当前位置: 首页 > 医学版 > 期刊论文 > 基础医学 > 感染与免疫杂志 > 2005年 > 第8期
编号:11409093
Hag Directly Mediates the Adherence of Moraxella catarrhalis to Human
http://www.100md.com 《感染与免疫杂志》
     Department of Medical Microbiology and Immunology, Medical College of Ohio, 3055 Arlington Ave., Toledo, Ohio 43614

    ABSTRACT

    Moraxella catarrhalis is a human pathogen that causes otitis media in young children and lung infections in patients with chronic obstructive pulmonary disease. In this study, the role of the surface protein Hag in the adherence of multiple M. catarrhalis strains was examined. The hag genes of four clinical isolates were disrupted with a spectinomycin resistance cassette, and the binding of isogenic mutants to primary cultures of human middle ear epithelial cells (HMEE), as well as A549 pneumocytes, was measured. These experiments revealed that the attachment of most mutants to both cell types was 10-fold less than that of their wild-type progenitors. To determine whether Hag directly mediates adherence to human cells, the hag genes from three M. catarrhalis isolates were cloned and expressed in a nonadherent Escherichia coli cloning strain. At least 17-fold more E. coli bacteria expressing Hag attached to HMEE cells than an adherence-negative control. Surprisingly, Hag expression did not increase the binding of recombinant E. coli to A549 monolayers. Our data demonstrate that the involvement of Hag in M. catarrhalis adherence to A549 and HMEE cells is conserved among isolates and that Hag directly mediates binding to HMEE cells.
, 百拇医药
    INTRODUCTION

    Moraxella catarrhalis is a gram-negative bacterium that causes respiratory tract infections in humans. In adults, this organism causes up to 35% of all infectious exacerbations in patients suffering from chronic obstructive pulmonary disease, contributing to the progression of this fourth leading cause of death in the United States (48, 49). M. catarrhalis also causes 20% of all cases of otitis media (middle ear infection) in children of developed countries (6, 9, 12, 23-26, 36, 45). More than 80% of children have at least one ear infection by the age of 3 years, and these episodes can cause significant delays in development of language and learning skills (24-26). More than 90% of M. catarrhalis strains isolated from patients are -lactam resistant (21, 27, 34), and there is currently no vaccine protective against this organism. Clearly, M. catarrhalis is a significant health concern and the development of a vaccine and novel therapeutic approaches is desirable.
, 百拇医药
    Adherence is a necessary step of pathogenesis by most infectious agents (4, 20, 51). The proteins mediating this adherence (adhesins) are surface located, making them attractive vaccine candidates. Studies with FimH, a major adhesin of uropathogenic Escherichia coli, have shown that vaccination with purified FimH and passive transfer of FimH-specific antibodies are protective in animal models of infection (31, 32). Moreover, the Bordetella pertussis adhesins FHA and Pertactin are components of three acellular pertussis vaccines currently licensed for use in the United States (7). Thus, adhesins are proven effective vaccine antigens. Several M. catarrhalis adhesins have been identified, including UspA1 (1, 28), UspA2H (28), OMPCD (18), and McaP (54), all of which have been shown to directly mediate adherence to human epithelial cells in vitro. Hag, a 200-kDa outer membrane protein, is critical for hemagglutination, autoagglutination, and binding of human immunoglobulin D by the M. catarrhalis isolate O35E (42). In addition, our laboratory has shown that Hag expression plays an important role in the binding of strain O35E to A549 human pneumocytes and to primary cultures of human middle ear epithelial (HMEE) cells (19). A purified and radiolabeled recombinant protein corresponding to residues 764 to 913 of MID, a Hag ortholog expressed by M. catarrhalis strain Bc5, was shown by Forsgren and coworkers to bind A549 monolayers. In addition, immunization with this MID764-913 polypeptide yielded antibodies that decreased adherence of M. catarrhalis to A549 cells by 65% (13). These data suggest that Hag mediates the binding of M. catarrhalis to A549 and HMEE cells.
, 百拇医药
    Building on past research, we have generated isogenic hag mutants in four M. catarrhalis clinical isolates of various geographic and clinical origins to determine whether Hag expression is important for adherence to HMEE and A549 cells. We also report the successful cloning and expression of hag genes in E. coli in order to demonstrate that this protein directly mediates binding to middle ear cells.

    MATERIALS AND METHODS

    Bacterial strains, plasmids, tissue culture cell lines, and culture conditions. A549 and primary HMEE cells were cultured as previously described (18, 19, 55). HMEE cells were kindly provided by Thomas DeMaria (Ohio State University). These cultures were obtained at passage 5, a stage at which they were shown to be free of fibroblast contamination via immunochemical staining with an antivimentin monoclonal antibody (MAb) (55), and aliquots were immediately frozen. At 4 to 6 days prior to performing adherence assays, aliquots of frozen HMEE cells were thawed and cultured in fresh minimal essential medium (MEM) as described by Tong et al. (55). When HMEE cells reached confluency, they were immediately seeded into the wells of 24-well tissue culture plates to perform adherence assays. HMEE cells were also passaged once into fresh medium in order to obtain more cells for repeating the adherence assays. When grown under these conditions, HMEE cells were shown to be free of fibroblast contamination (55).
, 百拇医药
    Table 1 lists the bacterial strains and plasmids used in the present study. M. catarrhalis was cultured by using Todd-Hewitt medium (TH; Difco) as previously described (18, 19, 54). For isogenic hag mutants, the medium was supplemented with spectinomycin at a concentration of 15 μg/ml with the exception of O35E.TN2, which was grown in the presence of 20 μg of kanamycin/ml. Recombinant E. coli clones were cultured using Luria-Bertani medium (LB; Fisher Scientific) supplemented with 15 μg of chloramphenicol/ml.
, 百拇医药
    M. catarrhalis cells used for genomic DNA preparation, whole-cell lysates, and adherence assays were grown on solidified medium (1.5% agar). For the preparation of whole-cell lysates, plasmid DNA, and adherence assays, recombinant E. coli were cultured overnight in 5 ml of broth. These cells were then seeded into 20 ml of fresh medium supplemented with Epicentre's copy control induction solution at a final concentration of 10x and subsequently grown for 2 h (adherence assays and whole-cell lysates) or 5 h (plasmid DNA) with vigorous shaking (300 rpm). All agar plates were incubated at 37°C and 7.5% CO2, whereas broth cultures were grown at 37°C with agitation only (no CO2 supplementation). Hag was found to be expressed under these conditions.
, 百拇医药
    Recombinant DNA methodology. Standard molecular biology techniques were implemented as described elsewhere (46). M. catarrhalis genomic DNA was obtained with the Invitrogen Easy-DNA kit according to the manufacturer's guidelines. Plasmid DNA was purified from E. coli recombinant clones with the QIAprep Spin Miniprep kit (QIAGEN) with minor variations from the manufacturer's instructions (two additional washes with PE buffer and an elution volume of 75 μl).

