Role of glucosyltransferase R in biofilm interactions between Streptococcus oralis and Candida albicans

Abstract

Streptococcal glucosyltransferases (Gtf) synthesize α-glucan exopolymers which contribute to biofilm matrix. Streptococcus oralis interacts with the opportunistic pathogen Candida albicans to form hypervirulent biofilms. S. oralis 34 has a single gtf gene (gtfR). However, the role of gtfR in single and mixed species biofilms with C. albicans has never been examined. A gtfR deletion mutant, purified GtfR, and recombinant GtfR glucan-binding domain were tested in single and mixed biofilms on different substrata in vitro. A mouse oral infection model was also used. We found that in single species biofilms growing with sucrose on abiotic surfaces S. oralis gtfR increased biofilm matrix, but not bacterial biomass. In biofilms with C. albicans, S. oralis encoding gtfR showed increased bacterial biomass on all surfaces. C. albicans had a positive effect on α-glucan synthesis, and α-glucans increased C. albicans accretion on abiotic surfaces. In single and mixed infection of mice receiving sucrose S. oralis gtfR enhanced mucosal burdens. However, sucrose had a negative impact on C. albicans burdens and reduced S. oralis burdens in co-infected mice. Our data provide new insights on the GtfR-mediated interactions between the two organisms and the influence of biofilm substratum and the mucosal environment on these interactions.

Fig. 1. S. oralis biofilms growing for 24 or 48 h on polystyrene.

Biofilms of wild-type (So34), gtfR mutant (So ∆gtfR) or complemented (So pgtfR) strains were grown in RPMI, 10%FBS, 10% BHI media supplemented with 1% sucrose or 1% glucose. a X–Y isosurfaces (top panel) and three-dimensional reconstructions (bottom panel) of representative confocal laser scanning microscopy images of biofilms. S. oralis (blue) was visualized after fluorescence in situ hybridization with a Streptococcus-specific probe conjugated to Alexa 405. Alexa Fluor 647-labeled dextran conjugate probe (red) was used to stain biofilm matrix (glucans). Scale bars, 50 µm (X–Y isosurfaces) and 70 µm (three-dimensional reconstructions). b Average total biovolumes (in µm³) for 24 and 48 h biofilms exposed to glucose (white bars) or sucrose (black bars) shown in (a) above. Biovolumes were measured in two different confocal laser scanning microscopy image stacks from two independent experiments. c Average biofilm thickness (in µm) in biofilms growing with 1%sucrose for 24 and 48 h. d Average S. oralis colony-forming units (CFU) in logarithmic scale for each 24 and 48 h biofilm. e Average matrix (α-glucan) biovolumes (in µm³) for 24 and 48 h biofilms. *p < 0.05, using the Bonferroni t-test. The error bars indicate standard deviations.

Fig. 2. Twenty-four and forty-eight hour biofilms of C. albicans (Ca) alone or with wild-type S. oralis (So34), gtfR mutant (So ∆gtfR) or complemented (So pgtfR) strains.

Biofilms were grown on polystyrene surfaces in RPMI, 10% FBS, 10% BHI media supplemented with 1% sucrose. a X–Y isosurfaces (top panel) and three-dimensional reconstructions (bottom panel) of representative confocal laser scanning microscopy images of biofilms. Please note overlap of green and red signals shown in yellow, suggesting close physical proximity between the two organisms. Scale bars 50 μm. b Average total biofilm biovolume (in µm³) for 24 and 48 h biofilms. c S. oralis CFU counts expressed as fold of mixed over single biofilms. d Average matrix biovolumes (in µm³). e Relative expression of gtfR gene levels assessed by RT-qPCR. Results represent mean fold change gene expression in C. albicans with S. oralis (CaSo, black bars) over S. oralis (So, white bars) biofilms alone, in three independent experiments. f C. albicans CFU counts expressed as fold of mixed biofilms with each of the three S. oralis strains (CaSo) over single biofilms (Ca). *p < 0.05, using the Bonferroni t-test. The error bars indicate standard deviations.

Fig. 3. Biofilms growing on titanium surfaces for 24 h with 1% sucrose.

S. oralis (So) (wild-type So34, ∆gtfR mutant, or complemented pgtfR strains) and C. albicans (Ca) growing alone or in combination. a X–Y isosurfaces (top panel) and three-dimensional reconstructions (bottom panel) of representative confocal laser scanning microscopy images of biofilms. Organisms and α-glucan-rich matrix were visualized by staining as above. Scale bars, 50 µm (X–Y isosurfaces) and 70 µm (three-dimensional reconstructions). b Average total, Candida or bacterial biovolumes (in µm³). Biovolumes were measured in two different confocal laser scanning microscopy image stacks from two independent experiments. c S. oralis CFU counts shown as mean fold of mixed biofilms over single biofilms in three experiments. d Candida CFU counts shown as fold of mixed biofilms over single biofilms. e Average matrix (α-glucans) biovolumes (in µm³) on S. oralis (WT) alone biofilms and C. albicans–streptococci mixed species biofilm. f Relative expression levels of gtfR gene in S. oralis strain 34 were analyzed by RT-qPCR. Results represent mean fold change gene expression in C. albicans with S. oralis (CaSo) over S. oralis (So) alone biofilms in independent experiments. *p < 0.05, using the Bonferroni t-test. The error bars indicate standard deviations.

