Deletion of core septin gene aspB in Aspergillus fumigatus results in fungicidal activity of caspofungin

Abstract

Septins are a family of GTP-binding proteins found in many eukaryotic lineages. Although highly conserved throughout many eukaryotes, their functions vary across species. In Aspergillus fumigatus, the etiological agent of invasive aspergillosis, septins participate in a variety of processes such as cell wall organization of conidia, septation, and response to cell wall stress. Previous studies determined that the ΔaspB strain had a greater sensitivity to anti-cell wall drugs, especially the echinocandin caspofungin, yet mechanisms behind this augmented sensitivity are unknown. We performed cell viability staining of the deletion strains post-caspofungin exposure and found that the ΔaspA, ΔaspB, and ΔaspC strains have significantly lower cell viability. Concomitant with the reduced viability, deletion strains are more susceptible to caspofungin on solid media. These results indicate that the septin cytoskeleton is important for A. fumigatus survival in the presence of caspofungin. Due to the potential of improved therapeutic outcome, we followed up using a neutropenic murine model of invasive aspergillosis. Animals infected with the ΔaspB strain and treated with caspofungin showed improved survival compared to the animals infected with akuB^KU80 wild-type or complemented strains. Additionally, histological analysis showed reduced fungal burden and inflammation in the ΔaspB infected, caspofungin-treated group. Affinity purification coupled with quantitative proteomics identified proteins involved in the septin-dependent response to caspofungin, includng four candidate interactors involved in cell wall stress response. Deletion of these candidate genes resulted in increased susceptibility to caspofungin and moderately reduced viability post-drug exposure. Taken together, these data suggest that septin AspB contributes to the fungistatic response to caspofungin.

Author Summary:

Invasive aspergillosis is a pulmonary disease caused by the fungus Aspergillus fumigatus that primarily occurs in immunocompromised patients. Invasive aspergillosis has a high mortality rate, ranging from 50–90%. Therapy options are limited due to few available drugs with fungicidal activity and growing global drug resistance. Treatment typically starts with triazoles, which target the fungal cell membrane. If unsuccessful, an echinocandin, which targets the cell wall, is given as a salvage therapy. Echinocandins, including caspofungin, are fungistatic against A. fumigatus, slowing growth of the fungus rather than killing it. Due to this, echinocandins have a high therapeutic failure rate. Previous work suggests that deletion of the cytoskeletal septin genes increases sensitivity to caspofungin. Here we describe our finding that the septin genes aspA, aspB, and aspC are involved in the fungal response to caspofungin. Additionally, the deletion of aspB results in fungicidal activity of this otherwise fungistatic drug. These findings show promise for novel therapy options that block the septin-mediated response to caspofungin.

Introduction:

Aspergillus fumigatus is a ubiquitous environmental mold responsible for a wide range of opportunistic systemic and allergic pathologies [1]. One such pathology is invasive aspergillosis (IA), an invasive systemic infection most common in immunocompromised patients [2]. Over 300,000 cases of IA are reported yearly, leading to a mortality rate ranging between 30–90% [1,3]. The at-risk patient population for IA is increasing due to increasing numbers of immunomodulating therapies, as well as emerging global diseases [4–6]. Additionally, IA has the highest per-patient cost of any invasive fungal disease, costing the United States an estimated $1.3 billion per year [7].

IA begins when the conidia, or asexual spores, are inhaled by an individual. The conidia are easily aerosolized and are prevalent in both indoor and outdoor environments [8,9]. Due to their small size (~2.5 μm diameter) and hydrophobic nature, the conidia can travel down the airway into the terminal alveoli [10]. In healthy individuals, the mucociliary escalator removes conidia [8]. However, any remaining conidia can then germinate within the lung. Epithelial cells, alveolar macrophages, neutrophils, and other immune cells can detect the germinating conidia, which leads to the release of cytokines and phagocytosis of conidia by immune cells [8]. In immunocompromised patients, the lack of immune responses leads to the progression of IA. Germinated conidia can then establish polarity, leading to the formation of hyphae that invades host tissue. IA can disseminate within the host via the release of hyphal fragments into the bloodstream. Neutropenic patients, such as those on anti-rejection medications post-transplant, are at high risk for IA [11]. In neutropenic patients, disease often presents with abundant hyphal growths, angioinvasion, and intra-alveolar hemorrhage [12,13]. Patients undergoing corticosteroid-induced immunosuppression are also at risk of IA, as glucocorticoids lead to a reduction of pattern recognition receptor (PRR) signaling and inhibit lymphocyte activation [14,15]. Pneumonia, inflammatory necrosis, and minimal hyphal growths are seen in cases of IA in patients under corticosteroid-induced immunosuppression [12,13]. Therapeutic options to effectively treat IA are limited. The frontline treatment for IA is the triazole class of drugs [16]. Triazole treatments are often prolonged and are now common to use as antifungal prophylaxis in at-risk populations. Nonetheless, cases of triazole-resistant A. fumigatus infections are becoming more common. This increase in the incidence of azole resistance led to the inclusion of A. fumigatus as a critical priority in the World Health Organization 2022 Fungal Priority Pathogens List [17].

Echinocandins are a class of antifungals that target the cell wall via β-glucan synthase [18]. In fungi such as C. albicans, echinocandins have a fungicidal effect [19]. In contrast, the echinocandins are fungistatic against A. fumigatus [18]. For immunocompromised patients, the echinocandins are often used as a salvage or secondary therapy in conjunction with triazoles as they cannot reap the benefits of solely fungistatic medications [16]. Slowing the growth of the fungi through fungistatic therapies then requires either the immune system or fungicidal drugs to kill the fungi and clear disease [20,21]. Because of its fungistatic nature, caspofungin use as a salvage treatment had a favorable response rate of only 45% [22]. Understanding the mechanisms within fungi that render these medications fungistatic rather than fungicidal can aid in developing new therapies that improve the efficacy of existing antifungal drugs.

Previous studies have shown that septins AspA, AspB, and AspC contribute to the Aspergillus spp. response to the echinocandin caspofungin [23,24]. Septins are a family of highly conserved eukaryotic GTP-binding proteins. In A. fumigatus, septins are primarily involved in septation, conidiation, and response to cell wall stress [24]. A. fumigatus has five septins: AspA, AspB, AspC, AspD, and AspE [25]. AspA-D are core mitotic septins involved in polymerizing into hexameric and octameric complexes. They are orthologs of Saccharomyces cerevisiae’s Cdc11, Cdc3, Cdc12, and Cdc10, respectively [26]. AspE is a group 5 septin that is present in filamentous fungi and other eukaryotic lineages [27]. Septins also can scaffold proteins involved in stress response pathways. Candida albicans septins and phospholipid PI(4,5)P₂ are regulated in response to cell wall stress induced from caspofungin exposure [28]. They relocalize at specific foci on the plasma membrane to activate the MAPK Mkc1, which then relocalizes chitin synthase and results in chitin deposition in the cell wall [28–30]. Similarly, Aspergillus nidulans septins have been associated with the cell wall integrity (CWI) pathway [23]. Double deletion strains of aspB and CWI pathway MAPK mpkA showed a novel phenotype when grown on caspofungin compared to the respective single deletion strains [23]. These data suggest AspB may have a role in pathways that maintain CWI.

