Fig 1.
Construction of a protein kinase disruption library in A. fumigatus by CRISPR/Cas9-mediated gene editing.
A) Miniaturized protoplast transformations were carried out in 96-well plates, with a final total volume of 200 μl per well, and each well representing an attempted disruption of a single protein kinase gene. B) After the transformation process, the entire contents of each well were spread onto individual sorbitol minimal medium (SMM) agar plates and allowed to recover overnight at room temperature before overlaying with hygromycin-containing top agar for selection. Following these transformation procedures, typically 10 to 30 transformant colonies were evident on each selection plate after 3 to 4 days of incubation at 37°C. However, due to the high efficiency of gene targeting with the CRISPR/Cas9 system, only 3 to 4 colonies per transformation were required to be isolated for genotypic screening. C) Putative transformants were subjected to genotypic analyses by PCR to confirm proper integration of the repair template for gene disruption. These PCR analyses included screens with allele specific primer sets PF/PR and PF/IPR, pictured above. PF = Forward screening primer. PR = Reverse screening primer. IPR = Internal reverse screening primer complementary to HygR sequence. HygR = Hygromycin Resistance cassette, utilized as the repair template for gene disruption. All kinase genes were targeted for disruption at the 5’ end of the gene (within the first exon, where possible), as indicated by the placement of the protospacer and protospacer adjacent motif (PAM, red bold font) above.
Fig 2.
Colony morphologies of selected A. fumigatus protein kinase disruption mutants.
96-hr colony morphologies of severely (A) and moderately (B) growth restricted protein kinase disruption mutants, as well as colony morphologies of mutants that are not growth restricted but display reduced conidiation (C). Ten thousand conidia from each strain were point inoculated onto the center of minimal media agar and cultured for 96 hrs at 37°C.
Fig 3.
Multiple protein kinases contribute to cell wall integrity in A. fumigatus.
A) Protein kinase gene disruption mutants displaying increased susceptibility to both cell wall disrupting agents, calcofluor white (CFW) and congo red (CR), by spot-dilution assay when compared to the wild type parent (CEA10). B) Protein kinase gene disruption mutants displaying hyper-susceptibility to only CR when compared to the parent strain. C) Protein kinase gene disruptants displaying increased resistance to CFW (pkaC1-1) or to varying concentrations of CR (kfsA-1, cmkA-1, and stk22-1). For each target protein kinase gene, the systematic name is listed with the strain name given in parentheses. Strain names were designed using either the previously published or putative (based on homology to Aspergillus nidulans) gene names with the addition of “-1” to indicate a disruption mutation of that gene. GMM = glucose minimal media with no CFW or CR added. For all assays, conidial inocula were applied at 104, 103, 102, and 101 total conidia and plates were incubated at 37°C for 72 hrs.
Fig 4.
The Septation Initiation Network (SIN) kinases are each required for hyphal septation and protection against cell wall damage in A. fumigatus.
A) The putative core SIN pathway in A. fumigatus based on signal transduction models constructed for Schizosaccharomyces pombe and Aspergillus nidulans. A protein kinase cascade, initiated by activation of the SepH kinase through interaction with the GTP-bound GTPase, Spg1, leads to downstream activation of the SepL and SidB kinases to eventually promote initiation of septation. SepL and SidB are shown with their putative regulatory binding partners, SepM and Mob1, respectively. B) Deletion of sepH (ΔsepH) phenocopies sepL and sidB disruption (sepL-1 and sidB-1, respectively) as evidenced by restricted colony size and loss of conidiation (i.e., white colony formation). Complementation of SIN activity in the sepL-1 disruption mutant by gene replacement (sepL-1-C’) results in full growth recovery and conidiation. Ten thousand conidia from each strain were spot-inoculated onto the center of a GMM agar plate and cultured for 96 hrs at 37°C. C) Loss of any single SIN kinase results in absence of growth in the presence of the cell wall destabilizing compounds CFW or CR. Conidia from each strain were spot inoculated in descending concentrations onto GMM alone or GMM containing either 40 μg/ml CFW or CR. D) Loss of any single SIN kinase results in the absence of septa in mature hyphae. Conidia from each strain were cultured to mature hyphae (16 hrs at 37°C) and subsequently stained with calcofluor white (CFW) and propidium iodide (PI) to visualize septa and nuclei, respectively. White arrowheads indicate septa in the CEA10 (parent) and sepL-1 complemented (sepL-1-C’) strains. No septa were evident in the ΔsepH, sepL-1 or sidB-1 mutants.
Fig 5.
The A. fumigatus SIN kinases are required for survival under echinocandin stress.
