The link between morphotype transition and virulence in Cryptococcus neoformans

Abstract

Cryptococcus neoformans is a ubiquitous human fungal pathogen. This pathogen can undergo morphotype transition between the yeast and the filamentous form and such morphological transition has been implicated in virulence for decades. Morphotype transition is typically observed during mating, which is governed by pheromone signaling. Paradoxically, components specific to the pheromone signaling pathways play no or minimal direct roles in virulence. Thus, the link between morphotype transition and virulence and the underlying molecular mechanism remain elusive. Here, we demonstrate that filamentation can occur independent of pheromone signaling and mating, and both mating-dependent and mating-independent morphotype transition require the transcription factor Znf2. High expression of Znf2 is necessary and sufficient to initiate and maintain sex-independent filamentous growth under host-relevant conditions in vitro and during infection. Importantly, ZNF2 overexpression abolishes fungal virulence in murine models of cryptococcosis. Thus, Znf2 bridges the sex-independent morphotype transition and fungal pathogenicity. The impacts of Znf2 on morphological switch and pathogenicity are at least partly mediated through its effects on cell adhesion property. Cfl1, a Znf2 downstream factor, regulates morphogenesis, cell adhesion, biofilm formation, and virulence. Cfl1 is the first adhesin discovered in the phylum Basidiomycota of the Kingdom Fungi. Together with previous findings in other eukaryotic pathogens, our findings support a convergent evolution of plasticity in morphology and its impact on cell adhesion as a critical adaptive trait for pathogenesis.

Figure 1. Overexpression of MAT2 causes constitutive activation of pheromone signaling.

(A) The C. neoformans pheromone signaling pathway. Pheromone signaling is triggered by environmental cues (mating cues) and it turns on the master regulator Mat2, which in turn activates pheromone signaling, thereby constituting a self-reinforcing system. Activated pheromone signaling determines the output mating-relevant behaviors (e.g. formation of shmoo cells and mating projections, and initiation and cell contact and cell fusion). (B) MF1α (pheromone) and other mating signal genes (not shown here) were highly induced in a x α cocultures under the mating-inducing condition (V8), but not under mating-suppressing conditions (YPD and Serum). Because unisexual mating is not observed in the wildtype strain, the induction of pheromone was evaluated during bisexual mating with the coculture of H99α and its congenic partner KN99a incubated on different media for 72 hr. The expression level of MF1α during bisexual mating on V8 medium was arbitrarily set as 1 for comparison. (C, D, E and F) In order to analyze the effect of the deletion or the overexpression of MAT2 on the key elements of the pheromone pathway and to avoid potential complication due to higher expression of these elements in the presence of a compatible mating partner under mating-inducing conditions, only α strains alone in the H99 background were used in the these assays. Overexpression of MAT2 constitutively activated pheromone signaling in single strain cultures under all the conditions tested. The expression patterns of MAT2 (C), MF1α (D), STE3 (pheromone receptor gene) (E), and STE6 (pheromone transporter gene) (F) in wildtype (H99), and its derived mat2Δ mutant and the P_GPD1-MAT2 strain were shown. Gene expression levels in the wildtype H99 grown on V8 medium were arbitrarily set as 1 for comparison. Cells were cultured on different media for 72 hr.

Figure 2. Activated pheromone signaling triggers formation of shmoo cells or mating projections, but not filaments under mating-suppressing conditions.

(A) Overexpression of MAT2 drove filamentation during unisexual mating (α cell culture alone) or bilateral mating (a-α cocultures) only under mating-inducing condition. (B) The pheromone overexpression strain P_GPD1-MAT2 was able to form shmoo-like cells irrespective of culture conditions. White arrows indicate shmoo-like cells and the red arrow indicates a hypha cell. (C and D) Activated pheromone signaling only induced the transcription of ZNF2 under mating-inducing condition either during unisexual mating (α cell culture alone) (C) or during bisexual mating (a-α cocultures) (D). Transcript levels of ZNF2 in the wildtype H99 grown on YPD agar were arbitrarily set as 1 for comparison. The cells were cultured on different media for 72 hr.