In this study, we found that septins AspA, AspB, and AspC contribute to fungal viability post-caspofungin exposure in A. fumigatus. The deletion of aspB in particular produced a fungicidal response. We also observed a higher chance of survival, reduced lung inflammation, and reduced fungal burden was associated with ΔaspB infected mice treated with caspofungin than ΔaspB strain treated with saline or akuB^KU80 and ΔaspB::aspB strains with either treatment in our neutropenic murine model of IA. To gain deeper mechanistic insights, we conducted a quantitative mass spectrometry-based proteomics analysis to identify candidate AspB-interacting proteins during exposure to caspofungin. Based on our proteomics analysis, we narrowed down six possible proteins involved in the fungal response to caspofungin that were significantly increased (by at least two-fold) upon caspofungin exposure. Gene ontology (GO) analysis indicated that these candidate protein interactors might have a role in cell wall function and organization, suggesting that AspB may mediate cell wall responses through its interactions with these candidates. Deletion strains for each candidate were generated and characterized to obtain an understanding of their biology at basal conditions. Strains were then tested against caspofungin, and four out of the six genes, bgt1, gel2, nsdD, and mapA, were implicated in fungal viability after caspofungin exposure.

Results:

Deletion of aspA, aspB, and aspC Reduces Viability Post-Caspofungin Exposure

Previous work demonstrated the hypersensitivity of A. fumigatus ΔaspB strain and sensitivity of ΔaspC strain to caspofungin on solid media [24]. In order to further confirm and characterize the response of septins to echinocandins, we conducted a spore dilution assay (10⁴-10¹ conidia, 1 μg/mL caspofungin), minimum effective concentration assay (MEC), and E-strip test (10⁶ conidia) to visualize the susceptibility to caspofungin. Deletion of the core septin genes, aspA, aspB, and aspC, resulted in increased sensitivity to caspofungin compared to the akuB^KU80 wild-type strain, ΔaspB::aspB complemented strain, and other septin deletion strains (Fig. 1A, Supplementary Table 4). Similarly, the ΔaspA, ΔaspB, and ΔaspC strains had a clearer and larger zone of effect on the E-strip test (Fig. 1D). To explore whether this effect was exclusive to caspofungin, we conducted an E-strip test to visualize the susceptibility to another drug in the echinocandin class, micafungin. The ΔaspA and ΔaspC strains showed a slight increase in sensitivity to micafungin, while ΔaspB had a more noticeable decrease in growth in the zone of effect (Fig. 1D). Thus, this increase in susceptibility of the ΔaspA, ΔaspB, and ΔaspC strains is not limited to caspofungin, albeit it is more pronounced.

Fig 1.

AspA, AspB, and AspC are involved in fungal response to caspofungin. The deletion of aspA, aspB, and aspC results in reduction of viability to echinocandins. (A) Spore dilution assay show an increase in susceptibility to caspofungin in the ΔaspA, ΔaspB, and ΔaspC strains. 10⁴-10¹ conidia were plated on GMM media supplemented with caspofungin for 48 hours at 37°C. (B, C) Deletion of aspB results in loss of viability when grown in caspofungin but not micafungin. 10⁴ conidia were incubated for 48 hours in GMM supplemented with (B) 1μg/mL caspofungin or (C) 1 μg/mL micafungin. Cells were incubated in CFDA for one hour then visualized. Viable and non-viable colonies were counted. The number of viable colonies was divided by the total number of colonies. Experiment was replicated three times. Error bar represent SEM. Student t-test were done in Graphpad Prism, with each strain compared against the akuB^KU80 wild-type. (D) E-strip plates of caspofungin and micafungin show clearer zone of effect in ΔaspA, ΔaspB, and ΔaspC strains. 10⁶ conidia were plated with beads and left to dry. E-strip was placed, and plates were grown at 37°C for 48 hours. All experiments were replicated three times. Representative images are shown in this figure.

AspB is Required for Fungistatic Response to Caspofungin in vitro

Since we observed an increase in susceptibility in all three of our drug susceptibility assays, we hypothesized that the absence of AspA, AspB, and AspC leads to a fungicidal effect of caspofungin. To determine viability of the deletion strains against caspofungin, we utilized CFDA to determine cell viability post-caspofungin exposure. In basal conditions, 10⁴ conidia were grown on coverslips in GMM liquid media for 24 hours, followed by 1 hour incubation in 1 μg/mL CFDA. All strains are equally viable, indicating that the increased clearance is not due to a reduction in viability of septin deletion strains (Fig. 2A). Strains were then grown for 48 hours in the presence of 1 μg/mL caspofungin, followed by incubation with CFDA. The aspA, aspB, and aspC deletion strains were significantly less viable (p<0.0001) after 48 hours of caspofungin exposure, with only 0.92% of ΔaspB (p<0.0001) being viable (Fig. 1B, 2B). ΔaspB also does not form many microcolonies in the presence of caspofungin. ΔaspA and ΔaspC form microcolonies in the presence of caspofungin; however, only 16.3% and 18.5% of the colonies, respectively, were viable at the time of microscopic examination (p<0.0001)(Fig. 1B, 2B). We repeated this assay with the echinocandin micafungin in order to test if this reduction in viability was specific to caspofungin or a general response to echinocandin exposure. In contrast to caspofungin treatment, the majority of colonies from all strains were viable after micafungin treatment (Fig. 1C). This indicates that the reduction in viability is specific to caspofungin.

Fig 2.

ΔaspB strain loses cell viability when grown in caspofungin. (A) All strains had viability when grown in GMM+UU. 10⁴ conidia were grown in 4 mL GMM+UU at 37°C for 24 hours. Cells were then treated with CFDA, which is hydrolyzed in living cells to a fluorescent ester, then visualized. (B) Loss of viability in ΔaspB strain is seen after growth in the presence of caspofungin. After 48 hours post exposure to 1 μg/mL caspofungin, cells were incubated in CFDA for one hour then visualized. Experiments were replicated three times. Scale bar is 500 μm.

