Influence of preformed antibody on the pathogenesis of experimental Candida albicans endocarditis

Infective endocarditis

Infective endocarditis (IE) is a rare, life-threatening disease that has long-lasting effects even among patients who survive and are cured. IE disproportionately affects those with underlying structural heart disease and is increasingly associated with health care contact, particularly in patients who have intravascular prosthetic material. In the setting of bacteraemia with a pathogenic organism, an infected vegetation may form as the end result of complex interactions between invading microorganisms and the host immune system. Once established, IE can involve almost any organ system in the body. The diagnosis of IE may be difficult to establish and a strategy that combines clinical, microbiological and echocardiography results has been codified in the modified Duke criteria. In cases of blood culture-negative IE, the diagnosis may be especially challenging, and novel microbiological and imaging techniques have been developed to establish its presence. Once diagnosed, IE is best managed by a multidisciplinary team with expertise in infectious diseases, cardiology and cardiac surgery. Antibiotic prophylaxis for the prevention of IE remains controversial. Efforts to develop a vaccine that targets common bacterial causes of IE are ongoing, but have not yet yielded a commercially available product.

A model of the transmission of micro-organisms in a public setting and its correlation to pathogen infection risks

AIM

Gastro-intestinal infections are widespread in the community and have considerable economic consequences. In this study, we followed chains of infection from a public toilet scenario, looking at infection risks by correlating the transmission of bacteria, fungi and viruses to our current knowledge of infectious doses.

METHODS AND RESULTS

Transmission of Escherichia coli, Bacillus atrophaeus spores, Candida albicans and bacteriophage MS2 from hands to surfaces was examined in a transmission model, that is toilet brush, door handle to water tap. The load of viable pathogens was significantly reduced during transfer from hands to objects. Nevertheless, it was shown that pathogens were successfully transferred to other people in contagious doses by contact with contaminated surfaces.

CONCLUSIONS

Our results suggest that infection risks are mainly dependent on current infectious doses of pathogens. For enteritic viruses or bacteria, for example Norovirus or EHEC, only a few particles or cells are sufficient for infection in public lavatories, thus bearing a high risk of infection for other persons. However, there seems to be only a low probability of becoming infected with pathogens that have a high infectious dose whilst sharing the same bathroom.

SIGNIFICANCE AND IMPACT OF THE STUDY

The transmission model for micro-organisms enables a risk assessment of gastro-intestinal infections on the basis of a practical approach.

Antibodies to a surface-exposed, N-terminal domain of aggregation substance are not protective in the rabbit model of Enterococcus faecalis infective endocarditis

The aggregation substance (AS) surface protein from Enterococcus faecalis has been implicated as an important virulence factor for the development of infective endocarditis. To evaluate the role of antibodies specific for Asc10 (the AS protein from the conjugative plasmid pCF10) in protective immunity to infective endocarditis, an N-terminal region of Asc10 lacking the signal peptide and predicted to be surface exposed (amino acids 44 to 331; AS(44-331)) was cloned with a C-terminal histidine tag translational fusion and expressed from Escherichia coli. N-terminal amino acid sequencing of the purified protein revealed the correct sequence, and rabbit polyclonal antisera raised against AS(44-331) reacted specifically to Asc10 expressed from E. faecalis OG1SSp, but not to other proteins as judged by Western blot analysis. Using these antisera, flow cytometry analysis demonstrated that antibodies to AS(44-331) bound to a surface-exposed region of Asc10. Furthermore, antibodies specific for AS(44-331) were opsonic for E. faecalis expressing Asc10 in vitro but not for cells that did not express Asc10. New Zealand White rabbits immunized with AS(44-331) were challenged intravenously with E. faecalis cells constitutively expressing Asc10 in the rabbit model of experimental endocarditis. Highly immune animals did not show significant differences in clearance of organisms from the blood or spleen or in formation of vegetations on the aortic valve, in comparison with nonimmune animals. Although in vivo expression of Asc10 was demonstrated by immunohistochemistry, these experiments provide evidence that immunity to Asc10 does not play a role in protection from experimental infective endocarditis due to E. faecalis and may have important implications for the development of immunological approaches to combat enterococcal endocarditis.

