We assessed the infection resistance of different implant materials after confrontation with E. faecalis by comparing them using various analyses. The materials were kindly provided by Biotronik (Berlin). Two thermoplastic polyurethanes (TPU) and liquid silicone rubber (LSR) were used, either with or without modification. Further details on the preparation are described in our previous work (Woitschach et al., 2021a; Woitschach et al., 2021b). After modification, all samples were subjected to a sterilization process with ethylene oxide (ETO). Thermanox, a well-established polyester, was used as a reference. All materials were round and had a diameter of approximately 16 mm. The following abbreviations were used: Thermanox = tmx, Thermoplastic polyurethane 55 = P55, Thermoplastic polyurethane 80 = P80, unmodified Liquid silicone rubber = LSR, LSR + Polymethylmethacrylate-2-methacryloyloxyethyl phosphorylcholine = PMMA-MPC, LSR + Polysulfone = PSU and LSR + Poly (1,4-phenylene-ehter-ether-sulfone) = PPSP.

Snap frozen Enterococcus faecalis (ATCC29212) was purchased from the American Type Culture Collection (ATCC). E. faecalis was grown in brain heart infusion broth (BHI) supplemented with 1% glucose (Kwasny and Opperman, 2010). For all experiments an inoculum with OD₆₀₀ of 0.1 (optical density measured with wavelength 600 nm) was prepared by adjusting the concentration of an overnight bacterial broth culture in BHI medium, this corresponds to 1 x 10⁸ CFU/ml (colony forming unit/ml). All experiments were carried out aerobically at 37°C in 24-well polystyrene plates (Nunc, Thermofisher scientific, Germany) with a culture volume of 2 ml unless otherwise stated and were performed in three independent replicates.

Bacterial viability was analyzed by generating growth curves using OD measurement and performing LIVE/DEAD fluorescent staining. In brief, 2 ml of the bacterial culture (1 x 10⁷ CFU/ml) were added to a 24-well plate containing the different rounded material disks and incubated for 6 h. Bacterial growth was monitored hourly by measuring OD₆₀₀ using a microplate reader (FLUOstar Omega, BMG Labtech). In addition, bacterial viability was determined after 6 and 24 h using the two-color fluorescence assay LIVE/DEAD™ BacLight™ Bacterial Viability Kit (Thermo Fischer, Waltham, Massachusetts, USA) according to the manufacturer’s instructions. The fluorescence was visualized using a fluorescence microscope (Axio Scope.A1, Zeiss, Germany) equipped with a camera (AxioCam MRc, Zeiss, Jena, Germany). The fluorescence intensity was determined using a microplate reader (FLUOstar Omega, BMG Labtech) to subsequently calculate the percentage of living cells using the adjusted dye ratio equation as follows: % living cells = (100 x SYTO^®/PI)/(1+ SYTO^®/PI) in accordance to Ou et al. (2016.). SYTO^® 9 green-fluorescent nucleic acid (SYTO^®) stain labels all bacteria in a population (live) and the red-fluorescent nucleic acid stain, propidium iodide (PI), is taken up by bacteria with damaged membranes (dead) only.

The disc diffusion test was used to determine the inhibition of bacterial growth by substances released by the materials. The agar plates were inoculated with 1 x 10⁸ CFU/ml of the bacterial culture. The different material discs were then placed in the middle of the inoculated agar surfaces. The plates were incubated overnight at 37°C and the diameters of the inhibition zones around the discs were measured.

Using the MATH/BATH test (Microbial Adhesion to Hydrocarbons/Bacterial Adherence to Hydrocarbons), the hydrophobicity of the bacterial cell surface was assessed according to Lather et al. (2016) and Di Ciccio et al. (2015), with minor modifications. For this purpose, the bacteria were centrifuged at 3000 g for 5 min. The pellet was washed three times in ice-cold phosphate buffer and resuspended in a specific amount of phosphate buffer to achieve an OD₅₀₀ of 0.5. The bacterial suspension was overlaid with 0.5 mL of xylene (Sigma Aldrich, Darmstadt, Germany). After shaking for 1 min, the phases were incubated at room temperature for 15 min for separation. Results were expressed as the percentage of cells excluded from the aqueous phase as determined by the following equation: % adherence = [(1 -A/A0)] × 100, where A0 and A are the initial and final optical densities of the aqueous phase, respectively. The classification was made as follows: >50% = strongly hydrophobic; 20 - 50% = moderately hydrophobic; <20% = hydrophilic. The measurements were performed twice with three technical replicates.

Crystal violet staining was used to quantify biofilm formation on the materials, as previously described (Kwasny and Opperman, 2010). Briefly, 2 ml of the bacterial suspension (1 x 10⁸ CFU/ml) was incubated on the materials for 24 h. The supernatant was removed, and the discs were washed three times with double-distilled water to remove bacteria that were not or were only loosely attached. The discs were incubated for 1 h at 60°C for heat fixation. 200 µl of 0.06% crystal violet was added to each well and incubated for 5 min. The crystal violet was then removed and the slices were washed three times with double distilled water. Stained biofilms on the discs were eluted with 200 µl of 30% acetic acid. The supernatants were measured at 600 nm in a multilabel microtitre plate (FLUOstar Omega, BMG Labtech).

