Evaluation of combined growth media for in vitro cultivation of oropharyngeal biofilms on prosthetic silicone
In the upper aerodigestive tract, biofilm deposits by oropharyngeal microbes can cause failure of medical polymer devices like voice prostheses. Previous studies on testing of inhibitive strategies still lack of comparability due to varying study protocols concerning growth media, microbial species and growth conditions. Goal of the study was therefore to test cultivation of a mixed biofilm of isolated oropharyngeal microbes under in vitro growth conditions using mixtures of common growth media. Mixtures of yeast peptone dextrose medium (YPD), fetal bovine serum (FBS), RPMI 1640, Yeast nitrogen base medium (YNB) and brain heart infusion (BHI) were tested to grow mixed biofilm deposits of Candida albicans, Candida tropicalis, Staphylococcus aureus, Streptococcus epidermidis, Rothia dentocariosa and Lactobacillus gasseri on medical grade silicone. Periodic assessment of living biofilm was performed over 22 days by a digital microscope and the cultivated biofilm structures were analyzed by scanning electron microscopy after completion of the study. Mixtures of BHI, YPD and FBS improved microscopic growth of multispecies biofilm deposits over time, while addition of RPMI and YNB resulted in reduction of visible biofilm deposit sizes. A mixtures of FBS 30% + YPD 70% and BHI 30% + YPD 70% showed enhanced support of permanent surface growth on silicone. Growth kinetics of in vitro multispecies biofilms can be manipulated by using mixtures of common growth media. Using mixtures of growth media can improve growth of longterm multispecies oropharyngeal biofilm models used for in vitro testing of antibiofilm materials or coatings for voice prostheses.
2 Materials and methods
2.1 Preparation of microbial strains
Microbial strains of C. albicans, C. tropicalis, S. aureus, S. epidermis and S. salivarius originated from a collection of explanted dysfunctional voice prostheses of laryngectomized patients visiting the Department of Phoniatrics-Logopedics of the Medical University Hospital. VPs with macroscopic biofilm infestation were withdrawn and sonicated in phosphate buffered saline solution (PBS, Morphisto, Frankfurt am Main, Germany) for 10 min to remove lose biofilm debris and vortexed in 5 ml PBS for 3 min before the specimen were isolated and identified on agar plates using standard microbiological methods. Bacterial strains for composition of the in-vitro biofilm were selected based on their frequent appearance on explanted prostheses and on reports in literature. Isolated R. dentocariosa and L. gasseri strains were provided from a collection by the Department of Microbiology of the Medical University Hospital. All specimens were stored at −80 °C and thawed before further use.
2.2 Preparation of silicone material samples
Platelets of 8 mm diameter and 1 mm thickness were punched out of blue colored medical grade silicone sheets (Websinger, Wolkersdorf, Austria) and a segment was cut off to mark the bottom side of each platelet. The platelets were mounted on surgical steel tips for incubation in a vertical position to avoid settlement of planktonic cells by gravity. The prepared samples were autoclaved for 20 min at 125 °C and placed sterile in well titer plates (CellStar Greiner bio-one, Kremsmünster, Austria). In each of the following growth media, mixed biofilms were grown on 12 platelets.
2.3 Preparation of growth media
YPD (Yeast extract peptone dextrose: yeast extract 1% (Sigma-Aldrich Life Science, St. Louis, USA), glucose 2% (Merck KGaA, Darmstadt, Germany), peptone water 2% (Oxoid LTD, Hamshire, England)) and FBS (Fetal Bovine Serum: (Gibco, Life Technologies Carlsbad, California, USA)) were used as control growth media and to prepare the following mixtures:
BHI 30% (Sigma-Aldrich life science, St. Louis, USA) + YPD 70%
FBS 30% + YPD 70%
FBS 30% + RPMI 70% medium + 2% glucose (RPMI 1640: 20,8 g RPMI-1640 (Sigma-Aldrich life science, St. Louis, USA), 69,06 g MOPS (Sigma-Aldrich life science, St. Louis, USA) 36 g glucose (Merck KGaA, Darmstadt, Germany)
FBS 30% + YNB 70% (Yeast nitrogen base 0,67% with ammonium sulphate without dextrose or amino acids (Sigma-Aldrich life science, St. Louis, USA), glucose 2% (Merck KGaA, Darmstadt, Germany)
YPD 50% + RPMI 1640 50%
2.4 Preparation of inoculum
The frozen candida strains were inoculated with sterile loops, plated out on Sabouraud-Dextrose agar (Becton Dickinson, New Jersey, USA) and Columbia 5% sheep blood agar (bioMerieux SA, Marcy l’Etoile, France) and incubated at 37 °C for 24 h. Single colonies of each candida species were picked from the Sabouraud-Dextrose agar, inoculated in each 20 ml of YPD and then incubated on an orbital shaker at 100 rpm for 24 h. The overnight candida cultures were centrifuged for 5 min and the supernatants discarded. The remaining cells were washed three times with PBS. The washed planktonic candida cells were used to prepare 1.0 McFarland standard (equaling 10 × 10–6 cfu/ml), inoculates in distilled water, which were then mixed into one microbial suspension. S. aureus, S. epidermidis and S. salivarius were suspended in PBS to a cell density of 10 × 6 cfu/ml and then added to the C. albicans suspension. One milliliter of each suspension was mixed into the final suspension and then a 10 fold dilution was performed.
