Abstract
The coding sequence for the surface protein hemagglutinin A (HagA) of Porphyromonas gingivalis 381 has previously been shown to contain four direct 1.35-kb repeats, designated repHA. This study was performed to determine if the number of repHA units in hagA is consistently 4 or if allelic polymorphism exists among strains and/or upon multiple passage of P. gingivalis. To this end, primers which were homologous to the regions directly 5′ and 3′ of the repeat domain in hagA were synthesized. PCR conditions which allowed amplification of the 8.4-kb repeat region between the primers in P. gingivalis 381 were established. Genomic DNA templates from 13 other P. gingivalis strains and 9 fresh clinical isolates from patients were analyzed under the same conditions as used above. Analysis of these PCR products demonstrated that the strains tested had different numbers (two to four) of repHA units in the respective hagA genes. The PCR products of 8.4, 7.0, and 5.7 kb represent four, three, and two repeats, respectively. One strain from each group (381, four repeats; W83, three repeats; and AJW4, two repeats) was also tested to determine if the number of repeats remained invariant upon passaging onto solid medium. No variability in the number of repeats in hagA within a strain was detected after 18 passages. P. gingivalis 381 was chosen for further testing in a mouse abscess model to determine if conditions of in vivo growth would select for deletions or duplications of the repeated sequences. Five days after infection, no change in the number of repeats was detected in cells recovered from either nonimmunized or preimmunized mice. This data indicates an interstrain variability of the number of repeat units and hence a size variability of the HagA protein of P. gingivalis, but unlike some surface antigens of other pathogenic species, the number of repeats remains relatively stable given the conditions of growth tested here.
Porphyromonas gingivalis, a gram-negative anaerobic bacterium, is considered to be one of the major etiologic agents of adult periodontitis. A strong link has also been established between P. gingivalis and rapidly progressive or refractory periodontitis (26).
Hemagglutinins are surface proteins that often function as adhesins, attaching bacteria to the host cells. The synthesis of hemagglutinating adhesins is an established virulence factor for many pathogenic bacterial species, including some of the most virulent human pathogens such as Vibrio cholerae (7), Salmonella spp. (14), Bordetella pertussis (20), and Escherichia coli (11).
By using functional assays, five genes encoding P. gingivalis hemagglutinins have been isolated in our laboratory (8, 15, 21, 22). These genes can be divided into two different groups based on their DNA sequence; one of these groups has homology to several other P. gingivalis proteins, including the cysteine proteases. The largest gene in this group, coding for hemagglutinin A (hagA), may be the largest prokaryotic gene described to date and is 7,887 bp in length in P. gingivalis 381. A dominant structural feature of hagA is the presence of four direct repeats, each 1,350 bp in length, within its open reading frame (Fig. 1). This motif of repeated sequences within the open reading frame of a surface-associated protein is common to several virulence-associated genes of some pathogenic species. The duplication or deletion of these repeats is one mechanism whereby these pathogenic species undergo antigenic variation.
FIG. 1.
Repeat structure of the hagA gene (7,887 bp, encoding a protein of 2,628 amino acids [AA]).
For this study, the sequence of the hagA gene from P. gingivalis 381 was used to develop a PCR method to examine the repeat structure of the hagA genes in several additional P. gingivalis strains and to determine if either in vitro passage or in vivo cultivation in mice resulted in a change in the number of repeats.
MATERIALS AND METHODS
Bacterial strains, media and growth conditions.
P. gingivalis strains including 381, A7A1-28, W50, W83, 33277, AJW3, AJW4, HG66, 9-14k-1, 10002, 10009, 10034, 10046, and 10048 were grown on Trypticase soy agar or in Todd-Hewitt broth (BBL Microbiology Systems, Cockeysville, Md.) supplemented as described previously (8).