Isolation and purification of mRNA was carried out using the Fungal/Bacterial RNA Miniprep Kit (Zymo Research, California, USA). The mRNA was extracted from E. faecalis biofilms after 6 and 24 h and quantified by spectrophotometric measurement (Colibri, Berthold Technologies, Bad Wildbad, Germany). An amount of 500 ng of total RNA was transcribed using High-Capacity cDNA Reverse Transcription Kit (ThermoFisher, Dreieich, Germany), according to the manufacturer’s instructions. Gene expression levels were determined using two-step quantitative reverse transcription PCR with 50 ng of cDNA in a reaction volume of 10 µl using TaqMan Universal MasterMix II (ThermoFisher, Dreieich, Germany). The primers used in this study were selected with regard to their involvement in biofilm formation: Gelatinase (GelE) F: 5’-GGTGACCCCGTATCATTGGT-3’ R: 5’-CCGTGTAAAGCAATTCCCGT-3’ Internal Oligo: 5’-ACGGAACATGTCGGTAGTGA-3’; vWF-binding protein (EbpA) F: 5’-CGCCAAGTAGCACTAGAGGA-3’ R: 5’-GTCCCGTAAAAGCGACCATC-3’ Internal Oligo: 5’-GTCGAGTTCAAACAGAGGCG-3’; LPXTG-anchored aggregation substance (AS) F: 5’-AGTCAAAATGGCGCCACAAT-3’ R: 5’-CAAACACAGACCATTGGCCA-3’ Internal Oligo: 5’-TACGCTGGTGTTGTGGAAGA-3’; Enterococcus faecalis antigen A (EfaA) F: 5’-TAGAAACAGGCGGAAATGGC-3’ R: 5’-AACCAAGCATGCGGATCTTC-3’ Internal Oligo: 5’-CAAGTGCCGGTCAAGAACAA-3’. 16S was used as an endogenous control F: 5’-GGGAATCTTCGGCAATGGAC-3’ R: 5’-CGCCCAATAAATCCGGACAA-3’ Internal Oligo: 5’-GAAAGTCTGACCGAGCAACG-3’. Primers and probes were purchased from Eurofins Genomics, Germany. Relative fold change in gene expression was determined using the 2^−ΔΔct method. Data were presented as the fold-change to tmx normalized to the reference gene expression level of 16S.

Statistical analysis was performed using GraphPad Prism 5.0 (GraphPad Software, La Jolla, CA, USA). Values are expressed as mean + SEM. Normal distribution was tested using the D’Agostino and Pearson Omnibus Normality Test. Non-normally distributed samples were compared using the Kruskal–Wallis test followed by a Dunn’s post hoc test. For all statistical analyses, p values < 0.05 were considered significant. *p < 0.05, **p < 0.01, ***p < 0.001, n.s. = not significant. Tmx served as control material.

The growth curves, which represent the density of cell populations in liquid culture over time, showed a comparable bacterial growth on all materials tested, represented by an increase in the first 5 h up to an OD of approximately 1.0, followed by a slight decrease ( Figure 1A ). To investigate whether inhibitory substances are released by the materials, we performed a disc diffusion test. The absence of inhibition zones indicated the non-toxicity of the materials ( Figure 1B ).

The ability of the materials to affect the viability of E. faecalis was analyzed using the LIVE/DEAD™ BacLight™ Bacterial Viability Kit. E. faecalis was seeded for 6 and 24 h on the different materials. The percentage of living bacteria was calculated as mentioned above. We obtained more than 95% living cells for at both time points with no discernible differences between the various materials ( Figures 2A ). In addition, the fluorescence microscope images of E. faecalis showed that the surface of the discs was covered by nearly the same amounts of adhered living bacteria (green) and dead bacteria (red) on all materials ( Figures 2B–H ).

Bacterial biofilm formation was quantified by crystal violet staining. After 6 and 24 h no pronounced differences were observed in terms of biofilm formation on the tested materials, as evidenced by the relatively similar intensity of staining, except for a slight decrease on P80 at both time points, as well as for P55 after 24 h ( Figures 3A, B ).

Also, we analyzed biofilm formation relevant gene expression of GelE, ebpA, efaA and AS after 6 and 24 h. We found a tendency towards lower expression levels for GelE after 6 h for all materials except of PMMA-MPC. This effect disappeared at later time points and, in terms of PMMA-MPC, even opposite gene expression level was reached ( Figure 4A ). In addition, for GelE significant differences between P55 and LSR, with an approximate reduction by half after 6 h ( Figure 4A ) and for ebpA between P55 and PSU ( Figure 4B ) were determined.

For further characterization of E. faecalis, we analyzed the hydrophobicity of the bacterial cell surface by the MATH/BATH (Microbial Adhesion to Hydrocarbons/Bacterial Adherence to Hydrocarbons) assay. Analysis of the hydrophobic properties of the cell surface of E. faecalis revealed with 39.74 ± 14.73% only moderate hydrophobicity.