A, B) Loss of sepH, sepL, or sidB increases susceptibility to echinocandins in a modified E-test assay. 1 x 106 total conidia in 500 μl sterile water from the wild type parent (CEA10), the sepH deletion (ΔsepH), or the sepL (sepL-1) or sidB (sidB-1) disruption strains were spread evenly over GMM agar plates. E-test strips for caspofungin (A) or micafungin (B) were applied and assays incubated for 48 hrs. Note the zone-of-clearance with no detectable growth for the ΔsepH, sepL-1 and sidB-1 mutants in the presence of either echinocandin. Insets show representative, drug-free minimal media culture plates onto which a single agar plug from the zone-of-clearance for each assay was sub-cultured. Multiple agar plugs (n = 10), taken from within 1 cm of the E-test strip and between the 32 and 0.25 ug/ml markers, were sub-cultured in the same manner for each assay. Note lack of growth for the SIN kinase mutant subcultures. CAS = caspofungin, MFG = micafungin. C) Quantitation of viability by CFDA staining of the CEA10 control and SIN kinase mutants in the absence of echinocandin stress. Conidia from each strain were germinated for 12 hrs and subsequently stained with 5-carboxyfluorescein diacetate (CFDA) to detect live hyphal elements. D) Quantitation of viability by CFDA staining of the strain set in the presence of caspofungin. Conidia from each strain were cultured for 12 hrs and 24 hrs at 37°C in the presence of 0.5 μg/ml caspofungin and subsequently stained with CFDA to detect live microcolonies. CFDA positivity was scored for 100 microcolonies in each experiment and all assays were completed in triplicate. Data were averaged for each strain and treatment. One-way ANOVA and Dunnett’s multiple comparisons post hoc analyses indicated differences in CFDA staining in the absence of micafungin were not significant (n.s.), whereas the ΔsepH, sepL-1 and sidB-1 mutants were significantly less viable after 12 and 24 hrs growth in the presence of caspofungin. ****p<0.0001; ***p = 0.0001. E) Echinocandin stress during early growth stages leads to death of the SIN kinase mutants. White arrowhead denotes example of a microcolony stained positive with CFDA (bright green). Black arrowhead denotes a CFDA-negative microcolony. Dash-lined arrows denote dead (CFDA-negative) hyphal compartments of CFDA-positive microcolonies only seen in the CEA10 control.
Fig 6.
Hyphae of SIN kinase mutants exhibit extensive damage in the presence of echinocandin.
Analysis of hyphal integrity using propidium iodide (PI) permeability as a measure of damage in response to echinocandin stress. Mature hyphae from the CEA10, ΔsepH, sepL-1, and sidB-1 strains were stained with PI (12.5 μg/ml) before (A) or after 2 hours (B) exposure to micafungin (0.5 μg/ml). Upper panels are brightfield images and lower panels are fluorescence acquired. Hyphae from all strains exhibited minimal staining with no exposure to echinocandin, suggesting intact cell walls (A). Limited staining of hyphal compartments was noted in the CEA10 parental strain after 2 hrs micafungin exposure, suggesting cell wall damage limited by the presence of septa (B, lower panel inset, white arrowheads denote hyphal compartment). In contrast, extensive PI staining was induced after micafungin treatment in each of the SIN kinase mutant strains (B). All fluorescence images were acquired at using identical exposure. Scale bar = 100 μm.
Fig 7.
SIN kinase activity is required for virulence in mouse models of invasive aspergillosis.
Mice (n = 8 / group for ΔsepH, sepL-1, sidB-1 and sepL-1-C’; n = 16 for CEA10) were chemotherapeutically immune suppressed with both cyclophosphamide and triamcinolone acetonide (A) or triamcinolone acetonide alone (B) and inoculated with 1 x 105 conidia of the indicated strain. Survival was followed for 15 days post-inoculation. Statistical analyses (Mantel-Cox Log-rank test) identified significantly reduced virulence for all SIN kinase mutant strains vs. the CEA10 control.
Fig 8.
Loss of virulence among the SIN kinase mutants is associated with lack of tissue invasion.
Low- and high-magnification photomicrographs of Gomori methenamine silver (GMS)-stained lung tissue sections from the CEA10, ΔsepH, sepL-1 and sidB-1 at day +4 post-inoculation. Mice were immune suppressed with triamcinolone acetonide and inoculated with each strain as described for the previous survival studies. Hyphae (black stained fungal elements) from the CEA10 strain were noted to invade lung tissue, forming fulminant lesions. In contrast, growth of each SIN kinase mutant was limited to the airways with minimal to no tissue invasion (white arrowheads). Rare tissue invasion was associated with loss of polarity maintenance (ΔsepH inset panel). Scale bar = 50 μm.
Fig 9.
Loss of virulence in the SIN kinase mutants is characterized by decreased fungal burden and host response to infection.