Figure 3. Znf2 is a master regulator of filamentation.

(A) Wildtype congenic pair H99α and KN99a, and their derived P_GPD1-ZNF2 strains were grown at 22°C on V8 juice agar medium (mating-inducing condition, scale bar: 200 µm) for 5 days or in YPD liquid medium (mating-suppressing condition, scale bar: 20 µm) (see Figure S1 for images of strains in serotype D backgrounds). (B) The mfα1-3Δ, mat2Δ, and ste12αΔ mutants in the JEC21 background and their transformants with the P_GPD1-ZNF2 construct were incubated on V8 agar medium for 2 days. Wildtype JEC21 can self-filament sporadically and poorly only after prolonged incubation (>1 week) under this condition. (C) For cell fusion assays, auxotrophic parental cells of either a or α mating type were cultured alone or together on V8 agar medium for 24 hrs. Cells were then collected, washed, and spotted onto rich YPD medium for growth control or onto minimal YNB medium to select prototrophic cell-fusion products. The images of cocultures on V8 agar medium are shown to the left. See Figure S2 and 3 for the effect of Znf2 and Mat2 on hyphal maintenance under host conditions, as well as expression of genes related to cell fusion and mating projection formation during mating. (D) The P_CTR4-2-ZNF2 strain and the wildtype H99 were incubated on YPD media that contain either BCS (inducer) or CuSO₄ (inhibitor) of varied concentrations. Cells scraped from the colony above were examined microscopically (shown below). 1: 25 µM CuSO₄, 2: 0 CuSO₄, 3: 5 µM BCS, 4: 25 µM BCS, 5: 50 µM BCS, 6: 100 µM BCS, 7: 200 µM BCS, 4, 8: 25 µM CuSO₄, and 9: 200 µM BCS.

Figure 4. Znf2 governs bi-directional morphological transition and morphotype maintenance.

Yeast cells of the P_CTR4-2-ZNF2 strain were incubated in YPD liquid medium supplemented with 200 µM BCS at 30°C. At 60 hrs, the filaments were collected, washed, and transferred to fresh YPD liquid medium containing 25 µM CuSO₄. The wildtype strain remained in the yeast form under such conditions. Images were taken at indicated time points.

Figure 5. Znf2 links morphogenesis and pathogenicity.

(A) Mice were infected intranasally with either the wildtype H99 or the P_GPD1-ZNF2 strain. Survival rate was plotted against days after inoculation. (B) Mice were infected with the wildtype H99, the znf2Δ mutant, and the P_GPD1-ZNF2 strain. Fungal burden in the lungs and the brains (Figure S4) was determined at DPI 10. Differences among the groups are statistically significant. (C) Lung tissues were stained with the Grocott Methenamine Silver stain. Fungal cells appear black or dark brown. Only tissues at DPI 12 are shown here. See Figure S4 for the correlation between brain fungal burden and Znf2 activity; see Figure S5 for the fungal cell morphology during the course of infection.

Figure 6. Znf2 regulates cell adhesion in Cryptococcus.

(A) The P_CTR4-2-ZNF2 strain and a strain transformed with the empty vector (control) were incubated on YPD agar medium that contains either BCS (inducer) or CuSO₄ (inhibitor). Cells scraped from the colony were examined microscopically (images below). (B) Wildtype H99 and the P_CTR4-2-ZNF2 strain were pre-grown in YPD medium containing 25 µM CuSO₄ (inhibitor) for 12 hrs (no cell aggregation). The yeast cells were washed twice, inoculated into fresh YPD medium containing 200 µM BCS (inducer) and grown for additional 4 hrs with shaking before they were allowed to settle. Cell concentration of the upper stagnate culture (OD₆₀₀) was measured every 30 min. (C) Cells from the bottom of the cultures were examined microscopically. (D) Overexpression of ZNF2 leads to the formation of hyphal bundles. Wildtype H99 and its derived P_CTR4-2-ZNF2 strain were grown on YPD BCS agar plate at 22°C for 5 days. Multiple hyphae were attached together forming bundles in the P_CTR4-2-ZNF2 strain (scale bar: 100 µm). (E) Znf2 controls invasive growth. Wildtype XL280 and its derived znf2Δ mutant and the P_GPD1-ZNF2 strain were grown on YPD agar medium at 22°C for 5 days. The left column shows the original colonies; the middle column shows invasive cells after surface cells were washed off; and the right column shows enlarged images of the remaining invasive cells.