AspB is Involved in Response to Caspofungin in Mature Mycelium

It is possible that most of these phenotypes that we observe are due to caspofungin acting as the mycelia germinates, as exemplified by the small microcolonies in the ΔaspB strain. Mature mycelia would be a more clinically relevant growth stage, and for this reason, we decided to determine the effect of caspofungin on mature mycelium. To test this, we used propidium iodide (PI) stain to determine hyphal damage. Strains were grown in GMM+UU for 24 hours at 37°C before being stained with PI solution and visualized. No hyphal damage was seen under basal conditions (Fig. 3A). Following the same procedure, we grew strains then incubated them with caspofungin (1 μg/ml) for 2 hours. Plates were washed and stained as before, then visualized. Similar to our viability assays, only the ΔaspB strain exhibits extensive hyphal damage after 2 hours of exposure to caspofungin (Fig. 3B). Taken together, these results suggest that AspB is needed for fungal response to caspofungin.

Fig 3.

ΔaspB strain has increased hyphal damage during caspofungin exposure. (A) No hyphal damage seen in basal conditions. 10⁴ conidia were grown in GMM at 37°C for 24 hours. Cells were then washed with PIPES (pH 6.7) for 5 minutes twice. Slides were then treated with propidium iodide (PI) solution, which stains nucleic acids. Coverslips were washed twice with PIPES, then prepared and visualized. (B) Hyphal damage was seen in ΔaspB strain treated with caspofungin. 10⁴ conidia were grown in GMM at 37°C for 24 hours. Coverslips were then incubated in caspofungin for 2 hours at 37°C. Cells were then washed with PIPES and treated with PI solution. Experiments were replicated three times. Scale bar is 500 μm.

Deletion of aspB improves caspofungin treatment efficiency in a neutropenic murine model of invasive aspergillosis

As we observed a strong fungicidal effect against the ΔaspB, we decided to determine if deletion of aspB led to improved survival in our murine model of invasive aspergillosis. As the ΔaspB strain has no discernable difference in virulence compared to the akuB^KU80 and ΔaspB::aspB strains with respect to fungal burden and lung inflammation, any effect on survival would be attributed to the role that AspB plays in fungal response to caspofungin [24]. To test this hypothesis, neutropenia was induced in 6-week old male CD-1 mice using 175 mg/kg cyclophosphamide and 40 mg/kg triamcinolone acetonide. Neutropenic mice were then intranasally infected with 4×10⁶ conidia of akuB^KU80, ΔaspB, and ΔaspB::aspB strains. They were subsequently treated with either 2 mg/kg caspofungin or an equivalent volume of the saline vehicle. All groups of mice treated with saline all had the first death on day 3 post-infection (Fig. 4A). Mice inoculated with ΔaspB strain and treated with caspofungin had a 70% probability of survival (p<0.0001)(Fig. 4B). Additionally, in this group the first death did not occur until day 12 post-infection. Compared with the next highest groups, ΔaspB strain treated with saline and akuB^KU80 treated with caspofungin had a 30% probability of survival (Fig. 4A). Lung histology was then performed to visualize inflammation and fungal lesions. Concurrent with the survival graph, animals infected with the ΔaspB strain treated with caspofungin had reduced inflammation and stunted fungal growth compared to animals from the other groups (Fig. 4D). Previous infection models also showed no difference in inflammation or fungal lesions between untreated akuB^KU80, ΔaspB, and ΔaspB::aspB strains [24]. This suggests the caspofungin treatment reduces growth or potentially has a fungicidal effect on the fungus without AspB in vivo, as well.

Fig 4.

The ΔaspB strain treated with caspofungin has reduced inflammation and fungal burden in immunosuppressed intranasal murine model of invasive aspergillosis. (A-B) Mice infected with ΔaspB strain and treated with caspofungin had a higher chance of survival. Ten mice per strain per condition were infected intranasally with 4×10⁶ conidia. Mice were treated with either saline (A) or caspofungin (B) on days +1 through +4 post-infection. Mice were monitored at least twice a day for 14 days. Survival is on a Kaplan-Meier curve with log rank pair-wise comparison (p<0.0001). (C) H&E stain of lungs after 3 days post-infection shows that there is a decrease in inflammatory response in ΔaspB infected mice treated with caspofungin. Silver stain shows less and smaller fungal lesions in ΔaspB infected mice treated with caspofungin. H&E images taken on an Echo Rebel Hybrid Light microscope using a 10x objective. Scale bar is 300 μm. Silver stain images taken on an Inverted Leica DMi8 with Leica K5 Microscope Camera using a 40x objective. Scale bar is 100 μm.

AspB Interactome Changes Post Caspofungin Exposure

Previous work determined that septins’ localization is altered by exposure to caspofungin in C. albicans and A. fumigatus [24,28]. Additionally, the protein interactome of septin AspB was altered after exposure to caspofungin in a qualitative proteomic experiment [31]. This work only detected for the presence or absence of interactant proteins between basal and caspofungin conditions, potentially missing interactions that occur in both conditions but change in abundance [31]. To gain a more mechanistic insight into how AspB contributes to the fungal response to caspofungin, we applied affinity purification coupled with quantitative proteomics. An AspB-eGFP expressing strain was grown in GMM and GMM supplemented with 1 μg/mL caspofungin for 24 hours. AspB-GFP was then purified using a GFP-Trap^® affinity matrix, and proteins bound to AspB-GFP were prepared and analyzed by LC-MS/MS to identify protein interactors in each condition. Principal Component Analysis (PCA) indicates that the GMM- and caspofungin-AspB interactome are distinctive from each other (Fig. S1A). A total of 226 proteins were significantly decreased (fold change (FC) < −2, p < 0.05) and 106 proteins were significantly increased (FC > 2, p < 0.05) upon caspofungin treatment (Fig. S1B). Among the proteins increased upon caspofungin exposure was PpoA, a characterized fatty acid monooxygenase. PpoA was also identified by a previous study investigating the AspB interactome post-caspofungin exposure, indicating that our analyses were able to confirm previously described caspofungin-specific AspB interactions [31]. Candidate protein interactors that met the screening criteria were analyzed using FungiFun, a gene ontology tool, to assess changes in biological processes [32]. Proteins that are known to be involved in hyphal growth were overrepresented among the list of proteins increased upon caspofungin treatment (Fig. S1C). In contrast, proteins that are known to be involved in protein folding were overrepresented in the list of proteins decreased upon caspofungin treatment (Fig. S1D). Using this information, we selected six genes that were significantly increased by at least two-fold (FC > 2, p < 0.05) after caspofungin exposure and had a role or putative role in cell-wall related functions for further investigation (Table S3).

