The Aspergillus fumigatus Phosphoproteome Reveals Roles of High-Osmolarity Glycerol Mitogen-Activated Protein Kinases in Promoting Cell Wall Damage and Caspofungin Tolerance

Abstract

The filamentous fungus Aspergillus fumigatus can cause a distinct set of clinical disorders in humans. Invasive aspergillosis (IA) is the most common life-threatening fungal disease of immunocompromised humans. The mitogen-activated protein kinase (MAPK) signaling pathways are essential to the adaptation to the human host. Fungal cell survival is highly dependent on the organization, composition, and function of the cell wall. Here, an evaluation of the global A. fumigatus phosphoproteome under cell wall stress caused by the cell wall-damaging agent Congo red (CR) revealed 485 proteins potentially involved in the cell wall damage response. Comparative phosphoproteome analyses with the ΔsakA, ΔmpkC, and ΔsakA ΔmpkC mutant strains from the osmotic stress MAPK cascades identify their additional roles during the cell wall stress response. Our phosphoproteomics allowed the identification of novel kinases and transcription factors (TFs) involved in osmotic stress and in the cell wall integrity (CWI) pathway. Our global phosphoproteome network analysis showed an enrichment for protein kinases, RNA recognition motif domains, and the MAPK signaling pathway. In contrast to the wild-type strain, there is an overall decrease of differentially phosphorylated kinases and phosphatases in ΔsakA, ΔmpkC, and ΔsakA ΔmpkC mutants. We constructed phosphomutants for the phosphorylation sites of several proteins differentially phosphorylated in the wild-type and mutant strains. For all the phosphomutants, there is an increase in the sensitivity to cell wall-damaging agents and a reduction in the MpkA phosphorylation upon CR stress, suggesting these phosphosites could be important for the MpkA modulation and CWI pathway regulation.IMPORTANCEAspergillus fumigatus is an opportunistic human pathogen causing allergic reactions or systemic infections, such as invasive pulmonary aspergillosis in immunocompromised patients. The mitogen-activated protein kinase (MAPK) signaling pathways are essential for fungal adaptation to the human host. Fungal cell survival, fungicide tolerance, and virulence are highly dependent on the organization, composition, and function of the cell wall. Upon cell wall stress, MAPKs phosphorylate multiple target proteins involved in the remodeling of the cell wall. Here, we investigate the global phosphoproteome of the ΔsakA and ΔmpkCA. fumigatus and high-osmolarity glycerol (HOG) pathway MAPK mutants upon cell wall damage. This showed the involvement of the HOG pathway and identified novel protein kinases and transcription factors, which were confirmed by fungal genetics to be involved in promoting tolerance of cell wall damage. Our results provide understanding of how fungal signal transduction networks modulate the cell wall. This may also lead to the discovery of new fungicide drug targets to impact fungal cell wall function, fungicide tolerance, and virulence.

Keywords: Aspergillus fumigatus; MpkC; SakA; caspofungin; cell wall integrity pathway; osmotic stress.

FIG 1

Phosphorylation profile of proteins from A. fumigatus wild type upon exposure to Congo red (CR 10 min). (A) Workflow of experimental design for phosphoproteome results. (B) Principal-component analysis plot for the A. fumigatus wild type without CR treatment (T = 0, T0WT; six replicates) and with 10-min CR treatment (CRWT; four replicates). (C) Number of phosphopeptides and unique proteins modified by phosphorylation during CR treatment. (D) Volcano plot of total phosphopeptides identified, showing a major increase of phosphorylation after CR treatment. FC, fold change. (E) Venn diagram of total 756 phosphopeptides indicating the profile of total phosphorylation (380) or total dephosphorylation (3) in wild-type strains. (F) Number of phosphosites in serine (S), threonine (T), or tyrosine (Y) for total phosphopeptides. (G) Number of phosphorylation sites for each peptide.