    PCR. Unless stated otherwise, all PCRs were performed by using the Platinum Pfx DNA polymerase as recommended by the manufacturer (Invitrogen). Amplicons containing entire hag genes were generated with the oligonucleotides Hag minus 2 (5'-GTC AGC ATG TAT CAT TTT TTA AGG-3') and Hag R4 (5'-TGA GCG GTA AAT GGT TTA AGT G-3'). These PCR products ranging in size from 7.0 to 8.5 kb were used in cloning experiments, for the construction of isogenic mutants, and as templates for sequencing. To identify isogenic hag mutants, Taq DNA polymerase (Invitrogen) was substituted for Pfx (see below). Genomic DNA was used as a template in all PCRs.
, http://www.100md.com
    Construction of isogenic hag mutations. A PCR product of 8-kb, corresponding to the entire O35E hag gene interrupted by the insertion of a spectinomycin resistance (Specr) cassette near the middle of the open reading frame (ORF), was amplified from the genomic DNA of strain O35E.ZCS by using primers Hag minus 2 and Hag R4. This amplicon was purified with Epicentre's PCR precipitation solution according to the manufacturer's recommendations, resuspended in sterile deionized water, and used to electroporate M. catarrhalis competent cells, which were prepared as previously reported (19). Upon electroporation, M. catarrhalis cells were spread onto agar plates supplemented with spectinomycin, and resistant colonies were analyzed by PCR with Taq DNA polymerase. The introduction of the Specr cartridge into the hag gene was demonstrated by the following three distinct PCRs. First, we used the primers HagF 3760 (5'-ACA TTG ACC AGT ACT GGC ACA G-3') and HagR 4153 (5'-GGC GTT GTC TTC ATT GAT GCC TTG-3'), which anneal to regions within the hag gene that are adjacent to the intended insertion site of the Specr cassette. In wild-type (WT) strains, these primers amplified DNA fragments of 0.4 kb. In contrast, PCR products of 1.6 kb were generated in isogenic mutants tested with HagF 3760 and HagR 4153 (data not shown). This 1.2-kb difference in the size of amplicons is consistent with the size of the Specr cassette (56). To confirm that the Specr cassette is inserted within the hag gene, we used primer sets in which one primer anneals within the hag gene, whereas the other binds to one of the terminal regions of Specr. The first primer pair, HagR 4153 and Spec 5' (5'-AGT CCA CTC TCA ACT CCT GAT CCA-3') generated an amplicon of 0.8-kb in isogenic mutants. As expected, no PCR products were amplified in WT strains with these primers (data not shown). These reactions demonstrated the insertion of the Specr cassette in the hag genes. These results were confirmed with a second set of primers, HagF 3231 (5'-CCA TCA CAG GAC TAA GCA ACA C-3') and Spec 3' (5'-TAT TGC GGG AAA TGC AGT GGC TGA-3'), which yielded a DNA fragment of 0.7-kb in mutants. No products were observed in WT strains (data not shown).
, 百拇医药
    Adherence assays. Adherence assays were performed as reported (18, 19, 54) with the exception that M. catarrhalis strains were incubated with monolayers of epithelial cells for 5 min prior to washing off unbound bacteria. For adherence assays with recombinant E. coli clones, bacteria were incubated with human cell lines for 1 h. Bacterial growth was found to be negligible during these incubation periods (data not shown).

    Protein preparation and analysis. Western blots were performed as previously described (18, 19, 42, 54). Briefly, whole-cell lysates were prepared by suspending bacteria in 5 ml of phosphate-buffered saline (PBS) plus 0.15% gelatin (PBSG) to an optical density of 300 Klett units. These suspensions were then centrifuged, and the pelleted bacteria were resuspended in 1 ml of PBS plus 0.5 ml of 3x sodium dodecyl sulfate (SDS) loading buffer. Lysates were electrophoresed on a 7.5% SDS-polyacrylamide gel, and proteins were then transferred to a polyvinylidene difluoride (PVDF) membrane (Millipore) via electrotransfer in a solution containing no alcohol. After transfer, membranes were incubated for 1 h in PBS supplemented with 0.05% (vol/vol) Tween 20 and 3% (wt/vol) skim milk. Detection of proteins was accomplished with mouse MAbs 5D2 (Hag specific) (42), 1D3 (OMPCD specific) (37), 24B5 (USPA1 specific) (11), and 17C7 (UspA1-, UspA2- and UspA2H-reactive) (2), followed by incubation with a secondary goat anti-mouse antibody that was conjugated to horseradish peroxidase (Southern Biotechnology Associates, Inc.). McaP expression was detected using polyclonal antibodies specific for the synthetic peptide PEP1 (see below). Signals were then detected by using the SuperSignal West Pico chemiluminescence kit from Pierce according to the manufacturer's recommendations.
, http://www.100md.com
    To obtain antibodies directed against the M. catarrhalis protein McaP (54), the synthetic peptide PEP1 (KDNNKTVETAVLSNRDYHRS) was synthesized and conjugated to keyhole limpet hemocyanin (Sigma Genosys). The sequence of PEP1 corresponds to residues 577 to 598 of the M. catarrhalis O35E-McaP. The keyhole limpet hemocyanin-PEP1 conjugate was used to immunize mice as previously reported (30). Murine antibodies were shown to specifically bind to McaP by Western blotting (data not shown).
, http://www.100md.com
    Nucleotide sequence analysis. Purified PCR products and plasmid DNA were used as templates in sequencing reactions. Sequencing services were provided by the University of Michigan sequencing core (http://seqcore.brcf.med.umich.edu/). Chromatograms were analyzed and assembled by using the ChromaTool software (BioTools, Inc.). Sequence analysis was conducted with the various tools provided by the ExPasy Proteomics Server (http://us.expasy.org/), as well as with the BioEdit analysis tool (16). The nucleotide sequences of the hag genes from strains P44 (AY862882), McGHS1 (AY862881), V1171 (AY862884), and TTA37 (AY862883) were deposited in GenBank under the indicated accession numbers.
, http://www.100md.com
    Statistical analysis. All statistical analyses were performed with the Mann-Whitney test in the GraphPad Prism 4.0 software. P values of <0.05 were considered statistically significant.

