Secretion of the siderophore rhizoferrin is regulated by the cAMP-PKA pathway and is involved in the virulence of Mucor lusitanicus

Abstract

Mucormycosis is a fungal infection caused by Mucorales, with a high mortality rate. However, only a few virulence factors have been described in these organisms. This study showed that deletion of rfs, which encodes the enzyme for the biosynthesis of rhizoferrin, a siderophore, in Mucor lusitanicus, led to a lower virulence in diabetic mice and nematodes. Upregulation of rfs correlated with the increased toxicity of the cell-free supernatants of the culture broth (SS) obtained under growing conditions that favor oxidative metabolism, such as low glucose levels or the presence of H₂O₂ in the culture, suggesting that oxidative metabolism enhances virulence through rhizoferrin production. Meanwhile, growing M. lusitanicus in the presence of potassium cyanide, N-acetylcysteine, a higher concentration of glucose, or exogenous cAMP, or the deletion of the gene encoding the regulatory subunit of PKA (pkaR1), correlated with a decrease in the toxicity of SS, downregulation of rfs, and reduction in rhizoferrin production. These observations indicate the involvement of the cAMP-PKA pathway in the regulation of rhizoferrin production and virulence in M. lusitanicus. Moreover, rfs upregulation was observed upon macrophage interaction or during infection with spores in mice, suggesting a pivotal role of rfs in M. lusitanicus infection.

Figure 1

Analysis of the toxicity of the cell-free supernatant of the culture broth (SS) of M. lusitanicus under different growth conditions. Quantification of the toxicity of the SS obtained from the wild-type strain R7B grown in YPG medium with 2% glucose (YPG-2%) after (A) aerobic growth (H-SS) or (B) anaerobic growth (Y-SS) at indicated times against Caenorhabditis elegans; hyphae and yeast cells obtained from 3 and 8 h, respectively, were recovered under the corresponding growth conditions and tested against the nematode. (C) Hyphae from a culture with 3 h of growth or its SS obtained from the YNB medium with 2% glucose (YNB-2%) after aerobic growth at indicated times were used to treat the nematode. Quantification of the toxicity of the SS obtained after aerobic growth for 48 h in (D) YPG medium and (E) YNB medium supplemented with different concentrations glucose (0.1, 0.5, 1, and 2%). A total of 15–20 nematodes were used per experiment and incubated at 18 °C for 72 h. The results presented are the average values obtained from four independent experiments. Data were statistically analyzed using the Mantel-Cox test. ***P < 0.01. When results were not considered significant, we did not provide an additional indication (P > 0.05).

Figure 2

The toxic factor in the cell-free supernatant of the culture (SS) of M. lusitanicus was susceptible to protease activity and had a low molecular weight. (A) SS from the R7B strain obtained under aerobic conditions (H-SS) in YNB medium supplemented with 0.1% glucose (YNB-0.1%) after 48 h of growth was passed through a molecular weight exclusion filter (cut off 3 kDa), obtaining a V_o fraction consisting of molecules with a weight lesser than 3 kDa and a V₁ fraction consisting of molecules with a weight greater than 3 kDa. (B) SS or V_o and V₁ fractions from panel A were treated with proteinase K (PK) for 2 h at 37 °C; SS or V_o and V₁ fractions from panel A were incubated for 2 h at (C) 37 °C or (D) 95 °C. All the SSs and fractions were used to treat Caenorhabditis elegans to test their toxicity. A total of 15–20 nematodes were used per experiment and incubated at 18 °C for 72 h. The results presented are the average of the values obtained from four independent experiments. The data were statistically analyzed using the Mantel-Cox test. ***P < 0.01. When results were considered not significant, we did not provide an additional indication (P > 0.05).

Figure 3

Addition of Fe²⁺ downregulates rfs expression and negatively influences the virulence of the cell-free supernatant of the culture (SS) from M. lusitanicus. (A) Quantification of the mRNA levels of rfs in the mycelium of MU636 grown for 48 h under aerobic conditions in Vogel´s medium supplemented with 2% glucose with or without supplementation with 200 μM Fe²⁺ or in the livers of diabetic mice infected previously with 3 × 10⁶ spores of the MU636 strain and euthanized 7 days post infection; a ∆Ct analysis was performed to compare the mRNA levels between samples. tfc1 was used as reference housekeeping gene (B) Effect of Fe²⁺ present at different concentrations in the YNB medium with 0.1% glucose (YNB-0.1%) on nematode survival. (C) Toxicity of the SSs obtained when the fungus (strain MU636) was aerobically grown (H-SS) for 48 h in YNB-0.1% supplemented with 22.5 μM or 45 μM Fe²⁺. All the SSs were used to treat Caenorhabditis elegans. A total of 15–20 nematodes were used per well and incubated at 18 °C for 72 h. Significance testing was performed using ANOVA with Fisher's exact test, *P < 0.05; **P < 0.01; ***P < 0.001. The virulence results presented are the average from four independent experiments. Data were statistically analyzed using the Mantel-Cox test. ***P < 0.01. When results were not considered significant, we did not provide an additional indication (P > 0.05).

