Serum resistance in Haemophilus ducreyi requires outer membrane protein DsrA

Haemophilus ducreyi is resistant to killing by normal serum antibody and complement. We discovered an H. ducreyi outer membrane protein required for expression of serum resistance and termed it DsrA (for "ducreyi serum resistance A"). The dsrA locus was cloned, sequenced, and mutagenized. An isogenic mutant (FX517) of parent strain 35000 was constructed and characterized, and it was found to no longer express dsrA. FX517 was at least 10-fold more serum susceptible than 35000. DsrA was expressed by all strains of H. ducreyi tested, except three naturally occurring, avirulent, serum-sensitive strains. FX517 and the three naturally occurring dsrA-nonexpressing strains were complemented in trans with a plasmid expressing dsrA. All four strains were converted to a serum-resistant phenotype, including two that contained truncated lipooligosaccharide (LOS). Therefore, serum resistance in H. ducreyi does not require expression of full-length LOS but does require expression of dsrA. The dsrA locus from eight additional H. ducreyi strains was sequenced, and the deduced amino acid sequences were more than 85% identical. The major difference between the DsrA proteins was due to the presence of one, two, or three copies of the heptameric amino acid repeat NTHNINK. These repeats account for the variability in apparent molecular mass of the monomeric form of DsrA (28 to 35 kDa) observed in sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Since DsrA is present in virulent strains, is highly conserved, and is required for serum resistance, we speculate that it may be a virulence factor and a potential vaccine candidate.     

Swine model of Haemophilus ducreyi infection.

Haemophilus ducreyi is a strict human pathogen that causes sexually transmitted genital ulcer disease. We infected domestic swine with H. ducreyi 35000, resulting in the development of cutaneous ulcers histologically resembling human chancroid lesions. Intraepidermal lesions progressed from pustules to ulcers containing polymorphonuclear leukocytes and were accompanied by a dermal inflammatory infiltrate containing T cells and macrophages. H. ducreyi was recovered from lesions up to 17 days after inoculation, and pigs did not develop immunity to reinfection with the challenge strain. Features of the model include inoculation through abrasions in the epidermis, ambient housing temperatures for infected pigs, the ability to deliver multiple different inocula to a single host, and the availability of monoclonal antibodies against porcine immune cells permitting immunohistochemical characterization of the host immune response to H. ducreyi infection.     

A hemoglobin-binding outer membrane protein is involved in virulence expression by Haemophilus ducreyi in an animal model.

Haemophilus ducreyi exhibits a requirement for exogenously supplied heme for aerobic growth in vitro. Nine of ten wild-type isolates of H. ducreyi were shown to contain a readily detectable hemoglobin-binding activity. Spontaneous hemoglobin-binding-negative mutants of two of these wild-type isolates lost the ability to express an outer membrane protein with an apparent molecular mass of approximately 100 kDa. Similarly, the single wild-type isolate that lacked the ability to bind hemoglobin also appeared to lack expression of this same 100-kDa protein. A monoclonal antibody (5A9) to this 100-kDa protein was used to identify a recombinant clone which possessed an H. ducreyi chromosomal fragment containing the gene encoding the 100-kDa protein; this protein was designated hemoglobin utilization protein A (HupA). Nucleotide sequence analysis of the hupA gene revealed that the predicted protein, with a calculated molecular mass of 108 kDa, was similar to TonB-dependent outer membrane proteins of other bacteria. Increasing the concentration of heme in the growth medium resulted in decreased expression of the HupA protein. Mutant analysis was used to prove that the HupA protein was essential for the utilization by H. ducreyi of both hemoglobin and hemoglobin-haptoglobin as sources of heme in vitro. In addition, it was found that an isogenic hupA mutant was less virulent than the wild-type parent strain in the temperature-dependent rabbit model for dermal lesion production by H. ducreyi.     