E. coli JM109 [recA1 endA1 gyrA96 thi-1 hsdR17 supE44 relA1 Δ(lac-proAB) (F′ traD36 proAB lacIqZΔM15)] was used for all cloning studies. A Rec+ E. coli strain, χ6097 [F− Δ(lac-pro) rpsLΔasdA4 Δ(zhf-2::Tn10) thi φ80 dlacZΔM15) was used to test the stability of the repeats and was cultured aerobically on Luria-Bertani agar (6, 25). E. coli transformants containing pBluescript (Stratagene Cloning Systems, La Jolla, Calif.) were grown in Luria-Bertani medium supplemented with 100 μg of ampicillin per ml.
Identification of P. gingivalis from clinical isolates.
Strains of black-pigmenting anaerobic bacteria obtained from periodontal pockets of patients with adult periodontitis were identified as P. gingivalis by their ability to hydrolyze N-benzoyl-dl-arginine-2-naphthylamide and p-nitrophenyl-α-glucoside, as described by Laughon et al. (13), and the synthetic fluorogenic substrate N-α-benzoyl-l-arginine-7-amido-4-methylcoumarin hydrochloride (Sigma, St. Louis, Mo.) (27). For confirmation that the fluorescent colonies were indeed P. gingivalis, fatty acid analysis was performed on an HP5890 gas chromatograph and analyzed with the Microbial Identification System vesion 4 software (MIDI, Inc., Newark, Del.).
Genomic DNA isolation and purification.
For isolation of genomic DNA, colonies were resuspended in 50 ml of lysis buffer (1% Triton X-100, 20 mM Tris [pH 8.5], 2 mM EDTA) (24), 1 ml of RNase (20 mg/ml; Sigma) and 1 ml of lysozyme (50-mg/ml stock; Sigma) were added, and the suspensions were left on ice for 15 min. After phenol-chloroform purification, 2 volumes of ethanol and 1 ml of glycogen (20-mg/ml stock; Boehringer Mannheim) were added. After the mixture had been cooled for 20 min at −20°C, the DNA was pelleted by centrifugation for 15 min in a microcentrifuge at room temperature. Finally, the pelleted DNA was dissolved in 10 ml of Tris-EDTA (TE), and 1 ml was used for PCR.
For the isolation of genomic DNA from liquid cultures of the 14 laboratory strains, the Wizard Genomic DNA purification kit (Promega, Madison, Wis.) was used as specified by the manufacturer.
PCR.
A PCR method was developed to detect allelic polymorphism in the number of repHA units in the hagA genes of various P. gingivalis strains. Two 22-mers were synthesized at the DNA Synthesis Core Lab, Interdisciplinary Center for Biotechnology Research, University of Florida. The upstream primer, 208 (5′-TTTCGCTCGCCGTCCTATTATC-3′), is located at nucleotides 23 to 44 of the hagA coding sequence (8). The downstream primer, 207 (5′-TTTCATCCTTTTCCCTATTTCA-3′), is positioned at nucleotides 8442 to 8463 after the start codon. The suitability of the primers was checked on Oligo 4.0 software. Reactions were performed in a final volume of 50 ml containing 2.35 mM MgCl2, 0.3 mM each primer, 0.4 mM each deoxynucleotide triphosphate, 1.25 U of Taq DNA polymerase (Promega), and 0.78 U of Pfu DNA polymerase (Stratagene). The amplifications were carried out in a PTC-100 thermal cycler (MJ Research, Watertown, Mass.) with a 1-min initial denaturation step at 94°C, followed by 30 cycles at 94°C for 30 s, 55°C for 30 s, and 68°C for 6 min, and with a final extension step of 10 min at 72°C.
Assaying repeat stability in a Rec+ E. coli host.
To attempt to induce repeat-involved recombination events, a Rec+ E. coli strain (χ6097) was transformed with two different plasmid clones of the hagA gene, a 3.2-kb partial clone ST2 (21) and a 10.1-kb full-length genomic clone (8). Two individual transformants were grown from each transformation and, plasmid minipreps were isolated and analyzed for plasmid size. In addition, both plate and liquid cultures of each transformant were grown at 30 and 37°C. Luria-Bertani agar plates were supplemented with 100 μg of ampicillin per ml and 50 μg of dl-α,ɛ-diaminopimelic acid per ml to complement the deficiency in the strain (6).
