Ethylzingerone, a Novel Compound with Antifungal Activity

Abstract

Preservatives increase the shelf life of cosmetic products by preventing growth of contaminating microbes, including bacteria and fungi. In recent years, the Scientific Committee on Consumer Safety (SCCS) has recommended the ban or restricted use of a number of preservatives due to safety concerns. Here, we characterize the antifungal activity of ethylzingerone (hydroxyethoxyphenyl butanone [HEPB]), an SCCS-approved new preservative for use in rinse-off, oral care, and leave-on cosmetic products. We show that HEPB significantly inhibits growth of Candida albicans, Candida glabrata, and Saccharomyces cerevisiae, acting fungicidally against C. albicans Using transcript profiling experiments, we found that the C. albicans transcriptome responded to HEPB exposure by increasing the expression of genes involved in amino acid biosynthesis while activating pathways involved in chemical detoxification/oxidative stress response. Comparative analyses revealed that C. albicans phenotypic and transcriptomic responses to HEPB treatment were distinguishable from those of two widely used preservatives, triclosan and methylparaben. Chemogenomic analyses, using a barcoded S. cerevisiae nonessential mutant library, revealed that HEPB antifungal activity strongly interfered with the biosynthesis of aromatic amino acids. The trp1Δ mutants in S. cerevisiae and C. albicans were particularly sensitive to HEPB treatment, a phenotype rescued by exogenous addition of tryptophan to the growth medium, providing a direct link between HEPB mode of action and tryptophan availability. Collectively, our study sheds light on the antifungal activity of HEPB, a new molecule with safe properties for use as a preservative in the cosmetic industry, and exemplifies the powerful use of functional genomics to illuminate the mode of action of antimicrobial agents.

Keywords: Candida albicans; HEPB; antifungal; cosmetics; ethylzingerone; hydroxyethoxyphenyl butanone; mechanism of action.

FIG 1

Antifungal activities of ethylzingerone, triclosan, and methylparaben. (A) Chemical structures of methylparaben (MPB), triclosan (TCS), and ethylzingerone (HEPB). (B) Representative killing curves of C. albicans strain SC5314 exposed to different concentrations of each preservative in YPD medium. x axis, exposure time (min) to the indicated concentrations of each preservative; y axis, percentage of CFU counts at each time point relative to CFU counts at time point 0. (■) control with solvent alone, () MPB at 5 mg/ml (1× MIC), (●) HEPB at 10 mg/ml (1× MIC), () HEPB at 20 mg/ml (2× MIC), and (□) TCS 0.062 mg/ml (1× MIC).

FIG 2

Transcript profiling in C. albicans exposed to ethylzingerone. (A) Heat maps of the 50 highest (left panel, red) and lowest (right panel, green) transcriptionally modulated genes (log₂-transformed ratios are shown and color scale indicates the maximum and minimum expression ratios, +/−10.08) following exposure of C. albicans strain SC5314 to 4 mg/ml (0.4× MIC) or 10 mg/ml (1.0× MIC) HEPB for 10, 30, and 60 min (combination of 2 or 3 biological replicates in each condition). The most upregulated (descending signal intensity, sorted by average expression under all conditions, left panel) or downregulated (ascending signal intensity, sorted by average expression under all conditions, right panel) genes in HEPB-treated versus untreated cells are indicated with their corresponding name or systematic nomenclature on the right side of each panel. Genes highlighted with a blue asterisk are those that are transcriptionally modulated by activation of transcription factor Tac1p (20), while genes highlighted with a red asterisk are those involved in amino acid biosynthesis. Heat maps were constructed using Genesis version 1.8.1 (50). (B) K-means profile plots of 2 selected clusters (cluster 1, 118 genes, upper panel and cluster 2, 83 genes, lower panel) out of 10 clusters generated through mining of the complete transcript profiling data set (Table S1) using Genesis version 1.8.1 (50). The expression dynamics of each gene (log₂-transformed ratios, gray line) are plotted on the y axis, whereas the experimental condition is indicated on the x axis (bottom). (C) GO term enrichment scores (black bars, representing the negative value of log₁₀ transformed P values shown on the x axis) of the significantly enriched functional categories (P value <0.05) among the 118 and 83 genes from K-means clusters 1 (upper chart) and 2 (lower chart), respectively. The GO terminologies are indicated on the y axis. The number of genes belonging to each GO terminology are indicated between parentheses.

