Fig 1.
The oak tree strains display significant differences in nutrient use, and Oak 2 shows resistance to amphotericin B.
Growth of all three oak tree isolates was assessed as growth curves at 30 °C in different media comprising specific carbon (A) or nitrogen (B) sources. The growth is shown as area under the curve relative to the reference strain SC5314 (% growth). Asterisks indicate significance compared to SC5314 calculated using a one-way ANOVA with Tukey’s multiple comparisons test (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001) (n = 3). (C, D) Resistance to fluconazole and amphotericin B was determined by performing a drop test of serially diluted C. albicans solutions on YPD agar plates supplemented with the indicated antifungal concentration. Pictures were taken (C) after 2 days at 30 °C and (D) after 6 days at 37 °C. For all pictures, contrast and brightness were increased by 20%.
Fig 2.
One oak tree C. albicans isolate shows an increased virulence potential on epithelial cells and augmented survival in an ex vivo blood model.
(A) Damage to oral epithelial cells by the environmental oak tree strains and clinical isolates was determined after 24 h by LDH release (n = 4). The lower dashed line indicates the uninfected control. The upper dashed line indicates the full lysis control. One-way ANOVA with Tukey’s multiple comparisons test (** p < 0.01, **** p < 0.0001). (B) Damage to intestinal cells was monitored 48 h post infection by quantifying LDH release (n = 3). One-way ANOVA with Tukey’s multiple comparisons test (* p < 0.05). (C) Adhesion of environmental strains was determined relative to SC5314 at 1 h post-infection of oral cells by staining and microscopy (n = 3). (D) Hypha formation on oral cells was determined by microscopic evaluation 3 h post infection. One-way ANOVA with Tukey’s multiple comparisons test (**** p < 0.0001) (for oak strains and SC5314: n = 3-6, clinical isolates: n = 2-3). (E) Invasion was measured by differential staining of oral cells 3 h post infection. Invasion was calculated as percentage of invading hyphae compared to all counted hyphae (n = 3). (F) Survival of C. albicans in whole human blood was assessed over the course of 4 h by plating surviving colony-forming units (4 donors). Survival is shown relative to the inoculum. Heat map displays fold change to SC5314 and p-values calculated with a two-way ANOVA with repeated measures and Dunnett’s multiple comparison test. Significant p-values are indicated in bold.
Fig 3.
ECE1 (Extent of cell elongation 1) characteristics of Oak 2 resemble the SC5314 reference strain.
ECE1 expression on oral (A) and intestinal epithelial cells (B) is shown as fold change relative to SC5314 at the indicated time points. The expression is normalized to ACT1 mRNA levels. Statistical significance was calculated using a one-way ANOVA with Tukey’s multiple comparisons test (* p < 0.05, **** p < 0.0001). (C) Predicted Ece1 sequences were aligned using Clustal Omega. Identical amino acids are indicated as asterisks (*) and similar ones with dots (. or:). Blue indicates the signal sequence and pink amino acids (KR, KK, R) mark the Kex protease cleavage sites of the polypeptide. (D) Secretion of Ece1 peptides and peptide fragments from the indicated strains into culture medium, evaluated by LC-MS/MS analysis. The mean peptide-spectrum matches (PSM) scores for each identified peptide are presented (n = 2); if no number is given, the peptide was not detected. Oak 3 was not included in this measurement.
Fig 4.
The oak tree strains show differences in their interaction with primary macrophages compared to the reference strain.
(A) Damage to monocyte-derived macrophages by the environmental oak tree strains was determined after 24 h by measuring LDH release. The lower dashed line indicates the uninfected control. The upper dashed line indicates the full lysis control. Each circlerepresents one donor. Significance was calculated by a one-way ANOVA with Tukey’s multiple comparisons test (* p < 0.05, ** p < 0.01) (n = 4-6 donors). (B) Intracellular survival of the C. albicans strains in human monocyte-derived macrophages was assessed 3 h post infection by lysis and plating of the intracellular fungal cells. Survival is depicted relative to the reference strain SC5314. Each circlerepresents one donor. Significance was calculated by a one-sample t-test against 100% (* p < 0.05) (n = 6 donors). (C) Microscopic pictures were taken either 3 h after infecting human monocyte-derived macrophages (37 °C, CO2) with a 63 × magnification with immersion oil (scale bar, 20 µm). Samples were stained with DAPI (nuclei staining, purple) and Concanavalin A-AlexaFluor647 (fungal staining, blue). Figure shows representative pictures from at least 4 donors.
Fig 5.
Experimental evolution with a host dietary sugar leads to antifungal resistance and metabolic flexibility.