Immunoblotting of cell fractions.
The cell pellet from a 20-ml saturated culture of P. gingivalis 381 grown in Todd-Hewitt broth was resuspended in 1 ml of phosphate-buffered saline (PBS) before the cells were sonicated in ice three times for 20 s (Kontes microsonicator). A culture of E. coli JM109 hosting the 8.6-kb hagA clone in pBluescript (8) and a control culture of E. coli with pBluescript only were grown and analyzed in a similar manner. Following a 10-min centrifugation at 16,000 × g, the soluble cellular fractions were separated from the membrane fractions. For preparation of a concentrated spent culture medium fraction, the supernatants from the cell cultures were filter sterilized through a 0.22-μm-pore size Acrodisc filter (Gelman Sciences, Ann Arbor, Mich.) and concentrated 26 fold with a Centriprep-10 concentration device (Amicon, Beverly, Mass.) at 3,900 × g. Gradient (4 to 15%) polyacrylamide gels were subjected to electrophoresis in Tris-sodium dodecyl sulfate buffer by the method of Laemmli, and the proteins were visualized with a zinc staining kit (Bio-Rad). After being destained, the proteins were transferred to nitrocellulose membranes (MSI, Westboro, Mass.) by standard procedures (3). The membrane was reacted for 1.5 h with monoclonal antibody 1A1 (5) diluted 1:20 in a blocking solution of 1% Carnation dry milk and 0.02% sodium azide in Tris-buffered saline (TBS). Alkaline phosphatase-conjugated goat anti-mouse immunoglobulin G (Fisher), diluted 1:500 in the blocking solution, was used as the secondary antibody.
Mouse abscess model.
Female BALB/c mice, 10 weeks old, were obtained from Charles River Laboratories, Inc. (Wilmington, Mass.). For experiments with nonimmunized animals, 10 mice, divided into five groups, were each injected with 0.2 ml of P. gingivalis 381 which had been grown in liquid medium, checked by Gram staining, washed twice with PBS, and resuspended in PBS to 5 × 1010 cells/ml. Following localized depiliation, 0.1 ml of the strain 381 preparation was injected subcutaneously into each of two sites 1 cm laterally from the midline on the dorsal surface of each mouse. During the 5-day time course, samples were recovered from each abscess site of two mice per day. The aspirate was treated with 1 ml of sterile water to lyse the erythrocytes, the preparation was centrifuged, and the pellets were resuspended in 0.2 ml of Todd-Hewitt broth–2% dimethyl sulfoxide and stored at −80°C. From group 5 (the samples taken on day 5 after inoculation), 10−3 dilutions in PBS were plated onto blood agar plate (BAP) medium plus gentamicin (50 μg/ml) for analysis of the number of repHA repeats in isolated individual colonies.
For immunized animals, on day 14 after the initial immunization, five mice were challenged in an identical way with twice the number of P. gingivalis 381 cells, or 2 × 1010 cells per animal. Five days after the challenge, samples were collected from the abscesses of the two surviving mice and dilutions of the abscess fluids were plated onto BAP medium plus gentamicin (50 μg/ml) plates for analysis of the number of repeats in isolated individual colonies.
RESULTS
PCR assay to detect allelic polymorphism in hagA size and number of repeats.
Our previous studies (8) have shown the presence of four 1.35-kb repeats in the hagA gene of strain 381. To determine the number of repeat sequences in other strains of P. gingivalis, a PCR-based assay was developed. The assay was first tested with strain 381 to verify its accuracy. To this end, oligonucleotide primers flanking the repeats in hagA from P. gingivalis 381 were chosen and a PCR assay was developed to amplify the 8.4-kb fragment in genomic DNA from strain 381 between the primers (Fig. 2, lane 2). This product includes all four direct repeats in the open reading frame.