Immunization with FimA protects against Streptococcus parasanguis endocarditis in rats

FimA, a surface-associated protein of Streptococcus parasanguis, is associated with initial colonization of damaged heart tissue in an endocarditis model (D. Burnette-Curley, V. Wells, H. Viscount, C. Munro, J. Fenno, P. Fives-Taylor, and F. Macrina, Infect. Immun. 63:4669-4674, 1995). We have evaluated the efficacy of recombinant FimA as a vaccine in the rat model of endocarditis and investigated in vitro the mechanism for the protective role of immunization. FimA-immunized and nonimmunized control animals were catheterized to induce heart valve damage and infected intravenously with 10(7) CFU of wild-type S. parasanguis FW213 bacteria. The presence of bacteria associated with platelet-fibrin vegetations 24 h postchallenge was evaluated. Immunized rats were significantly less susceptible to endocarditis (2 cases among 34 animals) than the control group (21 cases among 33 animals) (P < 0.001). Incubation of S. parasanguis FW213 with rabbit anti-FimA immune serum decreased the mean percent adherence (0.34% of added cells) to platelet-fibrin matrix in vitro compared with that of preimmune normal serum (5.04% of added cells; P < 0.001). Adsorption of immune serum with FimA-positive S. parasanguis FW213 yielded antiserum that failed to block adherence to the platelet-fibrin matrix. We assessed the vaccine potential of FimA as a common immunogen able to provide cross-protection in streptococcal endocarditis by determining the occurrence and expression of fimA in the viridans group streptococci and enterococci. We detected the presence of fimA homologs by Southern hybridization and PCR amplification analyses and determined by immunoblotting the expression of FimA-like proteins among a variety of streptococci and enterococci that frequently cause endocarditis. Eighty-one percent (26 of 32) of streptococcal and enterococcal strains isolated from bacteremic patients expressed proteins that comigrated with FimA and were reactive with polyclonal anti-FimA serum. Streptococcal DNA from strains that were positive by Western blot (immunoblot) analysis hybridized to the full-length fimA probe. Our studies suggest that FimA immunization results in antibody-mediated inhibition of bacterial adherence, a critical early event in the pathogenesis of endocarditis. Our data demonstrate that a majority of streptococcal strains associated with endocarditis have genes that encode FimA-like proteins. Taken together, these results suggest that FimA is a promising candidate for an endocarditis vaccine.

Effect of presensitization of Nocardia asteroides with specific antibody on the viability of the organism

Nocardia asteroides from various growth phases was treated in vitro with normal rabbit sera, immune rabbit sera containing nocardial polyclonal antibodies and a monoclonal antibody. At intervals, samples were grown in broth or on blood agar plates to determine their viability. Log and stationary phase cells were injected intra-peritoneally into female BALB/c mice and their survival rates in the liver and spleen were determined. Presensitization with antibodies reduced the viability of the log phase cells by 48% and that of the late stationary phase by 4%. The antibody-treated log phase organisms were less viable on the blood agar medium and in the spleen and liver than the control organisms. This indicates that pretreatment with antibody has a lethal effect on N. asteroides and affects its survival in vivo.

