A Novel Method of Quantifying Internalized Staphylococcus aureus

INTRODUCTION: Staphylococcus aureus is the major causative pathogen in osteomyelitis, responsible for approximately 50% of all cases. In some cases, S. aureus can internalize into non-phagocytic mammalian cells to allow persistent infection. Others have measured internalized bacteria after removal of non-internalized bacteria by a 2 hour treatment in gentamicin. Based on our knowledge of adherent bacteria and our experimental observations, this gentamicin treatment is inadequate as it fails to eradicate external bacteria. We have therefore developed a quantitative immunofluorescent assay to differentiate between internalized bacteria and adherent, external bacteria. We anticipate that this assay will supply an important tool for investigating how S. aureus and other pathogens are able to cause persistent orthopaedic infection. METHODS: Mouse embryonic fibroblasts (line FN +/H4) were cultured in DMEM + 10% FBS (complete media). All experiments were performed at cell densities of 5,000 cells/cm, and after cells had reached 90% confluence. For imaging analysis, S. aureus strain AH1710, a strain expressing GFP on a chloramphenicol resistance plasmid was used. For the lyse and plate analysis reported for literature internalization studies, S. aureus ATCC 25923 were also used; these same bacteria were used for the quantitative immunofluorescent (quantImmuno) method. All bacteria were grown overnight in trypticase soy broth (TSB), 37°C, with agitation; AH1710 media included 10μg/ml chloramphenicol. This overnight culture was subcultured for 2-4 hours, rinsed in phosphate buffered saline (PBS), and diluted to a concentration of 10 colony forming units (CFU)/ml in complete media. Prior to co-incubation, fibroblasts were rinsed with PBS, fresh media was added, and 10 CFU/ml bacteria added to the fibroblast culture. Bacteria and cells were co-incubated for 2 hours at 37°C, 5% CO2. For studies with gentamicin, at 2 h, cells were rinsed with PBS and incubated with complete media or complete media plus 100 μg/ml of gentamicin. In all experiments, bacterial inoculant was verified by serial dilution and plating on aerobiccount Petrifilms (3M, 6406). For experiments using the quantImmuno method, a standard curve of bacterial concentrations was prepared on a separate 96-well plate. The plate was incubated concurrently with the cellular experiments, and then treated as described below for quantImmuno. Lyse and plate method: fibroblasts were lysed in PBS + 0.3% Tween-20 with sonication, 10 min. Recovered bacteria in the lysates were serially diluted in PBS, and plated on Petrifilms, and incubated overnight at 37°C. QuantImmuno and microscopic visualization: cells were rinsed 4 times in PBS and then fixed in 4% paraformaldehyde for 10 minutes. To determine total number of bacteria (external + internal), fibroblasts were permeabilized, 30 minutes in 0.3% Triton-X 100, and rinsed with PBS 3X. To determine number of external bacteria, cultures were not permeabilized. After this step, bacterial and cellular alkaline phosphatase activity was inactivated by incubation at 72°C, 2 h. Plates were blocked with 0.3% nonfat dry milk in PBS for 30 minutes, followed by incubation with rabbit antiStaphylococcus aureus polyclonal IgG (1:100, Abcam, ab20920), 2 h, at room temperature. Plates were rinsed in PBS 4X, incubated in blocking buffer, 30 min, room temperature, and incubated with alkaline phosphatase-conjugated goat anti-rabbit IgG (H+L) (1:200, Invitrogen, G-21079) After 4 PBS rinses, alkaline phosphatase activity was analyzed using a PNPP substrate kit (Thermo Scientific Pierce, 37620). A405 was measured and values normalized to controls that had no primary antibody. Numbers of bacteria were calculated by comparison to the standard curve. To determine number of internalized bacteria, the external bacterial count (not permeabilized) was subtracted from the total bacterial count (permeabilized). Statistics were performed using the Student’s t-test, with significance at p* < 0.01. Experiments were each performed three times. Image analysis was performed on cultures stained as above, except primary antibody: 1:500, and secondary antibody: Alexa Fluor 568 goat anti-rabbit IgG (H+L) (1:500 Invitrogen, A-11011). Slides were mounted using Vectashield + DAPI (Vector Laboratories, H-1200). Digital images were recorded with a 20x lens. Composites were prepared using ImagePro 6.2. Figure 1. Cells that were coincubated with S. aureus AH 1710 for 2 h, rinsed, and incubated with complete media + 100 μg/ml gentamicin for 2 h; brightfield overlaid with GFP. Black arrows: bacteria not associated with a cell. Figure 2. Fibroblasts + S. aureus AH1710 after 2 h incubation followed by 2 h gentamicin treatment. Left shows antibody staining specific for extracellular S. aureus, middle shows the same field visualizing the GFP-expressing bacteria, and right is a merged image showing nuclei (DAPI), GFP and antibody fields. [DAPI in blue, GFP in green, and AlexaFluor 568 in red]. Red arrows indicate internalized bacteria which are GFP positive but AlexaFluor 568 negative. Figure 3. Number of internalized bacteria using lyse and plate method (CFU) or ImmunoQuant method. *p<0.01;