Feedback control of morphogenesis in fungi by aromatic alcohols

Abstract

Many fungi undergo a developmental transition from a unicellular yeast form to an invasive filamentous form in response to environmental cues. Here we describe a quorum signaling pathway that links environmental sensing to morphogenesis in Saccharomyces cerevisiae. Saccharomyces cells secrete aromatic alcohols that stimulate morphogenesis by inducing the expression of FLO11 through a Tpk2p-dependent mechanism. Mutants defective in synthesis of these alcohols show reduced filamentous growth, which is partially suppressed by the addition of these aromatic alcohols. The production of these auto signaling alcohols is regulated by nitrogen: High ammonia restricts it by repressing the expression of their biosynthetic pathway, whereas nitrogen-poor conditions activate it. Moreover, the production of these aromatic alcohols is controlled by cell density and subjected to positive feedback regulation, which requires the transcription factor Aro80p. These interactions define a quorum-sensing circuit that allows Saccharomyces to respond to both cell density and the nutritional state of the environment. These same autoregulatory molecules do not evoke the morphological switch in Candida albicans, suggesting that these molecular signals are species-specific.

Figure 2.

Aromatic alcohols stimulate FLO11 expression through a Tpk2p-dependent mechanism. (A) Tryptophol and phenylethanol induce the expression of FLO11. Wild-type (WT) haploid and diploid cells were inoculated at OD₆₀₀ 0.2 in SD with or without the specified aromatic alcohol (500 μM each). The cultures were incubated for 4 h at 30°C. RNA was extracted and analyzed by qRT–PCR. The haploid strain without the externally added aromatic alcohols was used as the baseline for comparison (Flo11 level was normalized to 1). (White bars) SD alone; (bars with horizontal stripes) tyrosol; (bars with vertical stripes) phenylethanol; (bars with diagonal stripes) tryptophol; (black bars) phenylethanol and tryptophol. The β-galactosidase assays using a FLO11∷lacZ reporter show the same results. (B) The autoinduction of FLO11 expression requires TPK2 and FLO8, but not elements of the MAPK pathway. Wild-type (WT) and mutant cells were inoculated at OD₆₀₀ 0.2 in SD with (black bars) or without (white bars) aromatic alcohol (500 μM phenylethanol and tryptophol). The cultures were incubated for 4 h at 30°C. RNA was extracted and analyzed by qRT–PCR. The relative levels of FLO11 transcript in mutant cells with or without the externally added alcohols were compared with that of wild-type cells without alcohols, which was normalized to 1. The data shown here were obtained from haploid strains. The diploid strains show the same pattern in FLO11 expression in response to aromatic alcohols.

Figure 3.

