Staphylococcus epidermidis uses distinct mechanisms of biofilm formation to interfere with phagocytosis and activation of mouse macrophage-like cells 774A.1

Abstract

Assembly of adherent biofilms is the key mechanism involved in Staphylococcus epidermidis virulence during device-associated infections. Aside from polysaccharide intercellular adhesin (PIA), the accumulation-associated protein Aap and the extracellular matrix binding protein Embp act as intercellular adhesins, mediating S. epidermidis cell aggregation and biofilm accumulation. The aim of this study was to investigate structural features of PIA-, Aap-, and Embp-mediated S. epidermidis biofilms in more detail and to evaluate their specific contributions to biofilm-related S. epidermidis immune escape. PIA-, Embp-, and Aap-mediated biofilms exhibited substantial morphological differences. Basically, PIA synthesis induced formation of macroscopically visible, rough cell clusters, whereas Aap- and Embp-dependent biofilms preferentially displayed a smooth layer of aggregated bacteria. On the microscopic level, PIA was found to form a string-like organized extracellular matrix connecting the bacteria, while Embp produced small deposits of intercellular matrix and Aap was strictly localized to the bacterial surface. Despite marked differences, S. epidermidis strains using PIA, Aap, or Embp for biofilm formation were protected from uptake by J774A.1 macrophages, with similarly efficiencies. In addition, compared to biofilm-negative S. epidermidis strains, isogenic biofilm-forming S. epidermidis induced only a diminished inflammatory J774A.1 macrophage response, leading to significantly (88.2 to 88.7%) reduced NF-κB activation and 68.8 to 83% reduced interleukin-1β (IL-1β) production. Mechanical biofilm dispersal partially restored induction of NF-κB activation, although bacterial cell surfaces remained decorated with the respective intercellular adhesins. Our results demonstrate that distinct S. epidermidis biofilm morphotypes are similarly effective at protecting S. epidermidis from phagocytic uptake and at counteracting macrophage activation, providing novel insights into mechanisms that could contribute to the chronic and persistent course of biofilm-related S. epidermidis foreign material infections.

Fig. 1.

Characterization of different S. epidermidis biofilm types. (A) Analysis of biofilm formation by S. epidermidis 1457 (PIA dependent), 5179-R1 (Aap dependent), and 1585v (Embp dependent), using a microtiter plate assay. Where indicated, dispersin B (Dsp; 1 μg/ml) or proteinase K (PK; 1 mg/ml) was added to the growth medium. Depending on the respective intercellular adhesin used, the biofilms were disrupted. (B) Macroscopic view of S. epidermidis 1457, 5179-R1, and 1585v grown in 6-well cell culture plates. After overnight growth, the medium was aspirated and biofilms were documented using a digital microscope (VHX-600; Keyence, Frankfurt, Germany) at a 5-fold magnification. (C) Microscopic analysis of S. epidermidis 1457, 5179-R1, and 1585v biofilms. Bacteria were grown overnight in cell culture dishes. After gentle washing and removal of nonadherent cells, biofilms were investigated using bright-field microscopy at a 100-fold magnification. Bars, 100 μm. (D) Detection of intercellular adhesins in native S. epidermidis biofilms. Bacteria were detected using a rabbit anti-S. epidermidis antiserum and Alexa 488-coupled anti-rabbit IgG (left column). PIA was detected using wheat germ agglutinin coupled to Alexa 568. Visualization of Aap and Embp was carried out by using anti-rAap domain B antiserum (34) and anti-rEmbp6599 antiserum (5), respectively, and anti-rabbit IgG coupled with Alexa 568 (second column). Biofilms were analyzed at a 630-fold magnification. After acquisition, pictures were edited and merged using Volocity and Adobe Photoshop software. White rectangles indicate enlarged regions presented in the right column. Open arrows indicate the string-like, organized, PIA-containing biofilm matrix of S. epidermidis 1457. Closed arrows demonstrate Embp located within the intercellular space. Bars, 6 μm.

Fig. 2.

Comparison of phagocytic uptake of biofilm-forming S. epidermidis 1457 and 5179-R1 and biofilm-negative 1457-M10 and 5179. (A) Invasion of J774A.1 macrophages into the biofilm matrix produced by S. epidermidis 1457. Macrophages were stained using a PKH fluorescent cell linker kit (Sigma, St. Louis, MO) and subsequently added to 24-h-old S. epidermidis 1457 biofilms, with PIA stained using Alexa 488-coupled WGA. The biofilm was analyzed at a 100-fold magnification, using a CLSM microscope. z stacks were collected at 5-μm intervals, and the images were compiled to generate three-dimensional renderings. CLSM z-stack processing was performed using Volocity software (Improvision, Lexington, MA). Green, PIA; red, macrophages. (B) Microscopic analysis of bacterial uptake by differential inside-outside staining. Four hours after addition of murine J774A.1 macrophages transfected with pmaxGFP to statically grown S. epidermidis 5179-R1 and biofilm-negative strain 5179, extracellular bacteria were detected using a rabbit anti-S. epidermidis antiserum and a Cy5-coupled goat anti-rabbit IgG (green). After cell permeabilization of macrophages, intracellular bacteria were detected by an anti-S. epidermidis antiserum and an Alexa 568-coupled goat anti-rabbit IgG (red). Macrophage cell boundaries (dotted white lines) were determined by detection of intracellular green fluorescent protein (GFP) (blue). Colorization was done with the Adobe Photoshop software package. Bars, 12 μm. (C) Quantitative analysis of S. epidermidis 1457, 1457-M10, 5179-R1, and 5179 uptake by J774A.1 macrophages. Columns represent the mean numbers of intracellular bacteria found in 45 macrophages counted in three independent experiments; error bars indicate the standard deviations. Differences between means for biofilm-forming and -negative S. epidermidis strains were found to be statistically significant by an unpaired t test with Welch's correction (significance level, 0.05). Results were similar to those obtained with S. epidermidis 1585v and M135 (5).