2.5 Biofilm growth
All silicon platelets were pre-coated with bovine serum for 24 h at 37 °C to improve initial microbial adhesion capability. Then one milliliter of the prepared microbial suspension was added to 9 ml of each growth medium. The growth media were then added to the silicone platelets in the micro titer plates. After 24 h of incubation at 37 °C at 150 rpm, the growth media were removed from the well plates and replenished with fresh growth media and planktonic cells. This reseeding procedure was performed daily for 22 days.
2.6 Biofilm analysis
2.7 Statistical analysis
Statistical calculation of the growth of biofilm deposits was performed using IBM SPSS Statistics (version 23, IBM Corporation, New York, United States) and graphs were illustrated using Graphpad Prism software (Version 7.0a, GraphPad Software, Inc., La Jolla, CA 92037 USA). The mean value of 12 platelets was calculated every two days for each growth medium. Percentages of the covered areas were arcsine transformed in order to remove the correlation between mean and standard deviation. Transformed data were analyzed by a general estimation equation model with an also autoregressive correlation structure and material as a group factor and day of measurement as within group factor. Comparison between the growth media were done by Bonferroni sequential tests. For all comparisons, a p-value < 0.05 was chosen as significance level.
3.1 Microscopic biofilm growth
3.2 Microscopic biofilm morphologies
3.3 Distribution of macroscopic biofilm deposits
Using the mapping function of the image analysis software, the areas of more permanent biofilm aggregation (red colors) can be distinguished from areas with less biofilm aggregation (green colors) over time on each platelet (see electronic supplementary material 1). It shows that the top edges of the platelets are areas of frequent and permanent colonization, which can be explained by the physical effect, that particles/cells tend to agglomerate near the surface of agitated fluids. The largest areas of permanent biofilm colonization were detected in FBS 30% + YPD 70%, BHI 30% + YPD 70%, FBS and YPD.
Multispecies biofilm formation by bacteria and fungi is a clinical problem of indwelling medical devices, if used in non sterile body compartments. They need to be replaced frequently due to biofilm induced loss of function [17, 18]. Silicone is the material of choice for modern standard VPs due to its biocompatibility and flexibility, which allows folding and atraumatic replacement . Since microbial colonization of the silicone material in VPs cannot be avoided due to the exposed location in the esophagus, protective strategies such as resistant materials, inhibitive coatings or repellent surfaces are needed to improve the device lifetime of standard VPs. Initial evaluation of such strategies using rapid and cost-effective in-vitro biofilm models is an essential step, before producing prosthesis prototypes, that meet the quality requirements of implantable products and can be studied in-vivo.