Outer membrane protein DsrA is the major fibronectin-binding determinant of Haemophilus ducreyi

The ability to bind extracellular matrix proteins is a critical virulence determinant for skin pathogens. Haemophilus ducreyi, the etiological agent of the genital ulcer disease chancroid, binds extracellular matrix components, including fibronectin (FN). We investigated H. ducreyi FN binding and report several important findings about this interaction. First, FN binding by H. ducreyi was greatly increased in bacteria grown on heme and almost completely inhibited by hemoglobin. Second, wild-type strain 35000HP bound significantly more FN than did a dsrA mutant in two different FN binding assays. Third, the expression of dsrA in the dsrA mutant restored FN binding and conferred the ability to bind FN to a non-FN-binding Haemophilus influenzae strain. Fourth, an anti-DsrA monoclonal antibody partially blocked FN binding by H. ducreyi. The hemoglobin receptor, the collagen-binding protein, the H. ducreyi lectin, the fine-tangle pili, and the outer membrane protein OmpA2 were not involved in H. ducreyi FN binding, since single mutants bound FN as well as the parent strain did. However, the major outer membrane protein may have a minor role in FN binding by H. ducreyi, since a double dsrA momp mutant bound less FN than did the single dsrA mutant. Finally, despite major sequence differences, DsrA proteins from both class I and class II H. ducreyi strains mediated FN and vitronectin binding. We concluded that DsrA is the major factor involved in FN binding by both classes of H. ducreyi strains.     

Differences in host susceptibility to disease progression in the human challenge model of Haemophilus ducreyi infection

With human volunteers inoculated at two sites with Haemophilus ducreyi, outcomes for a subject were not independent. In a reinfection trial, 2 of 11 previous pustule formers and 6 of 10 previous resolvers resolved all sites of infection. There was no correlation between serum bactericidal or phagocytic activity and outcome in the trial. These data indicate that different hosts are differentially susceptible to disease progression versus resolution in the model.     

Localization of Haemophilus ducreyi at the pustular stage of disease in the human model of infection

To localize Haemophilus ducreyi in vivo, human subjects were experimentally infected with H. ducreyi until they developed a painful pustule or for 14 days. Lesions were biopsied, and biopsy samples were fixed in 4% paraformaldehyde, and cryosectioned. Sections were stained with polyclonal anti-H. ducreyi antiserum or H. ducreyi-specific monoclonal antibodies (MAbs) and fluorescently tagged secondary antibodies and examined by confocal microscopy. We identified H. ducreyi in 16 of 18 pustules but did not detect bacteria in the one papule examined. H. ducreyi was observed as individual cells and in clumps or chains. Staining with MAbs 2D8, 5C9, 3B9, 2C7, and 9D12 demonstrated that H. ducreyi expresses the major pilus subunit, FtpA, the 28-kDa outer membrane protein Hlp, the 18-kDa outer membrane protein PAL, and the major outer membrane protein (MOMP) or OmpA2 in vivo. By dual staining with polyclonal anti-H. ducreyi antiserum and MAbs that recognize human skin components, we observed bacteria within the neutrophilic infiltrates of all positively staining pustules and in the dermis of 10 of 16 pustules. We were unable to detect bacteria associated with keratinocytes in the samples examined. The data suggest that H. ducreyi is found primarily in association with neutrophils and in the dermis at the pustular stage of disease in the human model of infection.     

Expression of peptidoglycan-associated lipoprotein is required for virulence in the human model of Haemophilus ducreyi infection

Haemophilus ducreyi expresses a peptidoglycan-associated lipoprotein (PAL) that exhibits extensive homology to Haemophilus influenzae protein 6. We constructed an isogenic PAL mutant (35000HP-SMS4) by the use of a suicide vector that contains lacZ as a counterselectable marker. H. ducreyi 35000HP-SMS4 and its parent, 35000HP, had similar growth rates in broth and similar lipooligosaccharide profiles. 35000HP-SMS4 formed smaller, more transparent colonies than 35000HP and, unlike its parent, was hypersensitive to antibiotics. Complementation of the mutant in trans restored the parental phenotypes. To test whether expression of PAL is required for virulence, nine human volunteers were experimentally infected. Each subject was inoculated with two doses (41 to 89 CFU) of live 35000HP and one dose of heat-killed bacteria on one arm and with three doses (ranging from 28 to 800 CFU) of live 35000HP-SMS4 on the other arm. Papules developed at similar rates at sites inoculated with the mutant or parent but were significantly smaller at mutant-inoculated sites than at parent-inoculated sites. The pustule formation rate was 72% (95% confidence interval [CI], 46.5 to 90.3%) at 18 parent sites and 11% (95% CI, 2.4 to 29.2%) at 27 mutant sites (P < 0.0001). The rates of recovery of H. ducreyi from surface cultures were 8% (n = 130; 95% CI, 4.3 to 14.6%) for parent-inoculated sites and 0% (n = 120; 95% CI, 0.0 to 2.5%) for mutant-inoculated sites (P < 0.001). H. ducreyi was recovered from six of seven biopsied parent-inoculated sites and from one of three biopsied mutant-inoculated sites. Confocal microscopy confirmed that the bacteria present in a mutant inoculation site pustule lacked a PAL-specific epitope. Although biosafety regulations precluded our testing the complemented mutant in humans, these results suggest that expression of PAL facilitates the ability of H. ducreyi to progress to the pustular stage of disease.     