    RESULTS

    Construction of isogenic hag mutants. Previous work demonstrated that Hag is involved in the binding of M. catarrhalis strain O35E to primary cultures of HMEE cells and to A549 human lung cells (19). To determine whether Hag's involvement in adherence is conserved among M. catarrhalis strains, isogenic hag mutations were constructed in four additional isolates. Mutagenesis was accomplished via homologous recombination of a PCR product, which consisted of the entire O35E-hag gene disrupted by a Specr cassette, into the chromosome of target strains. This 8-kb DNA fragment was amplified from strain O35E.ZCS and electroporated into M. catarrhalis isolates McGHS1, O12E, V1171, and TTA37. Specr colonies were selected, and the isogenic hag mutations were verified as described in Materials and Methods.
, 百拇医药
    To confirm the lack of Hag expression in isogenic mutants, whole-cell preparations were analyzed by Western blotting with the Hag-specific MAb 5D2 (Fig. 1). As expected, the WT isolates O12E, V1171, TTA37, and McGHS1 all expressed Hag, whereas none of the isogenic mutants did. We noted that the Hag proteins of strains TTA37, V1171, O12E, and McGHS1 migrated more slowly than O35E-Hag, suggesting larger molecules. We also observed that TTA37, V1171, and particularly McGHS1 appeared to express less Hag than strains O35E and O12E. Upon sequence analysis (see below), it was noted that the Hag peptide to which MAb 5D2 binds (e.g., DNADGNQVNIADIKKDPNSGSSSNR [42]) has a valine (underlined)-to-isoleucine substitution in McGHS1-Hag (data not shown). Thus, reduced antibody affinity or binding may account for the apparent lower level of Hag expression in McGHS1. The binding of polyclonal antibodies to the adhesin McaP was used to demonstrate that equivalent amounts of cell lysates were analyzed (Fig. 1). The panel of mutants was also shown to express wild-type levels of the adhesins UspA1, UspA2H (in the case of strain TTA37), and OMPCD by Western blotting with appropriate MAbs (data not shown).
, 百拇医药
    Hag is associated with the adherence of M. catarrhalis isolates to HMEE and A549 human cells. The adherence of the aforementioned isogenic mutants to HMEE cells was compared to that of their corresponding parent strain by the use of a quantitative viable cell count attachment assay. The four isogenic mutants constructed in the present study bound substantially less than their respective WT strains during a 5-min incubation with HMEE monolayers (Fig. 2A). O12E.Hag adhered 27-fold less than its parent strain O12E, and this was the greatest decrease observed among mutants (Fig. 2A). Disruption of the hag ORF in the isogenic strains TTA37.Hag and McGHS1.Hag resulted in 21-fold decrease in adherence to HMEE cells (Fig. 2A). V1171.Hag also showed substantially decreased binding (sevenfold), albeit to a lesser extent than what was observed for the other isogenic mutants (Fig. 2A). However, the adherence of the parent strain V1171 was also slightly less that than of other WT isolates (see Fig. 2A), possibly due to lower expression of Hag (see Fig. 1). The hag mutant O35E.TN2, which we previously reported to bind poorly to HMEE cells (19), was used as a negative control; it exhibited a 22-fold reduction in adherence compared to its parent strain O35E (Fig. 2A). All hag mutants bound poorly, with values ranging from 1.1 to 3.1% of input bacteria. These results indicate that Hag's involvement in adherence to HMEE cells is conserved among the M. catarrhalis strains that were tested.
, 百拇医药
    It has been previously demonstrated that Hag expression plays a role in the binding of strain O35E to A549 human type II alveolar pneumocytes (19) and that M. catarrhalis isogenic mutants lacking MID, a Hag ortholog, also exhibit decreased adherence to A549 cells (13). When we tested the binding of hag mutants to A549 monolayers, a substantial reduction in adherence was measured for three of the four isogenic strains (Fig. 2B). Interestingly, the WT strain V1171 bound poorly to A549 cells (<1.5%), and the adherence of the V1171.Hag mutant was not significantly lower (data not shown). TTA37.Hag, O12E.Hag, and the negative control O35E.TN2 bound poorly to A549 cells, with percent adherence ranging from 0.8 to 3.6%. McGHS1.Hag adhered at levels slightly higher (6.2%) than the other isogenic strains, but still exhibited a 6.1-fold decrease in binding compared to McGHS1 (Fig. 2B). These data demonstrate that the role of Hag in adherence to human pneumocytes is conserved in three of the M. catarrhalis strains that were tested.
, 百拇医药
    Hag directly mediates adherence to middle ear cells. To determine whether Hag directly mediates binding to A549 and HMEE cells, the hag genes from isolates O35E, O12E, and V1171 were cloned and expressed in the recombinant background of the nonadherent E. coli cloning strain EPI300. Recombinant plasmids were sequenced to verify that no mutations were introduced during PCR and that each ORF contained the appropriate number of guanine residues in their poly(G) tracts (see below). To ensure that all recombinant clones expressed their respective proteins, Western blots were also performed with the Hag-specific MAb 5D2 (Fig. 3).
, 百拇医药
    Quantitative attachment assays demonstrated that the binding of E. coli clones expressing O12E-Hag (pBBO12.Hag), O35E-Hag (pELO35.Hag), and V1171-Hag (pSV1171.Hag) was 17, 31, and 50 times greater, respectively, than that of the control strain (Fig. 4A). In contrast, Hag expression did not increase binding of E. coli to A549 pneumocytes. As an adherence-positive control, we used EPI300 cells carrying the plasmid pMHmcmA, which encodes a novel adhesin recently discovered by our laboratory (Fig. 4B) (29). The data clearly demonstrate that Hag directly mediates adherence to HMEE cells. In addition, our results suggest that Hag expression alone is not sufficient to enable recombinant E. coli to bind to A549 lung cells.
, 百拇医药