Fig. 3.

Activation of J774A.1 macrophages after contact with biofilm-positive and -negative S. epidermidis. (A) Induction of NF-κB activation. Sessile, biofilm-positive and -negative S. epidermidis strains were exposed to J774A.1 macrophages transfected with pELAM-NF-κB-luciferase. After 4 h, luciferase activity was measured as a function of NF-κB activation. The relative light units (RLU) obtained after macrophage-S. epidermidis contact were normalized against values obtained after 4 h of stimulation with LPS (500 ng/ml), which were set as 100%. Columns represent means for four independent experiments. Bonferroni's test for multiple comparisons after one-way analysis of variance (ANOVA) found NF-κB activation to be significantly higher after macrophage contact with biofilm-negative strains than after contact with biofilm-positive strains (significance level, 0.05). Values (%) represent reductions of macrophage activation after exposure to biofilm-forming S. epidermidis compared to that after exposure to the respective isogenic biofilm-negative strain. (B) Induction of IL-1β production. Sessile, biofilm-positive and -negative S. epidermidis strains were incubated with J774A.1 macrophages. After 4 h, macrophages were lysed and IL-1β concentrations in supernatants were measured by ELISA. Results were normalized against values obtained after 4 h of macrophage stimulation with LPS (500 ng/ml), which were set as 100%. Columns represent means for four independent experiments. IL-1β production was found to be significantly higher after macrophage contact with biofilm-negative strains than after contact with biofilm-positive strains (Bonferroni's test for multiple comparisons after one-way ANOVA; significance level, 0.05). Values (%) represent reductions of IL-1β production by macrophages after exposure to biofilm-forming S. epidermidis compared to that after exposure to the respective isogenic biofilm-negative strain.

Fig. 4.

Interference of S. epidermidis biofilms with LPS-induced NF-κB activation. J774A.1 macrophages transfected with pELAM-NF-κB-luciferase were exposed to biofilm-forming S. epidermidis 1457, 5179-R1, and 1585v. After 2 h, LPS was added to the medium at a final concentration of 500 ng/ml. After an additional 4 h, macrophages were lysed and luciferase activity was measured as a function of NF-κB activity. RLU obtained after LPS stimulation of macrophages in contact with S. epidermidis were normalized against values obtained after 4 h of macrophage stimulation with LPS (500 ng/ml) but without contact with S. epidermidis. These were set as 100%. Columns represent means for four independent experiments. NF-κB activation was found to be significantly lower when macrophages were stimulated with LPS after contact with S. epidermidis biofilms than after LPS stimulation without contact with biofilms (Bonferroni's test for multiple comparisons after one-way ANOVA; significance level, 0.05).

Fig. 5.

Macrophage interactions with disrupted S. epidermidis biofilms. (A) Biofilms of S. epidermidis 1457, 1585v, and 5179-R1 were scraped off a cell culture plate and mildly treated with ultrasound. Immunofluorescence microscopy using rabbit anti-PIA, anti-rAap domain B, or anti-rEmbp6599 antiserum proved the presence of the respective intercellular adhesins on the surfaces of the strains after disruption. Bound antibodies were detected using goat anti-rabbit IgG coupled to Alexa 488. Pictures were taken at a 1,000-fold magnification. (B) Microscopic analysis of bacterial uptake by differential inside-outside staining. Murine J774A.1 macrophages transfected with pmaxGFP were infected with sonified biofilm-forming S. epidermidis 1457, 5179-R1, and 1585v. After 4 h, extracellular bacteria were detected using a rabbit anti-S. epidermidis antiserum and a Cy5-coupled goat anti-rabbit IgG (green). After permeabilization of cells, intracellular bacteria were detected by an anti-S. epidermidis antiserum and an Alexa 568-coupled goat anti-rabbit IgG (red). Macrophage cell boundaries (dotted white lines) were determined by detection of intracellular GFP (blue). Colorization was done with the Leica TCS software package. Bar, 12 μm. (C) Quantitative uptake analysis of sonified biofilm-forming S. epidermidis 1457, 5179-R1, and 1585v and corresponding biofilm-negative strains. Columns represent the mean numbers of intracellular bacteria found in 30 macrophages counted in two independent experiments; error bars indicate the standard deviations. No significant differences (n.s.) in intracellular bacterial counts between sonified biofilm-forming and -negative S. epidermidis strains were detected (Bonferroni's test for multiple comparisons after one-way ANOVA; significance level, 0.05). (D) NF-κB induction by sonified S. epidermidis. Macrophages transfected with pELAM-NF-κB-luciferase were exposed to disrupted biofilms for 4 h. After cell lysis, luciferase activity was measured as a function of NF-κB activation. NF-κB activation was found to be significantly higher after macrophage contact with disrupted biofilms than after contact with intact biofilms (Bonferroni's test for multiple comparisons after one-way ANOVA; significance level, 0.05).