Fig. 4. Biofilms of C. albicans alone or in combination with S. oralis (WT and ∆gtfR strains) growing on organotypic mucosal surfaces for 6 or 24 h.

C. albicans (green) was visualized after staining with an FITC-conjugated anti-Candida antibody. S. oralis (red) was visualized after fluorescence in situ hybridization with a Streptococcus-specific probe conjugated to Alexa 546. a Tissue sections of mucosal biofilms with organisms stained as above, and mucosal cell nuclei counterstained with the nucleic acid stain Hoechst 33258 (blue, top panel). Corresponding haematoxylin and eosin-stained tissue sections are shown in the bottom panels. Scale bars 20 μm. b C. albicans (black bars) and S. oralis 34 (white bars) CFU counts expressed as fold of Candida and wild-type S. oralis mixed biofilms (CaSo34) over Candida and ΔgtfR mutant mixed biofilms (CaSoΔgtfR). c X–Y isosurfaces of representative confocal laser scanning microscopy images of mixed 24 h biofilms (green, Candida, blue, S. oralis) showing α-glucans (biofilm matrix) stained with Alexa Fluor 647-labeled dextran conjugate probe (red). Scale bars 50 μm. *p < 0.05, using the Bonferroni t-test. The error bars indicate standard deviation.

Fig. 5. C. albicans interactions with preformed S. oralis biofilms.

S. oralis biofilms were grown for 24 h using wild-type (So34) or ∆gtfR strains; media were supplemented with 1% sucrose or no carbohydrate. C. albicans was then added and incubated for 1 or 16 h. Unattached cells were washed and biofilms were stained. a X–Y isosurfaces (top panel) and three-dimensional reconstructions (bottom panel) of representative confocal laser scanning microscopy images of biofilms. C. albicans (green) was visualized after staining with an FITC-conjugated anti-Candida antibody. S. oralis (blue) was visualized after fluorescence in situ hybridization with a Streptococcus-specific probe conjugated to Alexa 405. Alexa Fluor 647-labeled dextran conjugate probe (red) was used to label biofilm matrix (α-glucans). Scale bars, 50 or 20 µm (X–Y isosurfaces) and 70 µm (three-dimensional reconstructions). Average Candida biovolumes (in µm³) after 1 h adhesion on S. oralis biofilms formed on polystyrene for 1 h (b) or 16 h (c). Similar experiments were performed on titanium surfaces and C. albicans biovolumes were quantified after 16 h (d). *p < 0.05 using the Bonferroni t-test. The error bars indicate standard deviations in triplicate experiments.

Fig. 6. C. albicans adhesion on polystyrene surfaces.

a Representative confocal images of C. albicans. Surfaces were precoated with purified GtfR (1 μg/ml) in the presence or absence of 1% sucrose for 1 h and C. albicans was inoculated after washing excess sucrose. C. albicans (green) was visualized 1 h post inoculation by staining with an FITC-conjugated anti-Candida antibody. b Average surface area covered by Candida cells as quantified by Image J analysis of three microscopic fields in each of two independent experiments. *p < 0.05 using the Bonferroni t-test. The error bars indicate standard deviations.

Fig. 7. Role of gtfR in mucosal biofilms in vivo.

Mice were inoculated with S. oralis wild-type (So34) or ΔgtfR strains, with or without C. albicans (Ca), and tongues were excised 5 days post inoculation at necropsy. a S. oralis mucosal burdens analyzed by qPCR using DNA extracted from tongues, and primers specific for the S. oralis 34 wefA-H gene. Cell (gene copy) numbers were calculated according to standard curves using known amounts of S. oralis 34 or ΔgtfR strain gDNA, and normalized over tissue weight. b C. albicans mucosal burdens as assessed by viable counts in tongue homogenates, normalized by tissue weight. Results of two independent mouse experiments, with 6–8 animals/group are shown. c Representative SEM images of biofilms forming on the tongue surface. Yellow arrows indicate the matrix-like material filling the spaces between filiform papillae in mice infected with wild-type S. oralis and C. albicans. Green arrows indicate yeast cells. *p < 0.05 using the Bonferroni t-test.

Fig. 8. Mucosal bacterial microbiome analyses based on high-throughput 16S rRNA gene sequencing.

a Beta diversity assessed by nonmetric multidimensional scaling (NMS) based on Bray–Curtis dissimilarities among the treatment groups. Shown are community structures in mice infected with C. albicans, in the presence or absence of added sucrose. Results represent bacterial community structure differences at the end of the experimental period (day 5). Communities clustered by type of treatment, indicating a significant effect of sucrose, which explained 52% of the variability (p < 0.02). b Relative abundance of bacterial 16S rRNA gene sequences corresponding to major mucosal genera in mice infected with C. albicans, in the presence or absence of added sucrose. c Relative abundance of endogenous Streptococcus, Enterococcus, and Lactobacillus in mice infected with C. albicans in the presence or absence of added sucrose, based on 16S rRNA gene sequences. *p < 0.05 for a t-test comparison of the two indicated groups.