Figures
Abstract
Morphogenic conversion of Candida from a yeast to hyphal morphology plays a pivotal role in the pathogenicity of Candida species. Both Candida albicans and Candida tropicalis, in combination with a variety of different bacterial strains and species, appear in biofilms on silicone-rubber voice prostheses used in laryngectomized patients. Here we study biofilm formation on silicone-rubber by C. albicans or C. tropicalis in combination with different commensal bacterial strains and lactobacillus strains. In addition, hyphal formation in C. albicans and C. tropicalis, as stimulated by Rothia dentocariosa and lactobacilli was evaluated, as clinical studies outlined that these bacterial strains have opposite results on the clinical life-time of silicone-rubber voice prostheses. Biofilms were grown during eight days in a silicone-rubber tube, while passing the biofilms through episodes of nutritional feast and famine. Biofilms consisting of combinations of C. albicans and a bacterial strain comprised significantly less viable organisms than combinations comprising C. tropicalis. High percentages of Candida were found in biofilms grown in combination with lactobacilli. Interestingly, L. casei, with demonstrated favorable effects on the clinical life-time of voice prostheses, reduced the percentage hyphal formation in Candida biofilms as compared with Candida biofilms grown in absence of bacteria or grown in combination with R. dentocariosa, a bacterial strain whose presence is associated with short clinical life-times of voice prostheses.
Citation: van der Mei HC, Buijssen KJDA, van der Laan BFAM, Ovchinnikova E, Geertsema-Doornbusch GI, Atema-Smit J, et al. (2014) Voice Prosthetic Biofilm Formation and Candida Morphogenic Conversions in Absence and Presence of Different Bacterial Strains and Species on Silicone-Rubber. PLoS ONE 9(8): e104508. https://doi.org/10.1371/journal.pone.0104508
Editor: Tom Coenye, Ghent University, Belgium
Received: February 6, 2014; Accepted: July 10, 2014; Published: August 11, 2014
Copyright: © 2014 van der Mei et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This study was entirely funded by the University Medical Center Groningen, Groningen, The Netherlands. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Verbal communication in patients after laryngectomy can be restored (Figure 1) by placing a silicone-rubber voice prosthesis into a surgically created puncture between the trachea and esophagus [1]. By closing the stoma with a finger, air from the lungs can be forced through the silicone-rubber valve to produce tissue vibration and accompanying voice. In the non-sterile environment of the esophagus, valves sides of voice prostheses become rapidly colonized by microorganisms, leading to increased airflow resistance or leakage of food and liquid [2], which results in frequent replacements of prostheses. Candida albicans and Candida tropicalis are regarded as the main fungal species in voice prosthetic biofilms, and seldom exist alone in a biofilm. Next to Candida species, several other bacterial members of the commensal oral and skin flora of the host have been detected in voice prosthetic biofilms [1], [3], mainly comprising streptococci, staphylococci and lactobacilli. Clearly, these commensal have easy access to prosthesis site. Recently it was shown that the diversity (i.e. number of bacterial species detected) of pathogens detected on voice prostheses correlated positively with the diversity of pathogens in the oral cavity [4].
Schematic drawing of voice restoration after laryngectomy using a tracheostoma (left) and SEM of a mixed species (yeast and bacteria) biofilm formation on the tracheostoma valve after insertion for 40 days into a patient (right, bar marker indicates 600 µm). Taken with permission from [7].
An important aspect of Candida pathogenicity [5] is the morphogenic conversion between yeast and filamentous forms of hyphal growth [6]. Hyphae have been associated with the invasive characteristics of Candida species and their growth into silicone-rubber, when exposed to cycles of feast and famine [7]. A mature biofilm on a voice prosthetic surface consists of densely packed Candida and bacteria, embedded within a matrix of exopolysaccharides [8] with a high degree of interspecies interaction [9] and co-adhesion occurring on a spatio-temporal basis. The presence of bacteria is essential for the integrity of voice prosthetic biofilms [10] and like in the oral cavity, metabolic cooperation among the organisms may be of importance to the establishment of stable biofilm communities [11]. Interestingly, combinations of different bacterial strains, like Rothia dentocariosa with C. albicans have been associated with a clinical life-time of voice prostheses lower than average, while combinations with lactobacilli have been suggested to prevail in prostheses with extended clinical life-times [1], [3]. It is currently unknown why certain combinations of bacteria with Candida are more or less harmful than others.