Antimicrobial susceptibility and characterization of outer membrane proteins of Haemophilus ducreyi isolated in Thailand.

The identity of each of 239 yeasts, encompassing 9 genera and 30 species, was determined with the Quantum II and API 20C identification systems. With API 20C results accepted as being correct, Quantum II proved to be 92% accurate in identification of common isolates, e.g., Candida albicans and Torulopsis glabrata, but only 73% effective with less frequently encountered yeasts, e.g., Trichosporon beigelii and Rhodotorula glutinis. Overall, Quantum II was 86% as accurate as API 20C for the yeast isolates tested. In addition, physical problems were encountered in inoculation of Quantum II cartridges and in the automated reading of some biochemical tests. We conclude that Quantum II is not yet suitable for routine identification of clinical yeast isolates.     

Characterization of an 18,000-molecular-weight outer membrane protein of Haemophilus ducreyi that contains a conserved surface-exposed epitope.

Identification of antigenically conserved surface components of Haemophilus ducreyi may facilitate the development of reagents to diagnose and prevent chancroid. A hybridoma derived from a mouse immunized with nontypeable Haemophilus influenzae produced a monoclonal antibody (MAb), designated 3B9, that bound to 35 of 35 H. ducreyi strains isolated from diverse geographic regions. The MAb 3B9 bound to a non-heat-modifiable H. ducreyi outer membrane protein (OMP) whose apparent molecular weight was 18,000 (the 18K OMP), and the 3B9 epitope did not phase vary at a rate of greater than 10(-3) in H. ducreyi. In immunoelectron microscopy, the 3B9 epitope was surface exposed, and there was intrastrain and interstrain variability in the amount of 3B9 labelling of whole cells. The MAb 3B9 cross-reacted with many species of the family Pasteurellaceae and bound to the 16.6K peptidoglycan-associated lipoprotein (P6 or PAL) of H. influenzae. Unlike P6, the 18K OMP did not copurify with peptidoglycan. In Western blots (immunoblots), five of seven serum samples obtained from patients with chancroid and four of five serum samples obtained from patients with other genital ulcer diseases at the time of presentation contained antibodies that bound to the 18K OMP. In a competition enzyme-linked immunosorbent assay, four of these serum samples inhibited the binding of 3B9 to H. ducreyi by more than 50%. We conclude that members of Pasteurellaceae expressed a conserved epitope on OMPs that sometimes had different physical characteristics. Patients with chancroid usually have antibodies to the 18K OMP and the 3B9 epitope that may have resulted from infection with H. ducreyi or previous exposure to other Haemophilus or Actinobacillus sp. strains.     

Cloning, overexpression, purification, and immunobiology of an 85-kilodalton outer membrane protein from Haemophilus ducreyi

We have identified an 85-kDa outer membrane protein that is expressed by all tested strains of Haemophilus ducreyi. Studies of related proteins from other pathogenic bacteria, including Haemophilus influenzae, Pasteurella multocida, Neisseria gonorrhoeae, Neisseria meningitidis, and Shigella dysenteriae, suggested a role for these proteins in pathogenesis and immunity. In keeping with the first such described protein from Haemophilus influenzae type B, we termed the H. ducreyi protein D15. The gene encoding the H. ducreyi D15 protein was cloned and sequenced, and the deduced amino acid sequence was found to be most similar to sequences of the D15-related proteins from other Pasteurella spp. The arrangement of the flanking genes was similar to that of H. influenzae Rd and suggested that D15 was part of a multigene operon. Attempts to make a null mutation of the D15 gene were unsuccessful, paralleling results in other D15 gene studies. Overexpression of H. ducreyi D15 in Escherichia coli resulted in a source of recombinant D15 (rD15) from which it was readily purified. rD15 was immunogenic, and it was found that immunization of rabbits with an rD15 vaccine preparation conferred partial protection against a virulent challenge infection. Antisera to an N-terminal peptide recognized all tested strains of H. ducreyi.     