FIG. 2.
Analysis of genomic DNA purified from 11 strains of P. gingivalis to determine the size and number of repeats found for hagA. The size variation of hagA divides the strains into three groups, according to the number of repeats. The product size is indicated by arrows. (W50, see Fig. 3B; lanes W.A7A1-28 and 33277 are not shown.) MW, BstEII-digested λ DNA. Positive control, hagA cloned in pUC18. Negative control, template-free reaction (results not shown).
Determination of the number of repHA units in the hagA gene of additional strains.
After the PCR assay was developed with strain 381, it was applied to DNA purified from additional strains to determine the polymorphism in size and number of repeats within their hagA genes. Previously, we showed that strains of P. gingivalis can be placed into nine different restriction fragment length polymorphism groups based on the heterogeneity of the hagA locus. Genomic DNA from 36 strains was digested, transferred to nitrocellulose filters, and hybridized to the hagA probe (16). For this study, a representative of each of these groups was chosen for the hagA repeat analysis. In addition, strains W50, W83, A7A1-28, and 33277 and nine fresh clinical isolates identified as P. gingivalis by confirmatory tests were analyzed. This analysis revealed that the hagA genes of the laboratory-passaged strains fall into three categories according to the repeat number, with two, three, or four repeats (Fig. 2, lanes 4 to 14; Table 1). The PCR products of 8.4, 7.0, and 5.7 kb represent four, three, and two repeats, respectively. The hagA genes of the fresh clinical isolates were found to contain either three or four repeats (data not shown). As an additional control, genomic DNA from Prevotella intermedia 17 (provided by W. Nesbitt) was amplified in a similar way; no related PCR products were obtained. There are probably no genes homologous to hagA in other black-pigmented pathogens. A serological study of 10 oral species has shown that they are antigenically distinct (19).
TABLE 1.
Comparative PCR analysis of hagA size and number of repeats in 14 P. gingivalis laboratory strainsa
| P. gingivalis strain | No. of repeats in hagA gene |
|---|---|
| 381 | 4 |
| A7A1-28 | 3 |
| HG66 | 4 |
| W50 | 3 |
| W83 | 3 |
| AJW3 | 3 |
| AJW4 | 2 |
| 9-14k-1 | 4 |
| 10002 | 3 |
| 10009 | 3 |
| 10034 | 4 |
| 10046 | 3 |
| 10048 | 3 |
| 33277 | 3 |
See Fig. 2.
Testing for size variability upon repeated passage in vitro.
Representatives of each group (P. gingivalis 381 with four repeats, W83 with three repeats, and AJW4 with two repeats) were chosen for a further study of size variability as a result of in vitro passage. The cells were transfered on solid BAP medium 18 times. Chromosomal DNA was isolated from individual colonies, and the repeat number was determined. PCR analysis after 8 (Fig. 3A) and 18 (Fig. 3B) passages was performed, as above. As can be seen in Fig. 3, during extended in vitro cultivation, no variability in size or repeat number was detected in any of the P. gingivalis strains examined.
FIG. 3.
(A) Analysis of the number of hagA repeats in the representative strains 381 (lane 1 to 3), W83 (lanes 4 to 6), and AJW4 (lanes 7 to 9) after six to eight passages. Lane 10 contains control template hagA cloned DNA. The sizes of the amplified hagA fragment are indicated by arrows. (B). Size of hagA in the representative strains after 18 passages. Strains 381 (lanes 1 and 2), W83 (lanes 3 and 4), AJW4 (lanes 5 and 6) are shown. Lane 7 contains the control hagA clone. MW, BstEII-digested λ DNA. Empty lane, negative control template-free reaction.
Testing cellular fractions for the presence of an epitope in hemagglutinin A.
