Use of Quantitative Scanning Electron Microscopy of S. aureus Biofilm Formation in vitro to Identify Strain and Implant Material Specificities

Size: px

Start display at page:

Download "Use of Quantitative Scanning Electron Microscopy of S. aureus Biofilm Formation in vitro to Identify Strain and Implant Material Specificities"

Kevin Fields
6 years ago
Views:

1 Use of Quantitative Scanning Electron Microscopy of S. aureus Biofilm Formation in vitro to Identify Strain and Implant Material Specificities Werasak Sutipornpalangkul, MD, PhD 1, Kohei Nishitani, MD, PhD 1, Karen L. de Mesy Bentley, MS 1, John J. Varrone, MS 1, Andrew D. Shubin 1, Paul T. Rubery, MD 1, Hiromu Ito, MD, PhD 2, Stephen L. Kates, MD 1, John L. Daiss, PhD 1, Edward M. Schwarz, PhD 1. 1 University of Rochester, Rochester, NY, USA, 2 Kyoto University, Kyoto, Japan. Disclosures: W. Sutipornpalangkul: None. K. Nishitani: None. K.L. de Mesy Bentley: None. J.J. Varrone: None. A.D. Shubin: None. P.T. Rubery: None. H. Ito: None. S.L. Kates: None. J.L. Daiss: 4; Telephus. 5; Telephus. E.M. Schwarz: 4; Telephus. 5; Telephus. Introduction: Infection remains a major complication of orthopaedic surgery, of which ~50-60% is caused by Staphylococcus aureus. Although biofilm formation on the implant is a critical event in the pathogenesis of orthopaedic infections, the natural history of bacterial adhesion, proliferation, and glycocalyx matrix production remains poorly understood due to the lack of quantitative methods to assess differences between strains or implant materials. While results from static in vitro biofilm assays are well documented, our scanning electron microscopy (SEM) results revealed that most strains do not produce glycocalyx in static cultures. Thus, we chose to investigate the natural history of S. aureus in vitro biofilm formation using a previously described flow chamber model (1). After confirming glycocalyx production in the flow chamber, we develop a quantitative SEM approach to characterize the kinetics and phenotype of biofilm formation by different S. aureus strains. We used bioluminescent S. aureus strains to better understand the nature of this biomarker of bacterial growth and biofilm formation. Finally, we compared biofilm formation on Kirschner (K) wires of both titanium (Ti) and stainless steel (SS) to test material resistance to S. aureus colonization. Methods: Three different S. aureus strains were used: 1) SH1000, 2) UAMS-1 (and its bioluminescent version Xen40), and 3) USA300 LAC (and its bioluminescent version USA300 LAC::lux). All strains were cultured in tryptic soy broth (TSB) and grown overnight at 37 O C. Then a 1:200 dilution was inoculated into media in a flow chamber that continuously circulated the culture with a flow rate of 0.2 ml/min at 37 C. The chamber contained a flat (0.5x0.2x7 mm) stainless steel ribbon that was used to assess differences between the strains. To assess differences between clinically relevant metal implants, the chamber contained round orthopaedic grade K-wires made of SS or Ti. All materials were coated with 20% human plasma at 4 O C overnight prior to placement in chamber. To assess the effects of host factors, the chamber was loaded with TSB media enriched with glucose and NaCl (TSBGN), or TSB with 10% human plasma (TSB+HP). After the desired incubation period, the implants were removed and analysed for bioluminescence intensity (BLI) and morphological appearance by SEM. BLI was assessed using an IVIS Lumina II. To quantify the BLI, the signal in a circular region of interest (ROI) set to 1cm diameter was captured and reported as the average radiance (p/s/cm 2 /sr) using the Living Image software. SEM analysis of the biofilm on the implants was performed. Quantitation of the biofilm on the pins was performed using an ROI analysis of a 150X SEM image, and the % surface covered by biofilm determined by Image J software. Statistical analyses were performed by ANOVA, in which p<0.05 was considered to be a significant difference. Results: Gross assessment of the SEM images revealed that S. aureus inoculation was required for biofilm formation, and that the phenotype of biofilm production after 24 hrs by the 3 strains markedly differed based on the culture media (Fig. 1A). SH1000 produced more biofilm in TSBGN, and its biofilm in TSB+HP did not contain fibronectin fibers. UAMS-1 biofilm formation in TSBGN was scant and limited to small clusters of bacteria without matrix, while its biofilm was robust in TSB+HP that were extensively covered by matrix containing fibronectin fibers as determined by immuno-gold labelling. USA300 produced robust biofilm in both culture media, however fibronectin fibers were only produced in TSB+HP media. SEM quantification confirmed these significant differences (Fig. 1B). The time course study of BLI and biofilm production by Xen40 and USA300 LAC::lux revealed that UAMS-1 adheres to the implant immediately via secondary attachment to the fibronectin fibers that form directly on the wire, with increasing BLI and biofilm growth until the peak at 9 hrs (Fig. 2). In contrast, biofilm formation by USA300 commences via direct bacterial adherence to the pin after 3 hrs, and achieves a rapid increase in BLI and biofilm formation at 6 hrs, with a similar peak at 9 hrs to that observed with UAMS-1 (Fig. 3). BLI and SEM results from the K-wire experiment revealed that in vitro growth and biofilm formation by UAMS-1 and USA300 on SS and Ti surfaces are virtually identical (Fig. 4).