Δbgt1, Δgel2, ΔnsdD, and ΔmapA Have a Reduction in Conidiation, While ΔnsdD and ΔpunA Have a Growth Defect

Based on our proteomics and gene ontology analyses, we performed the deletion of all six candidate genes (Table S3) to determine their role in septin-related phenotypes. We first characterized each deletion strain to understand how the gene affects growth during basal conditions. Major and biologically relevant radial growth defects are seen in ΔnsdD (37.8 mm average diameter compared to the 85 mm diameter of the akuB^KU80 strain) and ΔpunA (24.3 mm average diameter compared to the 57.3 mm average diameter of the akuB^KU80 strain at 30°C)(Fig. 5A, 5D, Fig.S2). The slight reduction of radial growth in ΔaspB was also seen in a previous study and deemed not biologically relevant, which is also the case for the Δgel2 strain [24]. Similarly, a previously noted defect in conidiation was seen in ΔaspB [24]. Δbgt1, Δgel2, ΔnsdD, and ΔmapA also have reduced conidiation, but not as severely as ΔaspB (Fig. 5B, 5E). As prior work noted that ΔaspB strain had delayed septation, we were interested in determining if any of the candidate gene deletion strains were also defective in septation [24]. To test this, we measured apical compartment length as an indirect method of measuring potential septation defects. Δbgt1 mutants have a similar increase in apical compartment length to ΔaspB, suggesting a delay in septation (Fig. 5C). In contrast, Δgel1 mutants have apical compartments similar to the akuB^KU80 wild-type strain (Fig. 5C), suggesting that Gel1 is dispensable for septa formation. Δgel2 (70.4 μm) and ΔmapA (67.7 μm) mutants exhibit hyperseptation. ΔnsdD (37.7 μm) has a more drastic reduction in apical compartment length, but this phenotype may be due to its growth defect (Fig. 5C).

Fig 5.

ΔnsdD and ΔpunA strains have a growth defect; Bgt1, Gel2, NsdD, and MapA are involved in conidiation; and Δbgt1, Δgel2, ΔnsdD, and ΔmapA strains have septation defects. (A, D) ΔnsdD and ΔpunA have a significant defect in radial growth after 5 days incubation. 10⁴ total conidia of each strain were placed on GMM agar. Plates were incubated for 5 days and radial growth was measured every 24 hours. Experiments were replicated three times. (B, E) Deletion of aspB, bgt1, gel2, nsdD, or mapA results in reduced conidial production. ΔpunA has increased conidial production. Conidia were collected on day 5 of growth on GMM media using 10 mL of 0.05% Tween-80. Conidia were counted using a hemocytometer and values were divided by the total area of growth to account for strains with growth defects. Experiments were replicated three times. (C) ΔaspB and Δbgt1 have an increase in length of the apical compartment. Δgel2, ΔnsdD, and ΔmapA have a decrease in length of the apical compartment. 10⁴ total conidia of each strain were inoculated onto coverslips immersed in GMM and incubated for 15 hours. Coverslips were stained with aniline blue and visualized. The length of the apical compartments (N=20) was measured using imageJ. (A, B, C) Experiment performed at 37°C. (D, E) Experiment performed at 30°C. Error bar represent SEM. One-way ANOVA with Tukey’s multiple-comparison test were performed in Graphpad Prism and were declared significantly different at a p-value of <0.05. Group means with different lowercase letters are significantly different.

Δbgt1, Δgel2, ΔnsdD, and ΔmapA Have Increased Sensitivity to Caspofungin Exposure

Since the candidate proteins showed increased interaction with AspB during caspofungin exposure, we explored their potential role in mediating the AspB-dependent fungal response to caspofungin. To test this, we conducted a spore dilution assay on GMM agar supplemented with 1 μg/mL caspofungin. 10⁴, 10³, 10², and 10¹ spores were plated and incubated at either 30°C or 37°C for 2 days. Deletion of the aspB gene displayed colony-level growth defects in the 10³, 10², and 10¹ concentrations (Fig. 6A, 6C). No other deletion strains displayed as severe of a sensitivity to caspofungin. Δbgt1, Δgel2, ΔnsdD, and ΔmapA lost full colony growth at 10² and 10¹ concentrations (Fig. 6A). In contrast, Δgel1 and ΔpunA are similar to the akuB^KU80 wild-type and ΔaspB::aspB complemented strains (Fig. 6A, 6C). The minimum effective concentration (MEC) of caspofungin for Δbgt1, Δgel2, ΔnsdD, and ΔmapA was determined to be lower than that of the wild-type, and equal to or lower than that of ΔaspB (Table S4). ΔpunA was not determined as CLSI standards require 37°C, and this strain cannot grow at that temperature.

Fig 6.

Deletion of Bgt1, Gel2, NsdD, and MapA increases susceptibility to caspofungin. (A, C) Spore dilution assays show Δbgt1, Δgel2, ΔnsdD, and ΔmapA strains have an increase in susceptibility to caspofungin. Conidia (10⁴-10¹) were grown on GMM media supplemented with 1 μg/mL caspofungin for 48 hours. (B, D) Deletion of bgt1, gel2, nsdD, and mapA results in reduced, but not loss of, viability when grown in caspofungin. Deletion of gel1 and punA does not affect viability when exposed to caspofungin. 10⁴ conidia were grown in GMM for 24 hours then incubated in CFDA and visualized. Experiment was replicated three times. Error bar represent SEM. One-way ANOVA with Tukey’s multiple-comparison test were performed in Graphpad Prism and were declared significantly different at a p-value of <0.05. Group means with different lowercase letters are significantly different. (A, B) Experiment performed at 37°C. (C, D) Experiment performed at 30°C.

Reduced Viability of Δbgt1, Δgel2, ΔnsdD, and ΔmapA Post-Caspofungin Exposure

Since aspB deletion leads to a fungicidal response to caspofungin, we investigated whether any candidate gene deletion strains also exhibited reduced viability. All strains showed equal viability in basal conditions (Fig. 7A, 7C). We then grew all strains for 48 hours in the presence of 1 μg/mL caspofungin and determined their viability with CFDA. The aspB deletion strain showed nearly no viability at both 30°C and 37°C, as previously shown in Figure 1B (Fig. 6B, 6D and Fig. 7B, 7D). No other strain demonstrated the complete loss of viability phenotype observed in the ΔaspB. The Δbgt1, Δgel2, ΔnsdD, and ΔmapA strains form microcolonies in the presence of caspofungin like the akuB^KU80 strain, but they had a reduction in viability (Δbgt1=37.4%, Δgel2=39.5%, ΔnsdD=27.8%, ΔmapA=26.1% compared to akuB^KU80 = 90.4%)(Fig 6B and Fig.7B). Δgel1 and ΔpunA did not have a reduction in cell viability when grown in caspofungin (Fig. 6B, 6D and Fig. 7B, 7D).

