Deletion of the Stress Response Gene DDR48 from Histoplasma capsulatum Increases Sensitivity to Oxidative Stress, Increases Susceptibility to Antifungals, and Decreases Fitness in Macrophages

Abstract

The stress response gene DDR48 has been characterized in Saccharomyces cerevisiae and Candida albicans to be involved in combating various cellular stressors, from oxidative agents to antifungal compounds. Surprisingly, the biological function of DDR48 has yet to be identified, though it is likely an important part of the stress response. To gain insight into its function, we characterized DDR48 in the dimorphic fungal pathogen Histoplasma capsulatum. Transcriptional analyses showed preferential expression of DDR48 in the mycelial phase. Induction of DDR48 in Histoplasma yeasts developed after treatment with various cellular stress compounds. We generated a ddr48∆ deletion mutant to further characterize DDR48 function. Loss of DDR48 alters the transcriptional profile of the oxidative stress response and membrane synthesis pathways. Treatment with ROS or antifungal compounds reduced survival of ddr48∆ yeasts compared to controls, consistent with an aberrant cellular stress response. In addition, we infected RAW 264.7 macrophages with DDR48-expressing and ddr48∆ yeasts and observed a 50% decrease in recovery of ddr48∆ yeasts compared to wild-type yeasts. Loss of DDR48 function results in numerous negative effects in Histoplasma yeasts, highlighting its role as a key player in the global sensing and response to cellular stress by fungi.

Keywords: DDR48; Histoplasma; amphotericin-B; ergosterol; hydrogen peroxide; ketoconazole; paraquat.

Figure 1

DDR48 gene expression levels are enriched in Histoplasma mycelia. (a) amino acid sequence of H. capsulatum DDR48 (top) and S. cerevisiae DDR48 (bottom) showing SYG repeats (red). (b) Expression of DDR48 in yeasts (Y) (blue bar) and mycelia (M) (grey bar) growth phases in liquid HMM measured by qRT-PCR (left) and representative northern blot (right) (n = 12) calculated relative to HHT1 transcript levels. All data generated were performed on three technical replicates and at least two biological replicates. Data is represented as the mean ± standard deviation (SD). Statistical analyses were performed using Student’s t test.

Figure 2

DDR48 transcript levels are increased in Histoplasma yeasts subjected to cellular stressors. Expression of DDR48 in yeast (blue bar) and mycelia (grey bar) growth phases measured by qRT-PCR (n = 9) at 0, 15, 30, and 60 min after the addition of various cellular stressors. The stressors include 0.1 µM paraquat (PQ)-superoxide anions, 2.5 mM hydrogen peroxide (HPO)-peroxides, 0.1 µg/mL amphotericin-B (AMB)-membrane disruption, 0.25 µg/mL ketoconazole (KTZ)-sterol synthesis inhibition, 50 µg/mL 5-fluorocytosine (5FC)-DNA/RNA biosynthesis inhibition, 1 mM methyl methanesulfonate (MMS)-DNA damage;methylating agent, 0.25 µM 4-nitroquinoline-1-oxide (4NQO)-UV-like damage, and heat-shock at 42 °C (HS). All data generated were performed on three technical replicates and at least two biological replicates. Data is represented as the mean ± standard deviation (SD). Statistical analyses were performed using one-way ANOVA with Tukey’s multiple comparisons test.

Figure 3

ddr48∆ yeasts possess decreased antioxidant activity, resulting in increased ROS levels and killing. (a) Survival (relative CFUs) of wildtype (DDR48(+)), ddr48∆, and ddr48∆/DDR48 strains measured (n = 8) after 4 days of growth in basal cell culture medium (HMM) supplemented with various concentrations of paraquat (PQ) or hydrogen peroxide (HPO). Survival was normalized to relative CFUs recovered from wildtype (DDR48(+)) yeasts with no treatment. (b) ROS levels measured (n = 8) in wildtype (DDR48(+)), ddr48∆, and ddr48∆/DDR48 yeasts by DCFDA fluorescence and flow cytometry. Data is shown as a representative histogram (left) and median fluorescence intensity (MFI) measurements (right). (c) ROS/Antioxidant pathway utilized by H. capsulatum yeasts with representative genes and their function. The type of oxidant generated by paraquat (PQ) and hydrogen peroxide (HPO) are shown in bold red. (d) Transcriptional profile of wildtype (WT), ddr48∆, and ddr48∆/DDR48 yeasts left untreated (Control) or treated with 0.1 µM PQ for 30 min (n = 9) measured by qRT-PCR relative to HHT1 levels. Expression of genes is presented as normalized row intensity (blue and red tiles) grouped by ROS or REDOX functions. (e) HPO destruction and (f) catalase activity of wildtype (WT), ddr48∆, and ddr48∆/DDR48 protein extracts measured (n = 9) 10 s after the addition of 5 mM HPO. (g) Superoxide dismutase (SOD) enzymatic activity (n = 6) of mid-log phase wildtype (WT), ddr48∆, and ddr48∆/DDR48 yeasts in HMM. Data is represented as the mean ± standard deviation (SD). Statistical analyses were performed using one-way ANOVA with Tukey’s multiple comparisons test.

Figure 4

Loss of DDR48 increases sensitivity to the antifungal drugs ketoconazole and amphotericin-B. Dose response curves (n = 10) of wildtype (WT), ddr48∆, and ddr48∆/DDR48 yeasts for (a) amphotericin-B (AMB) and (b) ketoconazole (KTZ). IC₅₀ values were determined by non-linear regression. Data is represented as the mean ± standard deviation (SD). Statistical analyses were performed using one-way ANOVA with Tukey’s multiple comparisons test.

Figure 5

DDR48-Depleted Histoplasma Yeasts Contain an Aberrant Ergosterol Biosynthesis Pathway That is Exacerbated by Antifungal Treatment. (a) Ergosterol biosynthesis pathway utilized by H. capsulatum yeasts with representative genes and their function. The step at which each class of antifungals interferes with the pathway are shown in bold red. (b) Transcriptional profile of wildtype (WT), ddr48∆, and ddr48∆/DDR48 yeasts grown in basal cell culture medium (HMM) and (c) left untreated (0) or treated with 0.1 µg/mL amphotericin-B (AMB; polyene) or 0.25 µg/mL ketoconazole (KTZ; azole) for 15, 30, and 60 min (n = 9) measured by qRT-PCR relative to HHT1 levels. Expression of genes is presented as normalized row intensity (blue and red tiles) grouped by chronological step in the ergosterol biosynthesis pathway. All data generated were performed on three technical replicates and at least two biological replicates.

Figure 6

Deletion of DDR48 decreases fitness of Histoplasma yeasts in macrophages. (a) RAW 264.7 macrophages were infected with wildtype yeasts (MOI 5:1) and DDR48 mRNA levels were measured (n = 9) from uninfected (Ctrl) yeasts or from yeasts 4 h and 24 h post-infection by qRT-PCR and calculated relative to HHT1 levels. (b) Resting (left) or stimulated (right; 10 ng/mL IFNγ) RAW 264.7 macrophages were infected with wildtype (WT), ddr48∆, or ddr48∆/DDR48 yeasts (MOI 5:1) and in vitro macrophage growth of each strain was determined by lysis of macrophages and comparison of levels of viable yeasts (CFUs) in the lysate 24 h post-infection. All data generated were performed on three technical replicates and at least two biological replicates. Data is represented as the mean ± standard deviation (SD). Statistical analyses were performed using one-way ANOVA with Sidak’s multiple comparisons test.