2 Figure 1A. Glucose/salt versus plasma requirements for in vitro biofilm formation by different S. aureus strains assessed by SEM at 24hr. Figure 1B Quantification of in vitro biofilm formation. The data are presented as the mean +/- SD (*p<0.05 vs. no inoculation in TSBGN media; # p<0.05 vs. no inoculation in TSB+HP p<0.05 TSBGN vs. TSB+HP for each strain).

Figure 2. The kinetics of UAMS-1 in vitro biofilm formation. Bioluminescent UAMS-1 (Xen40) was inoculated and were harvested at the indicated time (n=4).

3 Figure 2. The kinetics of UAMS-1 in vitro biofilm formation. Bioluminescent UAMS-1 (Xen40) was inoculated and were harvested at the indicated time (n=4). (A) Heat map images of the harvested pins. (B) Quantification of the BLI signal in the 1 cm 2 ROI. (C) Representative SEM images of the pin are shown at 80X (left), and at 5,000x (right).

4 Figure 3. The kinetics of USA300 in vitro biofilm formation. Bioluminescent USA300 (USA300 LAC:lux) was inoculated. (A) Heat map images of the harvested pins. (B) Quantification of the BLI signal in the 1 cm 2 ROI. (C) Representative SEM images of the pin are shown at 80X (left), and at 5,000X (right).

Figure 4. UAMS-1 and USA300 produce similar biofilm on SS and Ti K-wires. Bioluminescent UAMS-1 or USA300 were inoculated into the flow chamber containing K-wires made of SS or Ti.

5 Figure 4. UAMS-1 and USA300 produce similar biofilm on SS and Ti K-wires. Bioluminescent UAMS-1 or USA300 were inoculated into the flow chamber containing K-wires made of SS or Ti. (A) Heat map images of the harvested K-wires. (B) Quantification of the BLI signal in the 1 cm 2 ROI. (C) Representative SEM images of the biofilm on the implant are shown at 30X. Discussion: Biofilm formation on implants is a major cause of chronic infection following orthopaedic surgery. There is a great demand for anti-microbial coated implants that can inhibit biofilm formation. We have demonstrated that this problem is complicated by strain-specific biofilm phenotypes that must be accounted for in order to protect against all S. aureus infections. While anti-adhesin coatings may be effective for strains like SH1000 that directly bind to the surface of metal implants, this approach would likely be ineffective against MSSA strains like UAMS-1 that bind to the implant via host factors. Moreover, the need for a combined anti-microbial coating approach is highlighted by the highly virulent MRSA strain USA300, which is capable of forming biofilm via host factor dependent and independent mechanisms equally well on SS and Ti K-wires. The goal of this study was to develop a rapid in vitro biofilm assay system with quantitative measures of S. aureus growth and biofilm formation. While our model has morphological similarities to both animal models and implants retrieval analyses from patients with S. aureus infection, it is not a substitute for in vivo research. As a major limitation of this system is the absence of host immune mechanisms and other critical biological components of the bone micro-environment, follow up studies to confirm these findings in an appropriate in vivo model are warranted. Significance: We demonstrate a novel in vitro model of S. aureus biofilm formation with quantitative BLI and SEM outcome measures, and used this model to demonstrate strain specific phenotypes and its potential use to evaluate anti-microbial surfaces.

6 Acknowledgments: Financial support has been provided by the AO Trauma Clinical Priority Program on Bone Infection and NIH, NIAMS grant P30 AR References: 1. P. Chen et al, J Microbiol Methods 90, 115 (Aug, 2012). ORS 2014 Annual Meeting Poster No: 1056

Results: Conventional ELISA results demonstrated that sera from the experimentally infected mice contained higher antibody titers vs.

Results: Conventional ELISA results demonstrated that sera from the experimentally infected mice contained higher antibody titers vs. Development Of A Multiplex Serum-based Immunoassay For Early Diagnosis Of Deep Musculoskeletal Infection By Staphylococcus Aureus In A Clinical Pilot Of Infected Patients Vs. Non-infected Controls Kohei