Fig 7.

Δbgt1, Δgel2, ΔnsdD, and ΔmapA strains have reduced cell viability when grown in caspofungin. (A, C) All strains were fully viable when grown in GMM media. Conidia (10⁴) were grown in GMM for 24 hours on coverslips then incubated in CFDA. Slides were prepared and visualized on an Inverted Leica DMi8 with Leica K5 Microscope Camera using a 10x objective. (B, D) Candidate deletion strains did not lose total viability, but strains Δbgt1, Δgel2, ΔnsdD, and ΔmapA had a partial reduction in viability. Conidia (10⁴) were grown on coverslips immersed in GMM supplemented with 1 μg/mL caspofungin for 48 hours. Coverslips were then incubated in CFDA and visualized. (A, B) Experiment performed at 37°C. (C, D) Experiment performed at 30°C. All experiments were replicated three times. Scale bar is 500 μm.

Bgt1, Gel2, NsdD, and MapA are Involved in the Response to Caspofungin in Mature Mycelium

Next, we determined if any of our candidate gene deletion strains were susceptible to hyphal damage when exposed to caspofungin as mature mycelia. To visualize hyphal damage, we utilized a propidium iodide (PI) stain. No hyphal damage was observed in basal conditions, indicating that there is no defect in the cell wall at basal conditions (Fig. 8A, 8C). The experiment was then repeated with a 2-hour incubation in GMM supplemented with 1 μg/mL caspofungin prior to PI staining. Extensive hyphal damage was seen in Δbgt1, Δgel2, ΔnsdD, and ΔmapA strains similar to ΔaspB after exposure to caspofungin (Fig. 8B). Taken together, Bgt1, Gel2, NsdD, and MapA are involved in the fungal response to caspofungin in earlier stages of growth and mature hyphae.

Fig 8.

Δbgt1, Δgel2, ΔnsdD, and ΔmapA strains have increased hyphal damage during caspofungin exposure. (A, C) Hyphal damage was not observed in basal conditions. Conidia (10⁴) were grown on coverslips in GMM for 24 hours. Coverslips were washed in PIPES (pH 6.7) for 5 minutes twice then treated with propidium iodide (PI) solution. Coverslips were washed twice with PIPES, then visualized on an Inverted Leica DMi8 with Leica K5 Microscope Camera using a 10x objective. (B, D) Strains Δbgt1, Δgel2, ΔnsdD, and ΔmapA had hyphal damage when treated with caspofungin. Conidia (10⁴) were grown on coverslips in GMM for 24 hours, then incubated in caspofungin for 2 hours. Cells were then washed with PIPES and incubated with PI. (A, B) Experiment performed at 37°C. (C, D) Experiment performed at 30°C. Experiments were replicated three times. Scale bar is 500 μm.

Candidate Genes are not involved in the response to other Cell Wall Disrupting Agents or in the Caspofungin Paradoxical Effect

We further characterized the candidate gene deletion strains sensitivities to other cell wall disturbing agents. To test this, we plated conidia on GMM agar and GMM supplemented with either 1 μg/mL caspofungin, 100 μg/mL Congo red, 2 μg/mL nikkomycin Z, 5 μg/mL calcofluor white, or 10 μg/mL calcofluor white and incubated for three days. Additionally, we plated 4 μg/mL caspofungin and incubated for five days to observe whether the deletion strains were still capable of the caspofungin paradoxical effect. After incubation, ΔaspB shows increased susceptibility to caspofungin and Congo red, and slight increase in susceptibility to nikkomycin Z and calcofluor white (Fig. S3). We observed Δgel1 had a mild increase in susceptibility to Congo red compared to the akuB^KU80 wild-type (Fig. S3). Δbgt1, Δgel2, and ΔnsdD had a mild increase in susceptibility to 10 μg/mL calcofluor white, but not as pronounced as ΔaspB (Fig. S3). ΔpunA visually appeared to have a reduction in conidiation and abnormal colony morphology when grown on nikkomycin Z, but no differences in growth when compared to the wild type strain (Fig. S4). Taken together, these candidate genes might mediate fungal response to caspofungin while not being a part of a general fungal cell wall stress response.

Discussion:

The septins are a highly conserved family of cytoskeletal proteins with a variety of cellular roles. This includes roles in cell division, stress response, cytoskeleton organization, and scaffolding [24,33–35]. Previous work noted the role of septins in response to cell wall stress. In C. albicans, septins are required for proper cell wall regulation and mislocalize during exposure to caspofungin [29,36]. Additionally, septins also mislocalized during basal conditions in deletion strains of protein kinases involved in cell wall regulation, including Gin4, Kin3, Vps34, and Cbk1 [36]. In A. nidulans, C. albicans, and S. cerevisiae, chitin synthases are dependent on septins for proper localization [29,37,38]. Previous work in A. fumigatus determined that the septins have a role in response to cell wall stress, notably showing that ΔaspB exhibits a hypersensitive response to caspofungin exposure on solid media [24]. Here, we aimed to further investigate the septin response to caspofungin. We found that deletion of aspA, aspB, and aspC showed a reduction in fungal viability post-caspofungin exposure, while deletion of aspB elicited a fungicidal response. The deletion of aspB also increased hyphal damage in mature mycelia. The ΔaspB response to caspofungin was also recapitulated in an in vivo murine model, with ΔaspB infected mice showing higher survival rates, reduced inflammation, and delayed mortality upon treatment with caspofungin.

The Aspergillus spp. septins form heteropolymers consisting of hexamers (AspA-C) and octamers (AspA-D) [39]. Deletion of aspD still allows proper formation of hexamers but not octamers. In our work, we found that aspA, aspB, and aspC deletion strains had reduced viability to caspofungin (Fig. 1). While the aspD deletion strain was slightly more sensitive than the akuB^KU80 wild-type, ΔaspB::aspB, or ΔaspE strains, its increased susceptibility was not as pronounced as that of the aspA, aspB, and aspC deletion strains. This suggests that the presence of the hexamer is at least sufficient, if not necessary, for proper fungal response to caspofungin. It is unknown if only the sole presence of the octamer would also be sufficient, as deletion of aspA-C prevents the formation of both structures. Similarly, ΔaspB strain had a hypersensitive to fungicidal response to clinically relevant doses of caspofungin (Fig. 1), more drastic than the one observed in the ΔaspA or ΔaspC strains. It is possible that AspB specifically is needed to scaffold necessary proteins in response to caspofungin, but the loss of both the hexamer and octamer causes general dysfunction that would lead to the phenotypes seen in ΔaspA or ΔaspC. Proteomics analysis in Cryptococcus neoformans also supports the possibility of subunit-specific protein interactions, where the cdc3^aspB and cdc10^aspD septins are shown to have differences in interactomes after exposure to cellular stressors [40]. All of the septin deletion strains retain the caspofungin paradoxical effect, suggesting that individual septins and their heteropolymers act independently of the pathways that contribute to the caspofungin paradoxical effect.