Antibodies to capsular polysaccharides are not protective against experimental Staphylococcus aureus endocarditis

The protective efficacy of antibodies to the Staphylococcus aureus capsular polysaccharide was examined in a rat model of catheter-induced endocarditis. Capsular antibodies were induced either by active immunization with killed S. aureus or by passive immunization with hyperimmune rabbit antiserum to S. aureus. Control rats were injected with phosphate-buffered saline or passively immunized with normal rabbit serum or rabbit antiserum to a nonencapsulated strain. Animals with indwelling catheters were challenged intravenously with 5 x 10(4) to 4 x 10(6) CFU of the homologous S. aureus strain (capsular serotype 5 strain Reynolds or serotype 1 strain SA1 mucoid). Both immunized and control rats developed S. aureus endocarditis. The numbers of S. aureus cells recovered from the blood and aortic valve vegetations of immunized rats were similar to those of control rats, indicating that capsule-specific antibodies were not protective. To determine whether the presence of an indwelling catheter interfered with antibody-mediated protection against S. aureus endocarditis, catheters were removed 2 h after insertion in additional groups of rats. An inoculum of 10(8) CFU of strain Reynolds was needed to provoke endocarditis in rats catheterized for 2 h, compared with 5 x 10(4) CFU for rats with indwelling catheters. Passively transferred capsular antibodies were not protective since both immunized and nonimmunized animals developed endocarditis, and quantitative cultures of blood and valvular vegetations revealed no differences between immunized and control animals. The findings of this study indicate that antibodies to the capsular polysaccharide are not protective in the rat model of experimental S. aureus endocarditis.

Gastric colonization with Candida albicans

We adapted a rat model of gastrointestinal candidiasis for studies of in vivo gastric colonization with Candida albicans. Whereas normal rats cleared a single intragastric inoculum of 5 x 10(6) C. albicans from the stomach within 4 hours, rats pretreated with chloramphenicol and gentamicin achieved stable gastric colonization for at least 5 days after administration of this inoculum. We next used this model to study host modifications hypothesized to alter gastric colonization. A first group received dilute HCl 4 hr before yeast inoculation, to induce acute superficial gastric erosions; another group was treated with glucocorticosteroid beginning 12 days before yeast inoculation; and another group received famotidine therapy beginning 3 days before yeast inoculation, to neutralize gastric acidity. Recovery of yeasts from stomachs was significantly different from the control group only in rats treated with steroids; greater colonization was found in the rats so treated. In a final group of experiments, we attempted to inhibit in vivo gastric colonization with yeasts by preincubation of yeasts in vitro with a polyclonal antiserum raised in rabbits against heat-killed C. albicans. We were not able to demonstrate inhibition of gastric colonization by preincubation with this antiserum in this model system.

Adherence of Staphylococcus epidermidis to fibrin-platelet clots in vitro mediated by lipoteichoic acid

The adherence of two strains of Staphylococcus epidermidis to human fibrin-platelet clots in vitro was investigated. Both strains were noncapsulated, nonhemagglutinating, and nonslime producers. Binding was not related to surface charge, carbohydrate profile, or hydrophobicity of the bacteria. Adherence was reduced four- to sixfold (P less than 0.001) on pretreatment of bacteria with lipase, while neuraminidase, trypsin, phospholipase C, and sodium periodate did not alter their binding. Pretreatment of bacteria with substances known to bind lipoteichoic acid (LTA), such as human albumin and anti-LTA antibodies, also resulted in a fourfold (P less than 0.001) reduction in adherence. Prior incubation of clots with free LTA, but not with deacylated LTA, produced a fourfold (P less than 0.001) decrease in the adherence of homologous and heterologous strains of S. epidermidis. A similar reduction was also observed when LTAs derived from Staphylococcus aureus and Streptococcus pyogenes were used. These data provide evidence that the lipid moiety of LTA has a central role in the adherence of S. epidermidis to fibrin-platelet clots in vitro.