To test whether immunodetection of the hagA translation product can be used to detect variation in the size of the protein, P. gingivalis 381 cells were fractionated and the cellular fractions were analyzed by Western immunoblotting with monoclonal antibody 1A1 (5). In addition to whole cells and the soluble cellular fractions, the spent culture medium was analyzed after concentration. MAb 1A1 is the only characterized antibody specific for hemagglutinin A antigenic determinant presently available. Multiple bands of various sizes were detected with the whole-cell lysate (Fig. 4, lane 1), and fewer protein species from the soluble cellular fraction were recognized (lane 2). Numerous protein species from the 26-fold-concentrated spent culture medium also reacted with the antibody (lane 3). The high-molecular-weight fraction reacted with the antibody in the positive control expressing the HagA protein in E. coli JM109 (lane 4), but no reactive bands were detected from the same strain carrying the plasmid vector only (lane 5), confirming the specificity of the antibody to P. gingivalis.
FIG. 4.
Expression of monoclonal antibody 1A1-reactive proteins in P. gingivalis 381. Lanes: 1, P. gingivalis whole-cell lysate; 2, P. gingivalis soluble cellular fraction; 3, spent culture medium; 4, hagA cloned in pBluescript in E. coli, soluble cellular fraction; 5, pBluescript in E. coli, soluble cellular fraction.
Testing for hagA size variability upon in vivo growth and in immunized mice.
After dilution of aspirated samples from the localized abscesses of BALB/c mice, 21 individual colonies from the mice from day 5 postinoculation were grown on plates and the chromosomal DNA was isolated and analyzed by PCR to determine the number of hagA repeats. No change in the number of repeats was detected. To test cells recovered from preimmunized mice, 2 weeks after immunization the mice were challenged again. Twenty-seven individual colonies were recovered from two mice on day 5 postchallenge, and the chromosomal DNA was analyzed for the number of repeats in hagA. Again, no change in the number of repeats was detected compared to the control (data not shown).
Testing repeat number stability in a Rec+ positive E. coli host.
To assess whether the lack of variability in the number of repeat units in the hagA gene following in vitro or in vivo growth is a factor of the hagA gene itself or is due to factors of the genetic background in P. gingivalis such as reduced recombinational activity, hagA was cloned into a Rec+ background. Eight such individual colonies were grown, and plasmid minipreps were analyzed for plasmid size. No difference was found compared to the control plasmids (ST2 and a genomic clone of hagA isolated from the recombination-impaired host E. coli JM109) (data not shown).
DISCUSSION
The motif of several direct repeats within an open reading frame has been found in virulence-associated genes of several animal and human pathogens, including Rickettsia rickettsii (2), Streptococcus gordonii (17), Mycoplasma hominis (12), Haemophilus influenzae (10), M. hyorhinis (23), and Anaplasma marginale (1), as well as in the gene encoding the M protein of group A streptococci (18); all these genes contain domains with various numbers of tandemly repeated sequences.
In some species, a change in the number of repeats results in antigenic or size variation in surface antigens. Loss or gain of one or more repeat units is known to mediate phase variation of surface molecules of pathogenic bacteria. This facilitates evasion of host defenses and adaptation to the varying microenvironments of the host (9). Similarly, the size and antigenic diversity of the variable adherence-associated (Vaa) antigen, a major surface protein and a putative adhesin of M. hominis (30) and of the highly variable surface antigens of M. pulmonis and Ureaplasma urealyticum (29, 31), have also been reported. Thus, the potential for virulence-related diversity possessed by internal periodic structures of surface-exposed antigens is well established.
Given the precedence that bacterial surface antigens that contain multiple repeats function as important virulence factors and undergo rapid deletions or additions of repeat units, we were prompted to develop a technique for determining the number of repeats for the P. gingivalis hagA gene. In vitro or in vivo passage may result in alteration of surface antigens. These alterations should be the result of changes in the gene structure. We have developed a PCR assay that has high sensitivity and specificity for the determination of the interstrain variability of the number of repeat units in the hagA gene of P. gingivalis. This genomic DNA-based assay may be superior to surface variability assays based on alterations in expression, size, and surface presentation of membrane proteins. The assay has been used to determine interstrain variability of the number of repeat units in hagA and hence the size variability of the HagA protein.