A superoxide dismutase C mutant of Haemophilus ducreyi is virulent in human volunteers

Haemophilus ducreyi produces a periplasmic copper-zinc superoxide dismutase (Cu-Zn SOD), which is thought to protect the organism from exogenous reactive oxygen species generated by neutrophils during an inflammatory response. We had previously identified the gene, sodC, responsible for the production and secretion of Cu-Zn SOD and constructed an isogenic H. ducreyi strain with a mutation in the sodC gene (35000HP-sodC-cat). Compared to the parent, the mutant does not survive in the presence of exogenous superoxide (L. R. San Mateo, M. Hobbs, and T. H. Kawula, Mol. Microbiol. 27:391-404, 1998) and is impaired in the swine model of H. ducreyi infection (L. R. San Mateo, K. L. Toffer, P. E. Orndorff, and T. H. Kawula, Infect. Immun. 67:5345-5351, 1999). To test whether Cu-Zn SOD is important for bacterial survival in vivo, six human volunteers were experimentally infected with 35000HP and 35000HP-sodC-cat and observed for papule and pustule formation. Papules developed at similar rates at sites inoculated with the mutant or parent. The pustule formation rates were 75% (95% confidence intervals [CI], 43 to 95%) at 12 parent-inoculated sites and 67% (95% CI, 41 to 88%) at 18 mutant-inoculated sites (P = 0.47). There was no significant difference in levels of H. ducreyi recovery from mutant- and parent-inoculated biopsy sites. These results suggest that expression of Cu-Zn SOD does not play a major role in the survival of this pathogen in the initial stages of experimental infection of humans.     

Characterization of a mutant of Haemophilus influenzae type b lacking the P2 major outer membrane protein.

An isogenic mutant of Haemophilus influenzae type b (Hib) lacking the ability to express the P2 major outer membrane (porin) protein was constructed and characterized in various model systems. Linker insertion mutagenesis of a cloned Hib DNA insert containing the P2 structural gene was used in conjunction with a genetic transformation system to obtain a transformant unreactive with a P2-specific monoclonal antibody. This transformant was shown to lack detectable P2 protein by both protein staining and immunoblot methods. The P2 mutant exhibited a generation time in complex broth medium that was significantly longer than that of the wild-type parent strain. The P2 mutant was also unable to produce detectable bacteremia in infant rats after intraperitoneal challenge, while the wild-type parent strain produced bacteremia in all animals challenged with this strain. Reintroduction of a wild-type copy of the P2 gene into this mutant yielded a transformant strain that had a generation time in vitro identical to that of the wild-type parent strain and that was also fully virulent in the infant rat model. These findings suggest that the ability to synthesize the P2 protein may be necessary but not sufficient for full expression of virulence by this pathogen.     

Identification and purification of a conserved heme-regulated hemoglobin-binding outer membrane protein from Haemophilus ducreyi.

A hemoglobin-binding protein (HgbA) from Haemophilus ducreyi was identified and purified. The 100-kDa HgbA was detected in all strains of H. ducreyi tested, and a somewhat larger hemoglobin-binding protein was found in one strain of Haemophilus influenzae. HgbA was purified and the amino acid sequence of the N terminus of HgbA revealed no significant homologies with known proteins. Two different antisera to HgbA from H. ducreyi 35000 recognized HgbA proteins from all tested H. ducreyi strains; they did not recognize proteins from the H. influenzae strain. Expression of HgbA was regulated by the level of heme but not by iron present in the medium. Animal species of hemoglobin competed with iodinated human hemoglobin for binding to whole cells of H. ducreyi and supported the growth of H. ducreyi. The lack of immunological cross-reactivity and the differences in hemoglobin specificities between the H. ducreyi and the H. influenzae hemoglobin-binding proteins suggest that they are unrelated.     

The conserved 18,000-molecular-weight outer membrane protein of Haemophilus ducreyi has homology to PAL.

Haemophilus ducreyi expresses an 18,000-molecular-weight outer membrane protein that contains a conserved surface-exposed epitope recognized by monoclonal antibody 3B9. Monoclonal antibody 3B9 cross-reacts with proteins of similar molecular weight found in many Haemophilus sp. strains, including P6, a candidate vaccine for Haemophilus influenzae. The gene encoding the 18,000-molecular-weight outer membrane protein was identified by screening a lambdagt11 genomic library with 3B9. The coding sequence of the gene was localized to a 471-bp open reading frame, designated pal (peptidoglycan-associated lipoprotein). Translation of pal predicted a mature polypeptide with a molecular weight of 15,000 that had extensive homology with P6 and Escherichia coli PAL. The predicted signal peptide had features characteristic of a prokaryotic lipoprotein, and processing of PAL was sensitive to globomycin in H. ducreyi. The sequences encoding mature H. ducreyi PAL were subcloned into the vector pRSET B and expressed as a polyhistidine-containing fusion protein that bound 3B9. In Western blot (immunoblot) analysis, serum samples obtained from healthy subjects and patients with chancroid or other genital ulcer diseases contained antibodies to purified PAL. Antibodies that bound to PAL were removed by absorption with a lysate of Haemophilus sp. antigens, suggesting that patients with chancroid do not develop an H. ducreyi-specific antibody response to PAL.     