    The lack of Hag expression does not adversely affect the autoagglutination of all M. catarrhalis isolates. Previous studies have shown that hag mutants of strain O35E no longer autoagglutinate, suggesting that Hag is involved in this phenotypic trait of M. catarrhalis isolates (19, 42). We therefore tested whether disruption of hag in strains other than O35E would have the same effect. To this end, autoagglutination assays were performed by suspending the different M. catarrhalis strains and their respective hag mutants in PBS. The turbidity of these suspensions was then measured at various time points, with O35E and O35E.TN2 as positive and negative controls, respectively. Surprisingly, these assays revealed that all experimental isogenic hag mutants autoagglutinate at or near WT levels (Fig. 5). For the majority of strains, autoagglutination began to level out at 30 min after suspension. The V1171.Hag mutant stayed in suspension longer than most other isogenic strains but eventually autoagglutinated to the same extent as V1171 (Fig. 5).
, 百拇医药
    Nucleotide sequence analysis of hag from strains TTA37, McGHS1, P44, and V1171. PCR products of 7 to 8 kb and containing entire hag genes were generated with oligonucleotides Hag minus 2 and Hag R4. Sequence analysis predicted that the hag ORFs specified proteins of 1,964 to 2,335 amino acid (aa) residues (Table 2). V1171-Hag was found to be the largest, whereas O35E-Hag was the smallest of the predicted molecules (Table 2). The sequences of the O35E (GenBank accession no. AY077637) and O12E (GenBank accession no. AY077638) hag ORFs, reported by Pearson et al. (42), as well as that of the M. catarrhalis strain Bc5 mid gene product, described by Forsgren et al. (14), were included in our comparative analysis. The hag ORFs of strains TTA37, McGHS1, V1171, and P44 contain homopolymeric tracts of guanine residues that are 9, 5, 6, and 6 nucleotides (nt) in length, respectively (Table 2). Variation in the length of this poly(G) tract, which in each case is located 80 nt downstream of the predicted ATG translational start codon, has been shown by several investigators to cause phase-variable expression of Hag/MID (35, 42, 47, 52). In these studies, it was shown that only poly(G) tracts in multiples of three allow for Hag expression. McGHS1-hag appears to have circumvented this requirement by replacing what would be the sixth guanine with an adenosine, thereby keeping the gene in the correct reading frame (data not shown).
, 百拇医药
    The predicted amino acid sequences of the six Hag proteins, as well as that of Bc5-MID, were compared and found to share an average identity of 73%. O35E-Hag and Bc5-MID displayed only 56% identity, whereas V1171-Hag and McGHS1-Hag were 89.1% identical (Table 3). We noted that most of the similarity between molecules was clustered in their C termini, and these observations are summarized in Fig. 6A. The last 454 residues were particularly well-conserved among these proteins, with an average identity of 97% (see red box in Fig. 6A). In contrast, a region of 800 aa encompassing most of the N-terminal halves of these molecules displayed an average identity of only 35.8% (see green box in Fig. 6A). We also noted that O35E-Hag is missing a 326-aa domain that is highly conserved in all of the other proteins analyzed (97.5% identical; see gray box in Fig. 6A). Bc5-MID was found to be missing a moderately conserved region (64.8%) of 200 residues that is present in all other Hag proteins (aa 861 to 1,063 of O35E-Hag; see yellow box in Fig. 6A).
, 百拇医药
    Selected structural features of the hag gene products. Analysis with the PSIPRED secondary structure prediction method (22) revealed that the last 52 aa of Hag and MID proteins are predicted to form four -strands (data not shown), each connected by short loops of two to five residues (illustrated in Fig. 6B). This potential membrane-anchoring domain was immediately preceded by a helical region of 40 residues (Fig. 6B; aa 1862 to 1902 of O35E-Hag). Two putative signal sequence cleavage sites were detected at the N terminus of each molecule by using the SignalP 3.0 server (38, 39), and those are as listed in Table 2. Interestingly, analysis through the Protein Families database (Pfam) service (3) identified a region of 25 aa that exhibits a high level of similarity with a domain designated HIM (Pfam accession no. PF05662), which is described as being associated with bacterial adhesins. This HIM motif corresponds to aa 312 to 336 of O35E-Hag (Fig. 6B) and is well conserved at the N terminus of all Hag and MID proteins (76% identity).
, 百拇医药
    Upon closer examination, the last 14 aa residues of this HIM-like domain of Hag (e.g., AGxxxTDAVNVAQL) were found to be 78 to 85% identical to a region located in the middle of the so-called "neck" of the Yersinia enterocolitica adhesin YadA, with the last nine residues (e.g., TDAVNVAQL) being perfectly conserved (17, 40, 44). Of note, this YadA neck region is preceded by degenerate repeated motifs that are generally 14 residues in length and composed of the consensus sequence xxxSVAIGxxSxAx (40). These repeats, termed roll by Nummelin et al. (40), have been shown to be essential for the adhesive properties of YadA (17, 40, 44). When the region located directly upstream of the HIM-like domain of Hag was analyzed, 10 repeated motifs 14 residues in length and with the consensus GxxSIAIGxx[A/S]xAx were found (Fig. 6B; see aa 83 to 272).
, 百拇医药
    In contrast to YadA, which contains only one roll/neck (HIM) combination, all Hag and MID proteins contained a second HIM-like domain that is also immediately preceded by eight degenerate repeats (roll-like) and located directly prior to the C-terminal helical region of 40 residues (Fig. 6B; see aa 1680 to 1865 of O35E-Hag). Further analysis also revealed two to four additional HIM-like domains dispersed throughout the Hag/MID molecules. These additional HIM domains, however, were not associated with potential rolls. These observations are summarized in Table 2 and Fig. 6B.
, 百拇医药
    DISCUSSION