The results of the PCR assay applied to 14 different P. gingivalis laboratory strains are documented in Fig. 2 and Table 1. The multiple bands in the PCR assays are most probably caused by Taq polymerase slippage, as indicated by the ladder-type pattern and by their presence in both genomic and plasmid DNA templates.
The strains examined fall into three groups according to the number of repeats. The PCR products of 8.4, 7.0, and 5.7 kb represent four, three, and two repeats, respectively. The data indicate an interstrain polymorphism in the number of repeat units and hence a size variability of the HagA protein. Figure 3A and B depict the results of the amplification assay at two different time points. These data demonstrate that the internal periodic structure of HagA is conserved within the chosen representative strains upon both 8 and 18 passages. No change in the number of repeats was detected on plate-grown cultures or in response to in vivo stimuli in a mouse abscess model. It has been previously noted that separate colonies derived from a single clinical specimen or strains of H. influenzae passaged for several weeks on chocolate agar plates also did not change the lengths of the variable number of tandem repeats (28).
Immunoblotting analysis of the HagA protein of P. gingivalis cell lysates could also be informative for determination of the size variability of the polypeptide. The only available monoclonal antibody (5) specific for a peptide epitope found in the deduced HagA protein sequence was used for this purpose. The Western analysis showed the presence of multiple protein species of different sizes reactive with the antibody in cells and in culture medium. This supports the previous data showing that other P. gingivalis proteins contain the epitope for 1A1 (4, 5). To our knowledge, there is no other monoclonal or polyclonal antibody specific for another HagA antigenic determinant that can be used for analysis of protein size variation. Thus, the PCR-based assay is the most specific and hence the most useful assay for this purpose.
In conclusion, the experimental data presented here demonstrate that allelic polymorphism in the size of HagA exists among different laboratory and clinical P. gingivalis isolates. The data suggest that HagA repeat number variability is probably not connected to antigenic variation of P. gingivalis. While shuffling the number of repeats may play a role in evading the immune response, the main function of HagA may be adhesion to blood and tissue factors. This possibility is currently being investigated.
ACKNOWLEDGMENTS
We thank C. Walker for assistance with fatty acid analysis and J. Burks for excellent technical assistance. MAb 1A1 was kindly provided by M. Curtis. We appreciate the critical comments of the manuscript by J. Hillman.
This work was supported by NIH grant DE07496 to A. Progulske-Fox.
REFERENCES
- 1.Allred D R, McGuire T C, Palmer G H, Leib S R, Harkins T M, McElwain T F, Barbet A F. Molecular basis for surface antigen size polymorphisms and conservation of a neutralization-sensitive epitope in Anaplasma marginale. Proc Natl Acad Sci USA. 1990;87:3220–3224. doi: 10.1073/pnas.87.8.3220. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Anderson B E, McDonald G A, Jones D C, Regnery R L. A protective protein antigen of Rickettsia rickettsii has tandemly repeated, near-identical sequences. Infect Immun. 1990;58:2760–2769. doi: 10.1128/iai.58.9.2760-2769.1990. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Ausubel F M, Brent R, Kingston R E, Moore D D, Seidman J G, Smith J A, Struhl K. Current protocols in molecular biology. Vol. 1. New York, N.Y: Greene Publishing Associates and Wiley-Interscience; 1997. [Google Scholar]