Oral adhesion of yeasts

Oral adhesion of yeasts probably occurs by interaction between yeast cell adhesins and oral epithelial cell receptors. In Candida albicans mannoprotein, glucan, chitin, cell wall proteins, and lipids are possible adhesins. Mannoprotein appears as a fibrillar or floccular outermost layer in stationary-phase cells grown in sugar-rich medium. Preincubation of buccal epithelial cells (BECs) with concanavalin A inhibits adhesion, as does suppression of mannoprotein production by tunicamycin. Germ tubes adhere more easily to BECs and plastic than do blastospores. Methyl-alpha-D-mannoside may be analogous to the yeast adhesin or epithelial cell receptor because it inhibits adhesion of C. albicans to BECs. L-Fucose, N-acetyl-D-glucosamine, or D-mannose, having the same effect, may also function as epithelial cell receptors. Other factors affecting yeast adhesion may be fibronectin, hydrophobicity, s-IgA, and indigenous bacteria. Growth of yeasts to stationary phase in sugar-rich media promotes adhesion to acrylic, as do divalent cations and serum. Saliva, chlorhexidine, and Streptococcus salivarius inhibit adhesion of yeasts.

Autoreactive mechanisms in infective endocarditis

Infective endocarditis is not a simple interaction between a microbial agent and a cardiac valve. For the infection of a non-bacterial thrombotic vegetation, predisposition is required which is at least partially operative by a decreased suppressor T cell activity. During infection, peripheral blood natural killer cell activity is decreased, but normalizes under anti-microbial therapy. Non-major histocompatibility complex-restricted lymphocytotoxicity to isolated heart cells can be present in one third of patients. Circulating immune complexes normalize during therapy. They may be the cause of many clinical symptoms of infective endocarditis. Anti-bacterial and also anti-sarcolemmal antibodies which are cross-reactive to the bacterium are secreted in a polyclonal immune response. Anti-sarcolemmal antibodies which are cytolytic in vitro in the presence of complement may partly explain the myocardial factor of heart failure in patients with only marginal valve incompetence due to the endocarditic vegetation.

Influence of mucosal cell origin on the in vitro adherence of Candida albicans: are mucosal cells from different sources equivalent?

The influence of collecting mucosal cells from various anatomical sites, and varying the date of collection and cell donor on adhesion of Candida albicans to human epithelial cells was examined by using an in vitro adherence assay. Examination of buccal mucosal cells from twenty-four donors showed statistically significant differences in the number of attached yeasts between individuals. Sex did not exert a significant influence on adhesion. Examination of buccal mucosal cells from ten donors collected on five different dates revealed that yeast attachment to mucosal epithelial cells varied significantly within subjects across time. Epithelial cells from some donors manifested greater date-to-date variations in yeast adhesion than others. Adherence of Candida to mucosal cells from three anatomical sites (mouth, vagina and urinary tract) collected from ten different donors was also tested. Yeast adherence to buccal cells was highest, lowest using urinary tract cells, while vaginal epithelium was intermediate. Adherence to mucosal cells from three sites was significantly different both within and between individuals although some subjects manifested larger variations than others. These data suggest that the in vitro adherence of Candida albicans is influenced by mucosal cell donor, date of collection and body site of origin. Mucosal cells from different sources do not appear to be equivalent in receptiveness to C. albicans and this might explain some of the discrepancies observed when adhesion studies performed by different investigators are compared. The existing need for a more uniform methodology with which to pursue studies on fungal attachment to mucosal surfaces is emphasized.

Immunoblot analysis of the serological response in systemic candidosis

To investigate the heterogeneity in antibody response to the various antigenic determinants of Candida albicans in patients with disseminated candidosis 201 serial serum samples from 45 patients with proven systemic candidosis were examined by immunoblotting (Western blotting) for antibodies to C albicans type A NCTC 3153. 40 patients had detectable antibody. 26 different antigen bands, ranging from 104 kD to 23 kD, were identified. Despite wide variation in the antigens recognised by different patients, the antibody responses fell into six patterns (A to F). 13 patients infected during an outbreak of systemic candidosis gave the same response (A). Of a further 7 patients with this response, 2 had isolates indistinguishable from the outbreak strain. Group B responses were associated with leukaemias and lymphomas. Groups C-F responses were seen in 9 patients. Production of antibody to a 47 kD antigen, common to all six groups, occurred in all those who recovered from the infection and may therefore be of prognostic significance. Whether this antigen could be used as the basis of a vaccine remains to be determined.