Figure 4

rfs is essential for the toxicity of the cell-free supernatant of the culture (SS) of M. lusitanicus. The toxicity of the SSs obtained from the different strains after 48 h of aerobic growth (H-SS) in (A) YPG medium (YPG-0.1%) and (B) YNB medium (YNB-0.1%) supplemented with 0.1% glucose was assessed against the nematode. A total of 15–20 nematodes were used per experiment and incubated at 18 °C for 72 h. The results presented are the average of results obtained from four independent experiments. Data were statistically analyzed using the Mantel-Cox test. ***P < 0.01. When results were not considered significant, we did not provide an additional indication (P < 0.1). (C) Extracted ion chromatograms of rhizoferrin (m/z = 437.141) in SS obtained from the different M. lusitanicus strains after 48 h of growth in the YNB medium with 0.1% glucose. Chromatograms show the relative abundance of the most intense sample. Top, a representative mass spectrum of rhizoferrin. (D) Relative abundance of rhizoferrin in the corresponding SS from the different strains. Total RNA was obtained from mycelium grown for 8 h either in (E) YPG medium (with 2% glucose) or (E) YNB medium (with 2% glucose), and rfs expression was assayed by RT-qPCR. A ∆Ct analysis was performed to compare the mRNA levels between samples, using tfc1 as reference housekeeping gene. Significance testing was performed using ANOVA with Fisher's exact test, *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 5

Role of rfs in the regulation of M. lusitanicus virulence in mice. (A) Spores (3 × 10⁷) from the MU636, Δrfs, Δrfs + rfswt, and MU636 + rfswt strains were intraperitoneally injected into streptozotocin-induced diabetic mice, and survival was monitored daily for 15 days. The experiments were repeated two times in independent groups with n = 10. Data were statistically analyzed using the Mantel-Cox test. ***P < 0.01. When results were not considered significant, we did not provide an additional indication (P > 0.05). Mice surviving after 7 days post-inoculation were considered to (B) determine the change in the weight of mice since the initial infection. Animals were euthanized at 15 days post-inoculation, and the livers were removed to isolate nucleic acids to determine the (C) fungal burden in the mice using quantitative polymerase chain reaction (qPCR) for the expression of the tfc-1 gene from M. lusitanicus and Actb from Mus musculus. A ΔCt analysis was performed to compare the relative abundance of the fungal DNA compared with that of the murine DNA. The mRNA levels of the inflammation marker IL-6 in the liver of mice (D) infected with spores from the different strains and (E) inoculated with 200 μL of the V_o fraction obtained through a molecular weight exclusion filter (cut off 3 kDa) were measured using reverse transcription quantitative polymerase chain reaction (RT-qPCR). Data are presented as the average values obtained from four independent biological replicates; ± corresponds to standard error (SE). Significance testing was performed using ANOVA with Fisher's exact test, *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 6

Virulence, rfs transcript levels, and rhizoferrin accumulation are influenced by oxidative stress and macrophage interaction. (A) mRNA levels of rfs in Mucor lusitanicus were determined after interaction with macrophages (or not) for 6 h. (B) Spores of the MU636 strain were incubated with different concentrations of H₂O₂ for 1 h in YPG with 2% glucose (YPG-2%), following which the germinating spores were grown aerobically for 48 h in YPG-2% The toxicity of the cell-free supernatant of the culture (SS) obtained after 48 h of aerobic growth (H-SS) was assessed against the nematode. (C) mRNA levels of atp9, adh1, and rfs; (D) cit1 and spe1 expression was quantified in the mycelium of the MU636 strain grown for 8 h in YPG-2% in the absence or presence of 0.5 mM H₂O₂. (E) Relative abundance of rhizoferrin in SS obtained from the R7B strain of M. lusitanicus grown in YNB with 0.1% glucose (YNB-0.1%) in the absence or presence of 0.5 mM H₂O₂. (F) The virulence of the SSs obtained after 48 h of growth of different strains in YPG-2% in the absence or presence of 0.5 mM H₂O₂ was assessed in the nematode. The SSs were used to treat Caenorhabditis elegans. A total of 15–20 nematodes were used per experiment and incubated at 18 °C for 72 h. The results presented are the average of values obtained from four independent experiments. Data were statistically analyzed using the Mantel-Cox test for virulence assays. ***P < 0.01. Significance testing was performed using ANOVA with Fisher's exact test, *P < 0.05; **P < 0.01; ***P < 0.001. When results were not considered significant, we did not provide an additional indication (P > 0.05).