Figure 7. Znf2 regulates the expression of many extracellular Proteins.

(A) Classification of genes differentially expressed in znf2Δ mutants compared with wildtype. (B) Selected genes predicted to encode extracellular proteins were also differentially expressed in the ZNF2 overexpression strains by qPCR. (C) Gene overexpression strains in the H99 background were grown on YPD medium at 22°C for 3 days. (D) H99, the P_CTR4-2-ZNF2 strain, and the P_CTR4-2-CFL1 strain were pre-grown for 12 hrs in YPD liquid medium containing CuSO₄ (inhibitor). The yeast cells were washed twice and then incubated on YPD agar medium containing BCS (inducer) for 3 days. Cells scraped from the colony above were examined microscopically (shown below).

Figure 8. Cfl1 is a hypha-specific adhesin regulated by Znf2.

(A) Diagram of the reporter system. The znf2Δ mutant carries both the P_GAL10-ZNF2 construct (ZNF2 driven by the galactose-inducible promoter) and the P_CFL1-CFL1-m-Cherry construct (fluorescent Cfl1 driven by its native promoter). The output behaviors (fluorescence, wrinkled colony morphology, and filamentation) are determined by the input signals (inducer/galactose or inhibitor/glucose). (B) The reporter strain showed wrinkled colony morphology and filamentation under the inducing condition (galactose). Under the suppressing condition (glucose), only smooth yeast colony without fluorescent was observed. (C and D) Cfl1-mCherry under the control of CFL1 native promoter was detected in hyphae but not in yeasts. It was detected during both unisexual mating (characteristic unfused clamp cells, pointed by the black arrow) and bisexual mating (characteristic fused clamp cells, pointed by red arrow). Scale bars: 10 µm. (E) The coding region of the predicted N-terminal signal peptide of CFL1 was deleted in frame in the construct of P_CTR4-2-CFL1(sigPΔ)-mCherry. Both the P_CTR4-2-CFL1-mCherry strain and the P_CTR4-2-CFL1(sigPΔ)-mCherry strain showed cherry fluorescence under the inducing condition in the presence of BCS. Wrinkled colony morphology was observed in the P_CTR4-2-CFL1-mCherry strain but not in the P_CTR4-2-CFL1(sigPΔ)-mCherry strain. (F) The P_CTR4-2-CFL1-mCherry strain and the P_CTR4-2-CFL1(sigPΔ)-mCherry strain were grown in YPD liquid medium at 22°C for 7 days. The cultures with the same cell density were centrifuged and the supernatants were collected and filtered to remove the residual cells. Two microliters of the corresponding culture supernatants were spotted onto a glass slide and observed microscopically.

Figure 9. Cfl1 plays important roles in hyphal morphogenesis and pathogenicity in Cryptococcus.

(A) Cfl1 affects hyphal formation in the hyper-filamentous wildtype strain XL280 under both mating-inducing and mating-suppressing conditions. (B) Deletion of CFL1 reduced filamentation during bisexual mating. (C and D) Mice were infected intranasally with either wildtype H99 or the P_GPD1-CFL1 strain. Survival rate was plotted against days after inoculation. Fungal burden in the lungs at DPI 7 was shown. Differences among the groups are statistically significant (p<0.05).