Host carbon sources modulate cell wall architecture, drug resistance and virulence in a fungal pathogen

Abstract

The survival of all microbes depends upon their ability to respond to environmental challenges. To establish infection, pathogens such as Candida albicans must mount effective stress responses to counter host defences while adapting to dynamic changes in nutrient status within host niches. Studies of C. albicans stress adaptation have generally been performed on glucose-grown cells, leaving the effects of alternative carbon sources upon stress resistance largely unexplored. We have shown that growth on alternative carbon sources, such as lactate, strongly influence the resistance of C. albicans to antifungal drugs, osmotic and cell wall stresses. Similar trends were observed in clinical isolates and other pathogenic Candida species. The increased stress resistance of C. albicans was not dependent on key stress (Hog1) and cell integrity (Mkc1) signalling pathways. Instead, increased stress resistance was promoted by major changes in the architecture and biophysical properties of the cell wall. Glucose- and lactate-grown cells displayed significant differences in cell wall mass, ultrastructure, elasticity and adhesion. Changes in carbon source also altered the virulence of C. albicans in models of systemic candidiasis and vaginitis, confirming the importance of alternative carbon sources within host niches during C. albicans infections.

Fig. 1

Carbon source influences cell wall architecture. A. TEM images of the C. albicans RM1000 cell walls grown on glucose, lactate or a mixture of both (scale bar = 0.5 μm). B. The thicknesses of the chitin plus β-glucan and mannan layers were quantified from TEM images of individual cells. Means ± SEM for n > 20 cells for each growth condition are shown. C and D. Biochemical content (C) and cell wall biomass (D) of exponential C. albicans cells grown on glucose or lactate. Means ± SEM for three independent experiments are shown. Relative to glucose-grown cells: *P < 0.05.

Fig. 2

Carbon source affects biophysical properties of the cell wall. C. albicans RM1000 cells were grown to mid-exponential phase (OD₆₀₀ 0.4–0.6) prior to the analyses. A. The relative cell wall porosity of glucose or lactate-grown cells assayed via polycation-induced leakage of UV-absorbing compounds (De Nobel et al., 1990). B. Cell surface hydrophobicity of glucose- and lactate-grown C. albicans RM1000 cells assayed via the water contact angles between a liquid surface and these cells (n > 20 drops). C. Adhesion of C. albicans cells to a non-treated polystyrene plastic surface. Adhered cells were scraped from the surface, serially diluted, plated on agar and their numbers determined as CFUs (shown as percentage CFUs compared to glucose-grown controls). D. Adhesion force and adhesion energy of glucose- and lactate-grown cells examined by AFM (n = 6). E. Cell surface flexibility (indicated by Young's modulus) measured by AFM (n = 6). Results are presented as means ± SEM for at least three independent experiments. Relative to glucose-grown cells: *P < 0.05.

Fig. 3

Osmotic stress resistance and adaptation are affected by carbon source. A. Resistance of wild-type (RM1000), hog1Δ and mkc1Δ C. albicans cells (Table S1) to osmotic stress (2 M NaCl) during exponential growth on glucose, lactate or glucose plus lactate. Means ± SEM for four independent experiments are presented. Relative to glucose-grown cells: *P < 0.05. B. Phosphorylation and activation of Hog1 and Mkc1 in glucose- and lactate-grown cells following exposure to 1 M NaCl as revealed by Western blotting with phospho-specific antibodies. Similar results were obtained in four independent experiments. C. Dynamic changes in cell volume following hyperosmotic shock (1 M NaCl; n > 25 cells).

Fig. 4

Carbon source impacts upon resistance to other stresses and antifungal drugs. A. Sensitivity of C. albicans RM1000 to Calcofluor White (200 μg ml⁻¹) and Congo Red (300 μg ml⁻¹) when grown on glucose, lactate or a mixture of both. B. Sensitivity of C. albicans RM1000 to the antifungal drugs amphotericin B (Ambisome; 10 μg ml⁻¹), caspofungin (6.4 ng ml⁻¹), miconazole (25 μg ml⁻¹) and tunicamycin (4 μg ml⁻¹). Relative to glucose-grown cells: *P < 0.05.

Fig. 5

Carbon source impacts upon resistance of other pathogenic Candida species. Candida cells (Table S1) were grown to exponential phase on glucose or lactate, and their resistance to (A) hyperosmotic stress (2 M NaCl) and (B) amphotericin B (Ambisome; 10 μg ml⁻¹) was determined. Means ± SEM for at least three independent experiments are presented. Relative to glucose-grown cells: *P < 0.05.

Fig. 6

Impact of various carbon sources upon stress and drug resistance. A–C. Sensitivity of C. albicans RM1000 to (A) 2 M NaCl, (B) 10 μg ml⁻¹ amphotericin B and (C) 300 μg ml⁻¹ Congo Red following growth on different carbon sources. Data represent results from at least three independent experiments (means ± SEM). Relative to glucose-grown cells: *P < 0.05. D. Growth curves of C. albicans RM1000 on different carbon sources. Cells were grown in either 2% or 0.2% (for oleic acid) carbon source for 4 h at 30°C and the relative levels of growth were determined by monitoring the OD₆₀₀. Each curve represents the average of three biological replicates, with a maximum SEM ± 0.4.

Fig. 7

Growth on serum affects cell wall architecture. A. TEM of the C. albicans RM1000 yeast cell walls growing on 2% glucose, serum or serum plus 1% glucose (scale bar = 0.5 μm). B. The thicknesses of the chitin plus β-glucan and mannan layers were quantified from TEM images of individual cells. Means ± SEM for n > 20 cells for each growth condition are shown. Relative to glucose-grown cells: *P < 0.05.

Fig. 8

Growth on alternative carbon sources influences C. albicans virulence. C. albicans (RM1000 containing Clp20, Table S1) cells grown on different carbon sources were used to infect mice intravenously. A and B. Kidney burdens (A) and weight change (B) were measured 72 h post infection (means ± SEM; n = 6). C. Infection outcome scores were calculated after 72 h (means ± SEM; n = 6), higher outcome scores reflecting more severe infection (MacCallum et al., 2010). D. Mice were infected intravaginally with C. albicans (RM1000 containing Clp20) cells grown on different carbon sources. Fungal burdens in vaginal swabs were measured 72 h post infection (means ± SEM; n = 6). In all panels, the asterisk (*) indicates P < 0.05 relative to glucose-grown C. albicans infection.