We found that the septin response to echinocandins is not limited to caspofungin, but the fungicidal response of ΔaspB is specific to caspofungin. The echinocandin class includes four clinically approved drugs: anidulafungin, micafungin, caspofungin, and most recently approved, rezafungin [41,42]. These drugs all work through noncompetitive inhibition of the β−1,3-D-glucan synthase, but alterations in their side chains comprise the major differences between each one [18,43]. These changes in the side chains can lead to changes in how the drug impacts fungal physiology. For instance, caspofungin is the only echinocandin to induce the paradoxical effect [43]. Unlike micafungin, high concentrations of caspofungin increase the levels of cytosolic calcium [44,45]. This activates calmodulin-calcineurin signaling and leads to the caspofungin paradoxical effect [45]. Thus, our findings add to the growing evidence that even though the different echinocandins have a shared mechanism of action, how the fungus responds to echinocandins is drug specific.

Hyphal damage was not observed in any of our strains, indicating that there is no underlying cell wall defect that would cause damage in basal conditions. However, after 2-hour exposure to caspofungin, the ΔaspB strain showed significant hyphal damage. This suggests that AspB is involved in the fungal cell wall response to caspofungin in mature mycelium. In C. albicans, caspofungin has a fungicidal effect. Phosphatidylinositol-(4,5)-bisphosphate and septin regulation are needed for proper cell wall stress response [28]. Disruption of either party causes hypersusceptibility to caspofungin [28,29]. Based on our analyses, AspB is needed for proper cell wall response to caspofungin in A. fumigatus. Previous proteomic analyses and our own proteomics experiments show that MpkA, MkkA, and other cell wall integrity kinases co-immunoprecipitated with AspB [31]. It is possible that AspB interacts with the cell wall integrity pathway in A. fumigatus to facilitate fungal response to caspofungin [31]. In A. nidulans, deletion of aspB partially rescues both the growth defect and susceptibility to caspofungin of the ΔmpkA strain, further hinting a possible crosstalk between the septin and the CWI [23]. Further work will be needed to investigate the possible role of AspB in the cell wall integrity pathway.

Animals infected with the ΔaspB strain and treated with caspofungin had a 70% chance of survival, which is higher than that of animals infected with akuB^KU80 strain (30%). The first death of ΔaspB strain infected mice treated with caspofungin was recorded on day +12 post infection, compared to ΔaspB strain infected mice treated with saline or the other strains and conditions with first deaths on day +3 or day +4. It is possible that colonies that persist under caspofungin treatment are capable of growing and establishing infection after the drug pressure is released. Our histological analyses showed reduced inflammation and reduced fungal lesions in the lungs of ΔaspB infected mice treated with caspofungin. However, this assay is done on day +3 after infection when mice are still being treated with caspofungin. Nonetheless, this increase in survival makes AspB a prospective target for drug therapy developments, as use in conjunction with caspofungin elicits a fungicidal response in A. fumigatus. Currently, only the plant cytokinin forchlorfenuron (FCF) is known to disrupt septin organization [46]. This compound disrupts septins within both fungal and mammalian cells by interfering with their ability to bind and hydrolyze GTP [46–48]. It is possible that FCF could be modified and refined to target fungal-specific septins. FCF analogues have previously been developed to better target ovarian and endometrial cancers [49]. Thus, FCF analogues could be used in conjunction with caspofungin to create a new fungicidal therapy for treatment of IA. Further work is required to better understand the mechanisms of AspB’s response to caspofungin and to develop novel clinical therapies.

Our proteomic analysis uncovered six candidate protein interactors of septin AspB that were significantly increased upon caspofungin treatment and that have a potential role in cell wall functions. Bgt1, Gel1, and Gel2 are glucanosyltransferases in A. fumigatus [50]. These glucanosyltransferases work to remodel unorganized β−1,3-glucan chains in the periplasmic space to stabilize and modify the cell wall as needed [50]. Due to their role in cell wall homeostasis, previous work identified these three glucanosyltransferases as potential drug targets that required further investigation [50]. Bgt1 works by hydrolyzing a β−1,3-linked oligosaccharide and placing it on another molecule of β−1,3-glucan [51]. In C. albicans, deletion of BGL2, the bgt1 homologue, resulted in an increase in susceptibility to nikkomycin Z, a chitin synthase inhibitor [52]. In contrast, A. fumigatus Δbgt1 only has a slight increase in susceptibility to nikkomycin Z (Fig. S3). Deletion of BGL2 in S. cerevisiae increased the cellular chitin content [53]. In A. fumigatus, though, Δbgt1Δbgt2 double mutants do not have a change in chitin content [51]. A slight reduction in conidiation was observed in A. fumigatus Δbgt1 (Fig. 5A, 5B). A. fumigatus Δbgt1 susceptibility to calcofluor white and Congo red were similar to the akuB^KU80 wild-type, although compared to basal conditions there was a reduction in growth (Fig. S3) [54]. A potential mechanism for the increased susceptibility in A. fumigatus Δbgt1 and ΔaspB to caspofungin is that hyphal damage may be exacerbated by the lack of interaction between AspB and Bgt1 when either party is missing (Fig. 6B). Due to the role of septins in septation and scaffolding, the loss of AspB in particular may be more deleterious than just the loss of Bgt1 and thus result in the observed fungicidal effect (Fig. 1B). Further work is needed to uncover the relationship between glucanosyltransferase Bgt1 and the septins in septation and cell wall stress responses.

Gel1 and Gel2 are two GPI-anchored glucanosyltransferase members of the Glycoside Hydrolase Family 72 (GH72) [55]. Despite having the same enzymatic activity, deletion of gel1 did not result in a unique phenotype, but reduced mycelial growth and abnormal cell wall architecture was observed in Δgel2 during basal conditions (Fig. 5A–B) [50,51,56]. An increase in chitin is associated with deletion of gel2, a compensatory mechanism similar to deletion of bgt1 [56]. A. fumigatus Gel1 is dispensable for proper fungal response to caspofungin (Fig. 6 A–B). In contrast, Gel2 does appear to be involved in the response to caspofungin due to increased susceptibility to caspofungin and reduced viability in caspofungin compared to akuB^KU80 wild-type (Fig. 6A–B). Our proteomics analysis noted an increased interaction between Gel2 and AspB, but cellular levels of Gel2 were not explored. It is unknown if cellular levels of Gel2 are increased in A. fumigatus post-caspofungin exposure similar to C. albicans. Additionally, further work exploring overexpression of Gel2 may uncover a new mechanism of resistance to caspofungin in A. fumigatus.