A) Analysis of fungal burden by qPCR at day +4 post inoculation. Mice (n = 5 / group) were immune suppressed with cyclophosphamide and triamcinolone acetonide and inoculated with 1x 106 conidia from each strain. Data are represented as nanograms of A. fumigatus specific DNA in 500 ng of total DNA. * p < 0.02. Quantitation of IL-1β (B) and TNFα (C) revealed decreased host response in SIN kinase mutant infected mice. Mice (n = 5 / group) were immune suppressed as indicated for fungal burden analysis and lung tissue was removed at day +4 post-inoculation, homogenized and analyzed by ELISA. **p = 0.0024 for (B); **p = 0.0031 for (C). An in vitro IL-1β release assay uncovered decreased induction of inflammasome activation by the SIN kinase mutants. D) Conidia from each strain were co-incubated with phorbol 12-myristate 13-acetate (PMA)-activated THP-1 cells (MOI 10:1) for 16 hrs and supernatants analyzed by ELISA for IL-1β concentration. ***p < 0.0001; **p = 0.0014. E) Inflammasome dependence of IL-1β release was established by co-culturing PMA-activated WT (THP1-null), Nlrp3−/− (THP1-KO-NLRP3), and Asc−/− (THP1-KO-ASC) THP-1 cells with CEA10 conidia (MOI 10:1) as indicated for (D). ***p < 0.0001. F) Inflammasome dependence of Aspergillus-induced IL-1β release was further confirmed by repeating this assay in the presence of the inflammasome inhibitor, MCC950 (10 μM). ***p < 0.0001. All experiments were conducted in technical replicates (n = 4) and repeated independently in triplicate. Statistical comparisons in (A), (B), (C), and (D) were made by one-way ANOVA with Dunnett’s multiple comparisons test post hoc and represent comparison of each SIN kinase mutant to the CEA10 control. Statistical comparisons in (E) were made by one-way ANOVA with Dunnett’s multiple comparisons post hoc and represent the NLRP3-/- and ASC-/- versus WT control. The statistical comparison of MCC950 versus vehicle in (F) was made by unpaired T-test.
Fig 10.
Loss of hyphal septation improves echinocandin therapy characterized by clearance of fungal burden from lung tissue.
A) Survival analysis of mice infected with either the CEA10 or sepL-1 mutant strain with and without micafungin therapy. All mice were immune suppressed through intraperitoneal injection of cyclophosphamide on days -3, +1, +4, and +7 and a single subcutaneous injection of triamcinolone acetonide on day -1. Mice were inoculated with 1 X 106 conidia of the indicated strain suspended in 20 μl of sterile saline on day 0 and then received three separate intraperitoneal injections of micafungin at either 1 mg/kg (MFG1) or 2 mg/kg (MFG2) on days +1, +2 and +3. Statistical comparisons were made by Mantel-Cox log-rank test and represent each sepL-1 mutant experimental arm compared to its CEA10 control (i.e., CEA10 saline vs. sepL-1 saline, CEA10 MFG1 vs. sepL-1 MFG1, and CEA10 MFG2 vs. sepL-1 MFG2). B) CEA10 and sepL-1 residual lung tissue burden at day 0 and day 4 with and without micafungin 2 mg/kg therapy. Organ cultures are shown from two representative animals from each treatment group. Note that the sepL-1 infected mice treated with micafungin 2 mg/kg are culture negative at day 4 post-inoculation. MFG = micafungin.
Fig 11.
The A. fumigatus genes encoding myosin light chain (mlcA) and alpha-actinin (ainA) are required for hyphal septation and echinocandin resistance.
A) Characterization of septation by CFW staining of the wild type parental strain (CEA10) and the mlcA (ΔmlcA) and ainA (ΔainA) deletion mutants. Conidia from each strain were cultured to mature mycelial growth on sterile coverslips submerged in minimal media. Microscopic analysis of CFW-stained cultures revealed fully formed, normal septa in the CEA10 control strain (black arrow), whereas ΔmlcA formed aseptate hyphae with brightly stained puncta of cell wall material and ΔainA developed only aseptate hyphae. B) Anidulafungin E-test assays for each strain. Note the zone-of-clearance of the ΔmlcA and ΔainA strains which show the complete absence of growth.
Fig 12.
Loss of septation caused by deletion of mlcA or ainA results in avirulence associated with lack of tissue invasion.
Survival analysis and GMS-stained tissue histology from immune suppressed mice infected with the ΔmlcA isogenic strain set (A) and (B) or the ΔainA isogenic strain set (C) and (D), respectively. Mice (n = 8 / strain) were immune suppressed with a single, subcutaneous injection of triamcinolone acetonide (40 mg/kg) on Day -1. On Day 0, all mice were intranasally administered 100,000 conidia suspended in 20 μl of sterile saline. No deaths were recorded in sham treated mice (sterile saline alone, n = 8). Statistical comparisons were made by Mantel-Cox log-rank test. Black arrows denote non-invasive fungal growth identified only in the airways of ΔmlcA and ΔainA infected mice.