- 4.Barkocy-Gallagher G A, Han N, Patti J M, Whitlock J, Progulske-Fox A, Lantz M S. Analysis of the prtP gene encoding porphypain, a cysteine proteinase of Porphyromonas gingivalis. J Bacteriol. 1996;178:2734–2741. doi: 10.1128/jb.178.10.2734-2741.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Curtis M A, Aduse-Opoku J, Slaney J M, Rangarajan M, Booth V, Cridland J, Shepherd P. Characterization of an adherence and antigenic determinant of the ArgI protease of Porphyromonas gingivalis which is present on multiple gene products. Infect Immun. 1996;64:2532–2539. doi: 10.1128/iai.64.7.2532-2539.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Dusek D M, Progulske-Fox A, Brown T A. Systemic and mucosal immune responses in mice orally immunized with avirulent Salmonella typhimurium expressing a cloned Porphyromonas gingivalis hemagglutinin. Infect Immun. 1994;62:1652–1657. doi: 10.1128/iai.62.5.1652-1657.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Finkelstein R A, Hanne L F. Purification and characterization of the soluble hemagglutinin (cholera lectin) produced by Vibrio cholerae. Infect Immun. 1982;36:1199–1208. doi: 10.1128/iai.36.3.1199-1208.1982. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Han N, Whitlock J, Progulske-Fox A. The hemagglutinin gene A (hagA) of Porphyromonas gingivalis 381 contains four large, contiguous, direct repeats. Infect Immun. 1996;64:4000–4007. doi: 10.1128/iai.64.10.4000-4007.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Hood D W, Deadman M E, Jennings M P, Bisercic M, Fleischmann R D, Venter J C, Moxon E R. DNA repeats identify novel virulence genes in Haemophilus influenzae. Proc Natl Acad Sci USA. 1996;93:11121–11125. doi: 10.1073/pnas.93.20.11121. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Jarosik G P, Hansen E J. Identification of a new locus involved in expression of Haemophilus influenzae type b lipooligosaccharide. Infect Immun. 1994;62:4861–4867. doi: 10.1128/iai.62.11.4861-4867.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Johnson J R, Swanson J L, Barela T J, Brown J J. Receptor specificities of variant Gal(alpha1-4)Gal-binding PapG adhesins of uropathogenic Escherichia coli as assessed by hemagglutination phenotypes. J Infect Dis. 1997;175:373–381. doi: 10.1093/infdis/175.2.373. [DOI] [PubMed] [Google Scholar]
- 12.Ladefoged S A, Birkelund S, Hauge S, Brock B, Jensen L T, Christiansen G. A 135-kilodalton surface antigen of Mycoplasma hominis PG21 contains multiple directly repeated sequences. Infect Immun. 1995;63:212–223. doi: 10.1128/iai.63.1.212-223.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Laughon B E, Syed S A, Loesche W J. Rapid identification of Bacteroides gingivalis. J Clin Microbiol. 1982;15:345–346. doi: 10.1128/jcm.15.2.345-346.1982. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Lee M D, Curtiss R R, Peay T. The effect of bacterial surface structures on the pathogenesis of Salmonella typhimurium infection in chickens. Avian Dis. 1996;40:28–36. [PubMed] [Google Scholar]
- 15.Lepine G, Progulske-Fox A. Duplication and differential expression of hemagglutinin genes in Porphyromonas gingivalis. Oral Microbiol Immunol. 1996;11:65–78. doi: 10.1111/j.1399-302x.1996.tb00339.x. [DOI] [PubMed] [Google Scholar]
- 16.Lepine G, Savett D A, Progulske-Fox A. Restriction fragment length polymorphism analysis of two hemagglutinin loci, serotyping and agglutinating activity of Porphyromonas gingivalis isolates. Oral Microbiol Immunol. 1995;10:1–7. doi: 10.1111/j.1399-302x.1995.tb00110.x. [DOI] [PubMed] [Google Scholar]