    Previous research has demonstrated that the Hag protein of strain O35E forms elongated fibrillar structures covering the surface of M. catarrhalis (42), which are crucial for attachment to HMEE and A549 cells (19). In the present study, we sought to extend these findings by constructing isogenic hag mutations in four M. catarrhalis isolates of various geographical and clinical origins in order to test whether the involvement of Hag in adherence is conserved. We found that all mutants exhibited a substantial decrease in binding to HMEE cells (7- to 27-fold). Attachment assays with A549 pneumocytes yielded comparable results, with the exception of strain V1171.Hag. The parent strain V1171 did not attach well to A549 cells, and the basis for this low binding is under investigation. Our findings are consistent with those of Forsgren et al. with the M. catarrhalis protein MID, a Hag ortholog. When these investigators disrupted the mid gene of wild-type strains BBH18 and RH4, a two- to threefold reduction in attachment to A549 cells was measured (13). The role of Hag in adherence to HMEE and A549 cells is therefore conserved among M. catarrhalis strains. Furthermore, the very low binding capabilities of our isogenic mutants indicate that nonexpression of Hag has profound effects on attachment (Fig. 2).
, http://www.100md.com
    To determine whether Hag directly mediates adherence, the hag genes of three M. catarrhalis strains were cloned and expressed in E. coli, allowing recombinant bacteria to attach to HMEE cells. These experiments conclusively demonstrate that Hag is sufficient to mediate attachment to middle ear cells. In contrast, we discovered that Hag expression alone is not sufficient to permit recombinant E. coli to bind A549 cells. Our results are interesting in light of a previous report demonstrating that purified recombinant Bc5-MID binds to A549 monolayers (13). These observations suggest that Hag/MID expression by itself may not be sufficient for M. catarrhalis adherence to A549 pneumocytes. Alternatively, Hag's adherence epitope for A549 cells may not be properly displayed on the surface of E. coli. Adherence to A549 cells may also require posttranslational modification of Hag that is possibly not achieved in the heterologous genetic background of E. coli. For instance, glycosylation of the diarrheagenic E. coli protein AIDA modulates its ability to function as an adhesin and requires the expression of the heptosyltransferase aah gene, which is located directly upstream of the aidA ORF (5, 50). We believe, however, that binding to A549 pneumocytes requires coexpression of another M. catarrhalis protein. UspA1 has previously been proposed to act synergistically with Hag in attachment to A549 cells (13). Another adhesin potentially acting in concert with Hag is OMPCD, which was recently shown by our laboratory to mediate adherence to A549 monolayers. Interestingly, isogenic ompCD mutants have also been shown to bind poorly to these lung cells (18). The hypothesis that coexpression of OMPCD and Hag is necessary and sufficient for attachment to A549 cells is currently being investigated.
, 百拇医药
    Another phenotypic trait associated with Hag/MID expression is autoagglutination (19, 35, 42, 47). Pearson et al. were the first to show that an isogenic hag mutant of strain O35E no longer autoaggregated when suspended in PBS (42). Mollenkvist et al. also isolated nonfloculating variants of two M. catarrhalis strains that expressed very low levels of MID (35). We therefore tested the ability of hag mutants to fall out of solution when suspended in PBS. Surprisingly, the only mutant that did not flocculate was that of strain O35E; all other strains autoagglutinated at or near WT levels (Fig. 5). It should be noted that Mollenkvist et al. obtained their nonautoagglutinating variants after six serial passages over a period of 4 weeks, each time recovering nonflocculating variants (35). During this type of progressive selection, other mutations probably accumulated in the M. catarrhalis genome, and those mutations might have collectively caused the strong nonflocculating phenotype. In the case of strain O35E, it is possible that Hag does not mediate autoagglutination but rather lack of its expression affects the proper surface display of another molecule responsible for autoagglutination. This possibility is supported by the fact that none of our Hag-expressing E. coli strains gained the ability to autoagglutinate, including EPI300(pELO35.Hag), which we would have expected to flocculate based on the autoagglutination phenotypes of O35E and its hag mutant (Fig. 5) (42). Although these results suggest that Hag is not a flocculation factor, we cannot exclude the possibility that Hag-mediated autoagglutination requires posttranslational modification not achieved in E. coli. Our results suggest, however, that Hag is not the sole contributor to autoagglutination. This is consistent with data presented by Stutzmann Meier et al., in which six isogenic hag mutants were found to autoagglutinate at WT levels (52). Interestingly, M. catarrhalis was recently shown to express a type IV pilus (Tfp) (33), a structure associated with autoagglutination in other bacterial species (8, 41, 57). Whether Tfp is involved in autoaggregation by M. catarrhalis cells remains to be determined.
, 百拇医药
    Structural analysis of the Hag sequence revealed that it resembles members of a family of autotransporter proteins designated Oca (for oligomeric coiled-coil adhesion) (17). This family is represented by Yersinia enterocolitica YadA and also includes Haemophilus influenzae Hia (53), Neisseria meningitidis NadA (10), and the recently described Bartonella henselae proteins BadA (43) and VompA, VompB, and VompC (58). These molecules share the following structural features: (i) a C-terminal membrane anchor consisting of four -strands, (ii) a helical region of 40 residues that links the membrane anchor to a region of variable size, and (iii) an N-terminal signal sequence. Previous studies have suggested that the signal sequence cleavage site for Hag/MID is located between aa 66 and 67 (AYA66Q) (35, 42). Our analysis revealed that all Hag/MID proteins display a second potential cleavage site (AVA20E; see Table 2). Upon aligning Hag/MID sequences, it was also noted that the first 20 amino acids are perfectly conserved (data not shown). Furthermore, V1171-Hag does not have the potential signal sequence cleavage site AYA66Q due to an insertion of ten residues after aa 62 (data not shown). This insertion places the next potential signal sequence cleavage site at serine 76 (AVS76Q; see Table 2). The hypothesis that the Hag signal sequence cleavage site is located between residues 20 and 21 is being tested.
, http://www.100md.com
    The N-terminal region of several Oca adhesins has also been reported to contain a series of degenerate repeats, which in YadA have been shown to form a nine-coiled left-handed parallel structure termed roll (40). We found that the N terminus of all Hag/MID proteins exhibit a potential roll that contains 10 14-mer degenerate repeats with the consensus sequence GxxSIAIGxx[A/S]xAx. A related sequence was previously reported for O35E-Hag and O12E-Hag by Pearson et al. (IAIGxxxxxxxxxxIAIG) (42), and we believe that this sequence contains the end of one 14-mer repeat, as well as the beginning of a second one. This belief is based on the fact that the consensus sequences for the Hag (GxxSIAIGxx[A/S]xAx) and YadA (xxxSVAIGxxSxAx) repeats are very similar and that the glycine at position 8 is perfectly conserved (underlined). Following this potential roll near the N terminus of Hag, sequence analysis identified a region of amino acids that is similar to a Pfam domain designated HIM. YadA also contains a HIM domain that is found in a so-called "neck" region located immediately after the aforementioned roll. This roll-neck combination is essential for the adhesive properties of YadA (40, 44) and has been observed in other bacterial adhesins (17, 40, 43, 58). Based on these observations, we propose that this N-terminal region of Hag forms a similar roll-neck structure which could be important for adherence.
, http://www.100md.com
    It is also interesting that YadA has only one HIM-like domain, whereas the Hag/MID proteins have at least four interspersed throughout their entire lengths (Table 2 and Fig. 6B). The presence of multiple potential neck (HIM) regions in Oca adhesins has previously been observed (40, 43, 58). For instance, the B. henselae adhesin BadA displays as many as 24 potential neck (HIM) regions, which have been proposed to be important for the structural stability of this large molecule (340 kD) (43). Given its size, it is possible that the multiple HIM (neck) regions of Hag may serve a similar purpose. We also noted that one of these additional HIM domains, located near the C terminus of Hag/MID, was preceded by eight degenerate repeats resembling those associated with a roll (see Fig. 6B), an observation also made by Pearson et al. with the IAIGxxxxxxxxxxIAIG motif (42). This stretch of 112 residues is 97.3% identical in all proteins analyzed, and its contribution to the adhesive properties of Hag is under investigation.
, http://www.100md.com
    M. catarrhalis is one of the major causative agents of otitis media. Thus, the findings that Hag directly mediates adherence to human middle ear cells has important implications for vaccine development. The Hag/MID proteins exhibit several characteristics of potential vaccine antigen. Most isolates tested to date contain a hag/mid gene, and a majority express the proteins on their surface (35, 42, 52). However, our data and those of others (35, 42) clearly demonstrate that although certain domains are highly conserved (e.g., aa 1063 to 1964 of O35E-Hag, see Fig. 6A), other regions of Hag/MID vary substantially (e.g., aa 66 to 860 of O35E-Hag, see Fig. 6A). Furthermore, some regions appear to be completely deleted. For instance, a domain of 326 aa located near the C terminus of all other Hag/MID proteins analyzed in the present study is missing in O35E-Hag (see gray box in Fig. 6A). Interestingly, this domain that is missing from O35E has the highest degree of identity (97.5%) among the remaining proteins analyzed. A moderately conserved domain of 200 aa was also found to be missing in Bc5-MID but is present in all other Hag isotypes examined (see yellow box in Fig. 6A). Thus, a detailed structure-function analysis of Hag is warranted to identify domains of the molecules with the best vaccinogenic potential. The need for a functional study was highlighted by a recent report showing that immunization with MID764-963 (proposed adherence epitope for A549 cells) increased the clearance of M. catarrhalis from the lungs of mice in a pulmonary challenge model (15); this region of MID shares the highest degree of identity with O35E-Hag at 54.1%.
, 百拇医药
    Although Hag's participation in attachment to both A549 and HMEE cells is conserved, it is clear that the molecular basis for adherence differs for these cell lines, reemphasizing the need for a detailed structure-function analysis of the Hag protein in order to understand this important step in pathogenesis by M. catarrhalis. These studies will also facilitate the identification of Hag's ligands on the surface middle ear and lung epithelial cells, which may lead to novel therapeutic approaches.
, 百拇医药
    ACKNOWLEDGMENTS

