Aspergillus fumigatus RasA regulates asexual development and cell wall integrity

Abstract

The Ras family of proteins is a large group of monomeric GTPases. Members of the fungal Ras family act as molecular switches that transduce signals from the outside of the cell to signaling cascades inside the cell. A. fumigatus RasA is 94% identical to the essential RasA gene of Aspergillus nidulans and is the Ras family member sharing the highest identity to Ras homologs studied in many other fungi. In this study, we report that rasA is not essential in A. fumigatus, but its absence is associated with slowed germination and a severe defect in radial growth. The DeltarasA hyphae were more than two times the diameter of wild-type hyphae, and they displayed repeated changes in the axis of polarity during hyphal growth. The deformed hyphae accumulated numerous nuclei within each hyphal compartment. The DeltarasA mutant conidiated poorly, but this phenotype could be ameliorated by growth on osmotically stabilized media. The DeltarasA mutant also showed increased susceptibility to cell wall stressors, stained more intensely with calcofluor white, and was refractory to lysing enzymes used to make protoplasts, suggesting an alteration of the cell wall. All phenotypes associated with deletion of rasA could be corrected by reinsertion of the wild-type gene. These data demonstrate a crucial role for RasA in both hyphal growth and asexual development in A. fumigatus and provide evidence that RasA function is linked to cell wall integrity.

FIG. 1.

Deletion of A. fumigatus rasA. (A) Schematic of the rasA genomic locus from the wt, deletion (ΔrasA), and reconstituted (ΔrasA+rasA) strains. Probes used for Southern blot application are shown as small white boxes, and the relative positions of the EcoRI restriction sites are indicated. The dotted line shown in the ΔrasA+rasA strain represents plasmid sequence inserted during integration of the complementation construct. (B) Genomic Southern blot of wt, ΔrasA, and ΔrasA+rasA strains digested with EcoRI. (C) RT-PCR of the isogenic set using rasA gene-specific primers. Amplification control is A. fumigatus gpdA.

FIG. 2.

Deletion of rasA affects colony morphology and radial growth. (A) Colony morphology of the A. fumigatus wt, ΔrasA, and ΔrasA+rasA strains. Equal numbers of conidia were inoculated and allowed to grow for 48 h at 37°C on AMM. (B) Germination rates of the isogenic set. Conidia (10⁴) from each strain were inoculated onto coverslips and submerged in YG media at 37°C. The percent germination was scored as the number of conidia producing a visible germ tube among 100 total enumerated conidia. (C) Comparison of total biomass from the isogenic set. Each strain was grown for 24 or 48 h at 37°C in YG, and the total biomass was determined as described in Materials and Methods. All experiments were performed in triplicate. Measurements and error bars represent the mean ± the standard deviation (SD).

FIG. 3.

Deletion of rasA causes restricted radial growth. Colonies of the members of the rasA isogenic set were compared after 48 h of incubation at 37°C. The addition of 1.2 M sorbitol did not correct the radial growth defect (YG versus YG+sorb), but conidiation was restored. The addition of exogenous cAMP (as dibutyryl-cAMP [Db-cAMP]) to OSM did not remediate the growth phenotype, even in the presence of sorbitol (OSM+DbcAMP). Inhibition of the calcineurin pathway by the addition of 75 μM cyclosporine (Cs) to AMM resulted in radial growth inhibition of all members of the rasA isogenic set, implying that calcineurin is active in the rasA deletion mutant.

FIG. 4.

Microscopic morphology of wt and ΔrasA conidiophores on osmotically stabilized medium. Conidiophores from the wt (A and B) and ΔrasA (C to E) strains grown on YG plus 1.2 M sorbitol for 2 days at 37°C are shown. Note the septated conidiophore in panel B (arrowhead), seen in 15 to 20% of wt and ΔrasA+rasA conidiophores on sorbitol containing medium. A minority of mutant conidiophores were relatively normal (C), but even those showed blunted phialides. Many mutant conidiophores were branched (D) or bizarre and rudimentary (E). Conidiophores from the ΔrasA+rasA strain (not shown) were indistinguishable from the wt. All were photographed at 400× using differential interference contrast (panels A to C and panel E) or bright-field microscopy (panel D).

FIG. 5.

Deletion of rasA affects hyphal morphology. wt and ΔrasA hyphae grown for 16 h at 37°C in liquid AMM are shown. Higher-magnification images, shown as insets, highlight the aberrant polarized growth of the ΔrasA mutant. Scale bar of inset, 50 μm.

FIG. 6.

Deletion of rasA affects nuclear distribution in A. fumigatus. Fluorescent images of the wt and ΔrasA A. fumigatus strains grown for 16 h at 37°C and 250 rpm are shown. Arrowheads indicate the presence of septa. All strains were fixed as described in Materials and Methods and stained with propidium iodide to visualize nuclei and calcofluor white to visualize the cell wall. Note the relative chitin-bright staining of the ΔrasA cell wall. Scale bar, 20 μm.

FIG. 7.

Treatment of the rasA mutant with Congo red or Nikkomycin Z results in morphologically abnormal growth. Conidia (10⁵) from the wt and ΔrasA strains were grown in AMM with the addition of 75 μg of Congo red or 1 μg of Nikkomycin Z/ml overnight at 37°C. Images are shown from cultures with (w/) or without (w/o) 1.2 M sorbitol (Sorb) added to the medium as an osmotic stabilizer.

FIG. 8.

The ΔrasA mutant is hypersensitive to cell wall stress. Fungal growth was measured using cleavage of CMFDA by strains treated with increasing concentrations of SDS (A), Fungin (B), and caffeine (C). All experiments were performed in triplicate. Measurements and error bars represent the mean ± the SD.

FIG. 9.

Transcript abundance in members of the rasA isogenic set compared using RT-PCR. RNA from wt (□), ΔrasA (▪), and ΔrasA+rasA (░⃞) was harvested from hyphae grown overnight as described in Materials and Methods. Products of the RT-PCR were run on gels and stained with Sybr green, and the intensity in pixels was normalized to the signal from gpdA run in the same gel. These ratios were then analyzed by analysis of variance, and no significant differences were detected.