Saccharomyces mutants missing the ARO8 and ARO9 genes are defective in the production of aromatic alcohols and morphogenesis. (A) The biosynthetic pathways of aromatic alcohols. (B) The Saccharomyces aro8 aro9 double mutant is unable to utilize the nitrogen of aromatic amino acids and grows poorly on plates containing aromatic amino acids (400 μM each) as the sole nitrogen source. The growth defect of the aro8 aro9 strain is specific to aromatic amino acids. The mutant grows well on plates containing other amino acids (such as 400 μM Leu and Pro) as the sole nitrogen source. The aro8/aro8 aro9/aro9 diploid mutant shows the same growth defect as the aro8 aro9 haploid mutant. (C) The aro8 aro9 double mutant is defective in producing aromatic alcohols. Wild-type (WT) and aro8 aro9 mutant cells were incubated in YNB + 2% glucose in the presence of each ¹⁴C-aromatic amino acid. CM was prepared and subjected to TLC analysis. The aro8/aro8 aro9/aro9 diploid mutant shows the same defect in the production of aromatic alcohols as the haploid mutant. (D) The Saccharomyces aro8 aro9 mutant is defective in haploid invasive growth, and the defect can be suppressed by phenylethanol (PheOH). The strong adhesion by wild-type (WT) haploid cells in the absence of any alcohol presumably results from the presence of tyrosine and phenylalanine (100 μM each) in the plates, which may induce the production of endogenous aromatic alcohols. (E) The Saccharomyces aro8/aro8 aro9/aro9 diploid strain is defective in pseudohyphal growth, and the defect can be suppressed by tryptophol (TrpOH) and phenylethanol (PheOH). Wild-type (WT) and mutant diploid strains were grown on SLAD plates with or without the specified aromatic alcohols (100 μM each). (F) The aro8 aro9 strain has a lower level of FLO11 transcript than wild type, and the defect in FLO11 expression can be partially restored by the addition of tryptophol (TrpOH) and phenylethanol (PheOH). Wild-type (WT) and mutant cells were incubated at OD₆₀₀ 0.2 in SD with or without the specified aromatic alcohols (500 μM each) for 4 h. RNA was purified and analyzed by qRT–PCR. The wild-type strain without the addition of any aromatic alcohol was used as the baseline for comparison. Only the results from haploid strains are shown. The corresponding diploid strains have the same pattern of FLO11 expression. (White bars) SD alone; (bars with horizontal stripes) tyrosol; (bars with vertical stripes) phenylethanol; (bars with diagonal stripes) tryptophol; (black bars) phenylethanol and tryptophol.

Figure 4.

The production of aromatic alcohols is controlled by nitrogen source. (A) The concentration of aromatic alcohols accumulated in the media increases with decreasing ammonium concentration. Wild-type (WT) haploid cells were incubated in YNB + 2% glucose containing specified concentrations of ammonium sulfate for 4 h at 30°C before TLC analysis. The concentration of ammonium sulfate used in each lane (from left to right) is 37 mM; 5 mM; 500 μM; 50 μM; and none, but with 5 mM L-pro. Wild-type diploid cells show the same pattern as haploid cells. (B) High ammonium condition represses the expression of ARO9, ARO10, and PDC6 genes. Wild-type haploid or diploid cells were incubated in YNB + 2% glucose with either 37 mM (SD, white bars) or 50 μM NH₄⁺ (SLAD, black bars) at OD₆₀₀ 0.5 for 2 h before being collected for RNA preparation and qRT–PCR assay. The fold change in the specified gene expression was calculated as the ratio between SLAD and SD. The representative results from haploid cells are shown. The expression patterns of these genes in diploid cells are essentially the same.

Figure 5.