FIG 3

Comparative analysis of the transcriptomics data. Hierarchical clustering using average linkage WPGMA (clustering of both genes and conditions) showing the relationships between the distinct 18 compound treatments (top). Each gene is represented by a rectangle colored according to the level of upregulation (red) or downregulation (green) as indicated in the colored scale showing adjusted maximal (+5.0) and minimal (−5.0) log₂-transformed ratios. The relatedness between conditions is shown on the upper cladogram, whereas relatedness between gene expression profiles is indicated on the left cladogram. The hierarchical clustering heatmap was generated using Genesis version 1.8.1 (50).

FIG 4

Phenotypic profiling in S. cerevisiae links HEPB mode of action to tryptophan availability. (A) Histogram depicting the relative abundance of each group of S. cerevisiae mutants (histogram bins) measured as the log₂-transformed ratio of barcode signal intensity in HEPB-treated samples (n = 3) compared to untreated control sample (x axis). The number of strains per histogram bin are shown on the y axis. Mutants with significantly decreased abundance following HEPB treatment are shown on the left part of the histogram, whereas those with increased relative abundance are shown on the right part of the histogram. (B) Parental BY4742 (gray bar) and the trp1Δ mutant derivative (white bar) were grown in the absence (−) or presence (+) of 0.4 mg/ml tryptophan in YPD medium (YPD) supplemented (+2 mg/ml HEPB, +1 mg/ml HEPB) or not (YPD) with 2 mg/ml or 1 mg/ml HEPB. Generation times (in hours) of each strain under each condition are indicated on the y axis calculated as the mean of 3 independently grown cultures with error bars denoting standard deviations. (C) Growth curves of the trp1Δ mutant grown in YPD medium (YPD) or in YPD medium supplemented with 2 mg/ml HEPB (+2 mg/ml HEPB) are depicted in different colors depending on the identity of the amino acid being added to the growth medium. Turbidity (OD₆₀₀, y axis) was recorded every 5 min as a function of time (hours, x axis) in a Tecan Sunrise device.

FIG 5

Chemical-genetic interactions of TCS and MPB with S. cerevisiae mutants that are sensitive to HEPB. Fitness profiling matrix displaying the relative abundance of mutant strains sod2Δ, aro7Δ, vrp1Δ, sac1Δ, cin8Δ, dal81Δ, erg2Δ, gcn4Δ, sod1Δ, and trp1Δ following exposure to TCS (15 and 20 μg/ml), MPB (300 and 400 μg/ml), and HEPB (937 and 1,250 μg/ml). Fitness defect intensities (numerical values) are also displayed as colored squares, according to the color scale shown at the bottom of the panel. Negative values indicate decreased abundance of the corresponding mutant.

FIG 6

C. albicans trp1Δ/trp1Δ and gcn4Δ/gcn4Δ mutants are sensitive to HEPB treatment. (A) Parental CAI4 (TRP1/TRP1, gray bar) and the trp1Δ/trp1Δ mutant derivative (white bar) were grown in YPD medium supplemented (5 mg/ml HEPB) or not (control) with 5 mg/ml HEPB. Generation time (in hours) of each strain under each condition are indicated on the y axis, calculated as the mean of values from 3 independently grown cultures with error bars denoting standard deviations. Asterisk, P < 0.05 based on a Welch’s t test comparing mean values of the trp1Δ/trp1Δ mutant to those of the parental strain TRP1/TRP1 in the presence of HEPB (5 mg/ml HEPB). (B) HEPB susceptibility of strains DAY286, gcn4Δ/gcn4Δ (gcn4^−/−), and SC5314 was tested by spot assay on YPD plates supplemented (or not supplemented, left panel, control) with 12.5 mg/ml of HEPB (12.5 mg/ml HEPB, right panel). Plates were incubated at 30°C for 3 days.

FIG 7

Simplified schematic representation of the aromatic amino acid biosynthetic pathway. A sequence of enzymatic reactions encoded by many ARO and TRP genes are crucial for the biosynthesis of aromatic amino acids. Specific steps from the pentose phosphate pathway (top box, left) and glycolysis (top box, right) generate erythrose-4-phosphate and phosphoenolpyruvate, which are processed by the products of ARO and TRP genes to generate tryptophan (whose chemical structure is shown at the bottom right), tyrosine, and phenylalanine. Tryptophan can also be taken up from the medium owing to the activity of a low-affinity permease encoded by TAT1 (gray oval). Genes with a role in amino acid biosynthesis whose deletion strongly sensitizes S. cerevisiae to HEPB treatment are colored in red.