actinomycetemcomitans. J Bacteriol 181, 7298–7307.[Abstract/Free Full Text]

Towbin, H., Staehelin, T. & Gordon, J. (1979). Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci U S A 76, 4350–4354.[Abstract]

Umemoto, T., Nakatani, Y., Nakamura, Y. & Namikawa, I. (1993). Fibronectin-binding proteins of a human oral spirochete Treponema denticola. Microbiol Immunol 37, 75–78.[Medline]

Westerlund, B. & Korhonen, T. K. (1993). Bacterial proteins binding to the mammalian extracellular matrix. Mol Microbiol 9, 687–694.[Medline]

Westerlund, B., Kuusela, P., Risteli, J., Risteli, L., Vartio, T., Rauvala, H., Virkola, R. & Korhonen, T. K. (1989). The O75X adhesin of uropathogenic Escherichia coli is a type IV collagen-binding protein. Mol Microbiol 3, 329–337.[Medline]

Wilson, M. E. (1991). IgG antibody response of localized juvenile periodontitis patients to the 29 kilodalton outer membrane protein of Actinobacillus actinomycetemcomitans. J Periodontol 62, 211–218.[Medline]

Wolfgang, M. C., Lee, V. T., Gilmore, M. E. & Lory, S. (2003). Coordinate regulation of bacterial virulence genes by a novel adenylate cyclase-dependent signaling pathway. Dev Cell 4, 253–263.[Medline]

Zambon, J. J. (1985). Actinobacillus actinomycetemcomitans in human periodontal disease. J Clin Periodontol 12, 1–20.[Medline]

Received 20 February 2004; revised 23 April 2004; accepted 11 May 2004.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Mintz, K. P.
Articles citing this Article
PubMed
PubMed Citation
Articles by Mintz, K. P.
Agricola
Articles by Mintz, K. P.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2004 Society for General Microbiology.