The aim of this study was to investigate early biofilm formation on silicone-rubber by combinations of Candida (C. albicans or C. tropicalis) with different commensal bacterial strains and strains of lactobacilli. In addition, hyphal formation in C. albicans and C. tropicalis, as stimulated by R. dentocariosa or lactobacilli was evaluated, as clinical studies outlined that these bacterial strains have opposite results on the life-time of silicone-rubber voice prostheses in vivo [1], [3]. All strains and species used in this study, except the lactobacilli, were originally isolated from biofilms on explanted voice prostheses.
Materials and Methods
Silicone-rubber
Commercially available silicone-rubber tubes (inner diameter 4.0 mm) were used (Rubber B.V., Hilversum, The Netherlands). Silicone-rubber surfaces were characterized by water contact angles, taken at 25°C using the sessile drop (3 µL) technique and a homemade counter monitor. This monitor registers the contour of a liquid droplet based on grey value thresholding, after which contact angles are calculated from the height and base width of a droplet.
Biofilm Formation
Two yeast strains (C. albicans GBJ 13/4A and C. tropicalis GB 9/9) and seven bacterial strains (Staphylococcus aureus GB 2/1, Staphylococcus epidermidis GB 9/6, Streptococcus salivarius GB 24/9, R. dentocariosa GBJ 52/2B, Lactobacillus casei ATCC 393, Lactobacillus acidophilus ATCC 4356 and Lactobacillus crispatus ATCC 33820) were selected for this study. All strains, except for the lactobacilli, were clinical isolates from biofilms on voice prosthesis. Yeast and bacteria were cultured in a mixture of 30% brain heart infusion broth (OXOID, Basingstoke, UK) and 70% defined yeast medium (per liter: 7.5 g glucose, 3.5 g (NH4)2SO4, 1.5 g L-asparagine, 10 mg L-histidine, 20 mg DL-methionine, 20 mg DL-tryptophane, 1 g KH2PO4, 500 mg MgSO4.7H2O, 500 mg NaCl, 500 mg CaCl2.2H2O, 100 mg yeast extract, 500 µg H3BO3, 400 µg ZnSO4.7H2O, 120 µg Fe(III)Cl3, 200 µg Na2MoO4.2H2O, 100 µg KI, 40 µg CuSO4.5H2O). Each strain was grown overnight at 37°C for 24 h. Subsequently, cultures of a yeast and bacterial strain were mixed in one to one volume ratio and comprised approximately 3×107 yeast per mL and 2×109 bacteria per mL. The resulting dual species suspension was used for biofilm growth (see also Table 1).
Biofilms were grown in silicone-rubber tubes and maintained at a temperature between 36°C and 37°C during an experiment [1]. To grow biofilms, the silicone-rubber tubes were filled for 5 h with the dual species suspension.
After 5 h, adhering bacteria and yeast were allowed to grow into a biofilm on the silicone-rubber tubes during three days, by filling the tubes with growth medium. From day four till day seven, the tubes were perfused three times a day with 250 mL phosphate buffered saline (10 mM potassium phosphate and 150 mM sodium chloride, pH 6.8). Subsequently, the tubes were left in the moist environment. At the end of each day, the tubes were filled with growth medium during 30 min and left overnight in the moist environment of the drained tubes (see Figure 2A). This cycle of feast and famine was performed in order to mimic the pattern of eating and drinking of a patient and found necessary to stimulate ingrowth of hyphae into the silicone rubber (see also Figure 2B).
A. Time sequence of biofilm formation in silicone rubber tubes from mixed cultures of Candida species and a bacterial strain during exposure to feast and famine. B. Scanning electron micrograph of C. tropicalis hyphae penetrating into silicone-rubber after 12 days growth under conditions of feast and famine. Bar markers represent 10 µm and 2.5 µm for the left and right panel, respectively. Taken with permission from [7].