Expression of Haemophilus ducreyi collagen binding outer membrane protein NcaA is required for virulence in swine and human challenge models of chancroid

Haemophilus ducreyi, the etiologic agent of the sexually transmitted genital ulcer disease chancroid, has been shown to associate with dermal collagen fibers within infected skin lesions. Here we describe NcaA, a previously uncharacterized outer membrane protein that is important for H. ducreyi collagen binding and host colonization. An H. ducreyi strain lacking the ncaA gene was impaired in adherence to type I collagen but not fibronectin (plasma or cellular form) or heparin. The mutation had no effect on serum resistance or binding to HaCaT keratinocytes or human foreskin fibroblasts in vitro. Escherichia coli expressing H. ducreyi NcaA bound to type I collagen, demonstrating that NcaA is sufficient to confer collagen attachment. The importance of NcaA in H. ducreyi pathogenesis was assessed using both swine and human experimental models of chancroid. In the swine model, 20% of lesions from sites inoculated with the ncaA mutant were culture positive for H. ducreyi 7 days after inoculation, compared to 73% of wild-type-inoculated sites. The average number of CFU recovered from mutant-inoculated lesions was also significantly reduced compared to that recovered from wild-type-inoculated sites at both 2 and 7 days after inoculation. In the human challenge model, 8 of 30 sites inoculated with wild-type H. ducreyi progressed to the pustular stage, compared to 0 of 30 sites inoculated with the ncaA mutant. Together these results demonstrate that the collagen binding protein NcaA is required for H. ducreyi infection.     

Evaluation of humoral and cell-mediated inducible immunity to Haemophilus ducreyi in an animal model of chancroid.

To study the mechanisms of inducible immunity to Haemophilus ducreyi infection in the temperature-dependent rabbit model of chancroid, we conducted passive immunization experiments and characterized the inflammatory infiltrate of chancroidal lesions. Polyclonal immunoglobulin G was purified from immune sera raised against H. ducreyi 35000 whole-cell lysate or a pilus preparation and from naive control rabbits. Rabbits were passively immunized with 24 or 48 mg of purified polyclonal immunoglobulin G intravenously, followed 24 h after infusion by homologous titered infectious challenge. Despite titratable antibody, no significant difference in infection or disease was observed. We then evaluated the immunohistology of lesions produced by homologous-strain challenge in sham-immunized rabbits and those protectively vaccinated by pilus preparation immunization. Immunohistochemical stains for CD5 and CD4 T-lymphocyte markers were performed on lesion sections 4, 10, 15, and 21 days from infection. Lesions of pilus preparation vaccinees compared with those of controls had earlier infiltration with significantly more T lymphocytes (CD5+) and with a greater proportion of CD4+ T lymphocytes at day 4 (33% +/- 55% versus 9.7% +/- 2%; P = 0.002), corroborating earlier sterilization (5.0 +/- 2 versus 13.7 +/- 0.71 days; P < 0.001) and lesion resolution. Intraepithelial challenge of pilus-vaccinated rabbits with 100 micrograms of the pilus preparation alone produced indurated lesions within 48 h with lymphoid and plasmacytoid infiltration, edema, and extravasation of erythrocytes. We conclude that passive immunization may not confer a vaccine effect in this model and that active vaccination with a pilus preparation induces a delayed-type hypersensitivity skin test response and confers protection through cell-mediated immunity seen as an amplified lymphocytic infiltrate and accelerated maturation of the T-lymphocyte response.     

Clustering of an outer membrane adhesin of Haemophilus parainfluenzae.

Haemophilus parainfluenzae synthesizes an outer membrane protein adhesin which mediates binding to oral streptococci, salivary pellicle, and neuraminidase-treated erythrocytes. An indirect gold labeling technique and immunoelectron microscopy verified the location of this outer membrane protein. Further, a clustering of gold particles was observed in irregular patches at the cell surface.