    We thank Christine Akimana, Rachel Balder, Robert Blumenthal, Mark Wooten, and Darren Sledjeski (Medical College of Ohio) for critically reviewing the manuscript; Edward Toland (Medical College of Ohio) for technical assistance; Eric Hansen (University of Texas Southwestern Medical Center at Dallas) for providing antibodies and M. catarrhalis isolates; Timothy Murphy (University of Buffalo) for providing antibodies; Thomas DeMaria (Ohio State University) for providing HMEE cells; and George Hageage (Medical College of Ohio) for the M. catarrhalis isolate McGHS1.
, 百拇医药
    This study was supported by NIH/NIAID award AI051477 to E.R.L.

    REFERENCES

    1. Aebi, C., E. R. Lafontaine, L. D. Cope, J. L. Latimer, S. L. Lumbley, G. H. McCracken, Jr., and E. J. Hansen. 1998. Phenotypic effect of isogenic uspA1 and uspA2 mutations on Moraxella catarrhalis O35E. Infect. Immun. 66:3113-3119.

    2. Aebi, C., I. Maciver, J. L. Latimer, L. D. Cope, M. K. Stevens, S. E. Thomas, G. H. McCracken, Jr., and E. J. Hansen. 1997. A protective epitope of Moraxella catarrhalis is encoded by two different genes. Infect. Immun. 65:4367-4377.
, 百拇医药
    3. Bateman, A., L. Coin, R. Durbin, R. D. Finn, V. Hollich, S. Griffiths-Jones, A. Khanna, M. Marshall, S. Moxon, E. L. Sonnhammer, D. J. Studholme, C. Yeats, and S. R. Eddy. 2004. The Pfam protein families database. Nucleic Acids Res. 32(Database Issue):D138-D141.

    4. Beachey, E. H. 1981. Bacterial adherence: adhesin-receptor interactions mediating the attachment of bacteria to mucosal surface. J. Infect. Dis. 143:325-345.

    5. Benz, I., and M. A. Schmidt. 2001. Glycosylation with heptose residues mediated by the aah gene product is essential for adherence of the AIDA-I adhesin. Mol. Microbiol. 40:1403-1413.
, 百拇医药
    6. Cappelletty, D. 1998. Microbiology of bacterial respiratory infections. Pediatr. Infect. Dis. J. 17:S55-S61.

    7. Centers for Disease Control and Prevention. 2000. Pertussis. Centers for Disease Control and Prevention, Atlanta, Ga. [Online.] http://www.cdc.gov/nip/publications/pink/pert.pdf.

    8. Chiang, S. L., R. K. Taylor, M. Koomey, and J. J. Mekalanos. 1995. Single amino acid substitutions in the N terminus of Vibrio cholerae TcpA affect colonization, autoagglutination, and serum resistance. Mol. Microbiol. 17:1133-1142.
, 百拇医药
    9. Christensen, J. J. 1999. Moraxella (Branhamella) catarrhalis: clinical, microbiological and immunological features in lower respiratory tract infections. APMIS Suppl. 88:1-36.

    10. Comanducci, M., S. Bambini, D. A. Caugant, M. Mora, B. Brunelli, B. Capecchi, L. Ciucchi, R. Rappuoli, and M. Pizza. 2004. NadA diversity and carriage in Neisseria meningitidis. Infect. Immun. 72:4217-4223.