Evaluation of Biofilms: Amount of Biofilm
On day eight of the experiment, the silicone-rubber tubes were removed to assess the number of colony forming yeast and bacteria [1]. To this end, tubes were cut open and biofilms were removed by cotton swabs and sonication in Reduced Transport Fluid (per liter: NaCl 0.9 g, (NH4)2SO4 0.9 g, KH2PO4 0.45 g, Mg2SO4 0.19 g, K2HPO4 0.45 g, EDTA 0.37 g, L-Cysteine HCl 0.2 g, pH 6.8), after which the resulting suspension was serially diluted and plated on MRS (de Man, Rogosa and Sharpe) agar for yeast and blood agar for bacteria. Plates were incubated at 37°C in an aerobic incubator for 3 days prior to enumeration. In case of combinations with lactobacilli, the resulting suspension was plated on MRS plates with incubation in 5% carbon dioxide incubator, while Candida strains were plated on Sabouraud agar plates (incubation at 26°C).
In order to demonstrate the presence of a biofilm in the silicone-rubber tubes, biofilms were imaged using optical coherence tomography (OCT) and confocal laser scanning microscopy (CLSM). OCT allowed in situ imaging of biofilms within the silicone-rubber tubes without additional staining or sectioning using a spectral domain OCT (Ganymade, Thorlabs Inc., Munich, Germany) with an axial resolution of 5.8 µm and a lateral resolution of 8 µm. CLSM images were collected using a Leica TCS-SP2 CSLM (Leica Microsystems Heidelberg GmbH, Heidelberg, Germany) after taking transversal sections from the silicone-rubber tube. Sections were placed on a microscope glass slide and biofilm stained by live/dead stain (Baclight™, Molecular probes Europe BV) for 30 min in the dark, after which images were taken.
Evaluation of Biofilms: Hyphal Induction by Selected Bacterial Strains
Biofilms grown with combinations of Candida and R. dentocariosa or lactobacilli, were further examined immediately after suspending of the biofilm in RTF. The resulting microbial suspension was analyzed using phase-contrast microscopy in order to determine the percentage of fungi in a yeast or hyphal morphology after observing 500 Candida cells in total in one experiment.
Results
Water contact angles on the silicone-rubber tubing amounted 110±1 degrees, attesting to the hydrophobic nature of the material and corresponding with water contact angles on medical-grade silicone-rubber voice prostheses [12].
Examples of OCT and CLSM images of biofilms grown with different combinations of yeast and bacteria are shown in Figure 3. Note that biofilms consisting of a combination of C. tropicalis with L. crispatus are thicker than observed for a combination of C. albicans with R. dentocariosa regardless of the technique applied, although biofilms thicknesses obtained using OCT do not match numerically with those from CLSM. This is most likely related with the fact that tubes with biofilm were sectioned for CLSM, while OCT was done on in situ biofilms. Moreover, CLSM requires staining.
A. In situ cross section of a mixed species biofilm of C. tropicalis combined with L. crispatus by OCT. Average biofilm thickness is 80±24 µm. In situ cross section of a mixed species biofilm of C. albicans combined with R. dentocariosa by OCT. Average biofilm thickness is 26±8 µm. B. A cross section of a mixed species biofilm of C. tropicalis combined with L. crispatus by CLSM after live/dead staining. Average biofilm thickness is 109±29 µm. C. A cross section of a mixed species biofilm of C. albicans combined with R. dentocariosa by CLSM after live/dead staining. Average biofilm thickness is 7± µm. The bars in A and C denote 500 µm, while in B and D bars represent 75 µm.
Table 1 summarizes the total numbers of CFU/cm2 of the mixed species biofilm on the silicone-rubber tubes. In general, biofilms consisting of combinations of C. albicans with a bacterial strain comprised significantly less viable organisms than C. albicans biofilms grown in the absence of bacteria. C. tropicalis on the other hand, showed a higher number of viable organisms when grown in combination with a bacterial strain, except when grown in combination with S. salivarius. The percentage of Candida present in biofilms grown in combination with lactobacilli, was systematically higher than when grown in combination with other commensal bacterial strains. In general, biofilms comprising C. tropicalis harvested higher numbers of viable organisms than when C. albicans was involved, in line with the biofilm thicknesses observed using OCT and CLSM (compare Figure 3 and Table 1).
In C. albicans biofilms however, growth in presence of R. dentocariosa significantly stimulated hyphal formation (see Table 2), while especially L. casei significantly suppressed morphogenic conversion of C. albicans to its hyphal form. In C. tropicalis biofilms, we did not see a significant stimulation by R. dentocariosa on hyphal formation, while L. casei reduced hyphal formation.