NsdD (never in sexual development) is a putative GATA-type transcriptional activator with roles in A. nidulans for sexual development [57]. NsdD is responsible for activating sexual development [58]. Deletion of nsdD results in the inability to produce cleistothecia (fruiting bodies), while overexpression allows production of Hülle cells even in conditions which typically block sexual development [58]. Despite being a primarily asexual fungus, A. fumigatus is capable of and still holds a variety of functional sexual reproductive genes [59]. NsdD has been implicated in cell wall remodeling and hyphal fusion [60]. Deletion of nsdD results in reduced hyphal growth on minimal media (Fig. 5B) [57,60]. Additionally, mating done with ΔnsdD resulted in impaired heterokaryon formation [60]. Drug challenges show some sensitivity of ΔnsdD to Congo red and caspofungin (Fig. 6A–B, Fig. S3) [60]. Previous work also noted weakened hyphal tips in ΔnsdD, which is in accordance with the increased hyphal damage during cell wall stress seen via PI staining (Fig. 8B) [60]. We suspect that while NsdD may have independent roles in sexual development, it is possible that previous cell wall remodeling functions of NsdD seen during hyphal fusion for heterokaryon formation may have alternative functions in A. fumigatus cell wall remodeling outside of sexual development. Further work is required to understand the NsdD-mediated response to caspofungin exposure.

MapA (gene Afu4g06930) is a previously uncharacterized protein in A. fumigatus. In S. cerevisiae and C. albicans, cytosolic protein MAP2 is an ortholog that functions as a methionine aminopeptidase. Deletion of MAP2 results in slightly slower growth rate and some chemical sensitivities in S. cerevisiae and C. albicans [61,62]. It is possible that this methionine aminopeptidase can remove the methionine to promote maturation of protein involved in response to caspofungin.

PunA (gene Afu6g07470) is a previously uncharacterized protein in A. fumigatus. In yeast, PUN1 is a Sur7 family tetraspan that localizes in eisosomes/MCC (membrane compartment of Can1) [63,64]. These punctate membrane domains have roles in stable furrowing and promote stress resistance, including detecting and responding to changes in cell wall integrity [63]. Deletion of PUN1 resulted in decreased thermotolerance, abolished filamentous growth, reduced cell wall components, and decreased metal ion tolerance [65–67]. This decreased thermotolerance was also observed in our A. fumigatus ΔpunA strain, thus requiring the growth of ΔpunA and comparison strains at 30°C. Yeast deletion strains of SUR7, a closely related protein, saw little sensitivity to caspofungin but high sensitivity to the triazole fluconazole [68]. Interestingly, a previous study in A. fumigatus noted an increase in gene expression of punA post-exposure to the triazole voriconazole [69]. A. fumigatus ΔpunA had a similar phenotype to the akuB^KU80 wild-type when exposed to caspofungin or other cell wall stressors (Fig 6C–D, Fig. S4). It is possible that A. fumigatus PunA contributes to the fungal response to non-cell wall stressors, such as metal ion stress or membrane targeting drugs. It is also possible that PunA might contribute with septin localization post-caspofungin exposure; however, this possible role in septin localization does not contribute to A. fumigatus susceptibility to caspofungin.

There is increasing evidence that the septins are involved in the mediation of cell wall stress [23,24,28,29,31]. In this work, we demonstrate that the septin hexamer is sufficient for response to caspofungin and the deletion of the septin gene aspB results in a fungicidal effect. This phenotype continues in vivo, suggesting potential therapeutic applications. We also identified four AspB interactors that are involved, but not required, in the fungal response to caspofungin. Overall, we suggest that the deletion of aspB, and thus loss of all AspB’s interactions, results in the fungicidal effect of caspofungin demonstrated. Future studies directed towards developing drugs specific to fungal septins or in determining additional septin interactants that have key roles in the caspofungin response can help to better understand ways in which caspofugnin therapy can be improved for use against invasive aspergillosis.

Materials and Methods:

Strains, media, and conditions:

A. fumigatus akuB^KU80 served as the wild-type and control strain. ΔaspB::aspB complemented strain served as an additional control to determine that phenotypes were due to the deletion of the aspB gene. Septin deletion strains are described in (Table S1, [24,70]). The aspB-egfp strain from [24], with aspB-egfp expressed by the aspB promoter from the native loci, was used to pulldown AspB interactants. All cultures were plated on glucose minimal media (GMM) supplemented with 5 mM uracil and 5 mM uridine (GMM+UU) and incubated at 37°C, unless otherwise specified.

Drug susceptibility assays:

Spores were counted, diluted, and inoculated in rows onto agar plates containing caspofungin (1 μg/mL) in 10μL quantities with a total of 10⁴, 10³, 10², and 10¹ conidia. Plates were incubated for 48 hours at 37°C unless otherwise noted. For the minimum effective concentration assay, a quantity of 2.5×10⁴ spores were added to RMPI media with different concentrations of caspofungin and incubated for 48 hours at 37°C [71]. We were not able to test the aspD and punA deletion strains according to CLSI guidelines as ΔaspD, which is an uracil/uridine auxotroph, could not grow in RMPI, and ΔpunA, which has a temperature sensitivity, could not grow at 37°C.

Fungal Viability Assay:

Conidia of akuB^KU80, ΔaspA-E, and ΔaspB::aspB were diluted to 10⁴ spores and cultured on coverslips immersed in 4 mL of GMM+UU broth and GMM+UU+caspofungin (1 μg/ml) and incubated at 37°C for 24 and 48 hours, respectively. To examine cell viability, coverslips were incubated in 5-carboxyfluorescein diacetate (CFDA) (50 μg/mL 0.1 M MOPS pH 3) for 1 hour at 37°C and 70 rpm. Slides were prepared for microscopy and image on an Inverted Leica DMi8 with Leica K5 Microscope Camera using a 10x objective. Images were analyzed using imageJ [72]. Viable colonies were quantified by counting colonies manually and dividing fluorescent colonies over total colonies counted. A minimum of 50 colonies per strain per replicate were counted.