FIG 2

Functional categorization and identification of kinase substrates. (A) Phosphopeptide functional categorization of approximately 75% of proteins with known function. (B) Profile of distribution of kinase families that are able to phosphorylate the 756 phosphopeptides identified by LC-MS/MS as predicted by Group-based Prediction System 3.0 (GPS; with medium-stringency cutoff). (C) Consensus sequence of the site of phosphorylation of p38/Hog1 peptide substrates predicted by Group-based Prediction System 3.0 (GPS; with medium-stringency cutoff). (D and E) Phosphorylation distribution of all kinases (D) and transcription factors (TFs) (E) modulated by phosphorylation during Congo red response. Gray blocks indicate lack of the phosphopeptide, while green (P) indicates the presence of a phosphopeptide.

FIG 3

Phosphoproteomics allows the identification of novel transcription factors involved in the CWI pathway. A. fumigatus conidia (1 × 10⁵) were inoculated on solid minimal medium (MM) with different concentrations of Congo red (CR), calcofluor white (CFW), caspofungin (CASPO), and sorbitol and grown for 5 days at 37°C. All plates were grown in triplicate, and the average ± SD was plotted. A Student t test using Prism GraphPad (version 6) was applied to confirm the statistical significance difference between treatment and control (P value < 0.05, ***; < 0.01, **; < 0.001, *).

FIG 4

Phosphorylation profile of proteins from A. fumigatus and MAPK mutants during incubation with Congo red (CR 10 min). (A to C) Volcano plots of total phosphopeptides identified in ΔsakA mutant versus wild type (WT), ΔmpkC mutant versus WT, and the double ΔsakA ΔmpkC mutant versus WT, respectively. (D) Percentage of proteins with unknown and known function in all three phosphoproteomes ΔsakA (top); ΔmkpC (middle); and ΔsakA ΔmpkC (bottom). All of them showed more than 60% characterized proteins. (E to G) Functional categorization of approximately 75% of proteins with known function differentially phosphorylated in ΔsakA, ΔmpkC, and ΔsakA ΔmpkC strains during CR stress. (H) Venn diagram of total differentially phosphorylated proteins (n = 621) representing the number of common phosphoproteins between strains in each intersection (each strain has a set of differentially phosphorylated proteins after CR treatment compared to its untreated control). Thus, the WT strain has a total set of 431 phosphoproteins from which a unique set of 245 is not shared with any other condition, 69 are shared only with the ΔsakA mutant, 13 are shared only with the ΔsakA ΔmpkC mutant, and 38 are shared only with the ΔmpkC mutant, while 16 phosphoproteins are shared simultaneously by the four strains.

FIG 5

General A. fumigatus functional protein association network based on the protein phosphorylation profile during incubation with Congo red (CR 10 min). The whole set of differentially phosphorylated proteins from the wild-type (WT), ΔsakA, ΔmpkC, and ΔsakA ΔmpkC strains during CR stress was combined for the generation of a general protein association network. Each node represents a protein that is differentially phosphorylated in at least one strain under CR stress. Node colors are divided into gray nodes, for proteins which were not differentially phosphorylated, and heatmap-colored nodes (blue-white-red), for proteins which were differentially phosphorylated in each strain in relation to its control (the same strain without any stress condition). Node shapes represent molecular functions and are divided into triangles (phosphatases), squares (kinases), and circles (other functions). Each edge represents a functional protein association retrieved from the STRING server (medium confidence threshold of 0.4 for the interaction score), and node sizes represent the degree of each node (number of edges connected to the node). It is important to note that not all the differentially phosphorylated proteins are present in the network, as many proteins did not present any functional associations within the whole set. Changes in the color patterns of nodes from panels A to D represent changes in the differential phosphorylation of the proteins comprising the network in WT, ΔsakA, ΔmpkC, and ΔsakA ΔmpkC strains, respectively, during CR stress.