Discussion
The clinical life-time of silicone-rubber voice prostheses varies over extremely wide ranges within different laryngectomized patients. Sub-division of 692 failed voice prostheses into a short lifetime group (implantation-period less than 4 months) and an extended lifetime group (implantation-period over 9 months) revealed that R. dentocariosa and C. albicans and C. tropicalis were predominant strains in the short lifetime group. In the extended lifetime group R. dentocariosa was found with a fourfold lower isolation frequency and C. albicans was found with a twofold lower isolation frequency [3]. In a separate study, it was shown that daily consumption of a fermented milk containing L. casei yielded an increased prosthesis lifetime by a factor of 3.76 as compared to a historical control in the same group of patients [13].
In the present study we show that these observations on the clinical life-time of voice prostheses cannot be explained on the basis of numbers of microorganisms found in dual-species biofilms on silicone-rubber, or for that matter: the biofilm thickness. Combinations of lactobacillus strains with Candida yield less biofilm, but with an increased prevalence of Candida as compared with combinations involving other bacterial strains. At the same time, we show that L. casei, with demonstrated favorable effects on the clinical life-time of voice prostheses [13] and adhering in full grown clinical biofilms in close association with yeast (see Figure 4), reduced the percentage hyphal formation in Candida biofilms as compared with Candida biofilms grown in absence of bacteria by a factor of three. When compared with the percentage hyphal formation in Candida biofilms grown in combination with R. dentocariosa, a bacterial strain whose presence is associated with short clinical life-times of voice prostheses [3], this factor becomes ten. Therewith we here demonstrate that in voice prosthetic biofilms, morphogenic conversions in Candida biofilms depend on the bacterial strain present, corresponding with effects on the clinical life-time of the prostheses.
Overlay-images of a biofilm from an explanted voice prosthesis (life time 318 days) hybridized with the FITC-labelled EUK516 probe indicating all yeasts (with hyphae) and Cy3-labelled Lab158 probe illustrating presence of lactobacilli. The high magnification panel on the right clearly shows the association of lactobacilli (red) and yeasts (green). Bars equal 20 µm and 5 µm for the left and right panels, respectively. Taken with permission from [14].
The percentage hyphae in Candida biofilms grown in absence of bacteria amounts 0.7% to 3.0% (see Table 2). Other studies, in which Candida biofilms were grown in serum demonstrate much higher percentages of hyphae up to 100%, which is due to the hyphal transforming activity in serum associated with the signaling molecules prostaglandin and thromboxane [15]. Also the periods of nutritional feast and famine applied in the current study to mimic the conditions in the oropharynx more closely, may have a reducing effect on morphogenic conversions in Candida. More importantly, interactions between Candida and lactobacilli have been described to regulate Candida morphogenesis and therewith its virulence and invasiveness. Short chain fatty acids produced by different strains of lactobacilli (L. casei, Lactobacillus paracasei and Lactobacillus rhamnosus) inhibited hyphal formation, as well as culture supernatants of lactobacilli and live lactobacilli [15]. In the present study, we only find a significant reduction in C. albicans hyphal formation for L. casei with respect to Candida growth in absence of bacteria, while for C. tropicalis reductions observed are not statistically significant. For C. albicans however, the percentage hyphal formation when grown in presence of L. casei is tenfold lower (statistically significant at p<0.05) than when grown in presence of R. dentocariosa. A similar comparison exists for C. tropicalis but is not statistically significant.
Thus we conclude that R. dentocariosa stimulates morphogenic conversion in Candida toward a hyphal morphology, coinciding with short clinical life-times of voice prostheses from which R. dentocariosa was isolated [3]. In addition it is concluded that especially L. casei reduces hyphal formation in Candida, which coincides with clinical observations that the consumption of a fermented milk with L. casei extends the clinical life-time of voice prostheses in laryngectomized patients [13]. These conclusions are based on relatively small differences in hyphal formation. However, these differences were established over a time period of only eight days, whereas the above clinical observations were made over a period of several months to years. Likely, larger effects will develop over time. Moreover, a small number of hyphae may be sufficient to grow into the silicone-rubber and the close proximity of Candida and bacteria in pockets created in silicone-rubber [16] will enhance the effects of bacterial presence on morphogenic conversion of Candida. Since hyphal ingrowth is a major factor in the strong binding of initial Candida biofilms to silicone-rubber [16], [17], biofilm maturation after initial adhesion does not necessarily have to involve Candida in its hyphal morphology. This is in line with the high percentages of Candida in biofilms grown in the presence of lactobacilli, despite low conversions to the hyphal morphology (compare Tables 1 and 2).