- 17.McNab R, Jenkinson H F, Loach D M, Tannock G W. Cell-surface-associated polypeptides CshA and CshB of high molecular mass are colonization determinants in the oral bacterium Streptococcus gordonii. Mol Microbiol. 1994;14:743–754. doi: 10.1111/j.1365-2958.1994.tb01311.x. [DOI] [PubMed] [Google Scholar]
- 18.Miller L, Burdett V, Poirier T P, Gray L D, Beachey E H, Kehoe M A. Conservation of protective and nonprotective epitopes in M proteins of group A streptococci. Infect Immun. 1988;63:622–631. doi: 10.1128/iai.56.8.2198-2204.1988. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Okuda K, Ohta K, Kato T, Takazoe I, Slots J. Antigenic characteristics and serological identification of 10 black-pigmented Bacteroides species. J Clin Microbiol. 1986;24:89–95. doi: 10.1128/jcm.24.1.89-95.1986. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Parker, C. D., and S. K. Armstrong. 1988. Surface proteins of Bordetella pertussis. Rev. Infect. Dis. 10(Suppl. 2):S327–S330. [DOI] [PubMed]
- 21.Progulske-Fox A, Tumwarson S, Holt S C. The expression and function of Bacteroides gingivalis hemagglutinin gene in Escherichia coli. Oral Microbiol Immunol. 1989;4:121–131. doi: 10.1111/j.1399-302x.1989.tb00238.x. [DOI] [PubMed] [Google Scholar]
- 22.Progulske-Fox A, Tumwasorn S, Lepine G, Whitlock J, Savett D, Ferretti J J, Banas J A. The cloning, expression and sequence analysis of a second Porphyromonas gingivalis gene that codes for a protein involved in hemagglutination. Oral Microbiol Immunol. 1995;10:311–318. doi: 10.1111/j.1399-302x.1995.tb00160.x. [DOI] [PubMed] [Google Scholar]
- 23.Rosengarten R, Wise K S. The Vlp system of Mycoplasma hyorhinis: combinatorial expression of distinct size variant lipoproteins generating high-frequency surface antigenic variation. J Bacteriol. 1991;173:4782–4793. doi: 10.1128/jb.173.15.4782-4793.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Runnebaum I B, Syka P, Sukumar S. Vector PCR. BioTechniques. 1991;11:446–448. and 450–452. [PubMed] [Google Scholar]
- 25.Sambrook J, Fritsch E F, Maniatis T. Molecular cloning: a laboratory manual. 2nd ed. Vol. 2. Cold Spring Harbor, N.Y: Cold Spring Harbor Laboratory Press; 1989. [Google Scholar]
- 26.Shah H N, Mayrand D, Genco R J, editors. Biology of the species Porphyromonas gingivalis. Boca Raton, Fla: CRC Press, Inc.; 1993. [Google Scholar]
- 27.Slots J. Detection of colonies of Bacteroides gingivalis by a rapid fluorescence assay for trypsin-like activity. Oral Microbiol Immunol. 1987;2:139–141. doi: 10.1111/j.1399-302x.1987.tb00279.x. [DOI] [PubMed] [Google Scholar]
- 28.van Belkum A, Scherer S, van Leeuwen W, Willemse D, van Alphen L, Verbrugh H. Variable number of tandem repeats in clinical strains of Haemophilus influenzae. Infect Immun. 1997;65:5017–5027. doi: 10.1128/iai.65.12.5017-5027.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Watson H L, Zheng X, Cassell G H. Structural variations and phenotypic switching of mycoplasmal antigens. Clin Infect Dis. 1993;17:S183–S186. doi: 10.1093/clinids/17.supplement_1.s183. [DOI] [PubMed] [Google Scholar]
- 30.Zhang Q, Wise K S. Molecular basis of size and antigenic variation of a Mycoplasma hominis adhesin encoded by divergent vaa genes. Infect Immun. 1996;64:2737–2744. doi: 10.1128/iai.64.7.2737-2744.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Zheng X, Teng L J, Watson H L, Glass J I, Blanchard A, Cassell G H. Small repeating units within the Ureaplasma urealyticum MB antigen gene encode serovar specificity and are associated with antigen size variation. Infect Immun. 1995;63:891–898. doi: 10.1128/iai.63.3.891-898.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]