Host-pathogen interactions of clinical S. aureus isolates to induce infective endocarditis

Abstract

To evaluate potential pathomechanisms in the induction of infective endocarditis (IE), 34 Staphylococcus aureus (S. aureus) isolates, collected from patients with S. aureus endocarditis and from healthy individuals were investigated both in vitro and in vivo. S. aureus isolates were tested in vitro for their cytotoxicity, invasion and the association with platelets. Virulence factor expression profiles and cellular response were additionally investigated and tested for correlation with the ability of S. aureus to induce vegetations on the aortic valves in vivo. In an animal model of IE valvular conspicuity was assessed by in vivo magnetic resonance imaging at 9.4 T, histology and enrichment gene expression analysis. All S. aureus isolates tested in vivo caused a reliable infection and inflammation of the aortic valves, but could not be differentiated and categorized according to the measured in vitro virulence profiles and cytotoxicity. Results from in vitro assays did not correlate with the severity of IE. However, the isolates differed substantially in the activation and inhibition of pathways connected to the extracellular matrix and inflammatory response. Thus, comprehensive approaches of host-pathogen interactions and corresponding immune pathways are needed for the evaluation of the pathogenic capacity of bacteria. An improved understanding of the interaction between virulence factors and immune response in S. aureus infective endocarditis would offer novel possibilities for the development of therapeutic strategies and specific diagnostic imaging markers.

Keywords: MRI; Staphylococcus aureus; clinical isolates; endocarditis; pathomechanisms.

Figure 1.

Cytotoxic properties, invasiveness and the ability to form S. aureus-platelet-associates as mean ± SD of a S. aureus isolate collection (34 isolates) isolated from patients with approved IE (white, n = 24) as well as from nasal swabs of healthy individuals (gray, n = 10) (a). Principal component analysis (PCA) of pathogenic profiles using cytotoxicity, invasion and platelet-associates as principal components (PCs), explaining 100 % of the total variation. The analysis shows that the isolates of the two different collection sites (S. aureus IE and nasal swabs of healthy individuals) give comparable variation along the three PCs. Isolate 17, 30 and 33 differed most in their pathogenic profiles

Figure 2.

Gene expression patterns (RT-qPCR) of major virulence factors in the three selected isolates (17, 30, 33) 3 h post inoculation: general regulatory proteins, adhesion proteins (MSCRAMMs and SERAMs), immune evasion proteins, toxins, and proteases (a). Gene expression patterns (RT-qPCR) of major immunomodulatory (CCL5, CXCL10 and IL6) and proangiogenic (ICAM) factors in EA.hy926 cells induced by the three selected isolates (17, 30, 33) 8 h post infection (b). The results are displayed as bar charts representing the mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, One-way-ANOVA with Bonferroni posttest; n.d. not detected

Figure 3.

Representative MR (upper row) and macroscopic (lower row) images (a) of the five experimental groups (naïve controls (n = 5) and sham-operated mice (n = 5) and mice infected with S. aureus isolates 17 (n = 7), 30 (n = 8) or 33 (n = 5) in the IE model) 24 h post the induction of IE. MRI score (b) and clinical score (c) of the three S. aureus endocarditis infections (17, n = 11; 30, n = 14 or 33, n = 8) and the sham-operated group (n = 17) represented as mean ± SD. CFU counts of different organs including the aortic valves (d) for IE infections induced by the bacterial isolates 17 (n = 8), 30 (n = 9) and 33 (n = 6). The results are displayed as box and whiskers plots representing data between the first and third quartiles with the band standing for the second quartile (=median). Whiskers represent lowest and highest data within 1.5 interquartile ranges of the lower and upper quartile. *p < 0.05, **p < 0.01, Mann–Whitney-U-test

Figure 4.

Histopathological analysis of aortic valves in comparison to the five experimental groups 24 h post induction of IE: naïve controls, sham-operated mice and mice infected with S. aureus isolates 17, 30 or 33 in the IE model. Hematoxylin-eosin staining (a; left column: ×4 magnification; middle column: ×10 magnification) and Gram staining (a; right column: ×10 magnification) showing massive immune cell infiltration. Immunofluorescence staining using CD45 as pan-leukocyte marker (b, red fluorescence, left column: ×10 magnification) and F4/80 as mouse macrophage marker (b, red fluorescence, right column: ×10 magnification) in representative tissue samples of the aortic valve. Blue fluorescence represents DAPI nucleic acid staining. The scale bars in the images illustrate 100 µm

Figure 5.

Gene expression in the aortic valves of the IE mouse model 24 h post infection with three clinical isolates of S. aureus (isolate 17 – n = 3, 30 – n = 4 and 33 –n = 3). A panel of genes associated with inflammation and autoimmune response (a) as well as extracellular matrix and adhesion molecules (b) are altered in non-infected control mice (sham– n = 4) and mice after infection with S. aureus clinical isolates 17, 30 and 33 (FC ≥|2|, one-sided Fisher-Exact test). Ingenuity pathway z score analysis of cellular and molecular biological functions (c) predicted increased activity of leukocyte migration in mice after infection with clinical isolate 17 and 30. Angiogenesis was predicted to be more pronounced in mice after infection with clinical isolate 33. -Log₁₀(p-value) was given for the top three biological functions (d). Remodeling processes were more pronounced in mice after infection with isolate 33 compared to the other two