Biofilms on VPs typically comprise several cross-kingdom interacting communities of bacteria and fungi that evolve into valve blocking deposits and degrade/infiltrate the silicone after weeks (Fig. 1b) . The applied microbial composition has been selected from explanted voice prostheses of patients with a history of repeatedly short device lifetimes due to biofilm associated valve leakage or material infiltration and is consistent with existing literature on in-vivo colonization of VPs [3, 4]. To remain as close as possible to in-vivo conditions, microbial isolates from patients, who have repeatedly presented excessive biofilm infestation and aggressively infiltrated voice prostheses in-vivo, were used in this study. Pre-evaluation showed no difference in biofilm formation in-vitro between ATCC strains and the used strains. While strong biofilm formation is categorized in ATCC strains, to our knowledge, the ability of silicone material infiltration is not, but it is important for future testing of microbial degradation of prosthetic materials. Active interactions between C. albicans, oral streptococci and S. aureus have been described, though it is not fully understood, which co-factors lead to synergistic or antagonistic coexistence inside a biofilm [21, 22, 23]. R. dentocariosa and lactobacilli are discussed to play key roles in co-adhesion and transition to hypheal growth of C. albicans, and seem to be associated with early device failure [24, 25]. Using mixtures of standard growth media to support the microbial growth of all these species is plausible, but only limited data on in-vitro cultivation of oropharyngeal biofilms on silicone or VPs exists. A mixture of 30% BHI and 70% of a defined yeast medium has been used in a Modified Robins Device to illustrate the protective effect of probiotics, antibiofilm coatings or effects of co-incubation of candida with bacterial strains [10, 26, 27]. Wannemuehler et. al. investigated a vibratory stimulus on biofilms on VPs using a 1:1 mixture of BHI and YPD. In a previous study by our group, a two species biofilm of S. salivarius and C. albicans was incubated in RPMI 1640 on medical grade silicone for 140 days . However, such variant incubation protocols are a known problem for comparing the results of in-vitro simulations and the efficacy of antibiofilm measures . Simulation of complex multispecies formulations, such as oropharyngeal biofilms, in in-vitro models should be preceded by a pre-evaluation of growth media to optimally support growth of stable biofilm deposits, structures and cell morphologies and to achieve similar to in-vivo findings including a balanced proliferation of all involved species over a period of time that is representative for the clinical application of a particular implant or prosthesis. In the present model, initial evaluation of commonly applied single growth media, that have been recommended for the specific species or have been used in similar models, confirmed YPD and FBS as eligible in regard to biofilm mass, stability and cell growth forms, but differences in longterm performance and support of all species between the tested growth media have been noticed (data not published). Goal of this study was therefore to investigate, if combinations of these growth media could be used to further optimize longterm biofilm growth of oropharyngeal biofilm compositions on silicone.
The results show, that YPD, FBS and mixtures of FBS 30% + YPD 70% or BHI 30% + YPD 70% are eligible to generate long lasting oropharyngeal-like biofilm compositions on medical grade silicone for testing purposes. The mixture of YPD and FBS shows a synergy of accelerated onset of growth with the maximum achieved macroscopic biofilm cover. Notably for this longterm biofilm model, addition of RPMI and YNB to a mixture produced significantly less macroscopically visible biofilm mass. Further, the addition of RPMI or YNB to FBS showed no growth support of rod shaped lactobacilli in SEM analysis, whereas they were frequently identified in the other mixtures, including YPD 50% + RPMI 50%. Analysis of key morphological structures showed that all growth media except FBS 30% + YNB 70% produced solid macroscopic multispecies biofilms that are similar to findings on dysfunctional voice prostheses. In FBS 30% + YNB 70%, hypheal growth appeared to be reduced and an increased mass of bacteria could be identified. However, the platelets displayed abundant local morphological heterogeneity, which can be interpreted as a previously described organization in functional micro-consortia .
Growth media have impact on in-vitro formation of biofilms and need pre-evaluation to be adjusted for specific polymicrobial compositions. Mixtures of FBS 30% + YPD 70% and BHI 30% + YPD 70% proved to produce stable oropharyngeal-like biofilm deposits over weeks, and show similar microscopic cell morphologies to in-vivo findings of VPs. Application of these growth media mixtures in the presented biofilm model can be used to screen novel polymer materials, coatings or surface modifications that are intended to inhibit or slow biofilm formation on future VP designs.
The authors thank Prof. M. Kundi of the Department of Environmental Hygiene of the Medical University Vienna for supervision of the applied statistical analysis. This article does not contain any studies with human participants or animals performed by any of the authors.
Open access funding provided by Medical University of Vienna.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
- 8.Leunisse C, Van Weissenbruch R, Busscher HJ, van der Mei HC, Albers FW. The artificial throat: a new method for standardization of in vitro experiments with tracheo-oesophageal voice prostheses. Acta Otolaryngol. 1999;119:604–8.Google Scholar
- 27.van der Mei HC, KJDA Buijssen, BFAM Laanvan der, Ovchinnikova E, Geertsema-Doornbusch GI, Atema-Smit J, van de Belt-Gritter B, Busscher HJ. Voice prosthetic biofilm formation and Candida morphogenic conversions in absence and presence of different bacterial strains and species on silicone-rubber. PLoS ONE. 2014;9:e104508.Google Scholar
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.