Hyphal damage assay:

Conidia (10⁴) of akuB^KU80, ΔaspA-E, and ΔaspB::aspB were cultured on coverslips immersed in 4 mL of GMM+UU broth and incubated for 24 hours at 37°C. Coverslips were then incubated with GMM+UU or GMM+UU+caspofungin (1 μg/ml) for 2 hours at 37°C. To examine the damage of mature hyphae, coverslips were washed with 4 mL PIPES (pH 6.7) for 5 minutes. PIPES was then removed and washed again with 4 mL PIPES for 5 minutes. After removing the second PIPES wash, 500 uL of propidium iodide (PI) solution (12.5 ug/mL in 50 mM PIPES) was added on top of the coverslip and let incubate in the dark for 5 minutes. Slides were washed in 4 mL PIPES twice as described previously. They were then prepared for microscopy and imaged on an Inverted Leica DMi8 with Leica K5 Microscope Camera using a 10x objective. Images were analyzed in imageJ [72].

Neutropenic murine model of Invasive Aspergillosis:

Murine experiments followed previously established intranasal neutropenic models of IA [73,74]. Sixty 6-week old male CD1 mice (Charles River Laboratories, Raleigh, NC) were injected via intraperitoneal route with 175 mg/kg cyclophosphamide on days −2 and +3 and 40mg/kg triamcinolone acetonide subcutaneously on days −1 and + 6. Twenty mice per strain were infected intranasally with 40 μl of 10⁸ spores/ml conidia suspension of the akuB^KU80, ΔaspB, or ΔaspB::aspB strains on day +0. On days +1 through +4, mice were injected via intraperitoneal route with either 2 mg/kg caspofungin or saline. Mice were monitored until day +14 and humanely euthanized if they showed severe symptoms. Survival was plotted on a Kaplan-Meier curve with a log rank pair-wise comparison. Murine experiments were conducted in compliance with the SIU Institutional Animal Care and Use Committee Protocol 20–034.

Histopathology Analysis:

Mice were immunocompromised and treated as described above in [Section: Neutropenic murine model of Invasive Aspergillosis]. Lungs were harvested on day +3 after infection and tissue sections were stained using hematoxylin and eosin (H&E) stains to visualize inflammation and Gomori’s methenamine silver stain to visualize fungal hyphae.

Protein extraction, AspB-eGFP fusion protein purification, and LC-MS/MS Analysis

The aspB-egfp strain from [24] was grown in GMM liquid media and GMM liquid media supplemented with 1 μg/mL of caspofungin for 24 hours at 37°C. Protein extraction and pulldown were completed as described by [31,75]. Fungal mycelia were homogenized to obtain total cell lysate and AspB complexes were purified using the GFP-Trap^® affinity purification (Chromotek), as described [75]. GFP-Trap^® magnetic beads were equilibrated by washing beads three times in 500 μl ice-cold dilution buffer then resuspended in 100 μl ice cold dilution buffer. The resin suspension is then mixed with total cell lysate (10 mg total protein) and incubated at 4°C with gentle agitation for 2 hours. Beads were collected using a magnetic stand. Beads were washed in 500 μl ice-cold lysis buffer and five times with 500 μl of wash buffer. Beads were suspended in 50 μl wash buffer.

Samples were digested on-bead with trypsin followed by C18 desalting. Samples were analyzed via LC-MS/MS on a Thermo Easy nLC 1200-QExactive HF in technical duplicate. All mass spectra data was processed using MaxQuant (ver. 1.6.12.0) and searched against the Uniprot Aspergillus fumigatus proteome reviewed database (Proteome ID UP000002530). MaxQuant output was further processed via Perseus with filtering at 1% false discovery rate (FDR). Only proteins with >1 peptide were reported.

Prioritization of candidate genes and generation of deletion mutants

Candidate genes were chosen by prioritizing proteins with at least a two-fold increase, statistical significance, and known or putative roles in cell wall functions as listed in FungiFun GO term search for biological processes (Table S3) [32]. Deletions of bgt1 (Afu1g11460; fungidb.org), gel1 (Afu2g01170; fungidb.org), gel2 (Afu6g11390; fungidb.org), nsdD (Afu3g13870; fungidb.org), mapA (Afu4g06930; fungidb.org), and punA (Afu6g07470; fungidb.org) genes were obtained by replacing the gene with the 2.4 kb pyrG gene from Aspergillus parasiticus. Approximately 1 kb of promoter and terminator region of each gene were PCR-amplified from AF293 genomic DNA. Deletion constructs were generated by overlap fusion PCR and subsequently transformed into akuB^KU80 pyrG− strain, all as previously described by [73]. Primers used in transformation of all candidate strains are found in (Table S2). Transformants were validated via PCR screening.

Supplementary Material

1

Supplementary Figure 1. Caspofungin exposure alters AspB protein interactions.

(A) Principal component analysis (PCA) of the proteomics data reveals different protein composition from AspB pulldown from GMM and caspofungin treated samples. Blue color denotes those grown in GMM, red color denotes fungi grown in GMM+ 1 μg/mL caspofungin. (B) 226 significantly decreased (FC < −2, p < 0.05) proteins (blue) and 106 significantly increased (FC > 2, p < 0.05) proteins (red). Proteins under two-fold change or had a p-value greater than 0.05 are shown in black. Red horizontal line denotes p=0.05, red vertical lines denote ± 2 fold change. (C) Category of biological processes most enriched after exposure to caspofungin is hyphal growth. (D) Category of biological processes most decreased after exposure to caspofungin is protein folding. (C,D) Gene ontology enrichment for biological processes determined by using FungiFun [32].

Supplementary Figure 2. ΔnsdD and ΔpunA strains have a growth defect.

Radial growth assay shown per each day, 10⁴ conidia of each strain were inoculated on GMM agar. Plates were then incubated for 5 days, and radial growth was measured every 24 hours. (A) Assays performed at 37°C. (B) Assays performed at 30°C. All experiments done in triplicate.

Supplementary Figure 3. Δbgt1, Δgel2, and ΔnsdD have mild sensitivity to calcofluor white.

10⁴ conidia were inoculated into GMM agar supplemented with cell wall disrupting agents listed in the figure then incubated for 3 days at 37°C, with the exception of 4 μg/mL caspofungin for five days to ensure observation of the caspofungin paradoxical effect. Experiments were replicated three times.

Supplementary Figure 4. Deletion of punA results in changed colony morphology in nikkomycin Z.

10⁴ conidia were inoculated into GMM agar supplemented with the cell wall disrupting agents listed in the figure then incubated for 3 days at 30°C, with the exception of 4 μg/mL caspofungin for five days to ensure observation of the caspofungin paradoxical effect. Experiments were replicated three times and representative images are shown.