    11. Cope, L. D., E. R. Lafontaine, C. A. Slaughter, C. A. Hasemann, Jr., C. Aebi, F. W. Henderson, G. H. McCracken, Jr., and E. J. Hansen. 1999. Characterization of the Moraxella catarrhalis uspA1 and uspA2 genes and their encoded products. J. Bacteriol. 181:4026-4034.
, http://www.100md.com
    12. Faden, H. 2001. The microbiologic and immunologic basis for recurrent otitis media in children. Eur. J. Pediatr. 160:407-413.

    13. Forsgren, A., M. Brant, M. Karamehmedovic, and K. Riesbeck. 2003. The immunoglobulin D-binding protein MID from Moraxella catarrhalis is also an adhesin. Infect. Immun. 71:3302-3309.

    14. Forsgren, A., M. Brant, A. Mollenkvist, A. Muyombwe, H. Janson, N. Woin, and K. Riesbeck. 2001. Isolation and characterization of a novel IgD-binding protein from Moraxella catarrhalis. J. Immunol. 167:2112-2120.
, http://www.100md.com
    15. Forsgren, A., M. Brant, and K. Riesbeck. 2004. Immunization with the truncated adhesin Moraxella catarrhalis immunoglobulin D-binding protein (MID764-913) is protective against M. catarrhalis in a mouse model of pulmonary clearance. J. Infect. Dis. 190:352-355.

    16. Hall, T. A. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. :95-98.

    17. Hoiczyk, E., A. Roggenkamp, M. Reichenbecher, A. Lupas, and J. Heesemann. 2000. Structure and sequence analysis of Yersinia YadA and Moraxella UspAs reveal a novel class of adhesins. EMBO J. 19:5989-5999.
, http://www.100md.com
    18. Holm, M. M., S. L. Vanlerberg, I. M. Foley, D. D. Sledjeski, and E. R. Lafontaine. 2004. The Moraxella catarrhalis porin-like outer membrane protein CD is an adhesin for human lung cells. Infect. Immun. 72:1906-1913.

    19. Holm, M. M., S. L. Vanlerberg, D. D. Sledjeski, and E. R. Lafontaine. 2003. The Hag protein of Moraxella catarrhalis strain O35E is associated with adherence to human lung and middle ear cells. Infect. Immun. 71:4977-4984.
, 百拇医药
    20. Hultgren, S. J., S. Abraham, M. Caparon, P. Falk, J. W. St Geme III, and S. Normark. 1993. Pilus and nonpilus bacterial adhesins: assembly and function in cell recognition. Cell 73:887-901.

    21. Jacobs, M. R., S. Bajaksouzian, A. Windau, C. E. Good, G. Lin, G. A. Pankuch, and P. C. Appelbaum. 2004. Susceptibility of Streptococcus pneumoniae, Haemophilus influenzae, and Moraxella catarrhalis to 17 oral antimicrobial agents based on pharmacodynamic parameters: 1998-2001 US Surveillance Study. Clin. Lab. Med. 24:503-530.
, http://www.100md.com
    22. Jones, D. T. 1999. Protein secondary structure prediction based on position-specific scoring matrices. J. Mol. Biol. 292:195-202.

    23. Karalus, R., and A. Campagnari. 2000. Moraxella catarrhalis: a review of an important human mucosal pathogen. Microbes Infect. 2:547-559.

    24. Klein, J. O. 2000. The burden of otitis media. Vaccine 19(Suppl. 1):S2-S8.

    25. Klein, J. O. 1994. Otitis media. Clin. Infect. Dis. 19:823-833.
, 百拇医药
    26. Klein, J. O., D. W. Teele, and S. I. Pelton. 1992. New concepts in otitis media: results of investigations of the Greater Boston Otitis Media Study Group. Adv. Pediatr. 39:127-156.

    27. Klugman, K. P. 1996. The clinical relevance of in-vitro resistance to penicillin, ampicillin, amoxicillin and alternative agents, for the treatment of community-acquired pneumonia caused by Streptococcus pneumoniae, Haemophilus influenzae, and Moraxella catarrhalis. J. Antimicrob. Chemother. 38(Suppl. A):133-140.
, 百拇医药
    28. Lafontaine, E. R., L. D. Cope, C. Aebi, J. L. Latimer, G. H. McCracken, Jr., and E. J. Hansen. 2000. The UspA1 protein and a second type of UspA2 protein mediate adherence of Moraxella catarrhalis to human epithelial cells in vitro. J. Bacteriol. 182:1364-1373.

    29. Lafontaine, E. R., S. L. Vanlerberg, and M. M. Holm. 2003. Identification of a novel Moraxella catarrhalis gene product that confers adherence to human lung cells, abstr. 15. Presented at the Tenth Annual Midwest Microbial Pathogenesis Conference, Iowa City, IA.
, 百拇医药
    30. Lafontaine, E. R., N. J. Wagner, and E. J. Hansen. 2001. Expression of the Moraxella catarrhalis UspA1 protein undergoes phase variation and is regulated at the transcriptional level. J. Bacteriol. 183:1540-1551.

    31. Langermann, S., R. Mollby, J. E. Burlein, S. R. Palaszynski, C. G. Auguste, A. DeFusco, R. Strouse, M. A. Schenerman, S. J. Hultgren, J. S. Pinkner, J. Winberg, L. Guldevall, M. Soderhall, K. Ishikawa, S. Normark, and S. Koenig. 2000. Vaccination with FimH adhesin protects cynomolgus monkeys from colonization and infection by uropathogenic Escherichia coli. J. Infect. Dis. 181:774-778.
, 百拇医药
    32. Langermann, S., S. Palaszynski, M. Barnhart, G. Auguste, J. S. Pinkner, J. Burlein, P. Barren, S. Koenig, S. Leath, C. H. Jones, and S. J. Hultgren. 1997. Prevention of mucosal Escherichia coli infection by FimH-adhesin-based systemic vaccination. Science 276:607-611.

    33. Luke, N. R., A. J. Howlett, J. Shao, and A. A. Campagnari. 2004. Expression of type IV pili by Moraxella catarrhalis is essential for natural competence and is affected by iron limitation. Infect. Immun. 72:6262-6270.
, 百拇医药
    34. Manninen, R., P. Huovinen, and A. Nissinen. 1997. Increasing antimicrobial resistance in Streptococcus pneumoniae, Haemophilus influenzae, and Moraxella catarrhalis in Finland. J. Antimicrob. Chemother. 40:387-392.