Author Contributions
Conceived and designed the experiments: HM KB BL EO HB. Performed the experiments: KB EO BBG GGD JAS. Analyzed the data: HM KB EO GGD JAS BBG BL HB. Wrote the paper: HM KB BL HB. Editing of final paper: HM KB EO GGD JAS BBG BL HB.
References
- 1.
Buijssen KJDA (2012) Biofilms on silicone rubber voice prostheses – Importance of lactobacilli and surface modification. Thesis, Groningen, The Netherlands.
- 2. Mahieu HF, Van Saene HKF, Rosingh HJ, Schutte HK (1986) Candida vegetations on silicone voice prostheses. Arch Otolaryngol Head Neck Surg 112: 321–325.
- 3. Elving GJ, Van der Mei HC, Van Weissenbruch R, Busscher HJ, Albers FWJ (2002) Comparison of the microbial composition of voice prosthesis biofilms from patients requiring frequent versus infrequent replacement. Ann Otol Rhinol Laryngol 111: 200–203.
- 4. Bertl K, Zatorska B, Leonhard M, Rechenmacher-Strauss J, Roesner I, et al. (2013) Oral microbial colonization in laryngectomized patients as a possible cofactor of biofilm formation on their voice prostheses. J Clin Periodontol 40: 833–840.
- 5. Odda FC (1994) Pathogenesis of Candida infections. J Am Acas Dermatol 31: S2–5.
- 6. Gow NA, Brown AJ, Odds FC (2002) Fungal morphogenesis and host invasion. Curr Opin Microbiol 5: 366–371.
- 7. Busscher HJ, De Boer CE, Verkerke GJ, Kalicharan R, Schutte HK, et al. (1994) In vitro ingrowth of yeasts into medical grade silicone-rubber. Int Biodeter Biodegrad 33: 383–390.
- 8. Baillie GS, Douglas LJ (1999) Role of dimorphism in the development of Candida albicans biofilms. J Med Micrbiol 48: 671–679.
- 9. El Azizi MA, Starks SE, Khardori N (2004) Interactions of Candida albicans with other Candida spp. and bacteria in the biofilms. J Appl Microbiol 96: 1067–1073.
- 10.
Parker AJ, Ell SR, Davies CM, Clegg RT (1995) Do bacteria play a role in silastic speaking valve failure? In: Algaba J, editor. Proceedings of the 6th international congress on surgical and prosthetic voice restoration after total laryngectomy San Sebastiaan. pp. 83.
- 11. Bradshaw DJ, Homer KA, Marsh PD, Beighton D (1994) Metabolic cooperation in oral microbial communities during growth on mucin. Microbiology 140: 3407–3412.
- 12. Everaert EPJM, Mahieu HF, Wong Chung RP, Verkerke GJ, Van der Mei HC, et al. (1997) A new method for in vivo evaluation of biofilms on surface-modified silicone rubber voice prostheses. Eur Arch Otorhinolaryngol 254: 261–263.
- 13. Schwandt LQ, Van Weissenbruch R, Van der Mei HC, Busscher HJ, Albers FWJ (2005) Effect of dairy products on the lifetime of Provox2 voice prostheses in vitro and in vivo. Head Neck 27: 471–477.
- 14. Buijssen KJDA, Harmsen HJM, Van der Mei HC, Busscher HJ, Van der Laan BFAM (2007) Lactobacilli: important in biofilm formation on voice prostheses. Otolaryngol Head Neck Surg137: 505–507.
- 15. Noverr MC, Huffnagle GB (2004) Regulation of Candida albicans morphogenesis by fatty acid metabolites. Infect Immun 72: 6206–6210.
- 16. Neu TR, Van der Mei HC, Busscher HJ, Dijk F, Verkerke GJ (1993) Biodeterioration of medical-grade silicone rubber used for voice prostheses: a SEM study. Biomaterials 14: 459–464.
- 17. Ten Cate JM, Klis FM, Pereira-Cenci T, Crielaard W, De Groot PWJ (2009) Molecular and cellular mechanisms that lead to candida biofilm formation. J Dent Res 88: 105–115.