FIG 6

Subnetwork of functional protein associations based only on the phosphorylation profile of protein kinases and phosphatases during incubation with Congo red (CR 10 min). The subset of differentially phosphorylated kinases and phosphatases from the wild-type (WT), ΔsakA, ΔmpkC, and ΔsakA ΔmpkC strains during CR stress was extracted from the general protein association network. Each node represents a protein that was differentially phosphorylated in at least one strain under CR stress. Node colors are divided into gray nodes (proteins that were not differentially phosphorylated) and heatmap-colored nodes (blue-white-red; proteins which were differentially phosphorylated in each strain in relation to its untreated control). Node shapes represent molecular functions and are divided into squares (kinases) and triangles (phosphatases). Each edge represents a functional protein association retrieved from the STRING server (medium confidence threshold of 0.4 for the interaction score). Node size is representative of the number of edges connected to the node. Note that not all the differentially phosphorylated proteins are present in the network, as for many proteins no functional associations were available. Changes in the color patterns of nodes from panels A to D represent changes in the differential phosphorylation of the proteins comprising the network in WT, ΔsakA, ΔmpkC, and ΔsakA ΔmpkC strains, respectively, during CR stress.

FIG 7

Sites of phosphorylation (S or T) identified during response to Congo red (CR) in wild-type or three kinase null mutant strains. (A, C, and E) Schematics of the kinases showing their kinase domain (blue or green boxes) and position of phosphorylated amino acid residues highlighted in red. (B, D, and F) Phosphorylation sites identified in the phosphoproteome, according to the conditions indicated. For each residue, a score for a kinase site as predicted by Group-based Prediction System 3.0 (GPS; with medium-stringency cutoff) is indicated.

FIG 8

Phosphomutants for RckA serine phosphorylation sites. (A) Molecular modeling for RckA wild type and predicted phosphomutants. (B to F) Growth phenotypes of the wild type and ΔrckA, RckA^S481D, and RckA^S481A mutants grown for 5 days at 37°C on minimal medium (MM) or MM plus cell wall-damaging or osmotic stress agents (Congo red [CR], calcofluor white [CFW], caspofungin, and sorbitol). (G) Western blot analysis for MpkA phosphorylation in the wild type and ΔrckA, RckA^S481D, and RckA^S481A mutants grown for 16 h in MM and transferred to MM plus 300 μg/ml of Congo red. Signal intensities were quantified using the Image J software by dividing the intensity of phosphorylated MpkA (MpkA∼P) by MpkA.

FIG 9

Phosphomutants for MpkB and SrrB serine phosphorylation sites. (A to C) Phenotypes of the wild type and ΔmpkB, MpkB^Y183F, and SrrB^{S1466–1469A} mutants grown for 5 days at 37°C on minimal medium (MM) or MM plus cell wall-damaging or osmotic stress agents (Congo red [CR], calcofluor white [CFW], caspofungin, and sorbitol). (D) Western blot analysis for MpkA phosphorylation in wild type and ΔmpkB, MpkB^Y183F, and SrrB^{S1466–1469A} mutants grown for 16 h in MM and transferred to MM plus 300 μg/ml of Congo red. Signal intensities were quantified using the Image J software by dividing the intensity of MpkA∼P by γ-tubulin.

FIG 10

Differential phosphorylation of MAPK pathways during response to cell wall damage. The wild-type, ΔsakA, ΔmpkC, and ΔsakA ΔmpkC strains were exposed to CR for 30 min. Phosphorylated (green boxes) or dephosphorylated (red boxes) proteins in the presence of CR are shown in the different MAPK pathways. Superscript names correspond to the respective homologue genes in S. cerevisiae. Proteins with no phosphorylation modulation are represented with the S. cerevisiae corresponding name, adapted from the MAPK signaling pathway, yeast, Aspergillus fumigatus, in KEGG (Kyoto Encyclopedia of Genes and Genomes [https://www.genome.jp/kegg/]).