(A) Schematic overview of the evolution experiment in YNB medium supplemented with 2% galactose. Oak 1 and SC5314 overnight cultures were diluted 1:50 into the galactose-containing medium and were incubated for 6 weeks at 30 °C, with passaging them into fresh medium every 2-3 days when the early stationary phase was reached. Growth was measured as OD600nm over the whole time course. Samples were taken and frozen after every passage. (B) After the evolution experiment, growth of the sugar-adapted strain (Evo) was characterized in multiple carbon and nitrogen sources. The growth is shown as area under the curve relative to the reference strain SC5314 (% growth). Black asterisks indicate significance compared to SC5314 and blue asterisks indicate whether the sugar-adapted strain grew significantly different compared to its parental Oak 1 strain. Significance was calculated by using a one-way ANOVA with Tukey’s multiple comparisons test (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001) (n = 3). (C, D) Antifungal resistance of the galactose-evolved strain was assessed by performing growth curves in YPD supplemented with either 32 µg/ml fluconazole or 0.5 µg/ml amphotericin B at 30 °C for 72 h, and is depicted as area under the curve. Significance was calculated by using one-way ANOVA with Tukey’s multiple comparisons test (* p < 0.05, ** p < 0.01, *** p < 0.001) (n = 3).
Fig 6.
The sugar adaptation leads to transcriptomic remodeling.
(A) Principal component analysis of the transcriptomes for the SC5314, two oak tree isolates, and Evo on intestinal cells. Samples clustered together depending on their origin, and the sugar-adapted strain underwent a positive shift along the PC1 axis. (B) KEGG pathways enriched for genes that were significantly up- or down-regulated on intestinal epithelial cells 24 or 48 h post infection. % genes in pathway is indicated by the dot size and represents the number of genes within each pathway that were differentially regulated compared to the indicated reference strains (below the graph). The color scale indicates the adjusted p value.
Fig 7.
The sugar-evolved Evo strain shows increased adhesion, filamentation, invasion, and host cell damage during interaction with epithelial cells.
(A) Damage to oral cells and (B) intestinal cells was evaluated by quantifying epithelial LDH release 48 h post infection (oral n = 3, intestinal n = 4). The lower dashed line indicates the uninfected control. The difference between Oak 1 and Evo was not statistically significant in (A) and (B). (C) Adhesion of C. albicans strains was assessed by fixing and staining the fungal cells 1 h post infection of oral cells. The average number of attached C. albicans cells per microscopic image is plotted relative to the reference strain SC5314 (n = 3). (D) Hypha formation was determined by microscopic evaluation counting the hypha-forming cells relative to all C. albicans cells on oral cells 3 h post infection (n = 3). (E) Hyphal length was measured 3 h post infection of oral epithelial cells (n = 3). The difference between Oak 1 and Evo was not statistically significant. (F) Invasion was measured by infecting oral cells, fixing, and differentially staining 3 h post infection. Invasion was calculated as percentage of invading hyphae compared to all counted fungal cells (n = 3). (G) ECE1 expression was analyzed after 3 h of growth in hypha-inducing medium. The expression is normalized to ACT1 (ΔCt). The control condition is the yeast morphology of the reference strain SC5314 (ΔΔCt) (n = 3). The difference between Oak 1 and Evo was not statistically significant. (A-G) Significance by one-way ANOVA with Tukey’s multiple comparisons test (* p < 0.05, ** p < 0.01, *** p < 0.001). (H) Predicted Ece1 sequences were aligned using Clustal Omega. Identical amino acids are indicated as asterisks (*). Blue indicates the signal sequence, and pink-colored amino acids show the Kex protease cleavage sites of the polypeptide.
Fig 8.
Virulence of the parental Oak 1 isolate and the sugar-adapted Evo strain in a transwell assay and a complex gut-on-chip model.
(A) Translocation of the strains was measured in an intestinal transwell model by plating the translocated colonies 24 h post infection. Significance was calculated using a one-way ANOVA with Tukey’s multiple comparison test (** p < 0.01) (n = 3). (B) Barrier integrity was determined by measuring TEER (transepithelial electrical resistance) in an intestinal transwell model 24 h after infection. Significance was calculated using a one-way ANOVA with Tukey’s multiple comparison test (*** p < 0.001, **** p < 0.0001) (n = 3). (C) Percentage of adhesion to the gut epithelium in the gut-on-chip model was determined by plating the amount of non-attached C. albicans in the flow-through relative to the injected inoculum (n = 3). (D) Damage was measured as LDH release from both, the gut side and the vascular compartment of the gut-on-chip 24 h post infection (n = 3). (E) Fungal burden was determined by plating the intestinal dispersion (gut flow-through), intestinal fungal burden (intestinal lysate), vascular invasion (vascular lysate), and dissemination (vascular flow-through) 24 h post infection (n = 3). To show the variation between the different gut-on-chip experiments, the individual values of the biological replicates are shown in (D) and (E). Significance was calculated using a two-way ANOVA with Tukey’s multiple comparison test (** p < 0.01). The difference between Oak 1 and Evo was not statistically significant in panel (A), (B), (C), (D), and (E).