Figure 7

Role of Rfs in the mitochondrial membrane potential, reactive oxygen species generation, and M. lusitanicus virulence. Germinules formed at 3 h of growth in YNB with 0.1% glucose (YNB-0.1%) were visualized (A) in the presence of MitoTracker Green FM, following which the cells were visualized by confocal microscopy (120 ×), bars = 20 mm. The fluorescence signal was quantified using the (B) Mitotracker signal and (C) hydroxyl radical (OH⁻) concentration. The results presented are the average of values obtained from four independent experiments. Significance testing was performed using ANOVA with Fisher's exact test, *P < 0.05; **P < 0.01; ***P < 0.001. (D) Effect of 10 mM N- acetylcysteine (N-ace) or (E) 0.5 mM potassium cyanide (KCN) on the virulence of cell-free supernatants of the culture (SS) from MU636-, R7B-, and rfs-overexpression (MU636 + rfswt) strains aerobically grown (H-SS) in YNB-0.1%. The SSs were used to treat Caenorhabditis elegans. A total of 15–20 nematodes were used per experiment and incubated at 18 °C for 72 h. The results presented are the average of values obtained from four independent experiments. Data were statistically analyzed using the Mantel-Cox test. ***P < 0.01. When results were not considered significant, we did not provide an additional indication (P > 0.05).

Figure 8

Involvement of the cAMP-PKA pathway in the regulation of rfs, accumulation of rhizoferrin, and virulence of M. lusitanicus. (A) Effect of YPG supplemented with 0.1%, 2%, 4%, and 6% glucose (YPG-0.1%, YPG-2%, YPG-4%, and YPG-6%) on the virulence of the cell-free supernatant of the culture (SS) obtained from R7B- and rfs-overexpression (MU636 + rfswt) strains. (B) Effect of the SSs of MU636, R7B, and rfs-overexpression (MU636 + rfswt) strains obtained after aerobic growth (H-SS) in YNB containing 0.1% glucose (YNB-0.1%) in the presence or absence of 3 mM dibutryril cAMP (db-cAMP) on Caenorhabditis elegans. (C) Effect of SSs of adenylyl cyclase 1 (cyr1)- and phosphodiesterase 2 (pde2)-overexpressing M. lusitanicus on C. elegans. (D) Effect of SSs obtained under the aerobic growth of different M. lusitanicus mutants of the regulatory subunit of PKA on C. elegans survival. The results presented are the average of the values obtained from four independent experiments. Data were statistically analyzed using the Mantel-Cox test. ***P < 0.01. When results were not considered significant, we did not provide an additional indication (P > 0.05). (E) Quantification of rhizoferrin in the SSs obtained after the aerobic growth of R7B in YNB-0.1%, MU636, and rfs-overexpression (MU636 + rfswt) in the absence or presence of 3 mM db-cAMP, or MU636 overexpressing phosphodiesterase 2 (pde2). Significance testing was performed using ANOVA with Fisher's exact test, *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 9

Proposed model for the regulation of rhizoferrin levels via the cAMP-PKA pathway in M. lusitanicus. Presence of organic nitrogen (peptone) or high levels of glucose possibly stimulated the enzymatic activity of adenylyl cyclase (Cyr1) and inhibited phosphodiesterase (Pde2), eventually inducing the generation of high levels of cAMP, which binds to PkaR1 (regulatory subunit 1) of PKA (protein kinase A), thereby activating catalytic PKA (PkaC unknown). This stimulates fermentative metabolism and decreases oxidative metabolism, which negatively regulates reactive oxygen species (ROS) production and decreases the mRNA levels of Rfs, probably via the transcriptional regulation of SreA and Yap1. Lower oxidative metabolism correlates with the downregulation of citrate synthase 1 (cit1) and ornithine decarboxylase (spe1), which synthetize Rfs substrates. This leads to lower levels of rhizoferrin accumulation in the cell exterior, correlating with decreased virulence. In the same context, excess Fe²⁺ led to the transcriptional downregulation of rfs, which correlated with lower virulence. Rhizoferrin chelates extracellular iron and transports it to the intracellular space; intracellular rhizoferrin could store iron and help avoid free iron toxicity. Meanwhile, decreased cAMP, low levels of glucose, presence of oxidative stressors, such as H₂O_2, or stimulation of the activity of Pde2 produced the contrary effect, inducing oxidative metabolism and increasing rhizoferrin production, which correlates with increased virulence of M. lusitanicus.