    35. Mollenkvist, A., T. Nordstrom, C. Hallden, J. J. Christensen, A. Forsgren, and K. Riesbeck. 2003. The Moraxella catarrhalis immunoglobulin D-binding protein MID has conserved sequences and is regulated by a mechanism corresponding to phase variation. J. Bacteriol. 185:2285-2295.
, http://www.100md.com
    36. Murphy, T. F. 1996. Branhamella catarrhalis: epidemiology, surface antigenic structure, and immune response. Microbiol. Rev. 60:267-279.

    37. Murphy, T. F., C. Kirkham, E. DeNardin, and S. Sethi. 1999. Analysis of antigenic structure and human immune response to outer membrane protein CD of Moraxella catarrhalis. Infect. Immun. 67:4578-4585.

    38. Nielsen, H., J. Engelbrecht, S. Brunak, and G. von Heijne. 1997. Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Protein Eng. 10:1-6.
, http://www.100md.com
    39. Nielsen, H., J. Engelbrecht, S. Brunak, and G. von Heijne. 1997. A neural network method for identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Int. J. Neural. Syst. 8:581-599.

    40. Nummelin, H., M. C. Merckel, J. C. Leo, H. Lankinen, M. Skurnik, and A. Goldman. 2004. The Yersinia adhesin YadA collagen-binding domain structure is a novel left-handed parallel beta-roll. EMBO J. 23:701-711.
, 百拇医药
    41. Park, H. S., M. Wolfgang, and M. Koomey. 2002. Modification of type IV pilus-associated epithelial cell adherence and multicellular behavior by the PilU protein of Neisseria gonorrhoeae. Infect. Immun. 70:3891-3903.

    42. Pearson, M. M., E. R. Lafontaine, N. J. Wagner, J. W. St Geme III, and E. J. Hansen. 2002. A hag mutant of Moraxella catarrhalis strain O35E is deficient in hemagglutination, autoagglutination, and immunoglobulin D-binding activities. Infect. Immun. 70:4523-4533.
, 百拇医药
    43. Riess, T., S. G. Andersson, A. Lupas, M. Schaller, A. Schafer, P. Kyme, J. Martin, J. H. Walzlein, U. Ehehalt, H. Lindroos, M. Schirle, A. Nordheim, I. B. Autenrieth, and V. A. Kempf. 2004. Bartonella adhesin a mediates a proangiogenic host cell response. J. Exp. Med. 200:1267-1278.

    44. Roggenkamp, A., N. Ackermann, C. A. Jacobi, K. Truelzsch, H. Hoffmann, and J. Heesemann. 2003. Molecular analysis of transport and oligomerization of the Yersinia enterocolitica adhesin YadA. J. Bacteriol. 185:3735-3744.
, 百拇医药
    45. Ruuskanen, O., and T. Heikkinen. 1994. Otitis media: etiology and diagnosis. Pediatr. Infect. Dis. J. 13:S23-S26; S50-S54.

    46. Sambrook, J., and D. W. Russell. 2001. Molecular cloning: a laboratory manual, 3rd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y..

    47. Sasaki, K., L. Myers, S. M. Loosmore, and M. H. Klein. 1999. Control mechanism of the 200-kDa protein gene expression in Moraxella (Branhamella) catarrhalis, p. 89. Abstr. 99th Gen. Meet. Am. Soc. Microbiol. American Society for Microbiology, Washington, D.C.
, 百拇医药
    48. Sethi, S., N. Evans, B. J. Grant, and T. F. Murphy. 2002. New strains of bacteria and exacerbations of chronic obstructive pulmonary disease. N. Engl. J. Med. 347:465-471.

    49. Sethi, S., and T. F. Murphy. 2001. Bacterial infection in chronic obstructive pulmonary disease in 2000: a state-of-the-art review. Clin. Microbiol. Rev. 14:336-363.

    50. Sherlock, O., M. A. Schembri, A. Reisner, and P. Klemm. 2004. Novel roles for the AIDA adhesin from diarrheagenic Escherichia coli: cell aggregation and biofilm formation. J. Bacteriol. 186:8058-8065.
, 百拇医药
    51. St Geme, J. W., III. 1997. Bacterial adhesins: determinants of microbial colonization and pathogenicity. Adv. Pediatr. 44:43-72.

    52. Stutzmann Meier, P., P. Luescher, and C. Aebi. 2004. The role of hemagglutinin in autoagglutination of Moraxella catarrhalis isolates from children, poster B-188. Abstr. 104th Gen. Meet. Am. Soc. Microbiol. 2004. American Society for Microbiology, Washington, D.C.

    53. Surana, N. K., D. Cutter, S. J. Barenkamp, and J. W. St Geme III. 2004. The Haemophilus influenzae Hia autotransporter contains an unusually short trimeric translocator domain. J. Biol. Chem. 279:14679-14685.
, http://www.100md.com
    54. Timpe, J. M., M. M. Holm, S. L. Vanlerberg, V. Basrur, and E. R. Lafontaine. 2003. Identification of a Moraxella catarrhalis outer membrane protein exhibiting both adhesin and lipolytic activities. Infect. Immun. 71:4341-4350.

    55. Tong, H. H., Y. Chen, M. James, J. Van_Deusen, D. B. Welling, and T. F. DeMaria. 2001. Expression of cytokine and chemokine genes by human middle ear epithelial cells induced by formalin-killed Haemophilus influenzae or its lipooligosaccharide htrB and rfaD mutants. Infect. Immun. 69:3678-3684.
, 百拇医药
    56. Whitby, P. W., D. J. Morton, and T. L. Stull. 1998. Construction of antibiotic resistance cassettes with multiple paired restriction sites for insertional mutagenesis of Haemophilus influenzae. FEMS Microbiol. Lett. 158:57-60.

    57. Winther-Larsen, H. C., F. T. Hegge, M. Wolfgang, S. F. Hayes, J. P. van Putten, and M. Koomey. 2001. Neisseria gonorrhoeae PilV, a type IV pilus-associated protein essential to human epithelial cell adherence. Proc. Natl. Acad. Sci. USA 98:15276-15281.

    58. Zhang, P., B. B. Chomel, M. K. Schau, J. S. Goo, S. Droz, K. L. Kelminson, S. S. George, N. W. Lerche, and J. E. Koehler. 2004. A family of variably expressed outer-membrane proteins (Vomp) mediates adhesion and autoaggregation in Bartonella quintana. Proc. Natl. Acad. Sci. USA 101:13630-13635., 百拇医药(Brian Bullard, Serena L. Lipski, and Eri)