The production of aromatic alcohols is controlled by cell density. (A) The accumulation of aromatic alcohols in the medium increases with increasing cell density. Saccharomyces haploid cells were inoculated at 10⁵ cells/mL in minimum medium (YNB) containing 5 mM L-Pro as nitrogen source. At indicated time points the cell density (◆) was measured by OD₆₀₀ and the concentration of aromatic alcohols (●, phenylethanol; ▴, tyrosol; ◼, tryptophol) was determined by HPLC. The diploid cells have very similar growth and alcohol production curves to haploid cells. (B) The rate of aromatic alcohol production is controlled by cell density. Wild-type (WT) haploid cells were incubated in SLAD containing 50 μM specific aromatic amino acids either at low density (5 × 10⁵ cells/mL) or high density (5 × 10⁷ cells/mL) for 2 h. One milliliter of CM from the low-density culture and 10 μL of CM from the high-density culture were lyophilized and loaded on the TLC silica gel. Wild-type diploid cells show similar results. (C) High cell density up-regulates the level of ARO9 and ARO10 transcripts, and the induction requires ARO80. Wild-type (WT) and aro80 cells were incubated in liquid SLAD medium either at low (10⁵ cells/mL, white bars) or high (5 × 10⁷ cells/mL, black bars) cell densities for 60 min. Cells were collected by filtration and centrifugation. Total RNA was purified and subjected to qRT–PCR analysis. The transcript level in wild-type cells at low density was used as the baseline for comparison. The results shown here were obtained from haploid strains. The diploid strains show the same pattern. (D) Tryptophol, but not tyrosol and phenylethanol, induces the expression of ARO9 and ARO10 genes, and the autostimulation depends on ARO80. Wild-type (WT) and aro80 cells were cultured at OD₆₀₀ 0.2 in SD with or without the specified aromatic alcohol (500 μM) for 4 h. RNA was isolated and analyzed by qRT–PCR. The transcript level of the wild-type strain in SD was the baseline for calculating fold changes. Tryptophol has the same effects on the expression of ARO9 and ARO10 in both haploid and diploid strains. The results shown here were obtained from haploid strains. (White bars) SD alone; (light-gray bars) tyrosol; (dark-gray bars) phenylethanol; (black bars) tryptophol. (E) Tryptophol (TrpOH) stimulates the growth of a reporter strain P_ARO9-URA3 on uracil-minus (SD) plates, and the stimulation requires ARO80. Five microliters of 10-fold titrated Saccharomyces cells (L8199 for wild-type and L8220 for aro80 mutant) was spotted on specified plates for 2 d at 30°C. (F) Tryptophol up-regulates the transcript levels of ARO9 and ARO10 in cells at low density (only ARO9 is shown). Wild-type (WT) or aro80 cells were incubated in SLAD with (black bars) or without (white bars) 500 μM tryptophol at low (10⁵ cells/mL) or high (5 × 10⁷ cells/mL) cell densities for 60 min. RNA was isolated and analyzed by qRT–PCR. The transcript level of ARO9 in wild-type cells at low density without tryptophol was the baseline for comparison. Both haploid and diploid strains show the same pattern of gene expression. Only the results from haploid strains are shown. (G) The Saccharomyces aro80 mutant is defective in the biosynthesis of aromatic alcohols. Saccharomyces wild-type (WT) and aro80 cells were incubated in YNB + 2% glucose with either 37 mM (lanes 1,3) or 50 μM ammonium sulfate (lanes 2,4). CM was analyzed by TLC. The aro80/aro80 mutant shows the same defect in the production of aromatic alcohols as the aro80 haploid mutant. The results from haploid strains are shown.

Figure 6.

Aromatic alcohols evoke different morphological effects in C. albicans. (A) phenylethanol (PheOH) and tryptophol (TrpOH) at high concentrations (≥500 μM) inhibit the filamentous growth of Candida, whereas tyrosol (TyrOH) stimulates it. The final concentration of aromatic alcohols used in the assay was 1 mM. Bar, 25 μm. (B) Phenylethanol (PheOH) and tryptophol (TrpOH) at high concentrations (≥500 μM) inhibit biofilm formation in C. albicans, whereas tyrosol (TyrOH) promotes it. The final concentration of aromatic alcohols used in the assay was 1 mM. OD₄₂₀ reading of Candida cells without aromatic alcohols was normalized to 100%. The data represent the averages of three independent measurements. The error bars represent standard deviations.

Figure 7.

A quorum signaling pathway linking environmental sensing to morphogenesis and entry into stationary phase in S. cerevisiae. Aromatic alcohols are autosignaling molecules capable of stimulating morphogenesis in Saccharomyces by elevating the level of Flo11p through a Tpk2p-dependent mechanism. The production of aromatic alcohols is tightly controlled by environmental conditions. High ammonium represses their production by inhibiting the transcription of key genes in the biosynthesis pathway (such as ARO9 and ARO10), whereas low ammonium activates it. Cell density also controls the production of aromatic alcohols. High density stimulates the production by activating the expression of ARO9 and ARO10, whereas low cell density inhibits it. Moreover, the production of aromatic alcohols is under positive feedback control by tryptophol. Both the cell density-dependent and feedback regulation require transcription factor Aro80p. Aromatic alcohols also regulate the expression of many stationary-phase genes, suggesting a potential role in modulating the entry of Saccharomyces cells into stationary phase.